Measuring the Information Society Report

Transcript

1 Internati onal Measuring Telecommunicati on Union the Information Place des Nati ons CH-1211 Geneva 20 Switzerland Society Report ISBN: 978-92-61-21431-9 2016 6 4 0 4 3 3 9 6 1 2 1 4 2 1 9 7 9 8 Printed in Switzerland Geneva, 2016

2

3 Measuring the Information Society Report 2016

4 © 2016 ITU International Telecommunication Union Place des Nations CH-1211 Geneva Switzerland Original language of publication: English All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the International Telecommunication Union. ISBN: 978-92-61-21421-0 (paper version) 978-92-61-21431-9 (electronic version) 978-92-61-21581-1 (epub) 978-92-61-21591-0 (moby) Measuring the Information Society Report 2016 ii

5 Foreword It is my pleasure to present to you the latest Measuring the Information Society edition of the Report . This annual report presents a global and regional overview of the latest developments regarding information and communication technologies (ICTs), based on internationally comparable data and agreed methodologies. It aims to stimulate the ICT policy debate in ITU Member States by providing an objective assessment of how countries have performed in the field of ICT and by highlighting areas that need further improvement. One of the core features of the Report is the ICT Development Index (IDI). This year’s results show that nearly all of the 175 countries covered by the index improved their IDI values between 2015 and 2016. During the same period, stronger improvements have been made on ICT use than access, mainly as a result of strong growth in mobile-broadband uptake globally. This has allowed an increasing number of people, in particular from the developing world, to join the information society and benefit from the many services and applications provided through the Internet. This year, for the first time, the Report also shows countries’ rankings according to their improvement in IDI value. The results show strong improvements in performance throughout the world; a number of middle- income developing countries in particular are reaping the benefits of more liberalized and competitive ICT markets that encourage innovation and ICT uptake across all sectors. Despite these encouraging developments, we need to focus on the countries that are among the least connected in the world. Urgent action is required to address this persistent digital divide if we want to achieve the Sustainable Development Goals (SDGs) enshrined in the 2030 Agenda for Sustainable Development. For example, the Report shows that in some low-income countries, between 20 and 40 per cent of people still do not own a mobile phone and that the gender gap in mobile phone ownership is substantially higher. This year’s Report takes a closer look at barriers to Internet uptake. New data show that while 84 per cent of the world's people live in an area where mobile-broadband services are offered, only 47 per cent are actually using the Internet. While infrastructure deployment is crucial, high prices and other barriers prevent people from entering the digital world. The price of the service (and of the device) remains a critical determinant for whether people make use of ICTs. I am pleased to see that, globally, the prices for fixed and mobile communication services continued to fall over the past year. The reduction in mobile-broadband prices is particularly pleasing, as it leads not only to more people being connected to the Internet but also to more intense Internet usage among those who are already online. The availability and affordability of high-speed fixed-broadband services nevertheless remain a challenge in the majority of low-income countries. In the world’s least developed countries, a fixed-broadband plan with a minimum of 1GB of data per month still corresponds, on average, to over 60 per cent of GNI per capita. In addition, in those least developed countries where the service is offered, speed and quality are usually lower than in developed countries. This is a constraint not only for the domestic business sector but also in Measuring the Information Society Report 2016 iii

6 terms of using ICTs to accelerate the achievement of the Sustainable Development Goals (SDGs), through e-agriculture, e-health, e-education, e-governance, gender equality, just to mention a few. Education and income levels are strong determinants, not only of whether or not people use the Internet, but also of how they use it. The Report finds that Internet users with higher levels of education use more advanced services, such as e-commerce and online financial and government services, to a higher degree than Internet users with lower levels of education and income levels, who use the Internet predominantly for communication and entertainment purposes. In line with the more integrated development approach adopted in the 2030 Agenda for Sustainable Development, ITU is working in close cooperation with other United Nations agencies and the private sector to raise awareness of and harmonize development policy approaches in order to create an enabling collaborative environment. This - with no doubt - will help us leverage the full potential of ICTs for the achievement of socio-economic development for all. Brahima Sanou Director Telecommunication Development Bureau (BDT) International Telecommunication Union Measuring the Information Society Report 2016 iv

7 Acknowledgements The 2016 edition of the Measuring the Information Society Report was prepared by the ICT Data and Statistics Division within the Telecommunication Development Bureau of ITU. The team comprised Susan Teltscher (Head of Division), Fredrik Eriksson, Vanessa Gray, Esperanza Magpantay, Lourdes Montenegro and Ivan Vallejo. David Souter, consultant to ITU, provided substantive contributions to Chapters 1, 2 and 3. Halvor Sannæs and Josie Sephton from Strategy Analytics Ltd. provided substantive inputs to Chapter 4. Joss Gillet from GSMA Intelligence, Mariama Deen-Swarray from Research ICT Africa, Alexander Moler from InterMedia and Shazna Zuhyle from LIRNEasia provided useful comments and insights for data used in Chapter 5. Helpful inputs were received from Daniela Pokorna during her internship at ITU. Nathalie Delmas provided statistical and desktop publishing assistance. The work was carried out under the overall direction of Cosmas Zavazava, Chief, Project Support and Knowledge Management Department, Telecommunication Development Bureau. Halvor Sannæs and Josie Sephton from Strategy Analytics Ltd., André Wills, Fernando Callorda, Lisa Kreuzenbeck and Shazna Zuhyle contributed to the compilation of datasets on prices. The report includes data from Eurostat, OECD, IMF, the UNESCO Institute for Statistics, the United Nations Population Division and the World Bank, which are duly acknowledged. ITU also appreciates the cooperation of countries that have provided data included in this report. The report was edited by the ITU English Translation Section, led by Bruce Granger. The desktop publishing was carried out by the ITU Publication Production Service, led by Simon De Nicola, and the cover was designed by Jesús Vicente. Measuring the Information Society Report 2016 v

8

9 Table of Contents Foreword iii ... ... v Acknowledgements ... vii Table of Contents 1 ... Chapter 1. The ICT Development Index (IDI) – Global Analysis ... 3 Key findings 1.1 Introduction and overview ... 5 7 1.2 ... The ICT Development Index (IDI) ... 11 1.3 Global IDI analysis ... 1.4 The IDI and the digital divide 31 onclusion 35 Summary and c ... 1.5 ... 39 Chapter 2. The ICT Development Index (IDI) – regional and country analysis Key findings 41 ... Introduction ... 2.1 43 ... 43 2.2 Regional IDI analysis Summary and c 2.3 71 onclusion ... 75 ... Chapter 3. The role of ICTs in monitoring the SDGs Key findings 77 ... ... 79 3.1 Introduction ICTs and SDGs 3.2 79 ... tor framework ... 80 3.3 The SDG indica Summary and c ... 92 3.4 onclusion ... 97 Chapter 4. ICT prices ... 99 Key findings 4.1 Introduction ... 101 Mobile-cellular prices 105 4.2 ... Fixed-broadband prices 4.3 ... 115 ... 128 Mobile-broadband prices 4.4 Monitoring the price of bundled services ... 143 4.5 153 ... Chapter 5. Measuring mobile uptake Key findings ... 155 Introduction ... 157 5.1 Moving beyond subscriptions: phone owners and users ... 157 5.2 How many people actually own or use a mobile phone? ... 162 5.3 Who does not own or use a mobile phone? ... 166 5.4 Why do people not own or use mobile phones? 171 5.5 ... Conclusions 174 ... 5.6 177 ... Chapter 6. Internet user and activity trends ... 179 Key findings How the Internet has changed – and changed the world ... 6.1 182 6.2 Socio-economic f actors that determine Internet use ... 186 ... 6.3 199 The Internet is not living up to its potential 6.4 ... 209 Conclusions Measuring the Information Society Report 2016 vii

10 List of references ... 215 ... 223 Annex 1. ICT Development Index (IDI) methodology ... 223 1. Indicators included in the IDI 2. Imputation of missing data ... 225 3. Normalization of data ... 225 ... 226 4. Weighting and aggregation 5. Calculating the IDI ... 227 229 ... 6. Sensitivity analysis ... 231 . Annex 2 ICT price data methodology 231 ... ta collection and sources Price da The mobile-cellular sub-basket ... 231 The fixed-broadband sub-basket ... 234 Mobile-broadband prices ... 235 ... 239 Annex 3. Statistical tables of indicators used to compute the IDI Access indicators ... 240 ... Use indicators 244 ... 248 Skills indicators ... 252 Notes 257 ... Annex 4. ICT Prices ... 258 Fixed-broadband prices 2015 ... 262 Mobile-cellular prices 2015 (on-net) Mobile-cellular prices 2015 (off-net) ... 266 ... 270 Mobile-cellular prices 2015 (to fixed telephone; SMS) Notes ... 274 Measuring the Information Society Report 2016 viii

11 List of tables, figures, charts and boxes Tables IDI values and changes in value, 2015-2016 ... 11 .. Table 1.1: IDI rankings and values, 2016 and 2015 ... .. Table 1.2: 12 ... IDI access sub-index rankings and values, 2016 and 2015 13 .. Table 1.3: ... 14 Table 1.4: .. IDI use sub-index rankings and values, 2016 and 2015 ... 15 IDI skills sub-index, rankings and values, 2016 and 2015 .. Table 1.5: Most dynamic countries in IDI rankings and values, 2015-2016 ... 21 .. Table 1.6: .. ... 24 IDI value change, 2015-2016 Table 1.7: Table 1.8: .. ... 27 Access sub-index, most dynamic countries, 2015-2016 Access sub-index value change, 2015-2016 ... 28 Table 1.9: .. Use sub-index, most dynamic countries, 2015-2016 ... 29 Table 1.10: Use sub-index value change, 2015-2016 ... 30 Table 1.11: IDI by development status, 2016 and 2015 ... 33 Table 1.12: Table 1.13: ... 34 IDI values for LDCs compared with global values and with all developing countries ... 35 Table 1.14: IDI values by IDI quartile, 2015 and 2016 Table 2.1: IDI by region, 2016 and 2015 ... 43 .. Highest- and lowest-ranking countries by region, IDI 2016 ... .. 45 Table 2.2: Table 2.3: ... .. 48 IDI rankings for the Africa region, 2016 and 2015 Most dynamic countries by IDI ranking and IDI value, Africa, 2015-2016 ... Table 2.4: .. 49 .. IDI ranking and values, Arab States region, 2016 and 2015 ... 52 Table 2.5: .. Table 2.6: 53 Most dynamic countries by IDI ranking and IDI value, Arab States, 2015-2016 ... .. ... 56 IDI rankings and values, Asia and Pacific region, 2016 and 2015 Table 2.7: Table 2.8: Most dynamic countries by IDI ranking and IDI value, Asia and Pacific, 2015-2016 ... 58 .. .. IDI ranking and values, CIS region, 2016 and 2015 ... 61 Table 2.9: IDI ranking and values, Europe region, 2016 and 2015 ... 63 Table 2.10: Table 2.11: ... 64 Most dynamic countries by IDI ranking and IDI value, Europe region, 2015-2016 IDI ranking and values, Americas region, 2016 and 2015 68 ... Table 2.12: Table 2.13: ... 69 Most dynamic countries by IDI ranking and IDI value, Americas region, 2015-2016 The Sustainable Development Goals ... Table 3.1: .. 79 .. ICT indicators and related SDG targets ... 82 Table 3.2: .. Fixed broadband subscriptions per 100 inhabitants, per region, 2015 ... 90 Table 3.3: .. ... 107 Mobile-cellular sub-basket, 2015 Table 4.1: Table 4.2: .. ... 110 Domestic mobile minutes (left) and SMS (right), selected economies, 2014 and 2013 .. Top five countries with the cheapest mobile-cellular services in each region, PPP$, 2015 ... 112 Table 4.3: Table 4.4: Fixed-broadband sub-basket, 2015 ... 120 .. Table 4.5: .. Countries with the highest fixed-broadband prices in USD, 2015 ... 127 Top three countries with the cheapest mobile-broadband services in each region, PPP$, .. Table 4.6: 2015 135 ... .. Average mobile-broadband prices and ranges by region, as a percentage of GNI p.c., 2015 ... 135 Table 4.7: .. Mobile-broadband prices, prepaid handset-based, 500 MB, 2015 ... 136 Table 4.8: Table 4.9: .. ... 137 Mobile-broadband prices, postpaid computer-based, 1 GB, 2015 Most common bundle combinations found in OECD countries ... 147 Table 4.10: Table 5.1: .. Individuals who own a mobile-cellular telephone, broken down by children and adults, 2015 . 167 Table 5.2: .. Individuals who use a mobile-cellular telephone, by urban/rural, 2015 or latest available ... 170 year, % Table 6.1: .. Top Internet activities (latest data 2010-2015) ... 200 Proportion of developed and developing countries in which a particular activity is the top .. Table 6.2: Internet activity or among the top 3 or top 5 activities (latest data 2010-2015) ... 201 Annex Table 2.1: OECD mobile-cellular low-user call distribution (2009 methodology) ... 232 Measuring the Information Society Report 2016 ix

12 Figures Three stages in the evolution towards an information society ... 8 Figure 1.1: Figure 1.2: indicators, reference values and weights ... 9 ICT Development Index: Geographical distribution of IDI quartiles, 2016 ... 35 Figure 1.3: ... Facebook’s Free Basics around the world, as of June 2016 102 Figure 4.1: Figure 4.2: 114 Dynamic discounting systems ... ... 116 Dynamic discounting schemes applied in selected countries Figure 4.3: Figure 4.4: ... 146 Individual price structure vs. bundle price structure Main barriers to mobile-phone ownership and usage 173 ... Figure 5.1: Gender parity in tertiary education (2015 or latest available) 193 ... Figure 6.1: Boxes The ITU Expert Groups on Telecommunication/ICT Indicators and on ICT Household Indicators ... 10 Box 1.1: Box 1.2: ... ICT and IDI developments in the Republic of Korea ... 19 ... ICT and IDI developments in Iceland ... Box 1.3: 19 Box 1.4: ICT and IDI developments in Denmark ... 20 ... ... 22 ICT and IDI developments in St. Kitts and Nevis and other Eastern Caribbean countries ... Box 1.5: Box 1.6: ICT and IDI developments in Myanmar 23 ... ... ICT and IDI developments in Namibia ... Box 2.1: ... 51 ... ICT and IDI developments in Côte d’Ivoire ... 51 Box 2.2: ... ICT and IDI developments in Jordan ... 55 Box 2.3: ... ICT and IDI developments in Malaysia ... 59 Box 2.4: ... ... 60 Box 2.5: ICT and IDI developments in Bhutan Box 2.6: ICT and IDI developments in Romania ... 65 ... Box 2.7: ICT and IDI developments in Albania ... 67 ... ... ICT and IDI developments in Bolivia ... 70 Box 2.8: 103 ... Zero-rating and price-differentiation schemes ... Box 4.1: Box 4.2: Smarter and cheaper: Global smartphone prices continue to drop but remain high for low- ... income population groups ... 104 Mobile broadband takes off in Bhutan 134 ... Box 4.3: ... ... Definitions of selected indicators to measure mobile-cellular uptake included in the Core Box 5.1: ... 159 List of ICT Indicators of the Partnership on Measuring ICT for Development Box 6.1: ... How to bring seniors online ... 195 Box 6.2: Rise of social media ... 202 ... ... Research shows side-effects of too much Internet for children ... 211 Box 6.3: ... 227 Weights used for indicators and sub-indices included in the IDI Annex Box 1.1: Example of how to calculate the IDI value ... 228 Annex Box 1.2: Annex Box 2.1: Rules applied in collecting mobile-cellular prices ... 232 Annex Box 2.2: Rules applied in collecting fixed-broadband Internet price data ... 234 Charts 5 . Global changes in levels of ICT uptake per 100 inhabitants, key ICT indicators, 2005-2016* ... Chart 1.1: Chart 1.2: ICT penetration levels, 2016*, by geographic region ... 6 . Chart 1.3: . ... 7 ICT penetration levels, 2016*, by level of development . Distribution of IDI values between regions ... 17 Chart 1.4: Chart 1.5: IDI values for top-ranking countries, 2015 and 2016 ... 18 . Chart 1.6: . IDI values for most dynamic countries, 2015 and 2016 ... 21 Chart 1.7: . IDI and GNI p.c., 2016 ... 32 Chart 1.8: IDI values by development status, 2015 and 2016 ... 33 . Chart 1.9: . IDI values for LDCs compared with global values and with all developing countries ... 34 Chart 2.1: . IDI by region compared with global average, 2016 ... 44 Chart 2.2: Average IDI values for each indicator, world and regions, IDI 2015-2016 ... 46 . Chart 2.3: . IDI values, Africa region, 2016 ... 49 Measuring the Information Society Report 2016 x

13 Chart 2.4: . ... 50 IDI values, selected countries, Africa, IDI 2015-2016 . IDI values, Arab States region, 2016 53 Chart 2.5: ... IDI values, selected countries, Arab States region, 2015-2016 Chart 2.6: ... . 54 ... Households with Internet, 2006-2015 55 Chart Box 2.3: ... 57 Chart 2.7: . IDI values, Asia and Pacific region, 2016 ... 58 IDI values, selected countries, Asia and Pacific region, 2015-2016 . Chart 2.8: Bhutan – Active mobile-broadband subscriptions and Internet users (%) ... Chart Box 2.5: 60 . ... 61 Chart 2.9: IDI values, CIS region, 2016 IDI values, selected countries, CIS region, 2015-2016 62 ... Chart 2.10: IDI values, Europe region, 2016 ... Chart 2.11: 64 IDI values, selected countries, Europe, 2015-2016 ... 66 Chart 2.12: Growth in data traffic in Albania ... 67 Chart Box 2.7: Chart 2.13: ... 69 IDI values, Americas region, 2016 70 Households with Internet, households with a computer and Internet users, Bolivia, 2014 ... Chart Box 2.8: Chart 2.14: IDI values, selected countries, Americas region, 2015-2016 71 ... Percentage of schools with computers, selected African countries, various years ... 83 Chart 3.1: . Proportion of individuals with ICT skills, by type of skill, latest available year, 2012-2015 ... 85 . Chart 3.2: Chart 3.3: . Proportion of individuals with reported ICT skills, Sweden, Morocco and Zimbabwe, ... 85 2014/2015 Chart 3.4: Proportion of individuals with ICT skills, by type of skill, by sex, developed (left) and . ... 86 developing (right) countries, latest available year (2012-2015) Chart 3.5: . Proportion of individuals owning a mobile phone, by sex, 2014/2015 ... 88 . Mobile network coverage and evolving technologies, 2007-2016 ... 89 Chart 3.6: . Proportion of individuals using the Internet, by region and by development status, 2016* ... 91 Chart 3.7: Chart 3.8: . Fixed-broadband subscriptions per 100 inhabitants, by speed, 2015 ... 93 Chart Box 4.1: Percentage of data plans, by type of plan, by country ... 103 Chart Box 4.2: ... 104 Average selling price of smartphones, 2015 . Mobile-cellular sub-basket, as a percentage of GNI p.c. (top), in PPP$ (middle) and in USD Chart 4.1: ... 106 (bottom), 2008-2015 Chart 4.2: . Mobile numbers ported, developed (left) and developing (right) countries, 2014 ... 108 Mobile-cellular prices as a percentage of GNI p.c. (top), in PPP$ (middle) and in USD Chart 4.3: . 111 ... (bottom) by region, 2015 . Fixed-broadband sub-basket, as a percentage of GNI p.c. (top), in PPP$ (middle) and in USD Chart 4.4: ... 118 (bottom), 2008-2015 Chart 4.5: . Most common entry-level fixed-broadband speed, globally and by level of development 121 ... . ... 121 Chart 4.6: Fixed-broadband prices by region, 2015, in USD (left) and in PPP$ (right) . Fixed-broadband prices as a percentage of GNI p.c., speeds and caps, Africa, 2015 ... 122 Chart 4.7: . ... 123 Fixed-broadband prices as a percentage of GNI p.c., speeds and caps, Americas, 2015 Chart 4.8: Fixed-broadband prices as a percentage of GNI p.c., speeds and caps, CIS, 2015 ... 124 . Chart 4.9: Chart 4.10: Fixed-broadband prices as a percentage of GNI p.c., speeds and caps, Europe, 2015 ... 124 Chart 4.11: ... 125 Fixed-broadband prices as a percentage of GNI p.c., speeds and caps, Arab States, 2015 Fixed-broadband prices as a percentage of GNI p.c., speeds and caps, Asia and the Pacific, Chart 4.12: 2015 ... 126 Availability of mobile-broadband services by type of service, by level of development, 2012- Chart 4.13: 2015 129 ... 500 MB handse Chart 4.14: GB c omputer-based (right) mobile-broadband prices: t-based (left) and 1 as a percentage of GNI p.c. (top graph), in PPP$ (middle graph) and in USD (bottom graph), 2013-2015 130 ... Percentage of Internet users that used the Internet on the move, selected economies, 2013 Chart 4.15: and 2014 ... 131 Chart 4.16: Mobile data traffic per subscription per month, selected economies, 2012-2014 ... 131 Comparison of postpaid fixed-broadband prices and postpaid computer-based mobile- Chart 4.17: broadband prices (1 GB/month), in USD, by level of development, 2014 and 2015 ... 132 Comparison of prepaid mobile-cellular prices and prepaid handset-based mobile-broadband Chart 4.18: Measuring the Information Society Report 2016 xi

14 (500 ... 133 MB/month) prices, in USD, by level of development, 2014 and 2015 Mobile-cellular and mobile-broadband penetration in Bhutan, 2008-2015 134 Chart Box 4.3: ... Prepaid handset-based mobile-broadband prices (500 Chart 4.19: MB per month) as a percentage of 139 GNI p.c. and data volume (cap) included, in the Africa region, 2015 and 2014 ... Chart 4.20: MB per month) as a percentage of Prepaid handset-based mobile-broadband prices (500 140 ... GNI p.c. and data volume (cap) included, in the Arab States region, 2015 and 2014 Prepaid handset-based mobile-broadband prices (500 MB per month) as a percentage of Chart 4.21: 141 GNI p.c. and data volume (cap) included, in the Asia-Pacific region, 2015 and 2014 ... Prepaid handset-based mobile-broadband prices (500 MB per month) as a percentage of Chart 4.22: ... 142 GNI p.c. and data volume (cap) included, in the CIS region, 2015 and 2014 Prepaid handset-based mobile-broadband prices (500 MB per month) as a percentage of Chart 4.23: GNI p.c. and data volume (cap) included, in Europe, 2015 and 2014 ... 142 Chart 4.24: Prepaid handset-based mobile-broadband prices (500 MB per month) as a percentage of GNI p.c. and data volume (cap) included, in the Americas, 2015 and 2014 143 ... Chart 4.25: Availability of standalone offers (% of operators) by service, selected OECD economies (left), Proportion of households in the EU that subscribe to bundled (two or more) Chart 4.26: ... 145 telecommunication services, 2014 (right) Chart 4.27: 147 Voice and TV services included in fixed-broadband offers, OECD ... Price range over and above the price of the cheapest individual fixed-broadband offer, Chart 4.28: selected OECD countries, January 2016 148 ... Chart 4.29: Price difference for bundling fixed voice with fixed broadband, selected OECD/EU countries, ... March 2016 149 . Global mobile-cellular subscriptions and population coverage, 2008-2016* ... 158 Chart 5.1: . Mobile-cellular subscriptions and mobile-phone users, selected economies, 2015 ... 161 Chart 5.2: Chart 5.3: . Evolution of mobile-cellular penetration, 2010, 2015 ... 161 Chart 5.4: . Individuals using a mobile-cellular telephone, 2015 or latest available year ... 163 Chart 5.5: Individuals who own a mobile-cellular telephone, 2015 or latest available year 164 . ... Comparison between individuals who own a mobile-cellular telephone and individuals who . Chart 5.6: ... 165 use a mobile-cellular telephone, 2015 or latest available year Chart 5.7: . Individuals who own a mobile-cellular telephone, broken down by age group, 2015 or latest ... available year 166 Chart 5.8: . Individuals using a mobile-cellular telephone, broken down by age group, 2015 or latest ... 168 available year . Individuals owning a mobile-cellular telephone and using a mobile-cellular telephone, Chart 5.9: broken down by age group, India, 2015 169 ... Chart 5.10: Individuals who own a mobile-cellular telephone (left) and using a mobile-cellular telephone 169 (right), broken down by gender, 2015 or latest available year ... Individuals who do not own a mobile-cellular telephone, by gender and level of education, Chart 5.11: ... 171 2015 or latest available year Chart 5.12: Individuals who do not use a mobile-cellular telephone, by gender and rural/urban, 2015 or latest available year ... 172 . Proportion of Internet users accessing the Internet while mobile via a mobile-cellular Chart 6.1: telephone connected to a mobile phone network (2010 and 2015 unless otherwise specified) 183 ... Chart 6.2: . Top three most frequent locations for Internet use in developed (top) and developing (bottom) countries, as a percentage of individuals using the Internet by location; latest data ... 184 2012-2015 Chart 6.3: . ... 185 Time spent on mobile vs. desktop per content category in the United States . Internet and IP traffic ... 186 Chart 6.4: Chart 6.5: . Distribution of Wikipedia articles by language 2003-2016 ... 187 Chart 6.6: Proportion of individuals using the Internet by level of development (left) and by region . ... 187 (right) Chart 6.7: . Proportion of individuals in LDCs using the Internet, 2015 ... 188 . Internet usage by income distribution in OECD countries (2015 unless otherwise specified) .. 189 Chart 6.8: Measuring the Information Society Report 2016 xii

15 Chart 6.9: . Households with Internet access by household income distribution in Latin America (latest ... 189 data 2013-2015) Internet access by household income distribution in Latin America (selected countries) 191 Chart 6.10: ... Chart 6.11: Internet use by level of education in developed (top) and developing (bottom) countries (latest data 2013-2015) ... 192 Internet user gender gap (2013 and 2016) ... 193 Chart 6.12: Internet usage among individuals over the age of 74 compared with the general population Chart 6.13: ... (latest data 2013-2015) 194 ... 195 Chart 6.14: Trends in Internet usage rates by age group (Japan) Proportion of individuals in urban and rural areas using the Internet (latest data 2010-2015) . Chart 6.15: 197 Coverage of mobile-cellular networks in relation to world population and the number of Chart 6.16: Internet users (2007-2016) ... 197 Chart 6.17: Top barriers to household Internet access at home in developed and developing countries ... 198 (latest data 2013-2015) Chart 6.18: Trends regarding activities on the Internet (2006-2015; selected activities) 200 ... Monthly active accounts in social media (2009-2016*; in billions) ... Box Chart 6.2.1: 202 Monthly active accounts for instant-messaging services (2012-2016*, in billions) ... 203 Box Chart 6.2.2: Chart 6.19: Internet use by countries’ income levels (selected activities) ... 204 Proportion of Internet users downloading software (left) and seeking health information Chart 6.20: (right) (latest data 2013-2015). 205 ... Chart 6.21: Proportion of Internet users participating in social media (latest data 2013-2015) ... 206 Chart 6.22: Internet use by activity and education level (selected activities; latest data 2013-2015) ... 207 Adolescents’ (age 15-24) use of the Internet compared with that of the general population .. 210 Chart 6.23: Measuring the Information Society Report 2016 xiii

16

17 Chapter 1. The ICT Development Index (IDI) – Global Analysis

18

19 Key findings The ITU ICT Development Index (IDI) is a unique benchmark of the level of ICT development in countries across the world. The IDI combines eleven indicators on ICT access, use and skills, capturing key aspects of ICT development in one measure that allows for comparisons across countries and over time. The IDI 2016, which covers 175 economies worldwide and makes comparisons to IDI 2015, highlights both progress and persistent divides in the global information society. Nearly all countries improved their IDI values over the last year, but great disparities continue to exist between more and less connected countries. The average IDI value rose by 0.20 points to 4.94 points (out of 10), with smaller increases at the top and at the bottom of the list. The gap between the highest and lowest performing countries – one measure of the digital divide – remained almost unchanged, at 7.76 points in IDI 2016. The Republic of Korea tops the IDI rankings in 2016 for the second consecutive year. The top 10 countries also include two other economies in the Asia-Pacific region, and seven European countries. This reflects the high level of ICT investment and innovation occurring in developed and high-income developing economies. The majority of high-performing countries have liberalized and competitive ICT markets that encourage innovation. They also have populations with relatively high incomes and the skills needed to make effective use of ICTs. There is a strong association between economic and ICT development, with least developed countries at a particular disadvantage. The average IDI value for developed countries (7.40) is 3.33 points higher than that for developing countries (4.07), although developing countries improved their IDI value more than developed countries. There is also a strong association between least connected countries, countries that are in the bottom quartile of the IDI 2016 distribution, and least developed countries. Indeed, the bottom 27 countries are all least developed countries, and the gap in IDI values between these countries and higher-performing developing countries continues to widen. There has been greater improvement in ICT use than access. The use sub-index rose by an average 0.37 points, compared with an increase of 0.13 points in the access sub-index, making ICT use a greater factor of change in IDI outcomes between 2015 and 2016. The increase in the IDI use sub-index was mainly a result of strong growth in mobile-broadband subscriptions across the world. In most regions, the increase in ICT access mainly related to progress made in connecting more households to the Internet, while in Africa improvements in mobile-cellular penetration had a greater impact on the value of the IDI access sub-index. Countries from around the world showed strong improvements in performance. The greatest improvement was achieved by St. Kitts and Nevis, which rose from 54th place in 2015 to 34th place in 2016. Other countries showing substantial ICT progress include Myanmar, Algeria and Bhutan. The experiences in investment, policy and regulation of top-ranking and dynamic economies – discussed in further detail in this chapter – are a source of valuable insights for governments and businesses worldwide. Measuring the Information Society Report 2016 3

20

21 Chapter 1. The ICT Development Index (IDI) – Global Analysis The results of IDI 2016 are analysed in Chapters 1 1.1 Introduction and overview and 2 of this report. This introductory section gives a brief overview of recent progress in the context The period since the conclusion of the World 1 of developments since 2005. Summit on the Information Society (WSIS) in 2005 has seen rapid growth in access to 1.1 illustrates the long-term trend in Chart and use of information and communication penetration rates for various ICTs since 2005. technologies (ICTs) throughout the world. However, the potential impact of ICTs is still It shows that the steep rise in mobile-cellular constrained by digital divides between different subscriptions worldwide, which began early countries and communities. The International in this century, is now tailing off, as the global Telecommunication Union (ITU) documents the penetration rate approaches 100 subscriptions pervasiveness of ICTs and the extent of digital per 100 inhabitants (although it should be divides between regions and countries through noted that, because some people have multiple its annual ICT Development Index (IDI), which subscriptions, the proportion of unique mobile- aggregates quantitative indicators for ICT access, cellular subscribers is significantly lower (GSMA, ICT use and ICT skills in the large majority of world 2016c). At the same time, there has been a gradual economies. decline in the penetration rate for fixed-telephone subscriptions, owing to fixed-mobile substitution bal changes in levels of ICT uptake per 100 inhabitants, key ICT indicators, 2005-2016* Glo Chart 1.1: Notes: * ITU estimates. Source: ITU. Measuring the Information Society Report 2016 5

22 and the tendency for new users to prefer mobile the majority of developing countries and LDCs. over fixed lines. While penetration rates for mobile-cellular subscriptions are now high in all regions, and The growth in mobile-broadband subscriptions exceed 100 subscriptions per 100 inhabitants in worldwide has also been marked, and has four of them, they are still significantly lower in paralleled that of mobile-cellular subscriptions in the Asia-Pacific and Africa regions, and in LDCs. the last five years, albeit at a lower level, rising (In analysing these charts, it should be noted that from one fifth to one half of the penetration the developing-country category includes some rate for mobile-cellular subscriptions between OECD and high-income countries.) Internet and 2011 and 2016. This has helped to drive steady computer access as well as penetration rates growth in the percentage of individuals using the for broadband networks are also higher in the Internet (defined as those who have used the Europe, CIS and Americas regions, which are Internet at least once in the last three months) predominantly composed of developed countries and of households with Internet access. The latter and middle-income developing countries, than in indicator has now overtaken the percentage of the other regions. households with a computer. The results for LDCs on these ICT indicators These global figures, however, mask substantial are particularly poor, especially where fixed- differences between countries in different telephone and fixed-broadband subscriptions are regions and with different levels of development. concerned. The lowly position of LDCs reflects Chart 1.2 compares the 2016 figures for the seven the substantial digital divide between LDCs and ICT penetration indicators in Chart 1.1 between other countries, which remains an important issue the ITU’s six geographic regions, while Chart 1.3 and has particular significance for efforts to use compares the figures for developed countries, ICTs to support achievement of the Sustainable developing countries and least developed Development Goals (SDGs) adopted by the UN countries (LDCs). General Assembly in 2015 (see Chapter 3 of this report). These charts illustrate the continued and significant digital divide between regions, between The remainder of this chapter analyses the IDI developed and developing countries, and between results for 2016. Chart 1.2: ICT penetration levels, 2016*, by geographic region Notes: * ITU estimates. Source: ITU. Measuring the Information Society Report 2016 6

23 Chapter 1 ICT penetration levels, 2016*, by level of development Chart 1.3: Notes: * ITU estimates. Source: ITU. • Section 1.4 analyses the implications of IDI 1.2 The ICT Development Index (IDI) 2016 for measuring the digital divide, with reference to longer-term trends identified in The IDI is a composite index that combines 11 the assessment of progress between 2010 and indicators into one benchmark measure which can 2015 made in the 2015 edition of the Report be used to monitor and compare developments . It relates the IDI to GNI p.c., and (ITU, 2015) in ICT between countries and over time. The IDI considers the particular contexts of LDCs and was developed by ITU in 2008 in response to ITU least connected countries (LCCs). Member States’ request to establish an overall ICT index, was first presented in the 2009 edition Section 1.5 summarizes the chapter and draws • (ITU, 2009), and has been published of the Report 2 conclusions. annually since then. Regional outcomes from IDI 2016 are analysed in This chapter presents the findings for IDI 2016, Chapter 2. which is calculated using data at end 2015, and assesses progress by comparing these data with those for IDI 2015 (calculated using data at end Objectives 2014). The main objectives of the IDI are to measure: • This section, 1.2, describes the objectives, conceptual framework and methodology of • of ICT the level and evolution over time the IDI. developments within countries and their experience relative to other countries; Section 1.3 presents and analyses global • findings for IDI 2016, highlighting high- progress in ICT development developed • in both performing countries and most dynamic ; and developing countries countries (i.e. those displaying the largest improvements in their IDI over the year). Measuring the Information Society Report 2016 7

24 ree stages in the evolution towards an information society Th Figure 1.1: , i.e. differences between • the digital divide – reflecting the results/ ICT impact Stage 3: • outcomes of more efficient and effective ICT countries in terms of their levels of ICT use. development; and the development potential of ICTs and the Advancing through these stages depends on a • extent to which countries can make use of combination of three factors: the availability of ICT them to enhance growth and development in usage infrastructure and access , a high level of ICT , the context of available capabilities and skills. and the capability to use ICTs effectively, derived from relevant skills . These three dimensions – ICT The Index is designed to be global and to reflect ICT skills and access, ICT use – therefore form the changes taking place in countries at different framework for the IDI. levels of ICT development. It therefore relies on The first two stages correspond to two major a limited data set which can be established with • components of the IDI: and . ICT access ICT use reasonable confidence in countries at all levels of development. • Reaching the final stage, and maximizing the . ICT skills impact of ICTs, crucially depends on ICT – and other – skills determine the effective Conceptual framework use that is made of ICTs, and are critical to leveraging their full potential for social and The recognition that ICTs can be development economic development. Economic growth enablers, if applied and used appropriately, is and development will remain below potential critical to countries that are moving towards if economies are not capable of exploiting information- or knowledge-based societies, and new technologies and reaping their benefits. is central to the IDI’s conceptual framework. The IDI therefore also includes indicators The ICT development process, and a country’s concerned with capabilities within countries transformation to becoming an information which affect people’s ability to use ICTs society, can be depicted using the three-stage effectively. 2.1: model illustrated in Figure A single indicator cannot track progress in all Stage 1: ICT readiness – reflecting the level of • three of these components of ICT development. networked infrastructure and access to ICTs; It is therefore necessary to construct a composite index, which aims to capture the evolution of the ICT intensity Stage 2: • – reflecting the level of information society as it goes through its different use of ICTs in the society; and Measuring the Information Society Report 2016 8

25 Chapter 1 ICT Development Index: dicators, reference values and weights Figure 1.2: in Reference (%) ICT access value 1. Fixed-telephone subscriptions per 100 inhabitants 60 20 120 2. Mobile-cellular telephone subscriptions per 100 inhabitants 20 40 3. International Internet bandwith (bit/s) per internet user 20 976’696* 4. Percentage of households with a computer 100 20 5. Percentage of households with Internet access 100 20 Reference CT I (%) ICT use value evelopment D ndividuals using the Internet i 33 100 6. Percentage of I ndex 40 60 7. Fixed-broadband subscriptions per 100 inhabitants 33 100 33 8. Active mobile-broadband subscriptions per 100 inhabitants Reference (%) ICT skills value 5 33 9. Mean years of schooling 1 20 33 100 10. Secondary gross enrolment ratio 11. Tertiary gross enrolment ratio 100 33 Note: * This corresponds to a log value of 5.99, which was used in the normalization step. Source: ITU. stages of development, taking into consideration indicators (mean years of schooling, gross technology convergence and the emergence of secondary enrolment, and gross tertiary new technologies. enrolment). As these are proxy indicators, rather than direct measures of ICT-related Based on this conceptual framework, the IDI is skills, the skills sub-index is given less weight in divided into the following three sub-indices, which the computation of the IDI than the other two 3 are illustrated, with their component indicators, in sub-indices. Figure 1.2: The choice of indicators included in these sub- Access sub-index : This sub-index captures • indices reflects the corresponding stage of ICT readiness, and includes five infrastructure transformation to the information society. The and access indicators (fixed-telephone indicators in each sub-index may therefore subscriptions, mobile-cellular telephone change over time to reflect technological subscriptions, international Internet bandwidth developments related to ICTs and improvements per Internet user, households with a computer, in the availability and quality of data. For and households with Internet access). example, subscription to what was considered basic infrastructure in the past – such as fixed- • This sub-index captures ICT : Use sub-index telephone lines – is fast becoming less essential intensity, and includes three intensity and because of the growth in mobile networks usage indicators (individuals using the Internet, and fixed-mobile substitution. Similarly, while fixed-broadband subscriptions, and mobile- broadband has historically been considered an broadband subscriptions). advanced technology, and is therefore included as an indicator in the use sub-index, it is increasingly Skills sub-index : This sub-index seeks • considered essential and may become more to capture capabilities or skills which are appropriate to the access sub-index. important for ICTs. It includes three proxy Measuring the Information Society Report 2016 9

26 the IDI is a transparent, statistically credible and Methodology legitimate tool for improved policy-making. The IDI includes 11 indicators. A detailed definition The results of the assessment were summarized of each indicator is provided in Annex 1. 4 in the 2015 edition of the (ITU, 2015). Report It found that the IDI was developed using The indicators used to calculate the IDI were international quality standards and tested using selected on the basis of the following criteria: state-of-the-art statistical analyses. The three-level structure of the IDI was found to be statistically The relevance of a particular indicator in • sound in terms of coherence and balance and the contributing to the main objectives and IDI had high statistical reliability. Its added value conceptual framework of the IDI. For example, was seen to lie in its ability to summarize different the selected indicators must be relevant to aspects of ICT development in a more efficient and both developed and developing countries, economical manner than would be the case with and should reflect, so far as possible, the eleven separate indicators. framework’s three components as described above. While the core methodology of the IDI has remained the same since it was first published, • Data availability and quality . Data are required adjustments are made year on year in accordance for a large number of countries, as the IDI is a with the criteria listed above, while also reflecting global index. There is a shortage of ICT-related the dynamic nature of the ICT sector and related data, especially on usage, in the majority of data availability. developing countries. In addition, as indicators which are directly related to ICT skills are The indicators included in the IDI and its sub- not available for most countries, it has been indices are therefore regularly reviewed in ITU, necessary to use proxy rather than direct in consultation with experts. Indicator definitions indicators in the skills sub-index. and the IDI methodology are discussed in the ITU Expert Group on Telecommunication/ICT The results of various statistical analyses . • Indicators (EGTI) and the ITU Expert Group on Principal components analysis (PCA) is used to ICT Household Indicators (EGH) (see Box 1.1). In examine the underlying nature of the data and 2015, EGTI amended the skills sub-index of the explore whether their different dimensions are IDI by substituting the indicator ‘mean years of statistically well-balanced. schooling’ for the indicator ‘adult literacy rate’. As a result, the IDI values and rankings for 2015 and An assessment of the statistical approach taken 2016 included in this year's report are not directly to the IDI was undertaken for ITU during 2015 by comparable with those published in previous the Composite Indicators Research Group of the editions of the annual Measuring the Information European Commission’s Joint Research Centre. Society Report . The main goal of the exercise was to ensure that Box 1.1: The ITU Expert Groups on Telecommunication/ICT Indicators and on ICT Household Indicators Much of ITU’s work in the area of indicator definitions and methodologies is carried out through 5 and the its two expert groups: the Expert Group on Telecommunication/ICT Indicators (EGTI) 6 Created in 2009 and 2012, respectively, Expert Group on ICT Household Indicators (EGH). these expert groups review and revise ITU’s supply-side and demand-side statistics, and discuss methodological issues and new indicators. Both groups, which are open to all ITU members and to experts in ICT statistics and data collection, work through online discussion forums and occasional face-to-face meetings. They periodically report to the World Telecommunication/ ICT Indicators Symposium (WTIS), ITU’s main forum on ICT statistics. Interested experts are invited to join the EGTI and/or EGH discussion to share experiences, contribute to discussions and participate in the decision-making process. Measuring the Information Society Report 2016 10

27 Chapter 1 The average IDI value among the 175 economies Data for IDI 2016 were collected at the beginning included in IDI 2016 was 4.94. of 2016, and refer to the situation at end 2015. Data for IDI 2015 used for comparative purposes Summary data for the IDI and its three sub-indices in this report have also been adjusted to take account of corrections and updates to data 1.1 to 1.5. in 2016 and 2015 are set out in Tables Data for IDI 2015 in these and subsequent tables previously reported. have been recalculated to accommodate changes in the Index as described in section 1.2. IDI 2016 was computed using the same methodology as in the past, applying the following Table 1.1 shows that the average IDI value steps (see also Figure 1.2 and Annex 1): increased by 0.20 points over the year, from 4.74 Preparation of the complete data set. to 4.94, with a higher rate of growth in the average This step • use sub-index value (which rose by 0.37 points, included filling in missing values using a variety of statistical techniques. from 3.54 to 3.91) than in the average access sub- index value (which rose by 0.13 points, from 5.45 • to 5.58). The skills sub-index remained unchanged . This is required in order Normalization of data at 5.74, since, for reasons of data availability, the to transform the values of IDI indicators into same data have been used for both 2016 and 2015 the same unit of measurement. The chosen in this sub-index. normalization method is the distance to a reference value, either 100 or a value obtained The IDI results for all economies included in the through an appropriate statistical procedure. Index in 2015 and 2016 are set out in Table 1.2, . The data were rescaled on Rescaling of data while results relating to the access, use and skills • sub-indices are set out in Tables 1.3, 1.4 and 1.5, a scale from 0 to 10 in order to compare the 1.2 respectively. The economies listed in Table values of the indicators and the sub-indices. have been divided into four quartiles according to • their IDI rankings, as follows: . Weighting of indicators and sub-indices Indicator weights were chosen based on PCA The high quartile includes the 44 top-ranked • results. The access and use sub-indices were countries, from the Republic of Korea with an given equal weight (40 per cent each), while IDI value of 8.84 to Portugal with an IDI value the skills sub-index was given lesser weight (20 of 6.94. per cent) as it is based on proxy indicators. The upper-middle quartile includes the next • Global IDI analysis 1.3 42 countries in the rankings, from Saudi Arabia with an IDI value of 6.90 to Maldives with an The IDI 2016 results show that there continue IDI value of 5.04. to be significant differences in the levels of ICT development between countries and regions The lower-middle quartile includes the next • around the world. IDI values range from a low group of 45 countries, from Seychelles with an of 1.07 in Niger to a high of 8.84 in the Republic IDI value of 5.03 to Nicaragua with an IDI value of Korea (within a possible range from 0 to 10). of 2.88. IDI values and changes in value, 2015-2016 Table 1.1: IDI 2015 IDI 2016 Change in average Average Average value StDev Min. Max. Range CV Min. CV Max. Range StDev value* value* 2016-2015 4.94 1.07 8.84 7.76 2.22 44.95 IDI 4.74 1.00 8.78 7.78 2.23 47.01 0.20 Access 5.58 1.34 9.54 8.21 2.16 38.71 index - sub 0.13 5.45 1.28 9.49 8.21 2.18 40.08 3.91 0.12 8.91 8.78 2.47 63.23 Use sub-index 3.54 0.06 8.84 8.78 2.48 69.88 0.37 Skills sub-index 5.74 1.01 9.18 8.17 2.19 38.15 5.74 1.01 9.18 8.17 2.19 38.15 0.00 Note: *Simple averages. StDev= Standard deviation, CV= Coefficient of variation Source: ITU. Measuring the Information Society Report 2016 11

28 Table 1.2: IDI rankings and values, 2016 and 2015 IDI Rank IDI IDI Rank IDI Rank Rank Economy Economy 2016 2016 2015 2015 2015 2016 2015 2016 90 4.99 4.66 Iran (I.R.) 89 Korea (Rep.) 1 8.84 1 8.78 90 Mongolia 2 3 8.66 8.83 4.54 93 4.95 Iceland Denmark 4.62 8.77 2 8.74 3 Albania 91 4.92 92 Switzerland 96 4.87 92 Mexico 4.45 4 8.68 5 8.50 8.54 4 8.57 5 United Kingdom Panama 93 4.87 91 4.63 87 4.85 94 St. Lucia 4.68 Hong Kong, China 6 8.46 7 8.40 8.47 4.49 95 4.83 95 Tunisia 8.45 6 Sweden 7 Netherlands 4.26 8.36 8 8.43 8 Morocco 96 4.60 98 Cape Verde Norway 8.35 9 8.42 9 97 4.60 99 4.23 Japan 4.54 94 4.56 98 Ecuador 10 8.37 11 8.28 101 4.52 99 Jamaica 8.34 Luxembourg 11 8.36 10 4.20 Egypt 100 4.44 97 4.26 8.13 13 8.31 12 Germany 100 4.42 101 Peru 4.23 New Zealand 13 8.29 16 8.05 102 4.41 102 Fiji 8.18 Australia 14 8.19 12 4.16 Algeria 103 4.40 112 3.74 8.06 15 8.17 15 United States 105 4.30 104 Dominican Rep. 4.02 France 16 8.11 17 7.95 Viet Nam 8.08 8.11 4.02 104 4.29 14 105 Finland 17 Estonia 4.12 7.95 18 8.07 18 Palestine 106 4.28 103 Philippines Monaco 7.86 20 7.96 19 107 4.28 106 3.97 Botswana Singapore 7.88 19 7.95 20 108 4.17 109 3.79 Paraguay Ireland 7.73 21 7.92 21 109 4.08 107 3.88 Uzbekistan Belgium 7.69 22 7.83 22 110 4.05 110 3.76 Austria Bolivia 4.02 117 3.49 7.53 24 7.69 23 111 111 3.99 112 Ghana 3.75 Malta 24 7.69 25 7.49 108 Kyrgyzstan 7.55 3.99 113 Canada 25 7.62 23 3.85 Tonga 114 3.93 114 3.63 7.46 27 7.62 26 Spain 3.86 115 Indonesia Andorra 27 7.61 115 3.63 29 7.39 116 Sri Lanka 7.47 3.77 116 Macao, China 28 7.58 26 3.56 Bhutan 117 3.74 122 3.12 7.42 28 7.46 29 Bahrain 113 3.73 118 El Salvador 3.64 Israel 30 7.40 30 7.25 119 Belize 7.02 3.66 119 Belarus 31 7.26 33 3.32 Namibia 120 3.64 121 3.20 7.20 31 7.25 32 Czech Republic 118 3.52 121 Guyana 3.44 Slovenia 33 7.23 32 7.10 120 Syria 6.23 3.32 122 St. Kitts and Nevis 34 7.21 54 3.21 Guatemala 123 3.20 123 3.09 6.87 39 7.18 35 Barbados 126 3.12 124 Gabon 2.81 Greece 36 7.13 40 6.86 2.78 127 3.12 125 36 Cambodia 7.11 6.89 Italy 37 United Arab Emirates 3.00 6.96 35 7.11 38 Honduras 126 3.09 124 Vanuatu Lithuania 7.00 34 7.10 39 127 3.08 131 2.73 Timor-Leste Latvia 6.88 37 7.08 40 2.92 125 3.05 128 2.78 Kenya 129 2.99 129 Croatia 41 7.04 41 6.83 128 130 Samoa 6.69 2.95 Slovakia 42 6.96 44 2.78 6.79 42 6.95 43 Russian Federation Nicaragua 131 2.88 130 2.74 139 2.86 132 Côte d'Ivoire 2.43 Portugal 44 6.94 45 6.64 132 133 Zimbabwe 6.88 2.78 Saudi Arabia 45 6.90 38 2.73 6.78 43 6.90 46 Qatar Lesotho 134 2.76 138 2.47 133 2.73 135 Cuba 2.64 Uruguay 47 6.79 49 6.44 2.73 136 Swaziland Hungary 2.49 136 48 6.72 46 6.60 Nigeria 137 2.72 137 2.48 Bulgaria 49 6.69 50 6.43 138 2.50 135 2.69 India Poland 50 6.65 47 6.56 Sudan 139 2.60 134 2.56 Serbia 51 6.58 51 6.43 140 1.95 153 2.54 Myanmar Kazakhstan 52 6.57 52 6.42 Senegal 141 2.53 140 2.41 Kuwait 53 6.54 48 6.45 54 Cyprus Nepal 142 53 6.28 6.53 2.50 142 2.32 2.40 141 2.46 Argentina 55 6.52 56 6.21 143 Gambia Chile 57 6.11 56 Lao P.D.R. 6.35 144 2.45 144 2.21 57 59 6.03 6.30 Costa Rica Bangladesh 2.27 143 2.35 145 58 2.15 145 2.35 146 Pakistan 6.23 6.28 55 Azerbaijan Zambia 2.22 148 2.05 147 6.04 58 6.27 59 Oman Cameroon 6.26 60 Romania 148 2.16 146 2.07 5.92 60 61 6.22 2.00 66 149 2.14 149 Mali 5.64 Malaysia Rwanda 150 2.13 158 1.79 6.05 62 5.76 Montenegro 64 1.90 Brazil 63 5.99 65 5.72 154 2.12 151 Mauritania 5.80 Kiribati 152 2.06 147 5.98 63 Bahamas 64 2.07 Solomon Islands 65 5.97 62 5.82 1.99 TFYR Macedonia 2.04 153 150 5.91 2.03 154 Angola 1.95 152 Lebanon 66 5.93 61 151 Trinidad & Tobago 5.76 68 5.48 Yemen 155 2.02 67 1.96 Moldova 5.60 67 5.75 68 Liberia 156 1.97 161 1.73 5.14 Dominica 69 5.71 1.86 155 1.94 157 Uganda 77 70 Benin 156 158 5.45 Turkey 1.83 69 5.69 1.92 1.86 159 Togo 159 Armenia 1.78 5.34 71 5.60 71 1.85 72 5.59 72 Georgia Equatorial Guinea 157 1.82 160 5.33 Mauritius 5.55 73 5.27 73 160 1.82 161 Djibouti 1.73 4.97 82 5.43 74 Grenada 163 1.80 162 Burkina Faso 1.60 Antigua & Barbuda 75 5.38 70 163 Mozambique 1.60 164 1.75 5.41 162 76 Ukraine 5.21 Afghanistan 164 1.73 76 1.62 5.33 1.57 166 1.72 165 Guinea 5.33 77 Brunei Darussalam 5.25 74 St. Vincent and the Grenadines 1.57 165 1.69 166 Madagascar 78 5.32 78 5.07 Venezuela Tanzania 1.54 167 1.65 167 79 5.27 75 5.22 Bosnia and Herzegovina Malawi 1.49 168 1.62 168 5.03 80 5.25 80 China Ethiopia 4.80 81 5.19 84 169 1.51 172 1.29 79 82 Thailand 1.48 5.18 Congo (Dem. Rep.) 170 1.50 169 5.05 1.16 173 1.42 171 Burundi Colombia 83 5.16 81 4.98 83 5.09 84 Suriname 4.89 South Sudan 172 1.42 170 1.36 89 85 Jordan 1.34 5.06 Guinea-Bissau 173 1.38 171 4.67 1.00 175 1.09 174 Chad Maldives 86 5.04 88 4.68 85 5.03 87 Seychelles 4.77 Niger 175 1.07 174 1.03 88 South Africa 4.70 5.03 86 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 12

29 Chapter 1 Table 1.3: IDI access sub-index rankings and values, 2016 and 2015 Rank IDI IDI IDI IDI Rank Rank Rank Economy Economy 2016 2016 2015 2015 2016 2016 2015 2015 Thailand 5.24 89 5.50 Luxembourg 92 1 9.54 1 9.49 90 South Africa Iceland 9.42 2 9.35 5.26 90 5.46 2 United Kingdom 91 9.18 4 9.24 3 China 5.45 91 5.26 Hong Kong, China 89 5.42 92 Venezuela 5.44 4 9.16 3 9.20 9.17 5 9.09 5 Germany Palestine 93 5.35 95 5.12 93 5.30 94 Egypt 5.20 Malta 6 9.04 6 9.01 9 5.01 96 5.29 95 Tunisia 8.94 Netherlands 7 9.02 Korea (Rep.) 96 9.00 8 8.99 8 Mongolia 5.12 100 4.77 Mexico Switzerland 9.00 7 8.95 9 97 5.08 99 4.82 Japan 4.56 105 5.03 98 Algeria 10 8.80 11 8.75 97 5.02 99 Cape Verde 8.61 Singapore 11 8.70 14 4.89 Fiji 100 4.97 101 4.68 8.67 12 8.70 12 France 98 4.95 101 El Salvador 4.88 Sweden 13 8.69 10 8.77 94 4.90 102 Ecuador 8.65 Denmark 14 8.52 13 5.16 Jamaica 103 4.83 103 4.65 8.16 20 8.35 15 Austria 102 4.80 104 Peru 4.66 Belgium 16 8.34 15 8.34 Ghana 22 8.08 4.51 107 4.74 105 New Zealand 17 8.32 Israel 106 8.18 19 8.28 18 Albania 4.73 108 4.48 Indonesia United States 8.11 21 8.27 19 107 4.71 106 4.53 Philippines Barbados 7.86 28 8.24 20 108 4.70 109 4.46 Syria Australia 8.24 17 8.23 21 109 4.66 104 4.58 Viet Nam Norway 8.19 18 8.21 22 110 4.60 111 4.42 Ireland Paraguay 4.59 110 4.45 8.24 16 8.19 23 111 116 4.53 112 Uzbekistan 4.22 United Arab Emirates 24 8.14 25 7.94 115 Sri Lanka 8.03 4.51 113 Monaco 25 8.12 23 4.26 Guatemala 114 4.47 113 4.34 7.84 31 8.06 26 Andorra 4.43 115 Tonga Estonia 27 8.02 117 4.20 29 7.85 121 Dominican Rep. 7.98 4.38 116 Canada 28 7.99 24 4.12 Bolivia 117 4.37 120 4.13 7.88 27 7.93 29 Slovenia 118 4.33 118 Guyana 4.19 Portugal 30 7.93 33 7.77 114 Botswana 7.80 4.33 119 Spain 31 7.92 32 4.27 Namibia 120 4.25 112 4.35 7.76 34 7.91 32 Bahrain 119 4.25 121 Kyrgyzstan 4.16 Qatar 33 7.91 26 7.90 124 Cambodia 7.60 4.21 122 Greece 34 7.85 38 3.93 Honduras 123 4.17 122 4.04 7.85 30 7.83 35 Macao, China 123 4.08 124 Nicaragua 4.00 Belarus 36 7.80 36 7.68 128 7.50 4.02 125 Bhutan 40 3.75 St. Kitts and Nevis 37 7.72 Italy 126 7.64 37 7.69 38 Gabon 3.98 125 3.88 Timor-Leste Finland 7.73 35 7.69 39 127 3.91 126 3.87 Gambia Hungary 7.49 41 7.62 40 128 3.90 127 3.85 Croatia 3.44 131 3.79 129 Côte d'Ivoire 41 7.58 44 7.33 129 3.69 130 Belize 7.46 Kazakhstan 42 7.56 42 3.62 Vanuatu 131 3.66 132 3.40 7.44 43 7.46 43 Czech Republic 130 3.59 132 Senegal 3.51 Kuwait 44 7.40 45 7.31 136 3.54 133 Kenya 7.20 Latvia 45 7.38 46 3.30 Samoa 134 3.43 137 3.27 7.12 51 7.37 46 Oman 138 3.41 135 Lesotho 3.26 Saudi Arabia 47 7.29 39 7.51 135 3.39 136 Pakistan 7.15 Uruguay 48 7.25 50 3.30 Zimbabwe 137 3.35 139 3.22 7.19 47 7.23 49 Russian Federation 133 3.33 138 Sudan 3.35 Slovakia 50 7.22 53 7.08 3.32 139 India Serbia 3.15 140 51 7.22 48 7.18 Brunei Darussalam 3.31 134 3.30 140 Mali 52 7.21 49 7.16 Swaziland 3.11 141 3.28 141 Poland 53 7.09 52 7.11 Lao P.D.R. 142 Lithuania 54 3.21 3.03 142 7.08 54 7.01 3.16 Trinidad & Tobago 7.03 58 6.72 2.91 144 55 Nepal 143 56 Cyprus 7.02 Myanmar 6.97 55 144 3.08 159 2.45 57 6.90 60 6.65 Romania Bangladesh 145 3.06 143 2.91 Mauritania 2.99 145 2.88 146 6.59 65 6.86 58 Mauritius 2.82 6.86 59 Bulgaria Nigeria 147 2.96 147 6.85 56 2.82 Montenegro 60 6.85 57 6.74 146 2.90 148 Mozambique 6.81 Chile 61 61 6.65 2.63 154 2.87 149 Burkina Faso Benin 150 2.86 6.58 66 6.81 2.72 62 Seychelles 151 Azerbaijan 63 6.78 59 6.68 2.66 152 2.84 Zambia 151 Bahamas 6.77 Cameroon 152 2.77 148 2.82 64 6.62 64 6.63 2.77 149 2.76 6.77 65 Argentina 62 Equatorial Guinea 153 150 2.75 6.53 Malaysia 66 69 Angola 154 6.75 2.76 2.76 6.63 63 155 155 2.59 6.68 67 Liberia TFYR Macedonia 2.59 68 Solomon Islands 156 Moldova 156 70 6.50 2.73 6.64 157 Yemen 2.66 Lebanon 69 6.57 2.65 153 67 6.57 6.57 70 Armenia 2.42 6.29 Tanzania 158 2.65 161 73 Rwanda 6.43 71 158 71 Ukraine 2.54 2.65 159 6.48 St. enadines Gr the and Vincent 2.55 157 160 Togo 2.59 72 6.47 74 6.28 161 2.57 162 2.41 Guinea 72 Costa Rica 73 6.44 6.31 Djibouti 162 2.55 160 2.44 Brazil 6.28 75 6.42 74 163 2.51 163 Afghanistan 2.39 6.06 78 6.40 75 Dominica 2.29 165 2.41 164 Guinea-Bissau Antigua & Barbuda 76 6.34 68 6.56 165 Madagascar 2.21 166 2.39 77 Grenada 77 6.30 6.14 Uganda 164 2.34 2.37 166 6.29 Georgia 78 76 6.25 2.00 Cuba 167 2.17 169 5.97 81 6.26 79 Iran (I.R.) 1.84 Burundi 168 2.14 172 6.04 79 80 Maldives 6.22 Kiribati 6.00 Turkey 81 6.20 80 169 2.11 167 2.14 5.91 6.10 82 Jordan 82 170 2.11 171 1.85 Ethiopia 5.64 83 Morocco 6.07 87 170 Niger 171 2.04 1.99 84 Panama 5.99 83 5.72 172 2.03 2.01 Malawi 168 Suriname 5.67 86 5.88 85 173 Chad 1.74 1.94 174 5.70 85 5.83 86 Colombia Congo (Dem. Rep.) 173 1.83 1.83 174 5.71 Bosnia and Herzegovina 87 5.78 84 South Sudan 175 1.34 175 1.28 5.65 88 88 St. Lucia 5.52 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 13

30 Table 1.4: IDI use sub-index rankings and values, 2016 and 2015 IDI Rank IDI IDI Rank IDI Rank Rank Economy Economy 2016 2016 2015 2015 2015 2016 2015 2016 81 3.72 3.43 St. Lucia 89 Denmark 1 8.91 1 8.84 90 Mongolia 2 6 8.17 8.67 2.97 92 3.64 Switzerland Korea (Rep.) 2.94 8.42 2 8.57 3 Jamaica 91 3.55 94 Norway 90 3.51 92 Viet Nam 3.01 4 8.48 3 8.33 8.11 7 8.44 5 Iceland Dominican Rep. 93 3.41 91 2.97 110 3.40 94 Bhutan 2.13 Sweden 6 8.36 4 8.31 8.21 2.95 93 3.40 95 Morocco 8.18 5 Finland 7 Japan 2.94 7.96 10 8.14 8 Seychelles 96 3.37 95 Ecuador United Kingdom 8.09 8 8.09 9 97 3.31 89 3.01 Luxembourg 2.37 106 3.26 98 Botswana 10 8.05 9 8.05 96 3.24 99 Panama 7.66 New Zealand 11 8.03 16 2.92 Fiji 100 3.23 97 2.88 7.74 13 7.94 12 Hong Kong, China 104 3.20 101 Jordan 2.44 Monaco 13 7.92 11 7.78 100 3.14 102 Egypt 7.75 Estonia 14 7.87 12 2.78 Ghana 103 3.03 101 2.64 7.68 14 7.77 15 Netherlands 98 2.97 104 Brunei Darussalam 2.81 Australia 16 7.70 15 7.67 Paraguay 7.61 7.23 2.61 102 2.96 20 105 France 17 United States 2.61 7.45 19 7.57 18 Peru 106 2.94 103 Philippines Singapore 7.45 18 7.54 19 107 2.93 105 2.42 Algeria Macao, China 7.22 21 7.53 20 108 2.92 119 1.75 Namibia Germany 6.98 22 7.49 21 109 2.91 121 1.73 Iran (I.R.) Bahrain 7.54 17 7.48 22 110 2.74 108 2.19 Ireland Bolivia 2.72 123 1.65 6.85 23 7.38 23 111 112 2.59 112 Tonga 2.07 Belgium 24 7.10 24 6.76 109 Uzbekistan 6.65 2.58 113 Spain 25 6.93 27 2.17 Ukraine 114 2.57 107 2.31 6.68 25 6.85 26 Canada 2.55 115 Belize United Arab Emirates 27 6.82 117 1.79 26 6.66 115 Nigeria 6.28 2.28 116 Malta 28 6.75 31 1.81 Palestine 117 2.25 111 2.08 6.41 30 6.74 29 Andorra 113 2.25 118 Kyrgyzstan 2.00 Austria 30 6.67 28 6.48 124 Vanuatu 6.43 2.21 119 Czech Republic 31 6.55 29 1.60 Indonesia 120 2.19 116 1.79 4.31 64 6.53 32 St. Kitts and Nevis 126 2.09 121 Cambodia 1.53 Lithuania 33 6.40 32 6.23 131 Côte d'Ivoire 5.86 2.08 122 Slovakia 34 6.38 37 1.34 Kenya 123 2.05 118 1.76 6.03 33 6.32 35 Qatar 134 1.92 124 Gabon 1.23 Saudi Arabia 36 6.32 35 6.03 1.91 114 1.91 125 36 Zimbabwe 6.27 5.94 Latvia 37 Italy 1.73 5.73 40 6.25 38 Sudan 126 1.87 120 El Salvador Uruguay 5.43 45 6.20 39 127 1.87 122 1.72 Lesotho Kuwait 6.03 34 6.15 40 1.22 135 1.80 128 0.89 Myanmar 129 1.73 146 Croatia 41 6.13 38 5.85 128 130 Timor-Leste 5.75 1.70 Israel 42 6.02 39 1.42 5.47 44 5.88 43 Barbados Sri Lanka 131 1.70 127 1.44 125 1.65 132 Guyana 1.57 Belarus 44 5.88 47 5.40 129 133 Senegal 5.52 1.64 Russian Federation 45 5.87 42 1.42 4.63 61 5.86 46 Malaysia Swaziland 134 1.61 136 1.19 130 1.52 135 Syria 1.35 Bulgaria 47 5.84 49 5.21 1.47 136 Rwanda Costa Rica 0.73 151 48 5.80 48 5.24 Guatemala 137 1.40 133 1.25 Slovenia 49 5.71 46 5.43 138 1.29 132 1.38 Honduras Azerbaijan 50 5.70 41 5.66 Nepal 139 1.35 137 1.15 Portugal 51 5.67 53 5.07 140 0.85 147 1.29 Mauritania Brazil 52 5.60 54 5.07 Uganda 141 1.27 138 1.10 Lebanon 53 5.51 43 5.48 54 Serbia India 142 50 5.15 5.50 1.25 143 0.95 0.97 142 1.23 Cyprus 55 5.46 58 4.89 143 Samoa Greece 55 5.05 56 Zambia 5.46 144 1.17 144 0.93 57 60 4.81 5.45 Argentina Yemen 0.99 140 1.12 145 58 0.70 152 1.11 146 Lao P.D.R. 5.05 5.39 56 Oman Angola 1.10 145 0.91 147 5.13 51 5.35 59 Poland Pakistan 5.28 60 Hungary 148 1.09 153 0.69 5.12 52 61 5.17 1.02 59 139 1.06 149 Bangladesh 4.85 TFYR Macedonia Cuba 150 1.04 141 0.97 5.15 62 4.90 Kazakhstan 57 0.73 Romania 63 5.08 63 4.48 150 1.00 151 Nicaragua 4.48 Mali 152 0.97 156 4.91 62 Chile 64 0.61 Gambia 65 4.82 76 3.72 0.83 Dominica 0.91 153 148 3.99 0.90 154 Burkina Faso 0.63 155 Montenegro 66 4.61 70 162 China 4.58 73 3.79 Liberia 155 0.89 67 0.44 Trinidad & Tobago 4.13 68 4.53 68 Malawi 156 0.86 159 0.56 4.20 Suriname 69 4.48 0.55 160 0.84 157 Cameroon 66 70 Ethiopia 161 158 4.17 Bahamas 0.54 67 4.46 0.82 0.74 159 Equatorial Guinea 154 Thailand 0.66 4.27 65 4.33 71 0.73 77 4.30 72 Maldives Solomon Islands 149 0.75 160 3.59 Moldova 4.26 69 4.02 73 157 0.71 161 Djibouti 0.59 3.43 82 4.24 74 Mexico 158 0.65 162 South Sudan 0.57 Bosnia and Herzegovina 75 4.21 75 163 Guinea 0.42 164 0.62 3.74 170 76 Turkey 3.77 Mozambique 164 0.62 74 0.30 4.18 0.32 169 0.49 165 Togo 4.03 77 Cape Verde 3.24 88 Antigua & Barbuda 0.34 166 0.47 166 Afghanistan 78 4.02 71 3.89 South Africa Kiribati 0.44 163 0.45 167 79 4.00 86 3.37 Georgia Madagascar 0.33 167 0.44 168 3.40 83 4.00 80 Venezuela Burundi 3.80 81 3.95 72 169 0.42 175 0.06 85 82 Tunisia 0.36 3.95 Congo (Dem. Rep.) 170 0.41 165 3.37 0.32 168 0.40 171 Benin St. Vincent and the Grenadines 83 3.89 79 3.47 84 3.88 84 Albania 3.40 Tanzania 172 0.30 171 0.27 80 85 Armenia 0.10 3.85 Chad 173 0.14 173 3.47 0.10 174 0.14 174 Niger Colombia 86 3.85 78 3.52 99 3.78 87 Grenada 2.79 Guinea-Bissau 175 0.12 172 0.12 88 Mauritius 3.36 3.78 87 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 14

31 Chapter 1 Table 1.5: IDI skills sub-index, rankings and values, 2016 and 2015 IDI IDI IDI IDI Rank Rank Rank Rank Economy Economy 2016 2016 2015 2015 2016 2015 2016 2015 89 5.90 5.90 Dominican Rep. United States 1 89 9.18 1 9.18 Bolivia Australia 9.10 2 9.10 5.89 90 5.89 2 90 9.08 91 5.89 91 5.89 Korea (Rep.) 3 9.08 3 Brazil 5.89 92 5.89 92 China Greece 4 9.01 4 9.01 5.89 8.96 5 8.96 5 Belarus 93 5.89 93 Panama 5.87 94 5.87 94 Malaysia Denmark 8.87 6 8.87 6 enadines Vincent and the Gr St. 7 8.87 8.87 Slovenia 7 5.86 95 95 5.86 8.77 8 8.77 8 New Zealand 96 Oman 5.83 5.83 96 Norway 8.70 9 8.70 9 5.83 5.83 Jamaica 97 97 Finland 10 8.65 10 8.65 98 5.81 98 Belize 5.81 8.57 11 Ukraine 11 8.57 5.74 Mexico 99 99 5.74 8.56 12 8.56 12 Netherlands 100 Botswana 100 5.69 5.69 Lithuania 8.55 13 8.55 13 5.68 5.68 101 Fiji 101 Russian Federation 14 8.55 14 8.55 102 5.68 Tunisia 102 5.68 Estonia 15 15 8.54 8.54 Trinidad & Tobago 103 5.67 103 5.67 Ireland 16 8.48 16 8.48 5.66 104 5.66 104 Guyana Canada 8.44 17 8.44 17 5.63 5.63 United Arab Emirates 105 105 18 8.43 Andorra 18 8.43 5.61 Tonga 106 106 5.61 8.41 19 8.41 19 Spain 107 Kuwait 107 5.59 5.59 Iceland 8.40 20 8.40 20 5.52 108 5.52 108 St. Lucia Austria 8.38 21 8.38 21 Indonesia 109 5.48 109 5.48 8.38 22 8.38 22 Israel 5.46 110 Lebanon 110 5.46 Germany 8.36 23 23 8.36 Samoa 111 5.44 111 5.44 Poland 24 8.35 24 8.35 5.33 112 5.33 112 Egypt Chile 8.30 25 8.30 25 5.28 5.28 Paraguay 113 113 26 8.27 8.27 Belgium 26 5.25 Viet Nam 114 114 5.25 8.25 27 Czech Republic 115 Kiribati 115 5.18 5.18 8.25 27 5.02 El Salvador 116 5.02 116 Argentina 28 8.18 28 8.18 Cape Verde 4.89 4.89 117 117 United Kingdom 29 8.18 29 8.18 4.79 118 Seychelles 4.79 118 Sweden 8.17 30 8.17 30 4.72 Suriname 4.72 119 119 31 8.15 8.15 31 Switzerland 4.44 4.44 120 Ghana 120 8.12 8.12 32 32 Latvia Honduras 121 4.36 121 4.36 33 33 Hong Kong, China 8.11 8.11 8.04 8.04 34 Bulgaria 34 4.29 4.29 India 122 122 35 7.97 35 Japan 7.97 Guatemala 4.29 123 4.29 123 France 7.94 36 7.94 36 4.23 Nicaragua 124 4.23 124 7.82 37 37 Hungary 7.82 4.22 Syria 4.22 125 125 Croatia 38 7.79 38 7.79 4.15 126 4.15 126 Maldives Turkey 39 7.72 7.72 39 4.09 4.09 Morocco 127 127 40 7.70 7.70 Monaco 40 Timor-Leste 128 4.01 128 4.01 41 7.69 7.69 41 Italy 129 Swaziland 129 3.86 3.86 Barbados 42 7.69 42 7.69 Namibia 130 3.85 3.85 130 7.68 7.68 43 Cyprus 43 Bhutan 3.84 131 3.84 131 7.63 Venezuela 7.63 44 44 3.81 132 3.81 Gabon 132 7.57 45 7.57 45 Slovakia 3.76 133 Kenya 133 3.76 St. Kitts and Nevis 46 7.55 46 7.55 Vanuatu 3.65 134 3.65 134 47 7.51 47 Portugal 7.51 3.60 Cameroon 135 3.60 135 Serbia 48 7.48 48 7.48 Lao P.D.R. 3.60 136 3.60 136 Bahamas 49 7.43 7.43 49 137 137 3.51 3.51 Bangladesh Kazakhstan 7.41 7.41 50 50 Nepal 138 3.50 138 3.50 7.37 51 7.37 51 Romania Zimbabwe 139 3.38 139 3.38 52 7.36 52 Albania 7.36 Lesotho 3.37 140 3.37 140 Georgia 53 7.34 53 7.34 Solomon Islands 141 141 3.27 3.27 54 Montenegro Togo 142 3.16 142 3.16 54 7.34 7.34 143 3.13 143 3.13 Nigeria 7.30 55 7.30 55 Saudi Arabia 3.12 3.12 144 144 South Sudan 56 7.27 56 Singapore 7.27 Myanmar 145 3.06 145 3.06 7.25 Cuba 57 57 7.25 Zambia 3.06 146 3.06 146 Mongolia 58 7.23 58 7.23 Benin 3.06 147 3.06 147 Macao, China 59 7.19 59 7.19 3.01 148 Congo (Dem. Rep.) 148 Armenia 60 7.17 60 7.17 3.01 7.04 61 7.04 61 Costa Rica 3.00 149 Cambodia 3.00 149 62 Uruguay 62 7.02 7.02 150 2.78 150 Pakistan 2.78 6.99 63 Grenada 63 6.99 2.77 Madagascar 151 2.77 151 64 6.97 6.97 Moldova 64 Gambia 2.66 152 2.66 152 Iran (I.R.) 65 6.96 65 6.96 153 2.65 153 2.65 Afghanistan Kyrgyzstan 66 6.96 66 6.96 Sudan 2.62 154 2.62 154 67 6.86 67 Malta 6.86 155 155 2.59 Djibouti 2.59 68 Jordan 68 6.68 6.68 Côte d'Ivoire 2.57 156 2.57 156 6.60 69 Peru 6.60 69 2.56 157 2.56 157 Liberia 70 Luxembourg 6.59 70 6.59 2.54 Yemen 158 158 2.54 6.50 Bahrain 71 6.50 71 159 2.43 2.43 159 Angola Azerbaijan 6.47 72 6.47 72 2.43 Uganda 2.43 160 160 Mauritius 73 6.45 73 6.45 161 2.42 161 Rwanda 2.42 74 Colombia 74 6.44 6.44 Tanzania 2.33 162 2.33 162 Sri Lanka 75 6.41 75 6.41 2.30 Malawi 2.30 163 163 Ecuador 76 6.37 6.37 76 Equatorial Guinea 164 2.27 164 2.27 Brunei Darussalam 6.31 77 6.31 77 Guinea 165 2.19 2.19 165 78 6.27 6.27 78 Bosnia and Herzegovina 166 2.17 166 Senegal 2.17 6.23 79 6.23 South Africa 79 167 2.15 167 2.15 Mali Thailand 80 6.21 80 6.21 Mauritania 168 2.02 168 2.02 Palestine 169 169 Burundi 2.01 2.01 81 6.18 6.18 81 1.87 170 170 1.87 Guinea-Bissau 82 6.17 82 6.17 Antigua & Barbuda 171 Mozambique 1.74 171 1.74 6.13 83 TFYR Macedonia 6.13 83 1.71 172 1.71 172 Ethiopia Philippines 84 6.11 6.11 84 Burkina Faso 1.48 173 1.48 173 Dominica 85 6.11 85 6.11 174 Chad 1.30 174 1.30 Algeria 6.10 86 6.10 86 Niger 175 1.01 6.04 87 6.04 87 Uzbekistan 175 1.01 88 88 6.03 6.03 Qatar 2014) of the ITU Plenipotentiary (Rev. Busan, 99 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution Conference. Source: ITU. Measuring the Information Society Report 2016 15

32 • The low quartile contains the 44 least continent), one in the Americas and nine in the connected countries (LCCs), from Côte d'Ivoire Asia-Pacific region. with an IDI value of 2.86 to Niger with an IDI value of 1.07. This relative stability of the Index year on year reflects steady progress in almost all countries on the indicators for access and use. Only five countries registered a decline in their IDI 2016 Overall distribution of IDI rankings value, all by less than 0.05 points. Three of these are northern European countries near the top of The country topping the IDI rankings in 2016, the rankings (Denmark, Sweden and Finland); the as in 2015, is the Republic of Korea, with an IDI two others are developing countries (Antigua and 2016 value of 8.84, up from 8.78 in IDI 2015. Two Barbuda and Kiribati). further economies in the Asia-Pacific region – Hong Kong (China) and Japan – also rank in the The average improvement in IDI values across top ten, along with seven countries from Europe all economies was 0.20 points, for an average (Iceland, Denmark, Switzerland, the United IDI score of 4.94. The scale of improvement was Kingdom, Sweden, the Netherlands and Norway). highest in the middle of the distribution. The The differences in IDI 2016 values among these improvement in IDI values for the top ten countries highest-performing countries are relatively small, was on average just 0.07, partly reflecting the fact with only 0.47 points separating those in first and that they are already approaching the maximum tenth positions. This reflects the high level of ICT attainable figure under the current Index. The development that has been achieved in many average improvement for the high quartile was developed and some high-income developing 0.13, for the upper-middle quartile 0.24 and countries, which have experienced high levels of for the lower-middle quartile 0.27. For LCCs investment in ICT infrastructure and innovation as at the bottom of the distribution, the average well as high levels of adoption of new services by improvement was lower, at 0.15 points. consumers. These high-performing countries also rank towards the top of the rankings for GNI p.c. The distribution between developed and and other economic indicators. developing countries, and the particular challenges faced by LDCs, are discussed in There has been relatively little change in the section 1.4 below. The analysis suggests that the IDI rankings for most economies between gap between developed countries and higher- 2015 and 2016. Only one change took place income developing countries is diminishing, in the countries making up the top ten ranked partly because of higher gains among developing economies, for example, with Japan narrowly countries on some indicators where developed displacing Luxembourg in tenth position; and only countries have already attained high levels of one change in the composition of the ten lowest- performance. However, developed countries and a ranked countries, where Guinea has replaced few developing countries (such as the Republic of Madagascar. Only two countries (St. Kitts and Nevis Korea, Hong Kong (China) and Singapore) now have and Portugal) moved up into the highest quartile access to much higher broadband speeds and to from the one below. Only eight (St. Kitts and Nevis, more sophisticated digital services, which are not Myanmar, Algeria, Dominica, Grenada, Rwanda, reflected in the Index, than they did five years ago. Côte d’Ivoire and Bolivia) climbed more than five The gap between these higher-income countries places in the rankings, and only two (Saint Lucia and the majority of developing countries may thus and Saudi Arabia) fell by more than five places. be widening in terms of these higher speeds and more sophisticated services. In the lowest quartile, only four countries (Côte d’Ivoire, Myanmar, Rwanda and Liberia) improved The discussion in section 1.4 also shows, however, their position in the rankings by five or more that the gap between the majority of developing places, while two (Sudan and Kiribati) fell by five countries, on the one hand, and LDCs and LCCs, on places. Of the 44 countries ranked as LCCs, 30 are the other, is growing. This widening digital divide in the Africa region, including the ten countries is a cause of particular concern in the light of the with the lowest rankings, while four are in the Arab role which ICTs are expected to play in efforts to States region (three of which are on the African achieve the SDGs. Measuring the Information Society Report 2016 16

33 Chapter 1 tribution of IDI values between regions Dis Chart 1.4: Source: ITU. The distribution of IDI values between regions countries with very high IDI 2016 scores, the spider charts show high levels of attainment across all the is illustrated in Chart 1.4. The columns clearly highlight the preponderance of high- indicators, although there are some differences performing countries in the Europe region and that are worthy of note. of low-performing countries in the Africa region. Aspects of the Republic of Korea’s ICT These two regions are relatively economically performance are outlined in Box 1.2. There has homogeneous. All but three countries in the been very little change in individual indicators Europe region are developed countries, while all of in the country during the past year, the most those in the Africa region that are included in the significant being a marginal decline in fixed- IDI are developing countries, and 25 of them are telephone subscriptions matched by marginal LDCs. The CIS region, most countries of which are increases in mobile-cellular and fixed-broadband found in the upper half of the distribution, is also subscriptions and in the number of Internet relatively homogeneous. Other regions are more users. The country’s overall IDI performance is heterogeneous, with a wider spread including both boosted by very high values for fixed-telephone high- and low-income countries and one or more subscriptions and for tertiary enrolment. However, LDCs. These regional characteristics are discussed its IDI values for the percentage of households in Chapter 2. with a computer and for international Internet bandwidth per Internet user are notably lower than those of Iceland and Denmark, indicating that Top performing countries there is scope for further progress in performance in these areas. The top performing country in the 2016 IDI, as in IDI 2015, is the Republic of Korea, followed Aspects of the IDI performances of Iceland and by two Nordic countries, Iceland and Denmark, Denmark are outlined in Boxes 1.3 and 1.4. which have exchanged places during the year. Iceland's overtaking Denmark in the IDI 2016 Spider charts illustrating the performance of these rankings is attributable to movements in two three countries on all the indicators in the IDI are indicators within the Index. Mobile-broadband 1.5. As is to be expected of presented in Chart penetration in Iceland rose significantly during Measuring the Information Society Report 2016 17

34 IDI values for top-ranking countries, 2015 and 2016 Chart 1.5: Source: ITU. the year, while Denmark had already attained Internet users. Iceland’s performance in the access almost 100 per cent penetration by 2015. At sub-index significantly exceeds that of the Republic the same time, fixed-telephone penetration in of Korea (9.42 against 8.99), and lags only slightly Denmark dropped more substantially than it did in behind it in the use sub-index (8.44 against 8.57). Iceland. This fall in fixed-telephone subscriptions Denmark’s performance in relation to the two in Denmark may be the result of fixed-mobile other leading countries is primarily determined by substitution, as individuals and households in its low level of fixed-telephone penetration. Both some high-income developed countries no longer Nordic countries, however, fall well below the choose to obtain or maintain fixed as well as Republic of Korea in one of the proxy indicators mobile subscriptions. that make up the skills sub-index – tertiary enrolment – while Iceland also has a significantly Iceland and Denmark both display higher IDI values lower value for mean years of schooling. As a than the Republic of Korea on several indicators, result, they register much lower overall values in particularly the proportion of households with the skills sub-index (8.87 for Denmark and 8.40 for a computer, but also international Internet Iceland against 9.08 for the Republic of Korea). bandwidth per Internet user and the proportion of Measuring the Information Society Report 2016 18

35 Chapter 1 Box 1.2: ICT and IDI developments in the Republic of Korea The Republic of Korea has consistently ranked as one of the most connected countries in the IDI. New initiatives and developments in 2015 have further improved its ICT environment, reinforcing its position among the top performers in the Index. The government aims to improve competition for mobile subscribers further by licensing an additional operator and through legislation which seeks to increase the market share of low-price 7 service plans from 10 per cent to 12 per cent. New legislation should allow operators to launch new tariffs without approval, enabling them to respond more quickly to consumer demand. The country also remains ahead in new developments and technologies. In 2015, the Republic of Korea’s operator SK Telecom (SKT) launched what it claims to be the world’s first commercial tri- band LTE-Advanced (LTE-A) service, offering downlink speeds of up to 300 Mbit/s by aggregating 8 three component carriers in three different frequency bands. The government is actively promoting use of the Internet across the entire population in order to extend the benefits of its high ICT development to currently unconnected groups. Government initiatives, such as the “Development and Supply of IT Assistance Devices”, “Supply of Green PCs of Love” and “Telecommunication Relay Service”, designed for hearing- and speech-impaired people, are examples of responses to ensure that disadvantaged groups have equal opportunity to access information (KISA, 2015). While use of the Internet is increasing among women and girls, household data collected by the Ministry of Science, ICT and Future Planning (MSIP) and the Korea Internet and Security Agency (KISA) still show that there is a gender gap in Internet use in the Republic of Korea. This contrasts with many European countries, where the gender differences are minor, resulting in a higher overall Internet uptake. Household data further reveal that Internet use by the elderly (those aged 75 and over) is well below that of other countries with high Internet uptake. While every second elderly person in Japan uses the Internet, and one in four in Switzerland, only one out of eight elderly persons in the Republic of Korea state that they use the Internet. This might reflect the fast pace of ICT development, as well as the fact that economic developments in the country are more recent in comparison with other high-income economies. This is discussed in detail in Chapter 6. Box 1.3: ICT and IDI developments in Iceland I celand has overtaken Denmark to rank second in IDI 2016. The main reason for this is a significant increase (11 per cent) in the number of mobile-broadband subscriptions in the country, lifting Iceland to fifth place in the use sub-index in 2016, up from seventh in 2015. As in IDI 2015, Iceland ranks second in the access sub-index, mainly because of high levels of access to computers and the Internet. As many as 98 per cent of Icelandic households are estimated to have access to a computer, the highest ratio in the world. Iceland also has the highest share of population using the Internet worldwide, at 98.2 per cent, boosted by an increase in female Internet participation from 92 per cent in 2010 to 98 per cent in 2014. th in the skills sub-index, however, mainly because of a relatively low score on Iceland ranks 20 mean years of schooling (10.59) compared with other Nordic countries. Measuring the Information Society Report 2016 19

36 Box 1.3: ICT and IDI developments in Iceland (continued) In May 2016, the Icelandic operator Siminn (Iceland Telecom) announced that its LTE network covered 91 per cent of the country’s population, only two years after deployment in January 2014 9 (and three years since the operator Nova launched the country’s first LTE network). This follows an upgrade of Siminn’s LTE transmitters to allow for an increase in maximum download speeds over the 4G network from 100 Mbit/s in 2015. In 2015, the Icelandic Mbit/s at launch to 150 4 million for the development of high- parliament, the Althingi, approved expenditure of USD speed networks in 2016, with the aim of bridging the final digital divide in Iceland and allowing almost all households in the country to have access with at least a 100 Mbit/s connection by the year 2020 (Post and Telecom Administration in Iceland, 2015). The Post and Telecom Administration (Póst- og Fjarskiptastofnun) (PTA) has announced plans to MHz band by the end of increase competition by holding an auction for frequencies in the 700 2016. Box 1.4: ICT and IDI developments in Denmark D enmark has dropped one place in the 2016 IDI ranking to third, just behind the Republic of Korea and Iceland, with an IDI score of 8.74. The main reason for its lower ranking is a 10 p er cent decrease in fixed-telephone subscriptions, resulting in a decline in the access sub-index and thereby in the IDI as a whole. The number of fixed-telephone subscriptions has fallen by nearly p er cent since its peak in 2001, from 72.2 to 29.9 subscriptions per 100 inhabitants in 2015. 60 However, Denmark tops the use sub-index, mainly because of its high fixed-broadband er cent). p penetration (42.5 subscriptions per 100 inhabitants) and high Internet use (96.3 Denmark is also one of the few countries with a higher share of female Internet users than male (96.4 p er cent compared with 96.2 er cent). The country ranks sixth in the skills sub-index, and p first among the Nordic countries, largely because of its high score on mean years of schooling (12.7). Like other countries near the top of the IDI, Denmark is a leader in the adoption of new technologies. By the end of 2015, almost the entire population of Denmark was covered by an LTE network - just five years after the launch of TeliaSonera’s first commercial LTE service in 12 10 11 , TDC and Telenor ) all commenced December 2010. In 2015, the three largest operators (Telia deployment of 4G+ or LTE-A networks using carrier aggregation (CA) technology over several bit/s. M frequency bands. This new technology enables theoretical download speeds of up to 300 These developments are in line with Denmark’s national broadband strategy, which aims to enable all households and businesses to have access to at least 100 M bit/s download and 13 30 bit/s upload speeds by 2020. M countries in the IDI as a whole during the year Most dynamic countries 2015-2016 in terms of ranking and value. Countries’ movement within the IDI can be There is broad correspondence between countries measured in terms of changes in their IDI ranking registering the strongest improvements in ranking 1.6 sets out the most and/or their IDI value. Table and in value. Only two of the top performers dynamic gains which have been made by individual in terms of improved IDI ranking – Rwanda and Measuring the Information Society Report 2016 20

37 Chapter 1 st dynamic countries in IDI rankings and values, 2015-2016 Mo Table 1.6: Change in IDI ranking Change in IDI value (absolute) IDI rank IDI rank IDI value Country IDI rank 2016 Country change 2016 change 20 34 St. Kitts and Nevis 34 St. Kitts and Nevis 0.98 Myanmar 13 140 0.66 103 Algeria 103 Algeria 9 Bhutan 117 0.62 69 8 Dominica 0.59 Myanmar 140 74 Grenada 8 61 0.58 Malaysia 8 Rwanda 150 0.57 Dominica 69 Côte d'Ivoire 7 132 Bolivia 111 0.53 Bolivia 111 6 Grenada 74 0.46 117 Bhutan 5 Côte d'Ivoire 0.44 132 5 Malaysia 61 120 Namibia 0.43 Source: ITU Liberia – are not also in the top ten countries in St. Kitts and Nevis and Myanmar are outlined in Boxes 1.5 and 1.6. Analysis of the performance terms of improvements in IDI value. of other dynamic countries at a regional level is included in Chapter 2. Spider charts illustrating the IDI values of the three countries which have improved their position most The spider charts show a marked difference dynamically in the IDI 2016 rankings – St. Kitts between the experience of the most dynamic and Nevis, Myanmar and Algeria – are presented th 1.6. Aspects of the IDI performance of in Chart country, St Kitts and Nevis, which ranks in 34 Chart 1.6: IDI values for most dynamic countries, 2015 and 2016 Source: ITU. Measuring the Information Society Report 2016 21

38 Box 1.5: ICT and IDI developments in St. Kitts and Nevis and other Eastern Caribbean countries th The most dynamic country in IDI 2016 is St. Kitts and Nevis, which moved up 20 places to 34 in the rankings in this year’s Index. Very substantial gains were also made by two of its neighbours in the Eastern Caribbean, Dominica and Grenada. These countries have seen improvements in most of the indicators making up the IDI, although with different emphases. St. Kitts and Nevis experienced a very significant and rapid increase in active mobile-broadband subscriptions, building on a strong performance in mobile-cellular subscriptions. The mobile- broadband penetration rate increased from just 18.6 per 100 inhabitants in 2014 to 71.0 in 2015. A similar trend can be observed in Grenada, where mobile-broadband penetration rose from only 2.6 per 100 inhabitants in 2014 to 28.8 in 2015. Dominica saw an increase in active mobile- broadband subscriptions from 29.3 to 42.2 per 100 inhabitants in the course of the past year, as well as substantial improvements in fixed-broadband penetration, the proportion of Internet users, and households with access to the Internet. A number of factors appear to have contributed to this trend. In St. Kitts and Nevis, the cost of prepaid mobile broadband decreased from USD 13.7 in 2015. A number 51.9 in 2014 to USD of promotions and zero-rating options also became available, which may have had an impact. Digicel offered occasional promotions of free credit in 2014. At the end of 2014 and throughout 2015, Digicel offered free Wikipedia, free social media (Facebook, Twitter, Instagram, and later on WhatsApp), and triple credit promotions several times a month. The two main operators in the region, Digicel and Cable & Wireless Communications (C&W), have undertaken large infrastructure investments that support this improvement in local IDI 250 million to improve and upgrade values. C&W launched Project Marlin in 2014 to invest USD 14 its network across the region. According to C&W, this has enabled significant improvements in network resilience and speed, and its network is carrying 104 pe r cent more traffic on mobile and per cent more traffic on fixed networks in 2016 in comparison with previous years. Digicel has 42 also invested heavily in the region in the three years ending on 31 March 2015 (Digicel, 2015). position, in the high quartile of the IDI, and rise in the proportion of active mobile-broadband rd and Algeria and Myanmar, which rank in 103 subscriptions. th 140 positions, in the lower-middle and low (LCC) The shapes of the spider charts for the two other quartiles, respectively. highly dynamic countries illustrated above – St. Kitts and Nevis is a small island developing Algeria and Myanmar – differ markedly from that country in the Caribbean with a very high level of for St. Kitts and Nevis but resemble one another. Like many countries in the lower half of the overall mobile-cellular penetration, lower but substantial fixed-telephone penetration and a high level of distribution, these countries have relatively high values for mobile-cellular subscriptions but very international bandwidth per Internet user. It also low values for fixed-telephone subscriptions; displays relatively high levels of performance relatively high values for international Internet in the skills sub-index, including a degree of bandwidth per Internet user but low values for tertiary enrolment on a par with many developed fixed-broadband subscriptions; and relatively high countries. It has improved its performance in the levels of secondary enrolment but low values for use sub-index, which has propelled the country up tertiary enrolment. the IDI 2016 rankings as a whole. Its use sub-index value climbed from 4.31 to 6.53, a far bigger rise than any other country, and its ranking on that Algeria’s relative prosperity compared with nd th sub-index rose correspondingly, from 64 4 870 in Myanmar – it had a GNI p.c. of USD . to 32 2015, according to the World Bank, as against St. Kitts and Nevis’ improved performance on the 15 use sub-index has been driven by a spectacular – is 1 just USD 280 for Myanmar the year before Measuring the Information Society Report 2016 22

39 Chapter 1 Box 1.6: ICT and IDI developments in Myanmar The Government of Myanmar passed a new Telecommunications Law in 2013 (MCIT, 2013) which opened up the country’s telecommunication market. Since then, Myanmar has seen rapid improvement in ICT access and use, moving up 13 places in the global IDI rankings between 2015 and 2016. Myanmar has consistently been placed among the top five fastest growing telecom 16 markets in the world recently, and in early 2016 was ranked second after India. The most significant improvements in Myanmar are seen in mobile-cellular subscriptions and Internet access. The entry of two new operators in the market, Qatari Ooredoo and Norwegian Telenor, in 2014, proved a significant driver for mobile-cellular uptake. Competition led to a .5 significant decrease in prices, the cost of a SIM card falling from USD 50 in 2013 to just USD 1 1 in 2015 (A4AI, 2015). ITU’s ICT Price Basket for mobile-cellular subscriptions shows that the 4 .8 in 2014 to USD 1 .9 in 2015, while the cost of a cost of a subscription went down from USD .41 in 2015. Operators have mobile-broadband subscription fell from USD 1 0.16 in 2014 to USD 2 also been making zero-rating and/or cheaper data plans available, while handset costs have decreased significantly with Ooredoo launching a subsidized 3G phone for less than USD 1 5 (Ooredoo, 2015). Infrastructure roll-out in Myanmar has been very rapid of late. Telenor alone invested more than USD 7.6 billion between 2014 and 2015. Ooredoo also invested heavily in network expansion 17 with loan support from the Asian Development Bank and International Finance Corporation, prioritizing data with an initial 3G-only network roll-out. As of April 2016, its network covered per cent of the population. 85 These improvements in affordability and network availability help to explain the country’s fast Internet uptake, supported by two other trends that distinguish Myanmar from neighbouring per cent of phone owners in markets. Firstly, smartphone penetration is very high, reaching 66 18 early 2015 according to a survey by the independent research institute LIRNEasia. According to per cent of mobile-phone users opt for smartphones (Oxford Business Ooredoo, as many as 80 Group, 2015). Secondly, growth in data usage has overtaken growth in voice traffic. Data per cent between January and June 2015, from Telenor indicate that data usage grew by 196 per per cent for voice traffic (Oxford Business Group, 2015). As many as 52 compared with 93 cent of Telenor’s mobile-phone subscribers are active data users. subscriptions. It was these developments that reflected in its higher values across all indicators. enabled these countries to move ahead of others The principal difference between the two with comparable IDI values and rankings in 2015. countries’ spider charts, other than this, lies in the growth of mobile-cellular subscriptions over the year. This was marginal in Algeria, which had already attained close to 100 per cent penetration Improvements in IDI values of mobile subscriptions by 2015; but rose sharply in Myanmar, where mobile telephony only became As well as assessing economies’ performance in 1.6). widely available in 2014 (see Box terms of the IDI value itself, it is also important to assess the progress they are making in relation to The growth in IDI performance in both Algeria their own previous performance. Table 1.7 lists and Myanmar was otherwise propelled by economies in order of the improvement achieved improvements on the indicators for the proportion in their overall IDI value between IDI 2015 and of Internet users, households with Internet, IDI 2016. Tables 1.9 and 1.11 do the same for the and the penetration of mobile-broadband access and use sub-indices. Measuring the Information Society Report 2016 23

40 Table 1.7: IDI value change, 2015-2016 Rank Rank IDI IDI value IDI IDI value IDI IDI IDI IDI Economy Economy change change 2016 2015 2016 2015 2016 2016 8.31 8.13 12 Germany St. Kitts and Nevis 0.17 0.98 6.23 7.21 34 3.74 103 130 4.40 2.95 2.78 0.17 Samoa Algeria 0.66 Bhutan 94 117 3.74 3.12 0.62 Saint Lucia 0.17 4.68 4.85 Myanmar 2.22 2.05 0.16 Zambia 147 0.59 1.95 2.54 140 61 6.22 5.64 0.58 Malaysia Spain 0.16 7.46 7.62 26 8.83 8.66 0.16 Iceland 2 Dominica 0.57 5.14 5.71 69 4.02 16 8.11 111 7.95 0.16 France Bolivia 0.53 3.49 Grenada 106 74 5.43 4.97 0.46 Palestine 0.16 4.12 4.28 Mozambique Côte d'Ivoire 132 2.86 2.43 0.44 0.16 1.60 1.75 163 Namibia 51 6.58 6.43 0.16 Serbia 0.43 3.20 3.64 120 7.69 7.53 0.16 Austria 92 Mexico 0.42 4.45 4.87 23 Russian Federation 0.16 6.79 6.95 43 90 4.95 4.54 0.41 Mongolia 5.75 5.60 0.15 Moldova 68 China 0.39 4.80 5.19 81 5.97 5.82 0.15 TFYR Macedonia 85 Jordan 0.38 4.67 5.06 65 Israel 0.15 7.25 7.40 30 108 4.17 3.79 0.38 Botswana 7.11 6.96 0.15 United Arab Emirates 38 Cape Verde 0.37 4.23 4.60 97 0.15 Guinea 5.04 165 1.72 1.57 86 Maldives 0.35 4.68 Uruguay 52 47 6.79 6.44 0.35 Kazakhstan 0.14 6.42 6.57 Mali Morocco 96 4.60 4.26 0.35 0.14 2.00 2.14 149 Nicaragua Vanuatu 127 3.08 2.73 0.35 0.14 2.74 2.88 131 Belgium Romania 60 6.26 5.92 0.34 0.14 7.69 7.83 22 Kyrgyzstan Rwanda 150 2.13 1.79 0.34 0.14 3.85 3.99 113 Tunisia Slovenia 7.10 7.23 33 95 4.83 4.49 0.34 0.13 5.18 5.05 0.13 Thailand 82 Cambodia 0.34 2.78 3.12 125 3.05 Timor-Leste 88 2.92 0.13 South Africa 0.34 4.70 5.03 128 Malawi 0.13 1.49 1.62 168 119 3.66 3.32 0.33 Belize 5.33 5.21 0.13 Ukraine 76 Iran (I.R.) 0.33 4.66 4.99 89 6.90 Qatar 124 6.78 0.12 Gabon 0.32 2.81 3.12 46 Hungary 0.12 6.60 6.72 48 35 7.18 6.87 0.31 Barbados 2.53 2.41 0.12 Senegal 141 Argentina 0.31 6.21 6.52 55 8.07 Estonia 99 7.95 0.12 Jamaica 0.31 4.20 4.52 18 Madagascar 0.12 1.57 1.69 166 44 6.94 6.64 0.30 Portugal 7.58 7.47 0.12 Macao, China 28 Philippines 0.30 3.97 4.28 107 3.20 Guatemala 114 3.09 0.11 Tonga 0.30 3.63 3.93 123 United States 0.11 8.06 8.17 15 91 4.92 4.62 0.29 Albania 3.32 3.21 0.10 Syria 122 Montenegro 0.29 5.76 6.05 62 164 1.73 1.62 0.10 Afghanistan 134 2.76 Lesotho 0.29 2.47 Uzbekistan 167 110 4.05 3.76 0.29 Tanzania 0.10 1.54 1.65 Chad Trinidad & Tobago 67 5.76 5.48 0.28 0.10 1.00 1.09 174 Cuba Dominican Rep. 104 4.30 4.02 0.28 135 2.73 2.64 0.10 39 Lithuania 0.10 7.00 7.10 Mauritius 0.28 5.27 5.55 73 2.16 0.10 Cameroon 57 2.07 Costa Rica 0.27 6.03 6.30 148 105 4.29 4.02 0.27 Viet Nam Japan 0.09 8.28 8.37 10 7.96 7.86 0.09 Monaco 19 Brazil 0.27 5.72 5.99 63 1.82 0.09 Djibouti 87 1.73 Seychelles 0.27 4.77 5.03 161 36 7.13 6.86 0.27 Greece El Salvador 0.09 3.64 3.73 118 3.52 3.44 0.09 Guyana 121 Slovakia 0.26 6.69 6.96 42 1.83 0.09 Benin Armenia 158 1.92 0.26 5.34 5.60 71 Honduras 0.09 3.00 3.09 126 Burundi 0.26 1.16 1.42 171 0.09 159 1.86 1.78 Togo Bulgaria 0.26 6.43 6.69 49 Poland 0.08 6.56 6.65 50 Georgia 0.26 5.33 5.59 72 0.08 53 6.54 6.45 Kuwait Fiji 0.25 4.16 4.41 102 1.86 1.94 Uganda 0.08 Cyprus 0.25 6.28 6.53 54 157 Brunei Darussalam 112 3.75 3.99 Ghana 0.08 5.25 5.33 77 0.25 2.35 2.27 0.08 New Zealand 0.24 8.05 8.29 13 Bangladesh 145 Angola 156 1.73 0.24 Liberia 1.97 0.08 1.95 2.03 154 5.07 78 5.32 St. Vincent and the Grenadines 0.24 Norway 9 8.42 8.35 0.07 0.24 25 7.62 7.55 0.07 Canada 137 2.48 2.72 Nigeria Singapore 7.88 7.95 20 0.07 70 5.69 5.45 0.24 Turkey 0.07 6.04 0.24 Oman Netherlands 8.36 8.43 8 59 6.27 0.24 7.02 6 7.26 8.46 8.40 0.06 Hong Kong, China 31 Belarus Czech Republic 0.06 7.20 7.25 32 4.63 0.24 93 Panama 4.87 155 Chile 0.24 6.11 6.35 56 2.02 1.96 0.06 Yemen 136 Korea (Rep.) 0.06 8.78 8.84 2.49 2.73 Swaziland 0.23 1 Gambia 0.23 2.21 2.45 144 143 Lao P.D.R. 2.40 0.06 2.46 115 5.22 0.05 Venezuela 79 5.27 Indonesia 0.23 3.63 3.86 6.28 Italy 6.89 7.11 37 Azerbaijan 0.05 6.23 0.23 58 Mauritania 151 2.12 1.90 0.22 South Sudan 0.05 1.36 1.42 172 27 Andorra 0.22 7.39 133 2.78 2.73 0.05 Zimbabwe 7.61 0.22 Guinea-Bissau 1.38 0.05 169 Ethiopia 173 1.51 1.29 1.34 1.99 0.05 Solomon Islands 2.04 Bosnia and Herzegovina 153 80 5.25 5.03 0.22 2.56 2.99 2.78 0.21 Kenya Sudan 2.60 139 0.05 129 Croatia 6.83 7.04 41 0.21 7.46 7.42 0.04 Bahrain 29 40 7.08 6.88 0.20 Latvia 1.07 1.03 0.04 Niger 175 Sri Lanka 0.20 3.56 3.77 0.03 Equatorial Guinea 160 1.85 1.82 116 6.90 0.20 Burkina Faso 162 Saudi Arabia 0.03 6.88 1.80 45 1.60 8.54 0.02 United Kingdom 5.09 5 4.89 0.20 Suriname 8.57 84 Malta 11 8.36 8.34 0.02 Luxembourg 0.20 7.49 7.69 24 Pakistan Ecuador 98 4.56 4.54 0.02 0.19 2.15 2.35 146 Paraguay Congo (Dem. Rep.) 1.48 0.02 0.19 3.88 1.50 109 170 4.08 21 7.73 7.92 0.19 Ireland Lebanon 0.01 5.91 5.93 66 4.42 0.19 Peru 14 4.23 Australia 0.01 8.18 8.19 101 152 2.06 2.07 -0.01 Kiribati India 0.19 2.50 2.69 138 8.68 8.50 0.18 Switzerland 4 Sweden -0.01 8.47 8.45 7 2.50 0.18 Nepal 3 2.32 Denmark -0.03 8.77 8.74 142 17 8.08 8.11 -0.03 Finland Colombia 0.18 4.98 5.16 83 4.44 4.26 0.18 Egypt 100 Antigua & Barbuda -0.04 5.41 5.38 75 0.18 Bahamas 64 5.80 5.98 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 24

41 Chapter 1 The average improvement in IDI value over the All but one of the top ten economies in IDI 2016 1.7 year, as noted earlier, was 0.20 points. Table also rank in the top 15 in the access sub-index, shows that, of the 175 economies in the Index, 73 although the highest performer in the access sub- th in the overall exceeded this average improvement, 12 of them index, Luxembourg, only comes 11 IDI 2016. Other countries towards the top of the improving their IDI values by more than twice the access sub-index displaying significantly higher global average. No fewer than 52 of these have performance in that sub-index than in the IDI as a IDI rankings within the upper-middle and lower- whole are Germany, Malta, Singapore and France. middle quartiles. The highest 34 countries ranked All of the top ten economies in IDI 2016 also by improvement in their IDI values are developing rank in the top 15 in the use sub-index, in which countries. Denmark is the highest performing country. At the other end of the scale, 45 economies had There is similarly broad consistency between the IDI values which improved by less than half the access and use sub-indices at the bottom of the average (less than 0.10 points), with five countries distribution: 11 of the 15 countries at the bottom experiencing a fall in value. Of these, 16 came of the use sub-index are also in the bottom 15 from the highest quartile of the Index, including countries in the access sub-index. However, eight of the ten highest-ranking economies in IDI 2016, while 15 came from the lowest quartile. This greater variations between rankings in these two sub-indices occur in the case of individual supports the earlier suggestion that, at least in countries towards the middle of the distribution. terms of indicators that are included in the Index, middle-ranking countries are closing the gap on Twenty-one countries in IDI 2016 have rankings the most-connected economies towards the top of for access which are 20 or more places higher the Index while drawing away from LCCs. than their rankings for usage, the greatest differences being observed for Brunei Darussalam (52 places), Ukraine (43 places), Seychelles (34 The access, use and skills sub-indices places), the Islamic Republic of Iran (31 places) and Mauritius (30 places). These imbalances suggest As in previous years, significant differences can that there is scope for greater deployment and be identified between scores on the overall IDI policy interventions to stimulate demand in these 2016 and on the three sub-indices of which it countries. 1.2, the is composed. As described in section per access and use sub-indices each make up 40 Conversely, 15 countries have rankings in the use per cent cent of IDI 2016, with the remaining 20 sub-index which are 20 or more places higher than derived from the skills sub-index. While the their rankings for access, the highest differences access and use sub-indices are composed of being observed for Finland (32 places), Nigeria ICT-specific indicators, the skills sub-index uses and Bhutan (each 31 places). In Finland, this is proxy indicators which are essentially concerned due to the country’s low level of fixed-telephone with educational attainment. The skills sub-index subscriptions. In Nigeria, Bhutan and other is therefore less directly related to ICTs than the developing countries, the imbalance suggests other sub-indices. that strong demand for services is not currently matched by adequate high-quality infrastructure, Not surprisingly, given the composition of the and that policy interventions to stimulate the Index, there is a strong level of correspondence supply side of the market may be required. between rankings in IDI 2016 and in the access and use sub-indices, but greater disparity between the As already noted, there is much more variation overall IDI and the skills sub-index. between the overall IDI and the skills sub-index, which is derived from non-ICT-specific indicators. Economies that rank higher in the access sub-index Only three of the top ten ranking countries in than in IDI 2016 tend to rank lower in the use IDI 2016 (the Republic of Korea, Denmark and sub-index, and vice versa. In some cases, among Norway) come within the top ten countries in the economies towards the top of the IDI, it is possible skills sub-index; and five of the top ten performers that somewhat lower rankings in the access sub- th th and 35 in IDI 2016 are ranked down between 29 index relative to the use sub-index are the result of in the skills sub-index. fixed-mobile substitution. Measuring the Information Society Report 2016 25

42 th th it ranks 57 in the use sub- , as against 150 The access sub-index th in the access sub-index. Other index and 167 countries whose access sub-index rankings are Rankings and values in the access sub-index for significantly below their overall IDI 2016 rankings IDI 2015 and IDI 2016 are set out in Table 1.3 include Lithuania towards the top of the overall above. Table 1.9 below ranks countries according distribution, Costa Rica and Albania in the middle to the change in access sub-index value they have of the distribution, and Kiribati towards the bottom achieved during the course of the year. of the distribution. In these cases, the difference indicates that infrastructure limitations are the There has been less movement in the access principal constraint on ICT development. sub-index than in the use sub-index in the year between IDI 2015 and IDI 2016. The average value Equally, a number of countries have access sub- of the access sub-index, at 5.58, is significantly index rankings which are considerably higher higher than that of the use sub-index (3.91) and than their overall IDI 2016 rankings. These include the IDI as a whole (4.94). It has improved by Malta, the United Arab Emirates, Portugal and 0.13 points over the year, as compared with an Barbados towards the top of the distribution; improvement of 0.37 in the use sub-index and 0.20 Mauritius, Seychelles and Brunei Darussalam in in IDI 2016 overall. the middle of the distribution; and El Salvador, Gambia and Mozambique towards the lower end A total of 19 countries, all of them developing of the distribution. In these cases, the difference countries, improved their access sub-index value indicates that efforts to increase usage of available by more than twice the average increase (by 0.27 infrastructure would have most positive impact on points or higher), while 20 economies saw their ICT development. value in this sub-index fall over the year. Half of the latter group came from the high quartile of IDI Table 1.8 identifies the countries which have performers, where there is little scope for further recorded the biggest increases in their access improvement in mobile-cellular penetration and sub-index rankings and values in the year 2015- fixed-telephone penetration is now in decline. 2016. While some of these countries also appear 1.5, which identifies the most dynamic in Table The highest values achieved in the access sub- performers in IDI 2016, this is far from universally index are significantly higher than those in the use the case, illustrating how the use sub-index, whose sub-index. Luxembourg tops the access sub-index average value has changed almost three times as rankings for 2016 with a sub-index value of 9.54, much as that of the access sub-index, has more followed by Iceland (9.42), the United Kingdom influence on changes in overall IDI rankings. It (9.24) and Hong Kong (China) (9.16). Of the 20 is notable, in particular, that St. Kitts and Nevis, lowest ranking countries in the access sub-index, which tops Table 1.8, 1.6, does not figure in Table 19 are LDCs, all but six of them in the African while the second most dynamic country in the region. access sub-index rankings, Barbados, does not appear among the most dynamic countries overall. Not surprisingly, the rankings of most economies in the access sub-index resemble their rankings in Table 1.8 shows the importance of comparing the overall IDI 2016, though there are significant improvements in sub-index value alongside exceptions. As many as 41 economies have access movements in the rankings. Seven of the rankings which differ by ten or more places from ten countries which showed the strongest their IDI 2016 rankings. improvements in sub-index values do not appear among the top ten improvers in the rankings, Two countries - Finland and Cuba - have overall including the country with the third highest IDI rankings which are more than 20 places higher improvement in its access sub-index value, than their access sub-index rankings. In the case Morocco. of Finland, where the difference is 22 places, this is primarily due to its low penetration of fixed- All but one of the indicators making up the access telephone subscriptions, probably as a result of sub-index showed some increase for the majority fixed-mobile substitution. The overall IDI ranking of countries between 2015 and 2016. The most for Cuba, meanwhile, is boosted by the country’s substantial average increases in this sub-index high performance on the skills sub-index, where Measuring the Information Society Report 2016 26

43 Chapter 1 cess sub-index, most dynamic countries, 2015-2016 Ac Table 1.8: Change in access ranking Change in access value Access rank Access value Access rank Access rank Country Country change change 2016 2016 144 Myanmar 0.63 Myanmar 144 15 Barbados 20 0.47 98 Algeria 8 7 98 0.42 Morocco 83 Algeria 0.38 Barbados 58 20 Mauritius 7 129 0.36 Côte d'Ivoire 17 5 New Zealand 0.35 Andorra 5 96 Mongolia 26 116 0.34 Dominica Dominican Rep. 5 75 0.31 149 5 Burkina Faso Uzbekistan 112 Austria 15 55 Trinidad & Tobago 5 0.31 0.30 5 46 168 Burundi Oman Source: ITU. in most regions occurred in the proportion There is also much more consistency between the of households with access to the Internet. In top countries in terms of improvement in rankings Africa, however, the biggest average increase and in values for this sub-index than for the access was in mobile-cellular subscriptions, while in sub-index. Nine of the top ten countries in terms the CIS region it was in households with access of improved value are among the ten countries to a computer. There was a fall in the indicator with the most improved rankings for this sub- for fixed-telephone subscriptions in a majority index. of countries, including both LCCs, which have historically been characterized by low levels of Seven countries saw increases over the year of fixed lines, and highly connected countries, where more than one whole point in this sub-index, with the numbers of fixed-telephone subscriptions have a further ten countries experiencing increases that historically been high. This development raises were also more than twice the average for the questions about the viability of measuring fixed- sub-index. All of these are developing countries, telephone subscriptions as a long-term indicator of the largest improvement for a developed country th in the table for ICT development. being that of Bulgaria, ranked 24 improvements in sub-index values. Meanwhile, 27 countries improved their value for the use sub- index by less than 0.10 of a point, of which three The use sub-index registered falls in value. Rankings and values in the use sub-index for IDI The highest ranking economies in the use sub- 2015 and IDI 2016 are set out in Table 1.4 above. index closely resemble those in the overall IDI. Table 1.11 below ranks countries according to the With two exceptions – Finland and Luxembourg in change in use sub-index value they have achieved place of Hong Kong (China) and the Netherlands – during the course of the year. the ten highest-ranking economies are the same as those in IDI 2016, and there is only one difference As indicated above, the use sub-index has in the top twenty, Macao (China) taking the place witnessed more substantial improvements in of Germany. It is a similar story at the bottom of values over the year 2015-2016 than the access the distribution, where 17 of the 20 lowest-ranking sub-index. The average value for the use sub-index countries in IDI 2016 appear among the 20 lowest in 2016 is 3.91, up 0.37 points (more than 10 per in the use sub-index. cent) on the figure of 3.54 in 2015. As a result, the use sub-index has had more influence on movements in IDI 2016 as a whole. Measuring the Information Society Report 2016 27

44 Table 1.9: Ac cess sub-index value change, 2015-2016 IDI access IDI access Rank IDI Rank IDI IDI access IDI access IDI access IDI access value value access access sub-index sub-index sub-index sub-index Economy Economy sub-index sub-index change change 2016 2016 2015 2015 2016 2016 2015-2016 2015-2016 3.08 3.22 0.13 Zimbabwe 144 Myanmar 137 2.45 3.35 0.63 0.47 98 114 4.47 4.34 4.56 Algeria 0.13 Guatemala 5.03 Cape Verde 83 Morocco 6.07 5.64 0.42 0.13 4.89 5.02 99 Barbados 0.38 20 8.24 7.86 Costa Rica 0.13 6.31 6.44 73 Côte d'Ivoire 163 2.39 0.13 Afghanistan 129 3.79 3.44 0.36 2.51 Mongolia 31 7.92 7.80 0.13 Spain 0.35 4.77 5.12 96 123 Honduras 0.12 4.04 4.17 Dominica 0.34 6.06 6.40 75 7.68 7.80 36 112 4.53 4.22 Uzbekistan 0.31 Belarus 0.12 2.41 2.29 0.12 Guinea-Bissau 164 Trinidad & Tobago 0.31 6.72 7.03 55 168 146 2.99 2.88 0.11 Mauritania Burundi 1.84 0.30 2.14 Montenegro 4.97 100 4.68 0.29 Fiji 0.11 6.74 6.85 60 Uruguay Iran (I.R.) 79 6.26 5.97 0.28 0.10 7.15 7.25 48 Djibouti Cambodia 122 4.21 3.93 0.28 0.10 2.44 2.55 162 Rwanda Armenia 70 6.57 6.29 0.28 0.10 2.54 2.65 159 Panama 7.56 7.46 0.10 Kazakhstan 42 0.28 5.72 5.99 84 95 5.29 5.01 0.28 Tunisia Gabon 0.10 3.88 3.98 126 8.28 8.18 0.10 Israel 18 Mauritius 0.27 6.59 6.86 58 5.20 Egypt 125 0.10 94 Bhutan 0.27 3.75 4.02 5.30 0.09 6.68 6.78 63 Azerbaijan Thailand 0.27 5.24 5.50 89 4.38 Kuwait 0.09 4.12 7.40 44 116 Dominican Rep. 0.26 7.31 4.16 Kyrgyzstan 0.09 4.25 121 170 2.11 1.85 0.26 Ethiopia Vanuatu 8.12 8.03 0.09 Monaco 25 0.26 3.40 3.66 131 0.09 Singapore Greece 0.26 8.70 8.61 34 7.60 7.85 11 Syria 0.09 4.58 4.66 109 143 3.16 2.91 0.26 Nepal 3.39 3.30 0.09 Pakistan 136 Croatia 0.25 7.33 7.58 41 0.08 Nicaragua Mexico 4.82 5.08 97 124 4.08 0.25 4.00 Netherlands 6.90 57 6.65 0.25 Romania 0.08 8.94 9.02 7 Albania Mozambique 2.82 2.90 148 106 4.73 4.48 0.25 0.08 4.95 4.88 0.08 El Salvador 101 Oman 0.25 7.12 7.37 46 Senegal 132 3.59 3.51 0.07 New Zealand 0.25 8.08 8.32 17 Belize Bolivia 130 3.69 3.62 0.07 0.25 4.13 4.37 117 Sri Lanka Iceland 2 9.42 9.35 0.07 113 4.51 4.26 0.25 133 Kenya 0.24 3.30 3.54 Lithuania 0.07 7.01 7.08 54 4.46 4.70 108 0.24 Philippines 5.78 Bosnia and Herzegovina 0.07 5.71 87 9.18 9.24 3 Burkina Faso 0.23 2.63 2.87 149 0.06 United Kingdom 2.65 2.04 1.99 0.06 171 2.42 0.23 Tanzania 158 Niger Palestine 5.35 93 5.12 0.23 175 1.34 1.28 0.06 South Sudan Botswana Seychelles 4.33 119 0.23 6.58 6.81 62 4.27 0.06 0.23 37 7.72 7.50 29 St. Kitts and Nevis 7.93 7.88 0.06 Slovenia 0.23 6.48 4.51 Ukraine 4.74 105 6.43 0.05 71 Ghana 4.43 6.63 6.68 Tonga 0.05 TFYR Macedonia 67 0.23 4.20 115 56 7.02 6.97 0.05 0.22 7.84 8.06 26 Andorra Cyprus Japan 6.75 6.53 0.22 66 0.05 8.75 8.80 10 Malaysia Italy 38 0.05 7.69 7.64 85 5.88 5.67 0.21 Suriname 0.21 1.74 1.94 173 1 9.54 9.49 0.05 Luxembourg Chad 90 Gambia 0.05 3.85 3.90 128 5.46 5.26 0.21 South Africa United Arab Emirates 0.20 7.94 8.14 52 7.21 7.16 0.04 Brunei Darussalam 24 5.45 5.26 0.19 China 6.29 Georgia 0.04 78 91 6.25 0.19 3.91 15 8.35 8.16 3.87 Austria 0.04 Timor-Leste 127 Turkey Togo 6.20 81 0.19 0.04 2.55 2.59 160 6.00 Gr enadines St. Vincent and the 0.04 7.19 7.23 49 Russian Federation 6.47 72 0.19 6.28 7.18 51 7.22 0.04 Serbia 5.91 Jordan 0.19 6.10 82 12 8.70 8.67 0.03 France Madagascar 165 0.18 2.21 2.39 9.04 9.01 Malta 6 0.03 45 Latvia 0.18 7.20 7.38 2.34 Uganda 0.02 2.37 166 142 3.03 3.21 0.18 Lao P.D.R. Czech Republic 0.02 7.44 7.46 43 Maldives 80 6.22 6.04 0.18 Norway 22 8.21 8.19 0.02 Indonesia 0.18 4.53 4.71 107 Malawi 172 2.03 2.01 0.02 Zambia 0.17 2.66 2.84 151 6.86 Bulgaria 0.01 6.85 59 Cuba 167 2.17 2.00 0.17 7.90 0.01 Qatar 7.91 33 110 Viet Nam 0.17 4.42 4.60 2.65 Yemen 157 0.01 2.66 103 4.65 4.83 0.17 Jamaica 7.99 7.98 Canada 28 0.01 India 0.17 3.15 3.32 139 Equatorial Guinea 0.00 2.76 2.77 153 0.17 141 3.28 3.11 Swaziland 6.57 6.57 69 0.00 Lebanon 6.81 6.65 Chile 0.17 61 Angola 154 2.76 2.75 0.00 8.02 Estonia 0.17 7.85 27 0.00 Belgium 8.34 16 8.34 2.57 Guinea 161 2.41 0.17 174 1.83 0.00 1.83 Congo (Dem. Rep.) Portugal 0.16 7.77 7.93 30 Australia 21 8.23 -0.01 8.24 Liberia 0.16 2.59 2.76 155 8 -0.01 8.99 9.00 Korea (Rep.) 0.16 19 8.27 8.11 United States Mali -0.01 3.31 3.30 140 3.27 0.16 134 Samoa 3.43 7.09 Poland 53 -0.02 7.11 7.91 32 7.76 0.16 Bahrain 3.35 138 -0.02 Sudan 3.33 Grenada 0.15 6.14 6.30 77 Venezuela 5.44 5.42 92 -0.02 0.15 64 Bahamas 6.77 6.62 35 Macao, China -0.03 7.85 7.83 3.41 135 Lesotho 0.15 3.26 169 2.11 2.14 -0.03 Kiribati Brazil 74 6.42 6.28 0.15 9.20 9.16 -0.04 Hong Kong, China 4 3.06 2.91 145 Bangladesh 0.14 Moldova 68 0.14 6.50 6.64 2.77 2.82 -0.04 Cameroon 152 -0.04 7.73 7.69 39 Finland 4.80 4.66 0.14 Peru 104 Switzerland 8.95 9.00 -0.05 9 50 7.22 Slovakia 0.14 7.08 Ireland -0.05 8.19 23 8.24 Argentina 0.14 6.63 6.77 65 Germany -0.08 9.17 9.09 5 2.73 2.59 0.14 Solomon Islands 156 Sweden 8.69 8.77 -0.08 13 4.33 Guyana 0.14 4.19 118 120 4.25 4.35 -0.09 Namibia 2.96 2.82 0.14 Nigeria 147 14 Denmark 8.52 8.65 -0.13 111 4.45 0.14 Paraguay 4.59 47 -0.22 Saudi Arabia 7.29 7.51 40 7.62 0.13 Hungary 7.49 76 6.56 -0.22 Antigua & Barbuda 6.34 150 2.86 2.72 Benin 0.13 102 4.90 5.16 -0.26 Ecuador 88 5.65 5.52 0.13 Saint Lucia 86 Colombia 0.13 5.70 5.83 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 28

45 Chapter 1 Although there is less divergence between use sub-index rankings and values in the year sub-index rankings and overall IDI rankings than 2015-2016. This table more closely resembles there is between access sub-index rankings and the representation of dynamic countries in IDI 2016 in Table overall rankings, there are nevertheless a number 6 than does the corresponding Table 1.8 for the access sub-index, because the of countries which perform significantly better or higher value changes which are observed for this worse in the use sub-index than in IDI 2016 as a sub-index mean that it has more influence on whole. changes in the overall IDI. The very substantial improvement in the use sub-index achieved by St. Countries that perform disproportionately well in Kitts and Nevis – almost one whole point above the use sub-index include Finland, the United Arab that of the next most dynamic country in this Emirates, Qatar, Kuwait and Malaysia towards the sub-index – is what has propelled it to the top of top of the distribution; Lebanon, China, Mexico and Bhutan in the middle of the distribution; and Table 1.6. This is primarily attributable to growth in mobile-broadband penetration. Myanmar, on Côte d’Ivoire, Nigeria, Sudan, Rwanda, Uganda and the other hand, has seen significant improvements Malawi towards the bottom of the distribution. in both the access and use sub-indices. Other In these economies, limitations in the availability strong performers in the use sub-index, with of infrastructure and devices are likely to be the improvements in value exceeding one point, major constraint on ICT development, thus calling include Bhutan, Malaysia, Namibia, Algeria, for stronger policy focus on these areas. Dominica and Bolivia. Other countries perform less well in the use There have been significant average gains in two sub-index than in the access sub-index or in IDI of the three indicators in this sub-index – Internet 2016. These include Slovenia, Greece, Armenia, users and mobile-broadband subscriptions – in all Mauritius, Ukraine and Brunei Darussalam in the regions during the year between IDI 2015 and IDI upper half of IDI 2016; the Islamic Republic of 2016. The biggest increase in all regions has been Iran (the most affected country) and Nicaragua in in mobile-broadband subscriptions. All regions the lower-middle quartile of IDI 2016; and Cuba also experienced growth in fixed-broadband and Kiribati among the LCCs. In these economies, subscriptions, but at a lower rate, which raises demand-side policies to stimulate usage of some concerns for the long-term development available infrastructure are most likely to raise the of high-capacity networks and services in less level of ICT development. developed regions. Table 1.10 identifies the countries which have achieved the most significant rise in their use Table 1.10: Us e sub-index, most dynamic countries, 2015-2016 Change in use ranking Change in use value Use rank Use value Use rank Use rank Country Country change change 2016 2016 32 32 St. Kitts and Nevis 2.22 St. Kitts and Nevis 32 129 94 17 Myanmar Bhutan 1.27 94 46 1.23 16 Malaysia Bhutan Namibia 109 15 Malaysia 46 1.18 15 1.17 Rwanda Algeria 108 136 111 65 1.10 Dominica Bolivia 12 111 Grenada 87 12 Bolivia 1.08 0.99 Grenada 87 109 Namibia 12 Algeria 108 11 98 Botswana 0.89 Dominica 0.84 65 129 Myanmar 11 Note: Cape Verde also has 11-rank change. Source: ITU. Measuring the Information Society Report 2016 29

46 Table 1.11: Use sub-index value change, 2015-2016 IDI use IDI use Rank Rank IDI use IDI use IDI use IDI use value value IDI use IDI use sub-index sub-index sub-index sub-index Economy Economy sub-index sub-index change change 2016 2016 2015 2015 2016 2016 2015-2016 2015-2016 164 32 2.22 6.53 St. Kitts and Nevis 0.62 0.30 0.32 Mozambique 4.31 0.86 0.56 0.30 Malawi 2.13 1.27 Bhutan 94 3.40 156 4.63 0.30 3.01 3.31 97 Malaysia 1.23 Ecuador 5.86 46 0.30 0.95 142 India Namibia 1.25 1.18 1.73 2.91 109 4.17 0.29 Bahamas 2.92 108 70 4.46 Algeria 1.17 1.75 Kenya 65 4.82 3.72 1.10 Dominica 0.29 1.76 2.05 123 Saudi Arabia Bolivia 111 2.72 1.65 1.08 0.29 6.03 6.32 36 Qatar Grenada 87 3.78 2.79 0.99 35 6.32 6.03 0.29 89 Saint Lucia 0.29 3.43 3.72 Botswana 0.89 2.37 3.26 98 0.82 0.29 Ethiopia 129 0.54 Myanmar 0.84 0.89 1.73 158 74 4.24 3.43 0.81 Mexico Suriname 0.28 4.20 4.48 69 0.84 0.55 0.28 Cameroon 157 Cape Verde 0.79 3.24 4.03 77 1.70 0.28 Timor-Leste 67 1.42 China 0.79 3.79 4.58 130 39 6.20 5.43 0.77 Uruguay Spain 0.28 6.65 6.93 25 6.13 5.85 0.28 Croatia 41 Belize 0.76 1.79 2.55 115 5.71 5.43 0.28 Slovenia 0.76 Jordan 2.44 3.20 101 49 Israel 136 1.47 0.73 0.75 Rwanda 0.27 5.75 6.02 42 Nicaragua Côte d'Ivoire 122 2.08 1.34 0.74 0.27 0.73 1.00 151 Samoa Maldives 72 4.30 3.59 0.71 0.26 0.97 1.23 143 Burkina Faso Gabon 124 1.92 1.23 0.69 0.26 0.63 0.90 154 Ukraine Mongolia 90 3.64 2.97 0.68 0.26 2.31 2.57 114 Sri Lanka Argentina 57 131 1.70 1.44 0.26 5.45 0.64 4.81 0.25 Kazakhstan South Africa 0.64 4.90 79 5.15 3.37 4.00 62 Kyrgyzstan 0.25 2.00 2.25 118 47 5.84 5.21 0.63 Bulgaria 4.26 4.02 0.24 Moldova 73 Montenegro 0.62 3.99 4.61 66 1.17 0.93 0.23 Zambia 119 Vanuatu 0.61 1.60 2.21 144 Poland 0.23 5.13 5.35 59 91 3.55 2.94 0.60 Jamaica 1.64 1.42 0.22 Senegal 133 Romania 0.60 4.48 5.08 63 1.35 1.15 0.20 Nepal 80 Georgia 0.60 3.40 4.00 139 Guinea 0.20 0.42 0.62 163 51 5.67 5.07 0.59 Portugal 7.94 7.74 0.20 Hong Kong, China 12 Lesotho 0.58 1.22 1.80 128 6.67 6.48 0.19 Austria 82 Tunisia 0.57 3.37 3.95 30 Angola 0.19 0.91 1.10 147 55 5.46 4.89 0.57 Cyprus 8.14 7.96 0.18 Japan 8 Cambodia 0.56 1.53 2.09 121 1.52 1.35 0.17 Syria 48 Costa Rica 0.55 5.24 5.80 135 Uganda 0.17 1.10 1.27 141 110 2.74 2.19 0.55 Iran (I.R.) 0.49 0.32 0.17 Togo 165 Brazil 0.53 5.07 5.60 52 6.85 7.38 23 26 6.85 6.68 0.17 Canada Ireland 0.53 Lithuania 38 6.25 5.73 0.52 Italy 0.17 6.23 6.40 33 Slovakia Palestine 2.08 2.25 117 34 6.38 5.86 0.52 0.17 6.82 6.66 0.16 United Arab Emirates 27 Tonga 0.52 2.07 2.59 112 Hungary Philippines 60 5.28 5.12 0.16 0.52 2.42 2.93 107 Norway Switzerland 2 8.67 8.17 0.51 0.16 8.33 8.48 4 Germany 81 3.95 3.80 0.15 Venezuela 0.51 6.98 7.49 21 2.97 2.81 0.15 Brunei Darussalam 92 Viet Nam 0.50 3.01 3.51 104 Guatemala 0.15 1.25 1.40 137 84 3.88 3.40 0.49 Albania 1.87 1.72 0.15 El Salvador 127 Belarus 0.47 5.40 5.88 44 3 8.57 8.42 0.15 Korea (Rep.) 4.21 Bosnia and Herzegovina 0.47 3.74 75 Nigeria 1.81 7.78 7.92 13 Monaco 116 2.28 0.47 0.14 Sudan Malta 0.46 126 1.87 1.73 0.14 6.28 6.75 28 0.34 0.47 166 0.45 2.95 0.13 3.40 95 Afghanistan Morocco 78 0.89 0.44 0.45 155 4.02 3.89 0.13 Antigua & Barbuda Liberia 1.12 0.13 0.99 140 145 0.85 0.44 Mauritania Yemen 1.29 Dominican Rep. 2.97 3.41 93 Estonia 0.13 7.75 7.87 14 0.44 6.55 6.43 0.12 Czech Republic 96 3.37 2.94 0.44 Seychelles 31 161 64 4.91 4.48 0.42 0.71 0.59 0.12 Djibouti Chile United States 3.36 3.78 88 18 7.57 7.45 0.42 0.12 Mauritius 6.03 6.15 40 Kuwait 0.11 Swaziland 0.42 1.19 1.61 134 enadines the and St. Gr Vincent Madagascar 0.11 0.33 168 0.44 3.89 0.41 83 3.47 Honduras 0.10 1.29 1.38 138 2.17 2.58 113 0.41 Uzbekistan Netherlands 15 7.77 7.68 0.09 Greece 56 0.41 5.05 5.46 Gambia 0.91 0.83 0.09 153 5.88 43 5.47 0.41 Barbados 0.09 Singapore 19 7.54 7.45 Lao P.D.R. 0.41 0.70 1.11 146 0.09 Benin 171 0.40 0.32 4.18 Turkey 0.40 3.77 76 0.09 132 Guyana 1.57 1.65 Indonesia 0.40 1.79 2.19 120 0.08 159 0.74 0.66 Equatorial Guinea 0.69 148 0.40 1.09 Pakistan 0.57 0.07 0.65 South Sudan 162 0.40 Trinidad & Tobago 4.53 68 4.13 0.07 Cuba 150 1.04 0.97 103 0.39 Ghana 2.64 3.03 1 Denmark 8.91 8.84 0.07 Armenia 85 3.85 3.47 0.38 4.33 71 Thailand 0.06 4.27 7.23 7.61 France 17 0.37 0.05 8.31 8.36 6 Sweden 8.03 11 0.36 New Zealand 7.66 170 Congo (Dem. Rep.) 0.04 0.36 0.41 152 Mali 0.36 0.61 0.97 1.06 149 Bangladesh 0.04 1.02 5.15 5.50 54 Serbia 0.35 0.04 174 0.14 0.10 Niger 0.35 2.78 3.14 102 Egypt Azerbaijan 5.70 50 0.04 5.66 0.06 169 0.42 0.35 Burundi 173 0.10 0.04 Chad 0.14 3.23 2.88 0.35 Fiji 100 Russian Federation 0.35 5.87 45 5.52 16 7.67 0.03 Australia 7.70 105 Paraguay 0.35 2.61 2.96 53 5.51 5.48 0.03 Lebanon 5.39 5.05 0.35 Oman 58 0.27 Tanzania 0.02 172 0.30 6.76 24 0.34 Belgium 7.10 0.01 0.44 167 0.45 Kiribati Peru 0.34 106 2.61 2.94 Guinea-Bissau 0.01 0.12 0.12 175 6.41 0.33 6.74 Andorra 29 United Kingdom 0.00 8.09 8.09 9 5 Iceland 0.33 8.11 8.44 Zimbabwe 0.00 1.91 1.91 125 6.27 37 Latvia 0.32 5.94 Luxembourg 0.00 8.05 8.05 10 3.52 3.85 86 Colombia 0.32 Solomon Islands -0.02 0.75 0.73 160 7 8.21 -0.02 Finland 8.18 61 5.17 4.85 TFYR Macedonia 0.32 22 7.48 7.54 -0.06 Bahrain 99 3.24 2.92 0.32 Panama 20 Macao, China 0.32 7.22 7.53 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 30

47 Chapter 1 th th ranks 57 in IDI in the skills sub-index and 135 The skills sub-index 2016). The biggest negative differences which impair overall IDI performance are observed for The IDI skills sub-index was amended during 2015, th Luxembourg (which ranks 70 in the skills sub- 1.1), in following consideration by EGTI (see Box th index but 11 in IDI 2016, and first in the access order to provide a better gauge of ICT-relevant sub-index) and the United Arab Emirates (which skills. Two of the proxy indicators which had th th in IDI in the skills sub-index but 38 ranks 105 previously been included in the sub-index – 2016). It is possible that some of these differences secondary and tertiary enrolment – were retained. may be ascribed to different definitions used for A new indicator – mean years of schooling – was national data gathering. introduced in place of the previous indicator (adult literacy rate). Data in the current skills sub-index are not, therefore, directly comparable with those 1.4 The IDI and the digital divide for the sub-index published in earlier annual editions of the Measuring the Information Society The term "digital divide" is used to describe . Report differences in ICT development within and between countries, regions and socio-economic Data for the skills sub-index are provided by the groupings. ITU and other UN agencies are UNESCO Institute for Statistics (UIS). Revisions committed to bridging such digital divides, in order have also been made to the IDI 2015 data, as they to ensure that everyone is able to take advantage are published in this report, to ensure that they of the benefits of the emerging information reflect the three current indicators, rather than society and that these benefits thereby contribute the set that was included in last year’s report. This to sustainable development. The UN General accounts for differences in the data for this sub- Assembly reaffirmed this commitment in its ten- index between this report and the 2015 edition. year review of outcomes of the World Summit on 19 the Information Society (WSIS) in 2015. Data used for the calculation of IDI 2016 are the latest available for the three indicators now There has been growing concern that, while the included in the skills sub-index. Because of data digital divide in basic services between developed collection schedules, data available for this sub- and developing countries has narrowed since index are identical for 2015 and 2016. It is not WSIS, as a result of the spread of mobile-cellular therefore possible to analyse changes in the skills uptake in almost all economies, digital divides sub-index between 2015 and 2016, and the sub- in the availability of broadband networks and index has not contributed to changes in the overall services, on the other hand, have been widening; IDI for this year. and that LDCs in particular may be falling further behind other countries. As a composite index, the Although it carries less weighting than the IDI provides a useful tool for comparing differences access and use sub-indices, the skills sub-index between economies, and between regions which does nevertheless have an impact on overall IDI include countries with higher and lower levels of performance insofar as some countries perform economic and ICT development. particularly well or particularly badly where these proxy indicators are concerned. The highest performing country in the skills sub-index, for The relationship between IDI and GNI p.c. example, is the United States which, despite th th for access and 18 for use, lies in ranking only 19 th One important starting point for such comparative 15 place in the overall IDI. Australia, which comes analysis is the relationship between Gross National second in the skills sub-index, likewise ranks only st th th Income per capita (GNI p.c.) and IDI performance. for access and 16 for use but 14 in the 21 overall IDI. 1.7, which plots IDI 2016 values against Chart GNI p.c. data for 2014 (the latest year for which The biggest positive differences between the data are available), shows that there is a strong skills sub-index and the overall IDI rankings, which and significant correlation between the two, effectively boost IDI 2016 performance, are th suggesting that the level of GNI p.c. (an indicator of observed for Ukraine (which ranks 11 in the skills th a country's economic performance) has a bearing in IDI 2016) and Cuba (which sub-index and 76 Measuring the Information Society Report 2016 31

48 IDI and GNI p.c., 2016 Chart 1.7: Source: ITU. on ICT development. In most cases, it is likely that and developing countries and considering the GNI p.c. levels influence both consumer demand special circumstances of LDCs. for use of ICTs and infrastructure investment in access networks to meet that demand. It is important to be clear about the composition of development categories when interpreting Outliers – countries that display significantly data that distinguish between them. The better or worse IDI performance than might be developing-economies group, as defined in UN anticipated from their GNI p.c. rankings – are data sets, includes a number of economies with worth considering further, as their experience high GNI p.c., including several economies in may point to policy and investment choices East Asia as well as oil-exporting members of the which are more or less effective in leveraging ICT Gulf Cooperation Council (GCC). Some of these access and use. The Republic of Korea, which has economies (notably the Republic of Korea, Hong the highest IDI ranking, notably outperforms its Kong (China) and Singapore) have become ICT expected position on the trend line in Chart 1.7, as champions with very high rankings in the IDI. do Belarus, Estonia and Bahrain. Countries which Five countries defined by the UN as developing noticeably underperform in comparison with their countries – Chile, Israel, the Republic of Korea, peers with similar GNI p.c. levels include Brunei Mexico and Turkey – are also member countries Darussalam, Kuwait and the United Arab Emirates. of the Organisation for Economic Cooperation and Development (OECD). The developed-country grouping, by contrast, includes relatively few countries with GNI p.c. levels that are significantly The relationship between IDI and development lower than average, and only one country (Albania) status that is in the lower half of the IDI rankings. As a result, the upward effect on the developing- Another way to assess differences between country average IDI value exerted by outliers in the economic groupings is to view IDI rankings and developing-country grouping tends to be greater values in relation to development status, in than the downward effect on the developed- particular by differentiating between developed Measuring the Information Society Report 2016 32

49 Chapter 1 IDI by development status, 2016 and 2015 Table 1.12: IDI 2015 Change in IDI 2016 average Average Average value StDev CV Min. CV Max. Range Max. Min. Range StDev value* value* 2016-2015 0.20 4.94 1.07 8.84 7.76 2.22 44.95 World 4.74 1.00 8.78 7.78 2.23 47.01 7.40 4.92 8.83 3.91 0.98 13.29 Developed 7.25 4.62 8.77 4.15 1.03 14.26 0.15 4.07 1.07 8.84 7.76 1.85 45.56 Developing 3.85 1.00 8.78 7.78 1.83 47.41 0.22 Note: *Simple averages. StDev= Standard deviation, CV= Coefficient of variation Source: ITU. Chart 1.8: IDI values by development status, country average IDI value of outliers in the 2015 and 2016 developed-country category. Data for the period 2010-2015, reported in the , Measuring the Information Society Report 2015 showed that the average IDI values of both developed and developing countries increased substantially during that period, and more or less in step with one another, leaving the digital divide between developed and developing countries largely unchanged. However, they also indicated that there was a growing gap between the majority of developing countries and LDCs. IDI values by level of development for 2015 and 1.12 and depicted 2016 are summarized in Table in Chart 1.8. The gap between the developed- and developing-country groupings remains considerable, with developed countries having an average IDI value of 7.40 against an average of 4.07 for developing countries, a difference of 3.33 points. However, developing countries have improved their position slightly relative to developed countries during the year, raising their average IDI value by 0.22 points (an increase of 5.6 per cent) as against 0.15 points (an increase of 2.1 per cent) for developed countries. Developing countries as a group registered higher rates of improvement in both the access and use sub- indices (0.15 points as against 0.08 points, and 0.39 points as against 0.30 points, respectively). 20 Least developed countries Source: ITU. When it comes to the digital divide, the situation Table 1.13 and Chart 1.9 compare the IDI of LDCs is of particular concern, given the potential performance of LDCs in the period 2015-2016 role of ICTs in facilitating sustainable development. with that of all developing countries and with the The bottom 27 countries in the IDI rankings are all global average. The overall IDI performance of LDCs, as are 36 of the bottom (LCC) quartile, while LDCs had been poorer than that of higher- and a further eight do not appear in the Index. The th middle-income developing countries over the five place out of highest ranking LDC is Bhutan, in 117 years from 2010 to 2015, and this trend continued 175 economies. in 2015-2016, with LDCs recording an average Measuring the Information Society Report 2016 33

50 IDI values for LDCs compared with global values and with all developing countries Table 1.13: IDI 2015 IDI 2016 Development status Access IDI Skills Use Access Use Skills IDI 3.54 5.74 4.74 World 5.45 5.58 4.94 5.74 3.91 6.31 8.08 Developed 7.76 7.25 8.08 6.61 7.84 7.40 3.85 4.91 2.56 4.62 Developing 4.77 2.95 4.91 4.07 LDCs 1.91 2.69 0.75 2.67 2.69 2.80 1.01 2.07 Source: ITU. IDI values for LDCs compared Chart 1.9: improvement in their IDI value of 0.16 points, as with global values and with all developing against 0.22 points for all developing countries countries (including LDCs) and 0.24 points for developing countries other than LDCs. However, the average rate of improvement in LDC values was slightly more positive, at 8.4 per cent as opposed to 5.6 per cent for all developing countries. In the access sub-index, the average improvement in values for LDCs was marginally smaller than the average improvement for all developing countries (0.13 points as against 0.15 points), while in the use sub-index the margin was more substantial (0.27 points as against 0.39 points). As Chart 1.9 indicates, LDCs fare particularly poorly on indicators in the use sub-index, where their average value is just 1.01 compared with 2.95 for all developing countries and 6.61 for developed countries. These findings suggest that LDCs as a group are not making significant headway in catching up with other developing countries; indeed, they may in fact be falling further behind them in ICT development. IDI performance quartiles and least connected countries (LCCs) Another way of looking at the relationship between countries in the IDI is to divide the Index into four quartiles, representing high, upper- middle, lower-middle and low IDI values. The group forming the lowest of these quartiles is also 1.14 shows referred to in this report as LCCs. Table the range of values for each quartile in IDI 2015 1.3 plots the quartiles on a and 2016, while Figure Source: ITU. world map. inclusion may well also have fallen into the LCC There is a close coincidence between LDCs and quartile had they done so. LCCs. As noted above, 36 of the LCCs in the lowest quartile are also LDCs. This is the large majority 1.3 shows how strongly IDI The map in Figure of the 40 LDCs reflected in the IDI, while a further performance is related to geography as well eight LDCs which have not reported data for as to development status. As noted earlier in this chapter, most of the highest-performing Measuring the Information Society Report 2016 34

51 Chapter 1 IDI values by IDI quartile, 2015 and 2016 Table 1.14: IDI 2016 IDI 2015 Number Number Group StDev CV of Average Min. Max. Range StDev Range Max. CV Average of Min. countries countries 7.57 44 7.80 44 High 0.59 1.89 6.94 8.84 7.67 6.69 8.78 2.09 0.64 8.38 1.87 5.04 6.90 5.92 42 0.59 9.90 - Upper mid dle 43 5.66 4.68 6.64 1.96 0.60 10.60 4.05 2.15 0.66 45 2.88 5.03 16.35 Lower - mid dle 43 3.78 2.74 4.68 1.94 0.61 16.24 44 2.06 1.07 2.86 1.79 0.48 23.07 Low 1.93 1.00 2.73 1.73 0.46 23.64 45 7.76 4.94 44.95 175 2.22 1.07 8.84 4.74 1.00 8.78 7.78 2.23 47.01 Total 175 Source: ITU. Note: *Simple averages. StDev= Standard deviation, CV= Coefficient of variation ographical distribution of IDI quartiles, 2016 Figure 1.3: Ge Source: ITU. economies in the IDI are member countries of 1.5 Summary and conclusion OECD, particularly in Western Europe, North America, East Asia and Oceania. Most, but not all, Measuring progress towards the information of these are categorized as developed economies. society is a complex task which entails striking Countries in the upper-middle and lower-middle a balance between different dimensions of ICT quartiles in IDI 2016 are more likely to be found in experience in different countries. The IDI pulls Eastern Europe, Latin America and the Caribbean, together 11 indicators concerned with ICT access, Central Asia and the Arabian Peninsula, with a ICT use and ICT skills into a composite index scattering in other regions including Oceania and which reflects the diversity and complexity of a few in Africa. The majority of LCCs are located that experience. Reported annually in the ITU’s in the Africa region and in South Asia, with a few, Report , the Measuring the Information Society such as Yemen and Kiribati, in other regions. Measuring the Information Society Report 2016 35

52 IDI has become an important input to building also contributed to growth in household access understanding of the spread of ICTs and their to the Internet and in the percentage of the population using the Internet. Fixed-broadband impact on economies and societies. subscriptions generally grew more slowly than mobile-broadband subscriptions, raising some Analysis of IDI 2016, and of progress during the year since IDI 2015, shows that almost all concerns about the long-term development of of the 175 economies included in the Index high-capacity networks in some regions. have continued to improve their level of ICT development. The average improvement over the The findings reported in this chapter show the continued importance of efforts to address digital year was 0.20 points. Improvements have been divides. The IDI 2016 highlights very considerable most significant among countries in the middle differences between countries and regions around of the IDI rankings, many of which are middle- income developing countries. Some developed the world, with IDI values ranging from 8.84 out of 10 in the Republic of Korea to just 1.07 in Niger. and higher-income developing economies towards It is notable that, as in previous years, while most the top of the rankings experienced little further improvement in their IDI values. Many LDCs countries have improved their IDI values, overall IDI rankings have remained relatively stable. Few towards the bottom of the distribution likewise countries moved up by more than five places in saw little improvement in their performance. The region with the lowest average IDI values, as in the IDI rankings. previous years, was Africa. While this general improvement in ICT development is to be welcomed, the relatively More progress occurred during the year, on poor performance of LDCs remains a matter of average, in the indicators which reflect ICT use than on those related to ICT access, although the concern. There is a strong correlation between LDCs and LCCs in the bottom quartile of the IDI indicators concerning ICT use started from a lower distribution. Of the 44 LCCs, 36 are also LDCs, baseline in IDI 2015. while a number of other LDCs do not appear in the The most substantial improvements within the IDI data set. As recognized in last year’s report, access sub-index, in most regions, related to this suggests that many countries in this grouping the proportion of households with access to the are locked into persistent low performance Internet. In Africa, however, the largest increase in in the Index. Given that ICT development is this sub-index was in the penetration of mobile- increasingly cited as an important enabling factor cellular subscriptions – an indicator which had for progress towards sustainable development, already reached near-saturation levels in some poor IDI performance points to the need for other regions. There was a fall in the indicator for policy interventions by governments and other fixed-telephone subscriptions in the majority of stakeholders in order to improve levels of countries, including highly-connected developed achievement on the indicators making up the IDI. countries which are experiencing fixed-mobile substitution. As well as analysing IDI findings at global and regional levels, this chapter includes examples The most significant increases within the use of the experience of countries at the top of the sub-index were in the indicator relating to distribution and of countries which have witnessed mobile-broadband subscriptions, with particularly the most dynamic improvement in their IDI values marked gains in a number of countries in the and rankings over the past year. While each Caribbean and other developing regions (see country is different, these examples – and those Chapter of other dynamic countries reported in Chapter 2 2). Factors contributing to this include investment in new infrastructure, growing uptake – suggest approaches which may be valuable in of smartphones, and reductions in prices following other contexts and are worthy of consideration increased competition or regulatory intervention. by governments, communications businesses and The increase in mobile-broadband subscriptions development agencies. Measuring the Information Society Report 2016 36

53 Endnotes 1 https:// itu. int/ en/ ITU- D/ Sta tistics/ A fuller account of the information in this section can be found in ITU (2016), at www. facts/ actsFigures2016. pdf , and on the ITU website, at http:// www. itu. int/ en/ ITU- D/ Sta tistics/ Pag es/ st at/ Documents/ ICTF . aspx default. 2 www. itu. int/ en/ ITU- D/ Sta tistics/ Pag es/ publica tions/ anapub. aspx . Previous reports can be accessed online at http:// 3 Data on the indicators included in the skills sub-index are sourced from the UNESCO Institute for Statistics (UIS). See Annex 1 for more details on the definition of the indicators. 4 2. ITU (2015), p.43 and Annex 5 itu. int/ net4/ ITU- D/ ExpertGroup/ def ault. asp . http:// www. 6 itu. int/ net4/ ITU- D/ www. expertgr ouponhouseholds/ forum/ . http:// forum/ 7 www. Telegeography, ‘South Korea to accept applications for fourth MNO licence in August 2015’. Available at: https:// com/ products/ commsupda te/ articles/ 2015/ 06/ 26/ south- kor ea- to- accept - applica tions- for - fourth- mno- telegeography. in- licence- - 2015/ . august 8 teleg eography. Telegeography, ‘SK Telecom introduces tri-band LTE-A commercially’, 2015. Available at: com/ https:// www. commsupdate/ articles/ 2015/ 01/ 02/ sk- telec om- intr oduces- tri- band- lte- a- commer cially/ products/ . 9 telec ompaper. com/ Telecompaper, ‘Siminn says 4G population coverage reaches 91%’, 2015. Available at: www. http:// siminn- say s- 4g- population- co verage- news/ 91-- 1141852 . reaches- 10 com/ teleg Telegeography, ‘Telia embarks on LTE-A deployment’, 2015. Available at: www. products/ https:// eography. articles/ 2015/ 08/ 10/ telia- commsupdate/ on- lte- a- deploymen t/ . embarks- 11 https:// teleg eography. com/ products/ Telegeography, ‘TDC starts ‘4G+’ network deployment’, 2015. Available at: www. t/ articles/ 07/ 06/ tdc- st arts- 4g- netw ork- deploymen 2015/ . commsupdate/ 12 www. teleg https:// com/ Telegeography, ‘Telenor launches ‘4G+’ network in five cities’, 2015. Available at: eography. commsupdate/ articles/ 2015/ 12/ 08/ telenor - launches- 4g- netw ork- in- five- cities/ products/ . 13 coun ec. eu/ digital- single- marke t/ en/ europa. try- inf ormation- denmark . https:// 14 Mance, H., ‘CWC plans to fire up telecoms networks with $250m investment’, 2014, in Financial Times. Available at: http:// ft. com/ cms/ s/ 0/ 61fe7820- e0e6- 11e3- a934- 00144feabdc0. html#a xzz4DhgytUZm . www. 15 databank. worldbank. http:// / dat a/ download/ GNIPC. pdf . org 16 Ericsson (2015a, 2015b and 2015c). 17 Ooredoo, 'Ooredoo Myanmar secures USD 300 million combined funding from the Asian Development Bank and com/ en/ ooredoo. media/ http:// International Finance Corporation for network rollout', 7 February 2016. Available at: view/ ooredoo- my anmar- secures- usd- news_ million- combined- funding- from- the- asian- dev elopment- bank- and- 300- finance- corpor ation- for - netw ork- rollout/ international- . 18 p.9. Galpaya (2015), 19 Int workspace. unpan. org / sites/ http:// ernet/ Documents/ UNPAN96078. pdf . 20 http:// The current list of LDCs can be found at: www. un. org / en/ dev elopment/ desa/ policy/ cdp/ ldc/ ldc_ list. pdf . Measuring the Information Society Report 2016 37

54

55 Chapter 2. The ICT Development Index (IDI) – regional and country analysis

56

57 Key findings There is a strong association between national and regional levels of ICT development, as captured by the ICT Development Index (IDI), and the level of social and economic development. While the overall regional IDI values did not shift dramatically compared to 2015, some countries made significant progress as a result of infrastructure investment and changes in policy and regulation. Europe continues to lead the way in ICT development. It had the highest average IDI value among world regions (7.35 points). Albania is the only country in Europe falling – slightly – below the global average. This reflects the region’s high levels of economic development and ICT investment. Countries in Europe generally have liberalized communication markets with high levels of ICT access, use and skills. A number of countries in the Americas significantly improved their performance in the IDI. Three island countries in the Caribbean – St. Kitts and Nevis, Dominica and Grenada – featured among the most dynamic countries, with strong improvements in their IDI value and rank. Several countries in Latin America, notably Bolivia and Mexico, also made noticeable progress in their IDI performance. Similar to other regions, the growth of mobile-broadband subscriptions played a particularly strong part in these outcomes. The Commonwealth of Independent States (CIS) is the most homogeneous region in terms of ICT development. Nearly all countries in the CIS have IDI values above the global average, and all countries in the region improved their IDI values as a result of increases in mobile-cellular and mobile-broadband penetration. The Asia and the Pacific region is, by contrast, the most heterogeneous. The region's top seven economies have IDI values above 7.50 points and rank within the highest quartile of IDI 2016. The region also includes a number of countries that significantly increased their IDI value and rank over the year, including Bhutan, Myanmar and Malaysia. However, nine out of 34 countries in the region, including several with large populations, are least connected countries (LCCs). There is great diversity in ICT development across the Arab States. The five highest performing countries in the Arab States region are oil-rich high-income economies, but the region also includes a number of low-income countries, three of which are LCCs. This illustrates that the digital divide between the LCCs and the more prosperous countries in the region may be growing. Africa is the region with the lowest IDI performance. The average IDI 2016 value for the Africa region was 2.48 points, just over half the global average of 4.94. The majority of the 39 African countries in IDI 2016 are LCCs. This reflects the lower level of economic development in the region, which inhibits ICT development. The highest growth achieved was in the number of mobile-cellular subscriptions, in contrast to other regions, in which the number of mobile- broadband subscriptions experienced the highest growth. . Investment, policy and regulation influence the performance of individual countries A number of countries rank higher than expected on the IDI relative to their level of economic development. In most regions, a number of countries also significantly increased their IDI rankings in only one year. The experiences of these dynamic countries, several of which are illustrated in this chapter, are a source of insights for other governments and businesses within their regions. Measuring the Information Society Report 2016 41

58

59 Chapter 2. The ICT Development Index (IDI) – Chapter 2. The ICT Development Index (IDI) – regional and country analysis regional and country analysis region), two in the CIS region, and three, all small region), two in the CIS region, and three, all small 2.1 Introduction Introduction 2.1 states, in Europe. states, in Europe. Chapter 1 described the ICT Development Index Chapter 1 described the ICT Development Index Table 2.1 sets out the results of IDI 2016 for each 2.1 sets out the results of IDI 2016 for each Table (IDI) and compared global findings for IDI 2015 (IDI) and compared global findings for IDI 2015 of the six ITU regions, and compares them with of the six ITU regions, and compares them with and IDI 2016. This chapter extends the analysis and IDI 2016. This chapter extends the analysis the results for IDI 2015. Chart 2.1 shows the the results for IDI 2015. Chart 2.1 shows the by investigating IDI findings at the regional level. by investigating IDI findings at the regional level. distribution of average, minimum and maximum distribution of average, minimum and maximum It also explores findings in relation to a number It also explores findings in relation to a number IDI values in these regions, compared with the IDI values in these regions, compared with the of individual countries which stand out as having of individual countries which stand out as having global average. global average. improved their position in the overall IDI rankings improved their position in the overall IDI rankings dynamically in comparison with others in their dynamically in comparison with others in their As in previous years, the Europe region records As in previous years, the Europe region records regions. regions. the highest regional average IDI value, at 7.35, the highest regional average IDI value, at 7.35, and includes only one country, Albania, just below and includes only one country, Albania, just below the global average of 4.94. The regional average the global average of 4.94. The regional average 2.2 Regional IDI analysis Regional IDI analysis 2.2 value for the CIS region, at 5.74, is significantly value for the CIS region, at 5.74, is significantly higher than the global average (although it should higher than the global average (although it should ITU Member States are divided into six regions ITU Member States are divided into six regions be noted that two lower-income countries in this be noted that two lower-income countries in this – Africa, the Americas, Arab States, Asia and the – Africa, the Americas, Arab States, Asia and the region are not included in the Index). The average region are not included in the Index). The average Pacific, Commonwealth of Independent States Pacific, Commonwealth of Independent States (CIS) for the Americas slightly exceeds the global for the Americas slightly exceeds the global (CIS) and Europe. The distribution of countries and Europe. The distribution of countries between average, at 5.13, while the average IDI values for average, at 5.13, while the average IDI values for between regions differs in a number of respects regions differs in a number of respects from the the Arab States and Asia-Pacific regions, at 4.81 the Arab States and Asia-Pacific regions, at 4.81 from the regional distributions used in other UN regional distributions used in other UN data series, and 4.58, respectively, fall somewhat below. As in and 4.58, respectively, fall somewhat below. As in data series, most notably where the Europe and most notably where the Europe and Africa regions previous years, the Africa region records by far the previous years, the Africa region records by far the Africa regions are concerned, and this should be are concerned, and this should be borne in mind lowest average IDI value, at 2.48, little more than lowest average IDI value, at 2.48, little more than borne in mind when undertaking comparative when undertaking comparative analysis with other 1 half that of the next lowest region. 1 half that of the next lowest region. analysis with other data sets. data sets. There is much greater variation in some regions There is much greater variation in some regions The IDI 2016 data published in this volume are The IDI 2016 data published in this volume are than in others. The CIS region has the smallest than in others. The CIS region has the smallest derived from 175 economies, of which 39 are in derived from 175 economies, of which 39 are in range between its highest and lowest IDI values, range between its highest and lowest IDI values, the Africa region, 34 in the Americas, 18 in the the Africa region, 34 in the Americas, 18 in the 3.27 points, reflecting its relative economic 3.27 points, reflecting its relative economic Arab States region, 34 in Asia and the Pacific, 10 Arab States region, 34 in Asia and the Pacific, 10 homogeneity. Europe also has a relatively narrow homogeneity. Europe also has a relatively narrow in the CIS region and 40 in the Europe region. Of in the CIS region and 40 in the Europe region. Of IDI range, of 3.91 points, a figure which drops to IDI range, of 3.91 points, a figure which drops to the 21 ITU Member States for which data are not the 21 ITU Member States for which data are not 3.14 if the region's two lowest-ranking countries 3.14 if the region's two lowest-ranking countries available, five are in the Africa region, one in the available, five are in the Africa region, one in the (Albania and Bosnia and Herzegovina) are (Albania and Bosnia and Herzegovina) are Americas, four in the Arab States region, six in Asia Americas, four in the Arab States region, six in Asia excluded. excluded. and the Pacific (including five from the UN Oceania and the Pacific (including five from the UN Oceania IDI by region, 2016 and 2015 Table 2.1: Table 2.1: IDI by region, 2016 and 2015 Difference 2015-2016 IDI 2015 IDI 2016 Number of Region economies CV Average* CV Range Max. Min. Range Average* StDev CV Max. Min. Range Average* StDev 8.83 4.92 3.91 7.35 0.97 13.23 8.77 4.62 4.15 7.19 1.03 14.36 -0.24 0.16 -1.14 40 Europe CIS 10 7.26 3.99 3.27 5.74 1.10 19.15 7.02 3.76 3.26 5.56 1.12 20.10 0.01 0.18 -0.94 0.25 -0.46 The Americas 34 8.17 2.73 5.44 5.13 1.39 27.09 8.06 2.64 5.42 4.89 1.35 27.55 0.01 0.18 -1.95 Arab States 4.63 1.89 40.74 -0.05 4.81 1.87 38.79 7.42 1.73 5.69 18 7.46 1.82 5.64 0.23 -3.27 34 8.84 1.73 7.11 4.58 2.19 47.87 8.78 1.62 7.16 4.35 2.23 51.14 -0.05 Asia & Pacific Africa 0.18 -0.51 39 5.55 1.07 4.47 2.48 1.14 46.06 5.27 1.00 4.27 2.30 1.07 46.57 0.20 Note: *Simple averages. StDev = Standard deviation, CV = Coefficient of variation. Note: *Simple averages. StDev = Standard deviation, CV = Coefficient of variation. Source: ITU. Source: ITU. Measuring the Information Society Report 2016 1 Measuring the Information Society Report 2016 43

60 IDI by region compared with global average, 2016 Chart 2.1: Source: ITU. The IDI distribution in the Africa region is marginally in most regions in the year between more variable, but at much lower levels which IDI 2015 and IDI 2016, the largest variations being are consistent with the region’s economic observed in the Europe and Africa regions. In development. Here again, the distribution is Europe, the 0.24-point reduction in the range affected by outliers, in this case three relatively resulted from a higher rate of improvement by the high-performing countries (Mauritius, Seychelles lowest-ranking country, Albania, in comparison and South Africa); without these, Africa’s average with countries at the top end of the distribution IDI would drop from 2.48 to 2.26 and the IDI range which are approaching the Index’s maximum value. would shrink from 4.47 points to 3.53. In Africa, the 0.20-point increase in the range resulted from faster improvements by the country The range of IDI values is greater in the Americas, with the highest ranking, Mauritius, in comparison the Arab States and, particularly, Asia and the with those at the bottom of the distribution. Pacific, reflecting the economic heterogeneity Table 2.2 illustrates the five highest- and-lowest of these regions. The Americas region includes ranking countries in each region in IDI 2016, in high-income countries in North America as well as order to provide further insight into differences in developing countries to the south. The Arab States levels of ICT development. region includes oil-rich countries belonging to the Gulf Cooperation Council but also several least The similarities and differences between regions developed countries (LDCs). The Asia-Pacific region can be explored in more detail by comparing includes a number of top performers in the Index, spider charts of the average scores achieved in such as the Republic of Korea, Singapore and Hong the different regions on each of the 11 indicators Kong (China), alongside least connected countries making up the Index. These are presented in (LCCs) in South Asia. Chart 2.2, along with a world chart to enable comparison between regional and global average There were broadly consistent improvements in values. In considering these charts, it should be the average level of the IDI across all regions in the remembered that they do not reflect the range year between IDI 2015 and IDI 2016, the greatest of values within regions, which, as noted above, is improvements taking place in the Americas much wider in some regions than in others. and Asia-Pacific regions. The range between the highest and lowest IDI values changed only Measuring the Information Society Report 2016 44

61 Chapter 2 Table 2.2: Hi ghest- and lowest-ranking countries by region, IDI 2016 Global IDI Global IDI Regional Regional IDI Country Country IDI rank rank IDI rank IDI rank Europe Arab States 2 8.83 Iceland 1 Bahrain 29 1 7.46 3 8.74 Denmark 2 United Arab 38 7.11 2 Emirates 3 Switzerland 8.68 4 45 3 Saudi Arabia 6.90 5 4 United Kingdom 8.57 4 46 6.90 Qatar Sweden 5 7 8.45 53 5 Kuwait 6.54 36 Montenegro 6.05 62 14 122 3.32 Syria 37 65 5.97 TFYR Macedonia Sudan 15 139 2.60 70 38 Turkey 5.69 16 Mauritania 2.12 151 Bosnia and 80 5.25 39 Herzegovina 155 17 2.02 Yemen Albania 4.92 91 40 161 Djibouti 18 1.82 Asia & Pacific CIS 1 8.84 Korea (Rep.) 1 1 7.26 31 Belarus 8.46 Hong Kong, China 6 2 Russian Federation 2 6.95 43 3 Japan 8.37 10 52 3 Kazakhstan 6.57 13 4 New Zealand 8.29 4 6.28 58 Azerbaijan 14 5 8.19 Australia 68 5.75 Moldova 5 30 Bangladesh 2.35 145 71 6 Armenia 5.60 31 Pakistan 2.35 146 7 5.59 72 Georgia 32 2.06 152 Kiribati 8 76 Ukraine 5.33 Solomon Islands 153 2.04 33 110 9 Uzbekistan 4.05 34 164 1.73 Afghanistan 113 10 Kyrgyzstan 3.99 The Americas Africa 15 8.17 United States 1 1 Mauritius 5.55 73 Canada 2 25 7.62 5.03 Seychelles 2 87 3 St. Kitts and Nevis 34 7.21 88 5.03 South Africa 3 Barbados 35 7.18 4 97 Cape Verde 4 4.60 Uruguay 6.79 5 47 4.17 108 Botswana 5 3.52 30 121 Guyana 35 Burundi 171 1.42 3.20 Guatemala 31 123 36 172 1.42 South Sudan 126 Honduras 32 3.09 37 Guinea-Bissau 1.38 173 131 2.88 Nicaragua 33 Chad 38 1.09 174 135 34 2.73 Cuba 39 1.07 175 Niger Source: ITU. Internet users and of households with Internet As these spider charts indicate, there has been access. Increases in households with a computer little difference in average IDI performance across were more significant in regions displaying a regions over the year between IDI 2015 and IDI higher average overall performance (Europe, 2016 on the majority of indicators in the Index. CIS and the Americas) than in those with a lower The biggest change in most regions has been average overall performance (Africa, Asia and the in the proportion of active mobile-broadband Pacific and the Arab States region), reflecting the subscriptions, followed by the proportion of Measuring the Information Society Report 2016 45

62 Chart 2.2: Av erage IDI values for each indicator, world and regions, IDI 2015-2016 Source: ITU. Measuring the Information Society Report 2016 46

63 Chapter 2 relative importance of growth in mobile-cellular states of Mauritius and Seychelles, together with subscriptions in the latter regions’ lower-income South Africa – fall into the two upper quartiles of countries. Most regions showed fairly modest the IDI distribution or exceed the global average increases in fixed-broadband subscriptions and a value in IDI 2016, while only these and two other decline in fixed-telephone subscriptions. There has countries – Cape Verde and Botswana – exceed been no change in the three skills indicators over the average value of 4.07 for developing countries. the year since, for reasons discussed in Chapter 1, the same data set has been used for both years. By contrast, 29 out of 39 African countries in the Index rank as LCCs in the lowest quartile of The smoothest distribution of results across the distribution, and the region includes all ten the range of indicators – with relatively high countries at the bottom of the global rankings. A performance across the board – is observed in the number of African LDCs are not included in the Europe region. The distribution of indicator results Index, and it is likely that at least some of these becomes less smooth as overall IDI performance would also have IDI values within the lowest declines, the most significant contributors to this quartile if data were available. These findings phenomenon being differences between regions illustrate the extent to which Africa lags behind in the proportions of fixed-telephone and fixed- other regions in ICT development, and the broadband subscriptions. The spider charts for importance of addressing the region’s ongoing the CIS and Americas regions reveal stronger digital divide. performance overall than those for the Arab States and Asia-Pacific regions, but are broadly similar in All countries in the region showed some their overall shape, reflecting this distribution of improvement in IDI value between 2015 and indicator values. 2016, although in 11 countries this improvement was marginal (less than 0.10 points). The average The spider chart for the Africa region is much less improvement recorded was 0.18 points, less smooth than those for other regions. This reflects than the average improvement of 0.22 points particularly low indicator values in Africa for fixed- for developing countries. The ten countries telephone and fixed-broadband subscriptions and at the top of the African rankings achieved an for household Internet and computer access, as average improvement in their IDI values of 0.33 well as for enrolment in tertiary education. The points, well above the global average of 0.20, strongest results in the Africa region relate to while the remaining countries in the region, all mobile-cellular subscriptions and international of which are in the LCC quartile, managed an Internet bandwidth. These variations between average improvement of just 0.14 points. Only indicators have been influenced by the prevalence four countries in the lower half of the regional of mobile over fixed terrestrial infrastructure in distribution - Rwanda, Liberia, Ethiopia and Africa, the relatively high cost of fixed-broadband Burundi – raised their IDI value by more than 0.20 connections on the continent, and the increasing points. This supports the suggestion that LDCs number of submarine cables offering international may be falling further behind other developing connectivity. countries in ICT development. The following paragraphs describe the findings for Across the Africa region as a whole, the indicators each region in more detail, and explore the results making up the IDI which showed the greatest achieved by a number of individual countries improvement were mobile-cellular penetration which have performed better than others within and mobile-broadband penetration. A particularly their regions. strong improvement in mobile-cellular penetration was recorded in Burundi, Côte d’Ivoire, Ethiopia, the Gambia, Ghana, Guinea, South Africa and Tanzania, although this indicator also fell in nine Africa countries in the region, including by a substantial margin in Namibia and Mali. Various factors The IDI values and rankings for the Africa region may account for these reductions: in Mali, are set out in Table 2.3. As noted 2.3 and Chart for example, the reduction followed a change above, the Africa region registers by far the lowest in the law requiring identification of mobile regional average IDI performance. Only three 2 Particularly strong improvements subscribers. countries in the region – the Indian Ocean island Measuring the Information Society Report 2016 47

64 Table 2.3: IDI rankings for the Africa region, 2016 and 2015 Global rank Global rank Global rank Regional rank change IDI 2015 IDI 2016 Economy 2015 2016 2016 2016-2015 Mauritius 1 5.55 73 5.27 0 73 Seychelles 2 87 5.03 85 4.77 -2 South Africa 3 5.03 86 4.70 -2 88 Cape Verde 97 4.60 99 4.23 2 4 Botswana 5 108 4.17 109 3.79 1 Ghana 6 112 3.99 111 3.75 -1 3.20 Namibia 7 3.64 121 120 1 Gabon 8 3.12 126 2.81 2 124 9 2.99 129 2.78 0 Kenya 129 10 132 2.86 Côte d'Ivoire 2.43 7 139 Zimbabwe 11 133 2.78 132 2.73 -1 Lesotho 12 134 2.76 138 2.47 4 Swaziland 13 2.73 136 2.49 0 136 Nigeria 137 2.72 137 2.48 0 14 Senegal 15 141 2.53 140 2.41 -1 Gambia 16 143 2.46 141 2.40 -2 147 17 148 2.05 1 2.22 Zambia 2.16 148 18 Cameroon 146 -2 2.07 Mali 2.14 149 2.00 0 149 19 20 150 2.13 158 1.79 8 Rwanda 21 154 2.03 152 1.95 -2 Angola 1.97 22 156 5 161 1.73 Liberia Uganda 157 1.94 155 1.86 -2 23 Benin 24 158 1.92 156 1.83 -2 Togo 25 159 1.86 159 1.78 0 -3 Equatorial Guinea 26 160 1.85 157 1.82 1.80 163 162 27 Burkina Faso 1 1.60 Mozambique 28 1.75 164 1.60 1 163 Guinea 165 1.72 166 1.57 1 29 Madagascar 30 166 1.69 165 1.57 -1 Tanzania 31 167 1.65 167 1.54 0 Malawi 32 168 1.62 168 1.49 0 Ethiopia 169 1.51 172 1.29 3 33 Congo (Dem. Rep.) 34 170 1.50 169 1.48 -1 Burundi 35 171 1.42 173 1.16 2 South Sudan 172 36 -2 1.36 170 1.42 Guinea-Bissau 173 1.34 171 -2 1.38 37 1.09 174 38 1.00 Chad 1 175 Niger 39 175 -1 1.03 1.07 174 2.30 2.48 Average Source: ITU. 2.4 shows the most dynamic countries in the in mobile-broadband penetration were recorded Table Africa region in terms of IDI ranking and value. It in Botswana, Cape Verde, Côte d’Ivoire, Gabon, shows that there is a marked difference in which Namibia and Rwanda. The most significant countries are identified as most dynamic in the increases in fixed telephony and fixed broadband region according to whether this is measured by occurred in South Africa. Measuring the Information Society Report 2016 48

65 Chapter 2 IDI values, Africa region, 2016 Chart 2.3: Source: ITU. t dynamic countries by IDI ranking and IDI value, Africa, 2015-2016 Mos Table 2.4: Change in IDI ranking Change in IDI value (absolute) IDI rank IDI value Region IDI rank Region IDI rank Country Country change change 2016 2016 rank rank 8 0.44 Côte d'Ivoire 10 150 20 Rwanda 132 120 10 Côte d'Ivoire 132 0.43 Namibia 7 7 5 5 22 156 0.38 Botswana Liberia 108 4 4 Lesotho 12 134 97 Cape Verde 0.37 Rwanda 169 33 0.34 3 150 20 Ethiopia 0.34 88 3 South Africa Source: ITU. improvements in IDI values or by progress up the lower down the African distribution, such as Côte global rankings. d’Ivoire and Rwanda, made much bigger leaps up the global rankings because they were ranked The biggest improvements in IDI values in the alongside lower-income LCCs whose average IDI region were made by Côte d’Ivoire (up 0.44 points, improvement was much smaller. This discrepancy lifting it out of the LCC quartile), Namibia (0.43 in dynamism within the Index between IDI points), Botswana (0.38 points), Cape Verde (0.37 rankings and values again illustrates the growing points) and Rwanda and South Africa (each 0.34 gap between middle-ranking countries and LCCs already discussed in Chapter 1. points). 2.4 presents spider charts showing the Chart The countries which were in the top group performance on all the IDI indicators of three of of African performers secured only marginal the region’s strong performers – Namibia, Côte improvements in their global rankings in spite d’Ivoire and Rwanda – together, for purposes of of strong performances in terms of IDI values. comparison, with the lowest-performing country Namibia and Botswana moved up only one in the IDI, Niger. place in the global rankings, and Cape Verde two places, while South Africa actually dropped These four charts bear strong similarities with down two places in the global rankings in spite one another and with those of other developing of improvement in its IDI value that was well countries in the lower half of the IDI distribution. above average. This is because they were ranked All four countries show relatively strong alongside other middle-income developing performance on two of the access indicators countries which also achieved a relatively high (mobile-cellular subscriptions and international improvement in their average IDI. Countries Measuring the Information Society Report 2016 49

66 IDI values, selected countries, Africa, IDI 2015-2016 Chart 2.4: Source: ITU. th in terms of GNI p.c., outperforms Internet bandwidth per Internet user) coupled which ranks 175 th with very weak performance on fixed-telephone , on nine of them. Niger, Rwanda, which ranks 198 3 has some of subscriptions and on the proportion of households one of the world’s poorest countries, the lowest values of any country across all of the with a computer or with Internet. The three 4 indicators in the Index. improving countries illustrated show significant improvements in Internet users and mobile- As is the case for many developing countries in broadband subscriptions. All four countries are the IDI, in both Namibia and Côte d’Ivoire it is also characterized by very low performance on the the indicator for mobile-broadband subscriptions proxy indicators in the skills sub-index. that has registered the biggest increase between IDI 2015 and IDI 2016, followed by the indicators The principal difference between the four for Internet users and households with Internet. countries illustrated is in their overall level of Côte d’Ivoire has also taken mobile-cellular performance on indicators across the Index. This subscriptions close to the highest possible value, reflects their overall standing in terms of GNI p.c. th globally in terms of GNI though this indicator fell substantially in Namibia. Namibia, which ranks 120 p.c., outperforms the other three countries on ten 2.2 provide further information Boxes 2.1 and out of the 11 indicators in the Index. Côte d’Ivoire, about developments in these countries. Measuring the Information Society Report 2016 50

67 Chapter 2 Box 2.1: ICT and IDI developments in Namibia Namibia improved its IDI score from 3.20 in 2015 to 3.64 in 2016, making it the second most dynamic country in Africa in terms of IDI value after Côte d'Ivoire. The main reason for this improvement was substantial growth in the number of mobile-broadband subscriptions, pushing up Namibia's use sub-index value from 1.73 to 2.91, the largest increase in the use sub-index of all African countries, and resulting in a 12-place jump in the global ranking for this sub-index. The growth was stimulated primarily by reductions in tariffs and packages aimed at low-income users. In 2015, Namibia’s mobile-broadband penetration stood at 62 subscriptions per 100 inhabitants, the fourth highest in Africa after Cape Verde, Botswana and Ghana. Namibia’s overall IDI score would have improved further had it not been for a drop in its access sub-index value caused by a 9 per cent reduction in the number of mobile subscriptions, the result of a decrease in the use of multiple SIM cards following a merger between two operators. Namibia’s score on the access sub-index fell from 4.35 to 4.25. Namibia is one of the frontrunners in Africa in ICT development, as the government has encouraged modernization of the country’s telecom network. In 2014, Telecom Namibia started to construct a fibre-based network to connect the central government to the administrative capitals of all 14 regions in the country. The project aims to support government efforts towards 5 decentralization and make effective e-government available to the wider public. Mobile Telecommunications (MTC), the largest operator in Namibia, was also one of the first 6 7 and 2012 , operators in Africa to launch both commercial 3G and 4G networks (in 2006 respectively). In April 2016, MTC, in cooperation with Huawei, also announced the first commercial use of an LTE-Advanced (LTE-A) network in Africa, making Namibia the first country in 8 Africa to reach speeds of 1 Gbit/s. Box 2.2: ICT and IDI developments in Côte d’Ivoire Côte d'Ivoire achieved an improvement in its IDI value of 0.44 points between 2015 and 2016. It also moved up six places in the global IDI rankings. Both Internet uptake and mobile-cellular subscriptions improved during the year. The percentage per cent in 2014 to 17 of households with Internet access grew from 12 per cent in 2015. The percentage of individuals using the Internet also rose from 15 per cent to 21 per cent. This latter increase may be attributed to the significant increase in active mobile-broadband subscriptions, which climbed from 25 per 100 inhabitants in 2014 to 40 per 100 inhabitants in 2015. Continuous 2G and 3G network expansion help to explain the steady growth in mobile-cellular subscriptions as well as the recent growth in mobile-broadband subscriptions. In 2014, the operator MTN expanded its network with 252 2G and 105 co-located 3G sites, and in 2015 the company invested ZAR 833 million to roll out a further 132 2G and 339 co-located 3G sites (MTN, 2014 and 2015). 3G coverage in Côte d'Ivoire rose from 43.6 per cent of the population in 2014 to 71.0 per cent in 2015. Mobile money may have contributed to the increase in the number of mobile-cellular subscribers in the country. The subscriber base for MTN’s Mobile Money service grew from 1.4 million in million in 2015. 2013 to 2.5 million in 2014 and then to 2.9 Measuring the Information Society Report 2016 51

68 Box 2.2: ICT and IDI developments in Côte d’Ivoire (continued) Moov (Maroc Telecom) has also launched an international mobile money transfer service called Flooz (Maroc Telecom, 2015). Prices for calls and SMS were reduced following a decision by the regulator ARTCI in January 2015, which is also likely to have improved the affordability and uptake of mobile-cellular subscriptions (ARTCI, 2015). rankings, these countries all have IDI values above Arab States 6.50. A further three countries in the region (Oman, Lebanon and Jordan) have IDI values above IDI values and rankings for the Arab States region the global average of 4.94. are set out in Table 2.5 and in Chart 2.5, where they are compared with the global average and The region also includes a number of low-income with averages for developed and developing countries with lower IDI values. Four countries in countries. the region – Yemen, which has experienced civil conflict, and three countries located on the African There are marked variations in IDI rankings continent (Sudan, Mauritania and Djibouti) – are and values within the region. The five highest- in the bottom (LCC) quartile. With the exception performing countries in the region (Bahrain, of Mauritania, these countries have fallen further United Arab Emirates, Saudi Arabia, Qatar and behind middle-ranking economies in the region Kuwait) are oil-rich high-income economies which during the course of the year, adding to concerns are members of the Gulf Cooperation Council about a growing digital divide between more and (GCC). While only two of these (Bahrain and United less prosperous nations. Arab Emirates) are in the top quartile of the IDI IDI ranking and values, Arab States region, 2016 and 2015 Table 2.5: Regional Global Global Global rank rank IDI 2016 change rank rank IDI 2015 Economy 2016 2016-2015 2016 2015 1 7.46 28 7.42 -1 Bahrain 29 Arab Emirates 2 38 7.11 35 6.96 United -3 Saudi Arabia 45 6.90 38 6.88 -7 3 4 43 6.90 Qatar 6.78 -3 46 Kuwait 53 6.54 48 6.45 -5 5 6.04 6.27 Oman 6 59 -1 58 Lebanon 7 5.93 61 5.91 -5 66 8 89 5.06 Jordan 4.67 4 85 Tunisia 95 4.83 95 4.49 0 9 Morocco 10 96 4.60 98 4.26 2 Egypt 100 4.44 97 4.26 -3 11 Algeria 12 103 4.40 112 3.74 9 Palestine 13 106 4.28 103 4.12 -3 Syria -2 14 122 3.32 120 3.21 Sudan 15 2.60 134 2.56 -5 139 Mauritania 151 2.12 154 1.90 3 16 Yemen 17 155 2.02 151 1.96 -4 Djibouti 161 1.82 160 1.73 -1 18 4.81 4.63 Average Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 52

69 Chapter 2 Chart 2.5: IDI values, Arab States region, 2016 2014) of the Plenipotentiary (Rev. Busan, 99 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution Conference. Source: ITU. The most substantial average improvement in some improvement, all but six saw their IDI values the region on any individual indicator was in rise by less than the world average and only four mobile-broadband penetration, where there out of the 18 regional economies included in the were particularly strong performances in Algeria, IDI improved their position in the global rankings. Jordan, Morocco, Saudi Arabia and Tunisia. Growth Four countries (Saudi Arabia, Kuwait, Lebanon and in mobile-cellular penetration was much stronger Sudan) dropped by five or more places overall. in Jordan than in any other country in the region, followed by Bahrain and Kuwait. All countries The most dynamic improvements in values in in the region also improved their performance the region between IDI 2015 and IDI 2016 were on the proportion of households with Internet observed in middle-income countries – Algeria access (where there were particularly strong (which improved its overall IDI value by 0.66 points performances in Morocco, Algeria and Mauritania) and rose nine places in the global rankings), Jordan and on the proportion of Internet users (where (up 0.38 points), Morocco (up 0.35 points) and Algeria was, by some distance, the strongest Tunisia (up 0.34 points). Among lower-performing performer). Saudi Arabia registered a substantial countries, Mauritania was the biggest gainer, reduction in households with a computer. improving its overall IDI value by 0.22 points, just above the world average gain of 0.20 points. Five of the higher-performing countries in this 2.6 sets out the most dynamic countries Table region (Bahrain, United Arab Emirates, Saudi in the region in terms of both IDI values and Arabia, Oman and Lebanon) were among the most 2.6 presents spider charts for four rankings. Chart dynamic economies worldwide in terms of IDI countries in the region – Bahrain (the region’s top values and rankings in the period between 2010 performer), Algeria and Jordan (its most dynamic and 2015 (ITU, 2015). Between 2015 and 2016, performers in terms of both rankings and values) however, while all countries in the region saw t dynamic countries by IDI ranking and IDI value, Arab States, 2015-2016 Mos Table 2.6: Change in IDI value (absolute) Change in IDI ranking IDI value IDI rank IDI rank IDI rank Country Country Region rank Region rank change change 2016 2016 12 103 9 Algeria 0.66 103 Algeria 12 85 8 0.38 Jordan 85 4 Jordan 8 3 151 16 Mauritania 96 10 Morocco 0.35 0.34 96 Morocco 2 10 95 9 Tunisia 0.24 59 6 Oman Source: ITU. Measuring the Information Society Report 2016 53

70 DI values, selected countries, Arab States region, 2015-2016 I Chart 2.6: Source: ITU. The differences are equally marked between the and Mauritania (the only LCC in the region to middle-income countries Algeria and Jordan, on 2.3 includes improve its global ranking). Box the one hand, and low-income Mauritania, on further information concerning Jordan. the other. Algeria and Jordan have a significantly These charts illustrate the different patterns of higher penetration of mobile-cellular subscriptions ICT development within the region. The wealthiest than Mauritania, while the latter has very low country among the four depicted, Bahrain, has scores across the use sub-index, on households the highest IDI ranking and value. Its performance with a computer, and on all three proxy indicators exceeds that of the other three countries most in the skills sub-index. The indicators that markedly in the use sub-index, where it achieved contributed most strongly to the rise in IDI values a score of 7.48 in 2016 compared with 3.20 for in both Algeria and Jordan were those concerned Jordan, 2.92 for Algeria and 1.29 for Mauritania. with households with Internet, Internet users and Bahrain performs particularly strongly on mobile-broadband subscriptions. The increase household access to a computer and to the achieved by Mauritania was mostly due to Internet and on the proportion of Internet users. increases – from a much lower base – in Internet Bahrain had also recorded the second highest users and mobile-broadband subscriptions, fixed-telephone penetration among the Arab along with some increase in mobile-cellular States (after United Arab Emirates) and a much subscriptions. higher level of mobile-broadband subscriptions by the time of IDI 2015 than the other three countries. Measuring the Information Society Report 2016 54

71 Chapter 2 Box 2.3: ICT and IDI developments in Jordan Jordan improved its IDI value from 4.67 in IDI 2015 to 5.06 in IDI 2016. This included a strong performance in the use sub-index. Its percentage of households with Internet had been growing Box more rapidly than in other countries in the Arab States region since 2006 (see Chart 2.3) and 76 per cent, between 2014 and 2015. The percentage of per cent to increased further, from 69 per cent, between 2014 and 2015. per cent to 53 Internet users also rose strongly, from 46 Chart Box 2.3: Ho useholds with Internet, 2006-2015 Source: ITU. This increased access to the Internet was associated with strong growth in mobile-broadband per cent in 2014 and then to penetration, which rose from just 0.1 per cent in 2010 to 19.1 per cent by end 2015. The growth in mobile-broadband penetration can be attributed partly 35.6 to lower prices and new promotions, and to the launch of LTE services in 2015. Prices for mobile per broadband appear to have been stable between 2012 and 2014, but then dropped by 42 per cent for the cent for the ITU-defined prepaid handset-based sub-basket in 2015 and by 35 postpaid USB/dongle-based sub-basket. Following the investment of USD 270 million (with more than USD 100 million meant for network 10 9 these two operators launched commercial and USD roll-out) by Zain, 351.52 million by Orange, LTE services in 2015. The country’s third operator, Umniah, followed suit in the first quarter of 2016. Both Orange and Umniah have also invested in upgrading their existing 2G and 3G networks (Batelco, 2016). Zain has reported an increase in daily data volume from 76 TB in 2014 to 275 TB in 2015 (Zain, 2015 and 2016). other. IDI values and rankings for this region are Asia and the Pacific 2.7 and Chart set out in Table 2.7. Asia and the Pacific is the most diverse region in The top ten positions in the regional rankings for terms of ICT development, reflecting the marked 2016 are almost identical to those for 2015, with differences in levels of economic development China just displacing Thailand in tenth position. between OECD member countries and other high-income economies in East Asia and Oceania, The top seven economies in the region – the on the one hand, and a number of low-income Republic of Korea, which is the global top countries in the region, including LDCs, on the performer, Hong Kong (China), Japan, New Measuring the Information Society Report 2016 55

72 Table 2.7: IDI rankings and values, Asia and Pacific region, 2016 and 2015 Global rank Global rank Regional rank Global rank IDI 2015 change IDI 2016 Economy 2015 2016 2016 2016-2015 Korea (Rep.) 1 1 8.84 1 8.78 0 Hong Kong, China 2 6 8.46 7 8.40 1 1 8.28 Japan 3 10 8.37 11 New Zealand 8.29 13 4 16 3 8.05 Australia 8.19 12 8.18 -2 14 5 6 20 7.95 19 7.88 -1 Singapore Macao, China 7 28 7.58 26 7.47 -2 Malaysia 8 61 6.22 66 5.64 5 5.25 Brunei Darussalam 9 77 5.33 74 -3 China 4.80 84 5.19 81 3 10 -3 5.05 79 5.18 82 11 Thailand 12 Maldives 4.68 2 88 5.04 86 4.66 13 89 4.99 90 1 Iran (I.R.) 3 14 90 4.95 93 4.54 Mongolia 102 4.41 15 Fiji 102 0 4.16 16 105 4.29 104 4.02 -1 Viet Nam 107 Philippines 17 106 4.28 3.97 -1 Tonga 18 114 3.93 114 3.63 0 3.63 3.86 0 Indonesia 19 115 115 0 3.56 116 20 Sri Lanka 3.77 116 5 Bhutan 21 117 3.74 3.12 122 Cambodia 127 2 2.78 3.12 125 22 3.08 Vanuatu 23 127 131 2.73 4 -3 3.05 2.92 125 Timor-Leste 24 128 Samoa 25 -2 2.78 128 2.95 130 India -3 2.50 135 26 138 2.69 2.54 1.95 13 153 140 27 Myanmar 2.32 142 2.50 142 28 Nepal 0 Lao P.D.R. 0 2.21 2.45 144 29 144 2.27 Bangladesh 30 145 2.35 -2 143 -1 145 Pakistan 31 146 2.35 2.15 2.07 -5 Kiribati 152 147 2.06 32 -3 Solomon Islands 33 153 2.04 150 1.99 Afghanistan 34 164 1.73 162 1.62 -2 4.35 4.58 Average Source: ITU. economies, only one – New Zealand – improved its Zealand, Australia, Singapore and Macao (China) IDI value by more than the world average. – all have IDI values above 7.5 and sit in the high quartile of the IDI rankings. They are all There is a significant gap in IDI values and rankings high-income economies which have maintained between these seven economies and others in high IDI performance throughout the period the region. A further five countries – Malaysia, since the Index was first published. The average Brunei Darussalam, China, Thailand and Maldives improvement in IDI value for these economies – rank in the top half of the IDI, while two more during the year was just below 0.1 point, reflecting – the Islamic Republic of Iran and Mongolia – their position near the top of the Index, where occupy places just below half-way. Significantly there is limited scope for further improvement greater improvements in IDI values, however, as the IDI is currently constituted. Of these Measuring the Information Society Report 2016 56

73 Chapter 2 IDI values, Asia and Pacific region, 2016 Chart 2.7: Source: ITU. were achieved by a number of middle- and These findings for the Asia-Pacific region reveal lower-ranking countries than by those at the top greater improvement in IDI values among middle- of the regional rankings. The most substantial ranking countries than among countries in the improvement was made by Bhutan (up 0.62 top and bottom quartiles, suggesting that the points), followed by Myanmar (up 0.59), Malaysia region may be witnessing a reduction in the digital (up 0.58), Mongolia (up 0.41) and China (up 0.39). divide between developed countries and most developing countries alongside a worsening divide One of these countries, Myanmar, is among between the majority of developing countries and the nine countries in the region which fall the least connected. within the lowest (LCC) quartile in the rankings. These regional LCCs also include three of the Table 2.8 identifies the most dynamic countries most populous countries in the region – India, in the Asia-Pacific region in terms of IDI rankings Bangladesh and Pakistan. With the exception of and values. In both cases, these were Bhutan, Myanmar, countries from the region within the Malaysia and Myanmar. Chart 2.8 contains spider LCC quartile averaged an improvement of only charts showing the performance of these three 0.13 points in overall IDI value, with one country, countries, together with that of the region’s top Kiribati, recording a marginal fall. performer (and the global top performer), the Republic of Korea. As in other regions, the highest average improvement on any individual indicator in this These spider charts illustrate differences between region was for mobile-broadband penetration. The countries that have performed strongly in each most substantial improvements on this indicator of the four quartiles of the IDI distribution: the were recorded by Malaysia, Bhutan, New Zealand, Republic of Korea in the high quartile, Malaysia in Mongolia and Vanuatu, while only one country, the upper-middle quartile, Bhutan in the lower- Thailand, recorded a fall. Thailand registered a middle quartile, and Myanmar in the low (LCC) significant fall on the indicator for mobile-cellular quartile. penetration, perhaps because of the exclusion of inactive SIMs, while also recording much the The Republic of Korea has long achieved high greatest improvement in the region on households IDI values across the board, though there is still with Internet access. Substantial falls in the mobile- room for improvement in international Internet cellular indicator were also registered by Viet Nam bandwidth per Internet user, the proportion of and the People’s Democratic Republic of Laos. households with a computer and the penetration Myanmar recorded the strongest improvements of fixed-broadband subscriptions. Like many on the indicators for mobile-cellular penetration developed countries, the Republic of Korea has and for the proportion of Internet users. There very high penetrations of fixed-telephone and was, overall, a small decline in the indicator for mobile-cellular subscriptions. fixed-telephone penetration across the region. Measuring the Information Society Report 2016 57

74 t dynamic countries by IDI ranking and IDI value, Asia and Pacific, 2015-2016 Mos Table 2.8: Change in IDI ranking Change in IDI value (absolute) IDI value IDI rank IDI rank IDI rank Country Region rank Region rank Country change change 2016 2016 Bhutan 21 0.62 117 13 Myanmar 27 140 Myanmar 0.59 27 61 8 Malaysia 140 5 61 0.58 117 21 Bhutan 5 Malaysia 8 Vanuatu 23 0.41 Mongolia 14 90 127 4 10 81 81 10 China 3 0.39 China 90 3 Mongolia 14 3 4 13 New Zealand Source: ITU. Chart 2.8: IDI values, selected countries, Asia and Pacific region, 2015-2016 Source: ITU. and for households with a computer and with The shape of the chart for Malaysia is typical of Internet, than do lower-income developing those for middle-income developing countries. countries such as Bhutan and Myanmar, although Malaysia records a high value for mobile-cellular its value for fixed-broadband subscriptions is subscriptions but a much lower value for fixed- low compared with developed countries. Its telephone subscriptions than found in developed values in the skills sub-index fall between those countries and high-income developing countries for developed and developing countries. As in such as the Republic of Korea. It achieves many other countries, the biggest improvement substantially higher values in the use sub-index, Measuring the Information Society Report 2016 58

75 Chapter 2 Box 2.4: ICT and IDI developments in Malaysia Malaysia is one of the most dynamic countries in the Asia-Pacific region in IDI 2016, climbing five places in the global rankings on the strength of an improvement of 0.58 points in its IDI value during 2015. Malaysia has seen a significant improvement in Internet uptake. The proportion of the population using the Internet increased from 63.7 per cent in 2014 to 71.1 per cent at end 2015. The proportion of households with Internet access also increased from 64.1 per cent to 70.1 per cent over the same period. Malaysia achieved considerable growth in mobile-broadband penetration, from 58.3 per cent in 2014 to 89.9 per cent at end 2015, although it also registered a slight decrease in the penetration of fixed broadband, perhaps indeed associated with this improvement in mobile broadband. Data volumes have been increasing steadily. Maxis says that total data volume on its TB by end 2015. 000 TB in the first quarter of 2014 to 50 network increased from about 20 000 Subscribers have also been using more data on a monthly basis. According to Maxis, average data usage per prepaid subscriber on its network increased from 476 MB/month in the first quarter of 11 478 MB/month by end 2015. 2014 to 1 Such a substantial growth in mobile-broadband subscriptions can be attributed to three main developments in the local market – new data plans and promotions; upgrades and expansion in network coverage; and a high level of smartphone ownership in the country. Greater competition in the Malaysian telecom market has encouraged operators to introduce new data plans and promotions. Digi, for example, launched unlimited Facebook access and/or unlimited music streaming in 2014 (Digi, 2016). Celcom has offered a new SIM card, Xpax Magic SIM, which 12 free calls and SMS and many other benefits (Axiata, 2016). Maxis provides free basic Internet, has offered unlimited streaming on Spotify Premium, and cheaper data plans, as well as a programme ‘zerolution’ which enabled people to buy smartphones by instalments (Maxis, 2016). U Mobile has partnered with Berjaya Credit to offer a similar smartphone financing programme called U MicroCredit, and has also promoted a cheaper data plan including 1 GB of high-speed 13 Internet free every month. Strong competition has encouraged operators to invest in network deployment and improvement. Population coverage by LTE networks increased from 33 per cent in 2014 to 71 per cent in 2015 and continues to expand in 2016. In order to prioritize LTE roll-out into areas of high demand, Digi encouraged citizens to nominate priority areas for LTE deployment (Digi, 2016). The high level of smartphone ownership in Malaysia can be linked to specialized postpaid data plans which allow instalment payments for smartphones. According to Ericsson (Ericsson, 2015c), Malaysia ranks highly in smartphone subscription penetration within the region, behind Singapore and Australia but ahead of Thailand, the Philippines and Indonesia. Celcom reported 14 14 reported growth in smartphone users of almost 20 per cent in 2015 (Axiata, 2016), while Digi an increase in smartphone users from 26 per cent in 2012 to 59.2 per cent in 2015 (Digi, 2016). Maxis experienced an increase in smartphone penetration among prepaid subscribers from 38 per cent at the beginning of 2014 to 67 per cent by end 2015. The shapes of the charts for Bhutan and Myanmar in Malaysia’s IDI performance in 2015-2016 has are characteristic of those for low-income been in the penetration of mobile-broadband developing countries. Both show lower values subscriptions, supported by improvements in the of mobile-cellular subscriptions and very low proportions of Internet users and of households values for fixed-telephone and fixed-broadband with access to a computer and to Internet. Measuring the Information Society Report 2016 59

76 subscriptions. The two countries also display much significant contributing factor has been rapid lower levels of household access to computers growth in mobile-cellular subscriptions, following and the Internet than do developed and middle- the relatively recent introduction of mobile- income developing countries, as well as lower cellular networks to the country. This has also levels of performance in the skills sub-index. facilitated the growth of mobile-broadband subscriptions in the country. Improvements in the proportions of mobile- broadband subscriptions (especially), Internet Boxes 1.2 and 1.6 in Chapter 1 include further users and households with Internet and with information concerning ICT and IDI developments a computer have contributed to the overall in the Republic of Korea and in Myanmar. Boxes performance improvement of both countries. 2.4 and 2.5 below provide further information In the case of Myanmar, however, the most concerning Malaysia and Bhutan. ICT and IDI developments in Bhutan Box 2.5: Bhutan was the most dynamic country in terms of IDI value in the Asia-Pacific region during the year between IDI 2015 and IDI 2016. The most significant change in Bhutan occurred in the active mobile-broadband subscriptions per 100 inhabitants, which increased from 28.2 to 56.4 over the year, continuing a trend of rapid growth since 2012 (Box Chart 2.5). There was also an increase in the proportion of individuals using the Internet, from 4.5 per cent in 2006 to 39.8 per cent in 2015 (Box Chart 2.5), while the proportion of households with Internet rose from 24.0 per cent in 2014 to 31.7 per cent in 2015. Chart Box 2.5: Bh utan – Active mobile-broadband subscriptions and Internet users (%) Source: ITU. The growth in Internet usage, especially through mobile broadband, can be explained by the expansion of 3G coverage in the country during 2015, alongside network upgrades to LTE. TashiCell’s 3G network covered all 20 administrative districts as of July 2015, as compared with 15 14 in March 2015 and eight in 2014. Overall, population coverage by 3G networks climbed to 80 per cent by end 2015, with 40 per cent coverage by LTE. The roll-out of LTE network and expansion of 3G network enabled data users to move from EDGE to 3G and LTE in 2015 (BICMA, 2016). Steadily falling prices for mobile broadband over the last four years are also likely to have boosted growth in mobile-broadband subscriptions. The price of a postpaid mobile broadband subscription (USB/dongle) for one month stood at USD 4.81 in 2015, as against USD 5.6 in 2012. The prepaid handset-based mobile-broadband subscription had fallen to USD1.66 in 2015, compared with USD 3 in 2012. Measuring the Information Society Report 2016 60

77 Chapter 2 section does not therefore include an analysis of Commonwealth of Independent States (CIS) individual dynamic countries in the CIS region. The CIS region is economically relatively All countries in the region nevertheless heterogeneous. Ten countries within the region experienced improvements in their IDI value supply data for the IDI, the exceptions being during the year 2015-2016, all but one of them Tajikistan and Turkmenistan. Four countries in the (Azerbaijan) in the range between 0.1 and 0.3 region (Belarus, Moldova, the Russian Federation points, and with a regional average of 0.18 points, and Ukraine) are categorized as developed which is just below the global average. The countries, while the remainder are categorized as sharpest fall in global ranking was experienced developing countries. by Kyrgyzstan, in spite of its improving its IDI value by 0.14 points, because better performance IDI values and rankings for the CIS region are set improvements were made by other middle-income out in Table 2.9 and Chart 2.9. Two countries in the countries with similar IDI values. region (Belarus and the Russian Federation) rank in the high quartile of IDI 2016, while the remainder The strongest average improvements on individual of those reporting data appear in the two middle indicators in this region were for mobile-cellular quartiles of the rankings. All but two of these have penetration and mobile-broadband penetration, IDI levels above the world average. However, only and for the proportions of Internet users and one, Belarus, improved its position in the global households with access to the Internet. The rankings in the year under review. All others either regional averages for mobile-cellular penetration retained their global ranking or saw it decline. This and for mobile-broadband penetration were Table 2.9: IDI ranking and values, CIS region, 2016 and 2015 Global rank change Global rank Global rank Regional rank IDI 2015 IDI 2016 Economy 2016-2015 2015 2016 2016 1 31 7.26 33 7.02 2 Belarus 6.79 42 Russian Federation 2 43 6.95 -1 Kazakhstan 3 6.57 52 6.42 0 52 4 6.28 55 6.23 -3 Azerbaijan 58 5 68 5.75 Moldova 5.60 -1 67 Armenia 6 71 5.60 71 5.34 0 Georgia 7 72 5.59 72 5.33 0 Ukraine 8 5.33 76 5.21 0 76 Uzbekistan 110 4.05 110 3.76 0 9 Kyrgyzstan 10 113 3.99 108 3.85 -5 5.74 5.56 Average Note: Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. Source: ITU. Chart 2.9: IDI values, CIS region, 2016 Note: Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. Source: ITU. Measuring the Information Society Report 2016 61

78 IDI values, selected countries, CIS region, 2015-2016 Chart 2.10: Source: ITU. boosted by much higher improvements reported of differences in GNI p.c. The most significant for one individual country than for others, improvements in Uzbekistan during the last year namely Kazakhstan in the case of mobile- were in households with a computer and with 16 cellular penetration, and Georgia in the case Internet, and in the percentage of Internet users in of mobile-broadband penetration. The greatest the population. improvements on the indicators for households with access to Internet and for Internet users were made in Uzbekistan. All but one country in the Europe region (Belarus) saw a reduction in the indicator for fixed telephony. Europe is the region which boasts the highest average IDI value, at 7.35, just below the 18 Chart 2.10 presents spider charts for two countries No fewer developed-country average of 7.40. from the CIS region – Kazakhstan towards the than 29 of the region’s 40 countries are among top of the regional distribution, and Uzbekistan the 44 countries in the high quartile of the IDI towards the bottom. The relative position of these rankings, while only one country, Albania, is nd th and 110 , two countries in the IDI rankings – 52 outside the top half of the distribution. Albania is respectively – reflects their position in terms of also the only country in the Europe region with an nd nd 17 ). and 162 GDI p.c. (82 IDI value below the global average (4.92, as against a global average of 4.94). These charts illustrate the similarities and differences between middle- and lower-income IDI values and rankings for the Europe region are developing countries in this region. Kazakhstan set out in Table 2.10 and Chart 2.11. They show displays much higher performance levels than that, while all countries in the region are high- or Uzbekistan across the range of indicators, with relatively high-performing, there are also some the exception of secondary enrolment which, geographical differences in the distribution. throughout the region, is at or close to 100 Positions at the top of the regional rankings are per cent, and mean years of schooling. It has mostly occupied by countries in Northern and a much higher penetration of mobile-cellular, Western Europe, while those towards the bottom fixed-telephone, fixed-broadband and mobile- are mostly held by countries in Southern and broadband subscriptions, as well as Internet Eastern Europe. As in previous years, the five users and households with a computer or with Nordic countries – Denmark, Finland, Iceland, Internet, even though Uzbekistan has considerably Norway and Sweden – rank particularly highly, all improved its values on the latter three indicators within the top 20 worldwide. The lowest 16 places over the year. If Uzbekistan’s very low level of in the regional rankings are occupied by countries tertiary enrolment were excluded, the shape on the Mediterranean and in Eastern Europe. The of the two spider charts would be very similar, lowest five places are occupied by countries which the variation between them largely a reflection are not members of the European Union. Measuring the Information Society Report 2016 62

79 Chapter 2 Table 2.10: IDI ranking and values, Europe region, 2016 and 2015 Global rank change Global rank Regional rank Global rank IDI 2015 IDI 2016 Economy 2016-2015 2015 2016 2016 1 2 8.83 3 8.66 Iceland 1 Denmark 2 3 8.74 2 8.77 -1 Switzerland 3 4 8.68 5 8.50 1 United Kingdom 4 8.57 4 8.54 -1 5 Sweden 7 8.45 6 8.47 -1 5 Netherlands 6 8 8.43 8 8.36 0 Norway 7 9 8.42 9 8.35 0 8 Luxembourg 8.36 11 10 8.34 -1 Germany 9 8.31 13 8.13 1 12 10 8.11 17 7.95 1 France 16 11 17 8.08 Finland 8.11 -3 14 Estonia 12 18 8.07 18 7.95 0 Monaco 13 19 7.96 20 7.86 1 Ireland 14 7.92 21 7.73 0 21 Belgium 22 7.83 22 7.69 0 15 Austria 16 23 7.69 24 7.53 1 Malta 17 24 7.69 25 7.49 1 Spain 1 7.46 18 26 7.62 27 29 7.61 19 Andorra 27 2 7.39 Israel 20 30 7.25 0 30 7.40 21 7.25 31 7.20 -1 Czech Republic 32 22 33 7.23 32 7.10 -1 Slovenia 23 36 7.13 40 6.86 4 Greece Italy 24 37 7.11 36 6.89 -1 Lithuania 39 7.10 34 7.00 -5 25 Latvia 26 40 7.08 37 6.88 -3 Croatia 27 41 7.04 41 6.83 0 2 Slovakia 28 42 6.96 44 6.69 6.64 1 6.94 Portugal 29 44 45 Hungary 30 6.72 46 6.60 -2 48 Bulgaria 49 6.69 50 6.43 1 31 Poland 32 50 6.65 47 6.56 -3 Serbia 33 51 6.58 51 6.43 0 6.53 Cyprus 34 54 6.28 53 -1 Romania 35 6.26 60 5.92 0 60 Montenegro 62 6.05 64 5.76 2 36 TFYR Macedonia 37 65 62 5.82 -3 5.97 Turkey 38 70 5.69 69 5.45 -1 Bosnia and 0 5.03 5.25 39 80 80 Herzegovina 4.62 1 Albania 4.92 91 92 40 7.35 7.19 Average Source: ITU. distribution was much higher (0.21 points) than The average increase in IDI values in the region that in the upper half (0.11 points). This reflects in the year between 2015 and 2016 was 0.16 the fact that countries towards the top of the points, below the world average. However, the distribution are pushing against the ceiling of the average increase in the lower half of the European Measuring the Information Society Report 2016 63

80 IDI values, Europe region, 2016 Chart 2.11: Source: ITU. Index, which does not necessarily capture some by Romania (0.34 points), Portugal (0.30 points), of the developments in ICT access and usage and Montenegro and Albania (each 0.29 points). which have been taking place in highly developed In the upper half of the distribution, the most economies, such as the introduction of very high substantial gains were made by Switzerland (near fixed-broadband speeds, the predominance of the top of the distribution, 0.18 points) and Ireland smartphones, and the widespread use of cloud (0.19 points). Unusually for this Index, IDI values for computing driving higher data volumes. three countries, all in Scandinavia, fell marginally during the year (see above). The strongest improvement in performance over the year in this as in other regions was Chart 2.12 presents spider charts for three of the registered on the indicator for mobile-broadband countries which achieved higher than average penetration. Only three countries – Luxembourg, gains – Switzerland, Romania and Albania – in Estonia and the United Kingdom – reported a order to illustrate similarities and differences fall in this indicator, while the most substantial within the region. It also includes one of the improvements were recorded by Bulgaria, countries, Finland, whose overall IDI value fell Romania and Ireland. Estonia also reported a during the year. Analysis of the spider charts for substantial decline in mobile-cellular penetration, the region’s two highest-performing countries (and but this was due to a change in definition. All but the second and third global performers), Iceland six countries in the region saw a reduction in the and Denmark, can be found in Boxes 1.3 and 1.4 in indicator for fixed telephony. Chapter 1. Further information about Romania and Albania can be found in Boxes 2.6 and 2.7 below. Table 2.11 sets out the most dynamic countries in the region in terms of IDI rankings and values. The These charts differ from those in other regions in greatest gains in value for individual countries were a number of respects, on account of the higher made in the lower half of the regional distribution, overall IDI ranking of European countries. t dynamic countries by IDI ranking and IDI value, Europe region, 2015-2016 Mos Table 2.11: Change in IDI value (absolute) Change in IDI ranking IDI value IDI rank Region IDI rank IDI rank Region Country Country change change rank rank 2016 2016 4 0.34 Greece Romania 36 35 60 23 44 2 Andorra 19 27 29 Portugal 0.30 Slovakia 28 42 2 91 40 Albania 0.29 62 36 Montenegro 0.29 62 36 Montenegro 2 Greece 36 23 0.27 Source: ITU. Measuring the Information Society Report 2016 64

81 Chapter 2 Box 2.6: ICT and IDI developments in Romania Romania improved its IDI value by 0.34 points, from 5.92 in IDI 2015 to 6.26 in IDI 2016. The strongest improvements in the country’s performance were observed in Internet access and use. The percentage of households with a computer has grown steadily from 26.0 per cent in 2006 to 68.7 per cent by end 2015. The percentage of households with Internet access increased from 14.3 per cent to 67.7 per cent over the same period. A similar steady growth has taken place in the percentage of individuals using the Internet, which stood at 55.8 per cent at end 2015. Fixed-telephone subscriptions per 100 inhabitants decreased from 21.1 in 2014 to 19.8 at end 2015, in line with a widespread general trend. There was a slight increase in fixed-broadband subscriptions per 100 inhabitants, from 18.6 to 19.7 by end 2015. The most significant improvement in penetration rates was seen in mobile-broadband, which increased from 49.3 in 2014 to 63.5 per 100 inhabitants at end 2015. The growth in active mobile-broadband subscriptions can be explained by a combination of network upgrades, network sharing, the launch of LTE and VoLTE, promotions and lower prices. At the beginning of 2015, DiGiMobil (RCS & RDS) invested in its 3G network in order to increase 19 download speed (to up to 21.6 Mbit/s). Later in 2015, it launched LTE, the last operator in the 20 21 country to do so. In September 2015, Orange launched VoLTE services in the country. During 2013, Orange and Vodafone signed a network sharing agreement in order to improve 2G and 3G coverage, in respect of which 70 per cent of the resulting programme of work was completed by end 2015 (Orange, 2015 and 2016). Telekom Romania has also been investing in infrastructure to improve network coverage (OTE, 2015 and 2016). In 2015, overall network population coverage stood at 99.9 per cent for 3G and 72 per cent for LTE. Owing to the high levels of competition, operators have been offering a variety of promotions which may have boosted the proportion of active mobile-broadband subscribers. Since 2014, for example, Orange has included Internet access and international calls in all its contracts (Orange, 2016). Telecom Romania has opted for bundles of mobile, fixed and TV subscriptions (OTE, 2015). The prices for mobile broadband (handset-based, prepaid) have also decreased from USD 13 in 2012 to just USD 5.55 in 2015. The price of the ITU’s mobile-cellular sub-basket dropped from USD 24.52 in 2008 to USD 6.65 in 2015. The falling price for the mobile-cellular sub-basket can be attributed to a decision taken by the regulator ANCOM in 2014 (Decision No. 366/2014) which lowered the termination call rates for mobile from 3.07 eurocents/min to 0.96 eurocents/min. mobile-broadband subscriptions in the region, High-income developed countries in Europe reflecting its significantly lower GNI p.c. and the – with some exceptions (see below) – tend to 22 late launch of 3G in the country. have very high penetration levels for both fixed- telephone and mobile-cellular subscriptions, and both fixed- and mobile-broadband subscriptions. However, the charts for Romania and Albania show They also tend to display high values for other that the most significant improvements in values indicators in the access and use sub-indices. for both countries between 2015 and 2016 were Many middle-income countries in Eastern Europe, in mobile-broadband subscriptions and in the such as Romania, have a broadly similar shape to proportion of households with Internet, with a their spider charts, but lower values across the smaller but significant increase in Romania in the board, including lower values for fixed-broadband proportion of households with a computer. subscriptions. Albania has the lowest figure for Measuring the Information Society Report 2016 65

82 IDI values, selected countries, Europe, 2015-2016 Chart 2.12: Source: ITU. Higher rates of improvement were achieved by a The Americas number of South American developing countries and Caribbean countries. In South America, The Americas region, like the Asia-Pacific region, significant gains were made by both relatively high is highly diverse, including two high-income performers – Uruguay (up 0.35 points), Argentina developed countries in North America, large (up 0.31 points), and Costa Rica and Brazil (up middle-income developing countries in South 0.27 points) – and relatively low performers America, and smaller developing countries and – Mexico (up 0.42 points) and Bolivia (up 0.53 small island states in Central America and the points). However, some other countries made only Caribbean. marginal gains on the previous year. IDI values and rankings for the region are set out The greatest gain in the region was made by in Table 2.12 and Chart 2.13. At the top of the the Caribbean island state of St. Kitts and Nevis, rankings are the region’s two large developed which rose 20 places in the global rankings on countries, the United States and Canada, which the strength of an increase of 0.98 points in its rank within the top 25 countries worldwide. Like IDI value. This improvement on the IDI appears to developed countries in Europe and the Asia-Pacific have resulted from infrastructure improvements region, the improvement in their IDI level over and marketing initiatives by communications the year was below the global average. Two of operators (see Box 1.5 in Chapter 1). Substantial the region’s small island countries – St. Kitts and improvements in IDI value were also recorded Nevis and Barbados – are also in the high quartile by five other Caribbean countries – Dominica (up of the rankings, while only one country in the 0.57 points), Grenada (up 0.46 points), Belize (up region, Cuba, ranks in the lowest (LCC) quartile. 0.33 points), Barbados (up 0.31 points) and the The region’s only LDC, Haiti, does not provide data Dominican Republic (up 0.28 points). for the IDI. Measuring the Information Society Report 2016 66

83 Chapter 2 ICT and IDI developments in Albania Box 2.7: st Albania ranks 91 in IDI 2016, and has improved its IDI value from 4.62 in IDI 2015 to 4.92 in IDI 2016. The most significant progress in the country has been made in Internet uptake and in the growth of households with a computer, the latter having risen from just 4.9 per cent in 2006 to 25.7 per cent in 2015. The fixed-broadband penetration rate for Albania increased slightly from 6.5 per cent in 2014 to 7.6 per cent in 2015. The price of the ITU-defined fixed-broadband sub-basket fell from USD 29.79 per month in 2008 to USD 9.52 in 2015. The mobile-cellular sub-basket decreased in price from USD 32.13 in 2008 to just USD 6.35 in 2015. Mobile-broadband prices also declined between 2012 and 2015. The increase in mobile-broadband penetration, from 30.9 per cent in 2014 to 40.6 per cent in 2015, was influenced by the commercial launch of LTE and LTE-A. In 2015, the regulator AKEP 23 amended Law No. 9918 to allow spectrum refarming for LTE services. Telekom Albania (OTE) 24 was the first operator in Albania to launch LTE in July 2015 and LTE-A in September 2015, 25 followed by ALBtelecom and Vodafone which both launched LTE in September 2015. As a result of fast network roll-out and upgrade, data traffic has been approximately doubling year on year in Albania, as reported by Telekom Albania (Chart Box 2.7). owth in data traffic in Albania Chart Box 2.7: Gr Source: OTE (2016). threshold and so prompted mobile operators to The Americas region shared the general trend 26 purge their subscriber base of inactive SIM cards . in the IDI whereby the greatest improvements occurred in the indicator for mobile-broadband Table 2.13 sets out the most dynamic countries in penetration. Gains on this indicator were driven the Americas region in terms of IDI rankings and by large increases in a number of Caribbean values. In both cases, the most dynamic countries countries (St. Kitts and Nevis, Grenada, Belize and are three island states in the Eastern Caribbean – Jamaica) and four countries on the South American mainland (Bolivia, Uruguay, Argentina and Brazil). Kitts and Nevis, Dominica and Grenada – which St. As in other regions, there were also significant share a common communications regulatory improvements in average performance on the authority. The most dynamic country on the proportion of Internet users and the proportion American mainland, also in both cases, is Bolivia. of households with Internet. Ecuador registered Chart 2.14 contains spider charts showing the a sizeable reduction in the indicator for mobile- performance of these countries. Developments cellular penetration, apparently because of a new Kitts and Nevis are discussed in Box 1.5 in in St. telecommunication law passed in 2015 which 1. Box 2.8 below describes the ICT and IDI Chapter imposed a charge on each active line above a given environment in Bolivia. Measuring the Information Society Report 2016 67

84 IDI ranking and values, Americas region, 2016 and 2015 Table 2.12: Global rank Global rank Global rank Regional rank IDI 2015 change IDI 2016 Economy 2015 2016 2016 2016-2015 United States 15 8.17 15 8.06 0 1 -2 2 25 7.62 23 7.55 Canada 20 6.23 St. Kitts and Nevis 3 34 7.21 54 39 35 4 Barbados 7.18 4 6.87 2 5 47 6.79 49 6.44 Uruguay 1 Argentina 6 55 6.52 56 6.21 Chile 7 1 6.11 57 56 6.35 59 Costa Rica 8 57 6.30 6.03 2 2 65 5.72 Brazil 9 63 5.99 5.80 Bahamas -1 5.98 64 10 63 1 5.48 68 5.76 67 11 Trinidad & Tobago 12 Dominica 5.14 77 8 5.71 69 8 13 74 5.43 82 4.97 Grenada Antigua & Barbuda 14 75 5.38 70 5.41 -5 St. Vincent and the 5.07 78 5.32 15 78 0 Grenadines 16 79 5.27 75 5.22 -4 Venezuela Colombia 17 83 5.16 81 4.98 -2 Suriname 18 84 5.09 83 4.89 -1 4.87 4 4.45 Mexico 19 92 96 -2 4.63 91 Panama 4.87 93 20 -7 St. Lucia 21 94 87 4.68 4.85 4.56 -4 4.54 94 98 22 Ecuador 4.52 Jamaica 23 99 101 4.20 2 -1 4.42 4.23 100 Peru 24 101 Dominican Rep. 25 1 4.02 105 4.30 104 Paraguay -2 3.88 107 26 109 4.08 4.02 3.49 6 117 111 27 Bolivia 3.64 113 3.73 118 28 El Salvador -5 Belize 0 3.32 3.66 119 29 119 3.44 Guyana 30 121 3.52 -3 118 123 0 Guatemala 31 123 3.20 3.09 -2 Honduras 3.09 124 126 3.00 32 -1 Nicaragua 33 131 2.88 130 2.74 Cuba 34 135 2.73 133 2.64 -2 4.89 5.13 Average Source: ITU. middle-ranking developing countries, with low The spider charts highlight a number of similarities (in the case of Bolivia, very low) levels of fixed and differences between dynamic countries telephony and fixed broadband, relatively strong within the region. The chart for St. Kitts and Nevis values for international Internet bandwidth per has the more rounded shape associated with Internet user, and moderate values for other higher-income higher-ranked countries in the IDI, access indicators. with a relatively high value for fixed-telephone subscriptions, and stronger performance in the The principal gains made by all of these more use sub-index than lower-income lower-ranked dynamic countries in the Americas between IDI countries. The charts for Dominica and Bolivia 2015 and IDI 2016 were in mobile-broadband are more representative of those associated with Measuring the Information Society Report 2016 68

85 Chapter 2 Chart 2.13: IDI values, Americas region, 2016 Source: ITU. Table 2.13: Mos t dynamic countries by IDI ranking and IDI value, Americas region, 2015-2016 Change in IDI ranking Change in IDI value (absolute) IDI value IDI rank IDI rank IDI rank Country Region rank Country Region rank change change 2016 2016 3 St. Kitts and Nevis 0.98 20 St. Kitts and Nevis 3 34 34 69 12 Dominica 0.57 69 12 Dominica 8 74 111 27 Bolivia 0.53 13 Grenada 8 111 Bolivia 0.46 6 Grenada 13 74 27 35 4 Barbados 4 92 19 Mexico 0.42 92 4 19 Mexico Source: ITU. subscriptions, Internet users and households with By far the largest gain for any indicator in any Internet. This is consistent with findings in other country in this region within IDI 2016, however, dynamic developing countries. However, the two was observed on mobile-broadband penetration island states in the Eastern Caribbean, whose in St. Kitts and Nevis, which rose from 18.63 geography is more conducive to fixed-network per cent to 71.02 per cent over the year. Strong deployment than that of a large country with performance on this indicator was also a feature difficult terrain such as Bolivia, also registered of the improvement in IDI values and rankings significant gains in fixed-broadband subscriptions. achieved by Dominica and Grenada. Some of the factors which contributed to this are described in Box 1.5 in Chapter 1. Measuring the Information Society Report 2016 69

86 Box 2.8: ICT and IDI developments in Bolivia Bolivia moved up six places in the IDI ranking, thanks to an increase of 0.53 points in its IDI value, more than twice the world average IDI improvement. This Andean country significantly improved its performance in terms of the percentage of households with a computer and households with Internet, which rose to an estimated 33.1 per cent and 23.8 per cent, respectively, by end 2015. The increase in household ICT access was spurred by growth in mobile-broadband subscriptions, which almost tripled from 12.2 subscriptions per 100 inhabitants in 2014 to 33.8 at end 2015. This large increase in active mobile-broadband subscriptions resulted from the migration of 2G mobile data customers to 3G networks. The growth of household Internet access and mobile- broadband uptake also contributed to an increase in the proportion of Internet users from 34.6 per cent in 2014 to an estimated 45.1 per cent by end 2015. Higher Internet uptake was reflected in an increase in data consumed: mobile Internet data traffic doubled between 2014 and 2015 (ATT, 2015), and the traffic exchanged in the national Internet exchange point also grew 27 significantly. Although more people are coming online in Bolivia, there remains a large urban-rural divide. Most Internet users and households with Internet access are located in urban areas, while less than 10 per cent of the population in rural areas use the Internet (Box 2.8). In addition, 3G coverage remains limited to 27 per cent of the population, suggesting that most rural areas do not have access to mobile-broadband networks. Chart Box 2.8: Ho useholds with Internet, households with a computer and Internet users, Bolivia, 2014 Source: ITU. The extension of broadband networks to rural areas and the promotion of Internet uptake among rural communities is therefore one of the most important challenges that ICT policy- makers face in Bolivia. This has been a focus of policy attention, and several initiatives of the Programa Nacional de Telecomunicaciones de Inclusión Social have tried to bridge the gap, for example by providing Internet access to rural schools or deploying mobile-broadband-capable 28 base stations in rural areas. Measuring the Information Society Report 2016 70

87 Chapter 2 IDI values, selected countries, Americas region, 2015-2016 Chart 2.14: Source: ITU. changes in policy, infrastructure deployment and Summary and conclusion 2.3 other factors. The IDI continues to demonstrate the diversity of Each economy within the IDI faces different ICT environments within the world community, challenges, related to its geography, infrastructure from economies with high levels of ICT requirements and social and economic performance to least connected countries. structure, as well as the resources available to it. Policy interventions aimed at improving As indicated in Chapter 1, there is a broad the ICT environment need to be tailored to association between levels of economic those particular characteristics. Nevertheless, performance (as represented by GNI p.c.) and governments and ICT businesses can draw on the levels of performance in the IDI. This is reflected in experience of more dynamic countries in the IDI variations between and within different geographic when developing their plans for improving national regions. Regions which are relatively prosperous, ICT environments. The policies, infrastructure and such as Europe, have higher average IDI values service deployment approaches implemented in a than those which are less prosperous, such as number of these dynamic countries are illustrated Africa. Regions which are economically more in this chapter. Understanding how and why they diverse, such as Asia-Pacific and the Americas, have achieved higher rates of ICT development can show the greatest variations in IDI performance. help policy-makers and businesses elsewhere as Within each region, some countries have improved they pursue better ICT performance which can, in their IDI values more than others, as a result of turn, contribute towards sustainable economic and social development within their countries. Measuring the Information Society Report 2016 71

88 Endnotes 1 The countries included in each regional grouping of the ITU’s Telecommunication Development Bureau are listed at http:// itu. int/ en/ ITU- D/ Sta tistics/ Pag es/ definitions/ regions. aspx . Palestine is not an ITU Member State; the status of www. 99 Palestine in ITU is the subject of Resolution Busan, 2014) of the ITU Plenipotentiary Conference. (Rev. 2 amrtp- mali. org / index. php? option= com_ con tent& view= www. article& id= 136: lancement - de- la- campagne- de- See http:// l- identific ation- des- abonnes- aux- services- de- la- telephonie- et - de- l- int ernet& ca tid= 87: actualite ; and communication- sur- http:// ech. net/ 2016/ 01/ le- mali- identifie- les- abonnes- du- telephone- mobile . www. africt th 3 213 of 217 countries in the GNI p.c. rankings 4 databank. worldbank. org / dat a/ download/ GNIPC. pdf . GNI p.c. figures from http:// 5 Telegeography, ‘Telecom Namibia nears completion of second-phase government fibre project’ in Telegeography https:// www. teleg eography. com/ commsupda te/ articles/ 2015/ CommsUpdate, 2 February 2015. Available at: products/ 02/ telecom- namibia- nears- comple tion- of- second- phase- go vernment- fibre- 02/ index. html . project/ 6 5.8 million 3.5G network’, 18 https:// www. Telegeography, ‘MTC Namibia launches USD December 2006. Available at: 8- commsupda te/ articles/ 2006/ 12/ 18/ mt c- namibia- com/ usd5- products/ million- 3- 5g- netw ork/ . telegeography. launches- 7 Telegeography, ‘LTE arrives in Namibia: MTC launches 4G in capital’, 18 May 2012. Available at: https:// www. com/ products/ commsupda te/ articles/ 2012/ 05/ 18/ lte- arrives- in- namibia- mt c- launches- 4g- in- capit al/ . telegeography. 8 March 2015. Available at: Telegeography, ‘MTC Namibia, Huawei treble LTE speeds’ in Telegeography CommsUpdate, 15 mt www. eography. com/ products/ commsupda te/ articles/ 2016/ 04/ 15/ teleg c- namibia- huaw ei- treble- lte- speeds/ . https:// 9 Februar y 2015. Available at: Telegeography, ‘Zain launches Jordan’s first LTE network’ in Telegeography CommsUpdate, 17 www. teleg eography. com/ products/ commsupda te/ articles/ 2015/ 02/ 17/ zain- launches- jordans- firs t- lte- netw ork/ . https:// 10 Telegeography, ‘Orange inks deal with Huawei for June 3G launch’ in Telegeography CommsUpdate, 26 March 2016. Available at: www. teleg eography. com/ products/ commsupda te/ articles/ 2015/ 03/ 26/ orang e- inks- deal- with- https:// huawei- for- june- 4g- launch/ . 11 maxis. list edcompany. com/ misc/ Maxis, ‘Results 1Q 2015’ (Quarterly Presentation), 27 April 2015. Available at: http:// 1Q15_ presen tation_ slide. pdf ; Maxis, ‘Financial Results: 4Q 2015 & FY2015’ (Quarterly Presentation), Presentation/ v3_ maxis. list edcompany. com/ misc/ Pr esentation/ 4Q15_ presen tation_ slide. pdf ; Maxis, 4 February 2016. Available at: http:// maxis. list edcompany. ‘Financial Results: First Quarter 2016’ (Quarterly Presentation), 21 April 2016. Available at: http:// 1Q16_ misc/ tation_ slide. Presentation/ pdf . com/ presen 12 According to Celcom: 'All New Magic SIMTM from Xpax comes with free basic internet that are subject to a monthly Fair www. xpax. com. my/ magicsim_ Usage Policy of 500 MB with speed of up to 64kbps ("Free Basic Internet").' See: https:// tnc. 13 U Mobile, ‘U Mobile wows Malaysian consumers with market-best plans for the amazing iPhone 6 and iPhone 6 Plus, www. u. com. my/ http:// RM98 to take home iPhone 6 with U MicroCredit’ (Public Release), 19 March 2015. Available at: 6205 ; U Mobile, ‘Free 1GB high speed Internet for life with U Mobile’s brand new power prepaid pack’ release/ press- www. u. com. press- release/ 6585 . http:// my/ (Public Release), 23 July 2015. Available at: 14 digi. list edcompany. com/ Digi, ‘4th Quarter: 2014 Results’ (Quarterly Presentation), 9 February 2015. Available at: http:// presentation/ presen tation4Q14. pdf . misc/ 15 TashiCell, ‘3G coverage’ (Operator’s Official Facebook post), 31 March 2015. Available at: www. facebook. com/ https:// ; TashiCell, ‘3G coverage’ (Operator’s Official Facebook post), 3 July 2015. Available at: https:// www. facebook. TashiCell/ TashiCell/ ; Ministry of Information and Communications, Royal Government of Bhutan (MOIC, 2014), Annual com/ http:// www. moic. go v. bt/ wp- InfoComm and Transport Statistical Bulletin (5th Edition), March 2014. Available at: con tent/ uploads/ 2016/ 05/ 2014. pdf . 16 Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. 17 databank. worldbank. org / dat a/ download/ GNIPC. pdf . http:// Atlas methodology: 18 The Europe region includes three countries classified as developing countries – Cyprus, Israel and Turkey. 19 Telegeography. ‘RCS & RDS promises faster 3G speeds in 25 cities’ in Telegeography CommsUpdate, 10 March 2015. www. teleg eography. com/ products/ Available at: commsupda te/ articles/ 2015/ 03/ 10/ rcsr ds- promises- fas ter- 3g- https:// 25- cities/ in- speeds- . 20 Telegeography, ‘Digi entering 4G data segment via 2K Telekom licence takeover’ in Telegeography CommsUpdate, 21 August 2015. Available at: www. teleg eography. com/ products/ https:// te/ articles/ 2015/ 08/ 21/ digi- ent ering- commsupda 4g- data- segment - via- 2k- telec om- licence- tak eover/ . 21 Telegeography, ‘Orange Romania launches VoLTE; Wi-Fi calling to follow’ in Telegeography CommsUpdate, 14 September https:// www. teleg eography. com/ products/ commsupda te/ articles/ 2015/ 09/ 14/ 2015. Available at: orang e- romania- launches- volte- wi- fi- calling- to- follo w/ . Measuring the Information Society Report 2016 72

89 22 Albania was the last country in Europe to launch 3G services: the first 3G licence was awarded to Vodafone in December 2010, and the operator started offering 3G services in January 2011. For more information, see Cullen International (2012) and Vodafone’s press release: www. voda fone. al/ voda fone/ Voda fone_ i_ pari_ qe_ sjell_ http:// 3G_ teknologjine_ me_ Tiranen_ 799_ filluar_ php . duke_ 1. 23 Telegeography, ‘AKEP opens door for 4G launches via spectrum refarming’ in Telegeography CommsUpdate, 5 June 2015. Available at: www. teleg eography. com/ https:// commsupda te/ articles/ 2015/ 06/ 05/ akep- opens- door- for - 4g- products/ launches- via- spectrum- re farming/ . 24 Telegeography, ‘Telekom Albania switches on LTE-A network; offers download speeds of 225 Mbps’ in Telegeography https:// www. teleg eography. com/ products/ commsupda te/ articles/ 2015/ CommsUpdate, 2 September 2015. Available at: 09/ 02/ telekom- albania- swit ches- on- lte- a- netw ork- offer s- download- speeds- of- 225mbps/ . 25 http:// Vodafone Albania, ‘Vodafone Albania sjell internet super të shpejtë me 4G+’ (Public Release), 2015. Available at: vodafone. al/ voda fone/ Voda fone_ Albania_ sjell_ int ernet_ super_ te_ www. shpejte_ me_ 4G_ 4286_ 1. php . 26 See Article 34 of the Ley Orgánica de Telecomunicaciones approved in February 2015, available at: http:// www. telecomunicaciones. gob. ec/ wp- con tent/ uploads/ downloads/ 2016/ 05/ Ley - Org%C3%A1nic a- de- Telec omunicaciones. pdf . 27 en/ http:// www. pit. bo/ index. php/ Punto de Intercambio de Tráfico Bolivia, tra fico . 28 For more information on the initiatives undertaken under Bolivia’s Programa Nacional de Telecomunicaciones de Inclusión Social , see: https:// www. oopp. gob. bo/ vmt el/ index. php/ inf ormacion_ institucional/ con tent,1160. html . Measuring the Information Society Report 2016 73

90

91 Chapter 3. The role of ICTs in monitoring the SDGs

92

93 Key findings In 2015, the United Nations identified 17 Sustainable Development Goals (SDGs) and associated targets, which will guide international development between 2015 and 2030. To measure progress towards achievement of the SDGs, the United Nations Statistical Commission adopted a global framework of indicators. Several SDGs refer to ICTs and technology, and several ICT indicators were identified to help track SDGs 4, 5, 9 and 17. Monitoring access to computers and the Internet in schools . SDG 4 is concerned with inclusive and equitable educational opportunities for all. One of its targets is to ensure provision of appropriate and inclusive educational facilities. Available data on schools with access to computers and the Internet suggest that, while a number of developing countries have achieved 100 per cent access to computers in both primary and secondary schools, many other countries lag behind. . Another SDG 4 target is to enhance the skills needed for Monitoring ICT skills among youth and adults employment, decent jobs and entrepreneurship. This will be measured by the proportion of young people and adults with a range of ICT skills. Data show that the share of the population with specific ICT skills is considerably higher in developed countries than it is in developing countries. Monitoring the role of ICTs in women’s empowerment. SDG 5 is concerned with women’s empowerment. One of its targets is to enhance the use of ICTs to promote empowerment. Data on the gender gap in mobile the percentage of women and men who own a mobile phone show that . phone ownership and use is higher in lower-income and less connected countries Monitoring the growth of access to ICTs and the Internet. SDG 9 calls for increased access to ICTs, working towards "universal and affordable access to the Internet in least developed countries by 2020". One of its targets focuses on the need to increase access to ICTs and the Internet, as measured The proportion of by the percentage of the population covered by different mobile technologies. the population covered by a mobile-broadband network will reach 84 per cent in 2016 globally, but only 67 per cent in rural areas . Just over half of the global population is covered by LTE or higher networks and few of those living in rural areas. Monitoring the contribution of ICTs to science, technology and innovation . SDG 17 is concerned with revitalizing the global partnership for sustainable development. One of its targets is to improve cooperation in science, technology and innovation. This will be measured, in part, by monitoring the number and speed of fixed-broadband subscriptions. Data show that there are substantial differences between developed and developing countries, and within regions, in terms of both the proportion of the population with fixed-broadband subscriptions and the speeds delivered by these subscriptions. While some countries, such as the Republic of Korea, Denmark and France, have fixed-broadband penetration rates of around 40 per cent and almost exclusively high-speed connections of above 10 Mbps, many low-income economies have less than 2 per cent fixed-broadband penetration rates and exclusively lower-speed connections of below 2 Mbps. Monitoring the use of ICTs as an enabling technology. Another target under SDG 17 is to enhance society's use of technology, including ICTs. This is measured by the proportion of individuals using In 2016, Internet usage rates are about twice as high in developed countries as in the Internet. developing countries, and more than twice as high in developing countries, as a whole, than as in least developed countries. Measuring the Information Society Report 2016 77

94

95 Chapter 3. The role of ICTs in monitoring the SDGs the indicators (including two sub-indicators) 3.1 Introduction are explicitly concerned with ICTs. This chapter outlines each of these indicators, considers the 2030 Agenda for In September 2015, the availability of relevant data, and takes stock of Sustainable Development was agreed upon at the 1 current levels of achievement. United Nations Sustainable Development Summit. The Agenda sets out a comprehensive framework for international cooperation between 2015 and 3.2 ICTs and SDGs 2030 in support of sustainable development, covering its economic, social and environmental The role of ICTs in development has been under dimensions through 17 Sustainable Development discussion since at least 1984, when the report Goals (SDGs) and 169 targets. The SDGs, which of the ITU-led Maitland Commission advocated succeed the Millennium Development Goals that international cooperation to reduce inequalities in guided international development policy between access to communications. It formed a centrepiece 2000 and 2015, are summarized in Table 3.1. They of the World Summit on the Information Society are applicable to all countries and regions, and (WSIS), held in 2003 and 2005, which called for are intended to ensure that ‘no one is left behind’ ‘a people-centred, inclusive and development- in the course of progress towards sustainable oriented Information Society, ... enabling development. individuals, communities and peoples to achieve their full potential in promoting their sustainable In March 2016, the United Nations Statistical development and improving their quality of life.’ Commission (UNSC) agreed on a global indicator framework, including 230 indicators, to help There is now extensive experience in ICTs for monitor progress, identify challenges and guide development (ICT4D), which exploits the potential policy-makers in their efforts to implement of ICTs to achieve particular development goals the Goals and Targets (ECOSOC, 2016). Six of Table 3.1: Th e Sustainable Development Goals 1 End poverty in all its forms everywhere 2 End hunger, achieve food security and improved nutrition and promote sustainable agriculture 3 Ensure healthy lives and promote well-being for all at all ages 4 Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all 5 Achieve gender equality and empower all women and girls Ensure availability and sustainable management of water and sanitation for all 6 7 Ensure access to affordable, reliable, sustainable and modern energy for all Promote sustained, inclusive and sustainable economic growth, full and productive employment and decent work 8 for all 9 Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation 10 Reduce inequality within and among countries 11 Make cities and human settlements inclusive, safe, resilient and sustainable 12 Ensure sustainable consumption and production patterns Take urgent action to combat climate change and its impacts 13 14 Conserve and sustainably use the oceans, seas and marine resources for sustainable development Protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat 15 desertification, halt and reverse land degradation and halt biodiversity loss Promote peaceful and inclusive societies for sustainable development, provide access to justice for all and build 16 effective, accountable and inclusive institutions at all levels Strengthen the means of implementation and revitalize the global partnership for sustainable development 17 Source: UN. Measuring the Information Society Report 2016 79

96 while facilitating access to ICTs and their use by interconnectedness has great potential to individuals and communities. The decade since accelerate human progress, to bridge the digital 5 The WSIS has also seen remarkable developments in divide and to develop knowledge societies.’ Summit did not, however, adopt an SDG concerned the capabilities and reach of ICTs within developing specifically with ICTs, and only one of the 169 countries, including the spread of broadband 2030 Agenda targets in the networks and the emergence of mobile Internet, is explicitly concerned with their availability. Three other targets refer smartphones and tablets, social media and cloud computing, all of which are widely recognized as to the potential of ICTs for achieving other Goals, 2 and indicators adopted by UNSC for these four enablers of sustainable development. targets are discussed in this chapter. Several At the same time, there has been increasing other Goals and targets refer more generally to concern about digital divides between developed the importance of technology and to the role and developing countries, particularly LDCs, and of information in enabling the achievement of between different sections of communities within sustainable development. individual countries; about the extent to which these enablers have facilitated development in Concern has been expressed by a number of practice; and about the impact of the digital divide stakeholders about the need to develop a more on other development divides. Data concerning comprehensive framework for assessing the the overall extent of digital divides can be found role of ICTs within the SDGs, including stronger in the opening section of Chapter 1 and in other measurement of the adoption and use of ICTs 3 ITU publications. themselves. An expert group meeting organized by World Development In its 2016 the UN Department of Economic and Social Affairs , the World Bank Report on Digital Dividends in 2015, for example, highlighted three ways (2016) noted that ‘the effect of technology on in which ICTs are likely to have a major impact global productivity, expansion of opportunity for on sustainable development as they become the poor and the middle class, and the spread of more pervasive and more sophisticated over accountable governance has so far been less than the period of SDG implementation: by changing expected,’ and that, to date, ‘the better educated, the underlying characteristics of economic and well connected, and more capable have received most of the bene ts. fi ’ It emphasized the need for social development, supporting the delivery of specific Goals and targets, and facilitating more policy-makers to focus on analogue complements 6 effective measurement of all Goals and targets. to digital development, in particular ‘a favorable WSIS Action Line facilitators have developed a business climate, strong human capital, and good matrix juxtaposing WSIS Action Lines and SDGs, governance.’ 7 Big-data which will inform SDG implementation. analysis is expected to play a substantial role in the The importance of ICTs in sustainable development 8 Agenda measurement of some of the Goals in the . was recognized by the UN General Assembly following its overall review of the implementation of WSIS outcomes which concluded in December 3.3 The SDG indicator framework 2015. The outcome document from that review called for ‘close alignment’ between the WSIS In March 2016, UNSC adopted a global indicator , highlighting the 2030 Agenda process and the framework for the SDGs and Targets which ‘crosscutting contribution of information and 2030 Agenda for Sustainable are set out in the communications technology’ to the SDGs and 9 . Development This framework, which includes 230 poverty eradication,’ and called on governments individual indicators, was developed by the UN and international organizations to integrate ICTs in Inter-Agency and Expert Group on SDG Indicators their implementation of sustainable development. (IAEG-SDGs) in the course of a year-long process The review expressed concern about continued of consultation with international agencies and digital divides which could slow progress towards Member States. It will provide the basis for annual sustainable development, with particular reference 4 monitoring, follow-up and subsequent review to the gender digital divide. of implementation of the at the 2030 Agenda national, regional and global levels. The IAEG-SDGs 2030 Agenda for Sustainable Development The is continuing its work to establish baselines for also recognized that ‘The spread of information and communication technology and global Measuring the Information Society Report 2016 80

97 Chapter 3 indicators, develop and review methodologies and Indicators for Target 4.a: address data gaps. Proportion of schools with access to computers, and Proportion of schools with access to the Substantial data challenges are raised by the Internet for pedagogical purposes indicator framework. Global monitoring will, to the extent possible, be based on comparable SDG Goal 4 aims to ‘ensure inclusive and equitable official statistics produced by national statistical quality education and promote lifelong learning systems (NSS). Where feasible, and in line with the opportunities for all.’ It includes ten targets, Agenda ’s pledge that ‘no-one will be left behind,’ two of which will be measured by ICT indicators. data will be disaggregated according to income, Target 4.a seeks to ensure provision of educational gender, age, disability and other characteristics in facilities which are ‘child, disability and gender order to improve the granularity of monitoring. It sensitive’ and to provide ‘safe, non-violent, is recognized, however, that this will be difficult to inclusive and effective learning environments achieve. The resources available to NSS in many for all.’ As well as two ICT sub-indicators – the developing countries are limited. In many cases, availability of computers and the Internet – it will national data sets are not collected regularly, be measured by sub-indicators concerned with the are not comparable or are not disaggregated. availability of electricity, adapted infrastructure UNSC has therefore called for steps to be and materials for students with disabilities, basic taken to standardize indicators internationally, drinking water, single-sex basic sanitation facilities, 11 strengthen national capacity to collect data, and and basic handwashing facilities. improve reporting mechanisms. Nevertheless, it is recognized that it will be difficult to obtain high- The sub-indicators concerned with computers and 12 quality data for some indicators in a number of indicators. The definition the Internet are Tier I/II 10 countries for some time. of a computer for the purpose of indicator 4.a is ‘a programmable electronic device that can Indicators in the framework have been grouped store, receive and process data, as well as share into three tiers reflecting variations in data information in a highly-structured manner,’ 13 standards and availability: including desktop, laptop and tablet devices. Internet access, as defined for the indicator, • Tier I – indicators for which an established includes fixed-narrowband access at download methodology exists and data are already speeds of less than 256 Kbit/s, as well as fixed- widely available. broadband access at speeds higher than 256 Kbit/s and mobile-broadband access of at least 3G Tier II – indicators for which a methodology • standard. has been established but for which data are not easily available. Data concerned with the provision of computers and the Internet in schools are collected by the • Tier III – indicators for which an internationally UNESCO Institute for Statistics (UIS) from different agreed methodology has not yet been national sources , including ministries of education, developed. national statistical offices and other specialized agencies. Data concerning schools with computer Six of the 230 indicators in the framework (one access were available in 2015 for 66 countries and of which is divided into two sub-indicators) are territories, along with data for 40 countries and concerned explicitly with ICTs. These indicators, territories concerning schools with Internet access. the targets with which they are concerned and the No data were available for countries in Europe, agency responsible for data-gathering at global where computer and Internet access in schools is level are shown in Table 3.2. generally high. Some UIS data currently available are relatively old: The following sections of this chapter discuss the relationship between these targets and indicators, Two countries supplied data concerned with • data sources and availability, as well as current computers in schools from 2014, with a levels of achievement where these indicators are further seven supplying data from 2013. concerned. Measuring the Information Society Report 2016 81

98 Table 3.2: ICT indicators and related SDG targets Ta rge t Indicator Agency** Tier 4.a Build and upgrade education facilities that I/II* UIS Proportion of schools with access to are child, disability and gender sensitive computers for pedagogical purposes and provide safe, non-violent, inclusive and Proportion of schools with access to the UIS I/II* effective learning environments for all Internet for pedagogical purposes 4.4 By 2030, substantially increase the number II Proportion of youth/adults with ICT skills, by ITU of youth and adults who have relevant type of skills skills, including technical and vocational skills, for employment, decent jobs and entrepreneurship 5.b Enhance the use of enabling technology, in Proportion of individuals who own a mobile ITU II particular information and communication phone, by sex technology, to promote the empowerment of women 9.c Significantly increase access to information ITU I Percentage of the population covered by a and communications technology and strive to mobile network, by technology provide universal and affordable access to the Internet in least developed countries by 2020 I Fixed Internet broadband subscriptions, by ITU 17.6 Enhance North-South, South-South and speed triangular regional and international cooperation on and access to science, technology and innovations, and enhance knowledge-sharing on mutually agreed terms, including through improved coordination among existing mechanisms, in particular at the United Nations level, and through a global technology facilitation mechanism ITU I Proportion of individuals using the Internet 17.8 Fully operationalize the technology bank and science, technology and innovation capacity- building mechanism for least developed countries by 2017 and enhance the use of enabling technology, in particular information and communications technology Note: *Since Target 4.4 includes several sub-indicators (some collected by UIS, and others by other agencies), this tier reference refers to the different sub-indicators. Both UIS indicators were classified as Tier I/II since data for both indicators are collected by UIS, based on an established methodology. While data exist for a large number of countries, UIS does not yet have a regular, global data collection, and for most countries data exist for only one year. ** UIS = UNESCO Institute for Statistics; ITU = International Telecommunication Union. Source: ITU, adapted from ECOSOC (2016). schools, some have combined these while others Data for the remaining 57 countries and • have provided data for only one or the other territories were derived from 2012 or earlier, category. with those from 29 of them being derived from 2009 or 2010. Substantial variations between countries are evident from those data which are available. • In the case of data on Internet in schools, Some countries from across different regions – three countries supplied data from 2013 including Armenia, Barbados, Malaysia, Mongolia, and 2014, with the remaining 37 countries Oman and Uruguay – reported achieving 100 per supplying data from 2012 or earlier, 18 of cent access to computers in both secondary and them from 2009 and 2010. primary schools as early as 2012. Others reported significantly lower figures at such earlier dates. Most of the available data, therefore, do not Where figures for secondary and primary schools allow for effective assessment of performance are reported separately, a higher proportion of against this target in 2015. In addition, while the former is usually reported to have computer most countries and territories that have provided data have disaggregated secondary and primary Measuring the Information Society Report 2016 82

99 Chapter 3 rcentage of schools with computers, selected African countries, various years Pe Chart 3.1: Source: UIS. seeking data on these indicators during 2016, access. The most recent data for this indicator are results from which are expected in 2017. 3. 1. derived from Africa, and are illustrated in Chart No data comparable to these are available for Internet access in Africa. However, data from Indicator for Target 4.4: before 2013 show a similar pattern in other Proportion of youth/adults with ICT skills, by regions to that for computer access – with some type of skills countries achieving 100 per cent access at least for secondary schools by 2012, and generally Target 4.4 within SDG Goal 4 seeks, with a target higher levels of access in secondary than in date in 2030, to ‘substantially increase the primary schools. number of youth and adults who have relevant skills, including technical and vocational skills, for The principal challenge concerning these two employment, decent jobs and entrepreneurship.’ indicators is the need to gather comprehensive recent data, rather than relying on information ICT skills enhance the level of use that can which significantly pre-dates adoption of the be made of ICTs by individuals, businesses target. UIS issued a questionnaire to all countries and organizations, while the lack of ICT skills is generally considered one of the principal Measuring the Information Society Report 2016 83

100 barriers preventing people from deriving full except for Europe. However, the data currently available do give a broad indication of current benefit from ICT availability. In particular, the UN outcomes and set a benchmark for the future. General Assembly’s WSIS+10 review asserted that ‘differences in individuals’ capabilities to both Chart 3.2 shows the average level of attainment use and create information and communications technologies represent a knowledge divide that reported among countries for the nine different skills included in the indicator across the whole perpetuates inequality’ (UNGA, 2015). dataset, as a percentage of the population. It also distinguishes between developed and developing The indicator assigned for this target is the countries. However, it is not weighted according to proportion of youth and adults with such skills, the populations of the different countries involved. disaggregated by type of skill. Information derived from this Tier 2 indicator should enable governments and other stakeholders to improve Chart 3.2 shows that, with the exception of the link between ICT usage and impact, targeting programming, the share of the population with specific ICT skills is considerably higher improvements in ICT literacy and proficiency while in developed countries than it is in developing paying particular attention to the gender digital divide, differences between people in different age countries. The difference between highly- connected countries (in the top quartile of the groups, and the needs of vulnerable groups such IDI distribution) and less-connected countries is as the unemployed and people with disabilities. even more marked. Only eight countries reported This relatively new indicator was added to the figures of over ten per cent of adults with list of core indicators of the Partnership on programming skills – Iceland (the highest, at 18.2 14 Measuring ICT for Development per cent), Croatia, Denmark, Luxembourg, Sweden and endorsed and Norway in Europe, together with Bahrain and by UNSC in 2014, since when a corresponding Morocco in the Arab States region – while two question has been included in ITU’s annual countries reported figures below 1 per cent for questionnaire to National Statistical Offices (NSOs). this indicator (Azerbaijan and Zimbabwe). Nine skills are included in the questionnaire and indicator, including copying/moving files and Chart 3.3 presents illustrative spider chart for documents, using e-mail, connecting devices, three individual countries which illustrate the using spreadsheets and presentation programmes, diversity of reported ICT skill levels between configuring software and programming. At countries at different levels of development – national level, data are collected through national Sweden, a highly developed and highly connected household surveys, based on self-reported member of the OECD; Morocco, a middle- answers which are not independently verified. income developing country in North Africa; and Zimbabwe, a landlocked developing country, A total of 51 economies have provided data for which is at the top of the least connected group of this indicator, almost all of them supplying data countries in the IDI. from 2014 or 2015. Of these economies, 36 are in Europe, five in the Asia-Pacific region, five in the Out of the total of 51 countries reporting data, Arab States region, two in the CIS region, two in 41 provided gender-disaggregated data for this the Americas and only one in Africa. They include indicator, of which 33 are developed countries and 34 developed and 17 developing economies. All eight are developing countries. (One developed but six of those providing data fall within the top country provided gender-disaggregated data for half of rankings for ITU’s ICT Development Index only one skill.) No gender-disaggregated data (IDI) (see Chapter 1), while only one (Zimbabwe) were provided by countries in the lowest quartile falls into the lowest quartile of the IDI (least of the IDI distribution. connected countries), and none is an LDC. There are some differences in the age ranges applied Chart 3.4 shows the average level of attainment in these surveys which may have an impact on 15 reported for each of the nine skills included in interpretation. the indicator in these 41 countries, by proportion The limited geographical range of countries of their male and female populations, in those reporting data means that it is not yet possible developed and developing countries that reported to produce regional aggregates for this indicator, findings for this indicator. Measuring the Information Society Report 2016 84

101 Chapter 3 Chart 3.2: oportion of individuals with ICT skills, by type of skill, latest available year, 2012-2015 Pr Note: Based on simple averages of 51 countries that reported data (34 developed and 17 developing countries). Source: ITU. Chart 3.3: oportion of individuals with reported ICT skills, Sweden, Morocco and Zimbabwe, 2014/2015 Pr Source: ITU. women had acquired ICT skills in all of the skill In 25 of the 41 countries reporting data for this indicator, a higher proportion of men than categories concerned. The discrepancy between Measuring the Information Society Report 2016 85

102 oportion of individuals with ICT skills, by type of skill, by sex, developed (left) and developing Pr Chart 3.4: (right) countries, latest available year (2012-2015) Note: Based on simple averages of 41 countries that reported data (33 developed and eight developing). Source: ITU. men and women was more marked in hardware wider evidence base is required for this indicator and software skills (installing and configuring to support implementation of SDG target 4.4, devices and software) than in usage skills (such particularly gender-disaggregated data from LDCs as e-mail, spreadsheets and presentations). At and less-connected countries. Achieving this will the most basic level, only one country (Slovenia) require support to NSOs to improve the frequency recorded a higher proportion of women than and reliability of household surveys. men able to move files and folders. At the more advanced level, only one country (Qatar) recorded a higher proportion of women than men with Indicator for Target 5.b: skills in connecting and installing new devices, Proportion of individuals who own a mobile installing and configuring software, or computer phone, by sex programming. No country recorded a higher proportion of women with skills in transferring files SDG 5 aims to ‘achieve gender equality and and folders between devices of different types. empower all women and girls.’ It includes nine targets, addressing different aspects of gender The difference in the male/female skills gap equality. Target 5.b seeks to ‘enhance the use of between developed countries in Europe and the enabling technology, in particular information small number of developing countries reporting and communications technology, to promote the data for this indicator is not particularly high. empowerment of women.’ The outcome document However, the small number of developing from the UN General Assembly’s review of the countries reporting data and absence of any World Summit on the Information Society similarly LDCs or least connected countries mean that this called for ‘immediate measures to achieve gender apparent finding is not generalizable. equality in Internet users by 2020,’ a goal which is 16 An also included in ITU’s Connect 2020 Agenda. The data currently available, and reported here, assessment of efforts to measure ICT and gender provide a useful starting point for assessing this was published by the Partnership on Measuring indicator. However, the majority of available data ICT for Development in 2014 (UNCTAD, 2014). are currently derived from developed countries in Europe, which have high levels of connectivity, Mobile phones have spread rapidly to become, educational attainment and ICT skills. A much for many people, the principal means of business Measuring the Information Society Report 2016 86

103 Chapter 3 and interpersonal communications and Internet At present, however, data on the proportion of use. Although the number of mobile phone individuals who own a mobile phone are available subscriptions now exceeds the global population, for only a small number of countries, and are there are many duplicate subscriptions, with the often not gender-disaggregated. Only 21 countries GSM Association estimating there to have been and territories supplied data on mobile phone 4.7 billion unique mobile subscribers in 2015 (see ownership through the 2015 questionnaire, with Chapter 5) (GSMA, 2016). Furthermore, some of 15 of them having provided data relating to 2015 those subscribers own one or more SIM cards but itself. Only 12 of these countries and territories do not own a phone device. supplied sex-disaggregated data. Results for these 12 countries are set out in Chart 3.5. While all are Mobile phone ownership is valuable in tracking developing countries, only one (Burundi) is an LDC gender equality since mobile phones are personal or LCC. Three other countries (Egypt, Indonesia devices that can provide women and girls with and Morocco) also fall within the lower half of the greater independence and autonomy in their IDI rankings (see Chapter 1). It should be noted social and economic lives. A number of studies that there are some variations in the age groups have identified a significant gender gap in mobile covered by these data in different countries. phone ownership and use that is broadly but not exclusively associated with other differences These data are consistent with findings in other between men and women’s life experience, research which suggest that there is a gender gap particularly where income and educational in mobile phone ownership and use, and that this 17 attainment are concerned. ITU has reported that varies between different countries. Although there there is a gender gap of 12 per cent in Internet is insufficient evidence in the current dataset to use worldwide in 2016, and that the gap is confirm this, other evidence suggests that this considerably higher (30.9 per cent) in LDCs (ITU, gender gap may be higher in lower-income, less- 21 2016). Building a stronger evidence base around connected countries. mobile phone ownership should help to determine The most appropriate means of data collection whether there is a comparable gender gap in for this indicator is through national household mobile ownership. surveys. While some countries include questions on mobile phone ownership in household surveys, The indicator assigned for this target is the many do not yet do so. This partly explains why proportion of individuals who own a mobile phone, only a small number of countries provided data by sex. This is a Tier II indicator, with established for this indicator following its inclusion in the ITU international measurement standards. It was annual questionnaire for the first time in 2015. ITU developed by the Task Group on Gender of the is encouraging all countries to add this indicator to Partnership on Measuring ICT for Development, such surveys, where these are undertaken, in the approved by the World Telecommunication/ hope that this will increase the availability of data ICT Indicators Symposium in 2014, and has been 18 in the near future. Other surveys of mobile phone added to the Partnership’s Core List of Indicators. It seeks to measure the number of individuals with ownership undertaken by commercial businesses a mobile phone device and at least one active SIM and non-governmental organizations provide 19 for personal use, either prepaid or postpaid, card valuable information to supplement this indicator, 22 but does not include those who own a SIM card and should also be taken into account. 20 without a mobile device. Measurement of this indicator should help Indicator for Target 9.c: governments and other stakeholders to design Percentage of the population covered by a policies that can address the gender digital divide mobile network, by technology and support initiatives aimed at other aspects of gender inequality. Data for the indicator will be SDG 9 aims to ‘build resilient infrastructure, collected by ITU through an annual questionnaire promote inclusive and sustainable industrialization to NSOs, which was issued for the first time in and foster innovation,’ and includes eight targets. 2015. Target 9.c seeks to ‘significantly increase access to information and communications technology and strive to provide universal and affordable access to Measuring the Information Society Report 2016 87

104 oportion of individuals owning a mobile phone, by sex, 2014/2015 Pr Chart 3.5: Source: ITU. the Internet in least developed countries by 2020.’ on an internationally agreed definition and 2030 Agenda This is the only target within the methodology and included in the Partnership which is specifically concerned with ICT networks on Measuring ICT for Development’s Core List of and services. Indicators. Information from this indicator should help governments and businesses to design The proportion of the population covered by regulatory frameworks and business models for a high-speed mobile-cellular network can be broadband deployment that will maximize benefits considered a useful indicator for ICT access. Over to communities that are currently underserved. the last decade, mobile-cellular networks have expanded rapidly, overcoming many of the access ITU collects data for this indicator through limitations of fixed terrestrial networks and thus an annual questionnaire sent out to national extending inclusion in basic telecommunications. communications regulators and ICT ministries, While 2G (narrowband) networks offer basic which in turn obtain data from licensed mobile- access, particularly to voice-based services, which cellular operators. Data on 2G networks were is valuable, effective Internet access requires available for 144 countries in 2015, while data for access to 3G or higher (broadband) networks. 3G networks were available for 135 countries. In Access to broadband networks is therefore crucial some countries, the available data refer only to the to Internet inclusiveness, as well as to the more operator with the most extensive network, which sophisticated services that foster innovation and may underestimate total coverage. enable online business. Chart 3.6 shows the evolution of mobile-network The indicator assigned to this target is the coverage worldwide since 2007. The proportion percentage of the population covered by a high- of the world’s population living in areas without speed mobile network (i.e. those whose home mobile coverage is now small, but still significant. is within range of a mobile signal), irrespective The proportion covered by a mobile-broadband of whether they are mobile phone users or network will reach 84 per cent in 2016, but only 67 subscribers, disaggregated between those per cent in the case of the rural population. Just with access to 2G, 3G and LTE or higher-speed over half (53 per cent) of the global population is networks. This is a Tier 1 indicator that is based now covered by LTE or higher networks. Measuring the Information Society Report 2016 88

105 Chapter 3 bile network coverage and evolving technologies, 2007-2016 Mo Chart 3.6: Note: * Estimate. Source: ITU. The Internet has become an increasingly important Indicator for Target 17.6: resource providing access to information, Fixed Internet broadband subscriptions, by enhancing knowledge-sharing and facilitating speed international cooperation in science, technology and innovation. Reliable broadband access is SDG 17 aims to ‘strengthen the means of essential in order to use more sophisticated ICT implementation and revitalize the global applications, including those required for scientific partnership for sustainable development.’ This collaboration. While mobile-broadband networks broad goal includes 19 targets addressing different are increasingly widely available in both developed aspects of global development, two of which have and developing countries, many developing assigned ICT indicators. countries have only limited fixed-broadband availability, which is considered preferable Target 17.6 seeks to ‘enhance North-South, South- for high-volume, time-critical applications. South and triangular regional and international Those networks which are available also vary cooperation on and access to science, technology considerably in terms of the speed of access they and innovations, and enhance knowledge- can provide, presenting barriers to international sharing on mutually agreed terms, including cooperation. through improved coordination among existing mechanisms, in particular at the United Nations The Tier I ICT indicator for this target, which level, and through a global technology facilitation concerns fixed-Internet broadband subscriptions, mechanism.’ It has two assigned indicators, differentiated by speed of access, is based on an one concerned with science and technology internationally-agreed definition and methodology. agreements (which lies outside the scope of this Fixed broadband subscriptions per 100 inhabitants, chapter), the other with fixed Internet broadband a related but distinct indicator, is included in the subscriptions. Partnership on Measuring ICTs for Development’s Core List of Indicators and forms part of the IDI, Measuring the Information Society Report 2016 89

106 the latest results from which are reported in Europe is 29 per cent, while that in other regions Chapters 1 and 2. Information concerning this is much lower. Within the African region it is just indicator will help governments and Internet 0.5 subscriptions per 100 inhabitants (including businesses to target public and commercial the region’s two highest performers, the small resources in areas that will enhance scientific and island States Mauritius and Seychelles, which had technical collaboration, with anticipated onward subscription levels of 15.7 and 14.3 respectively), benefits for productivity and economic growth. while only two other countries in the region – South Africa and Cape Verde – had rates above ITU collects data for this indicator through 2.0. With the exception of Europe, each region an annual questionnaire sent out to national includes one or more countries with very low communications regulators and ICT ministries, levels of fixed-broadband penetration. which obtain data from Internet service providers 23 (ISPs). Data and/or ITU estimates for the overall Table 3.3: xed broadband subscriptions per 100 Fi number of fixed-broadband subscriptions inhabitants, per region, 2015 per hundred inhabitants are available for 205 Highest Lowest economies for 2015 (including 19 territories not Weighted performing performing included in ITU regions), while data differentiated average country country by speed are available for 115 economies 15.8 0.5 0.0 Africa 24 (including two such territories). 0.1 22.8 4.2 Arab States Asia & Pacific 8.9 40.2 0.0 Since most ISPs offer broadband plans linked 0.1 31.4 14.8 CIS to download speed, the indicator is relatively 47.5 7.6 29.2 Europe straightforward to collect, but data collected at The Americas 18.4 36.4 0.0 national level do not all follow the same tiers of broadband capacity. ITU has therefore sought Source: ITU. to collate information in three bands: between kbit/s (the lowest data-transfer rate to be 256 Most fixed-broadband subscriptions in developed considered broadband) and 2 Mbit/s; between countries now offer higher advertised speeds. 2Mbit/s and 10 Mbit/s; and above 10 Mbit/s. All broadband subscriptions in the Republic of Korea are now reported to offer speeds of 10 There are very substantial differences between Mbit/s and above. The number of broadband developed and developing countries, and within subscriptions at speeds at or above 10 Mbit/s in 25 regions, in terms both of the proportion of is 21.8 per hundred inhabitants, more Europe inhabitants with fixed-broadband subscriptions than three-quarters of the total with fixed- and the speeds that these subscriptions prove broadband subscriptions in that region. Thirty- (Chart 3.8). While some countries, such as the two countries in Europe and sixteen economies Republic of Korea, Denmark and France have outside that region reported that more than half fixed-broadband penetration rates of around of their subscriptions have speeds at or above 10 40 per cent and almost exclusively high-speed Mbit/s, while only one country in Africa (Mauritius) connections of above 10 Mbps, many low-income reported more than 1 per cent of its subscriptions economies have less than 2 per cent fixed- at that level. broadband penetration rates, and exclusively lower-speed connections of below 2 Mbps. Table Datasets for this indicator are relatively well 3.3 illustrates the weighted average for fixed- established, with historic data for a substantial broadband subscriptions within each region, number of countries, and estimates for others, together with figures for the highest and lowest available back to 2008. While data on fixed- performing countries in each region. It also broadband subscriptions are available for the includes ITU’s estimate for the number of fixed- large majority of economies, more data on broadband subscriptions for each region in 2016, different broadband speeds are needed, especially including estimates for other countries which have for developing countries and LDCs. Greater not provided data. standardization in reporting might also facilitate monitoring and analysis. In addition, as the Table 3.3 shows that the number of fixed- capacities of broadband networks continue to broadband subscriptions per 100 inhabitants in increase during the SDG implementation period, Measuring the Information Society Report 2016 90

107 Chapter 3 it will from time to time be necessary to include categories and different social groups, should help higher-speed categories within this indicator. governments and other stakeholders to target resources in order to encourage affordable and effective use of the Internet. Indicator for Target 17.8: The proportion of individuals using the Internet is Proportion of individuals using the Internet defined as the proportion who have used it from any location in the last three months. The Tier I The second target within SDG 17 to which an indicator for this target is one of the Partnership ICT indicator is assigned is Target 17.8 which on Measuring ICT for Development’s Core List of seeks to ‘fully operationalize the technology Indicators, is based on an internationally-agreed bank and science, technology and innovation methodology, and is included in the IDI, the latest capacity-building mechanism for least developed results for which are discussed in Chapters 1 and countries by 2017 and enhance the use of 2. It was also used to support the measurement enabling technology, in particular information of Millennium Development Target 8.F, which was and communications technology.’ No indicator concerned with making available the benefits of has as yet been assigned for the first part of this new technologies, especially ICTs. target. The indicator which has been selected for the second part of the target, concerned with Data for this indicator are collected by ITU through enhancing the use of enabling technology, in an annual questionnaire sent out to NSOs. In most particular ICT, is the proportion of individuals using developed countries and a growing number of the Internet. developing countries, NSOs obtain data for this indicator through national household surveys. This target recognizes the Internet’s substantial Such data are available for 100 countries from at and growing importance in all aspects of least one survey in the period 2011-2015. Where sustainable development (economic, social and NSOs have not collected data, ITU uses a variety environmental), and in particular its importance of techniques to estimate the percentage of as an enabler of development for individuals, individuals using the Internet, including hot-deck communities and countries. The Internet provides imputation, regression models and time series extensive and growing access to information, forecasting. Hot-deck imputation uses data from services and applications which add value to countries with “similar” characteristics, such as people’s lives, enhance their productivity and GNI per capita and geographic location. enable them to access new opportunities. Lack of Internet access and use can exacerbate existing Chart 3.7 illustrates the differences between disadvantage. Understanding gaps in access to regions and between developed countries, and usage of ICTs between and within countries, developing countries and LDCs for this indicator. and between women and men, different age Pr Chart 3.7: oportion of individuals using the Internet, by region and by development status, 2016* Note: *Estimate. Source: ITU. Measuring the Information Society Report 2016 91

108 Chart 3.7 shows clearly that Internet usage Summary and conclusion 3.4 rates, using the definition for the indicator, are about twice as high in developed countries as in The United Nations has adopted 17 SDGs, developing countries, and more than twice as high supported by 169 targets and 230 indicators, in developing countries as a whole than they are to guide international development policy in LDCs. Europe, the CIS and the Americas have and practice between 2015 and 2030. Six of much higher Internet usage rates overall than the the indicators directly concern ICTs, and their African, Arab States and Asia/Pacific regions. In the measurement will provide important evidence case of developed countries, moreover, it is worth concerning progress towards implementation noting that higher bandwidth available to users, 2030 Agenda for Sustainable of the UN’s lower broadband access costs in relation to GNI . While a substantial evidence base Development p.c. (see Chapter 4) and generally higher levels of exists for several of those indicators, concerned educational attainment make it likely that people with ICT infrastructure and adoption, the evidence in developed countries are more intensive users of base for others – particularly those concerned the Internet than those in developing countries. with ICTs in education, ICT skills and gender equity – is less substantial. ITU is working with the international statistical community and national statistical systems to improve the coverage and quality of the required data. Measuring the Information Society Report 2016 92

109 Chapter 3 Chart 3.8: Fi xed-broadband subscriptions per 100 inhabitants, by speed, 2015 Measuring the Information Society Report 2016 93

110 Chart 3.8: Fixed-broadband subscriptions per 100 inhabitants, by speed, 2015 (continued) Source: ITU. Measuring the Information Society Report 2016 94

111 Endnotes 1 2030 Agenda can be found at http:// ww w. un. org / ga/ search/ view_ doc. asp? symbol= A/ RES/ 70/ 1& Lang= E. The 2 Experience to date is analysed in: UN Commission on Science and Technology for Development (2015), World Bank (2016); and Broadband Commission for Sustainable Development (2015). 3 itu. int/ en/ ITU- D/ Sta tistics/ Pag es/ st at/ de fault. aspx. ww w. See ITU (2016) and http:// 4 workspace. unpan. org / sites/ Int Documents/ UNPAN96078. pdf . http:// Resolution A/RES/70/125, at ernet/ 5 , http:// www. un. org / ga/ search/ view_ doc. asp? symbol= A/ RES/ 70/ 1& Lang= E, para. 15. The 2030 Agenda 6 http:// unpan. org / sites/ Int ernet/ Documents/ UNPAN94615. pdf . workspace. 7 wsis- itu. int/ net4/ wsis/ sdg/ Cont ent/ www. sdg_ matrix_ document. pdf. https:// 8 See, for example, UN Secretary-General’s Expert Advisory Group on Data Revolution (2014). 9 Report of the Inter-Agency and Expert Group, loc. cit. 10 ibid. , paras 26-31. 11 unstats. un. org / sdgs/ met adata/ files/ Met adata- 04- http:// 0A- 01. pdf . 12 Since the Target 4.4 includes several sub-indicators (some collected by UIS, and others by other agencies), this tier reference refers to the different sub-indicators. Both UIS indicators were classified as Tier I/II since data for both indicators are collected by UIS, based on an established methodology. While data exist for a large number of countries, UIS does not yet have a regular, global data collection and data exist for only one year for most countries. 13 ibid. 14 ww w. A collaboration between United Nations and other international agencies, established following WSIS: see http:// int/ en/ itu. D/ Sta tistics/ Pag es/ Intlc oop/ partnership/ def ault. aspx. ITU- 15 In Europe, these relate to the adult population aged 16-74. Data for developing countries vary, in some cases referring to the whole population, in others to the adult population only or to a wider age range than European data. The fact that younger children are less likely to have acquired ICT skills than older children may exacerbate the difference between developed and developing countries. However, this may be offset by the higher proportion of the European population which falls into older age groups, which were educated before the prevalence of computers and the Internet. 16 org / sites/ Int ernet/ Documents/ UNPAN96078. pdf , para. 27; http:// http:// w. unpan. itu. int/ en/ connect2020/ workspace. ww default. aspx. Pages/ 17 For example, Gillwald, A., Milek, A. & Stork, C. (2010) and GSMA (2015). 18 e- www. itu. int/ en/ ITU- D/ Sta tistics/ Documents/ cor eindicators/ Cor List - of- Indica tors_ March2016. pdf . EGH has http:// agreed that this indicator should also be included in the ITU data collection. 19 An active SIM card is one that has been used within the last three months. 20 unstats. un. org / sdgs/ files/ met adata- compila tion/ http:// Met adata- Goal- 5. pdf , p. 43. 21 GSM Association, ; For further discussion of the gender gap in mobile phone and Internet access and use, see op. cit. ww we forum. org / agenda/ 2016/ 05/ smartphones- w. are- closing- the- digital- divide- and- these- A4AI (2015). See also https:// hav e- made- the- most - countries- progr ess? utm_ con tent= buffer c5a03& utm_ medium= social& utm_ source= twitter . com& utm_ campaign= buffer . 22 See, for example, GSMA (2015); Gillwald, A., Milek, A. & Stork, C. (2010), Zainudeen, A. and Galpaya, H. (2015). 23 2015 data on the number of fixed-broadband subscriptions for 43 economies were estimated. 24 ITU does not estimate data for fixed-broadband subscriptions broken down by speed. 25 Disaggregated data are available for only 38 of the 42 European countries in the sample. Measuring the Information Society Report 2016 95

112

113 Chapter 4. ICT prices

114

115 Key findings Many people continue to be excluded from the global information society, and the relatively high cost of ICT services remains one of the main barriers to ICT uptake. Monitoring prices is critical for developing policies that aim to make ICT services affordable for all citizens. . For the Mobile-cellular prices continued to fall in 2015, and more steeply than in previous years first time, the average cost of the mobile-cellular basket (which includes 100 SMS and 30 mobile calls per month) in developing countries accounted for less than 5 per cent of GNI per capita. Least developed countries (LDCs) saw a 20 per cent drop in mobile-cellular prices, the strongest decrease in five years. The price drop is linked to the growing availability of prepaid packages that bundle SMS and local calls. Innovative pricing schemes, such as dynamic discounting, are also helping to make the service more affordable for low-income groups. The Asia and the Pacific region has the lowest average PPP$ price for mobile-cellular services of all . It is home to the countries with the lowest mobile-cellular price baskets worldwide: Sri Lanka regions and Bangladesh, where prices stand out at PPP$ 2.45 and PPP$ 4.14 per month. Fixed-broadband prices continued to drop significantly in 2015 but remained highest – and clearly unaffordable - in a number of LDCs. Globally, the price of a basic fixed-broadband connection fell from around USD 80 per month in 2008 to USD 25 in 2015, corresponding to a drop in the ratio of price to average GNI per capita from over 90 per cent to 14 per cent. In LDCs, a fixed-broadband plan with a minimum of 1GB of data per month still corresponds to over 60 per cent of GNI per capita. The service is sold at over USD 300 a month in Uganda, Chad and the Central African Republic, and remains very expensive and clearly unaffordable in some of the small island developing States. . In developed People in most low-income countries get lower speeds and quality for their money countries, the minimum speeds of entry-level fixed-broadband packages have increased considerably in recent years. Developing countries, on the other hand, are only gradually upgrading broadband infrastructure to offer higher speeds. In 2015, not a single developed country offered an entry-level broadband connection with speeds below 1 Mbit/s, but a large majority of LDCs did. These differences in available speeds have an impact on the types of services and applications that users can access and benefit from. Mobile-broadband is cheaper and more widely available than fixed-broadband, but is still not deployed in the majority of LDCs. Globally, handset-based mobile-broadband prices fell from an average of PPP$ 29 per month in 2013 to PPP$ 18 in 2015. Mobile-broadband services are offered in only 38 per cent of the LDCs; however, in those countries where the service is offered, handset-based prices more than halved in PPP terms between 2012 and 2015 and currently account for 11 per cent of GNI per capita. Still, mobile-broadband cannot always replace fixed-broadband Internet access, especially in the business sector, and a growing number of applications require higher speeds and better connection quality. The decrease in mobile-broadband prices goes hand in hand with an increase in the intensity of use . Figures on mobile Internet traffic show that the amount of data consumed by each subscription is increasing in most countries for which data are available. This suggests that the reduction in mobile- broadband prices contributes not only to connecting more people but also to fostering more intense Internet usage among those who are already online. Measuring the Information Society Report 2016 99

116

117 Chapter 4: ICT prices models, including zero-rating, refer to agreements Introduction 4.1 between operators and content providers, including Facebook, Google and Wikipedia, which Many people continue to be excluded from the offer clients access to restricted content, at no or global information society, and the relatively reduced cost. high cost of ICT services remains one of the key barriers to ICT uptake. Survey-based data that ITU ‘Zero-rating’ or ‘price-differentiation’ schemes collects from national statistical offices confirm have become a widely discussed topic, and that, next to the availability of access and the supporters and opponents highlight both their relevance of services, affordability is one of the benefits and their risks. What is the cost of key factors that continue to determine whether allowing ICT users to have access not to the wider or not people will use ICTs. A number of recent only to selected Internet content, Internet but studies on ICT developments also confirmed these 1 which is determined by the provider? Opponents findings. Monitoring prices is therefore a critical point to the threat to net-neutrality and consumer step towards better policies to make ICT services choice and evoke anti-competitive behaviour, more affordable. while supporters emphasize the benefits of making services more affordable, or even free, The need to provide affordable access to ICTs has and bringing more people in developing countries been clearly recognized by policy-makers at the online. Advocates suggest that access to some national and international level. The World Bank’s information is better than none, and that greater 2015 World Development Report states that demand for general Internet access can encourage collecting Internet price data and benchmarking is 3 investments in infrastructure, whereas detractors the first step towards better regulation for lower 2 lament the lack of empirical evidence on the risks prices. but also on the effectiveness of zero-rated or otherwise free but restricted services, suggesting Furthermore, making ICT services more affordable that it is still difficult at this stage to make and increasing the number of ICT users will play informed policy decisions. In the meantime, zero- a key role in the context of the 2030 Agenda rating type services have been banned in several for Sustainable Development. This new global European countries, as well as in Chile, Japan and, development agenda, which was adopted by the more recently, India and Egypt, while free access United Nations in September 2015, recognizes the to certain content or applications continues to be immense potential of ICTs to “ accelerate human offered by service providers in many developing ” and specifically refers to the need to progress 4 As of June 2016, for and developed countries. “ significantly increase access to information and instance, Facebook’s Free Basics was available in communications technology and strive to provide some 40 countries worldwide (Figure 4.1). ..” universal and affordable access to the Internet. (UNGA, 2015c). Although much debate has focused on zero-rating offers, a recent study highlights that data-specific The main objective of this chapter is to investigate services may actually be the most popular way of the price and affordability of all key ICT services, providing cheaper, but restricted Internet access benchmark countries and regions and highlight key (Box 4.1). trends over time. Electricity and device costs Free, or low-cost, access – at what price? The affordability of services is important, but A current debate – and controversy – among the cost of electricity for charging a device and policy-makers has been driven by a number of the one-time purchase price of the device, in initiatives that provide lower-priced or free access particular a mobile phone, are equally important to service-specific data plans. These data service Measuring the Information Society Report 2016 101

118 Figure 4.1: cebook’s Free Basics around the world, as of June 2016 Fa ve- we where- ory/ st en/ org ernet. int o. inf Source: https:// launched/ / and can constitute important barriers (Facebook, in the affordability of mobile-cellular prices. 2015). With the growth in mobile-broadband Recent price and pricing trends will be examined, services and applications, smartphones provide highlighting changes in prepaid offers and bundled an excellent opportunity to access voice and services, which have an impact on the affordability data services, including in many rural and remote of services. areas. As smartphones become more affordable, consumer demand increases, and by September This will be followed by a more in-depth 2014 more smartphones than traditional mobile analysis of prices in the fixed-broadband and 6 handsets were sold in developing countries. mobile-broadband markets. Country rankings While smartphones are becoming smarter, will be presented for the fixed-broadband more widely available and more affordable, they and mobile-broadband sub-baskets, with the remain expensive for many of the world’s poorest latter including both prepaid and postpaid population groups (Box 4.2). packages and computer-based and handset- based plans. The analysis of fixed-broadband prices will include 2008-2015 price trends and a discussion on changes in broadband speeds About this chapter (offered for minimum broadband plans) as well as developments in terms of the data volume This chapter will look first at the evolution of included in broadband offers. A regional analysis 20 mobile-cellular prices over the period 2008 15, - will be provided for both fixed- and mobile- in absolute and relative terms, in USD, in broadband services. international dollars (PPP$) and as a percentage of GNI p.c., for both developed and developing This chapter will also look at the growing trend countries. It will include the presentation of the of offering bundled telecommunication services 2015 mobile-cellular sub-baskets and country and highlight some efforts to monitor the price of rankings, and show some regional differences bundles, in particular in OECD countries. Measuring the Information Society Report 2016 102

119 Chapter 4 Zero-rating and price-differentiation schemes Box 4.1: There are different ways of providing free or discounted access to restricted and selected services and applications, both over fixed as well as mobile networks. So called ‘zero-rating’ refers to services that make certain content or applications available at no, or no additional, cost to the customer, and data volumes used to access the specified site or application do not contribute towards the customer’s data usage. A user of a service provider offering Wikipedia Zero, for example, has unlimited, no-cost access to everything in the online encyclopaedia. Facebook’s Free Basics provides clients of certain mobile-network providers free access to a limited number of websites and applications. A November 2015 comparison of available data plans in eight developing countries (Kenya, 5 Nigeria, Ghana, Bangladesh, Philippines, India, Colombia and Peru) showed that zero-rated plans were offered in all of them, but not by all carriers. While the three largest carriers in Kenya offered at least one zero-rated service, zero-rating was only proposed by one operator in Nigeria. The most common plan generally offered was the service-specific data plan. Service-specific plans offer data bundles at discounted rates that give users access only to specific applications and sites, over a given web browser, and/or for a specified period of time. They are often part of operators’ marketing strategies to increase their customer base by providing discounted access to popular sites and applications. In Bangladesh, Kenya and Colombia, customers can also ‘earn’ extra data, for example by watching a certain video, or by buying a specific device. Overall, however, these earned data plans are not very frequent. Finally, full-cost plans are offered in all the countries studied, although they are not always the most common offering. While in Ghana seven out of 12 plans were offered at full cost, only one out of a total of 12 plans in the Philippines was a regular, full-cost plan (Chart Box 4.1). Chart Box 4.1: Pe rcentage of data plans, by type of plan, by country Note: Based on a simple count of all plans offered by the top carriers in each of the selected countries. A total of 12 plans were included in the comparison for all the countries except Bangladesh, where 13 plans were included. Source: ITU, adapted from A4AI (2015b). Measuring the Information Society Report 2016 103

120 Box 4.2: Smarter and cheaper: Gl obal smartphone prices continue to drop but remain high for low-income population groups While smartphones are becoming ‘smarter’, with increasing functionalities and processing power, prices are going down. As highlighted by IDC, the average smartphone price (ASP) continues to drop, falling to below USD 300 by end 2015. Relatively higher prices in developed regions reflect the use of more sophisticated and expensive phones (Chart Box 4.2). Lower prices in developing regions are also the result of handset manufacturers’ efforts to offer increasingly affordable entry-level smartphones for low-income users. Many budget (but smart) phones are on sale for less than USD 200, and producers in India and China are promising even lower prices. Even at these prices, however, many people in the world will not be able to own a smartphone. Chart Box 4.2: Av erage selling price of smartphones, 2015 Note: The average selling price for smartphone handsets is calculated by region as the total spent on smartphones divided by the total number of units sold. Source: ISOC (2015) based on data from IDC. As a percentage of countries’ monthly GNI Prices in this chapter are expressed in three • 8 Prices are expressed as a p.c. (Atlas method). complementary units: percentage of GNI p.c. in order to show them relative to the size of the economy of each • In USD, using the IMF annual rates of exchange. country, thus pointing to the affordability of each ICT service at country level. In international dollars (PPP$), using • purchasing power parity (PPP) conversion The prices collected for each service correspond factors instead of market exchange rates. The to the cheapest plan offered by the dominant use of PPP exchange factors helps to screen operator that fulfils the usage requirements of out price and exchange-rate distortions, thus each basket. The methodological details of the IPB providing a measure of the cost of a given and the collection of mobile-broadband prices can service taking into account the purchasing 7 be found in Annex 2. power equivalences between countries. Measuring the Information Society Report 2016 104

121 Chapter 4 lower income levels, such as Estonia and Lithuania 4.2 Mobile-cellular prices (Table 4.1). All these economies have in common very high mobile-cellular penetrations (more than 135 subscriptions per 100 inhabitants). Mobile-cellular prices continued to decrease in 2015 Overall, mobile-cellular services are quite affordable in a majority of countries: the cost Mobile-cellular prices continued to decrease in of the service represents less than 1 per cent of 2015, and the price drop was stronger than in GNI p.c. in 61 countries. Nevertheless, there are per cent in purchasing power previous years: 6 47 countries where the price still corresponds to parity (PPP) terms, twice as much as in 2014 more than 5 per cent of GNI p.c., most of them (Chart 4.1). In USD the decrease was even bigger, LDCs and/or low-income African countries. The in part due to the exchange-rate fluctuations examples of Bhutan, Bangladesh and Myanmar – with countries in the Euro area. By end 2015, a all of them with prices representing less than 2 per mobile-cellular basket cost approximately the cent of GNI p.c. – show that affordable mobile- same on average in developed, developing and cellular services are also achievable in low-income least-developed (LDCs) countries: the equivalent of LDCs. around PPP$ 21 per month. The decrease in prices led to an improvement Prepaid packages are driving prices down in the affordability of mobile-cellular services and, for the first time, the average cost of the Although mobile-cellular subscription growth mobile-cellular basket in developing countries 10 there is still a has slowed in most countries, per cent of GNI per corresponded to less than 5 9 significant proportion of the global population capita. Although LDCs are still far from achieving that do not use and/or own a mobile phone (see this milestone, prices in terms of GNI p.c. fell Chapter 5). To reach these people as well as to by 20 per cent in LDCs in 2015, the strongest retain current customers, competition is increasing decrease in the last five years. The historical trend and thus exerting downward pressure on prices. highlights the progress achieved in LDCs: the This is even more true in countries where average cost of 100 SMS and 30 mobile calls per mobile number portability has been effectively 1 per month has fallen at a steady rate of USD implemented (Chart 4.2). One strategy that 15.8 in 2008 to year (except in 2013), from USD mobile operators are employing to retain prepaid 9.1 in 2015. In parallel, GNI p.c. has increased USD customers without cannibalizing their revenue per cent in LDCs during the by more than 40 flows is that of offering lower priced value-for- same period. Combined, these developments money packages of bundled services. have made mobile-cellular services much more affordable than before in LDCs, at an average price Indeed, the drop in mobile-cellular prices in 2015 per cent of GNI p.c. in 2015. corresponding to 11 is linked to the increasing availability of prepaid packages that include bundles of SMS and local The list of the top ten countries with the most calls. When purchasing a package, the customer affordable mobile-cellular services includes high- obtains a discount in the price per unit and, in income economies such as Macao (China), Austria, exchange, has to pay upfront the cost of the Singapore, Hong Kong (China) and the United consumption included in the package. Arab Emirates, but also countries with much mobile-cellular sub-basket , which refers to the To monitor mobile-cellular prices, ITU uses the price of a standard basket of 30 outgoing calls per month (on-net/off-net to a fixed line and for peak and off-peak times, in predetermined ratios), plus 100 SMS messages. It is calculated as a percentage of a country’s average monthly GNI per capita, and also presented in USD and PPP$. The mobile-cellular sub-basket is based on prepaid prices, although postpaid prices are used for countries where prepaid subscriptions make up less than 2 per cent of all mobile-cellular subscriptions. Measuring the Information Society Report 2016 105

122 Chart 4.1: Mo bile-cellular sub-basket, as a percentage of GNI p.c. (top), in PPP$ (middle) and in USD (bottom), 2008-2015 Note: Simple averages. Based on 140 economies for which data on mobile-cellular prices were available for 2008-2015. Source: ITU. Measuring the Information Society Report 2016 106

123 Chapter 4 Table 4.1: Mo bile-cellular sub-basket, 2015 Mobile-cellular sub- Mobile-cellular sub- Tax rate Tax rate basket basket GNI p.c., GNI p.c., Economy included Economy Rank Rank included USD* USD* as% of as% of (%) (%) USD USD PPP$ PPP$ GNI p.c. GNI p.c. 7.05 17.26 20.0 4,020 0.09 5.68 0.0 7.36 1 Macao, China 98 Armenia 2.10 76,270 9.65 0.16 6.69 6.97 20.0 49,670 2 1,570 14.4 Austria 2.76 2.11 India 99 100 15.0 7.0 9.10 7.96 0.17 Singapore 55,150 Antigua & Barbuda 2.11 23.40 28.62 13,300 3 0.18 4 101 Nigeria 2.12 5.25 10.17 5.0 2,970 Hong Kong, China 6.02 7.52 0.0 40,320 10.38 23.5 1,590 44,600 0.0 102 Ghana 2.16 2.87 5 United Arab Emirates 0.18 6.70 9.15 15,451 6 0.21 18.11 14.57 25.0 103,630 103 Barbados 2.18 28.11 22.62 - Norway 104 7 17.0 29.48 10.23 2.24 Algeria 5,490 19,030 Estonia 0.21 3.33 4.67 20.0 Sweden 0.23 11.74 10.78 25.0 61,610 5.96 31.3 3.21 2.25 Sudan 105 1,710 8 10,210 0.0 25.02 18.68 0.24 Qatar 92,200 - - 20.72 2.43 Equatorial Guinea 106 9 5,150 18.0 25.29 10.59 2.47 TFYR Macedonia 107 15,430 21.0 5.59 3.22 0.25 Lithuania 10 2,090 11 0.27 11.05 9.98 24.0 108 Finland 20.0 - 4.52 2.60 Uzbekistan 48,420 109 Argentina 2.74 30.81 - 0.27 21.0 13,480 14.75 12.65 10.0 64,540 Australia 12 110 47,640 Germany 0.28 11.08 12.15 19.0 13 Dominican Rep. 2.79 14.02 28.62 30.0 6,040 15.24 0.0 37,663 Brunei Darussalam 14 0.29 9.04 111 Viet Nam 2.81 4.42 10.92 10.0 1,890 Sri Lanka 112 14.27 30.96 20.0 5,820 15 2.94 0.30 0.86 2.45 27.5 3,460 Serbia 0.31 1.85 5.43 9.0 7,113 16 113 Kyrgyzstan 3.08 3.21 10.56 17.0 1,250 Iran (I.R.) 26.71 17 8.56 114 Bosnia and Herzegovina 3.18 12.63 6.88 Cyprus 0.31 17.0 4,760 19.0 26,370 0.34 16.7 2,560 13.46 18.0 13,220 Russian Federation 3.69 18 115 Moldova 21.51 6.89 3.23 0.0 49,300 116 11.66 15.10 15.0 4,260 3.29 19 Kuwait 0.35 14.19 22.83 Tonga Latvia 117 Kenya 3.36 3.62 7.96 26.0 1,290 15,280 21.0 6.84 4.43 0.35 20 Luxembourg 75,990 15.39 3.37 Angola 118 21 20.47 5.0 0.36 22.81 21.15 5,476 17.0 119 41,070 15.0 12.17 13.25 0.39 New Zealand 22 6,090 12.0 29.53 17.17 3.38 Ecuador 120 23 0.39 7.71 10.40 22.0 23,580 Slovenia Bulgaria 3.42 21.73 50.01 20.0 7,620 3.42 22,657 37.8 10.03 7.76 0.41 Greece Philippines 121 24 3,500 12.0 22.90 9.99 122 25 0.42 15.28 12.60 20.0 43,430 2,720 16.0 15.61 8.19 3.61 Congo (Rep.) United Kingdom 26 123 13.0 5.51 3.87 0.46 Costa Rica 10,120 Dominica 3.66 21.13 28.42 15.0 6,930 Switzerland 0.47 34.29 21.77 8.0 88,032 27 124 Grenada 3.97 26.17 35.06 15.0 7,910 0.56 21.57 19.43 24.0 Iceland 46,304 28 125 Nepal 3.98 2.42 7.85 24.3 730 126 Malaysia 0.56 5.21 12.62 6.0 11,120 Morocco 3,070 20.0 24.22 10.24 4.00 29 127 51,630 13.0 24.57 25.07 0.58 3,936 30 Guyana 4.01 13.15 - 16.0 Canada 12.06 14.0 4.08 Swaziland 12.0 14.02 5.96 0.60 Kazakhstan 31 128 32.89 3,550 11,850 129 0.62 8.74 16.80 0.0 16,853 Fiji 4.18 16.97 28.56 15.0 4,870 Oman 32 9,630 13.0 26.39 13.80 4.22 El Salvador 3,920 15.0 9.18 5.05 0.63 Mauritius 33 130 5,999 - - 21.53 4.31 Cuba 21,039 34 Bahrain 0.63 11.12 18.74 0.0 131 2,870 132 4.48 10.73 22.95 13.0 Bolivia 35 Italy 0.64 18.30 19.99 22.0 34,270 Netherlands 36 28.02 28.19 21.0 33.33 133 0.65 St. Lucia 4.54 27.46 15.0 51,890 7,260 37 China 0.65 4.01 St. Vincent and the 6.63 0.0 7,400 134 4.68 25.77 34.66 6,610 15.0 Grenadines 38 7.63 15.19 23.0 13,690 Poland 0.67 135 - - 4.27 4.75 Tajikistan 1,080 Saudi Arabia 14.13 0.68 0.0 25,115 39 28.56 22.95 16.55 4.89 Samoa 136 4,060 15.0 25.0 61,310 28.45 35.22 0.69 Denmark 40 137 4.94 6.84 17.41 10.0 1,660 Lao P.D.R. 41 4.28 0.70 Belarus 7,340 20.0 - 3,060 Palestine 5.69 14.51 22.12 16.0 138 29.10 28.54 0.72 42 Belgium 21.0 47,260 5.70 Zambia 1,680 16.0 23.52 7.97 139 4.89 0.73 Turkmenistan 43 15.0 8,020 - 22.22 - 0.0 4,390 Marshall Islands 140 6.07 44 35.73 35.73 8.9 55,200 United States 0.78 141 15.0 8.70 3.09 6.74 Ethiopia 550 0.80 17.98 Korea (Rep.) 10.0 27,090 45 22.20 7.43 Honduras 142 2,270 15.0 27.40 14.05 21,360 19.51 14.22 0.80 Portugal 46 23.0 7.76 Yemen 143 1,299 5.0 - 8.39 7.26 0.82 4,280 10.0 47 2.92 Mongolia 5.0 25.79 8.70 7.85 Lesotho 144 1,330 48 Andorra 0.83 30.99 - - 45,033 8.00 38.11 - 7.0 5,720 Tuvalu 145 49 Romania 0.84 6.65 13.64 24.0 9,520 1,020 10.0 17.37 6.82 8.02 Cambodia 146 0.84 42,000 8.0 31.23 29.44 Japan 50 147 Cape Verde 8.30 23.87 50.57 15.5 3,450 51 46,550 23.0 30.70 33.27 0.86 Ireland 148 8.32 21.91 20.81 12.5 3,160 Vanuatu 52 18,370 21.0 22.93 13.34 0.87 Czech Republic 20.76 1,670 149 S. Tomé & Principe 8.37 11.65 5.0 25.0 53 Croatia 0.90 9.74 16.23 12,980 150 Kiribati 8.47 20.83 - - 2,950 Slovakia 54 0.91 13.42 21.77 20.0 17,750 8.62 151 Belize 31.23 54.02 12.5 4,346 55 5,630 15.0 9.54 4.33 0.92 Namibia 152 South Sudan 8.66 7.00 - 13.0 970 5,780 56 Thailand 0.94 4.51 11.81 7.0 32.5 6.65 920 Tanzania 153 17.93 8.68 0.94 France 20.0 34.77 33.66 57 42,960 1,350 19.3 26.07 10.32 9.18 Cameroon 154 3.35 4,230 0.95 23.0 Tunisia 58 8.52 16.04 680 0.0 5.43 9.58 Afghanistan 155 6.36 Libya 59 0.98 7,820 0.0 - 3,430 12.0 50.84 27.37 9.58 Guatemala 156 8.10 0.98 Mexico 60 9,870 19.0 13.90 25.82 3,200 Micronesia 9.68 0.0 - 157 20,070 15.0 19.69 16.47 0.98 Trinidad & Tobago 61 Rwanda 700 18.0 15.48 5.85 10.04 158 1.00 2.96 18.27 20.0 3,560 62 Ukraine 10.26 159 Solomon Islands 15.64 15.56 10.0 1,830 Bahamas 15.39 1.00 17.48 63 20,980 - 6.10 160 30.0 670 10.93 Uganda 17.53 1.11 29.25 17.0 Israel 64 32.53 35,320 8.46 890 18.0 21.95 11.40 Benin 161 20,979 65 18.0 1.12 26.52 19.65 Malta Côte d'Ivoire 162 1,450 35.81 14.15 18.0 11.71 14,100 15.0 20.77 13.24 1.13 Seychelles 66 163 26.53 12.04 Papua New Guinea 10.0 2,240 22.47 67 Maldives 1.14 6.08 7.86 6.0 6,410 12.25 5.10 20.17 Gambia 22.3 500 164 68 Jordan 4.96 10.21 40.0 5,160 1.15 790 0.0 - 8.93 13.56 Comoros 165 12.0 10.41 12.67 1.21 Venezuela 69 12,615 10.0 166 Haiti 13.61 9.30 20.37 820 11.94 70 Brazil 1.24 21.19 40.2 11,530 167 470 18.0 11.76 5.43 13.87 Guinea 71 15.81 7.07 1.25 14.0 South Africa 6,800 18.0 39.43 15.04 14.21 Mauritania 168 1,270 17.04 72 Uruguay 1.25 23.36 22.0 16,350 1,050 34.65 13.84 15.82 Senegal 169 23.0 73 Azerbaijan 1.27 8.03 - 18.0 7,590 170 17.0 Mozambique 16.20 600 8.10 19.99 Spain 1.29 74 37.95 21.0 29,440 31.60 171 Burkina Faso 16.71 9.75 26.23 18.0 700 2.57 2,370 5.0 7.96 1.30 Bhutan 75 1,870 172 17.31 Nicaragua 26.98 69.38 15.0 4.93 4,400 10.0 10.45 1.34 Paraguay 76 840 173 Zimbabwe 18.39 12.87 - 20.0 21.30 12.64 77 Panama 1.36 7.0 11,130 Chad 980 18.0 33.61 174 15.75 19.29 Georgia 27.1 12.49 4.47 3,720 1.44 78 700 15.0 31.05 12.01 20.59 Sierra Leone 175 79 8.77 7,240 12.0 17.71 1.45 Botswana 32.85 22.85 12.38 176 Mali 650 18.0 80 23.50 18.96 1.52 St. Kitts and Nevis 14,920 - 177 12.33 30.38 Guinea-Bissau 15.0 550 26.91 81 7,320 19.0 18.96 9.70 1.59 Montenegro 25.31 410 178 Niger 27.75 9.48 - 14,910 82 Chile 1.61 20.03 32.15 19.0 570 18.0 34.16 13.33 28.07 To g o 179 1.53 83 15.0 4.14 1.70 Bangladesh 1,080 180 18.0 20.38 8.00 35.56 Burundi 270 84 Albania 1.71 6.35 14.08 20.0 4,450 Madagascar 41.49 15.21 440 55.80 20.0 181 85 Myanmar 1.76 1.87 7.33 0.0 1,270 320 19.0 11.50 - 43.13 Central African Rep. 182 86 Pakistan 2.09 7.04 33.5 1,400 1.79 20.68 14.08 45.66 Liberia 183 370 14.0 15.88 87 1.80 5.45 10.0 3,630 Indonesia 184 250 26.5 37.46 10.95 52.55 Malawi 88 0.0 19.49 6,500 10.05 1.86 Iraq Somalia** - 2.91 - 10.0 - 89 10,030 10.0 - 15.73 1.88 Lebanon Congo (Dem. Rep.)** - - 13.0 - 9.77 3,050 90 Egypt 1.90 4.83 16.78 15.0 Djibouti** - 11.26 18.99 7.0 - 91 27.0 13,340 1.93 Hungary 42.06 21.48 - 14.04 - - - Timor-Leste** 13.10 16.5 8.54 1.99 Jamaica 92 5,150 - 19.25 - Nauru** - - Colombia 93 2.01 13.33 29.68 20.0 7,970 - 24.07 27.96 0.0 - San Marino** 10,830 34.46 18.15 2.01 Turkey 94 43.0 - 8.0 - 30.13 - Liechtenstein** 9,950 8.0 29.31 17.00 Suriname 95 2.05 40.03 - 20.0 - - Monaco** 96 9,720 - Gabon 26.74 16.66 2.06 - - - 95.32 - Syria** 2.07 Peru 6,360 18.0 21.30 10.97 97 Note: * Data correspond to the GNI per capita (Atlas method) in 2014 or latest available year adjusted with international inflation rates.** Country not ranked because data on GNI p.c. are not available. Source: ITU. GNI p.c. and PPP$ values are based on World Bank data. Measuring the Information Society Report 2016 107

124 bile numbers ported, developed (left) and developing (right) countries, 2014 Mo Chart 4.2: Note: * 2013 data. Source: ITU. For instance, in Namibia, MTC offers a weekly 300 SMS and 50 on-net minutes; extra calls are 0.31 per minute, whereas the basic prepaid package including 50 minutes, 150 charged at USD 1.1, while the cost of SMS and 50 0.51 per minute. pay-per-use tariff costs USD MB for USD consuming the same amount of voice and SMS on Operators are increasingly offering these new a pay-per-use basis would be USD 6.6. prepaid plan arrangements with the dual objective Operators may also offer free minutes and SMS of fostering customer loyalty and ensuring more in exchange for large prepaid refills, which in stable revenue streams from their prepaid practice has the same effect on prices as a prepaid customer base. As a result, prepaid subscriptions package. For example, in Sweden, Telia offers 100 are acquiring some of the features of postpaid subscriptions, such as the requirement of a GB to be used within 30 minutes, 500 SMS and 0.5 days with each prepaid top-up of USD 11.7, which minimum expenditure per month. However, some is much cheaper than the cost of consuming the of the fundamental characteristics of prepaid same minutes and SMS on a pay-per-use basis subscriptions that make them attractive to lower- (USD 49.1). income groups are retained: prepaid packages do not require a commitment period and the A third variant is the hybrid prepaid plans offered customer can opt out without a penalty. by some operators that include a package of minutes and SMS as well as preferential pay-per- Prepaid packages with a significantly lower cost use rates for any extra consumption. For instance, than pay-per-use plans are driving down prices the incumbent operator in Venezuela, Movilnet, in developed countries such as Albania, Bulgaria, offers the “Optimo” prepaid plan, which requires Croatia, Greece, Italy, Latvia, New Zealand, 8.5 and includes a monthly payment of USD Romania, the Russian Federation, Slovenia and Measuring the Information Society Report 2016 108

125 Chapter 4 Sweden. In several cases, voice and SMS are Regional analysis of mobile-cellular prices packaged with data services (e.g. Mobitel’s prepaid packages in Bulgaria and Vodafone’s prepaid A regional analysis of mobile-cellular prices reveals packages in New Zealand). In a few countries, some differences across and within regions prepaid packages offer unlimited local SMS (e.g. 4.3): (Chart Text packages in Canada) and voice & Rogers Talk calls (e.g. Tele 2 in Latvia). These examples confirm Africa: the trend towards the commoditization of voice and SMS, and suggest that data is becoming the Mobile-cellular prices range from USD 25 3 to USD main element in determining mobile prices in per month in African countries, and the average developed countries. 10 per month) is similar to price in the region (USD that in Asia and the Pacific. In the developing world, the situation is slightly different because prepaid packages tend to have In PPP (and in GNI p. c.) terms, the range is wider 11 and they are seldom the shorter validity periods and the average (PPP$ 24 per month) is the cheapest option for a continuous monthly usage. second highest of all regions, below only that of There are, however, some exceptions, such as the Americas, reflecting the lower incomes in Brazil, where operator Vivo offers the Smart Vivo the region. Kenya and Ethiopia stand out as the Controle plans starting at USD 10.5 per month and countries with the lowest mobile-cellular prices including packages of on-net voice and SMS, as in PPP terms in the region (Table 4.3). These well as data. In Morocco, the incumbent operator countries have very different market conditions: Maroc Telecom offers the “Jawal Pass” including Kenya is a vibrant mobile market with three a large bundle of voice minutes, SMS and data for operators (the incumbent and two transnational USD 10.2 per month. These examples suggest that, per cent mobile-cellular operators) and 81 if operators in most developing countries were penetration; Ethiopia is an LDC at the early stages to offer prepaid monthly bundles, prices could be per cent of mobile-cellular development (43 further reduced, particularly for those customers penetration) and the incumbent operator, Ethio with a continuous monthly usage of mobile- Telecom, retains the monopoly in the mobile cellular services. market. The other two African countries with mobile-cellular prices below PPP$ 10 are Namibia Indeed, traffic data show that the average voice and Mauritius. usage per subscription is above 60 minutes per month in most countries, including in the An analysis of prices relative to GNI p.c. levels developing world (Table 4.2). Moreover, mobile shows that in more than two thirds of African 12 voice traffic is increasing in most economies. countries the cost of the mobile-cellular basket Although the number of SMS sent per subscription per cent of GNI p.c., and represents more than 5 13 in many countries is decreasing globally, the service thus remains unaffordable for large the average is still above 20 SMS per month segments of the population. This is particularly (Table 4.2). These figures suggest that prepaid the case in African LDCs, in all of which the customers in many developed and developing mobile-cellular basket corresponds to more than countries could benefit from lower mobile- 5 per cent of GNI p.c., except oil-rich Angola and cellular prices by subscribing to prepaid packages Equatorial Guinea. tailored to their monthly consumption. Taking into account that the availability and uptake of mobile Arab States: bundled packages in developing countries is not as widespread as in developed countries, there is Mobile-cellular services cost between USD 3 and an opportunity for further mobile-cellular price 20 per month in all Arab States except Syria, USD reductions in the developing world. 14 Excluding where prices are significantly higher. 10 per month) is Syria, the regional average (USD comparable to that of other developing regions such as Africa and Asia and the Pacific. Measuring the Information Society Report 2016 109

126 Table 4.2: Do mestic mobile minutes (left) and SMS (right), selected economies, 2014 and 2013 Average SMS sent Average domestic mobile minutes per subscription per month per mobile subscription per month Economy Economy Difference Difference 2014 2013 2013 2014 2013-2014 2013-2014 12 397 409 238 333 Iran (I.R.) 95 Canada Venezuela 314 304 Colombia 365 325 10 40 Namibia 317 United States 386 305 305 -70 0 268 257 Jordan Jamaica 252 276 -11 -23 Romania 254 251 France 231 243 -3 12 Turkey 235 196 New Zealand* 238 224 14 -39 237 229 Bahrain Pakistan 190 206 -16 -8 186 168 Portugal Pakistan 228 194 34 -18 Latvia 186 222 Belgium 163 176 36 -12 227 214 Cyprus Turkey 144 212 -13 -68 Costa Rica 188 210 Armenia 142 227 22 -85 134 Lithuania 129 Algeria 194 189 6 5 415 129 Ireland 163 190 China* -226 -34 188 France 247 Paraguay 129 171 -118 17 116 Jamaica 192 137 United Kingdom 188 -4 -20 Argentina TFYR Macedonia 172 186 104 174 -70 15 112 101 Denmark Korea (Rep.) 182 167 -11 15 165 Georgia 97 Sweden* 180 93 3 15 Serbia 175 98 92 Norway 179 4 -6 170 178 Ireland Malaysia 91 149 9 -58 Cyprus 176 163 Greece 90 101 -11 13 Saudi Arabia 191 175 Trinidad & Tobago 55 89 -15 35 174 168 Belarus Sweden* 88 116 5 -28 Latvia Iceland 173 177 87 77 -4 10 168 India Luxembourg 87 169 97 1 -10 Albania 166 150 86 86 Norway 0 16 86 145 163 Portugal Slovenia 77 17 8 162 99 200 Uruguay 83 Honduras -16 -38 Bangladesh 146 81 Korea (Rep.) 161 81 15 -1 Malta 159 Croatia 78 88 145 -10 13 155 Azerbaijan 193 76 77 Poland -37 1 153 Lithuania 61 139 Romania 71 9 14 52 68 Kenya 152 Tunisia 152 16 0 Croatia 151 149 Russian Federation 60 62 3 -2 146 United Kingdom 76 60 142 Singapore -16 4 144 Denmark 58 China* 37 137 7 20 49 Czech Republic 142 140 56 Slovenia -6 2 Trinidad & Tobago Italy 126 140 49 50 -1 15 59 48 Mauritius Bulgaria 139 129 -11 10 47 50 Iceland Georgia 134 132 2 -3 Hungary 132 133 Albania 45 38 2 7 115 132 Kazakhstan Italy 40 67 16 -27 Bhutan Austria 131 133 37 8 -1 29 37 123 Serbia Morocco 131 22 8 15 118 Bolivia 37 43 Spain 131 -6 13 Finland 128 50 36 130 Slovakia 2 -14 126 125 Estonia 35 37 Bosnia and Herzegovina -2 1 126 Germany 32 28 Slovakia 88 37 4 123 Moldova 38 Greece 31 113 10 -7 Czech Republic 37 29 Austria 121 108 -9 13 Ecuador 28 28 111 120 Poland 1 9 Switzerland 32 26 Mexico 118 112 -6 6 Moldova 26 17 Venezuela 118 112 9 6 28 25 Seychelles Guatemala 112 87 -2 25 Mauritius 25 127 109 Russian Federation 32 -18 -6 25 Azerbaijan 25 Tanzania 109 106 1 3 34 23 95 107 Malta India -11 11 19 Oman 29 103 106 Saudi Arabia -10 4 17 Namibia 104 86 19 Estonia 2 18 103 Germany 19 103 32 Costa Rica -13 0 17 90 99 Cuba 18 Hong Kong, China 9 1 Senegal 96 78 Panama 17 32 -15 18 99 Madagascar 96 United Arab Emirates 16 16 1 -3 Tunisia 20 Belgium 94 94 15 -4 0 Paraguay 98 21 94 Jordan 15 -4 -6 South Africa 15 6 97 93 Uruguay 9 -4 Andorra Morocco 14 90 77 15 12 0 14 9 Thailand 88 Oman 99 5 -11 13 87 74 New Zealand* 13 Hungary 13 0 80 TFYR Macedonia 12 Nigeria 86 15 -3 5 Hong Kong, China 15 Zambia 84 12 61 23 -3 Luxembourg 83 Colombia 11 26 85 -2 -14 39 Thailand 81 8 11 Chad 3 42 Switzerland 81 88 Senegal 10 14 -4 -7 80 United Arab Emirates South Africa 10 40 12 -2 40 10 9 Zimbabwe Rwanda 82 80 -3 1 9 9 Armenia Ecuador 78 80 1 -2 Bangladesh 78 76 Kenya 7 9 2 -2 85 75 9 St. Lucia Mali 2 -10 7 6 71 70 Panama 9 Sudan 2 -1 72 Côte d'Ivoire Lao P.D.R. 68 9 4 -4 4 8 9 61 Bosnia and Herzegovina 63 Bulgaria 0 -2 St. Lucia 42 8 60 Seychelles 8 0 18 10 8 Macao, China Cuba 58 53 -2 5 Andorra 9 7 Zambia 52 51 -1 2 Bahrain 13 7 Chad 41 39 2 -6 32 35 Bolivia Algeria 7 6 3 0 8 7 Guatemala 33 34 Benin 0 -1 33 32 Congo (Dem. Rep.) 7 6 Kazakhstan -1 1 Burkina Faso 6 7 30 32 Burkina Faso -1 3 27 Egypt Dominican Rep. 31 6 5 1 3 Zimbabwe 29 32 Kyrgyzstan 5 8 -2 -3 18 9 4 4 Lao P.D.R. Dominican Rep. 9 0 6 16 16 Mali 4 Spain -2 0 2 Madagascar 2 Congo (Dem. Rep.) 3 12 0 -9 Nigeria 1 1 0 Note: * 2014 and 2012 data. Source: ITU. Measuring the Information Society Report 2016 110

127 Chapter 4 Mo Chart 4.3: bile-cellular prices as a percentage of GNI p.c. (top), in PPP$ (middle) and in USD (bottom) by region, 2015 Note: Each horizontal dash represents the price in one country in the region. The yellow marks signal the regional average. Source: ITU. Measuring the Information Society Report 2016 111

128 Table 4.3: To p five countries with the cheapest mobile-cellular services in each region, PPP$, 2015 Europe Asia & Pacific The Americas PPP$ PPP$ Country PPP$ Country Country Estonia 4.67 5.51 Costa Rica 2.45 Sri Lanka 10.41 Lithuania 5.59 Bangladesh 4.14 Venezuela 6.84 Latvia Iran (I.R.) 5.43 Paraguay 10.45 6.97 Austria China 6.63 Jamaica 13.10 13.90 8.56 Pakistan 7.04 Mexico Cyprus Africa CIS Arab States Country PPP$ Country PPP$ Country PPP$ 5.96 Kyrgyzstan 10.56 Kenya 7.96 Sudan 8.52 12.49 Georgia Ethiopia 8.70 Tunisia United Arab Emirates 9.15 Russian Federation 13.46 Mauritius 9.18 Kazakhstan 14.02 10.21 Namibia 9.54 Jordan Egypt 16.78 Armenia 17.26 Nigeria 10.17 Note: Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. Source: ITU. In PPP terms, both the price range (PPP$ 6 – Asia and the Pacific: 40) and the average (PPP$ 20) are twice as PPP$ The region has the widest range of mobile-cellular high as the USD values. Sudan stands out as the prices, varying from USD 1 per month in Sri Lanka country with the lowest mobile-cellular prices in the region (PPP$ to USD 6 per month). Despite being an 38 per month in Tuvalu. 15 LDC, Sudan has a very competitive mobile market In PPP terms, Asia and the Pacific has the lowest and this is also reflected in the low handset-based average of all regions, highlighting that despite the mobile-broadband prices (see Section 4.4). The diversity of prices mobile-cellular services are on other Arab States where mobile-cellular services 10 per month are Tunisia and cost less than PPP$ average the least expensive of all regions. Indeed, the United Arab Emirates. the region is home to the lowest mobile-cellular prices worldwide, Sri Lanka and Bangladesh 5 per month. Prices relative to GNI p.c. levels are moderately standing out with prices below PPP$ Other Asian countries that display remarkably low affordable in most countries in the region, and the mobile-cellular prices include the Islamic Republic United Arab Emirates stands out as being among 5.4), China (PPP$ of Iran (PPP$ 6.6) and Pakistan the top five countries with the most affordable 7). (PPP$ mobile-cellular prices worldwide (Table 4.1). The Arab States in which prices correspond to more When the GNI p.c. of each country is taken than 5 per cent of GNI p.c. include Comoros, into consideration, Macao (China) has the most Mauritania and Yemen, which are LDCs with the affordable mobile-cellular services worldwide, lowest GNI p.c. levels in the region. Comoros saw representing 0.09 per cent of GNI p.c., and a significant decrease in mobile-cellular prices in per cent GNI p.c.) and Hong Kong Singapore (0.17 2015 (34 per cent in USD terms) in parallel with a (China) (0.18 per cent GNI p.c.) are also among the series of measures undertaken to pave the way global top five. The mobile markets in these three for some competition in the country’s mobile 16 economies are characterized by the presence of whereas in Mauritania and Yemen monopoly, prices remained the same or even increased. at least three strong operators, despite their small Although Palestine has higher GNI p.c. levels population size, and very high mobile-cellular than the region’s LDCs, mobile-cellular prices are penetration rates (more than 140 subscriptions relatively high compared with most Arab States per 100 inhabitants). Considering that mobile number portability is only widely employed 14.5 per month) and, as a result, the cost of (USD in Hong Kong (China), competition is probably per cent of GNI p.c. the service represents 5.7 spurred by multi-SIM ownership, thus fostering low mobile-cellular prices. Overall, mobile-cellular Measuring the Information Society Report 2016 112

129 Chapter 4 Moreover, the Baltic States (Estonia, Lithuania per cent of GNI p.c. in prices represent less than 5 and Latvia) stand out in the global comparison as most countries in Asia and the Pacific, including having some of the lowest mobile-cellular prices in several LDCs, such as Bangladesh, Bhutan, Lao PPP terms worldwide. PDR, Myanmar, Nepal and Samoa. In other Asian LDCs, such as Afghanistan and Cambodia, although When the GNI p.c. of each country is taken into per cent prices in terms of GNI p.c. are above the 5 account in order to assess the affordability of threshold, the low mobile-cellular prices achieved mobile-cellular services, the average price in terms 5.4, respectively) confirm that (USD 6.8 and USD of GNI p.c. in Europe is the lowest of all regions. low mobile-cellular prices are possible even in low- Most European countries have mobile-cellular income countries. prices corresponding to less than 1 per cent of GNI p.c., and all of them are below the 5 per cent of Commonwealth of Independent States (CIS): GNI p.c. threshold. Albania and Bulgaria were the last European countries to reach this milestone Mobile-cellular prices range in the CIS from USD 3 thanks to the price reductions achieved through to USD 8 per month. Indeed, CIS is the most prepaid bundled packages. As in the past, Austria homogeneous region when it comes to mobile- features among the global top five countries with cellular prices, which is explained by the relatively the most affordable mobile-cellular prices, and 17 small number of countries the region comprises the country also stands out for having some of the and by the prevalence of transnational operators world’s most affordable mobile-broadband prices such as MTS and VimpelCom, which offer their (see Section 4.4). services in several CIS countries. In PPP terms, prices are significantly higher, with Americas: an average of PPP$ 15 per month, although Mobile-cellular prices range from USD 36 4 to USD the CIS still has on average the second lowest in the Americas region, and the average is the PPP-adjusted prices of all regions, after Asia 18. highest of all regions at USD and the Pacific. No CIS country stands out for having particularly low prices in the global PPP In PPP terms the range is wider, and the average comparison, the lowest mobile-cellular prices being those offered in Kyrgyzstan (PPP$ 10.5) and 27. The remains the highest of all regions at PPP$ 18 Georgia (PPP$ high average price in the Americas is explained, 12.5). on the one hand, by the relatively high pay- A regional comparison of prices relative to GNI p.c. per-use prices in countries such as Argentina, levels shows that the CIS is the second region with Belize, Guatemala and Nicaragua, where prepaid the most affordable mobile-cellular services, after packages of bundled services with long validity Europe. The Russian Federation and Kazakhstan periods are not available. On the other hand, in are the CIS countries with the most affordable high-income countries such as Canada and the United States, good value-for-money offers are prices, representing 0.34 and 0.6 per cent of GNI typically all-inclusive family plans (voice, SMS and p.c., respectively. All CIS countries have mobile- data), whereas individual prepaid packages have cellular prices corresponding to less than 5 per a higher price. In the global comparison, Costa cent of GNI p.c., including Tajikistan, the last CIS Rica is the only country in the Americas that country to achieve this milestone in 2015. stands out for the affordability of mobile-cellular 5.5). Indeed, the service in Costa Rica prices (PPP$ Europe: costs almost half as much as in Venezuela and Paraguay, the second and third countries with the 3 and Mobile-cellular prices vary between USD lowest mobile-cellular prices in PPP terms in the 40 per month in Europe, with an average of USD Americas, respectively. 18 per month, the highest after the Americas. USD Variations are narrower when considering the An analysis considering purchasing power parity GNI p.c. of each country, and most countries in factors reveals that prices in Europe are on the region have prices that represent less than average similar to those in the Arab States and 5 per cent of GNI p.c. This suggests that, although Africa, despite significant income differences. there is room for further reductions of mobile- Measuring the Information Society Report 2016 113

130 cellular prices in most countries in the Americas, 2010). The technical and commercial implications the service is already relatively affordable in most of dynamic discounting are manifold and concern countries of the region. The countries where operators, customers and regulators. Dynamic mobile-cellular prices correspond to more than discounting: 5 per cent of GNI p.c. remain unchanged in the Enables operators to distribute traffic more Americas since 2014, namely Honduras, Belize, 1. Guatemala, Haiti and Nicaragua. In addition evenly location-wise and time-wise and thus optimize the utilization of existing cell- to some specific initiatives to reduce prices in towers. As a result, quality of service may be these countries, such as the dynamic discounts improved. offered by the operator Tigo in Guatemala, some additional and sustained private-led initiatives Provides an opportunity for operators to 2. or regulatory and policy interventions would be required to achieve lower mobile-cellular prices increase their market share and/or customer in these countries, particularly in Belize and Haiti, base without the need for upgrading 19 the countries with the lowest mobile-cellular the existing network in the short term. penetration in the region, together with Cuba. Therefore, it allows operators to increase revenue generation at very low incremental cost. Dynamic tariffs in mobile-cellular services Helps make mobile-cellular services affordable 3. for the lowest-income segments of the process of offering Dynamic tariffing refers to the ‘ population for which standard rates may not automated variable pricing of mobile services be affordable. Hence, it may contribute to ’ based on real-time analysis of network utilization increasing mobile-cellular uptake. (Smyk, 2011). An optimization and analytics system (Figure 4.2) analyses network traffic and computes Allows operators to foster customer loyalty, 4. location-specific discounts at any given time which and existing customers may enjoy lower rates are then broadcast to customers (Piscataway, N.J., in selected time and geographical zones. Figure 4.2: Dy namic discounting systems Source: Digitata (2016). Measuring the Information Society Report 2016 114

131 Chapter 4 5. Makes mobile-cellular pricing schemes more 2011), while increasing operator revenues at an complex and, as a result, may complicate almost zero incremental cost. However, caution operators’ accountability vis-à-vis customers must be exercised when analysing the overall and regulators. impact of dynamic discounting on account of the time and geographical constraints it imposes Does not allow customers who are benefitting 6. on customers. Moreover, dynamic discounting from dynamic discounts to know in advance makes mobile pricing schemes more complex: how much they are going to pay for a specific customers may not always understand the price consumption in a given month. they are being charged, and they cannot know in advance how much a given consumption will cost Operators offer dynamic discounts in markets each month. These factors may impair customers’ with a majority of prepaid customers who are control over their mobile-cellular spending. very price sensitive (Piscataway, 2010). Discounts per cent on call and SMS may be as much as 99 Fixed-broadband prices 4.3 prices, depending on the time and the location of the customer. In 2007, MTN Swaziland was the The price of fixed-broadband services has dropped first operator to launch dynamic discounting. To substantially and become much more affordable test this method, MTN first ran a five-week pilot since 2008, when ITU first started collecting per cent for which offered discounts of up to 99 comprehensive price data for this service. Globally, on-net calls. As a result, peak traffic was reduced the price of a basic fixed-broadband connection per cent on over-utilized cell towers, while by 14 25 80 in 2008, to USD has fallen from around USD overall network utilization increased by 11 per 20 in 2015, corresponding to a decrease in price cent. Network utilization was more evenly relative to average GNI p.c. from over 90 per cent spread over time and location, and congestion er cent. The price drop resulted mainly from p to 14 was significantly reduced (Digitata, 2016). MTN a substantial decrease in developing countries, Swaziland then launched the dynamic discounting 21 200 to USD where prices fell from around USD 26. Following offer, MTN Zone, in August 2007. this initial deployment, other operators in many Although the USD price of the service is countries have implemented dynamic discounting, approaching similar levels across both developed particularly in Africa but also in Asia and the and developing regions, the service nonetheless Americas (Figure 4.3). remains unaffordable for large parts of the population living the world’s LDCs. These The benefits of dynamic discounting in terms discrepancies are also highlighted in terms of of expanding mobile-cellular penetration and PPP-adjusted prices of the service, with PPP$ fostering mobile usage are well illustrated by the prices almost twice as high in the LDCs as in the case of MTN Uganda. In July 2008, the operator developing countries (Chart 4.4). launched MTN Zone in the country, and within three months more than 2 million customers had per cent of its subscribed to this offer, i.e. about 60 Recent trends – comparing developed, customer base (Digitata, 2016). MTN Uganda has also noted a 70 per cent increase in voice traffic developing and least developed countries since the launch of MTN Zone. The estimated average discount offered was 45-50 per cent Contrary to previous years, 2014 fixed-broadband 22 during the day and 95 per cent at night. price averages showed that the service had become less affordable. However, this price hike In conclusion, dynamic discounting may benefit was due mainly to increases in a small number of both customers and operators, insofar as it helps countries and stagnating or zero price drops in 23 to distribute network utilization more evenly, many others. In 2015, fixed-broadband services thus providing a better quality of service. In saw a renewed and significant drop in the price for addition, lower prices may make access to mobile- the service. A price comparison in terms of USD cellular services affordable for the lowest-income and PPP-adjusted prices and as a percentage of segments of the population (Africa Telecoms, GNI p.c. highlights the following recent trends: Measuring the Information Society Report 2016 115

132 Dy namic discounting schemes applied in selected countries Figure 4.3: Note: The maximum saving per month is calculated as the difference between the plan selected for the mobile-cellular sub-basket and the maximum discounted rates, considering the call distribution of the mobile-cellular sub-basket. It should be noted that the discounted rates are often time and geographically limited. As a result, the maximum discounted rates may not be available for the call distribution specified in the mobile-cellular sub-bas - ket (i.e. the number of peak, off-peak and weekend minutes) and may be concentrated in a few days, rather than having a 30-day validity as in the plans considered in the ITU price data collection. Source: ITU. Measuring the Information Society Report 2016 116

133 Chapter 4 To monitor fixed-broadband prices, ITU uses the fixed-broadband sub-basket , which refers to the price of a monthly subscription to an entry-level fixed-broadband plan. It is calculated as a percentage of a country’s average monthly GNI per capita, and also presented in USD and PPP$. For comparability reasons, the fixed-broadband sub-basket is based on a monthly data usage of (a minimum of) 1 Gigabyte (GB). For plans that limit the monthly amount of data transferred by including data volume caps below 1 GB, the cost for the additional bytes is added to the sub- basket. The minimum speed of a broadband connection is 256 kbit/s. Percentage of GNI p.c. respectively). At the global level, PPP- adjusted prices fell by about 10 per cent • At end 2015, fixed-broadband prices from 2014 to 2015, the same decrease as were more affordable than at end 2014 in in developing countries (Chart 4.4, middle). both developed and developing regions (Chart 4.4, top). At 1.2 per cent of GNI USD prices p.c., the service remains very affordable in developed countries, but is still relatively Between 2014 and 2015, USD prices for • expensive in developing countries, where fixed-broadband services decreased in the monthly subscription to an entry-level developed and developing regions, and per service corresponded to close to 20 most strongly in the LDCs, thus reducing cent of GNI p.c. differences in the absolute USD price. By 2015, the price of a fixed-broadband • Globally, the average price of an entry- 26 service stood at USD 23 and USD level fixed-broadband subscription as a in developed and developing regions, percentage of GNI p.c. fell from close to 38 in the respectively, compared to USD 21 per cent in 2014 to 14 per cent in 2015. LDCs. Differences in terms of USD are In LDCs and other developing countries, relatively small in comparison with PPP$ prices dropped by one third, while the and GNI p.c. prices, and would be even prices in developed countries decreased smaller in the case of LDCs if the two at a lower rate. However, by end 2015 an outliers were not included in the average entry-level fixed-broadband subscription (Chart 4.4, bottom). per cent of still represented close to 61 GNI p.c. in LDCs, making it unaffordable for Global averages provide an important indication of a large portion of the population. trends over time, and are useful for understanding broad differences between geographic regions and those at different stages of development. Purchasing power parity At the same time, averages tend to hide major Purchasing power parity prices confirm differences between countries within a given • that fixed-broadband services remain region, in particular for very diverse regions unaffordable in the world’s LDCs. in terms of income levels and development. Although PPP-adjusted prices in the LDCs In some cases, and in regions with a relatively 130 in 2014 dropped from a high of PPP$ small number of countries, just a few outliers 100 in 2015, the service still to PPP$ will have a disproportionately large impact on remained on average more expensive regional averages, especially since price data than in 2013. However, the average for are not capped and remain exorbitantly high in LDCs was significantly influenced by the a few economies. These outliers are often from very high prices in two countries, Rwanda within the group of low-income economies within and Uganda. Considering only the other a region, where fixed-broadband services are LDCs included in the price comparison, not intended for residential users, and where 60, and the average for 2015 was PPP$ broadband penetration rates remain particularly there was a slight but sustained decrease low. between 2013 and 2015 (8 per cent, and 4 Measuring the Information Society Report 2016 117

134 Chart 4.4: Fi xed-broadband sub-basket, as a percentage of GNI p.c. (top), in PPP$ (middle) and in USD (bottom), 2008-2015 Note: Simple averages. Based on 144 economies for which data on fixed-broadband prices were available for the years 2008-2015. It should be noted that the 2014 price hike in the LDCs is partially the result of very substantial price increases in only two countries (Uganda and Rwanda), which had a sizeable impact on the LDC average (especially because complete price data for the period 2008-2015 are only available for 25 LDCs). The dotted line in the chart shows the evolution of the average in LDCs without these two countries. Source: ITU. Measuring the Information Society Report 2016 118

135 Chapter 4 of households and even of most (at least smaller) Benchmarking countries 300 in businesses. Services are sold at over USD Uganda, Chad and the Central African Republic, Country price data rank countries based on the and are also very expensive in some of the small affordability of fixed-broadband services, but also island developing states (SIDS), such as the show USD and PPP$ values and indicate price 188). Solomon Islands (USD 275) and Kiribati (USD relative to their GNI p.c. There is a strong link Cuba (USD 180) also stands out for its very high between income levels and the affordability of fixed-broadband prices. fixed-broadband services, and the service tends to be more affordable in high-income economies and less affordable in low-income economies. This is not surprising, given that that affordability of fixed- Cheaper and faster – how entry-level broadband is calculated on the basis of countries’ broadband speeds are evolving GNI p.c. levels. Some countries, however, stand out in that they offer relatively affordable fixed- Since 2008, so as to be able to make comparisons broadband services despite relatively low incomes. between countries, ITU has collected prices A comparison of USD prices points to these for the so-called entry-level fixed-broadband positive outliers (Table 4.4): . This refers to an Internet connection of service GB of a minimum of 256 kbit/s, with at least 1 In most countries of the world, fixed-broadband data included, a benchmark that has remained services cost between USD 10 and 40. A number unchanged. However, a comparison of the speeds of developing countries offer lower prices, though, of entry-level fixed-broadband packages highlights making the service relatively affordable, in that the minimum speeds on offer have risen particular given their relatively low income levels. considerably over the last eight years. While Countries that offer fixed-broadband services at in 2008 only about 30 per cent of all countries 10 include Mauritius (USD 2.9), the below USD Mbit/s, by offered entry-level speeds above 1 6.6), the Islamic Republic Russian Federation (USD per cent of countries offered 2015 close to 80 2.7), Brazil of Iran (USD 3.8), but also Ukraine (USD entry-level speeds of 1 Mbit/s or above. Indeed, 8.9) and Turkey (USD 8.8). These economies (USD by 2015, not a single developed country offered rank among the top 40 countries (i.e. in the top a connection with speeds below 1 Mbit/s and the quartile) in terms of the most affordable fixed- majority of plans were based on advertised speeds broadband sub-basket. Mauritius in particular of above 10 Mbit/s. This indicates that while the stands out for very low and affordable prices: this price of connections has decreased, speeds have African country ranks fourth in the sub-basket and, increased, on average, but not equally for all at USD 2.9, has the third lowest USD prices in the subscribers (Chart 4.5). world. Higher speeds reflect changes in the types of Other countries that stand out for having low services and applications that Internet users absolute fixed-broadband prices are Belarus access and providers offer, and which result in an 4.2) and Sri Lanka (USD 7.2), Tunisia (USD (USD 4.1), increase in data traffic. At the same time, speeds 2.8) and Bangladesh but also Viet Nam (USD have not increased equally in all countries and 4.4). (USD regions of the world, and developing countries are only gradually upgrading broadband infrastructure Economies with relatively high incomes but also to deliver higher speeds. By 2012, over 50 per expensive high-speed Internet services are Canada cent of all countries were still offering services at (USD 49.9) and New Zealand 49.4), Ireland (USD speeds below 1 Mbit/s, and 10 Mbit/s remained (USD 45.3), as well as Hong Kong (China), although the exception for entry-level fixed-broadband 51 subscribers in Hong Kong (China) have for USD packages. The distribution of offers between four Mbit/s Internet access through an advertised 100 speed categories is shown in Chart 4.5, which connection. reveals that in 2015 more than 50 per cent of countries continue to offer speeds of 2 Mbit/s or The most expensive fixed-broadband connections less, and in LDCs the large majority of entry-level are on sale in some of the poorest countries in the plans are still offering speeds of below 1 Mbit/s. world, where the service is clearly out of the reach Measuring the Information Society Report 2016 119

136 Table 4.4: xed-broadband sub-basket, 2015 Fi Fixed-broadband Fixed-broadband Tax Tax Speed Speed sub-basket sub-basket Cap per Cap per rate rate GNI GNI in in Economy month Rank Economy month Rank inclu- inclu- p.c., p.c., as % of as % of Mbit Mbit in GB in GB ded ded USD* USD* PPP$ GNI GNI PPP$ USD USD /s /s (%) (%) p.c. p.c. 3.84 11.90 33.23 10.00 Unlimited 18.0 3,720 1 Kuwait 0.5 10.54 Unlimited 96 49,300 0.26 Georgia 16.95 0.0 20.69 6.0 2 Macao, China 4.00 0.27 17.28 22.40 26.76 15 1.5 Unlimited 0.0 76,270 3.87 Maldives 97 6,410 Thailand 3.89 18.71 48.96 10.00 3 Unlimited United States 55,200 5,780 0.35 16.32 16.32 2 Unlimited 8.9 98 7.0 23.98 4.00 Unlimited 20.0 3,070 0.36 0.5 2 15.0 9,630 Mauritius 2.85 5.19 99 Morocco 3.96 10.14 4 100 5 0.42 15.28 12.60 17 10 20.0 43,430 Ecuador 3.97 20.16 34.67 3.00 Unlimited 12.0 6,090 United Kingdom Dominican Rep. 3.98 20.05 40.95 1.00 Unlimited 30.0 6,040 6 Andorra 0.46 17.39 - 0.5 2 4.5 45,033 101 102 7 40.80 32.83 6 Unlimited 25.0 103,630 2,370 0.47 Bhutan 4.14 8.17 25.31 2.00 4 5.0 Norway 8.0 Unlimited 5 22.43 35.33 0.48 Switzerland 8 28.30 16.82 4.14 Fiji 103 10.00 15.0 4,870 10.00 88,032 Grenada 4.30 28.33 37.96 2.00 Unlimited 15.0 7,910 0.51 104 18.02 19.12 Japan 12 900 8.0 42,000 9 75,990 10 2 8 30.33 32.72 105 Moldova 4.41 9.41 29.36 50.00 Unlimited 16.7 2,560 Luxembourg 0.52 17.0 20.0 4.47 2.00 Unlimited 8.0 9,950 Suriname 106 49,670 63.90 Unlimited 8 24.46 23.46 0.57 Austria 11 37.06 107 Botswana 4.75 28.65 57.86 0.50 Unlimited 12.0 7,240 Russian 13,220 12 18.0 0.60 6.56 23.93 30 Unlimited Federation Bangladesh 4.92 4.43 11.96 0.25 4 15.0 1,080 108 Singapore 13 0.63 29.02 33.18 200 Unlimited 7.0 55,150 Antigua & 15.0 109 4.96 54.94 67.20 1.00 Unlimited 13,300 3 14 Barbuda Iran (I.R.) 0.64 0.25 11.13 3.79 9.0 7,113 42,960 1,570 France 0.65 23.29 24.06 15 Unlimited 20.0 15 India 110 5.11 6.69 23.39 2.00 1.5 14.4 10.0 64,540 16 Australia 0.70 37.53 32.18 8 100 6,500 111 Iraq 5.22 28.27 54.81 0.25 Unlimited 0.0 5 24.0 46,304 17 Iceland 0.74 28.43 25.60 12 5.46 112 Jamaica 5,150 16.5 Unlimited 1.00 35.93 23.42 Unlimited 26.95 10 32.20 5.58 Dominica Unlimited 48,420 24.0 15.0 113 6,930 2.00 18 43.30 Finland 0.74 29.83 Pakistan 19 38.50 31.10 25 Unlimited 25.0 61,310 114 0.75 5.70 6.64 22.33 1.00 10 14.0 1,400 Denmark 0.50 20 Bahrain 0.76 13.30 22.42 2 20 0.0 21,039 115 Namibia 5.83 27.35 60.30 Unlimited 0.0 5,630 21.0 116 5.98 36.20 43.95 2.00 Unlimited 15.0 47,260 St. Lucia 100 50 31.10 30.50 0.77 Belgium 21 7,260 0.25 Unlimited 13.0 2,870 26,370 19.0 117 Bolivia 5.99 Unlimited 3 22.16 17.79 0.81 Cyprus 22 14.33 30.65 El Salvador 118 0.83 64.01 85.76 1 Unlimited 0.0 92,200 Qatar 6.50 21.23 40.61 2.00 Unlimited 18.0 3,920 23 36.04 119 24 Netherlands 0.83 51,890 21.0 Unlimited 20 36.26 Palestine 6.73 17.16 26.16 4.00 Unlimited 15.0 3,060 43.75 61,610 25.0 Unlimited 100 40.17 0.85 Sweden 25 120 Yemen 6.77 7.33 - 0.25 4 5.0 1,299 26 Guatemala 36.15 10 1 20.46 8.70 0.88 Kazakhstan 1.00 121 6.81 Unlimited 19.46 12.0 3,430 11,850 12.0 Brunei Darussalam 27 37,663 1,330 5.0 1 1.00 22.99 7.76 7.00 Lesotho 122 0.0 45 1 47.82 28.37 0.90 123 13,690 Jordan 23.0 Unlimited 7.04 30.28 62.38 1.00 10 8.0 5,160 10 20.61 10.35 0.91 Poland 28 9,520 24.0 14.84 7.24 0.91 Romania 100 Unlimited 29 St. Vincent and 7.33 Unlimited - 40.37 6,610 124 9.00 54.28 the Grenadines 30 3,560 20.0 Unlimited 5 16.92 2.74 0.92 Ukraine 7.38 24.21 125 - 1.50 Unlimited Guyana 16.0 3,936 18,370 31 Czech Republic 0.93 14.19 24.39 2 Unlimited 21.0 7.53 126 3,500 12.0 Unlimited 3.00 50.34 21.95 Philippines Unlimited 1 15.88 11,530 0.93 Brazil 32 40.2 8.95 28.63 60.73 1.00 7.81 Unlimited 10.0 127 Paraguay 4,400 33 12.09 18.69 5 Unlimited 21.0 15,280 0.95 Latvia 8.43 40.19 - - Unlimited 7.0 5,720 128 Tuvalu 22.0 34,270 34 Italy 0.97 27.62 30.16 7 Unlimited 14.0 129 Kyrgyzstan 8.58 8.94 29.36 0.50 Unlimited 1,250 0.98 38.76 42.51 Unlimited 19.0 47,640 35 Germany 16 Angola 130 5,476 - Unlimited 0.25 54.22 40.78 8.94 36 8.82 16.75 1 1 23.0 10,830 Turkey 0.98 4,260 Tonga 9.08 32.22 41.70 - 5 15.0 131 12.82 15,430 22.30 37 Lithuania 1.00 100 Unlimited 21.0 South Sudan 132 970 13.0 2 0.50 - 7.66 9.48 19.48 22,657 23.0 Unlimited 4 25.17 1.03 Greece 38 Indonesia 28.75 83.87 10.00 Unlimited 9.51 10.0 3,630 133 39 Slovenia 1.07 21.07 28.42 1 Unlimited 22.0 23,580 1,270 14.0 Unlimited 0.25 28.20 10.76 10.17 Mauritania 134 1.09 29,440 21.0 5 1 32.08 26.71 Spain 40 Nepal 10.75 0.50 6.54 21.20 7 135 730 13.0 United Arab 55.39 1.09 41 40.57 44,600 0.0 Unlimited 0.25 Emirates 38.66 11.43 Samoa 136 4,060 15.0 2.00 53.62 3 1 1 6.0 11,120 Papua New 42 Malaysia 1.11 10.31 24.97 137 21.67 25.59 - 1.17 10.0 2,240 11.61 Guinea 43 20.0 Unlimited 10 24.88 17.75 1.12 Estonia 19,030 Equatorial Guinea 11.92 101.45 - 0.25 Unlimited 10,210 138 - 44 Slovakia 1.12 16.64 26.99 2 300 20.0 17,750 12.38 Unlimited 139 Micronesia 33.00 - 0.25 0.0 3,200 29.84 1.13 35,320 17.0 Unlimited 15 33.19 Israel 45 140 1,080 12.44 - 11.20 - 0.25 10 Tajikistan 51,630 Canada 1.15 49.43 48.44 15 50 13.0 46 141 Nigeria 13.23 32.74 63.38 1.00 2,970 5 5.0 47 Belarus 1.17 7.16 3 Unlimited 20.0 7,340 0.50 142 Lao P.D.R. 13.31 18.41 46.90 Unlimited 10.0 1,660 Tunisia 48 4,230 23.0 Unlimited 4 10.68 4.21 1.19 13.65 4,390 0.0 Unlimited 0.25 - 49.95 Marshall Islands 143 3 1 23.68 15.10 1.28 14,100 15.0 Seychelles 49 20 23.5 4.00 144 Ghana 13.85 18.35 66.44 1,590 46,550 50 Ireland 1.29 49.91 46.05 100 Unlimited 23.0 12.00 30.56 145 Unlimited 10.0 1,020 Cambodia 14.12 4.00 27,090 Unlimited 50 36.01 10.0 29.17 1.29 Korea (Rep.) 51 15.0 Unlimited 0.50 61.70 15.39 Nicaragua 146 1,870 23.99 0.0 52 Oman 1.30 18.21 34.99 4 Unlimited 16,853 147 Honduras 16.39 31.00 60.47 1.00 Unlimited 12.0 2,270 53 2 20,070 15.0 Unlimited 26.07 21.80 1.30 Trinidad & Tobago 0.25 148 Swaziland 16.61 49.14 134.02 6 14.0 3,550 Uruguay 54 16,350 22.0 30 - 24.57 17.93 1.32 1,270 5.0 Unlimited 0.50 71.51 18.21 17.20 Myanmar 149 15.0 55 New Zealand 1.32 45.33 41.62 - 80 41,070 5.0 S. Tomé & Principe 19.92 27.73 49.41 1.00 12 150 1,670 12,980 1.37 56 14.79 24.65 4 1 25.0 Croatia 151 Zambia 20.69 28.96 85.44 2.00 10 16.0 1,680 57 12 1.37 24.39 33.46 Portugal Unlimited 23.0 21,360 12.5 Unlimited 0.50 54.93 20.86 3,160 Vanuatu 152 52.16 58 4.12 11.76 4 3.5 12.2 3,460 1.43 Sri Lanka 153 840 2 1.00 - 15.00 21.43 Zimbabwe 15.0 1.49 Malta 59 26.06 35.16 30 Unlimited 18.0 20,979 154 550 15.0 2 0.50 33.95 12.06 26.31 Ethiopia 60 0.0 Unlimited 100 64.20 51.34 1.53 Hong Kong, China 40,320 18.0 Unlimited 0.25 88.11 34.83 1,450 155 Côte d'Ivoire 28.82 61 1.54 9.76 Azerbaijan 7,590 18.0 Unlimited 1 920 Unlimited 0.50 60.91 22.60 29.47 Tanzania 156 32.5 62 Bulgaria 10.09 23.22 20 Unlimited 20.0 7,620 1.59 0.50 157 600 17.0 Unlimited 40.13 16.26 32.51 Mozambique Bosnia and 1.67 6.63 14.03 2 2 17.0 4,760 63 Senegal 158 1,050 23.0 Unlimited 1.00 76.19 30.43 34.78 Herzegovina 159 5,999 0.0 Unlimited 0.25 - 180.00 36.00 Cuba 64 20,980 - Unlimited 1 26.40 29.99 1.72 Bahamas Congo (Rep.) 160 36.07 81.77 155.80 0.25 Unlimited 16.0 2,720 65 Panama 1.73 16.04 27.03 1 Unlimited 7.0 11,130 161 Afghanistan 36.72 20.81 61.50 0.25 Unlimited 0.0 680 66 1,890 10.0 1.00 2.5 6.98 2.83 1.79 Viet Nam 21.00 50 1,290 162 Kenya 40.65 43.70 96.14 26.0 10.0 Unlimited 1 16.42 6.61 1.85 Mongolia 4,280 67 163 41.82 24.40 63.07 1.00 5 15.0 700 Sierra Leone 68 Saudi Arabia 1.90 39.73 80.33 10 Unlimited 0.0 25,115 164 1,350 19.3 Unlimited 0.25 127.67 50.55 44.94 Cameroon 41.57 13,340 27.0 Unlimited 1.28 21.23 1.91 Hungary 69 10.26 165 10.00 Malawi 2 26.5 250 49.26 35.12 2.03 Unlimited Colombia 70 1 30.04 16.0 13.49 7,970 166 71 2.07 10.99 21.35 0.25 Unlimited 18.0 6,360 Peru Unlimited 500 Gambia 22.3 50.56 21.05 0.25 83.26 500 8 25.97 2.09 19.0 14,910 Chile 72 41.69 0.50 56.99 Benin 167 109.72 890 42.27 18.0 Unlimited 10,030 10.0 40 2 - 17.51 2.10 Lebanon 73 18.0 Unlimited 0.25 89.29 33.65 62.12 650 Mali 168 10,120 13.0 Unlimited 74 Costa Rica 2.17 18.33 26.10 1 Comoros 41.71 - 0.50 Unlimited - 790 63.35 169 8,020 20.0 75 Turkmenistan 2.19 14.63 - 1 Unlimited 37.20 170 18.0 Unlimited 0.25 63.77 Burkina Faso 100.06 700 2.19 Venezuela 76 Kiribati 2,950 171 - Unlimited 0.25 - 187.82 76.40 12,615 18.96 1 Unlimited 12.0 23.07 172 570 79.36 37.70 96.56 0.25 Unlimited 18.0 To g o 2.28 2 10 14.0 6,800 77 South Africa 12.93 28.92 120.77 106.98 Guinea-Bissau 173 0.25 Unlimited 15.0 550 49.03 2.33 78 Barbados 30.00 24.14 2 Unlimited - 15,451 51.10 0.25 Unlimited Guinea 18.0 470 110.64 174 130.46 79 Armenia 2.50 8.37 4,020 20.0 Unlimited 4 20.50 138.52 50.79 Madagascar 175 440 186.31 8.00 Unlimited 20.0 Montenegro 2.51 15.31 1 1 19.0 7,320 80 29.90 50.30 147.22 Unlimited Niger 176 19.0 0.25 134.25 410 81 TFYR Macedonia 2.51 10.79 25.74 30 18.0 5,150 4 700 18.0 256.69 97.09 166.44 Rwanda 177 Unlimited 10.00 0.0 3,050 10 1 22.61 6.50 2.56 Egypt 82 178 180.48 Solomon Islands 275.22 273.87 0.25 6 10.0 1,830 83 9.52 4,450 20.0 4 2 21.10 2.57 Albania Burundi 179 82.90 211.13 0.25 Unlimited 18.0 368.46 270 84 22.02 9,870 16.0 Unlimited 5 37.80 2.68 Mexico 537.31 180 300.00 861.68 0.25 Unlimited 18.0 670 Uganda 14,920 - Unlimited 2 43.62 35.19 2.83 St. Kitts and Nevis 85 4.06 1,710 30.0 86 Sudan 2.85 7.54 0.5 2 181 Unlimited 18.0 613.19 1068.38 980 500.77 0.25 Chad - 9,720 87 Gabon 2.92 23.67 37.99 0.5 Unlimited 182 - Unlimited 0.25 320 488.63 1832.36 Central Afr. Rep. - 14.24 30.89 Serbia 10 Unlimited 88 20.0 5,820 2.94 - 18.47 San Marino** 21.45 20.00 Unlimited 0.0 - China 89 3.12 19.27 31.81 2 Unlimited - 7,400 0.50 - 30.00 - Somalia** - 10.0 10 7,820 0.0 20 0.5 - 21.74 3.34 Libya 90 31.61 7.0 0.25 5 53.30 - - Djibouti** Algeria 91 44.54 15.45 3.38 35.33 - Liechtenstein** 5,490 - 8.0 Unlimited 5.00 - 17.0 Unlimited 0.5 92 Belize 3.45 12.50 21.62 0.25 Unlimited 0.0 4,346 - - - 6 2.00 - 49.00 Timor-Leste** 93 5 12 21.10 9.96 3.46 Cape Verde 15.5 3,450 - 20.0 Unlimited - 100.00 55.34 - Monaco** 13,480 94 Argentina 3.66 41.16 - 3 Unlimited 21.0 0.0 Nauru** - 60.10 - 0.50 10 - - 0.25 - 6.59 3.78 2,090 Uzbekistan 1.17 95 - - 80.18 - 0.50 3 Syria** - Note: * Data correspond to the GNI per capita (Atlas method) in 2014 or latest available year adjusted with international inflation rates.** Country not ranked because data on GNI p.c. are not available. Measuring the Information Society Report 2016 120

137 Chapter 4 Chart 4.5: Mos t common entry-level fixed-broadband speed, globally and by level of development Note: Based on 144 economies for which 2008-2015 data on fixed-broadband prices were available. Source: ITU. Source: ITU. GNI p.c. and PPP$ values are based on World Bank data. where prices remain prohibitively expensive, which Regional analysis of fixed-broadband prices influence these regions’ averages (Chart 4.6). A regional comparison of fixed-broadband prices, speeds and data caps highlights important Africa differences between, as well as within, regions. Africa remains the region with the greatest remained the region with the In 2015, Africa divergence in absolute and relative price for highest absolute and relative fixed-broadband Internet access, while the CIS region has relatively prices (Chart 4.6), and the only region where low, and similar, prices. Africa, but also Asia and the average price of the service, at 119 per cent, the Pacific and the Americas, have some outliers, exceeded GNI p.c. levels. The regional average is Chart 4.6: Fi xed-broadband prices by region, 2015, in USD (left) and in PPP$ (right) Note: Each horizontal dash represents the price in one country in the region. The yellow marks indicate the regional average. *In Africa, the price of a fixed-broadband subscription was above USD300 (at USD489 and at USD500) in two countries. ** In Africa, the price of a fixed-broadband subscription was above PPP$300 (at PPP$862) in one country. Source: ITU. Measuring the Information Society Report 2016 121

138 influenced by nine countries where the price of Although a fairly large number of African countries a fixed-broadband connection exceeded average have unlimited fixed-broadband plans, none of the GNI per capita levels, but prices still remain unlimited offers come with high speed and at an unaffordable in many countries. affordable price, suggesting that subscribers will not be able to take full advantage of data-intensive In two thirds of all African countries for which services or applications. price data are available, the fixed-broadband er cent p service corresponds to more than 20 Americas of GNI p.c. In only six countries – Mauritius, Seychelles, South Africa, Gabon, Cape Verde and In the Americas, the cost of a fixed-broadband Botswana – is the entry-level offer below 5 r pe connection has dropped to about USD 30 .7). The unaffordability of 4 cent of GNI p.c. (Chart (PPP$ 37) and at end 2015 the region is close to fixed broadband in Africa goes hand in hand with per cent of GNI p.c. benchmark meeting the 5 the very low fixed-broadband penetration levels in terms of affordability of the service. Fixed observed in the region. broadband is most affordable in the United States and Brazil, where it corresponds to as little as 0.4 More than half the countries included in the price- and 0.9 per cent of GNI p.c. In about two thirds data collection continue to offer fixed-broadband of the countries in the Americas fixed-broadband services at speeds of 512 kbit/s, or below. In some services cost less than 5 per cent of GNI p.c., countries, including Kenya, Malawi and Rwanda, but most other countries, including Jamaica and entry-level fixed-broadband services offer high lower middle-income economies such as Bolivia, Mbit/s, but at high prices, speeds of above 10 El Salvador and Guatemala, have made the suggesting that these offers are aimed more at service relatively affordable. The exceptions are businesses than residential users. Very low fixed- Nicaragua, Honduras, and Cuba, where the service broadband penetration rates in these countries is still the most expensive and remains relatively support this conclusion. Cape Verde, which only unaffordable (Chart 4.8). recently graduated from the list of LDCs, stands out for its entry plan that offers theoretical speeds In the majority of countries in the Americas, entry- Mbit/s and includes 5 GB of data, at a of 12 level fixed-broadband connections are offered at relatively affordable price. speeds between 512 kbit/s and 2 Mbit/s. Lower speeds are offered only in Bolivia, Belize, Cuba, xed-broadband prices as a percentage of GNI p.c., speeds and caps, Africa, 2015 Fi Chart 4.7: Note: Broadband speeds and caps/month refer to the advertised speeds and the amount of data included in the entry-level fixed-broadband subscrip - tion. Source: ITU. GNI p.c. values are based on World Bank data. Measuring the Information Society Report 2016 122

139 Chapter 4 xed-broadband prices as a percentage of GNI p.c., speeds and caps, Americas, 2015 Fi Chart 4.8: - Note: Broadband speeds and caps/month refer to the advertised speeds and the amount of data included in the entry-level fixed-broadband subscrip tion. Source: ITU. GNI p.c. values are based on World Bank data. the CIS. In Kyrgyzstan, Tajikistan and Uzbekistan, Nicaragua and Peru. A handful of countries, speeds vary between 256 kbit/s and 512 kbit/s, including Argentina, Chile, Ecuador and Mexico, followed by higher speeds of up to 2 Mbit/s offer connections with speeds of between 2 in Azerbaijan, Kazakhstan, and Turkmenistan. and 10 Mbit/s. Canada’s entry-level plan offers Mbit/s. 15 Armenia, Belarus and Ukraine have entry-level plans with speeds of between 2 and 10 Mbit/s. The Americas is the region with the largest number The highest speeds, and unlimited data plans, are offered in Georgia, Moldova and the Russian of unlimited data plans. Only Canada, Chile and Uruguay cap the amount of data that subscribers Federation. The number of fixed-broadband can download, and even then the caps are very subscriptions in these countries has increased GB, respectively. high, at 50, 30 and 500 steadily over the last years, and by end 2015 penetration rates stood at around 15 per cent in per cent in the Georgia and Moldova, and at 19 24 Commonwealth of Independent States (CIS) Russian Federation. Of all the regions, the CIS has on average the In Kazakhstan and Tajikistan, entry-level fixed- cheapest fixed-broadband services, both in terms GB of data, and only broadband plans included 10 of USD and PPP$ (Chart 4.6). Apart from Europe, GB. All Uzbekistan had a much lower limit, of 1.2 it is also the region with the most affordable other countries in the region offer unlimited fixed- fixed-broadband services, which correspond to broadband services. per cent of GNI p.c. In the Russian Federation, 3.6 Kazakhstan and Ukraine, the price of the service amounts to less than 1 per cent of GNI p.c., and Europe only Kyrgyzstan and Tajikistan, where the service per cent of GNI p.c., lie represents 8.6 and 12.4 Europeans benefit from the most affordable entry- above the 5 per cent GNI p.c. benchmark (Chart level fixed-broadband services globally, usually 4.9). at relatively high speeds and with unlimited data. Although the service is not among the cheapest There are large variations in terms of the speeds in terms of absolute prices – a fixed-broadband offered for entry-level fixed-broadband services in 25 connection in Europe costs on average USD Measuring the Information Society Report 2016 123

140 xed-broadband prices as a percentage of GNI p.c., speeds and caps, CIS, 2015 Fi Chart 4.9: Note: Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. Broadband speeds and caps/ month refer to the advertised speeds and the amount of data included in the entry-level fixed-broadband subscription. Source: ITU. GNI p.c. values are based on World Bank data. 27) – high income levels make the service (PPP$ Andorra, Norway, Switzerland and Luxembourg, 4.6). By 2015, the average per cent of GNI p.c. but represents less than 5 very affordable (Chart price of fixed broadband as a percentage of GNI in all countries. In Serbia – the country with the per cent per cent (down from 1.3 p.c. stood at 1.1 highest relative price for high-speed Internet per cent in the a year earlier), compared to 3.6 access – the plan represents 2.9 per cent of GNI p. CIS, 5.4 c. (Chart 4.10) per cent per cent in the Americas, 6.6 in the Arab States, 13.6 per cent in Asia and the Pacific and 119.1 per cent in Africa. The service Fixed broadband in Europe is not only cheap but is particularly affordable in the United Kingdom, also relatively fast. Only Andorra continues to Fi Chart 4.10: xed-broadband prices as a percentage of GNI p.c., speeds and caps, Europe, 2015 Note: Broadband speeds and caps/month refer to the advertised speeds and the amount of data included in the entry-level fixed-broadband subscription. Source: ITU. GNI p.c. values are based on World Bank data. Measuring the Information Society Report 2016 124

141 Chapter 4 offer an entry-level plan of 512 Almost half of countries in the Arab States region kbit/s. Speeds Mbit/s are offered in another nine of below 2 for which price data are available offer entry-level fixed-broadband plans with speeds of between countries, including the Czech Republic, Turkey, kbit/s and 512 256 kbit/s, which is a greater Slovenia, Spain, Hungary and Albania. The majority of European countries offer plans at speeds of proportion than in other regions, except for Africa. Mbit/s and above, and entry-level plans in 2 With the exception of Saudi Arabia, where the entry-level speed is 10 Mbit/s, all GCC members Belgium, Sweden, Romania, Lithuania, Ireland and Malta provide speeds of at least 30 Mbit/s. offer relatively low-speed plans. Entry-level plans in Tunisia, Oman, Morocco and Palestine, on the Fixed broadband in Europe is affordable and fast, other hand, come with speeds of 4 Mbit/s. and two thirds of all countries offer unlimited data Two thirds of countries in the Arab States region plans. One third of countries continue to impose offer unlimited data plans, and caps are relatively data caps, ranging from a high of 300, 100 and 30 GB) and Libya GB), Bahrain (20 high in Lebanon (40 GB in Slovakia, Belgium and TFYR Macedonia, GB). Egypt and Jordan limit the amount of (20 respectively, to more restrictive caps in Andorra, Luxembourg, Iceland, Bosnia and Herzegovina and data included in the entry-level broadband plan to Albania. A 1 GB, and Yemen and Sudan’s data limits are 4 GB GB cap is applied only in Turkey, Spain, 10 GB) and Montenegro. and 2 GB, respectively. Croatia, Hungary (1.3 Arab States Asia and the Pacific Fixed broadband in Asia and the Pacific varies The average relative and absolute prices of greatly between countries in terms of absolute fixed-broadband services in the Arab States conceal sizeable variations between countries, and relative prices, and in terms of speed and data allowance. These differences reflect wide in part owing to wide variations in income levels variations that also characterize the region in between the region’s oil-exporting members of terms of income, infrastructure, population size the Gulf Cooperation Council (GCC) and others. and density, and which make Asia and the Pacific The 2015 price of a fixed-broadband connection one of the most diverse regions in the world. per cent of GNI p.c. in represented less than 1 Kuwait, Bahrain and Qatar, and fell below the 5 per cent benchmark in most countries in the region. Very affordable fixed-broadband services are Prices remain relatively high in Palestine, Yemen, offered in several of the high-income economies Jordan and Mauritania, and very high in Comoros of the region, including Japan, Singapore and Australia, but also in the Islamic Republic of Iran, (Chart 4.11). Chart 4.11: xed-broadband prices as a percentage of GNI p.c., speeds and caps, Arab States, 2015 Fi Note: Broadband speeds and caps/month refer to the advertised speeds and the amount of data included in the entry-level fixed-broadband subscrip - tion. Source: ITU. GNI p.c. values are based on World Bank data. Measuring the Information Society Report 2016 125

142 where the 2015 price of the service corresponds to and Thailand offer high-speed Internet access at 0.6 per cent of GNI p.c. Malaysia and Sri Lanka, but between 10 and 30 Mbit/s. Entry-level plans with speeds of above 30 also Viet Nam and Mongolia, have lower incomes Mbit/s are only offered in the yet relatively affordable fixed-broadband services, high-income, highly connected nations Singapore and the Republic of Korea. corresponding to less than 2 per cent of GNI p.c. (Chart 4.12). Prices in the region’s most populous countries, China and India, have become relatively About half of the economies in the region offer affordable, corresponding to 3.1 and 5.1 per cent entry-level broadband packages that include of GNI p.c., respectively. unlimited data volumes. These include some of the high-income economies with faster speeds, Overall, almost half of all countries in the including Singapore, Australia and the Republic region offer prices below the 5 per cent GNI of Korea, but also low-income economies with slower broadband connections, such as Kiribati, p.c. benchmark, including the LDCs Bangladesh Afghanistan and Myanmar. In these latter and Bhutan. The service remains less affordable economies, however, the limited speeds are likely in many of the region’s other LDCs and SIDS – to restrict the type and amount of services and including in the Solomon Islands, Kiribati and applications that subscribers can use in practice. Vanuatu – where infrastructure barriers and limited international Internet bandwidth often keep prices high. Why some of the poorest countries continue to In the Asia and the Pacific region broadband have the highest fixed-broadband prices speeds and data caps vary significantly between countries, but most of the LDCs and SIDS with Unlike mobile-cellular prices, the highest entry- the most unaffordable fixed-broadband services level fixed-broadband prices are found in are also those which offer relatively low-speed developing countries, and in particular in some of connections. Exceptions include Cambodia and the world’s least developed countries (LDCs). Samoa, where the ISP offers entry-level speeds of Mbi t/s, respectively. Relatively high- 4 Mbit/s and 2 By the end of 2015, a fixed-broadband plan Mbit/s are speed connections of between 2 and 10 with a minimum of 1 GB of data per month cost also offered in the Philippines, Maldives, Viet Nam more than USD 80 per month in ten developing and Sri Lanka. Papua New Guinea, Indonesia, Fiji countries (Table 4.5). Eight of those countries were xed-broadband prices as a percentage of GNI p.c., speeds and caps, Asia and the Pacific, 2015 Fi Chart 4.12: Note: Broadband speeds and caps/month refer to the advertised speeds and the amount of data included in the entry-level fixed-broadband subscription. Source: ITU. GNI p.c. values are based on World Bank data. Measuring the Information Society Report 2016 126

143 Chapter 4 Co untries with the highest fixed-broadband prices in USD, 2015 Table 4.5: Mobile broadband Total houshehold Fixed broadband (computer-based) expenditure** Development Economy USD per Prices Prices status capita/month USD/month USD/month LDC 501 Chad 17 58 32 N/A 489 Central African Rep. LDC Uganda 300 11 41 LDC LDC - 275 Solomon Islands 73 LDC Kiribati 188 56 - Cuba 180 N/A 308* non-LDC Equatorial Guinea 101 N/A 272 LDC 97 8 39 LDC Rwanda LDC 83 N/A Burundi 18 Congo (Rep.) 82 17 107 non-LDC Note: N/A means the service is not available. “-“ means that the information is not available. * 2014 data. ** Calculated by dividing the indicator “household final consumption expenditure (current USD)” by the population of the country. Source: ITU. Data on household final consumption expenditure sourced from the World Bank. (Orange WiMAX offer in the Central African LDCs in which the total household consumption Republic) or low contention ratios (CBINET ADSL expenditure per capita ranged from USD 18 to USD 25 28 offer in Burundi), This highlights how unaffordable is added by default to entry- 58 per month. fixed broadband is in these countries, especially level fixed-broadband plans. Normally, operators considering the international comparisons: the would offer these extra features for a higher price, highest entry-level fixed-broadband prices in the but also offer basic plans to residential customers. developed world are recorded in Ireland, at USD This is not the case in several LDCs, and therefore 50 per month. This is significantly lower than in basic fixed-broadband plans become unaffordable all the countries listed in Table 4.5, even though for residential customers. 26 and the entry- income in Ireland is much higher level plan has a speed of 100 Mbit/s, whereas in Another element that may explain the high prices most LDCs the entry-level speed is 256 kbit/s. in some countries is the technology used in the fixed-broadband network. The traditional fixed- Entry-level fixed-broadband plans cost less line network (copper wire) has limited reach than USD 15 per month in a number of LDCs, in most LDCs and ADSL services rely on this including Bangladesh, Bhutan, Ethiopia, Cambodia, infrastructure. As a result, ADSL plans are often Mauritania, Malawi, Lesotho, South Sudan and only offered by the incumbent operator (i.e. the Sudan. However, fixed-broadband uptake is also only operator having access to the legacy fixed- 29 and at very high very low in these countries, with the exception line infrastructure in most LDCs) 27 prices. Fixed wireless technologies, such as fixed Therefore, the much of Bangladesh and Bhutan. WiMAX, are often a more affordable alternative higher prices in other LDCs must have specific for extending the reach of the fixed-broadband supply-side causes which, if addressed, could significantly contribute to making fixed broadband network in countries with limited basic fixed-line infrastructure and reduced or sparse demand. more affordable in these countries. Significant investment is needed in LDCs to extend the basic wired-line infrastructure and making the In LDCs with very high fixed-broadband prices, operators often market fixed-broadband services appropriate technological choice in each situation as a premium or business service. For instance, could help streamline the limited investment flows Foris Telecom in Uganda and Airtel in DR Congo allocated to fixed services. advertise fixed Internet offers to business customers, whereas households are only offered Uganda is a good example of how the technology mobile-broadband services. Even in some cases may affect the price of entry-level fixed-broadband where it is not specifically stated, the inclusion services. The local ISP Foris Telecom offered of some features typical of business broadband WiMAX plans at 512 kbit/s for USD 14 per month services, such as a minimum guaranteed speed in 2013. The service was discontinued in 2014 Measuring the Information Society Report 2016 127

144 and, in that year, the cheapest fixed-broadband ICT opportunities are linked to areas requiring plan advertised on the incumbent’s website was high connectivity, such as big data analytics and based on ADSL and cost USD 300 per month for a the Internet of Things (IoT). Moreover, developing speed of 256 kbit/s. In 2015 the plans offered by countries, and LDCs in particular, could benefit the 30 the incumbent operator Uganda Telecom were the Therefore, most from these ICT developments. same as in 2014. However, in June 2016, Uganda policy-makers and regulators in these countries should not disregard the issue of very high Telecom advertised a WiMAX plan at 256 kbit/s fixed-broadband prices, but rather address the for USD 37 per month. This suggests that, if the concrete commercial and infrastructure-related WiMAX offer is maintained, prices in Uganda may problems mentioned above that make fixed become much lower in 2016 than in 2015. broadband a premium service that is unaffordable for residential customers and small/micro Other infrastructure elements have an impact undertakings. on the underlying costs of fixed-broadband provision in LDCs. These include limited and expensive international connectivity and backhaul 4.4 Mobile-broadband prices connections, as well as deficiencies in the power grid. However, these factors are to a large extent common in the broadband infrastructure chain, Overview of global trends in mobile-broadband and therefore also affect mobile-broadband prices. An analysis of mobile-broadband prices prices in countries with very high fixed-broadband prices reveals that mobile-broadband is much Mobile-broadband services are becoming available cheaper, thus suggesting that the infrastructure in more and more countries, including LDCs, elements common to fixed and mobile broadband where the availability of prepaid handset-based are not the main determinant of the very high plans almost doubled in the period 2012-2015, fixed-broadband prices. Instead, the regulatory and tripled in the case of postpaid computer- challenges faced in the fixed-broadband market based plans (Chart 4.13). In 2015, eight developing and the resulting limited competition in some countries started offering mobile-broadband LDCs (ITU, 2013) may better explain some of the services. differences in fixed and mobile-broadband prices. In addition to 3G mobile broadband, mobile- Another distinct element in LDCs that may have an broadband networks based on LTE and other impact on the fixed-broadband prices is the way in advanced technologies are being deployed and which prices are communicated. Price information per cent of countries are now available in 70 is not always available on operators’ websites worldwide. However, the availability of LTE but can sometimes be obtained by phone, e-mail broadband networks varies across development or paper advertisements. As a result, bespoke levels: LTE technologies have been deployed only prices and/or one-off offers may be common and in 38 per cent of LDCs, as against 58 per cent of information on prices is more difficult to obtain, developing countries and 91 per cent of developed 32 even for telecommunication regulators. For countries. This suggests that the speed and instance, MTN Rwanda does not advertise fixed- capacity enjoyed by mobile-broadband users may broadband prices on the website, but the small differ significantly across countries. alternative operator Hai advertises fibre-optic packages starting at USD 97 per month for 10 Apart from the increasing availability of the Mbit/s. None of them publishes prices for fixed- service, another key factor for the uptake of wireless broadband plans, although data from the mobile broadband has been the drop in prices. Rwanda Utilities Regulatory Authority show that Globally, handset-based mobile-broadband prices most fixed-broadband subscriptions in the country 23 in 2013 to have fallen from an average of USD relied on fixed-wireless technologies in 2014. 13 in 2015 (Chart 4.14). In parallel, average USD prices for computer-based mobile-broadband Fixed-broadband Internet access cannot always 21 to USD services have decreased from USD 16. be replaced by mobile-broadband access, The decrease has been remarkable in LDCs, where particularly for users requiring high capacity and handset-based prices have more than halved in high speed. Some of the most promising future both USD and PPP terms in the period 2012-2015, Measuring the Information Society Report 2016 128

145 Chapter 4 Chart 4.13: Av ailability of mobile-broadband services by type of service, by level of development, 2012- 2015 Note: A mobile-broadband service is counted as being available if it was advertised on the website of the dominant operator or if prices were provided 31 to ITU through the ICT Price Basket Questionnaire. Source: ITU. while there has been a 40 per cent reduction in broadband may be having a stronger impact on how people go online in the developing world. computer-based prices. Indeed, the percentage of users accessing the Internet on the move tripled in Egypt between Nevertheless, prices in LDCs still represent on average 11 per cent of GNI p.c. for handset-based 2013 and 2014, and doubled in Brazil. As Internet services and 17 pe r cent for computer-based plans. usage continues to grow in Brazil and Egypt, a significant proportion of new Internet users may This suggests that the service, and particularly go online exclusively through mobile networks. computer-based mobile broadband, is still The situation is most likely to be the same in unaffordable for large segments of the population other developing countries, in view of the growing in LDCs. In developing countries, handset- proportion of households with Internet and based mobile broadband is also significantly the low fixed-broadband subscription figures in more affordable than computer-based mobile per cent against 7.6 per cent, on broadband (5.1 most developing countries. This highlights the average, in 2015). This is in stark contrast with importance of affordable mobile-broadband the situation in developed countries, where both services to expand Internet usage in the handset-based and computer-based services are developing world. very affordable and correspond on average to less In addition to the increase in Internet users than 1 per cent of GNI p.c. (i.e. mobile broadband is more affordable than fixed-broadband and connecting through mobile networks, the mobile-cellular services in developed countries). decrease in mobile-broadband prices goes hand in hand with an increase in the intensity of use. The increasing availability of mobile-broadband Indeed, the statistics on mobile Internet traffic services and the decrease in prices is changing show that the amount of data consumed by each the way people access the Internet: a growing subscription is increasing in most countries for number of Internet users are connecting through which data are available (Chart 4.16). This suggests mobile networks (Chart 4.15). Available data show that the reduction in mobile-broadband prices that in a majority of developed countries Internet contributes not only to connecting more people users are increasingly connecting to the Internet but also to fostering more intense Internet usage 33 while on the move. The limited data available among those who are already online. from developing countries suggest that mobile Measuring the Information Society Report 2016 129

146 B handset-based (left) and 1 as a B computer-based (right) mobile-broadband prices: G Chart 4.14: M 50 0 percentage of GNI p.c. (top graph), in PPP$ (middle graph) and in USD (bottom graph), 2013-2015 Note: Simple averages. Based on 153 and 147 economies for which 2013-2015 data on handset-based and computer-based mobile-broadband prices are available, respectively. Source: ITU. , ITU collects data for (a) prepaid handset-based mobile- mobile-broadband prices To monitor MB per month, and (b) postpaid computer-based broadband plans with a data allowance of 500 GB per month. The plan selected in each mobile-broadband plans with a data allowance of 1 country for each service is not necessarily the one with the cap closest to 500 MB or 1 GB, but MB/1 the one from the dominant operator that is cheapest while including a minimum of 500 GB. The validity period considered for the plans is 30 days or four weeks. and postpaid computer-based mobile-broadband Comparison of mobile-broadband, fixed- prices shows that mobile broadband is significantly broadband and mobile-cellular prices less expensive (Chart 4.17). Indeed, in developing countries fixed broadband costs on average twice A comparison between postpaid fixed-broadband Measuring the Information Society Report 2016 130

147 Chapter 4 Chart 4.15: Pe rcentage of Internet users that used the Internet on the move, selected economies, 2013 and 2014 Note: * 2012 and 2013 data. Chart 4.15 refers to use of the Internet while mobile via a mobile phone or other mobile access devices, for example, a laptop computer, tablet or other handheld device. For developing countries, it refers to Internet use through the above mentioned devices connected to a mobile phone network and if the location is away from “home”, “work”, “place of education”, “another person’s home” and “community and commercial access facilities”. For - European countries, it refers to Internet use through the above-mentioned devices “away from home and work”. For more information on the defi nitions of Internet use by location, see page 55 in Manual for Measuring ICT Access and Use by Households and Individuals 2014 http:// available at: tistics/ Pag es/ publica D/ tions/ manual2014. aspx . ITU- Sta en/ www. itu. int/ Source: ITU and Eurostat for European countries. bile data traffic per subscription per month, selected economies, 2012-2014 Mo Chart 4.16: Note: Mobile data traffic does not include traffic offloaded onto fixed networks through WiFi. Source: ITU. as much as mobile broadband. In LDCs, a fixed- mobile-broadband services were 40 per cent broadband plan cost four times as much as a cheaper than fixed broadband in 2015. mobile-broadband plan in 2014 and this ratio did A more detailed analysis at the country level not improve much in 2015, despite the decrease reveals that in 13 developing countries fixed- in fixed-broadband prices. Even in developed broadband services are prohibitively expensive countries, where fixed-broadband prices are significantly lower than in the developing world, (above USD 50 per month), and in three LDCs the service costs more than USD 200. Measuring the Information Society Report 2016 131

148 Chart 4.17: mparison of postpaid fixed- Co In these countries, computer-based mobile- 40 to USD broadband prices and postpaid computer-based broadband prices range from USD 10 GB/month), in USD, mobile-broadband prices (1 per month, making mobile broadband the only by level of development, 2014 and 2015 affordable means of connecting to the Internet with a computer. The only two countries in which both fixed broadband and computer-based mobile broadband cost more than USD 50 per month are the Solomon Islands and Syria. Caution must be exercised when comparing the prices of fixed-broadband plans and computer- based mobile-broadband services, however, because of the different characteristics of the two services. In particular, a majority of the fixed- broadband plans (two thirds of the total in the 2015 ITU data collection) include an unlimited data allowance, whereas most mobile-broadband GB per month. As a result, the plans include 1 intensity of use may be higher in fixed-broadband users, as the available data on Internet data traffic 34 seem to confirm. A comparison of handset-based mobile-broadband prices and mobile-cellular prices sheds some light on the cost of mobile broadband relative to other mobile services that are complementary and often contracted together (see Section 4.5). Available data show that average mobile-cellular prices and handset-based mobile-broadband prices are converging: in 2015, handset-based 1.5 per month mobile broadband was only USD more expensive than mobile-cellular services in developing countries, and USD 1 per month in LDCs, whereas the difference had been twice as much in 2014 (Chart 4.18). In developed countries the situation was the opposite, with handset-based mobile broadband per cent cheaper than mobile- more than 20 cellular services. This finding highlights the low mobile-broadband prices available in many developed countries, and the fact that in many of them the average consumption per subscription is above 500 MB per month (Chart 4.16). The fact that handset-based mobile-broadband prices are comparable to mobile-cellular prices in the developing and the developed world, as well as in LDCs, suggests that the growth in mobile- Note: Simple averages. Based on 161 economies for which 2014 and cellular subscriptions witnessed in the last decade 2015 data on computer-based mobile-broadband and fixed-broadband prices are available. could be replicated in the mobile-broadband Source: ITU. market. Measuring the Information Society Report 2016 132

149 Chapter 4 Chart 4.18: Co mparison of prepaid mobile-cellular Indeed, affordable prepaid handset-based plans were a major driver for the uptake of mobile prices and prepaid handset-based mobile- voice and SMS services, and they could have a MB/month) prices, in USD, by broadband (500 level of development, 2014 and 2015 similar effect in promoting handset-based mobile- broadband services. Nevertheless, the cost of the service is not the only price component to be considered in relation to mobile-broadband services, and other factors such as the cost of a smartphone may be a decisive factor for future 35 uptake. Mobile-broadband prices in 2015 An analysis of prices in local currency shows that in about 70 per cent of countries mobile-broadband 43 prices decreased or remained the same in 2015. Moreover, a drop in prices of more than 10 per cent was recorded in 35 per cent of countries per cent of for handset-based plans, and in 30 countries for computer-based services. Despite the per general flat or downward trend, in about 15 cent of countries mobile-broadband prices went per cent. up by more 10 These findings provide a more nuanced view of mobile-broadband price trends in 2015. A more detailed analysis requires the examination of country data, and is presented in this section on the basis of the 2015 mobile-broadband prices. The price of a prepaid handset-based service with MB monthly data allowance corresponds a 500 to less than 0.15 per cent of GNI p.c. in Norway, Sweden, Austria, Estonia and Ireland, the countries with the most affordable services (Table 4.8). European nations dominate the list of the top ten countries with the most affordable handset-based mobile broadband, which also includes two Asian per cent GNI p.c.) and economies: Singapore (0.16 the Republic of Korea (0.22 per cent). All these countries have in common high income levels, advanced mobile networks (all have deployed LTE), strong competition in the mobile-broadband market (three or more operators) and high mobile- per cent). broadband penetration rates (above 65 Apart from high-income developed countries, the list of economies with relatively affordable handset-based mobile-broadband services (where Note: Simple averages. Based on 186 economies for which 2014 and the cost of the service corresponds to less than 2015 data on handset-based mobile-broadband and mobile-cellular prices are available. 1 per cent of GNI p.c.) includes 28 developing Source: ITU. countries. Some of these are countries with relatively low income per capita levels, such Measuring the Information Society Report 2016 133

150 Box 4.3: Mobile broadband takes off in Bhutan 000 inhabitants with a mountainous and rugged geography. A majority Bhutan is a country of 776 36 These demographic and geographic of the population in the country lives in rural areas. constraints pose a challenge for the deployment of telecommunication networks, which has been partially overcome by the deployment of wireless networks in the country, especially using low- frequency bands (800 and 700 MHz) that are better suited to Bhutan’s topology. Bhutan Telecom launched 3G services in 2008 in parallel with the launch of GPRS and EDGE data services. GPRS was soon available in most of the country, and EDGE in the major towns, but 37 3G services were only available in the capital city, Thimphu, until 2011. The second operator, 38 TashiCell, only started offering mobile services in 2008, and 3G services at the end of 2013. Therefore competition in the mobile market was focused on traditional mobile-cellular services (voice, SMS, narrowband data) in the period 2008-2013 and that resulted in high mobile-cellular subscription growth and lower prices (Chart Box 4.3). Chart Box 4.3: bile-cellular and mobile-broadband penetration in Bhutan, 2008-2015 Mo Source: ITU. Mobile broadband took off in Bhutan in 2013, coinciding with Bhutan Telecom’s expansion of the service to 15 out of 20 districts in the country, and the launch of 3G services by TahsiCell. Mobile-broadband subscriptions have more than tripled between 2013 and 2015, in parallel with the extension of 3G coverage in the country (from 54.7 per cent to 80 per cent during the same per cent decrease between 2013 and 2015). period) and the drop in prices (35 39 Bhutan Telecom launched LTE services in some areas of the capital city in October 2013, 40 41 although its LTE coverage has remained very limited. TashiCell launched LTE services in 2016. As the two Bhutanese operators are engaging in campaigns to promote the upgrade of 3G 42 customers to LTE, future LTE developments could play an important role in driving future mobile-broadband growth (in terms of both subscriptions and intensity of use). Measuring the Information Society Report 2016 134

151 Chapter 4 p three countries with the cheapest mobile-broadband services in each region, PPP$, 2015 To Table 4.6: Prepaid handset-based 500MB Africa Europe Asia & Pacific The Americas Arab States CIS PPP$ Country PPP$ Country PPP$ Country PPP$ Country PPP$ Country PPP$ Country 2.55 Cambodia 3.11 Estonia Sudan 5.01 Uruguay 3.67 Georgia 3.69 6.09 Rwanda 6.71 Moldova 10.15 Jordan 9.79 Chile 4.11 Sri Lanka 4.34 Lithuania 5.87 Liberia Antigua & 4.71 Bhutan 5.15 10.40 Iceland 6.17 Mozambique 11.32 Egypt Kazakhstan 10.50 Barbuda Postpaid computer-based 1GB Europe Asia & Pacific The Americas Arab States CIS Africa PPP$ Country PPP$ Country PPP$ Country PPP$ PPP$ Country PPP$ Country Country Trinidad & 5.88 Denmark Kenya Sri Lanka 10.57 Egypt 11.30 Georgia 6.15 12.02 4.64 Tobago Cambodia 6.82 5.09 Uruguay 10.93 Tunisia Austria 12.18 Tanzania 10.20 Kyrgyzstan 16.18 South Bangladesh 16.95 Sudan 7.44 14.48 Barbados Iceland 10.96 14.90 10.50 Kazakhstan Africa Note: Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. Source: ITU. Table 4.7: Av erage mobile-broadband prices and ranges by region, as a percentage of GNI p.c., 2015 Prepaid handset-based 500MB Postpaid computer-based 1GB % change % change Region Average Min. Max. Min. Average Max. avg. avg. 2015 2015 2015* 2015* 2015 2015 2014/15 2014/15 0.14 0.59 1.67 0.65 17% 2.16 -26% 0.07 Europe -56% CIS 0.30 12.29 2.68 -22% 0.45 12.29 2.82 The Americas 0.27 12.99 3.05 -11% 0.36 28.86 3.96 -12% 5.77 47.64 0.31 -20% 3.25 20.71 0.16 Asia & Pacific -20% 5.24 Arab States 0.29 29.73 4.15 -16% 0.21 29.73 -25% 20.75 -22% Africa 0.70 27.89 9.47 -27% 1.06 114.29 Note: *Simple averages based on 155 countries for which 2014 and 2015 price data for all mobile-broadband services were available. Source: ITU. The list of countries with the most affordable per cent of GNI p.c.), Georgia as Belarus (0.41 (0.43 per cent) and Bhutan (0.84 per cent), the postpaid computer-based services with a 1 GB monthly data allowance is also dominated by only LDC with handset-based mobile-broadband er cent of GNI p.c. p prices representing less than 1 European countries, with Denmark, Austria, (Box 4.3). Luxembourg and Norway at the top, having prices that represent less than 0.2 per cent of GNI p.c. At the end of the table, most of the countries (Table 4.9). The top 20 also include some non- with the least affordable handset-based mobile- per cent European countries such as Qatar (0.2 broadband prices are LDCs from Africa and Asia per cent), Australia of GNI p.c.), Singapore (0.3 and the Pacific. Indeed, in nine LDCs the cost of per cent) and the United States (0.4 (0.3 per cent). the service corresponds to more than 20 per cent There are two developing countries that stand out of GNI p.c., thus making it unaffordable for most of as offering relatively affordable computer-based the population in these countries. This is reflected mobile broadband despite their low income levels, in the low mobile-broadband penetration achieved namely Sri Lanka (0.56 r cent of GNI p.c.) and the pe in these countries (less than 25 subscriptions per Islamic Republic of Iran (0.63 per cent of GNI p.c.). 100 inhabitants). Measuring the Information Society Report 2016 135

152 Table 4.8: Mobi M B, 2015 le-broadband prices, prepaid handset-based, 500 Mobile-broadband, Mobile-broadband, prepaid handset-based prepaid handset-based Monthly data Monthly data GNI p.c., GNI p.c., (500 MB) (500 MB) allowance allowance Rank Economy Rank Economy USD* USD* (MB) (MB) as % of as % of PPP$ USD USD PPP$ GNI p.c. GNI p.c. 6,360 4.89 94 6.08 0.07 Norway 1 103,630 500 18.30 9.42 1.78 Peru 500 1.81 14.89 25.56 800 9,870 5.81 0.11 1,000 5.33 61,610 2 Sweden Mexico 95 3 0.13 5.55 5.78 1,024 96 Austria 500 19.93 10.28 1.90 Iraq 49,670 6,500 Gabon 1.98 16.06 25.78 500 9,720 97 Iceland 0.14 5.23 4.71 500 46,304 4 5 0.14 2.22 3.11 500 19,030 98 Morocco 2.00 5.12 12.11 4,096 3,070 Estonia 99 46,550 1,200 16.30 7.69 2.10 Paraguay 4,400 500 5.11 5.53 0.14 Ireland 6 1,200 14.55 8.65 2.13 Fiji 7 4,870 Singapore 0.16 7.27 8.31 1,024 55,150 100 8 48,420 0.18 6.61 500 Finland 7.32 5,630 800 22.29 10.11 2.16 Namibia 101 1,024 Montenegro 13.20 25.78 6,144 7,320 15,430 2.16 4.34 2.50 0.19 Lithuania 9 102 27,090 1,890 10 Korea (Rep.) 0.22 4.86 6.00 30,720 675 8.55 3.46 2.20 Viet Nam 103 13,690 104 4.99 9.50 500 2,720 Congo (Rep.) 2.20 1,000 5.28 2.65 0.23 Poland 11 2,048 7.56 9.17 0.25 United Kingdom 12 105 Colombia 2.25 14.95 33.29 7,970 43,430 500 0.27 3.66 5.01 512 16,350 1,270 550 13 2.41 2.28 Myanmar 106 Uruguay 9.46 14 Switzerland 0.27 19.74 12.54 600 Sudan 2.30 3.28 6.09 500 1,710 88,032 107 15 108 47,640 500 12.04 10.98 0.28 Germany Tonga 2.31 4,260 1,024 10.60 8.19 16 109 64,540 2.37 82.98 88.02 500 42,000 700 Japan Australia 0.28 15.03 12.88 110 Rwanda 17 Belgium 2.38 700 47,260 500 3.67 1.39 0.28 11.07 11.29 500 0.29 2.42 5.79 12.39 Bolivia 18 Qatar 111 21.98 29.45 3,000 92,200 2,870 500 112 0.29 12.26 600 40,320 7,240 800 29.72 14.71 2.44 Botswana 9.80 19 Hong Kong, China 20 113 2,250 11.96 3.28 0.30 Russian Federation 13,220 St. Lucia 2.45 14.81 17.98 1,000 7,260 13.31 13.39 500 51,890 Netherlands 21 0.31 114 Armenia 2.50 8.37 20.50 4,500 4,020 Spain 0.32 7.76 9.32 525 29,440 22 St. Vincent and the 6,610 1,000 19.92 14.81 115 2.69 Grenadines 23 Macao, China 0.35 22.29 28.89 500 76,270 116 3,160 510 7.45 7.85 2.98 Vanuatu 41,070 500 24 New Zealand 0.39 13.25 12.17 5,120 43.83 9,950 3.07 Suriname 117 25.42 25 Kuwait 0.40 16.62 26.74 5,120 49,300 118 1,570 India 3.09 4.04 14.12 1,024 500 7,340 Belarus 26 - 2.51 0.41 119 1,080 1,024 7.57 Bangladesh 2.80 3.11 3.69 0.43 27 Georgia 1.32 500 3,720 120 1,670 600 8.07 4.53 3.25 S. Tomé & Principe 500 28 United Arab Emirates 0.44 16.34 22.30 44,600 5,476 500 19.94 14.99 3.29 Angola 121 500 11.97 8.87 23,580 0.45 Slovenia 29 7,910 1,024 29.77 Grenada 22.22 3.37 122 1,024 10.50 30 Kazakhstan 0.45 4.46 11,850 5,150 123 Jamaica 3.49 14.97 22.97 2,048 Bahrain 0.46 7.98 13.45 1,024 31 21,039 124 Venezuela 3.49 36.70 30.15 800 12,615 0.47 32 Portugal 8.31 11.40 500 21,360 20.00 6,090 1,000 34.39 3.94 Ecuador 125 14,910 Chile 0.49 6.10 9.79 500 33 3.96 5.24 18.98 126 600 1,590 Ghana 1,200 3,460 34 Sri Lanka 0.50 1.44 4.11 3,936 800 - 13.56 4.13 Guyana 127 35 Latvia 0.52 6.64 10.27 600 15,280 128 4,346 4.14 15.00 25.95 1,024 Belize 36 Canada 0.52 51,630 22.48 22.03 500 129 El Salvador 14.00 4.29 3,920 2,000 26.78 37 Slovakia 0.52 7.75 12.58 700 17,750 10.91 21.13 600 2,970 130 4.41 Nigeria 13.88 35,320 500 15.44 0.52 Israel 38 5,000 1,660 6.14 131 4.44 Lao P.D.R. 15.63 39 0.54 5.83 9.72 500 12,980 Croatia 15.29 1,024 1,250 132 4.47 Kyrgyzstan 4.65 Turkey 10,830 500 9.77 5.15 0.57 40 3,430 133 Guatemala 4.52 12.93 24.02 1,500 41 34,270 1,024 18.17 16.64 0.58 Italy 3,072 134 Dominican Rep. 4.96 24.97 50.99 6,040 37,663 42 Brunei Darussalam 0.59 18.62 31.39 500 600 500 6.17 2.50 5.00 Mozambique 135 43 Azerbaijan 0.62 3.90 600 7,590 12.02 5.08 700 Kenya 5.46 1,290 136 22.92 France 600 42,960 44 0.62 22.18 36.23 7,820 500 - 137 Libya 5.56 Luxembourg 45 75,990 4,608 37.01 39.93 0.63 3,060 Palestine 5.73 14.62 22.29 1,024 138 7,113 Iran (I.R.) 0.63 3.76 11.03 1,000 46 5.74 2,090 Uzbekistan 139 10.00 - 1,000 47 7.12 13.95 500 13,340 Hungary 0.64 140 19.52 1,650 4,060 5.77 Samoa 27.08 0.66 24.94 48 2,000 45,033 Andorra - Dominica 141 6,930 5.90 34.07 45.82 1,000 49 5.59 0.70 10.16 800 9,630 Mauritius 3,550 48.95 17.95 6.07 Swaziland 142 500 50 9,520 1,024 11.36 5.55 0.70 Romania 1,680 143 Zambia 6.62 9.27 27.34 500 13.31 17.96 1,200 0.76 20,979 51 Malta 1,870 1,500 28.31 11.01 7.06 Nicaragua 144 Brazil 0.78 7.48 13.27 600 11,530 52 145 Lesotho 7.84 23.22 550 1,330 7.07 0.78 China 53 7,400 500 7.95 4.82 1,450 800 25.24 9.98 8.26 Côte d'Ivoire 146 20.90 12.16 54 Czech Republic 0.79 500 18,370 147 470 500 7.23 3.34 8.53 Guinea 55 United States 0.83 38.11 38.11 500 55,200 9.02 148 Cameroon 25.62 1,350 2,000 10.14 Antigua & Barbuda 56 0.84 9.26 11.32 700 13,300 149 1,050 800 19.98 7.98 9.12 Senegal 57 Bhutan 0.84 1.66 5.15 512 2,370 - 150 Kiribati 9.17 22.54 500 2,950 58 3,560 500 15.54 2.52 0.85 Ukraine 2,270 18.24 9.64 Honduras 151 1,024 35.57 61,310 59 Denmark 0.87 44.59 36.02 500 9.82 7.53 152 2,048 920 Tanzania 20.30 1,500 19.71 8.14 11,120 0.88 Malaysia 60 500 16.24 4.10 Gambia 153 500 9.86 61 Greece 0.88 16.64 21.50 600 22,657 154 1,000 19.63 6.05 9.95 Nepal 730 1,400 62 Argentina 0.89 9.96 - 13,480 600 Benin 155 890 7.61 10.26 19.75 63 Saudi Arabia 0.89 18.67 37.74 500 25,115 5.83 17.22 680 1,000 156 Afghanistan 10.28 13.00 64 Oman 0.93 24.99 1,024 16,853 157 500 18.04 8.45 10.35 Chad 980 65 0.95 4.60 Serbia 9.97 800 5,820 670 6.17 17.73 500 11.05 Uganda 158 500 66 Costa Rica 0.98 8.22 11.71 10,120 Micronesia 11.81 31.50 - 2,048 3,200 159 67 Moldova 1.01 2.15 6.71 500 2,560 160 Tajikistan 12.29 11.06 - 1,080 1,000 27.63 600 22.18 1.01 Cyprus 68 26,370 161 1,024 - 13.50 12.47 Yemen 1,299 Bulgaria 69 600 15.61 6.78 7,620 1.07 Papua New Guinea 12.77 23.84 28.15 1,500 2,240 162 1,024 70 St. Kitts and Nevis 1.10 13.70 16.99 14,920 370 4.00 5.87 163 Liberia 12.97 500 4.93 500 5,160 1.15 Jordan 71 10.15 820 2,500 19.44 8.87 12.99 Haiti 164 1,020 500 72 Cambodia 1.18 1.00 2.55 840 500 10.00 14.29 Zimbabwe 165 - 73 Egypt 1.18 2.99 10.40 500 3,050 8.45 700 1,024 22.74 14.49 Burkina Faso 166 74 9.68 4.37 1,000 1.18 Albania 4,450 14.67 167 Mali 7.95 21.09 650 500 500 6,410 8.33 6.44 1.21 Maldives 75 500 17.12 11.27 790 - 168 Comoros 15.20 5.81 1.21 Thailand 76 5,780 1,536 21.66 1,024 8.45 17.80 To g o 169 570 77 3,450 7.46 3.52 1.22 Cape Verde 500 500 1,830 31.43 31.59 20.71 Solomon Islands 170 Bahamas 78 1.23 21.50 18.93 800 20,980 21.05 9.65 27.16 1,000 550 171 Ethiopia 1,024 1.25 79 TFYR Macedonia 5,150 5.38 12.85 700 172 Sierra Leone 23.35 13.62 35.22 500 500 8,020 80 Turkmenistan 1.28 8.57 - 410 Niger 173 1,600 21.48 8.05 23.56 Tunisia 4,230 1,024 11.65 81 1.30 4.59 25.13 5.23 17.91 Malawi 500 174 250 600 17.18 5,490 82 Algeria 1.30 5.96 - 20.34 970 South Sudan 175 512 25.16 83 Lebanon 1.32 11.00 - 500 10,030 27.89 176 440 1,500 37.51 10.23 Madagascar 11.98 84 Indonesia 1.36 4.11 3,630 1,024 177 1,270 500 82.47 31.46 29.73 Mauritania 85 6,800 South Africa 1.37 7.76 17.35 500 Guinea-Bissau 127.27 58.33 178 1,024 550 143.68 Bosnia and Herzegovina 1.43 5.67 11.99 600 4,760 86 - 600 - 10.00 - Timor-Leste** 16.91 87 600 Seychelles 1.44 14,100 26.52 - Monaco** - 12.20 - 1,024 10.03 Philippines 1.50 750 4.37 88 3,500 - Congo (Dem. Rep.)** - 15.00 - 500 2,000 15,451 89 1.55 19.98 16.07 Barbados 560 - 15.40 - Somalia** - 13.89 5.59 1.57 Mongolia 90 4,280 1,126 - 1,024 32.21 27.73 - San Marino** 2,048 91 Trinidad & Tobago 1.61 26.88 32.14 20,070 - 500 - 267.26 - Syria** 2,048 25.26 14.99 11,130 1.62 Panama 92 1.95 1.67 Pakistan 1,000 93 1,400 6.54 Note: * Data correspond to the GNI per capita (Atlas method) in 2014 or latest available year adjusted with international inflation rates. ** Country not ranked because data on GNI p.c. are not available. Source: ITU. GNI p.c. and PPP$ values are based on World Bank data. Measuring the Information Society Report 2016 136

153 Chapter 4 Table 4.9: Mobi le-broadband prices, postpaid computer-based, 1 GB, 2015 Mobile-broadband, Mobile-broadband, Monthly Monthly postpaid computer-based postpaid computer-based GNI p.c., data data GNI p.c., (1 GB) (1 GB) Economy Rank Economy Rank allowance USD* allowance USD* as % of as % of (GB) (GB) USD PPP$ USD PPP$ GNI p.c. GNI p.c. 61,310 0.14 Denmark 1 1 24.91 2 2.37 Iraq 5.88 7.28 6,500 92 12.85 2 Austria 0.16 6.54 6.82 1 49,670 4.81 14.92 1 2,370 Bhutan 93 2.44 94 3 0.18 11.09 10.28 1 75,990 Lebanon 2.50 20.90 - 2 10,030 Luxembourg 103,630 3 36.52 21.18 2.55 Suriname 9,950 4 Norway 0.19 16.00 12.87 1 95 2.58 17.04 22.83 1 7,910 Grenada 96 5 46,304 1 7.44 8.26 0.21 Iceland 97 6 16.48 22.08 1 92,200 6,360 0.21 Peru 2.67 14.13 27.44 1 Qatar 13,300 98 2.77 30.72 37.57 Antigua & Barbuda 6 10.78 11.74 0.23 Sweden 7 1 61,610 St. Lucia 2.82 17.04 12.83 Netherlands 20.68 2 7,260 12.75 0.29 1 51,890 99 8 0.31 11.45 2 42,960 France 9 9,720 1 37.99 11.08 100 Gabon 2.92 23.67 2 14.47 0.31 101 Fiji 2.96 16.55 20.21 3 4,870 55,150 Singapore 10 12.01 3.10 10.20 1 1,250 2 7.72 4.44 0.35 Lithuania 11 15,430 102 Kyrgyzstan 2.98 1,570 64,540 India 2.98 3.90 13.63 1 103 12 Australia 0.35 18.78 16.11 1 1 Jamaica 2 5,150 12.83 104 55,200 19.69 16.50 16.50 0.36 United States 13 2.99 14 0.37 14.41 13.29 3 46,550 Ireland 1 15.48 7.24 3.03 Bolivia 105 2,870 12.11 11.09 0.39 34,270 Italy 15 2 St. Vincent and 24.90 106 3.36 18.52 2 6,610 the Grenadines 16 0.39 7.71 10.40 1 23,580 Slovenia 34.76 17.02 3.38 Dominican Rep. 107 1 6,040 17 - 15.07 0.40 Andorra 45,033 1 Belize 3.45 12.50 21.61 1 4,346 108 18 17.57 1 47,640 16.03 0.40 Germany 1,890 5.54 3.52 Viet Nam 109 2 13.68 19 1 14.93 16.53 0.41 Finland 48,420 6,410 3 110 Maldives 19.46 25.16 3.64 1 43,430 United Kingdom 20 15.28 0.42 12.60 5,630 111 Namibia 3.66 17.16 37.84 2 0.43 Korea (Rep.) 27,090 1 12.00 9.72 21 6,930 1 29.89 22.22 3.85 Dominica 112 Kazakhstan 10.50 4.46 0.45 1 11,850 22 4,400 Paraguay 3.93 14.41 30.57 2 113 Romania 23 9,520 5 7.67 3.74 0.47 3.96 1,590 1 18.98 5.24 Ghana 114 13,220 20.94 5.74 0.52 3 Russian Federation 24 3.96 3,070 15 23.98 10.14 Morocco 115 0.52 11.65 8.31 19,030 2 25 Estonia 26.76 4.28 3,920 2 13.99 El Salvador 116 1 20,070 8.84 26 0.53 Trinidad & Tobago 10.57 117 4.34 19.86 57.25 2 5,490 Algeria 0.53 21.35 47,260 27 20.94 2 Belgium Ecuador 4.41 22.40 38.52 1 6,090 118 40.52 88,032 10 25.73 0.55 Switzerland 28 10.96 119 Bangladesh 4.51 4.06 1 1,080 11.73 25,115 1 23.72 0.56 Saudi Arabia 29 120 1 Guyana 4.72 3,936 15.50 - 1 4.64 3,460 0.56 Sri Lanka 30 1.63 121 Pakistan 4.88 5.69 19.13 3 1,400 7.98 31 Uruguay 0.59 10.93 1 16,350 5 6.75 1,660 17.20 4.88 Lao P.D.R. 122 32 Spain 0.59 14.42 17.32 1 29,440 3,060 1 123 19.22 4.94 Palestine 12.61 33 Latvia 0.61 7.82 12.09 15,280 2 124 12.02 1 Kenya 1,290 5.08 5.46 34 41,070 19.84 21.61 0.63 New Zealand 1 5.55 Venezuela 125 12,615 1 47.97 58.39 1 11.03 3.76 0.63 Iran (I.R.) 35 7,113 5.56 126 Cape Verde 15.99 33.88 6 3,450 Malta 36 14.96 20,979 5 11.09 0.63 2,090 5.74 Uzbekistan 127 10.00 - 1 34.33 37,663 2 20.36 0.65 Brunei Darussalam 37 12.18 4.52 5.89 Tanzania 128 920 1 56.48 38 Macao, China 0.69 43.58 1 76,270 28.03 129 1 20.21 4,060 5.97 Samoa 39 Poland 0.70 7.96 15.85 25 13,690 16.95 9.13 6.41 Sudan 130 1,710 5 40 0.71 1 6.15 2.20 Georgia 3,720 3,500 Philippines 5 131 45.10 19.67 6.74 4 41 Belarus 0.72 4.39 - 7,340 36.15 19.46 6.81 Guatemala 132 3,430 2 42 United Arab Emirates 0.73 26.96 36.80 1 44,600 7.27 28.55 1 1,270 133 6.87 Myanmar 0.75 17,750 2 17.97 11.08 Slovakia 43 Nigeria 2 134 35.21 18.19 7.35 2,970 26,370 1 20.72 16.64 0.76 Cyprus 44 7.46 1 2,720 Congo (Rep.) 16.91 135 32.22 45 21,360 0.81 14.41 19.77 4 Portugal 1,870 30.86 12.00 7.70 Nicaragua 136 1 Turkey 46 1 13.89 7.32 10,830 0.81 21.16 8.45 9.66 Senegal 137 1,050 2 27.86 40,320 1 34.84 47 0.83 Hong Kong, China 41.23 138 Zambia 9.98 13.98 1 1,680 16.64 22,657 2 21.50 0.88 Greece 48 Afghanistan 139 10.28 5.83 17.22 1 680 49 Kuwait 0.89 36.56 58.82 100 49,300 1 140 Swaziland 10.57 31.27 85.28 3,550 2.75 3,560 1 Ukraine 50 16.95 0.93 141 11.00 66.36 134.05 1 7,240 Botswana 0.93 51 16,853 1 24.99 13.00 Oman 5 Honduras 11.63 22.00 142 42.91 2,270 52 18,370 2 24.39 14.19 0.93 Czech Republic 143 1,350 Cameroon 12.02 13.53 34.16 3 Canada 0.94 40.46 53 3 51,630 39.66 12.23 6.12 15.10 1 Mozambique 600 144 Bulgaria 6.23 7,620 2 14.33 0.98 54 145 1 Tajikistan 1,080 12.29 11.06 - 55 Hungary 1.05 13,340 11.66 22.83 3 146 12.47 Yemen 1,299 1 - 13.50 56 Bahrain 1.06 18.62 31.39 10 21,039 24.95 Nepal 12.65 7.70 1 730 147 57 Mauritius 1.06 8.53 15.51 1 9,630 36.93 3,160 1 35.07 14.02 Vanuatu 148 Croatia 58 12,980 2 19.20 11.52 1.06 700 Rwanda 14.03 8.18 21.64 3 149 7,590 Azerbaijan 1.08 6.83 - 1 59 150 Burkina Faso 14.49 8.45 22.74 1 700 60 Israel 1.09 32.13 28.89 1 35,320 151 15.56 17.24 51.09 2 1,330 Lesotho 14.44 1.16 St. Kitts and Nevis 61 1 17.91 14,920 152 To g o 570 17.80 8.45 21.66 1 1.18 6.66 62 1 6,800 South Africa 14.90 18.83 13.97 36.25 1 890 Benin 153 Albania 63 4,450 1.18 4.37 9.68 1 154 1 31.02 10.80 19.34 670 Uganda 64 7,320 7.20 1.18 Montenegro 2 14.06 1,670 S. Tomé & Principe 155 20.92 29.11 51.88 3 1 5,780 65 Thailand 1.21 5.81 15.20 156 Côte d'Ivoire 4 1,450 25.36 64.16 20.99 Japan 2 42,000 46.38 43.72 1.25 66 Ethiopia 157 27.16 1 9.65 21.05 550 11,530 1 21.32 12.01 1.25 Brazil 67 158 2 46.44 39.34 21.07 Papua New Guinea 2,240 12.85 1 5.38 1.25 TFYR Macedonia 68 5,150 159 Chad 21.37 17.45 37.23 1 980 3.25 1.28 Egypt 1 69 3,050 11.30 160 56.34 - 2 2,950 Kiribati 22.92 70 1 13.25 8.03 1.30 China 7,400 23.41 12.68 161 1 650 Mali 33.65 1.32 12.21 29.56 2 11,120 Malaysia 71 162 9.35 20.25 3 470 Guinea 23.88 5 72 Serbia 1.32 6.42 13.94 5,820 24.74 8.45 Niger 2 410 163 22.56 1.33 Costa Rica 6 73 10,120 15.98 11.22 164 1 - 16.91 25.68 Comoros 790 1.36 74 Indonesia 4.11 11.98 2 3,630 Haiti 165 820 7 43.21 19.72 28.86 Barbados 75 1.40 18.00 14.48 2 15,451 166 1,270 1 82.47 31.46 29.73 Mauritania 76 Mongolia 1.42 5.08 12.63 1 4,280 167 5,476 20 199.35 149.92 32.85 Angola 77 Argentina 1.45 16.25 - 1 13,480 250 168 37.93 7.90 27.04 1 Malawi 1 29.20 14,910 78 Chile 1.46 18.19 58.87 22.77 700 Sierra Leone 169 1 39.03 79 Bahamas 1.54 26.88 23.66 1 20,980 41.37 170 440 1 55.64 15.17 Madagascar 29.29 80 Seychelles 1.59 18.68 14,100 1 171 47.64 Solomon Islands 1,830 1 72.29 72.65 7,970 1 81 Colombia 1.64 10.90 24.28 172 South Sudan 53.05 42.88 - 1 970 82 Libya 1.67 10.87 - 1 7,820 173 370 5 35.25 24.00 77.84 Liberia 14.03 4,760 83 Bosnia and Herzegovina 1.67 6.63 3 174 - 80.00 114.29 Zimbabwe 840 1 5 84 Tunisia 1.81 6.37 16.18 4,230 143.68 550 1 175 Guinea-Bissau 127.27 58.33 1 9,870 1.91 85 15.71 26.97 Mexico 11.09 - - 1 San Marino** 176 12.88 3 4.30 2.02 Moldova 2,560 86 13.42 - Timor-Leste** 177 - - 12.50 1 5,160 6 87 Jordan 2.12 9.13 18.80 - Congo (Dem. Rep.)** 178 - 1 - 20.00 Panama 88 33.62 1 11,130 2.15 19.95 179 - 1 - 25.00 - Somalia** 2 89 Armenia 2.19 7.32 17.94 4,020 - 180 - Liechtenstein** 1 - 30.13 - 15.71 2.35 Turkmenistan 90 8,020 1 181 Monaco** - 43.25 - 10 - 91 2.00 2.35 Cambodia 1,020 5.09 1 Syria** 182 133.63 - 1 - - Note: * Data correspond to the GNI per capita (Atlas method) in 2014 or latest available year adjusted with international inflation rates. ** Country not ranked because data on GNI p.c. are not available. Source: ITU. GNI p.c. and PPP$ values are based on World Bank data. Measuring the Information Society Report 2016 137

154 The countries with the least affordable computer- Sri Lanka and Cambodia stand out for having • based mobile broadband are mainly LDCs. Indeed, the lowest postpaid computer-based mobile- 17 out of the 19 countries in which computer- broadband prices worldwide. Moreover, based mobile-broadband plans correspond to the prices in these two countries decreased more than 20 from 2014, when they were already among per cent of GNI p.c. are LDCs. Most of these countries have in common very the lowest in the world. The sustained low low income levels and a limited proportion of prices for both handset-based and computer- households with a computer (a prerequisite based mobile-broadband services in Sri Lanka for using a computer-based mobile-broadband go hand in hand with the very low mobile- 44 Even in some countries with higher income plan). cellular prices (see Section 4.2), and confirm levels, such as Angola, Kiribati and Papua and New the competitive environment in the Sri Guinea, the high cost of computer-based mobile- Lankan mobile market. Cambodia also enjoys intense competition in the mobile-broadband broadband plans (more than USD 35 per month) 46 makes them unaffordable for a majority of the but its benefits are still to be market, translated into lower mobile-cellular prices. population. On the basis of the price comparison taking account of the purchasing power of local Regional analysis of mobile-broadband prices currencies, some countries can be highlighted for having the lowest PPP mobile-broadband prices in The aggregate analysis of prices in terms of GNI each region (Table 4.6). The following observations p.c. shows that, on average, mobile broadband can be made based on the 2015 prices: became more affordable in all regions in 2015 (Table 4.7). All regional averages saw double-digit The lowest prepaid handset-based mobile- • drops in 2015, and the strongest improvement was broadband prices are found in Cambodia, recorded in the CIS, where computer-based prices Estonia, Rwanda and Georgia. In all of these as a percentage of GNI p.c. dropped by more than countries, mobile-broadband services are per cent. 50 4 per month, offered for less than PPP$ 5 whereas in 2014 prices below PPP$ continues to be the region with the most Europe were only found in Europe. This shows affordable mobile-broadband services, and also that countries from different regions have the one with the smallest differences across succeeded in reducing mobile-broadband countries in terms of GNI p.c. The CIS and the prices to very low levels. Americas have similar ranges for handset-based mobile broadband, but the average for the CIS is In the Arab States and the Americas, the • lower and decreasing faster. When considering lowest handset-based mobile-broadband postpaid computer-based services, the Americas prices are higher than the lowest prices has an average price in terms of GNI p.c. that is offered in other regions. Nevertheless, prices well above that of the CIS. Moreover, the Americas are relatively low in countries such as Uruguay saw the smallest reduction in the average price of and Sudan. In Africa, Liberia stands out for all regions. having some of the lowest prices globally, despite being a country with very low income region Asia-Pacific Prices across countries in the 45 levels. display a wide range, especially for postpaid computer-based plans, thus reflecting the • The lowest postpaid computer-based mobile- different stages of development of the mobile- broadband prices are significantly higher broadband markets in the region. Furthermore, than the lowest handset-based prices in the the relatively high average for postpaid computer- Americas, the Arab States and Africa. In Europe based prices in terms of GNI p.c. suggests that and Asia and the Pacific, there are a number providing affordable Internet access with large , which 10 of countries with prices below PPP$ data allowances remains an issue in several illustrates the lower computer-based mobile- countries in the region. broadband prices found in these two regions. The Arab States region has the widest range of handset-based mobile-broadband prices in terms Measuring the Information Society Report 2016 138

155 Chapter 4 of GNI p.c. of all regions. This is explained by the Africa: contrast between a few Arab LDCs where the service remains largely unaffordable and the high- Prepaid handset-based mobile-broadband prices income GCC countries where prices represent less per cent of GNI p.c. in 13 represent less than 5 African countries, including several countries in than 1 per cent of GNI p.c. The range is similar for postpaid computer-based services, although they which prices were above that threshold in 2014, are more expensive than handset-based mobile- such as Congo (Rep.), Rwanda, Botswana and São Tomé and Principe (Chart 4.19). This confirms that broadband in most Arab States. the service is becoming affordable in more and Africa is the only region in which the average more African countries. price of handset-based mobile-broadband plans Despite the progress made in reducing handset- represents more than 5 per cent of GNI p.c. This underlines that the service remains unaffordable based mobile-broadband prices in some low- for several segments of the population in many income African countries, such as Chad, Zimbabwe and Sierra Leone, in two thirds of the countries in African countries, although there has been a large reduction in prices in terms of GNI p.c. in 2015. per the region prices correspond to more than 5 cent of GNI p.c. Therefore, the service remains Postpaid computer-based services have much rather unaffordable in these countries. higher prices in Africa, with an average price per cent of GNI corresponding to more than 20 Apart from the African countries with the p.c. in the region. This average is four times that highest GNI p.c. levels (Mauritius, South Africa, of any other region and, coupled with the high Seychelles, Gabon and Namibia), there are other fixed-broadband prices in Africa (see Section 4.3), African countries with much lower income levels highlights that computer-based broadband access where handset-based mobile broadband is fairly remains largely unaffordable in the region. affordable, such as Cape Verde, Congo (Rep.) and Rwanda. A closer look at prepaid handset-based mobile- broadband prices as a percentage of GNI p.c. All African countries with handset-based mobile- provides additional insights into the differences 47 Based on a broadband prices that represent more than 10 per in affordability within each region. regional comparison, the following points can be cent of GNI p.c. are LDCs, except Zimbabwe. The highlighted: service is unaffordable for large segments of the population in these countries, and this may be an B per month) as a percentage of GNI epaid handset-based mobile-broadband prices (500 Chart 4.19: Pr M p.c. and data volume (cap) included, in the Africa region, 2015 and 2014 Note: The caps indicated refer to the 2015 prices. Source: ITU. GNI p.c. values based on World Bank data. Measuring the Information Society Report 2016 139

156 obstacle for the take-off of mobile broadband, Arab States that saw a significant fall in prices in considering that mobile-broadband penetration in 2015 include Jordan, Egypt, Algeria and Morocco. these LDCs is below 20per cent. The availability of data add-ons that can be attached to popular mobile prepaid bundles (e.g. Maroc Telecom’s data add-ons to the Jawal pass) Arab States: and flexible mobile bundles (e.g. Vodafone’s Flex prepaid plans in Egypt) are driving prices down in Prepaid handset-based mobile-broadband prices these countries. correspond to less than 5 per cent of GNI p.c. in most Arab States, the only countries where prices are clearly above that threshold being Mauritania, Asia and the Pacific: Comoros and Yemen (Chart 4.20). These three countries, together with Iraq and Libya – two Prepaid handset-based mobile-broadband plans per cent of GNI p.c. that represent less than 5 countries suffering ongoing armed conflict – have are offered in a majority of economies in the Asia the lowest mobile-broadband penetrations in the and the Pacific region, including Myanmar and Arab States (all below 25 subscriptions per 100 Vanuatu, the two countries that witnessed the inhabitants). This suggests that the affordability largest drop in prices in the region in 2015 (Chart of mobile-broadband services remains a major 4.21). barrier for the uptake of the service in Mauritania, Comoros and Yemen. In several SIDS, mobile-broadband remains unaffordable as prices correspond to more than High-income Arab States belonging to the GCC, 5 per cent of GNI p.c. Indeed, in countries such as such as Qatar, Kuwait and the United Arab Micronesia, Papua New Guinea and the Solomon Emirates, have the most affordable mobile- Islands, prepaid mobile broadband costs more broadband prices in the region. Moreover, in Qatar than USD 20 per month, and thus represents more and Kuwait, entry-level packages offer large data allowances (3 GB, respectively). Bahrain, GB and 5 per cent of GNI p.c. in these countries. than 10 despite having much lower income levels than the High mobile-broadband prices in terms of GNI high-income GCC countries, has achieved equally p.c. go hand in hand with the limited mobile- affordable mobile-broadband services thanks to broadband uptake in these countries (less than the relatively low prices offered in the country 15 mobile-broadband subscriptions per 100 GB). (USD 8 per month for 1 inhabitants). The example of other SIDS with lower prices and higher mobile-broadband uptake, Chart 4.20: Pr epaid handset-based mobile-broadband prices (500 M B per month) as a percentage of GNI p.c. and data volume (cap) included, in the Arab States region, 2015 and 2014 Note: The caps indicated refer to the 2015 prices. Source: ITU. GNI p.c. values based on World Bank data. Measuring the Information Society Report 2016 140

157 Chapter 4 Chart 4.21: B per month) as a percentage of GNI M Pr epaid handset-based mobile-broadband prices (500 p.c. and data volume (cap) included, in the Asia-Pacific region, 2015 and 2014 Note: The caps indicated refer to the 2015 prices. Source: ITU. GNI p.c. values based on World Bank data. such as Maldives, Fiji and Vanuatu, highlights Conversely, the increase of the data allowance GB in Uzbekistan drove prices from 500 MB to 1 the importance of affordable prices for fostering 6 to USD 10 per month) and thus the up (from USD mobile broadband. per cost of the service represented more than 5 cent of the country’s GNI p.c. in 2015. Some of the world’s most affordable mobile- broadband prices are found in the Asia and the The most affordable mobile-broadband services Pacific region, in countries such as Singapore and in the CIS were found in the Russian Federation. the Republic of Korea. In the latter, entry-level There were five additional countries (Belarus, mobile-broadband plans include very large data 48 Kazakhstan, Azerbaijan and Ukraine) allowances (above 30 GB). This shows that high- Georgia, where the cost of the service represented less capacity mobile broadband at very affordable prices (less than USD 5 per month) is possible. than 1 per cent of GNI p.c. and was therefore Other countries that stand out for having relatively affordable. affordable mobile-broadband prices despite their low GNI p.c. levels include Sri Lanka, Bhutan, In Ukraine, the roll-out of 3G networks in 2015 is Cambodia and Pakistan, all of them offering making mobile-broadband services available to a prepaid mobile-broadband plans at prices below larger share of the population, which previously 2 per month and representing less than 2 per USD only had narrowband mobile Internet access 49 cent of GNI p.c. Coupled with the (through GPRS technology). relatively affordable price of the service, the extension of 3G networks in Ukraine is expected Commonwealth of Independent States (CIS): to boost mobile-broadband adoption, which was among the lowest in the region in 2015 at less than Prepaid handset-based mobile-broadband prices 10 subscriptions per 100 inhabitants. correspond to less than 5 per cent of GNI p.c. in all CIS countries except Uzbekistan and Tajikistan (Chart 4.22). Thanks to the decrease in prices Europe: 8 per month in recorded in Kyrgyzstan (from USD 4.7 per month in 2015), prices were 2014 to USD Europe has the most affordable prepaid handset- per of GNI p.c. threshold brought below the 5 based mobile-broadband prices of all regions. there for the first time, while the data allowance Most countries in the region have prices that correspond to less than 1 per cent of GNI p.c. and GB per month. was increased from 500 MB to 1 Measuring the Information Society Report 2016 141

158 Chart 4.22: epaid handset-based mobile-broadband prices (500 MB per month) as a percentage of GNI Pr p.c. and data volume (cap) included, in the CIS region, 2015 and 2014 Note: Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. The caps indicated refer to the 2015 prices. Source: ITU. GNI p.c. values based on World Bank data. Norway, Sweden, Austria, Iceland, Estonia and powerful downward pressure on mobile prices Ireland have the most affordable prices worldwide, exerted by competition is thus also having an representing less than 0.15 per cent of GNI p.c. impact on the affordability of prepaid handset- (Chart 4.23). based mobile-broadband services. Indeed, more than one third of European countries saw a Competition in the European mobile markets reduction in prices in 2015. The price reduction continues to be strong and data is the main was greatest in Ireland, Israel, the Czech Republic, determinant in the pricing of mobile services, Cyprus and Bosnia and Herzegovina. which are often contracted as a bundle. The Chart 4.23: B per month) as a percentage of GNI Pr epaid handset-based mobile-broadband prices (500 M p.c. and data volume (cap) included, in Europe, 2015 and 2014 Note: The caps indicated refer to the 2015 prices. Source: ITU. GNI p.c. values based on World Bank data. Measuring the Information Society Report 2016 142

159 Chapter 4 countries suggest that lower prices could be Mobile-broadband services are affordable in possible if smaller data add-ons were offered. Europe not only because of the high income levels in the region, but also because of the low price of Uruguay and Chile stand out for having the the service. Indeed, in countries such as Estonia, most affordable mobile-broadband services in Lithuania, Poland, Serbia and Albania prepaid 7 per month, the region, with prices below USD mobile-broadband plans were offered at prices 4 per month in 2015. corresponding to less than 0.5 per cent of GNI p.c. below USD The high-income countries in North America have relatively affordable mobile-broadband services The Americas: (representing less than 1 per cent of GNI p.c.) despite much higher absolute prices (USD 22 per Prepaid handset-based mobile-broadband plans month in Canada and USD 38 per month in the priced at less than 5 per cent of GNI p.c. are United States). Brazil and Costa Rica, the countries offered in a majority of countries in the Americas in the region with the highest mobile-broadband region. These now include the Dominican Republic, penetration after the United States, have mobile- where the price reduction between 2014 and broadband prices that also correspond to less 2015 drove prices down below that threshold than 1 per cent of GNI p.c., thus confirming the 4.24). (Chart link between affordable prices and high mobile- broadband uptake. Haiti stands out for having the least affordable mobile-broadband services in the region, as well as very low mobile-broadband penetration (less than 4.5 Monitoring the price of bundled one subscription per 100 inhabitants). Honduras services and Nicaragua are in a similar situation, with prices representing more than 5 per cent of GNI p.c. Not only prices, but also pricing plans and models, and low mobile-broadband uptake (17 and seven change over time, usually to adapt to user subscriptions per 100 inhabitants, respectively). needs, address specific income structures and The relatively large data allowances in these three Chart 4.24: Pr epaid handset-based mobile-broadband prices (500 M B per month) as a percentage of GNI p.c. and data volume (cap) included, in the Americas, 2015 and 2014 Note: The caps indicated refer to the 2015 prices. Source: ITU. GNI p.c. values based on World Bank data. Measuring the Information Society Report 2016 143

160 For the purposes of collecting bundled telecommunication subscriber data, ITU has defined the notion of bundled telecommunication services as a prepaid or postpaid subscription that meets all of the following criteria: 1. A commercial offer that includes two or more of the following services: fixed telephony, mobile voice, fixed broadband, mobile broadband, pay TV. Marketed as a single offer, with a single invoice and single price for the set of services 2. included in the bundle. 3. Subscribed under conditions that cannot be obtained by adding single-play offers together. affordability issues, and ultimately attract new (EC, 2014, see Chart 4.26) . The same survey customers and increase operator revenues. highlighted the fact that domestic Internet access is more likely to come as a bundled service, and Over the last decade, the marketing and sale that triple-play bundles have increased by ten of telecommunication service packages have percentage points since 2007. become increasingly common. The move to higher broadband speeds and convergence over the According to the OECD, which started to monitor Internet Protocol (IP) are enabling operators to bundled services in 2010, bundling can “increase offer a range of services in one single package. In competition if it brings more choices, higher an attempt to offer customers a better deal, while quality, or lower prices to consumers from the at the same time increasing customer loyalty by facilities-based networks providing bundled offering a more comprehensive package, service offers” (OECD, 2015c). At the same time, bundling providers are putting together bundles of related – raises concerns about price transparency, since and in some cases unrelated – services and selling customers cannot easily compare how much they them as packages at a price that is lower than the have to pay for a specific service. It also carries the combined prices of the individual services. risk of consumer ‘lock in’ because bundles make it more complicated to switch from one operator Bundling can refer to both fixed and mobile to another for parts of the package. Bundling services and can include two, three or more may also lead to commercial practices considered services, including fixed and mobile voice anticompetitive, for instance by allowing telephony, broadband data and pay television. customers to buy a given telecommunication The packages created are often called bundles or, service only if purchased together with another 50 more specifically, when referring to the number of one. services involved, double-play, triple-play or quad- Thus, while monitoring the evolution of bundled play offers. They are also often generically known offers and prices is relevant from a regulatory as “multiplay offers”. and consumer-protection perspective, comparing the prices of bundled services can also be very A 2015 OECD report which looked at 12 major challenging. OECD economies found that an increasing number of operators no longer offer standalone services. In an increasing number of countries outside While most operators still offered standalone the OECD, operators are offering bundles that fixed-telephone services, “only 23 in 38 provide are similar to those available in OECD countries. standalone broadband services and the number For example, Latin America’s two largest mobile drops to only twelve operators if a standalone operators, América Móvil and Telefónica, both ” (OECD, 2015c, pay-television offer is requested offer double- and/or triple-play services combining see 4.25). On the demand side, a 2014 European fixed telephony, fixed broadband and pay TV in Union survey showed that almost half of all several Latin American countries, including Brazil, households in the European Union subscribed to Chile, Colombia, Ecuador, Honduras, Mexico bundled services, up from 38 per cent in 2009 Measuring the Information Society Report 2016 144

161 Chapter 4 Chart 4.25: Av ailability of standalone offers (% of operators) by service, selected OECD economies Pr oportion of households in the EU that subscribe to bundled (two or more) (left), Chart 4.26: telecommunication services, 2014 (right) Note: The left chart relates to the following 12 OECD countries: Australia, Canada, France, Germany, Italy, Japan, Republic of Korea, Mexico, Nether - lands, Spain, United Kingdom and United States. Source: OECD (2015c) (left chart), Eurobarometer (right chart). and Peru. Other incumbent operators in the ... and that the same types of service are b. region, such as CNT in Ecuador and Oi in Brazil, included in all bundles (e.g. all have fixed broadband and fixed voice) also offer double- and triple-play packages. In some cases, quadruple-play bundles including 51 ... and that all the included services have Although in a c. mobile services are also on offer. similar properties (e.g. all fixed-broadband number of countries with very limited fixed-line services fall within a given speed range). infrastructure, including some least developed countries, bundled services based on the fixed 2. Comparing multiplay services network are offered, mobile-based bundles are 52 likely to be more relevant for such countries. d. By building a “super-basket” with requirements and definitions for all services covered by the bundles. How to measure bundles For each bundle, by ensuring that all e. Since bundles are created from a set of individual requirements are fulfilled services, these may retain their original price structure, as shown in Figure 4.4. f. ... and, where a bundle does not offer all the services required, by filling the gaps Benchmarking prices for bundled services is far with the best possible service that can be more challenging than it is for individual services, bought individually. as the structures and properties of all the bundled services must be considered together. Basically, Methodologies have been developed for the two there are two main ways to ensure a proper like- types of comparison listed above, but they do for-like comparison: give different types of result and should be used for different purposes. The “comparison of similar 1. Comparing similar structure bundles structure bundles” is the simpler approach, and is useful for comparing the prices of bundles only. a. By ensuring that the bundles included in This method can be used with or without inclusion the comparison have more or less the of the individual service usage element, i.e. it is same structure (e.g. all are double play) possible to compare only the fixed prices, or also Measuring the Information Society Report 2016 145

162 Individual price structure vs. bundle price structure Figure 4.4: Source: Strategy Analytics based on research in OECD countries. to include usage elements in respect of voice Typical bundle structures calls, data, TV channel packages, etc. However, this type of comparison allows for analysis of only Bundles are most commonly made possible by one specific type of bundle at a time, potentially the utilization of an underlying network, whereby requiring a multitude of different, parallel analyses several, if not all, of the bundled services benefit that may create different outcomes, depending on from a common transport network. This can, which bundle combination is being considered. for example, include fixed-broadband and fixed- telephony services delivered over the copper line Where the objective is to compare the end- of traditional fixed-line operators; or fixed-voice, user cost for a specific set of services and to fixed-broadband and TV services delivered over compare across different types of offer, the the coaxial cable of a cable TV network. These more comprehensive “multiplay comparison” examples show that the types of service offered could be used, as this takes more elements into in bundles may vary with the type of network account and will yield more detailed results. operated by the provider. A cable TV provider will This type of approach is more flexible and can typically approach bundling from a different angle be used to compare different types of bundle; than a provider with a copper-pair based network. however, it does require a much broader and The most typical combinations found in OECD more detailed set of data, and its application to countries are shown in Table 4.10. broader international benchmarking exercises will be limited owing to the amount of data that With the development of LTE mobile networks, needs to be collected and processed. Data from the distinction between fixed and mobile services the majority of providers in a given country are is becoming less important, and there are now typically included, across all services. five-play offers available that are based either on mobile networks only, with fixed-location services based on mobile networks, or on a combination of fixed and mobile networks. Measuring the Information Society Report 2016 146

163 Chapter 4 Table 4.10: Mos t common bundle combinations found in OECD countries Computer-based Fixed TV Mobile voice Category Fixed voice Network mobile broadband broadband   Double play Fixed, copper    Triple play   Double play Cable TV    Triple play   Mobile Double play    Triple play Note: * Fixed voice under mobile bundles will typically mean fixed-location services based on mobile networks. Source: Strategy Analytics based on research in OECD countries. Also, for developing countries, where fixed- and cable TV providers. While the methodology network services may be more limited, typical adopted in 2010 focused on individual fixed- bundles may be based on mobile networks, with broadband services, it has become increasingly double-play combinations of mobile voice and difficult to subscribe to fixed broadband on its mobile broadband. Mobile-based bundles may also own. However, to facilitate further analysis, increasingly contain fixed-location services based the fixed-broadband information collected also on mobile networks. contains indicators for bundling with fixed-voice and TV services, which allows a basic analysis of the availability and pricing of double- and triple- play bundles based on fixed networks. From individual to bundled fixed-broadband price benchmarking Chart 4.27 shows the proportion of basic fixed- broadband offers that include either voice or The OECD has for many years been benchmarking TV services in the OECD countries. In 2016, 41 fixed-broadband services as offered by fixed-line Chart 4.27: Voi ce and TV services included in fixed-broadband offers, OECD Note: Total fixed-broadband offers include standalone/‘naked’ fixed-broadband offers, as well as offers that include fixed-broadband services combined with voice services and fixed-broadband services combined with TV services. Source: OECD/Strategy Analytics Ltd. Measuring the Information Society Report 2016 147

164 per cent of all fixed-broadband offers (including Chart 4.28: Pr ice range over and above standalone) across the OECD countries include the the price of the cheapest individual fixed- broadband offer, selected OECD countries, fixed-voice service element, as compared to 33 January 2016 per cent in 2011. Basic TV services are included in 24 per cent of 2016 fixed-broadband offers, as compared to 19 per cent in 2011. Not all providers in OECD countries offer bundles, but the vast majority of fixed-broadband providers will have some kinds of bundle that include a combination of fixed broadband with either fixed voice or TV services, or both. Price comparisons for bundled services Note: FBB refers to fixed broadband. FV refers to fixed voice. The pricing of bundled services in the OECD varies TV refers to television. Out of a total 40 OECD/EU countries, immensely. In some countries, the inclusion of this chart covers 27 countries included in the FBB / FV data, 30 countries in the FBB / TV data, and 36 countries in the FBB / FV fixed voice on top of fixed broadband is “free”, i.e. / TV data. only the voice calls are payable in addition to the Source: OECD/Strategy Analytics Ltd. basic fixed broadband prices. In other countries, it is more common to pay an additional 10 to 30 service in selected countries. It should be noted per cent or more on the fixed price in order to that in many countries the fixed-voice service is have the fixed-voice service in addition to fixed already included in the cheapest fixed-broadband broadband. offer. Chart 4.28 shows the additional price, over and above the best fixed-broadband price, for Adapting international ICT price comparisons adding a fixed-voice service and/or a TV service. The additional price is shown as a percentage The above analysis highlights the fact that, above the price of the cheapest fixed-broadband because of technological changes and new service in the country, for a broadband speed commercial practices in an increasing number above 1.5 Mb/s and at least 5 GB data usage. The of countries, telecommunication services are fixed-broadband service may differ between the bundled. Although countries with limited fixed three scenarios, as higher speeds are required to networks are likely to offer more mobile-network- support TV services. based bundles, standalone services will become less common as more countries move towards Among the OECD and EU countries, the actual high-speed networks and services. price of the cheapest fixed-broadband offer varies by country, and ranges from 9 PPP$ per This will make price comparisons of individual month (Lithuania, 2 Mbit/s) to 49 PPP$ per month services less relevant, and the monitoring of (Spain, 20 Mbit/s). At the same time, the cheapest ICT price developments more challenging, in all fixed-broadband offers may also include different parts of the world. Indeed, as more countries services, making comparisons more difficult. The offer bundled-only services, international price Spanish offer, for example, includes a fixed-voice comparisons will have to take this trend into service, while in Lithuania the addition of fixed account and the corresponding methodologies voice increases the cost considerably. The range of will need to be reviewed. One first step could be speeds for the cheapest fixed-broadband services to carry out an analysis of the types of bundle found in each country varies considerably, from 2 available in all parts of the world, so as to Mb/s to 100 Mb/s. understand and monitor the trend from individual to bundled telecommunication services. Chart 4.29 shows the added cost of bundling a fixed-voice service on top of the fixed-broadband Measuring the Information Society Report 2016 148

165 Chapter 4 ice difference for bundling fixed voice with fixed broadband, selected OECD/EU countries, Chart 4.29: Pr March 2016 Note: Where there is only a small marker on top of the fixed-broadband bar, the additional cost of having a fixed-voice service on top of the fixed-broadband service is zero, i.e. the cheapest offer already includes fixed voice. The OECD fixed broadband price benchmarking is based on 3 providers per country, with one xDSL provider, one cable TV provider, and the next large provider. However, the type of provider may vary as not all countries have extensive cable TV networks. Source: OECD/Strategy Analytics Ltd, March 2016. Measuring the Information Society Report 2016 149

166 Endnotes 1 See, for example, Facebook (2015) and ISOC (2015). Both reports highlight access/infrastructure, content and affordability as key barriers to ICT access and use. 2 World Bank (2016), page 218: “ Up-to-date price data will allow for comparisons, both within the country (between operators and over time) and between countries, using appropriate comparators. Armed with data, the next step is to work out in which part of the value chain for the supply of internet the market may be failing .” 3 See, for example, Facebook (2015) as well as Drossos, A. (2015), Eisenach, J.A. (2015) and The Economist (2015). 4 See Telecom Regulatory Authority of India (2016) and Abutaleb, Y. and Menn, J. (2016). 5 Since the study was published, India has banned all discriminatory tariffs for data services. 6 See ISOC (2015), p. 16. 7 For example, if country A and country B have the same price in USD for any given ICT service, but in country A prices of other products are in general cheaper (in USD), then applying PPP exchange rates to the price of the ICT service in country A will make this service more expensive. That is so because, compared to country B, in country A the same amount of USD (exchanged into national currency at market exchange rates) can buy more products or services. Therefore, the ICT service in country A is more expensive in terms of what could be bought with that amount in each country. The International Comparison Program (ICP) is the major global initiative to produce internationally comparable price levels. For more information on the PPP methodology and data, see icp. worldbank. http:// . org 8 GNI takes into account all production in the domestic economy (i.e. GDP) plus the net flows of factor income (such as rents, profits and labour income) from abroad. The Atlas method smooths exchange-rate fluctuations by using a three- year moving average, price-adjusted conversion factor. See: data. worldbank. org / indica tor/ http:// NY. GNP. PCAP. CD . 9 In the 2015 targets set by the Broadband Commission for Digital Development, 5 per cent of monthly income was the reference set for . The ITU Connect 2020 Agenda incorporates a similar reference value making broadband affordable Broadband services should cost no more than 5% of average monthly income in developing countries in Target 2.3.B: “ per cent of GNI p.c. is used as a rule of thumb to determine the affordability of ”. Throughout this section, the 5 by 2020 mobile-cellular prices. 10 per cent globally, the lowest growth rate in the last 10 years. Mobile-cellular subscriptions increased by 2 11 For example, in Senegal, Orange offers a series of prepaid packages called “Illimix”. These packages include bundles of GB of data. Most of these services ranging from 60 on-net and 10 off-net minutes to unlimited on-net calls, SMS and 1 packages have a validity of one day, except the top package that is valid for a week. 12 In parallel to the increase in mobile voice minutes, fixed-telephone minutes are decreasing in most countries, thus suggesting that there may be a fixed-mobile substitution effect in voice usage. 13 The number of SMS sent per subscription is decreasing in most developed countries, suggesting that many customers are substituting SMS with instant messages, using applications such as WhatsApp that operate on top of the Internet and require a mobile data connection. In the developing world, the downward trend in the volume of SMS is not so strong: there are only a few more countries in which the number of SMS sent is decreasing than countries in which it is increasing. 14 Mobile-cellular prices in Syria have been above the global and regional averages since ITU started publishing the ICT Price Basket. Indeed, the cost of the mobile-cellular basket in Syria has been above USD 80 per month since 2008, reaching a 95 per month in 2015. maximum of USD 15 Three transnational operators offer mobile-cellular services in Sudan: MTN, Sudatel and Zain. Unlike in other mobile markets in the Arab States where the incumbent retains a very large market share, none of the three operators in Sudan accounts for more than 45 per cent of total subscriptions. As a result, the Sudanese mobile market is very competitive, with a Herfindahl-Hirschman Index of 0.34, on a scale of 0 to 1, where 0 denotes perfect competition and 1 a monopoly. Data source: GSMA Intelligence, Q4 2015 data. 16 In December 2015, a second mobile licence was awarded to the operator Telma as a first step in the process of market liberalization. In addition, a public-private partnership has been created with the fibre-optic assets of Comoros Telecom in order to separate the incumbent’s retail and the wholesale operations. Together with the deployment of an additional undersea cable (FLY-LION) in a project coordinated by the World Bank, these initiatives aim to ease the international connectivity bottleneck and prepare the telecommunication market for its effective liberalization. Sources: “How the WDR16 Policy Framework is applied in the Union of Comoros”, World Bank ICT4D Blog, 13 January 2016; “Union des Comores : un procesus d’attribution de licence de communications électroniques réussi”, Press release from the Autorité Nationale de Régulation des TIC de Comores, 22 January 2016. 17 Data on mobile-cellular prices were available for 12 CIS countries in 2015. This compares with mobile-cellular price data available for 43 countries in Africa, 22 in the Arab States, 39 in Asia and the Pacific, 42 in Europe and 35 in the Americas. Measuring the Information Society Report 2016 150

167 18 Georgia exited the Commonwealth on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. 19 one2many . eu/ en/ portfolio/ dynamic- tariffing . www. http:// Source: One2Many’s Dynamic Tariffing available at: 20 www. Source: Digitata’s Case Studies available at: ta. com/ about- digita ta/ case- studies . http:// digita 21 Ibid. 22 Ibid. 23 It should be noted that in 2014 the price of fixed-broadband services fell in only six LDCs, remained the same in more than half of all LDCs, increased slightly in two LDCs, and increased substantially in two others (Uganda and Rwanda). The high prices in these latter two countries had a sizeable impact on the average, especially because complete price data for the period 2008-2015 are only available for 25 LDCs. In the remaining LDCs, fixed-broadband services were not available or not advertised during one or more years in that period. While in 2015 prices remained high in Uganda, they dropped substantially in Rwanda, as well as in a number of other countries, including Zambia and Mali. 24 Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. 25 The only exception being oil-rich Equatorial Guinea, in which the household final consumption expenditure per capita was USD 272 per month in 2015. Household final consumption expenditure is an indicator produced in the context of national accounts and therefore does not reflect income and consumption inequalities. As a result, depending on the distribution of income/consumption within the population, the actual economic wealth of most households may be significantly lower than the average value derived from the national accounts. Data from household income and expenditure surveys provide better indicators to measure household economic wealth, but data availability is limited in developing countries. For more information, see pp. 140-146 in ITU (2014a). 26 Measured in terms of household final consumption expenditure per capita, income levels are seven times higher in Ireland than in Equatorial Guinea, the LDC with the highest household final consumption expenditure per capita (of all those with data available). 27 Of 44 LDCs with data available on fixed-broadband prices in 2015, 37 had a fixed-broadband penetration rate below one subscription per 100 inhabitants. In Bangladesh, there were 2.4 fixed-broadband subscriptions per 100 inhabitants, and 3.6 in Bhutan. 28 In Burundi, CBINET offers contention ratios for ADSL services of 1:4. Contention ratios for common residential fixed- broadband plans are around 1:15 in most countries. 29 For an example of the issues relating to ISP access to the wired local loop in many LDCs, see ITU (2013). 30 For more details on the development opportunities that IoT brings, see Chapter 5 in ITU (2015), and ITU and Cisco (2016). 31 Data on mobile-broadband prices were collected from 2010 to 2014 through the ICT Price Basket Questionnaire, which is sent out annually to all ITU Member States/national statistical contacts. Since 2015, data on mobile-broadband prices have been collected by ITU from operators’ websites. 32 Source: ITU calculation based on GSMA data on LTE deployments. 33 “On the move” refers to use of the Internet while mobile via a mobile cellular telephone or other mobile access devices, for example, a laptop computer, tablet or other handheld device. For developing countries, it refers to Internet use through the above mentioned devices connected to a mobile phone network and if the location is away from “home”, “work”, “place of education”, “another person’s home” and “community and commercial access facilities”. For European countries, it refers to Internet use through the above mentioned devices “away from home and work”. For more information on the definitions of Internet use by location, see page 55 in Manual for Measuring ICT Access and Use by tions/ www. itu. int/ en/ ITU- D/ Sta tistics/ Pag es/ publica http:// manual2014. Households and Individuals 2014 available at: aspx. 34 Of the 25 countries (16 developing and nine developed) that reported mobile and fixed Internet data traffic in the period 2013-2014, 80 per cent reported more fixed-broadband Internet traffic than mobile-broadband Internet traffic, despite the fact that in all of them there were more mobile-broadband subscriptions than fixed-broadband subscriptions. 35 See, for instance, OECD (2013) for an analysis of how the cost of the smartphone affects the total cost of ownership in OECD countries. Further information on the recent average selling prices of smartphones in Sub-Saharan Africa and their impact on affordability of mobile services in the region are presented in GSMA (2016c). 36 According to the World Bank, 38 per cent of the population in Bhutan lived in urban areas in 2014. 37 Source: Bhutan Telecom, http:// www. bt. bt/? page_ id= 183 . 38 posts/ http:// www. tashicell. com/ Source: TashiCell, tashicell- 3g- services . 39 Source: Bhutan Telecom, www. bt. bt/? http:// page_ id= 3683 . Measuring the Information Society Report 2016 151

168 40 By the end of 2015, Bhutan Telecom’s LTE coverage was below 1 per cent of the population of the country. 41 tashicell- tashicell. com/ posts/ Source: TashiCell: www. 4g- services . http:// 42 http:// tashicell. com/ posts/ 3g- See for instance TaahiCell’s campaign to exchange 3G data cards with 4G USD devices, data- card- exchange- offer . 43 The change in mobile-broadband prices between 2014 and 2015 is analysed in local currency so that the effects of exchange-rate fluctuations or changes in GNI p.c. are screened out. 44 ITU’s definition of computer includes desktop computers, laptop (portable) computers and tablets (or similar handheld computers). Smartphones are not considered as computers. 45 Liberia has the second lowest GNI p.c. of all countries for which data are available on mobile-broadband prices, higher only than Malawi’s. 46 For more information on the Cambodian mobile-broadband market, see Box 4.7 in ITU (2015). 47 Prepaid handset-based mobile broadband is selected for the analysis because it is the mobile-broadband service which holds the greatest potential for development. Indeed, handset-based subscriptions are much more widespread than computer-based subscriptions, and most handset-based subscriptions in the world are prepaid. This suggests that the affordability of prepaid handset-based mobile-broadband services will be a key enabling factor if the “mobile miracle” (i.e. the mass uptake of regular mobile-cellular services) is to be replicated in the broadband arena. 48 Georgia exited the CIS on 18 August 2009 but is included in the ITU/BDT administrative region for the CIS countries. 49 http:// For more information on 3G deployments in Ukraine, see Kyvistar’s press release of 27 May 2016 at kyivs tar. www. ua/ ru/ kr- 620/ press_ cent er_ new/ new s/? id= 58604 . 50 For a discussion on bundling in the European Union, including market trends and regulatory issues regarding the bundling of services, see BEREC (2010) and OECD (2015c). 51 See, for instance, the quadruple-play offer from Claro in Brazil: http:// www. combomulti. com. br . 52 In Senegal, for example, Orange’s Home+ offer includes fixed-telephony, fixed-broadband Internet and TV services. See . http:// www. home. sn/ Orange Senegal Home+ at: Measuring the Information Society Report 2016 152

169 Chapter 5. Measuring mobile uptake

170

171 Key findings Mobile phone adoption has largely been monitored based on mobile-cellular subscription data since these are widely available and regularly collected and disseminated by regulators and operators. At the end of 2016, there are almost as many mobile-cellular subscriptions as people on earth and 95 per cent of the global population lives in an area that is covered by a mobile-cellular signal. However, since many people have multiple subscriptions or devices, other metrics need to be produced to accurately assess mobile uptake, such as the number of mobile phone users or mobile phone owners. Many people still do not own or use a mobile phone . Household data from developing countries show that a significant part of the population does not use mobile-cellular services at all. In developing economies where recent household data is available, close to 20 per cent of the population, on average, are still not using a mobile phone. The proportion of mobile-phone ownership is even lower, especially in large developing economies such as Bangladesh, India, Indonesia and Pakistan, where more than 40 per cent of the population do not own a mobile phone. Most people who do not own or use a mobile phone are among the youngest (5-14 years old) and the oldest (>74 years old) segments of the population . Usage and ownership penetration rates amongst these age groups are much lower than amongst the rest of the population. Among the 15-74 age group, 85 per cent or more of the population owns or uses a mobile phone in the countries where data are available. Significant gender gaps exist in mobile-phone adoption and the gap is larger for mobile- . Many women in developing countries rely on phone ownership than for mobile-phone use someone else’s mobile phone or SIM card to access mobile-cellular services. The gender divides are associated with differences in income and educational attainment, and reflect other types of social divides. Most people not owning or using a mobile phone have lower incomes and are less educated. People living in rural areas are less likely to own or use a mobile phone than people in urban areas. In several developing countries, sizeable segments of both the urban and the rural population do not yet own or use a mobile phone. Although basic mobile infrastructure is available for most of the global population living in rural areas, rural populations tend to have lower incomes and lower education levels, which are in turn linked to lower mobile-phone ownership and usage. . It is the cost of the handset, rather Affordability is the main barrier to mobile-phone ownership than the cost of the service itself, which is often reported as the main barrier to owning a mobile phone. Another important barrier is the lack of perceived benefits. In communities where overall mobile uptake is low, mobile phone use is perceived to have fewer benefits since fewer community members are also using this mode of communication. Other barriers include poor network quality and lack of ICT skills necessary for accessing the Internet through a mobile phone. Universal use of mobile-cellular services has not been achieved yet. Policy-makers and the telecommunication industry in developing countries should focus on targeted policies for promoting mobile adoption. As the 2030 Agenda for Sustainable Development has pledged, ICTs can be a strong empowerment tool, and no one should be deprived of their benefits because of economic, educational, social or technical barriers. Measuring the Information Society Report 2016 155

172

173 Chapter 5. Measuring mobile uptake development enablers for those at the bottom of Introduction 5.1 the pyramid. Mobile-cellular services have witnessed Data on the population living in an area covered by unprecedented growth over the last 15 years, and a mobile signal and data on the number of mobile- have taken a prominent place among the world’s cellular subscriptions show that the two are almost most ubiquitous technologies: in some countries, equal to the global population (Chart 5.1). Country- more people have access to mobile-cellular specific disaggregated data, however, show that services than to a bank account, electricity or some segments of the population do not yet use clean water (World Bank, 2012). In the time-span or own a mobile phone, for example numerous covered by the Millennium Development Goals women in low- and middle-income countries (MDGs, 2000-2015), the number of mobile-cellular (GSMA, 2015; GSMA and LIRNEasia, 2015) and subscriptions has increased ten-fold, from 738 the lowest-income segments of the population in million to over 7 billion. developing countries (InfoDev 2012a, 2012b; CKS 1 Consulting, 2012; Galpaya, H. et al., 2015). This phenomenon has been described as the “mobile miracle”, and has driven broad societal Despite the large numbers of mobile-cellular and economic transformations. Indeed, mobile- subscriptions at global level, hundreds of millions phone usage has changed not only the way people of people in the world do not use or own a mobile communicate, but also the way they plan their phone today. This chapter analyses the available daily lives, organize themselves socially, and access data on mobile-phone ownership and usage, educational, health, business and employment provides insights into who does not own or use opportunities (Castells et al., 2007; GSMA et al., a mobile phone today, and highlights some of 2010; PewInternet, 2012; Vodafone, 2013). the main barriers preventing such people from connecting to the basic mobile network. Developing countries have embraced mobile technologies following a “mobile first” approach, These data will contribute to informing some insofar as other ICTs have very limited reach in of the targets identified in the 2030 Agenda the developing world. This has spurred genuine for Sustainable Development, for example ICT innovation from the developing world in the those under Goal 5, which calls for “the use of mobile arena, in forms such as low-cost and multi- enabling technology, in particular information SIM mobile phones, low-value prepaid refills and and communications technology, to promote the mobile-money services (World Bank, 2012). These empowerment of women” (United Nations, 2015). innovations have contributed to making mobile- This chapter also aims to raise awareness among cellular services more pervasive and inclusive, thus ICT policy-makers regarding the current status transforming mobile technology into a powerful of mobile-phone uptake and the challenges that development tool for empowering entrepreneurs, must be overcome in order to strengthen digital women, young people, vulnerable groups – in fact, inclusion through universal mobile-cellular uptake, virtually anyone (Broadband Commission, 2013; particularly in developing countries. UNDP, 2012). As the international community moves on from 5.2 Moving beyond subscriptions: the MDG timeframe to that of the sustainable phone owners and users development goals (SDGs, 2015-2030), the question remains as to whether universal mobile- The most widely available indicator for measuring phone usage has been achieved as a means the uptake of mobile-cellular services is the of fulfilling the pledge of the 2030 Agenda for number of mobile-cellular subscriptions (Box 5.1). Sustainable Development that “No one must be This indicator is reported by telecommunication left behind”. This question is even more relevant operators and therefore provides information given the role that mobile technologies can play as Measuring the Information Society Report 2016 157

174 obal mobile-cellular subscriptions and population coverage, 2008-2016* Gl Chart 5.1: Note: * Estimates. Source: ITU. from the supply side. In contrast to demand- The mobile-cellular-subscription indicator is thus 2 as it refers to registered SIM side statistics, which are established based on becoming obsolete cards rather than people, and should therefore national household surveys that include questions be interpreted with caution when drawing on mobile-phone ownership and usage, supply- conclusions on the uptake of mobile-cellular side statistics are less expensive to establish services. Specifically, the following issues should because their collection does not entail the costs be considered: associated with conducting surveys. Indeed, supply-side statistics simply require administrative one person can own and Double counting: • notifications from mobile operators and are often use multiple subscriptions. For instance, a publicly disclosed by listed operators as part of single subscriber may have one subscription their annual or quarterly reports. at home and one at work, or decide to have several subscriptions with different operators The subscription (i.e. a SIM card in most in order to benefit from special offers or lower cases) is the basic revenue-generating unit for 3 The latter is more common on-net prices. mobile operators, and data on mobile-cellular in mobile markets that are predominantly subscriptions have therefore traditionally been prepaid and in which mobile termination rates used by the mobile industry to gauge size and are relatively high. trends in mobile markets. When the metric was originally introduced over a decade ago, most According to GSMA Intelligence, unique mobile mobile users had a single subscription and it subscribers tend to use on average 1.45 SIM was therefore statistically valid to assume that cards globally, while subscribers in countries subscriptions equaled subscribers. However, as such as South Africa, UAE, Saudi Arabia, the the price of handsets and services fell and prepaid Russian Federation or Côte d’Ivoire use more services became popular and coverage ubiquitous, 4 than two SIM cards each on average. These it became common in many markets for users to findings are in accordance with available data have multiple SIM cards and mobile devices, from on dual-SIM handsets. Indeed, more than handsets to tablets and other data-centric devices. Measuring the Information Society Report 2016 158

175 Chapter 5 Definitions of selected indicators to measure mobile-cellular uptake included in the Box 5.1: Core List of ICT Indicators of the Partnership on Measuring ICT for Development Supply side: Mobile-cellular subscriptions : Number of subscriptions to a public mobile-telephone • service that provide access to the public switched telephone network (PSTN) using cellular technology. The indicator includes the number of postpaid subscriptions, and the number of active prepaid accounts (i.e. that have been used during the last three months). The indicator applies to all mobile-cellular subscriptions that offer voice communications. It excludes subscriptions via data cards or USB modems, subscriptions to public mobile data services, private trunked mobile radio, telepoint, radio paging, M2M and telemetry services. Demand side: : proportion of individuals who Proportion of individuals using a mobile-cellular telephone • used a mobile telephone in the last three months. A mobile (cellular) telephone refers to a portable telephone subscribing to a public mobile telephone service using cellular technology, which provides access to the PSTN. This includes analogue and digital cellular systems and technologies such as IMT-2000 (3G) and IMT-Advanced. Users of both postpaid subscriptions and prepaid accounts are included. : An individual owns a mobile cellular Proportion of individuals who own a mobile phone • phone if he/she has a mobile-cellular phone device with at least one active SIM card for personal use. It includes mobile-cellular phones supplied by employers that can be used for personal reasons (to make personal calls, access the Internet, etc.) and those who have a mobile phone for personal use that is not registered under his/her name. It excludes individuals who have only active SIM card(s) and not a mobile phone device. Source: ITU. active mobile-cellular subscriptions in 2014, as 50 per cent of Android users in developing against 2.1 million registered subscriptions in countries such as Nigeria, Bangladesh and the same year. In Benin, there were 8.7 million Tanzania have a dual-SIM handset, as against active mobile-cellular subscriptions in 2014, as fewer than 5 per cent in the United Kingdom against 10.6 million registered subscriptions. and the United States (OpenSignal, 2015). GSMA Intelligence estimates that 7 per cent of global subscriptions (excluding M2M) were Disparities in multiple SIM usage and inactive in 2015. ownership make the relationship between mobile-cellular subscriptions and mobile New regulations concerning the taxation users different across countries, and may also 5 of registered subscriptions have prompted conceal inequalities within countries. operators in some countries to clean up the • subscriber base. For example, a new tax on Counting of inactive subscriptions: it is numbering resources approved in Guatemala difficult to track active mobile-cellular in 2014 prompted operators to return subscriptions accurately in markets that are 6 The maintenance of several numbers, and caused a 22 per cent predominantly prepaid. a prepaid subscription does not necessarily reduction in the number of active mobile- 7 imply a payment and there tend to be cellular subscriptions reported. Likewise, a significantly more registered subscriptions new law passed in Ecuador in 2015 imposing than active subscriptions. For example, in the taxation on the number of active mobile Central African Republic, there were 1.2 million lines produced a 20 per cent drop in the Measuring the Information Society Report 2016 159

176 8 number of mobile-cellular subscriptions. New Moreover, these issues mean that data on mobile- regulations concerning SIM card registration cellular subscriptions are often not comparable also tend to have the side-effect of cleaning across countries, as they may have different up the subscription base. This was the case, impacts in each country depending on national for instance, in the Lao People’s Democratic circumstances. The latter include consumer Republic, where the enforcement of behaviour, such as sharing and multi-SIM mandatory SIM card registration in 2015 led ownership patterns, as well as market conditions to a 19 per cent decrease in the number of and regulation, such as the off-net pricing policies mobile-cellular subscriptions . of operators, taxation of numbering resources and mobile termination rate regulation. The difficulties in collecting accurate data on active mobile-cellular subscriptions may lead A comparison between data from household to overestimation of actual mobile-cellular surveys on mobile-phone usage and data from uptake in some countries. telecommunication operators on mobile-cellular subscriptions for countries in which both metrics • Non-consideration of shared subscriptions: are available (Chart 5.2) allows the following sharing of mobile phones and mobile- the conclusions to be drawn: cellular subscriptions is not uncommon in the The fact that there are more mobile-cellular lowest income segments of the population, 1. subscriptions than inhabitants does not mean but it is not reflected in mobile-cellular that everyone in a country uses a mobile subscription figures. phone. Studies conducted in developing countries in The number of mobile-cellular subscriptions 2. Africa and Asia show that numerous people per 100 inhabitants provides a significant over- may not have a subscription but still use estimation of the actual number of mobile- mobile-cellular services by sharing someone phone users. else’s subscription and/or phone (Galpaya, H. et al., 2015; James, J., 2010). The relationship between mobile-cellular 3. subscriptions per 100 inhabitants and mobile- Other studies have shown that mobile-phone phone users varies enormously between sharing decreases as the percentage of phone countries. owners increases (Wesolowski, A. et al., 2012; InfoDev, 2012a), and that people at the bottom As more and more economies reach a situation in of the pyramid share their mobile phones, which there are more mobile-cellular subscriptions mainly with family members – usually the male 9 This than inhabitants in the country (Chart 5.3), data head of the household with the spouse. suggests that mobile phones may be used as on mobile-cellular subscriptions provide little household devices in some contexts. additional information on progress made in terms 10 Even in countries of mobile-cellular uptake. with fewer subscriptions per 100 inhabitants, The different sharing patterns are not captured subscription data are of limited use in identifying in the data on mobile-cellular subscriptions, people that do not yet use or own a mobile phone. which may therefore not correctly reflect mobile-phone usage in some segments of the It can thus be concluded that the traditional way of population. calculating mobile-cellular penetration – dividing total mobile-cellular subscriptions in a country These three different issues may apply by the number of inhabitants – has become simultaneously to the data on mobile-cellular obsolete. It should be complemented or replaced subscriptions in a specific country, and have by indicators using data relating to individuals, contradicting effects. For instance, the counting thus ranging from 0 to 100 per cent of the total of inactive subscriptions may overestimate the population. In addition, the target population actual number of subscribers, but not counting the should be considered carefully, because there may sharing of subscriptions may underestimate the be population groups that cannot use a mobile number of users. Measuring the Information Society Report 2016 160

177 Chapter 5 Chart 5.2: Mo bile-cellular subscriptions and mobile-phone users, selected economies, 2015 Note: For mobile-phone users, the age scope of each survey is indicated in brackets. Source: ITU except Myanmar, sourced from LIRNEasia. olution of mobile-cellular penetration, 2010, 2015 Chart 5.3: Ev Source: ITU. individuals who use a mobile-cellular telephone. phone or subscribe to mobile services (e.g. the 11 These indicators have been defined according very young). to international standards by the Partnership on Measuring ICT for Development and adopted by The following sections look at two of the metrics the United Nations Statistical Commission. collected from household surveys: individuals who own a mobile-cellular telephone, and Measuring the Information Society Report 2016 161

178 in Cuba since 2013, which increased from 13 5.3 How many people actually own or per cent to 33 per cent in 2015. Nevertheless, use a mobile phone? mobile uptake on the island remains among the lowest in the world. Data on mobile-phone usage and ownership are collected through national household surveys. In the countries included in Chart 5.4, there • In these surveys, individuals are asked whether were a total of about 525 million people they have used a mobile phone and, in a separate not using a mobile phone, corresponding to question, whether they own a mobile phone. 18 per cent of the total population in these Based on the answers from the respondents, economies. Nevertheless, mobile-network the totals for the country are estimated. Since coverage reached 94 per cent of the total questions included in household surveys are population in these countries. This indicates addressed directly to people, they sidestep that lack of coverage is not the main barrier the methodological pitfalls relating to inactive to mobile-phone usage, at least in these subscriptions and the double counting of countries. subscribers. Moreover, they allow the shared use of mobile phones to be taken into consideration. Mobile-phone ownership is related to mobile- phone usage, although there are some Since 2005, ITU has been collecting data on differences between the two indicators. Owning mobile-phone usage based on nationally a mobile phone is usually linked to greater representative household surveys carried out by privacy, convenience and security for the user. national statistics offices. The data available for Furthermore, mobile phone ownership can the period 2013-2015 are presented in Chart 5.4 also help increase economic and professional together with data from the Financial Inclusions opportunities, especially for entrepreneurs or the 12 When comparing the results Insights Program. self-employed. For those at the bottom of the across countries, it is important to consider the pyramid, owning a mobile phone may be a way of different age scopes of the surveys, because most having a personal address providing access to a of the people who do not use or own a mobile bank account, microfinance and basic information phone belong to the youngest or oldest segments on health, agriculture or education (GSMA et al., of the population (see section 5.4) and are 2010; GSMA, 2015; UNCTAD, 2014). unequally represented in the surveys. The importance of mobile-phone ownership as The following conclusions can be derived from an empowerment tool has been recognized in these data: the SDGs. Indeed, the global indicator framework agreed by the United Nations Statistical • In most economies with data available, more Commission in 2016 (ECOSOC, 2016) includes than 80 per cent of the population use a the indicator “Proportion of individuals who own mobile phone, and almost universal usage has a mobile telephone, by sex” to monitor SDG been reached in Qatar, the Republic of Korea, 5 (“Achieve gender equality and empower all Bahrain and Hong Kong (China). women and girls”). Data are available for only three least • ITU has been collecting data on mobile-phone developed countries (Bangladesh, Myanmar ownership from national statistics offices since and Uganda) and two low-income countries 2015. Based on the limited data available (Chart (Tanzania and Uganda). It is therefore 5.5), the following conclusions can be drawn: impossible to draw conclusions on mobile- phone usage in the world’s poorest nations. In half of the economies with available data, • more than 75 per cent of the population owns • Less than 70 per cent of the population used a mobile phone. In countries such as Bahrain, a mobile phone in Cuba, the Islamic Republic UAE, the Republic of Korea and Malaysia, of Iran, Myanmar and Puerto Rico in 2013. almost everyone owns a mobile phone. Cuba stood out with only 11 per cent of the population using a mobile phone in 2013. Over 70 per cent of the population are mobile- • More recent data on mobile-phone ownership phone owners in Azerbaijan, Colombia, show that significant progress has been made Measuring the Information Society Report 2016 162

179 Chapter 5 Individuals using a mobile-cellular telephone, 2015 or latest available year Chart 5.4: Note: The age scope of each survey is indicated in brackets. * 2014 data. ** 2013 data. Source: ITU for all countries except India, Indonesia, Kenya, Nigeria, Pakistan, Tanzania and Uganda, for which data are sourced from Financial Inclusion Insights, and Myanmar, sourced from LIRNEasia. include Asian countries with large populations, Uruguay and Palestine, whereas in the such as Bangladesh, India, Indonesia and remaining countries less than 60 per cent of Pakistan. the population own a mobile phone. These Measuring the Information Society Report 2016 163

180 Individuals who own a mobile-cellular telephone, 2015 or latest available year Chart 5.5: Note: The age scope of each survey is indicated in brackets. * 2014 data. ** 2013 data. - Source: ITU for all countries except Bangladesh, India, Kenya, Nigeria, Pakistan, Tanzania and Uganda, for which data are sourced from Financial Inclu sion Insights, and Myanmar, sourced from LIRNEasia. • in 2013, 39 per cent in Myanmar in 2015, 40 In the four LDCs with data available, large per cent in Bangladesh in 2015, and 55 per segments of the population do not own cent in Uganda in 2015. In other countries a mobile phone: only 13 per cent of the from sub-Saharan African with data available, population owned a mobile phone in Burundi Measuring the Information Society Report 2016 164

181 Chapter 5 the percentage of the population owning a (about 7 percentage points), suggesting that in mobile phone is greater: about 75 per cent these countries only a fraction of mobile users of the population in Cameroon, Kenya and do so using someone else’s device or SIM card. Tanzania, and 85 per cent in Nigeria. More data would be needed to understand the In all other countries with data available, overall mobile-phone ownership situation in differences between usage and ownership 13 Africa, as well as in LDCs. are large (14 percentage points or more), indicating that many people access mobile- Chart 5.6 presents the comparison between • cellular services by sharing a device and/or mobile-phone ownership and mobile-phone SIM card. Differences between mobile-phone usage in the few countries for which both ownership and usage are particularly large in indicators are available for the same year. Bangladesh and India, where around half of Differences between ownership and usage the mobile-cellular users do so using someone are small in Morocco (less than 2 percentage else’s SIM card or device. points) and moderate in Oman and Nigeria Co Chart 5.6: mparison between individuals who own a mobile-cellular telephone and individuals who use a mobile-cellular telephone, 2015 or latest available year Note: The age scope of each survey is indicated in brackets. * 2014 data. ** 2013 data. Source: Financial Inclusion Insights for all countries except Colombia, Morocco and Oman, for which the source is ITU, and Myanmar, sourced from LIRNEasia. Measuring the Information Society Report 2016 165

182 Data from the 11 countries included in Chart phone ownership and usage might be larger. 5.6 appear to confirm previous findings on the Likewise, differences within countries may inverse relationship between mobile-phone also be revealed by studying sub-national data ownership and the sharing of devices and/or (Wesolowski et al., 2012). 14 Indeed, in contexts of low mobile- SIM cards. phone ownership, many people who do not 5.4 Who does not own or use a own a mobile phone use the service through sharing; however, as mobile-phone ownership mobile phone? increases, sharing decreases and mobile- phone usage tends to be through owned devices and SIM cards. Age These findings suggest that a significant part An analysis of mobile-phone ownership across of the population in developing countries may different age groups indicates that most non- be using a mobile phone without owning it, owners belong to the youngest and the oldest and that in some developing countries the segments of the population, as could be expected mobile market may still have ample space (Chart 5.7). to evolve towards higher mobile-ownership rates. These changes in the way people Indeed, for the few countries with data available, access and use mobile-cellular services the percentage of individuals owning a mobile can only be monitored through data from phone exceeds 85 per cent in the 25-74 age group, nationally representative household surveys and it also exceeds that value in the 15-24 age including ICT questions, which are lacking group in all economies with data available except in most countries. In particular, more data Palestine. On the other hand, fewer than 45 per would be needed to determine the situation cent of children up to the age of 14 own a mobile in low and lower-middle income countries, phone in these countries, Costa Rica being the in which the differences between mobile- only exception. Nevertheless, even in Costa Rica Individuals who own a mobile-cellular telephone, broken down by age group, 2015 or latest Chart 5.7: available year Note: * 2014 data. ** 2013 data. Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 166

183 Chapter 5 the proportion of 5-14 year-olds owning a mobile 15 do not use a mobile phone. In the remaining phone (64 per cent) is well below that of people countries, the percentage of non-users is also owning a mobile phone in the 15-74 age group. significant: in Belarus, Brazil and Hong Kong (China), about 20 per cent of children below the Low mobile-phone ownership among children age of 15 do not use a mobile phone, and in the in these developing countries may indicate that other countries over 35 per cent do not. This either they do not use mobile-cellular services suggests that a significant proportion of children or they access them using their parents’ mobile in developing countries do not use mobile-cellular phone and SIM card, rather than having their own services. subscription and device. Mobile-phone usage is also limited among Another age group with low mobile-phone individuals over the age of 74: in most countries ownership is that of people over the age of 74: with data available, fewer than 60 per cent of fewer than half of them own a mobile phone in the senior adults use mobile-cellular services. The only countries with data available. This suggests that, economy that stands out with a higher level of despite the spread of mobile-cellular services, mobile-phone usage among the eldest is Macao many older people still do not benefit from mobile (China), where almost 80 per cent of individuals 15 communications in their everyday lives. Ten aged 75 or over use a mobile phone. years ago, when mobile-cellular services took off in these countries, these individuals were 65 years Data from GSMA Intelligence suggest that children old and may have lacked the ICT literacy skills below the age of 18 are two times less likely to to use mobile phones, or simply may not have use a mobile phone than adults in developed perceived the benefits of doing so. Section 5.5 countries, and four times less likely in developing analyses in more detail the different barriers to countries. Data from Financial Inclusion Insights mobile-phone ownership and use. confirm that older age groups are less likely to use mobile-cellular services (see, for instance, FII data ITU data are available for only a few developing on India in Chart 5.9). countries from the Americas and the Arab States region, and therefore may not be representative of other regions. However, a survey carried out Gender by GSMA Intelligence, covering 24 developing countries and 30 developed countries, also finds The disaggregation of data on mobile-phone usage that mobile-phone ownership is low among by gender shows that the percentage of male children (Table 5.1). users is higher than that of female users in most countries (Chart 5.10), although differences are small in most economies (less than 4 percentage Table 5.1: Individuals who own a mobile-cellular points). telephone, broken down by children and adults, 2015 Countries where the gender gap is slightly greater Mobile-phone include Mauritius, Nigeria, Morocco, Oman and Developed Developing ownership Uganda. Four countries stand out as having large Children (5-17) 39% 17% gender gaps: Pakistan (64 per cent of female 90% Adults (18+) 95% mobile users as against 81 per cent of male mobile users in 2015), the Islamic Republic of Iran (56 Source: GSMA Intelligence. as against 78 per cent, 2013), Bangladesh (71 as against 82 per cent, 2015) and India (79 as against An analysis of data on mobile-phone usage reveals 90 per cent, 2015). differences in uptake across age groups similar to those observed for mobile-phone ownership On the other hand, the percentage of female (Chart 5.8), although from a higher baseline mobile users is slightly higher than that of male because usage is overall higher than ownership. mobile users in some countries of the Americas region, such as Colombia, Brazil, Panama and In half of the countries with data available, 50 Jamaica. per cent or more of children below the age of Measuring the Information Society Report 2016 167

184 Chart 5.8: Individuals using a mobile-cellular telephone, broken down by age group, 2015 or latest available year Note: * 2014 data. ** 2013 data. Source: ITU. Measuring the Information Society Report 2016 168

185 Chapter 5 Chart 5.9: Individuals owning a mobile-cellular telephone and using a mobile-cellular telephone, broken down by age group, India, 2015 Source: ITU based on Financial Inclusion Insights. Chart 5.10: Individuals who own a mobile-cellular telephone (left) and using a mobile-cellular telephone (right), broken down by gender, 2015 or latest available year Note: The age scope of each survey is indicated in brackets. * 2014 data. ** 2013 data. The percentages of female/male mobile-phone users/owners are calculated as a proportion of total female/male population in each age group. - Source: ITU for all countries except Bangladesh, India, Kenya, Nigeria, Pakistan, Tanzania and Uganda, for which data are sourced from Financial Inclu sion Insights, and Myanmar, sourced from LIRNEasia. Measuring the Information Society Report 2016 169

186 Data show larger gender gaps in mobile-phone Urban/rural ownership than in mobile-phone usage. Indeed, in all countries with large gender gaps in mobile- Another factor linked to ICT uptake is the phone usage the gap is even larger in mobile- individual’s geographical location, in particular phone ownership (e.g. men are twice as likely as whether people live in urban or rural areas (see women to own a mobile phone in Bangladesh, Chapter 6 for an analysis of how this affects India and Pakistan). Other countries with large Internet uptake). Individuals living in rural areas gender gaps in mobile-phone ownership include are less likely to use mobile phones than those Burundi, Indonesia and Myanmar (more than 10 in urban areas (Table 5.2). Unlike with other ICT percentage points), and sizeable gender gaps exist services for which infrastructure may not be in in the Republic of Korea, Oman and Thailand (more place, basic mobile infrastructure is available than 5 percentage points). The fact that countries for most of the global population living in rural 17 with very different income levels display significant areas. However, rural populations tend to have differences in mobile-phone ownership between lower incomes and lower levels of education, men and women suggests that low mobile-phone which in turn are linked to lower mobile-phone ownership among women may not only be an usage. In addition, certain minority groups, such issue in poorer countries. as indigenous people, may be more represented in rural areas (UNDESA, 2009) and require targeted Because of the limited data available, only cautious policies in order to embrace mobile technologies. conclusions can be drawn on the current gender gap in mobile uptake in the low-income countries Table 5.2: Individuals who use a mobile-cellular and LDCs. However, data for the four LDCs with telephone, by urban/rural, 2015 or latest available data available (Bangladesh, Burundi, Myanmar and year, % Uganda) show that women are far less likely to Rural Urban own a mobile phone than men. 82 84 Oman** (5+) All these findings coincide with the results of Bangladesh** 79 87 (N/A) recent research in low- and middle-income 72 54 Iran (I.R.)** (all) countries pointing to the fact that in these countries over one billion women do not own a 4 13 Cuba** (6+) mobile phone (GSMA, 2015). Belarus* (6+) 94 85 78 Bolivia* (5+) 53 74 88 Brazil* (10+) Level of education 77 Colombia* (5+) 85 82 El Salvador* (10+) 74 Recent research carried out in Myanmar found 92 Jamaica* (14+) 89 that the gender gap in mobile-phone ownership 66 89 Panama (10+) varies according to the income level of the Note: The age scope of each survey is indicated in brackets. * 2014 household, being higher among lower-income data. ** 2013 data. The percentage of urban/rural mobile-phone users households (GSMA and LIRNEasia, 2015). ITU data is calculated as a proportion of the total urban/rural population in each on mobile-phone ownership disaggregated by age group. Source: ITU. level of education show that the large majority of non-owners (men or women) belong to those Despite lower mobile-phone usage in rural areas, segments of the population with the lowest some countries have larger urban than rural educational attainment (Chart 5.11). These findings populations and consequently most individuals highlight the fact that gender gaps are associated not yet using a mobile phone in these countries with differences in income and educational live in urban areas (Chart 5.12). This suggests that attainment, and therefore reflect other types of initiatives to promote digital inclusion through 16 social divides. mobile uptake should target both urban and rural population segments and focus on the underlying causes of non-ownership or non-use of mobile- cellular phones. A summary of these causes is presented in the following section, based on the Measuring the Information Society Report 2016 170

187 Chapter 5 Individuals who do not own a mobile-cellular telephone, by gender and level of education, Chart 5.11: 2015 or latest available year Note: The age group scope of each survey is indicated in brackets. * 2014 data. ** 2013 data. Source: ITU. 2. Determining the characteristics (e.g. age, results of recent quantitative and qualitative research on the main barriers to mobile uptake. gender, location) of those not yet owning or using a mobile phone. Why do people not own or use 5.5 Identifying the specific barriers to mobile- 3. cellular uptake faced by those not yet owning mobile phones? or using a mobile phone. As reflected in the structure of this chapter, the Reliable information on the barriers to mobile- statistical efforts to inform policy-makers in the cellular uptake is essential to ensuring the success field of mobile-cellular uptake can be divided into of policies in this area, because it constitutes three interlinked stages: the basis for determining what kind of initiatives can be most effective in making the benefits 1. Gathering quantitative evidence showing to of mobile-cellular services available to a larger what extent mobile-phone ownership and proportion of the population. usage are not yet universal. Measuring the Information Society Report 2016 171

188 Individuals who do not use a mobile-cellular telephone, by gender and rural/urban, 2015 or Chart 5.12: latest available year Note: The age scope of each survey is indicated in brackets. * 2014 data. ** 2013 data. Source: ITU. As could be expected, affordability is a greater Figure 5.1 presents a summary of the findings of obstacle for low-income individuals not three studies on barriers to mobile-cellular uptake. owning a mobile phone than for the rest of These studies combine quantitative research, the population not owning a mobile phone such as nationally representative surveys including (InfoDev, 2012a, 2012b). From a gender questions on barriers to mobile uptake, and perspective, cost in particular is a barrier for qualitative research, such as face-to-face in-depth women, because they often have less financial interviews with selected respondents and focus- independence (GSMA, 2015). group discussions. The following conclusions can be drawn based on the findings of these studies: Other surveys focusing on mobile Internet Affordability is the main barrier to mobile uptake have confirmed that affordability is also • uptake (this finding is consistent across a relevant barrier. For mobile Internet uptake, all surveys). In those surveys that provide however, affordability is as important as ICT information on the different cost elements, skills and less of an obstacle than relevant local the mobile device is mentioned as the main content (GSMA, 2016a, 2016b). cost barrier along with, to a lesser extent, credit recharge. The SIM card is considered Relevant local content is not mentioned as a the most affordable of all cost elements. This major obstacle to mobile-phone ownership indicates, contrary to the predictions of some and usage because, most probably, those not supply-side estimations (Facebook, 2015), that yet using and owning a mobile phone would the main cost barrier for new mobile-phone start by using voice and SMS services which owners is the cost of the device. are less dependent on local content. Lack Measuring the Information Society Report 2016 172

189 Chapter 5 in barriers to mobile-phone ownership and usage Ma Figure 5.1: Note: “% respondents” refers to the proportion of respondents that identified a given issue as a barrier to mobile-phone ownership or to both mobile- phone ownership and usage, depending on the survey. Source: ITU based on GSMA and LIRNEasia (2015), Galpaya et al. (2015), GSMA (2015) and data from Research ICT Africa. of mobile money and other development- of ICT skills is mentioned as an obstacle to enabling applications) or simply as mobile mobile-phone ownership and usage, but is phones become more available in a given not identified as prominently as for mobile community, the perceived economic value Internet uptake, probably because new mobile of a phone for work, business and even the phone owners and users will most likely use coordination of household affairs may increase basic mobile services which require fewer and convince more people to become mobile- ICT skills than accessing the Internet using a phone owners (CKS Consulting, 2012; GSMA mobile phone. and LIRNEasia, 2015). The of mobile- lack of perceived benefits • phone ownership among some segments of • Despite the high mobile-network coverage the population is one of the reasons most reported in most countries, including in the cited for not owning a mobile phone. In some developing world, respondents across different cases, the head of the household may own a as poor network quality countries identify mobile phone and the other members may a barrier to mobile uptake. Indeed, dropped not feel the need for it, as the device may calls and long waiting times for establishing a be considered a family or common property call are cited as a common problem not only (GSMA and LIRNEasia, 2015). Through in several African countries but also in some information campaigns (e.g. on the benefits Latin American and Asian economies. Although Measuring the Information Society Report 2016 173

190 poor network quality is more often a problem Conclusions 5.6 for rural respondents, it is also mentioned by urban dwellers. Data presented in this chapter show that, although there are almost as many mobile- These findings point to a rather surprising cellular subscriptions as people globally and paradox: while most of the regulatory and almost everyone lives in an area that is covered press attention today is focused on the roll-out by a mobile-cellular signal, hundreds of millions of long-term evolution (LTE) networks and of people do not yet own a mobile phone. This discussions on 5G, the quality of the basic apparent paradox is explained by the fact that mobile network may still be an issue in several many people have multiple SIM cards, whereas developing countries. others own neither a single SIM card nor a mobile device. Lack of ICT skills is a barrier for some • individuals who do not own a mobile phone, Some of those who do not own a mobile phone and slightly more of an issue for women than use mobile-cellular services by borrowing men owing to the former’s lower education someone else’s phone or SIM card (with the levels and lower confidence with technology. constraints this implies in regard to usage). or broken devices are consistently Phone theft However, even taking account of users who do cited as a barrier to use of the service, not own a mobile phone, a significant portion of although less frequently than other issues the population still does not use mobile-cellular such as affordability or the lack of perceived services at all. benefits. Lastly, trust in the operator or agent is also sometimes cited as an issue, particularly Policy-makers and operators should therefore be among rural, less-educated segments of the aware that universal use of mobile-cellular services population who are not yet mobile-phone has not yet been achieved, and that specific owners. initiatives should be undertaken to ensure that no one is left behind. This section has focused on the main barriers that face those who do not yet own or use a The first step in designing efficient policies in mobile phone at all. Beyond these entry barriers, this area is to clearly understand who should be however, other, significant obstacles may affect targeted, i.e. the characteristics of those who the intensity and type of use. For instance, lack do not yet own or use a mobile phone. Although of ICT skills is less of a barrier to making use of these may vary from one country to another, a the basic features of a mobile phone (e.g. making number of common characteristics emerge from and receiving calls), but a major obstacle when it the global analysis. comes to more sophisticated uses, such as Internet access using a smartphone. Most of those who neither own nor use a mobile phone belong to the youngest (age 5-14) and As more and more people own and use a mobile oldest (age >74) segments of the population. In phone, focus is shifting to the barriers preventing parallel, there are large gender gaps in mobile- some mobile-phone owners from making full use cellular uptake in some countries. These gender of the device. For example, it has been observed divides are associated with differences in income that in low-income countries female mobile-phone and educational attainment, and therefore reflect owners use their device less frequently and for less other types of social divide. Indeed, most people sophisticated services than men (GSMA, 2015). who neither own nor use a mobile phone belong Similar usage gaps have also been found in other to the less-educated and lower-income segments segments of the population, such as low-income of the population. Lastly, although people living in individuals (InfoDev, 2012b). These usage gaps rural areas are less likely to own or use a mobile will require particular consideration from both phone than people in urban areas, sizeable industry players and policy-makers in order to segments of both the urban and rural populations ensure that everyone is able to fully benefit from of several developing countries do not yet own or mobile-cellular and mobile-broadband services. use a mobile phone. Measuring the Information Society Report 2016 174

191 Chapter 5 A key element in designing efficient policies to services. One barrier that may be relevant in some foster mobile-cellular uptake is to have reliable countries is poor network quality, which may information on the specific barriers to mobile- deter mobile-phone usage. In addition, lack of ICT phone ownership and use. Even though the skills may be an obstacle in specific population specific relevance of each barrier varies from one segments, especially for accessing the Internet country to another, some of them are consistently using a mobile phone. identified in surveys from different countries. Policy-makers and the telecommunication Affordability is the greatest barrier to mobile- industry in developing countries could build phone ownership and usage. Rather than the on the evidence presented in this chapter, cost of the service itself, the cost of the handset complemented with national data, with a view to is more often considered the main obstacle to adopting targeted policies for promoting mobile ownership. Another major barrier is the lack of uptake. As the 2030 Agenda for Sustainable perceived benefits. This obstacle can be greater Development has pledged, mobile services can be in communities where lower overall mobile a strong empowerment tool, and no one should be uptake makes the service less valuable because deprived of their benefits on account of economic, of network effects. In other cases, it may simply educational, social or technical barriers. be linked to lack of information on mobile Measuring the Information Society Report 2016 175

192 Endnotes 1 References to income groups are based on the World Bank classification by income group, available at: datahelpdesk. worldbank. org / knowledg ebase/ articles/ 906519- world- bank- coun try- and- lending- group s . https:// 2 GSMA Intelligence: Measuring mobile penetration www. gsmain telligence. com/ resear ch/ 2014/ 05/ measuring- https:// penetr ation/ 430/ . mobile- 3 An analysis of the Ukrainian regulator regarding the high number of mobile-cellular subscriptions in Ukraine concluded that: “the number of mobile subscriptions is higher than the population in the country. This situation refers to the fact that one person has several SIM-cards of different operators. However, there are still the residents having no mobile phone in Ukraine, mainly they are children and seniors. One of the main reasons of buying several SIM-cards is the substantial difference between the tariffs for on-net calls and off-net calls. This led to the fact that nearly 94% of mobile outgoing traffic falls on on-net calls” (NCCIR, 2013). In Kenya, a survey from Research ICT Africa found that the main reason for people at the bottom of the pyramid (i.e. with the lowest incomes) having more than one SIM card was to reduce inter-network calling costs (InfoDev, 2012a). 4 Source: GSMA Intelligence Database 2016. 5 For instance, Chinese mobile users in rural areas use 1.18 mobile-cellular subscriptions on average, whereas those in urban areas use on average 2.03 subscriptions (GSMA, 2014). 6 75 per cent of all mobile-cellular subscriptions were prepaid at the end of 2015. 7 , which in Chapter 2, Section 1, See Decree 22-2014 of the Congress of the Republic. It includes the Ley de Ajuste Fiscal imposes a tax of GTQ 5.00 per telephone line (fixed and mobile) on operators. 8 See Article 34 of the Ley Orgánica de Telecomunicaciones approved in February 2015, available at http:// www. gob. ec/ wp- con tent/ uploads/ downloads/ 2016/ 05/ Ley - Org%C3%A1nic a- de- Telec omunicaciones. telecomunicaciones. . pdf 9 An ICT household survey carried out in Kenya in 2012 found that 25 per cent of the people at the bottom of the pyramid owning a mobile phone shared it with a family member, usually the spouse (InfoDev 2012a). A similar study carried out in South Africa revealed that 10 per cent of mobile phone owners at the bottom of the pyramid shared their mobile phones on a weekly basis (InfoDev, 2012b). 10 In the ICT Development Index, for instance, a reference value of 120 subscriptions per 100 inhabitants is set for the mobile-cellular-subscription indicator. Beyond that value, all countries are given the same score, because differences beyond that threshold are not considered to be indicative of the actual mobile-cellular uptake, but rather of the market structure. 11 www. gsmain telligence. com/ GSMA Intelligence: Measuring mobile penetration resear ch/ 2014/ 05/ measuring- https:// mobile- ation/ 430/ . penetr 12 The Financial Inclusion Insights (FII) Program is a partnership between InterMedia and the Bill and Melinda Gates Foundation. Under the FII Program, nationally representative household surveys are conducted annually for Bangladesh, India, Indonesia, Kenya, Nigeria, Pakistan, Tanzania and Uganda. The FII questionnaires include specific questions on http:// mobile-phone usage and ownership in these countries. More information on the FII Program is available at: org . Data presented in this chapter from FII have been calculated combining the survey questions on mobile- finclusion. phone usage and ownership with those on SIM card use and ownership. Mobile-phone ownership is defined as having “a mobile-cellular phone device with at least one active SIM card for personal use”. Likewise, mobile-phone usage is defined as using both a mobile phone and an active SIM card. 13 In its 2011 ICT household and individual survey, Research ICT Africa (RIA) included questions on mobile-phone ownership which made it possible to produce an overview of the situation in sub-Saharan Africa (see Research ICT Africa, 2012). However, given the growth recorded in most African mobile-cellular markets over the last five years, the present situation may well be very different from that reflected in the 2011 RIA surveys. This is confirmed by the data from Cameroon, Nigeria and Tanzania, where less than 50 per cent of the population owned a mobile phone in 2011 according to RIA’s surveys, but where mobile-phone ownership today exceeds 75 per cent according to recent ITU and FII data. 14 For the 11 countries in Chart 5.6, the Pearson correlation between mobile-phone ownership and individuals using a mobile-phone without owning it renders a result of -0.76. The Pearson correlation ranges from -1 to 1, being -1 a total negative correlation, 0 no correlation and 1 a total positive correlation. 15 For an example of how ICTs can assist in creating better conditions of life for older adults, see the Active and Assisted www. aal- http:// europe. eu . Living Programme in Europe, 16 For a discussion of gender divide in ICT access and usage within the perspective of other digital divides, see for instance www. apc. org / https:// en/ blog/ inside- inf ormation- society - how - digital- divide- has . 17 ITU estimates that over 95 per cent of the world’s total (i.e. both urban and rural) population is covered by a mobile signal. Measuring the Information Society Report 2016 176

193 Chapter 6. Internet user and activity trends

194

195 Key findings In 2016, people no longer online, they are online. An increasingly ubiquitous, open, fast and go content-rich Internet has changed the way many people live, communicate and do business, delivering great benefits for individuals, governments, organizations and the private sector. Yet many people are still not using the Internet, and many users do not fully benefit from its potential. A better understanding is needed of who is online and who is not, and how people are using the Internet, in order to create a more inclusive information society. The offline The benefits of the Internet are still unavailable to over half the world’s population. population – 3.9 billion people globally – is disproportionately female, elderly, less educated, lower income and rural. To bring more people online, it is important to focus on reducing overall socio-economic inequalities. Education and income levels are strong determinants of whether or not people use the Internet. Most people have access to Internet services but many do not actually use them. The spread of 3G and 4G networks across the world had brought the Internet to more and more people. In 2016, mobile-broadband networks covered 84 per cent of the world’s population, yet with 47.1 per cent Internet user penetration, the number of Internet users remains well below the number of people with network access. While infrastructure deployment is crucial, high prices, poor quality of service and other barriers are serious obstacles to getting more people to enter the digital world. The full potential of the Internet remains untapped, especially for low-income and less Internet users with higher levels of education make greater use of more educated users. advanced services, such as e-commerce and online financial and government services, than Internet users with lower levels of education and income, who use the Internet predominantly for communication and entertainment purposes. This suggests that many people do not benefit fully from the opportunities of the Internet. Indeed, the Internet is liable to reinforce existing inequalities, instead of addressing them. Access to the Internet is not enough; policy-makers must address broader socio-economic inequalities and help people acquire the skills they need to take full advantage of the Internet. This is in line with a more integrated development approach, like that adopted in the 2030 Agenda for Sustainable Development, which highlights that development challenges are linked and cannot be met in isolation. A data revolution is needed to better understand who uses the Internet, where and how. Reliable and valid data on Internet use are currently not available for many developing countries, and almost non-existent for least developed countries. This lack of data is a serious challenge for ICT policy-makers, investors and content producers. The United Nations has called for the use of new data sources, including big data, to supplement official statistics. ITU is responding to this call and has recently launched a new project, “Big Data for Measuring the Information Society”, which explores how big data from the ICT industry can help enhance data collections, benchmarks and methodologies for measuring the information society. Measuring the Information Society Report 2016 179

196

197 Chapter 6. Internet user and activity trends An increasingly ubiquitous, open, fast and and to identify barriers faced by non-users. It content-rich Internet has changed the way many will analyse Internet use and socio-economic people live, communicate and do business. variables, such as age, gender, income and level of Internet uptake has been found to bring great education, and examine the types of activity that benefits for people, governments, organizations different users engage in online. and the private sector. It has opened up new The results of the analysis demonstrate a link communication channels, provided access to between lower levels of educational attainment information and services, increased productivity and fostered innovation. It has also created a new and lower Internet penetration rates among Internet economy or digital economy, and new specific groups, including women and the elderly. Internet players like Google and Facebook have The offline population remains disproportionately illiterate, poor, rural, elderly, and female. This become some of the world’s leading businesses in 1 just over a decade. chapter also finds that three decades into the World Wide Web and almost half a century after By the end of 2016, close to half of the world’s the first e-mail was sent, communication, in population will be using the Internet. This particular the use of social media, is the key activity 4 Many Internet users, in compares to less than two per cent two decades of many of those online. particular those with lower levels of education and ago, when people connected to the Internet using income, make only very limited use of the Internet a modem that would take time to dial-in via a and are not able to exploit its full potential. telephone line. Access to the Internet was then limited mainly to e-mail and chat services and very 2 These findings suggest that the Internet is liable to limited amounts of content. reinforce existing inequalities and leave the most Nevertheless, not everyone has benefited equally vulnerable population groups even further behind. from the rapid expansion of the digital economy While the mobile phone has (rightly) been hailed and its opportunities. Globally, 3.9 billion people, as a development enabler that provides crucial more than half the world’s total population, are communication channels, access to information still offline. In addition, being online does not and new services to large population groups, necessarily mean that people are able to take full including the poor and less privileged, the full advantage of the potential benefits of the Internet. potential of the Internet remains largely untapped. An important step in bringing more people into the digital economy is to understand the profile of To turn the Internet into a truly universal tool current users and the ways in which they use the for development, policy-makers must tackle not Internet. It is also important to identify barriers to only the supply-side challenges of the Internet, connecting the other half of the world’s population including infrastructure deficiencies and high – those who currently do not use the Internet and prices, but also the demand-side barriers that remain excluded from the information society. exist outside the ICT ecosystem. This means addressing broader socio-economic inequalities. At the same time, increased access to and use Above all, people need to acquire not only the of the Internet come with a growing number of necessary digital skills but also analogue skills, such challenges, with debates increasingly focused on as basic literacy and numeracy, in order to exploit the negative effects of ICTs, and on how to make the potential of the Internet. ICT policy-makers the Internet safer and protect users’ privacy. must act as part of a larger Internet ecosystem Spending large amounts of time online has been in order to empower people and make Internet linked to depression, decreased social skills, and content easily accessible to disadvantaged groups. other neurological complications (Cash, H. et al., ICT policies must also be linked to investments 3 2012). in education in order to develop the necessary human skills and raise levels of education, and thus The aim of this chapter is to develop a better bring more people online. understanding of how people use the Internet, Measuring the Information Society Report 2016 181

198 This chapter uses ITU data and information from Internet access device and location complementary sources to analyse Internet usage across gender, age, level of education and The outstanding success of the Internet economy other variables. It points to persistent data gaps, and Internet user growth have been triggered especially in developing countries, and highlights by increased competition in the telecom market, the need to produce more and granular data, and the resulting decrease in the price of Internet to exploit and understand the potential of new access services and devices, and the expansion of sources of data, in particular big data. and technological advances in mobile broadband 8 . The increasing computing power of networks smartphones has enabled the growth of new How the Internet has changed – 6.1 applications and functionalities – including the sharing of photos and videos and accessing of and changed the world information on health, education and location services – while the rapid spread of 3G and 4G Over a relatively short period of time, the Internet networks around the world has helped transform ecosystem has changed and grown considerably the use of the Internet from fixed locations, in terms of technologies, players (including such as home, work and schools, to anywhere Internet users, content providers, ISPs and 9 with network coverage . ITU data show that network operators) and content. Technological an increasing number of people are accessing developments, the open nature of its governance the Internet while mobile using a mobile device and the technical architecture of the Internet have connected to a mobile phone network (Chart 6.1). profoundly changed the Internet, both in terms of In many developing countries, and in particular the its spread and pervasiveness. least developed countries (LDCs), Internet access is almost exclusively via mobile networks. online, they go In 2016, many people no longer are online. Internet access in many parts of the (mostly This trend has not only led to new business developed) world is fast, ubiquitous and mobile. models, such as the on-demand economy which Internet users read, shop, bank and date online, uses a person’s location to offer convenient access thanks to a growing number of websites, services to goods and services – such as the transportation and applications that did not exist a decade ago. service company Uber, and restaurant delivery services that find and deliver food from At the same time, the Internet has had an restaurants located within a person’s proximity; important economic impact, including in terms mobile connectivity has also impacted how people of the new businesses and business models it use the Internet with new devices, services and has brought about. In 2016 the FT 500 ranking, applications especially tailored to a mobile lifestyle which ranks companies based on their revenues, – for example, smart watches for exercising, and included several of the new Internet companies, th reality games such as Pokémon Go. ), Alphabet such as Amazon.com (ranked 18 th ), Facebook (which includes Google, ranked 36 th th 5 Although “home” remains the place where people ) and eBay (ranked 300 The FT 500 ). (ranked 157 most frequently use the Internet, in particular included over 50 companies from the technology in developed countries (Chart 6.2), “in mobility” and telecommunication sector, including Microsoft th th st is the second most important access location, (25 ). Apple, ), and IMB (31 ), Cisco Systems (54 followed by access at work. Developing countries whose key products, including its Iphone, build on show greater diversity in where people access the success of the Internet, ranked third in terms the Internet, and available data suggest that in of revenues, but was touted the world’s most 6 countries with lower income levels, schools and valuable company in terms of market value. universities remain important Internet access locations. In Egypt, while most people access the Rankings based on market capitalization (or stock Internet from home, more than half of all Internet market value) show that while in 2006 Microsoft users also go online at school or at universities. In was the only IT/Internet company ranked in the several countries in Latin America, such as Mexico, top 10, by 2016 no fewer than six of the top 10 – Peru and Venezuela, commercial facilities are Apple, Alphabet, Microsoft, Amazon, Facebook and 7 among the most frequent access locations. China Mobile – were IT/Internet companies. Measuring the Information Society Report 2016 182

199 Chapter 6 Chart 6.1: Pr oportion of Internet users accessing the Internet while mobile via a mobile-cellular 10 telephone connected to a mobile phone network (2010 and 2015 unless otherwise specified) Note: For developing countries, this chart refers to Internet use via a mobile-cellular telephone connected to a mobile phone network when the person’s location is away from “home”, “work”, “place of education”, “another person’s home” and “community and commercial access facilities”. For European countries, it refers to Internet use via a mobile-cellular telephone “away from home and work”. As such, for European countries, it could include Internet use via WiFi at other locations. Source: Eurostat and ITU. It should be noted that data are available only for when more web properties adapt their interfaces a limited number of developing countries. Data to the mobile platform. are lacking for those with very low income levels, including the LDCs. Internet traffic and content Location and the type of device used to access the Internet are important to policy-makers, content The change in the Internet ecosystem and its rise service providers and related businesses, as in popularity are accompanied by a large growth in evidence shows that Internet use varies according Internet traffic. Cisco, which tracks global IP traffic to the platform used. Data from Comscore on and related measures, predicts that by the end of Internet users in the United States, for example, 2016 annual global IP traffic will pass the zettabyte show that smartphone users spend most of the (ZB), and reach 2.3 ZB per year by 2020. time on photos, using maps, gaming and social networks (Comscore, 2015). Those accessing With faster speeds, cheaper mobile subscription the Internet on a computer spend most of the plans and devices and more data allowance, video time getting information through online portals, streaming on mobile phones is expected to rise in 11 and consulting business and finance as well as the coming years. This trend is also facilitated by entertainment and news websites (Chart 6.3). new apps and developments made by the major It should be noted that some content providers video-streaming companies. Smartphone traffic have been quicker to adapt their user interfaces is expected to exceed computer traffic by 2020, to mobile platforms; behaviour could thus change while traffic from wireless and mobile devices will then account for two-thirds of all IP traffic. Measuring the Information Society Report 2016 183

200 p three most frequent locations for Internet use in developed (top) and developing (bottom) Chart 6.2: To countries, as a percentage of individuals using the Internet by location; latest data 2012-2015 Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: Eurostat and ITU. Measuring the Information Society Report 2016 184

201 Chapter 6 me spent on mobile vs. desktop per content category in the United States Ti Chart 6.3: Note: Tablets included in mobile. From comScore’s 2015 US “Digital Future in Focus” whitepaper. Source: comScore Media metrix Multi-Platform, US, Dec 2014. Cisco predicts that by 2020, IP video traffic will Although an increasing amount of content is accessed via mobile and wireless devices, fixed account for 82 per cent of all IP traffic, up from networks continue to dominate in terms of 70 per cent in 2015. Growth is being driven by Internet video surveillance traffic, virtual reality global traffic. Most traffic using mobile devices is offloaded onto the fixed network via WiFi (Chart traffic, consumer video-on-demand (VoD), 6.4). Internet video to TV and Internet gaming traffic. To illustrate the impressive growth in video traffic, Other measures of content that include the Cisco highlights that: “It would take more than 5 number of webpages and domain name million years to watch the amount of video that networks each month in 2020. will cross global IP registrations (ITU, 2014b) show that content Every second, a million minutes of video content on the Internet is growing and becoming more will cross the network by 2020” (Cisco, 2016). diverse. At the end of the first quarter of 2016, there were an estimated 326.4 million top- The rapid growth in video traffic is also reflected level domain (TLD) registrations worldwide, in the number of video-streaming services, such representing an increase of 11 per cent from as YouTube, which has become the leading video- 294 million registrations the previous year, and 14 The total streaming platform globally with over one billion nearly doubling the figure for 2008. 12 number of country-code top level domain (ccTLD) users. The growth in subscription-based video-streaming registrations, which has been used as a proxy indicator for the availability of local content (ITU, services, such as Netflix, accounts for a large share 2014b), was estimated at 148.2 million at the end of the total IP traffic. Starting in the United States of the first quarter of 2016, an increase of 18 per in 1997 as a postal DVD rental service, Netflix cent (or 23.2 million registrations) from 2013. moved towards video streaming on the Internet in This trend is also observed with Wikipedia, the 2007. Over the last decade, it has vastly expanded its membership base across the globe and by 2016 largest user-generated online encyclopedia. The number of Wikipedia articles, which are available was available in over 190 countries, catering to 13 over 83 million subscribers. in 293 languages, has increased ten-fold over the last decade, from 3.9 million in 2006 to almost 40 million in 2016. Measuring the Information Society Report 2016 185

202 Internet and IP traffic Chart 6.4: Note: Fixed Internet traffic refers to traffic through fixed network providers on different platforms. Mobile Internet traffic refers to traffic through mobile-cellular networks. IP traffic refers to the sum of fixed and mobile Internet traffic (denoting all IP traffic crossing an Internet backbone) as well as non-Internet IP traffic (e.g. IP WAN, IP transport of TV and video-on-demand). Source: ITU based on Cisco and company reports. 16 The The Internet is also becoming more multilingual. number and types of applications, or apps. The increase in Internet users outside the primarily commercial value of this information has created English- or Chinese-speaking world is further new business models with private companies diversifying the languages used on the Internet, gathering, analysing and selling data for revenue reflected by what appears to be a relative decline optimization, for example using Internet users’ in the use of English and Chinese on the Internet. content history to target advertisements to a Estimates published by Internet World Statistics certain type of online user. However, because of suggest that 47 per cent of the world’s Internet the business value of such data, not all information on how people use the Internet is necessarily users are now English or Chinese speakers, down 15 Wikipedia can also be freely available to the public, even though such from 51 per cent in 2011. used as an indication of the availability of content information is important to understanding how in different languages (Chart 6.5). In 2003, two more people can be brought online to benefit years after Wikipedia was founded, 60 per cent of from the Internet’s opportunities. Using big data all articles were in English. By 2016, this proportion from the ICT industry could provide such insights had decreased to 13 per cent. The number of in the future, and ITU recently launched an ITU articles on Wikipedia has grown by 50 per cent project on “Big Data for Measuring the Information from 2013 to 2016, with five out of every six new Society”, which is exploring ways to use big data to articles written in languages other than the six help understand who uses the Internet, and where official UN languages (Arabic, Chinese, English, and how, as well as the benefits it delivers. French, Russian and Spanish). In 2016, 73 per cent of all articles on Wikipedia were written in Socio-economic factors that 6.2 languages other than the official UN languages. determine Internet use Many other indicators can be used to highlight the spread and growth of the Internet, and Although the number of Internet users is the Internet economy. These include the increasing continuously in all regions and countries growing number of social media outlets and of the world, major differences remain. In the users, such as Facebook accounts, number of world’s developed countries about 80 per cent of tweets, online searches, and the increase in the the population is online, as against only about 40 Measuring the Information Society Report 2016 186

203 Chapter 6 Chart 6.5: Dis tribution of Wikipedia articles by language 2003-2016 Note: The Internet users by language data are from Internet World Statistics, which assigns a single language to each individual in order to add up to the total world population; however, it is unclear how it assigns people’s first language in countries where large proportions of the population are bilingual or multilingual. htm otal. TablesArticlesT , accessed 26 May 2016, and Internet World Statistics. EN/ stats. Source: Wikipedia statistics at http:// wikimedia. org / per cent in the developing countries and less than Available data show that although Internet usage 15 per cent in LDCs (Chart 6.6). Globally, 47 per in LDCs has tripled in the past five years, Internet cent of the world’s population is using the Internet. penetration levels in LDCs today have reached the level enjoyed by developed countries in 1998, oportion of individuals using the Internet by level of development (left) and by region (right) Pr Chart 6.6: Note: * Estimate. Source: ITU. Measuring the Information Society Report 2016 187

204 oportion of individuals in LDCs using the Internet, 2015 Pr Chart 6.7: Source: ITU. suggesting that the LDCs are lagging nearly 20 Internet uptake is strongly linked to income and years behind the developed countries. At the same education time, the LDC average itself hides large differences, with some LDCs doing much better than others Internet access and usage vary significantly within (Chart 6.7). countries and available data suggest a strong link between household income and level of Internet Internet uptake in LDCs has increased significantly use, even in developed countries. The difference in the past years, driven by strong growth in a in Internet use between a country’s poorest and few LDCs in Asia, such as Bangladesh, Bhutan, richest segments is considerable (Chart 6.8). Data Cambodia and Myanmar, but also in a few LDCs from Eurostat and OECD show that, while over in Africa, such as Ethiopia, Lesotho, Mauritania, 90 per cent of individuals living in high-income Rwanda and Senegal. However, no LDC currently households use the Internet, Internet use among reaches the global average of Internet penetration. people in the poorest quartile is far from universal. In addition, in several of Africa’s most fragile and In some European countries, fewer than half of the poorest countries, still only one person in 10 uses people in the poorest quartile use the Internet. the Internet. Internet access in Latin America is even more Those LDCs with the lowest Internet penetration closely linked to household income than in OECD levels are those that already face other countries. Inequalities in Internet access in Latin development challenges, such as low incomes and America mirror pronounced inequalities in income levels of education and limited health services. distribution across the region. The two countries Since Internet access and use can deliver better with the highest income inequalities measured by communication channels, more services and the Gini coefficient, Colombia and Mexico, also applications and higher levels of productivity and present the largest differences in Internet access innovation, the lack of access in LDCs will reinforce between the poorest and richest quintiles. In existing inequalities. Mexico, nearly 70 per cent of the richest quintile of society has Internet access at home, while only 3 per cent of households in the poorest quintile are connected to the Internet (Chart 6.9). Measuring the Information Society Report 2016 188

205 Chapter 6 Chart 6.8: Internet usage by income distribution in OECD countries (2015 unless otherwise specified) Source: Eurostat and OECD. Chart 6.9: Ho useholds with Internet access by household income distribution in Latin America (latest data 2013-2015) Source: ECLAC (2015), Regional Broadband Observatory ORBA. Measuring the Information Society Report 2016 189

206 While there is a long way to go to achieve universal gender gap of 12 per cent. Only in certain select access to the Internet in Latin America, data from countries, in Europe and the Americas in particular, the Economic Commission for Latin America and are more women than men online, proportionally. the Caribbean (ECLAC) show that the Internet is Data also point to significant differences between slowly beginning to reach the poorest segments developed and developing countries (Chart 6.12). of society. However, the progress is unequal across countries. Costa Rica made significant Differences in levels of education and school progress towards equality in Internet access from enrolment are important factors that could explain 2010 to 2014, but progress has been slower in why more men than women use the Internet. several other countries of Latin America. In a few Some of the countries in which more women than countries, such as Bolivia and Peru, household men are Internet users, including the Bahamas, access to the Internet remains an amenity only Jamaica, New Zealand and Sweden, are also for the richest quintiles. Chart 6.10 presents the countries that do well on the gender parity index distribution of households with Internet access (GPI), which measures parity between girls and based on their income level. The straighter the boys in terms of school enrolment ratios. The line, the more equal is household access to the gender equality in these countries is also reflected Internet across income levels. by a high proportion of women in the labour force. Household income is often linked to level of Gender parity in tertiary education can also educational attainment. ITU data show that explain some of the differences in regional gender level of education is one of the most important gaps (Figure 6.1). The smallest Internet user indicators of whether or not people are Internet gender gap is observed in the Americas, where users (Chart 6.11). While level of education is a countries also score highly on GPI in tertiary 18 While Internet penetration among key factor explaining Internet usage in developing education. men and women is roughly the same in several countries, the same relationship is also observed in countries of North and South America, such as nearly all developed countries. Whereas Internet Brazil, Canada, Paraguay, Uruguay and the United usage in most developed countries is almost States, the link between gender parity in Internet universal among people with tertiary education, a usage and gender parity in tertiary education is large proportion of citizens with lower educational especially strong in Caribbean countries. In the attainment remain unconnected, despite similar Caribbean, there are an average of two females access to infrastructure and services. for every male attending tertiary education, and in several Caribbean countries, such as Cuba and The strong link between educational attainment Jamaica, Internet usage is higher among women and income level may help explain the fact that in than among men. developing countries the imbalances in Internet usage across groups of people with different This is in contrast with other regions with large levels of educational attainment are even more gender gaps in Internet usage, especially in Africa pronounced. In the Islamic Republic of Iran, over and Asia and the Pacific, where many countries 90 per cent of people with tertiary education use suffer from lower gender parity at higher levels the Internet as against 40 per cent of people with of education. Among developing countries, the upper secondary education, and fewer than 20 per largest Internet gender gaps are found in countries cent of people with a lower level of educational with low levels of gender parity in tertiary attainment. Similar trends are seen in Bangladesh, education, such as Bangladesh, Burundi and Egypt and Thailand. Ghana. The gender divide Seniors online Data on Internet usage broken down by gender points to a very clear gender divide. In the vast One age group with a proportionally lower majority of countries, the proportion of men Internet user penetration rate is the elderly. ITU data confirm that older population groups have using the Internet is higher than the proportion of women. These findings are reflected at global much lower Internet penetration levels than the level, where ITU reports a 2016 Internet user overall population (Chart 6.13). In most countries, Measuring the Information Society Report 2016 190

207 Chapter 6 Chart 6.10: Internet access by household income distribution in Latin America (selected countries) Note: The Lorentz curve presents the distribution of households with Internet access based on households’ income level. The red and black lines show the cumulative percentage of households with Internet access (y-axis) and the cumulative percentage of household income (x-axis) in 2009-2011 and 2013-2014. The straighter the line, the more equal is household access to the Internet across income levels. The grey diagonal line represents full equality in Internet access. Source: ECLAC (2015), Regional Broadband Observatory ORBA. Measuring the Information Society Report 2016 191

208 Internet use by level of education in developed (top) and developing (bottom) countries (latest Chart 6.11: data 2013-2015) Note: Data for most European countries are available only for upper secondary and tertiary education. Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: Eurostat and ITU. economic development, such as Hong Kong Internet user penetration for people over the (China), the Republic of Korea and Singapore. In age of 75 remains well below 10 per cent. The these countries, Internet usage among adolescents differences across age groups are especially large in countries which have experienced rapid Measuring the Information Society Report 2016 192

209 Chapter 6 Chart 6.12: Internet user gender gap (2013 and 2016) Note: 2016 are estimates. The gender gap is the difference between the Internet user penetration rate for males and females in relation to the Internet user penetration rate for males, expressed as a percentage. Source: ITU. Ge nder parity in tertiary education (2015 or latest available) Figure 6.1: Note: The darker the colours, the larger are the gender differences in enrolment in tertiary education. Dark blue indicates more than 1.2 women per man enrolled in tertiary education while dark red indicates less than 0.8 women per man. Source: eAtlas of Gender Inequality in Education; UNESCO Institute for Statistics (UIS). The elderly, however, are not a homogeneous and adults is almost universal; the elderly, however, lag behind. group, and although their overall Internet usage is significantly lower than that of the general Measuring the Information Society Report 2016 193

210 Internet usage among individuals over the age of 74 compared with the general population Chart 6.13: (latest data 2013-2015) Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: Eurostat and ITU. population, their reasons for not using the Internet Internet usage across all age groups. The user- are numerous. Possible explanations include friendliness of tablets and mobile devices has not only socio-economic factors, such as lower especially contributed to a spread of Internet incomes, educational attainment and literacy usage among young children, but also offers the levels, but also social and psychological barriers, possibility of increasing Internet usage among the such as computer and Internet anxiety, “feeling elderly. Chart 6.14 compares Internet usage rates too old”, and social isolation (Van Deursen and by age group in Japan between 2004 and 2014; Helsper, 2015). A United Kingdom study on the the largest increases can be observed in both the use of the Internet by the elderly highlights that lower the higher age groups. and “adults without an Internet connection at home are more likely to be older (particularly those Although many factors explain why the elderly over retirement age), have no formal educational tend to be late adopters of new technology, qualifications or have lower annual household including physical or health-related circumstances incomes.” (Milligan, C. and Passey, D., 2011). While that may limit their ability to learn or use new young people are fast to embrace technology, technologies without assistance, educational the challenge is to get seniors to go online for the attainment does much to explain which seniors first time. Research from the United States shows actually use the Internet. A United States survey that the elderly that begin to use the Internet stay confirms that education and income levels online (Pew Internet, 2014). (similarly related) make an important difference, highlighting that “...affluent and well-educated The possibility of accessing the Internet from seniors adopt the Internet and broadband at almost anywhere and from a multitude of different substantially higher rates than those with lower devices – compared with only accessing the levels of income and educational attainment. Fully Internet from desktops and laptops two decades 87 per cent of seniors with a college degree go ago – has also contributed to the spread of online, and 76 per cent are broadband adopters. Measuring the Information Society Report 2016 194

211 Chapter 6 Chart 6.14: Tr ends in Internet usage rates by age group (Japan) Note: Age 6 and over. Source: The Statistical Handbook of Japan 2016 (Statistics Bureau, Ministry of Internal Affairs and Communications). Available at: http:// www. go. st at. a/ htm english/ dat jp/ handbook/ c0117. Among seniors who have not attended college, services, health resources and opportunities for 40 per cent go online and just 27 per cent have social support. Many economies with a large broadband at home.” (Pew Internet, 2014). proportion of older population groups, and where the elderly remain the main population group Identifying specific barriers to Internet use excluded from the information society, have amongst older generations will be particularly a particular interest in bringing senior citizens 19 Several studies have been carried out to important in countries with large groups of older online. identify barriers and ways of increasing Internet people, since Internet access can provide them 20 penetration in these countries (Box 6.1). with valuable news and information, government How to bring seniors online Box 6.1: S tudies carried out in economies with high Internet penetration rates but relatively larger proportions of older population groups that remain offline suggest that ICT skills, but also awareness raising, relevant content and an accompanied introduction to the Internet, are crucial in bringing seniors online. In Germany, where in 2016 one out of two senior citizens is online, there are various efforts and projects by the public and private sectors to encourage the older generation to join the 21 There are also a number of studies and surveys aimed to provide an information society. understanding of the barriers that keep seniors from going online. A recent German study shows that two-thirds of non-users over the age of 65 indicate that they do not need the Internet. Just over half say they do not have the necessary technical skills and 40 per cent do not know what the Internet is about, or do not want to make the effort to use the Internet (BITKOM, 2014). Specific training courses, and accompanied introductions to the Internet, are seen as the main way of encouraging more seniors to join the information society, and are organized through 22 various groups and associations, including telecommunication operators. Measuring the Information Society Report 2016 195

212 Box 6.1: How to bring seniors online (continued) Another country that has taken action to get senior citizens online is Norway, the country with the highest Internet penetration rate among the elderly. In 2015, one of its main operators, Telenor Norway, started to offer free tablet and smartphone courses to its senior citizens in a bid to significantly raise Internet use among the elderly. Part of its “Internet for all” programme was implemented in co-operation with the Red Cross and also involves a mentor programme, as well 23 as online learning. A recent study of Internet use by elderly people in Hong Kong, province of China, concluded that to overcome the digital age divide it is imperative to provide more elderly-friendly websites for reading the news and a mobile messaging application for communication, the two main Internet activities undertaken by the elderly, and to create more elderly-friendly digital training courses 24 for seniors. 25 use and uptake. Since access to the Internet can Internet access and use amongst rural provide rural population groups with previously population groups unavailable services, open up new markets for agricultural products and increase productivity There is a strong link between Internet use and and income levels, it is particularly important for geographical place of residence. Based on data policy-makers to better understand the barriers in from 35 countries, Internet use in rural areas is order ultimately to address and overcome them. significantly lower than in urban areas (Chart 6.15). A number of factors make rural areas particularly vulnerable: their remoteness, limited access to Understanding key barriers to Internet use services (including electricity), and often difficult, i.e. mountainous or rugged, terrain. For network Identifying barriers that keep more people from operators, this means that the cost of providing joining the information society helps policy- connectivity is proportionally higher, and the makers – but also businesses, including network expected return on investment lower. Even the operators and content developers – to identify most developed economies in the world struggle concrete steps that they need to take bring more with connecting their rural and remote areas, and people online. Especially in developing countries, specific policies are usually adopted to encourage where Internet user penetration remains low, it is and provide incentives to operators to roll out very important for policy-makers to understand infrastructure to less profitable areas, where the barriers to Internet use in order to address economies of scale are absent. them, and thus allow more people to join the information society. Nevertheless, the urban-rural gap cannot solely be explained by lack of infrastructure. With 3G ITU collects data on barriers that households face and 4G networks being rolled out in more and in adopting the Internet, but the availability of more countries, network coverage is increasing. In these data, which are based on surveys carried 2016, the number of people covered by a mobile 26 out by national statistical offices, is limited. broadband network is much higher than the Data from 45 countries, including 25 developed number of Internet users (Chart 6.16). and 20 developing countries, suggest that most households are not yet online because the cost The affordability of services has often been of services and equipment is too high. The main highlighted as a key barrier, since households in reasons differ, however, across developed and rural areas tend to have lower income levels. At developing countries. While the cost of services the same time, levels of education in rural areas and equipment appears to be the key barrier lag behind those in urban areas, suggesting that, in developed countries, people in developing as for other population groups, levels of schooling, countries face other challenges. The most often- training and skills are important factors in Internet Measuring the Information Society Report 2016 196

213 Chapter 6 Chart 6.15: Pr oportion of individuals in urban and rural areas using the Internet (latest data 2010-2015) Note: The higher figure for Internet use in rural areas than in urban areas in Israel stands out and could be explained by the fact that the income and education levels of the rural population in Israel tend to be very similar to those of the urban population, because most of the small rural localities (fewer than 2 000 inhabitants) are cooperative or collective settlements, with a relatively high socio-economic level. It should also be noted that Israel’s rural population is very small. Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Confer - ence. Source: ITU. Chart 6.16: Co verage of mobile-cellular networks in relation to world population and the number of Internet users (2007-2016) Source: ITU. Measuring the Information Society Report 2016 197

214 cited response is that people “do not need the Similar findings derive from a McKinsey report, Internet”, which suggests that non-users are which warns about the “costs of the digital divide” either not aware of the information, services and and the risk of “leaving substantial portions of applications available over the Internet, or that the global population at a disadvantage that they there is not sufficiently available relevant content might never overcome”. The report looks at the for specific user groups. Lack of confidence, factors that deter individuals from going online knowledge and skills is another important and ) and Internet Barriers Index (compiled into an often-cited barrier, pointing to the importance concludes that the main factors are incentives, of raising levels of education to allow people to low incomes and affordability, user capability, benefit from the opportunities of being online and infrastructure. The user capability barrier (Chart 6.17). includes both digital literacy and language literacy. The report highlights that large proportions of Policy-makers and development experts are not the offline population are illiterate, and the need the only ones to see great potential in bringing for countries to invest in their education systems more people online. Internet content providers (McKinsey, 2014). and other companies operating in the digital market, and especially those Internet companies ICTs also provide humanitarian organizations, only whose clients are found amongst those who such as the UN High Commissioner for Refugees are online, have a clear interest in connecting the (UNHCR), with novel tools for development. Better world, and in understanding barriers that keep information on who is online and who is not their potential clients from using the Internet. allows organizations on the ground to innovate effectively, provide better services, and adapt the 2015 State of Connectivity Report, Facebook’s way humanitarian services are delivered to those for example, focused on the key reasons why that are connected. In this context, UNHCR has people are not using the Internet. The report highlighted the opportunities offered by ICTs for highlighted availability, affordability, but also the world’s 65 million refugees to obtain vital and relevance – which includes lack of content – and often life-saving information, have access to basic readiness as key barriers. The readiness barrier services, keep in touch with family and friends, and regroups a number of barriers, including lack of create new links in their new environments and skills, lack of awareness and cultural barriers. The local communities (UNHCR, 2016). report highlighted that Internet users needed not only digital skills but also basic literacy (reading To develop a new strategy, a UNHCR survey was and writing) skills, which according to the report carried out to provide information on refugees’ at least one billion people, mainly in developing level of ICT connectivity and Internet use. The countries, did not have (Facebook, 2015). survey suggests that refugees often spend up To p barriers to household Internet access at home in developed and developing countries Chart 6.17: (latest data 2013-2015) Top barrier Second barrier Source: Eurostat and ITU. Measuring the Information Society Report 2016 198

215 Chapter 6 2006 and 2015 in European countries. While to one-third of their disposable income on ICTs, engagement in all activities has increased, wide in order to remain connected. It also looked at discrepancies persist across countries, especially barriers to Internet usage and found that, next to regarding online services which are highly costs, low levels of literacy constituted the second- contingent on countries’ financial infrastructure, biggest barrier. Many also lacked content in local such as Internet banking and e-commerce. languages and had difficulties understanding and using ICTs, or were simply not aware of the Based on data and analysis by the web analytics benefits of Internet access. The availability of a firm Alexa, nearly all of the most visited websites network and electricity was also a challenge. globally in 2016 are either search engines or 27 social networks. Search engines dominate the The Internet is not living up to its 6.3 top-20 list; however, as search engines are the path to find information available elsewhere, potential and as limited aggregated or internationally comparable information is available on what kind Internet activities reflecting the transformation of information users search for, search engines towards a digital economy by increasing cannot be compared with other websites designed productivity, improving access to finance, for a specific service, such as communication, expanding citizens’ skills and facilitating more entertainment or e-commerce. Besides Facebook, effective interaction with the public sector are several other social media websites are among becoming increasingly available and used in the most visited websites, for example LinkedIn, developed countries. In developing countries, Twitter and Weibo. There are only two online however, the Internet is still mainly used for retailers among the most visited websites: the communication and entertainment purposes, United States company Amazon.com and Taobao, falling short of its potential benefits. Furthermore, the Chinese giant Alibaba's consumer-to-consumer education level and gender also seem to influence portal. the type of activity in which users engage, with implications for their potential gains. While use of the Internet and online services is constantly expanding to new horizons, it is still The types of activity that Internet users engage in mostly limited to communication with family and on the Internet have evolved over the last decade friends. Sending and receiving e-mails represents (Chart 6.16), and vary greatly across different the most frequent activity in nearly all of the 67 groups and depending on socio-economic factors, countries with household data on Internet use. in particular education and income levels. These While e-mailing is still increasing in many parts often also explain other differences, for example of the world, however, it is being complemented, in terms of gender and rural/urban Internet access and even replaced, by other forms of online and usage. communication through social media and instant messaging. In Turkey, for example, the proportion While most Europeans already used e-mail in of Internet users stating that they frequently sent 2006, other activities, such as online shopping and received e-mails fell from 69 per cent in 2010 or Internet banking, were less common. For to 58 per cent in 2013. In 2013, 84 per cent of example, while three-quarters of the Internet Turkish Internet users stated that they regularly users in Norway and Finland already used participated in social media. In the Republic of e-banking in 2006, only one out of four Internet Korea, more people indicate that they participate users in southern and central Europe did their in social media or frequently access chat sites, banking online. A decade later, the use of Internet blogs, newsgroups or online discussions than banking services had doubled in most countries of people indicating that they send and receive southern Europe, and had nearly tripled in most e-mails. In Europe, nine out of ten Internet users central European countries. Similar trends are seen regularly send and receive e-mails, whereas two across many types of online activities. out of three participate in social media. Many Internet users still only use the Internet Table 6.1 shows the number of countries in which for entertainment or communication. Chart 6.18 a particular activity is the top Internet activity or shows the change in the proportion of Internet among the top three or top five activities. users engaging in particular activities between Measuring the Information Society Report 2016 199

216 Chart 6.18: ends regarding activities on the Internet (2006-2015; selected activities) Tr Sending and receiving e-mails Purchasing or ordering goods or services Reading or downloading newspapers, magazines or books Internet banking Source: ITU, based on data from Eurostat. To p Internet activities (latest data 2010-2015) Table 6.1: Reading Getting Social media, Entertainment Telephoning Buying or sending or newspapers, informaiton Interacting with (number of countries) blogs and online (movies/music/ selling goods over the reviewing e-mail magazines or about goods or the government discussions games) or Services Internet/VoIP books services Top activity 31 13 8 5 4 2 1 1 36 Among top 3 activities 51 31 25 2 17 5 5 10 Among top 5 activities 54 44 30 22 43 6 62 Downloading Education or Professional Information Watching web software/ Listening to web Cloud storage other learning E-banking (number of countries) networks or related to TV managing radio or computing activities job search health website Top activity 1 Among top 3 activities 15 10 2 1 1 Among top 5 activities 26 20 3 4 2 12 Note: Based on data from 67 countries. Source: Eurostat and ITU. Measuring the Information Society Report 2016 200

217 Chapter 6 for both developed and developing countries. The rise of social media Besides the importance of Internet access as a communication tool, data also reveal a significantly Similar patterns can be identified across developed higher proportion of developing countries in and developing countries, but significant which entertainment activities, such as streaming divergences too. In developed countries, media and playing games, rank among the top 3 proportionally more citizens use the Internet to and top 5 Internet activities. Users in developing a greater degree to read newspapers, magazines countries read newspapers, magazines and books and books, interact with government and perform online less than in developed countries (Table 6.2). banking services online. The availability and However, education or other learning activities are promotion of e-government services in countries more popular in developing countries. This could in northern Europe have resulted in placing be linked to the fact that the place of education interaction with government among citizens’ top (school, university) remains an important access online activities. location, particularly in low-income economies (as discussed above and shown in Chart 6.2). In developing countries, activities relating to social media are particularly popular, and social The use of social media has spread at an media rank as the top Internet activity in far impressive pace across the world and the more developing than developed countries. impact of social media stretches far beyond While e-mailing is the top activity in two-thirds of communication. It has created new business developed countries, it is the top activity in only models and influenced politics across the world one out of four developing countries. E-mailing as citizens are finding new ways to organize features high amongst the top 3 and 5 activities themselves beyond traditional parties and Pr Table 6.2: oportion of developed and developing countries in which a particular activity is the top Internet activity or among the top 3 or top 5 activities (latest data 2010-2015) Developing countries Developed countries top 5 Top activity Top 3 Top 5 Top activity top 3 84% 95% 23% 67% 90% sending or reviewing e-mail 65% Social media, blogs and online discussions 5% 77% 30% 37% 67% 54% 3% 11% 14% - 3% 17% Telephoning over the Internet/VoIP Communication - - - - 7% 10% Listening to web radio - 7% 3% Watching web TV - - - Entertainment Entertainment (movies/music/games) 14% 19% 10% 40% 77% 3% - 16% 22% 3% 30% 50% Education or other learning activities 23% 49% Reading newspapers, magazines or books 11% 50% 78% 7% 13% Information related to health - - 22% - - Infomation Learning and finding Getting informaiton about goods or services 5% 57% 95% 20% 50% 63% 3% 14% - - Buying or selling goods or Services 3% 5% - 24% 43% - 3% 13% E-banking e-service Interacting with the government 5% 11% 46% - 3% 17% E-commerce and - 3% Downloading software/managing website - - - 13% - - - - - - Professional networks or job search - Cloud storage or computing - - - - - Ohter activities 81 <= 100% <= 20% 21 <= 40% 41 <= 60% 61 <= 80% Source: Eurostat and ITU. Measuring the Information Society Report 2016 201

218 Rise of social media Box 6.2: The social media revolution is best symbolized by the rapid success of companies such as Facebook, LinkedIn, Twitter and Qzone, which did not exist 15 years ago. Facebook, for example, managed to attract a fifth of the world’s population in only ten years and its presence and influence have made the company a powerhouse in the global economy. Box Chart 6.2.1 shows the strong rise in the number of monthly active users across different social media platforms, including the popular Chinese platforms Weibo and Qzone. nthly active accounts in social media (2009-2016*; in billions) Mo Box Chart 6.2.1: Note: *2016 refers to quarter 1 only. Monthly active accounts are also often referred to as monthly active users (MAUs), with one active account being treated as one active user. However, some people and organizations may have set up more than one account and some accounts used by organizations are used by many people within the organization. Source: ITU based on annual and quarterly reports and official public announcements from Facebook, LinkedIn, Sina, Twitter and Tencent. In recent years there has also been an increase in instant messaging applications, with Facebook’s purchase of WhatsApp in 2014 for USD 21.8 billion one of the largest acquisitions of its kind (and 29 The rapid increase in more than 20 times the amount Facebook paid for Instagram in 2012). monthly active users of WhatsApp from 300 million in August 2013 to 1 billion in February 2016 highlights the potential of instant messaging applications, which are quickly replacing mobile network-based short message services (SMSs). According to data from App Annie, a business intelligence firm specialized in app statistics, WhatsApp was the most downloaded app in 2015 worldwide followed by Facebook messenger 30 Other large instant-messaging services include Weixin/WeChat and QQ (Box Chart 6.2.2). developed by the Chinese Internet company Tencent. With nearly as many users as Facebook and WhatsApp, QQ is immensely popular in China. However, despite its efforts to expand to other countries with QQ International – a version available in six languages and designed for the global audience – it has not experienced the same user growth internationally as WhatsApp. of 34 OECD countries had a Twitter account and 21 governments target new avenues to connect with 28 had a Facebook account.” (OECD, 2015a). their citizens. According to a recent OECD study, “as of November 2014, the office representing the top executive institution (head of state, head of As with overall Internet penetration rates across government, or government as a whole) in 28 out countries, there is a clear link between countries’ gross national income (GNI) per capita level and Measuring the Information Society Report 2016 202

219 Chapter 6 Box 6.2: Rise of social media (continued) Mo nthly active accounts for instant-messaging services (2012-2016*, in billions) Box Chart 6.2.2: Note: *2016 refers to quarter 1 only. WhatsApp figures for 2013 and 2014 refer to Q3. Monthly active accounts are also often referred to as monthly active users (MAUs). Source: ITU based on annual and quarterly reports and official public announcements from Kakao, Naver, Rakuten, Microsoft, Tencent and Facebook. differences observed in these types of activity are their citizens’ activities online. While citizens of more pronounced in developed than in developing low- and middle-income countries participate countries. actively in social networks to a similar degree as citizens of high-income countries, they use Most other online activities show minor gender financial digital services such as Internet banking differences, with female participation in social far less. This could also be linked to the lack of media and educational activities somewhat higher availability of these kinds of online service in such than male participation. In contrast, the data countries, or to factors such as quality of service. suggest that men use e-banking services and read For example, while citizens of developed countries newspapers online more than women. There are, have been able to use sophisticated online banking however, wide discrepancies across countries. for more than a decade, Internet banking is less 31 Chart Consequently, there is a clear need for all countries deployed in many developing countries. 6.19 shows participation in social networks and to collect more gender-disaggregated data to use of e-banking across countries as a proportion unveil differences in Internet use among women of the total online population. and men and support policy-makers in finding the most appropriate responses. Although the gender gap in Internet penetration is minor in many developed countries, there are Overall, women tend to use the Internet for social some differences in what men and women do media more than men (Chart 6.21). In nearly all online. Despite differences across countries, some developed countries, women are more active on clear trends are visible. In nearly all countries, both social media, with the largest gender differences developed and developing, men tend to download in the countries of northern Europe such as software or applications to a larger extent than Estonia, Iceland, Norway and Sweden. On average women. Women, on the other hand, tend to in developed countries, participation in social be more active in seeking health information networks among female Internet users is seven 32 (Chart 6.20). The data also suggest that gender per cent higher than among male Internet users. Measuring the Information Society Report 2016 203

220 Internet use by countries’ income levels (selected activities) Chart 6.19: Source: Eurostat and ITU. In many developing countries the opposite trend activities which require trust in the system, such as is observed. This is especially the case in a few Internet banking, e-government, and purchasing economies in northern Africa and the Middle East, and selling goods and services. such as Palestine, Egypt and Qatar. A recent OECD study showed that “the breadth of Internet activities carried out by users with tertiary education is, on average, 58 per cent larger than Education determines how people use the for those with lower secondary education and Internet 33 While there is a clear below” (OECD, 2015a). relationship between level of education and use The previous section has illustrated that aside of e-banking services, the proportion of Internet from communication most Internet users do users participating in social media depends on not take advantage of the enormous offer country contexts. In the Republic of Korea and of commercial and public services available Brazil, social media participation increases with online. For this to happen, policy environment, education; however, the opposite is observed in infrastructure and skills matter. It is crucial the Russian Federation and Paraguay. to address socio-economic challenges within societies that stretch beyond the digital world. In The same pattern is observed for the purchasing particular, available data show a close link between or ordering of goods or services and the certain socio-economic characteristics and the broad entertainment category “streaming or way citizens use the Internet. While people with downloading images, movies, videos or music, lower levels of education mainly tend to use the playing or downloading games”. While people Internet for communication and entertainment with higher levels of education shop online more purposes, people with higher levels of education frequently than people with lower levels of use the Internet more diversely (Chart 6.22). education, the inverse is seen in several countries This is especially the case for more sophisticated for the entertainment category. Measuring the Information Society Report 2016 204

221 Chapter 6 Chart 6.20: Pr oportion of Internet users downloading software (left) and seeking health information (right) (latest data 2013-2015). Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: Eurostat and ITU. Measuring the Information Society Report 2016 205

222 oportion of Internet users participating in social media (latest data 2013-2015) Pr Chart 6.21: Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: Eurostat and ITU. The importance of raising education and tracking citizens’ potential to take full advantage of income levels and skills to allow more people ICTs, and in particular the Internet. to benefit from ICTs highlights the need for an integrated development approach, which is also Besides raising education levels, it is also recognized in the larger development debate. The important to note that online services, such as recently adopted 2030 Agenda for Sustainable e-commerce and e-banking, cannot function Development recognizes deep interconnections without trust in public administration and stable and interlinkages and the integrated nature of infrastructure and delivery chains. Educating the 17 sustainable development goals (SDGs). citizens regarding the broader benefits of the Just as eradicating poverty and ensuring quality Internet is another key element in enabling more education cannot be seen or achieved in isolation people to participate in the digital economy. but must be part of the larger development agenda, ICT progress is tied to progress in other key development domains. Likewise, other Improving skills begins at school development domains, such as education, health and climate change, need ICTs in order to achieve Overall, young people, and particularly teenagers, progress. tend to be more ICT-savvy, learn more quickly and can be brought online more easily than other The importance of education levels and skills in age groups. In addition, there is growing evidence order to effectively use and benefit from ICTs also on the benefits that Internet access provides underpins the use of the skills indicators and skills younger people. In particular, young people with sub-index to calculate the ITU ICT Development access to the Internet are often seen as having a Index (IDI, see chapter 1). The three skills indicators competitive advantage over their non-Internet- 34 (mean years of schooling, gross secondary using classmates. enrolment and gross tertiary enrolment) have indicators since they been described as proxy Most countries do not collect household data track education levels, rather than ICT skills more on children and young teenagers’ use of the specifically. However, the analysis in this chapter Internet, and even if they do, the age range differs suggests that they are particularly relevant in widely across countries. ITU household data show Measuring the Information Society Report 2016 206

223 Chapter 6 Internet use by activity and education level (selected activities; latest data 2013-2015) Chart 6.22: Participating in social networks Streaming or downloading images, movies, videos or music, playing or downloading games Note: Data for most European countries are only available for upper secondary and tertiary education. For Eurostat countries, the activity streaming or downloading images, movies, videos or music, playing or downloading games relates to playing/downloading games, images, films or music . Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Confer - ence. Source: Eurostat and ITU. Measuring the Information Society Report 2016 207

224 Chart 6.22: Internet use by activity and education level (selected activities; latest data 2013-2015) (continued) Purchasing or ordering of goods or services Internet banking Note: Data for most European countries are only available for upper secondary and tertiary education. For Eurostat countries, the activity streaming or downloading images, movies, videos or music, playing or downloading games relates to playing/downloading games, images, films or music . Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Confer - ence. Source: Eurostat and ITU. Measuring the Information Society Report 2016 208

225 Chapter 6 that from the age at which children begin lower The study shows that traditional education is secondary school, the proportion of adolescents crucial to increasing the ability of students to use using the Internet exceeds the Internet usage rate ICT tools for learning purposes. Reading content for the general population in nearly all countries on the Internet requires the same skills as reading across the globe (see Chart 6.23). a book or newspaper. While it is important to integrate Internet into education, results from Adolescents’ use of the Internet is already nearly the OECD Programme for International Student universal in developed countries and the younger Assessment (PISA) show that students in the population in many developing countries is twice highest-performing countries in digital reading as connected as the general population. In many were “not more exposed to the Internet at school developing countries, schools and universities than are students in other OECD countries” (OECD, 36 remain the entry-way for children and young adults 2015b). into using computers and the Internet. Further As much as access to and use of the Internet connecting schools is essential in order to increase have been linked to positive outcomes, and Internet usage and ensure that today’s youth will as much as a growing number of children and have the relevant skills for future employment. adolescents spend much of their time online, it is also important to acknowledge and understand However, basic education is crucial to opening up the negative side-effects of “too much Internet”. the possibilities of the Internet for those not yet The number of available studies looking into online. As described previously in this chapter, the possible side-effects of extensive use of poor literacy remains one of the key barriers to ICT the Internet is relatively limited, and given the connectivity and Internet usage. UNESCO’s Global novelty of the Internet it is also too early to study Education Monitoring Report 2016 estimates long-term impacts. Existing evidence, however, that nearly 61 million children of primary school suggests that children and teenagers who spend age and 202 million adolescents of secondary large amounts of time online are more at risk of school age did not go to school in 2014, many of experiencing different forms of mental distress them living in conflict-affected areas. In addition, (Box 6.3). 758 million adults, 63 per cent of them women, have not acquired even minimal literacy skills 35 (UNESCO, 2016). The same report highlights onclusions C 6.4 an annual shortfall of at least USD 21 billion in low-income countries in regard to achievement Over the past decade, the Internet has spread of the education targets in the 2030 Agenda for rapidly and, by the end of 2016, 3.5 billion people Sustainable Development. Strengthening global – or close to 50 per cent of the world’s population efforts to improve basic education for all people – are using the Internet, driven by the expansion of is a crucial component of the SDGs, a prerequisite mobile networks and falling prices. An increasingly for connecting the last billion and allowing more ubiquitous, open, fast and content-rich Internet people to take advantage of the opportunities has changed the way many people live, opened up by Internet access. communicate, and do business, delivering great benefits for people, governments, organizations A relationship between what people do on and the private sector. the Internet and socio-economic status is also observed among children and adolescents. An online. are Many people no longer online, they go OECD study shows that wealthier students are The Internet has opened up new communication more likely to use the Internet for educationally channels, provided access to information and advantageous activities such as gathering services, increased productivity and fostered information and reading the news, while innovation. It has also created an Internet poorer students are more likely to use it for economy, and a number of new leading businesses communication and playing games (OECD, 2016). whose clients are exclusively online. The same study suggests that inequalities exist even in countries with almost universal Internet Nevertheless, the Internet and its benefits have access. Lack of knowledge and familiarity in use of spread unequally and many people have not the Internet to find information can hamper young been able to benefit from the potential of the people in their studies and job-finding prospects. Measuring the Information Society Report 2016 209

226 Chart 6.23: Ad olescents’ (age 15-24) use of the Internet compared with that of the general population Note: Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU. Measuring the Information Society Report 2016 210

227 Chapter 6 Research shows side-effects of too much Internet for children Box 6.3: Internet usage does not always have positive effects. The rapid evolution in children’s use of the Internet will most likely have an impact on their overall health and development; however, research on the actual effects is scarce. Rapidly increasing attention is being paid to the importance of measuring and understanding children’s access to and use of the Internet. This coincides with increasing concern for the health of children spending a lot of time in front of tablets, computers and other screens. A recent government report from the United Kingdom suggests that children and adolescents who spend more time in front of screens tend to experience higher levels of emotional distress, anxiety and depression (Public Health England, 2013). Other studies show negative effects relating to sleep patterns, physical activity and social well-being. An OECD study from 2015 found that “students who spend more than six hours on line per weekday outside of school are particularly at risk of reporting that they feel lonely at school, and that they arrived late for school or skipped days of school in the two weeks prior to the PISA test” (OECD, 2015b). In addition, media reports on cyberbullying through social media and other platforms have also augmented the need for more research and information on how children use the Internet. While online access and use has the potential to augment young people’s knowledge and learning abilities, more research is required to better understand the full impact of children and adolescents’ increasing presence in the online world. Internet. The offline population – 3.9 billion people To overcome these challenges, policy-makers must globally – is disproportionally female and elderly, address broader socio-economic inequalities and less educated, has lower incomes and lives in in particular help people acquire the necessary rural areas. While Internet infrastructure, access skills, including analogue skills, to take full and quality of service remain important barriers advantage of the Internet. This is in line with a to uptake, more people have access to Internet more integrated development approach, like services than actually use them. Broader socio- that adopted in the 2030 Agenda for Sustainable economic factors that are not ICT-related need Development, which highlights that development to be addressed to bring more people online. challenges are linked and cannot be achieved in Education levels in particular, and accompanying isolation. income levels, are strong determinants of whether or not people use the Internet. Finally, this chapter has highlighted the need for more and better data. The lack of data remains Socio-economic factors also play an important role an important challenge to ICT policy-makers, in how the Internet is used and to what extent its investors and content producers. Data that provide potential is maximized. Existing data suggest that in-depth information on how exactly people use communication and social media are key activities the Internet and what people do online are scarce, for Internet users. Unlike Internet users with and often outdated. Most of the information higher levels of education, many Internet users on what people do online today stems from ICT with lower education and income levels tend to household surveys, which provide reliable data use the Internet predominantly for communication disaggregated by socio-economic characteristics, and entertainment purposes. This suggests that but are also costly and time-consuming to they do not benefit fully from the opportunities produce. Moreover, ICT household data are of the Internet and that the Internet is liable to currently not available for many developing become a driver of greater inequalities, instead of countries, and nearly non-existent for the world’s addressing them. LDCs. These constraints lead to information gaps and possible misinformation for policy-makers. In some cases, public and private institutions carry Measuring the Information Society Report 2016 211

228 out ad-hoc surveys to fill these data gaps, but which is exploring ways of using big data from generally these are not representative and are the ICT industry to help understand who uses therefore of limited use or, worse, produce wrong the Internet, and where and how, and to better information. understand the benefits it delivers. Based on a number of pilot studies, it will show how big data Within this context, the United Nations has called from the ICT industry (including mobile operators for the use of new data sources, including big and the over-the-top (OTT) market) can be used 37 A number data, to complement official statistics. as a source for new and existing ICT indicators in of efforts are being made to use big data to order to enhance data collections, benchmarks address and eventually overcome this data gap, and methodologies for measuring the information including a recently launched ITU project on “Big society. Data for Measuring the Information Society”, Measuring the Information Society Report 2016 212

229 Endnotes 1 Google went live only in 1998, LinkedIn launched in 2003, Facebook in 2004, and YouTube in 2005. Twitter started in 2006, just ten years ago. 2 For an account of what the Internet looked like in 1996, see slat e. com/ articles/ technology/ technology/ 2009/ www. http:// html . 02/ web. jurassic_ 3 According to (Kuss, D.J. and Lopez-Fernandez, O., 2016), a growing amount of research on the topic is emerging. 4 Tim Berners-Lee, a British scientist at CERN, invented the World Wide Web (WWW) in 1989. The web was originally conceived and developed to meet the demand for automatic information-sharing between scientists in universities and cern/ http:// birth- web . See also: home. topics/ institutes around the world, see: itu. int/ net/ itunew s/ issues/ 2009/ 10/ http:// 34. aspx . www. 5 Alphabet, which was created in October 2015, is the parent company of Google. It is, according to its website and co- founder Larry Page “...mostly a collection of companies. The largest of which, of course, is Google. This newer Google is a bit slimmed down, with the companies that are pretty far afield of our main internet products contained in Alphabet instead”, see: xyz/ abc. https:// 6 beta. fortune. See fortune500/ list/ http:// com/ 7 These were: Apple (1), Alphabet (2), Microsoft (3), Amazon (4), Facebook (5) and China Mobile (10), see: http:// www. com/ new s/ special- report/ 21707048- small- group- giant - economist. companiessome- old- some- new are- once- again- dominating- global 8 http:// For data on the rapid expansion of mobile networks, see page 2 of the ITU Facts and Figures 2016, available at: en/ ITU- D/ Sta tistics/ Documents/ itu. facts/ int/ ICTF actsFigures2016. pdf www. 9 For a discussion of the growing spread of the smartphone and its improved processing power, see: http:// www. com/ new s/ leaders/ 21645180- smartphone- ubiquitous- addictive- economist. trans formative- planet - phones and- 10 Chart 6.1 refers to use of the Internet while mobile via a mobile-cellular telephone connected to a mobile phone network. For developing countries, it refers to Internet use via a mobile-cellular telephone connected to a mobile phone network if the location is away from “home”, “work”, “place of education”, “another person’s home” and “community and commercial access facilities”. It does also not include WiFi connectivity. For European countries, it refers to Internet use via a mobile-cellular telephone “away from home and work”. As such, for European countries, it could include Internet use via WiFi at other locations. For more information on the definitions of Internet use by location, see page 55 of the Manual for Measuring ICT Access and Use by Households and Individuals 2014 available at: itu. http:// en/ www. int/ D/ ITU- Pag es/ publica tions/ manual2014. aspx . Statistics/ 11 Cisco expects the number of connected devices in 2020 to be more than triple the 2020 global population. 12 https:// www. youtube. com/ See: yt/ press/ st atistics. html 13 https:// netflix. com/ en/ press- releases/ netflix - is- now - av ailable- around- the- world and https:// ir. netflix. com/ See: media. cfm results. 14 Quarterly reports from VeriSign, the registry operator for .com and .net, which include data from the consultancy blog. verisign. com/ domain- names/ int ernet- gro ws- to- 326- 4- ZookNIC and Centralized Zone Data Service. See https:// domain- names- in- the- firs t- quarter - of- 2016/ . million- 15 Internet World Statistics assigns a single language to each individual in order to add up to the total world population; however, it is unclear how it assigns people’s first language in countries where large proportions of the population are st www. int ernetworldstats. com/ http:// ats7. htm . bilingual or multilingual. See 16 By June 2016, Facebook had 1.71 billion monthly active users, spread across almost every country of the world, see: newsroom. fb. com/ http:// compan y- inf o/ 17 By September 2016, Twitter, which launched its services just 10 years ago (in 2006), had 313 million monthly active users y about. twitter . com/ compan and 1 billion unique monthly visits to sites with embedded tweets, see: https:// 18 There are on average 1.28 women for each man in tertiary education in Latin America and the Caribbean and 1.37 women for each man in North America. 19 http:// www. bloomberg. com/ new s/ articles/ 2016- 02- 02/ mapping- the- oldest - coun tries- in- the- world See 20 See, for example, Barbosa Neves, B. and Amaro, F. (2012) 21 See for example, the “Golden Internet prize” project, which is sponsored by the Ministry of Justice and Consumer Protection and several private sector associations. This project presents a prize to people aged 60 and over who use int www. goldener the Internet and who help to bring other senior citizens online. See: https:// ernetpreis. de/ . For other - https:// www. sicher- im- de/ netz. projects designed to encourage seniors to use and benefit from the Internet, see: www. kompass- handreichung- 1 and http:// downloads/ digital- weg eausdereinsamkeit. de/ . Measuring the Information Society Report 2016 213

230 22 http:// gruppe. de/ table t- pcs- See, for example, the “Project Table PC for Seniors” which is funded by Telefónica: eplus- senioren/ . fuer- 23 www. telenor . com/ media/ articles/ 2015/ helping- http:// the- elderly- to- mast er- table t- and- smartphone/ 24 ajgg. org / AJGG/ V10N1/ 2014- 175- OA . http:// pdf 25 source= http:// worldbank. org / dat a/ reports. aspx? databank. For example, see World Bank Education Statistics, available at ~- all- statistics- indica tors . Research from the OECD’s PISA study also suggests that: “In most countries and education- economies, students who attend schools in urban areas tend to perform at higher levels than other students”, see www. https:// org / pisa/ pisaproducts/ pisainf ocus/ pisa%20 in%20 focus%20 n28%20 (eng)-- FINAL. pdf oecd. 26 The question put to households is: Why does this household not have Internet access? (multiple responses possible). For more information, see the Manual for Measuring ICT Access and Use by Households and Individuals 2014 available at: http:// www. itu. int/ en/ ITU- D/ Sta tistics/ Pag es/ publica tions/ manual2014. aspx . 27 . http:// alex a. com/ top sites www. See 28 OECD (2015a), page 52. 29 The acquisition initially valued at USD 19 billion increased to USD 21.8 billion because of a rise in Facebook stock value. 30 go. appannie. com/ report - app- http:// 2015- re trospective annie- 31 It should be noted that most of the world’s unbanked people are in developing countries, which is a reason why m-banking is increasingly successful in such countries. 32 This is based on the simple averages of the proportion of women and men participating in social media in 33 developed countries. It is calculated by dividing the average proportion of female Internet users participating in social media (68.3%) by the average proportion of male Internet users participating in social media (63.8%), minus one. 33 OECD (2015a), page 51. 34 Research from Michigan University suggests that the use of the Internet can improve the mental well-being of retired older adults, reducing the probability of depression by one third, see: psychsocgerontology. oxf http:// org / ordjournals. content/ early/ 2014/ 03/ 25/ ger onb. gbu018. full 35 UNESCO (2016), page 73. 36 In the 2009 and 2012 Programme for International Student Assessment (PISA) assessments, OECD assessed reading on digital media separately from reading printed text. For more information, see the PISA 2012 Assessment and Analytical http:// www. oecd- ilibrar y. org / educa pisa- 2012- assessment - and- analytic al- frame work/ tion/ Framework, available at: reading- framework_ 9789264190511- 4- en 37 The United Nations has recognized the opportunities that new data sources, including big data, offer in filling data gaps to track the 2030 Agenda for Sustainable Development. See: United Nations, A World that Counts. Mobilizing the Data Revolution for Sustainable Development. Report prepared at the request of the United Nations Secretary-General, by the Independent Expert Advisory Group on a Data Revolution for Sustainable Development, November 2014, see: http:// undatarevolution. www. org / report/ . In recognition of the opportunities offered by big data to support the monitoring of the post-2015 development goals, the UN Statistical Commission also set up the UN Global Working Group on Big Data / unstats. un. org http:// unsd/ bigda ta/ . for Official Statistics, see Measuring the Information Society Report 2016 214

231 List of references List of references Egypt blocked Facebook Internet service over surveillance Abutaleb, Y. and Menn, J. (2016), Exclusive: - sources . Reuters. 1 w r euters. c om/ a rticle/ u s- f acebook- e gypt- i dUSKCN0WY3JZ? http:// April 2016. ww. fe te chnologyNews . edName= feedType= RSS& https:// . Africa Telecoms, 12: 68-69. Africa Telecoms (2011), The History and Future of Dynamic Tariffing om/ m agazine1/ d ocs/ i ssue_ 1 2_/ 7 1? z oomed=& z oomPercent=& z oomX=& z oomY=& n oteText=& issuu. c vi ewMode= mag azine . noteX=& noteY=& , Alliance for Affordable Internet (A4AI) (2015a), Case Study: Delivering Affordable Internet in Myanmar 4ai. o rg/ w p- c ontent/ u ploads/ 2 015/ http:// 0 3/ M yanmar- C ase- S tudy. p df . a March 2015. The impacts of emerging mobile data services Alliance for Affordable Internet (A4AI) (2015b), https:// a 4ai. o rg/ w p- c ontent/ u ploads/ 2 015/ 1 1/ . November 2015. in developing countries Re pd f . searchBrief1. MeasuringImpactsofMobileDataServices_ The 2015-2016 Affordability Report . Alliance for Affordable Internet (A4AI ) (2015c), o a http:// 4ai. rg/ port/ affordability- re port/ 20 15/ . re Décision No. 2015-0031 du Conseil de Régulation de l’Autorité de Régulation des ARTCI (2015), Télécommunications/TIC de Côte d’Ivoire en date du 08 janvier 2015 portant fixation des plafonds des tarifs http:// w ww. a rtci. c i/ i mages/ s tories/ p df/ d de terminaison d’appel fixe et mobile pour l’année 2015 . ecisions_ eg/ d ecision_ 2 015_ 0 031_ c onseil_ r egulation. p df . conseil_ r . End-2015 report on Internet in Bolivia w ww. a tt. g ob. b o/ https:// c ontent/ s ituación- d el- i nternet- ATT (2015), en- bolivia . Axiata (2016), . http:// axiata. Shaping the Future Towards a Digital Company: Annual Report 2015 c m isc/ a r2015. p df . listedcompany. om/ Barbosa Neves, B. and Amaro, F. (2012), Too old for technology? How the elderly of Lisbon use and perceive . Centre for Public Administration and Policies (CAPP). Institute of Social and Political Sciences of ICT c j ournal. n et/ i ndex. p hp/ http:// iej/ i- a rticle/ v iew/ 8 00/ 9 04 . c Technical University of Lisbon (ISCSP-UTL). . http:// w ww. b atelcogroup. c om/ m Batelco (2016), 6 0887/ Annual Report 2015: Cultivating Convergence edia/ r_ 2 015_ e nglish. p df . batelco_ a Telecommunication Market Profile: Internet Bhutan InfoComm and Media Authority (BICMA) (2016), Subscribers . w ww. b icma. g ov. b t/ http:// b icmanew/? p age_ i d= 5 55 . Senioren in der Digitalen Welt. Berlin. December 2014. https:// w ww. b itkom. o rg/ P BITCOM (2014), resse/ Anhaenge- a n- P Is/ 2 014/ D ezember/ 1 41212- B ITKOM- P raesentation- S enioren- i n- d er- D igitalen- W elt- 1 2- 1 2- 2014. pdf Body of European regulators for Electronic communications (Berec) (2010), BEREC report on impact of bundled offers in retail and wholesale market definition . BoR(10)64. b erec. e uropa. e u/ e ng/ d ocument_ http:// o ubject_ m atter/ b erec/ r eports/ 2 09- register/ erec- r eport- o n- i mpact- s f- b undled- o ffers- i n- r etail- a nd- b wholesale- ma rket- finition . de Measuring the Information Society Report 2016 215

232 Broadband Commission for Digital Development (2013), Doubling Digital Opportunities: Enhancing the . http:// www. bro adbandcommission. or g/ Do cuments/ Inclusion of Women & Girls in the Information Society groups/ - do ubling- dig ital- 20 13. pd f . working- bb The State of Broadband 2015 Broadband as a Broadband Commission for Sustainable Development (2015), www. bro adbandcommission. or g/ pub Pa ges/ SO B- http:// lications/ Foundation for Sustainable Development. . 2015. aspx Cash, H., Rae, C., Steel, A. and Winkler, A. (2012), Internet Addiction: A Brief Summary of Research and National Center for Biotechnology Information, Current Psychiatry Rev. 2012 Nov; 8(4): 292–298. Practice. 10.2174/157340012803520513 . Published online 2012 Nov. doi: Mobile Communication and Society: Castells, M., Fernandez-Ardevol, M., Linchuan Qiu, J. and Sey, A. (2007), , MIT Press. A Global Perspective n/ http:// w ww. c isco. c om/ c / e . White Papers. June 2016. Cisco (2016), The Zettabyte Era—Trends and Analysis yperconnectivity- s c ollateral/ s ervice- p rovider/ v isual- n etworking- i ndex- v ni/ v ni- h olutions/ w p. h tml . us/ [email protected]: A Qualitative Study http:// c ks. i . Colombo: LIRNEasia. w p- c ontent/ CKS Consulting (2012), n/ 2 012/ uploads/ 6/ T eleuse- a t- B oP- 4 . c ompressed. p df . 0 ComScore (2015), Mobile App Report. The 2015 U.S. www. co mscore. co m/ In sights/ Pre sentations- https:// S- W 2 015/ T he- 2 015- U hitepapers/ M obile- A pp- R eport . and- Annual Report 2015 . http:// d igi. l istedcompany. c om/ m isc/ a r/ a r2015. Digi (2016), p df . Digicel (2015), Amendment No.4 to Form F-1 Registration Statement under the Securities Act of 1933 A , 2 October 2015. w ww. s ec. g ov/ rchives/ e dgar/ d ata/ (Securities and Exchange Commission Filing) https:// 0 00119312515335867/ d 946689df1a. 1645826/ h tm . Digitata (2016), (e-Brochure). Digitata: Dynamic Tariffing w ww. d igitata. c http:// w p- c ontent/ u ploads/ om/ 2016/ 0 2/ D igitata- D ynamic- T ariffing- e - B rochure- 2 016. p df . Drossos, A. (2015), The real threat to the open Internet is zero-rated content . The Digital Fuel Monitor eal_ http:// fmonitor. e u/ d ownloads/ W ebfoundation_ g uestblog_ T he_ r d t hreat_ o pen_ i nternet_ (Rewheel). pdf . zerorating. ECLAC (2015), State of the Broadband in Latin America and the Caribbean 2015 ( original title: Estado de la e- http:// w ww. c epal. o rg/ e s/ p ublicaciones/ 3 8605- e stado- d . banda ancha en América Latina y el Caribe 2015) a ncha- e n- a merica- l atina- anda- y - e l- c aribe- 2 015 . la- b om/ http:// w ww. n era. c The Economics of Zero Rating . NERA Economic Consulting. Eisenach, J.A. (2015), ner a/ pu blications/ content/ 15/ dam/ Ec onomicsofZeroRating. pd f . 20 Ericsson (2015a), Ericsson Mobility Report on the Pulse of the Networked Society: August 2015 (Interim Update). https:// w ww. e ricsson. c om/ r es/ d ocs/ 2 015/ e ricsson- m obility- r eport- a ugust- 2 015- i nterim. p df . Ericsson (2015b), Ericsson Mobility Report on the Pulse of the Networked Society: November 2015 . https:// e ricsson. c om/ r es/ d ocs/ 2 015/ m obility- r eport/ e www. m obility- r eport- n ov- 2 015. p df . ricsson- Ericsson (2015c), Ericsson Mobility Report: South East Asia and Oceania , November 2015. https:// www. ericsson. c om/ r es/ d ocs/ 2 015/ m obility- r eport/ e mr- n ov- 2 015- r egional- r eport- s outh- e ast- a sia- a nd- o ceania. pdf . Measuring the Information Society Report 2016 216

233 List of references Ericsson (2016), . https:// www. Ericsson Mobility Report on the Pulse of the Networked Society: June 2016 c r es/ d ocs/ 2 016/ e ricsson- m obility- r eport- 2 016. p df . ericsson. om/ European Commission (EC) (2014), Special Eurobarometer 414: E-communications and telecom single market household survey e c. e uropa. e u/ p ublic_ o pinion/ a rchives/ e bs/ e bs_ 4 14_ e n. p df . . March 2014. http:// c http:// ewsroom. fb . n om/ . Facebook (2015), State of Connectivity 2015: A Report on Global Internet Access 016/ 0 2/ s tate- o f- c onnectivity- 2 015- a - r eport- o n- g news/ I nternet- a ccess/ . 2 lobal- Galpaya, H., Zainudeen, A. and Suthaharan, P. (2015), A baseline survey of ICT and knowledge l irneasia. n et/ w p- c ontent/ u ploads/ 2 015/ 0 7/ L IRNEasia_ http:// . August 2015, p.9. access in Myanmar De scriptiveStats_ V1 . pd f . MyanmarBaselineSurvey_ , Research ICT Gender Assessment of ICT Access and Usage in Africa Gillwald, A., Milek, A. & Stork, C. (2010), Africa. w ww. i ctworks. o rg/ s ites/ d efault/ fi les/ u ploaded_ p ics/ 2 009/ G ender_ P aper_ S ept_ 2 010. p df . http:// . Untangling 'subscribers', 'mobile phone owners' and 'users'. GSMA (2014), Measuring mobile penetration ww. g smaintelligence. c om/ r esearch/? le= a afdf6d1736603f2494b61c33cf1de2f& d ownload . https:// w fi Bridging the Gender Gap: Mobile Access and Usage in Low- and Middle-income Countries. GSMA (2015), w ww. g sma. c om/ m obilefordevelopment/ w p- c ontent/ u ploads/ 2 016/ http:// 2/ G SM0001_ 0 3232015_ 0 WGRAYS- We b. pd f . GSMAReport_ NE www. https:// . Connected Society. Digital inclusion in Latin America and the Caribbean GSMA (2016a), co m/ re search/? fil e= gsmaintelligence. 89 5f6c0a1efa7a25f5d6b4ff874e92f1& do wnload . Connected Society. Consumer barriers to mobile Internet adoption in Asia. GSMA (2016b), https:// www. co m/ re search/? fil gsmaintelligence. c5 2d213ec6288da6b31248df71e370a3& do wnload . e= GSMA (2016c), . https:// w ww. g smaintelligence. c om/ r esearch/? fi le= The Mobile Economy 2016 do . 97928efe09cdba2864cdcf1ad1a2f58c& wnload . GSMA and LIRNEasia (2015), lirneasia. net/ wp - http:// Mobile phones, internet, and gender in Myanmar ploads/ 2 016/ 0 4/ G SMA_ u M yanmar_ G ender_ R 2_ S preads. p df . content/ Women and Mobile: A Global GSMA, Cherie Blair Foundation and Vital Wave Consulting (2010), Opportunity . w ww. g sma. c om/ m obilefordevelopment/ w http:// c ontent/ u ploads/ 2 013/ 0 1/ G SMA_ W omen_ p- obile- A _ G lobal_ O pportunity. p df . and_ M rg/ . w ww. i nfodev. o Mobile Usage at the Base of the Pyramid in Kenya i nfodev- fi les/ InfoDev (2012a), https:// k enya_ b op_ s tudy_ w eb_ j an_ 0 2_ 2 013_ 0 . p final_ . df InfoDev (2012b), . Mobile Usage at the Base of the Pyramid in South Africa w i nfodev. o rg/ i nfodev- https:// ww. fi w s outh_ a frica_ b op_ s tudy_ eb. p df . files/ nal_ Global Internet Report 2015: Mobile Evolution and Development of the Internet Society (ISOC) (2015), http:// www. in ternetsociety. or g/ . gl obalinternetreport/ . Internet 2015 Measuring the Information Society Report 2009 . ITU (2009), w ww. i tu. i nt/ e http:// I TU- D / S tatistics/ P ages/ n/ publications/ mi s2009. asp x . nt/ / w ww. i tu. i https:// e n/ I TU- D ITU (2013), Study on International Internet Connectivity in Sub-Saharan Africa . F arket/ D ocuments/ I IC_ A frica_ Regulatory- inal- e n. p df . M Measuring the Information Society Report 2016 217

234 ITU (2014a), Measuring the Information Society Report 2014 http:// w ww. i tu. i nt/ e n/ I TU- D / S tatistics/ . p m is2014/ M IS2014_ w ithout_ A nnex_ 4 . p df . Documents/ ublications/ . w ww. i tu. i nt/ Final WSIS Targets Review: Achievements, Challenges and the Way Forward ITU (2014b), http:// I TU- D / S tatistics/ P ages/ p ublications/ w sistargets2014. a spx . en/ S w ww. i tu. i nt/ e n/ I TU- D / http:// tatistics/ Measuring the Information Society Report 2015 . ITU (2015), p m isr2015/ M Documents/ ublications/ w 5. p df . ISR2015- ICT Facts and Figures 2016. https:// w ww. i tu. i nt/ e n/ I TU- D / S tatistics/ D ocuments/ f acts/ ITU (2016), f ICTFactsFigures2016. . pd Harnessing the Internet of Things for Global Development . w ww. i tu. i nt/ e n/ ITU and Cisco (2016), https:// b D ocuments/ H arnessing- I oT- G lobal- roadband/ D evelopment. p df . action/ Sharing mobile phones in developing countries: Implications for the digital divide, James, J. (2010), Technological Forecasting and Social Change , Volume 78, Issue 4, May 2011, Pages 729-735, ISSN 0040- 1625. d x. d oi. o rg/ 1 0. 1 016/ j . t echfore. http:// 2 010. 1 1. 0 08 . KISA (2015), . http:// i sis. k isa. o r. k r/ e ng/ e book/ E ngWhitePaper2015. p df Korea Internet White Paper . Kuss, D.J. and Lopez-Fernandez, O. (2016), Internet addiction and problematic Internet use: A systematic . National Center for Biotechnology Information. See comment in PubMed review of clinical research World J Psychiatry. 2016 Mar 22;6 (1):143-76. doi: 10.5498/wjp.v6.i1.143. Commons below Key Facts & Figures: Edition of June 30, 2015 . http:// w ww. Maroc Telecom (2015), am. m a/ L ists/ P ublication/ i Attachments/ 5 0/ M aroc%20 T elecom%20 e n%20 % 20Bref%20 S 1%202015 % 20Version%20 a nglaise_ E N. p df . d . m axis. m y/ a r2015/ i mages/ http:// ownload/ p df/ m axis_ a r2015. p df Maxis (2016), Annual Report 2015 . Offline and falling behind: Barriers to Internet adoption Mc Kinsey (2014), , October, 2014. http:// www. c om/ ~ / m edia/ m ckinsey/ d otcom/ c lient_ s ervice/ h igh%20 t ech/ p dfs/ o ffline_ a nd_ f alling_ b ehind_ mckinsey. full_ ash x . report. cit. The Telecommunications Law w ww. m http:// g ov. MCIT (2013), , The Pyidaungsu Hluttaw Law No.31, 2013. s ites/ d efault/ fi mm/ T elecom%20 L aw%20 E nglish%20 V ersion_ 0 . p df . les/ Milligan, C. and Passey, D. (2011). Nominet Trust. Ageing and the use of the Internet. Current engagement er cent20SoA per cent20- NT p https:// www. nom inettrust. or g. uk / si tes/ de fault/ fil es/ . and future needs per cent20Ageing per cent20and per cent20the per cent20use per cent20of per cent20the per cent20Internet_0.pdf . Ministry of Information and Communications, Royal Government of Bhutan (MOIC) (2014), Annual ontent/ http:// w ww. m oic. g ov. b t/ w c p- (5th Edition), March 2014. InfoComm and Transport Statistical Bulletin 016/ 0 uploads/ 2 2 014. p df . 5/ oup_ Gr https:// www. mt n. co m/ Su stainability/ Do cuments/ MT N_ . Annual Report 2014 MTN (2014), 014. eport_ Integrated_ R p df . 2 MTN (2015), Annual Report 2015 . https:// w ww. m tn. c om/ I nvestors/ F inancialReporting/ D ocuments/ esults_ ANNUALREPORTS/ 2 015/ B ooklet/ A nnual_ r b ooklet_ 2 015. p df . Measuring the Information Society Report 2016 218

235 List of references National Commission for the State Regulation of Communications and Informatization (NCCIR) (2013 ), . Annual report of the National Commission for the State Regulation of Communications and Informatization e n. n krzi. g ov. u a/ i mg/ z stored/ F ile/ 2 013_ 0 3/ n krzi- e n. p df . http:// (2013), OECD Digital Economy Papers , No. 224, OECD Publishing, OECD Mobile Handset Acquisition Models. d http:// d oi. o rg/ 1 0. 1 787/ 5 k43n203mlbr- e n . x. Paris. DOI. w ww. o ecd. o rg/ e conomy/ o ecd- d igital- e conomy- OECD Digital Economy Outlook 2015. OECD (2015a), http:// 015- 9 outlook- e n. h tm . 2 789264232440- http:// w ww. o ecd- i library. o rg/ OECD (2015b), Students, Computers and Learning: Making the Connection . 789264239555- c s a nd- l earning_ 9 tudents- e n . education/ omputers- OECD Science, Technology OECD (2015c), Triple and Quadruple Play Bundles of Communication Services. and Industry Policy Papers , 23. OECD Publishing: Paris. d x. d oi. o rg/ 1 0. 1 787/ 5 js04dp2q1jc- e n . http:// http:// Are there differences in how advantaged and disadvantaged students use the Internet? OECD (2016), o ecd- i library. o rg/ e ducation/ a re- t here- d ifferences- i n- h ow- a dvantaged- a nd- d isadvantaged- s tudents- www. t he- i nternet_ 5 jlv8zq6hw43- e n . use- . Ooredoo (2015), Annual Report 2015 http:// m/ en/ in vestors/ . ooredoo. co January 2015. http:// opensignal. The Global Prevalence of Dual-SIM Android Devices, OpenSignal (2015), r eports/ 2 015/ 0 1/ a ndroid- d evices- d ual- s im . com/ Registration document: Annual financial report 2014 . Orange (2015), e w o range. c om/ n/ c ontent/ http:// ww. 29884/ 83 4878/ ve rsion/ 3/ fil e/ download/ 20 14+Registration+document. pd f . c http:// w ww. o range. om/ e n/ c ontent/ . Orange (2016), Registration document: Annual financial report 2015 3 6161/ 1 131283/ v ersion/ 2 / download/ fi le/ O range%20 - %20 D DR%202015 _ VA. p df . OTE (2015), . 2014 Annual Report https:// c osmote. g r/ fi xed/ d ocuments/ 1 0280/ 7 0481091/ E N_ A nnual_ w ww. 4 2 p df/ 5 73e1f16- 1 8cf- bbc- 9 d27- a 692197dbbbb . Report_ 014. . https:// w ww. c osmote. g r/ fi xed/ d ocuments/ 1 0280/ 1 38879181/ O TE_ OTE (2016), 2015 Annual Report R 2 015_ e ng. p df/ d bb4a5ed- e 053- 4 fb9- 9 c94- f 9d2913955de . Annual_ eport_ . Myanmar witnesses increased demand for mobile communications https:// Oxford Business Group (2016), oxfordbusinessgroup. co m/ ov erview/ po wering- ahe ad- my www. wi tnessing- ma ssive- inc rease- anmar- mobile- co mmunications . demand- ld- http:// w ww. p ewinternet. o rg/ fi les/ o . Pew Internet (2012), The Best (and Worst) of Mobile Connectivity F iles/ R eports/ 2 012/ P IP_ B est_ W orst_ M obile_ 1 13012. p df . media// Older Adults and Technology Use, Pew Internet (2014), w ww. p ewinternet. o rg/ 2 014/ 0 4/ April 2014. http:// older- adults- an d- te chnology- us e/ . 03/ Telecordia Dynamic Pricing Solution Helps Mobile Operators Increase Revenue Piscataway, N.J. (2010), http:// w ww. p rnewswire. c om/ n ews- r eleases/ t elcordia- . 18 May 2010. Through Network Optimization p ricing- s olution- h elps- m obile- o perators- i ncrease- r evenue- t hrough- n etwork- o ptimization- dynamic- . 94043729. html Annual Report 2015 . https:// w ww. p fs. i s/ l ibrary/ S krar/ Post and Telecom Administration in Iceland (2015), English/ bout- P TA/ P TA_ A nnual_ A R eport_ 2 015. p df Measuring the Information Society Report 2016 219

236 Public Health England (2013), How Healthy Behaviour Supports Children’s Wellbeing https:// . August 2013. g u k/ g overnment/ u ploads/ s ystem/ u ploads/ a ttachment_ d ata/ fi le/ 2 32978/ S mart_ R estart_ 2 80813_ www. ov. p . web. df Research ICT Africa (2012), Internet Going Mobile – Internet access and usage in 11 African countries , www. res earchictafrica. ne t/ pub lications/ Co untry_ http:// Research ICT Africa Policy Briefs, no. 2, 2012. a olicy_ B riefs/ I nternet_ Specific_ g oing_ m obile_-_ I nternet_ a ccess_ P nd_ u sage_ i n_ 1 1_ A frican_ c ountries. pdf . Smyk, D. (2011), Optimization of Dynamic Pricing in Mobile Network: Deriving greater value out of existing C- http:// w ww. e ricsson. c om/ es/ o urportfolio/ p df/ t elcordia_ w hite_ p apers/ M r (White Paper). network assets W 0 36v2. p df . COR- P- Prohibition of Discriminatory Tariffs for Data Services Telecom Regulatory Authority of India (2016), Regulations, 2016 . 8 February 2016. w ww. t rai. g ov. i n/ W riteReadData/ http:// hatsNew/ D ocuments/ W ata_ S ervice. D p df . Regulation_ To be continued: The rules of the road on the internet will always be a work in The Economist (2015), ws/ http:// www. ec onomist. co m/ ne bus iness/ 21 641257- ru les- progress . 31 January 2015. The Economist. ork- road- w ill- a lways- b e- w i p rogress- b e- c ontinued . nternet- UNESCO (2016), Global Education Monitoring Report 2016. Education for People and Planet: Creating Sustainable Futures for All . http:// u nesdoc. u nesco. o rg/ i mages/ 0 024/ 0 02457/ 2 45752e. p df . United Nations (2015), Transforming our World: the 2030 Agenda for Sustainable Development https:// . un or g/ co ntent/ doc uments/ 212 52030 Agenda for Sustainable Development web. sustainabledevelopment. . pdf . ), Implementing WSIS United Nations Commission on Science and Technology for Development (2015 http:// u nctad. o rg/ e n/ P . d tlstict2015d3_ e n. p df ublicationsLibrary/ Outcomes . United Nations Conference on Trade and Development (UNCTAD) (2014), Measuring ICT and Gender: An . http:// u nctad. o rg/ e n/ P ublicationsLibrary/ w Assessment e n. p df . ebdtlstict2014d1_ United Nations Department of Economic and Social Affairs (UNDESA) (2009), The State of the World’s w ww. u n. o rg/ d evelopment/ d esa/ i ndigenouspeoples/ p ublications/ Indigenous Peoples. 009/ https:// 0 9/ 2 state- of- the- wo rlds- indi genous- pe oples- fir st- vo lume . United Nations Development Programme (UNDP) (2012), Mobile Technologies and Empowerment: rg/ http:// w ww. u ndp. o c ontent/ d am/ Enhancing human development through participation and innovation. undp/ De mocratic Governance/Access to Information and E-governance/Mobile Technologies and library/ . Empowerment_EN.pdf Report of the Inter-Agency and Expert Group United Nations Economic and Social Council (ECOSOC) (2016), , E/CN.3/2016/2/Rev.1*, 19 February 2016. on Sustainable Development Goal Indicators http:// unstats. un. org/ u nsd/ s tatcom/ 4 7th- s ession/ d ocuments/ 2 016- 2 - I AEG- S DGs- R ev1- E . p df . United Nations General Assembly (UNGA) (2015a), Transforming our world: the 2030 Agenda for Sustainable http:// Development (Resolution adopted by the General Assembly on 25 September 2015). A/RES/70/1. www. u n. o rg/ g a/ s earch/ v iew_ d oc. a sp? s ymbol= A / R ES/ 7 0/ 1 & L ang= E . United Nations General Assembly (UNGA) (2015b), Outcome document of the high-level meeting of the General Assembly on the overall review of the implementation of the outcomes of the World Summit on Measuring the Information Society Report 2016 220

237 List of references the Information Society wo rkspace. un pan. or g/ si tes/ In ternet/ Do cuments/ , 16 December 2015. http:// pd UNPAN96078. f. Connectivity for Refugees. How Internet United Nations High Commissioner for Refugees (UNHCR) (2016), and Mobile Connectivity can Improve Refugee Well-Being and Transform Humanitarian Action . Geneva, tml w u nhcr. o rg/ p ublications/ o perations/ 5 770d43c4/ c onnecting- r efugees. http:// h ww. . 2016. A World That Counts. United Nations Secretary-General’s Expert Advisory Group on Data Revolution (2014), r ww. u ndatarevolution. o http:// eport/ . w rg/ A nuanced understanding of Internet use and non-use amongst older Van Deursen and Helsper (2015), http:// e prints. l se. a c. u k/ 5 9995/ 1 / H elsper_ a _ n uanced_ u nderstanding_ o f_ I nternet_ u se. p df . . adults Vodafone (2013), Connected Worker. How mobile technology can improve working life in emerging economies . w ww. v odafone. c om/ c ontent/ d am/ s ustainability/ p dfs/ v odafone_ c https:// w orker. p df . onnected_ Heterogeneous Mobile Phone Ownership and Wesolowski, A., Eagle, N., Am, N., Rw, S., and Co, B. (2012), rnal. . 13 71/ . PLoS ONE 7(4):e35319. doi: 10 po ne. 00 35319. jou Usage Patterns in Kenya Information and Communication for Development 2012: Maximizing Mobile. http:// go. World Bank (2012), g/ worldbank. 0J 2CTQTYP0 . or World Development Report 2016 - Digital Dividends. documents. wo rldbank. or g/ World Bank (2016), http:// R e 8 96971468194972881/ p df/ 1 02725- P UB- eplacement- P UBLIC. p df . n/ curated/ . Annual Report 2014: Towards a Wonderful Digital World Zain (2015), w ww. z http:// c om/ m edia/ i mages/ ain. AIN_ A nnual_ R eport_ Z 014_ E N. p df . resumes/ 2 w ain_ http:// ww. z ain. c om/ m edia/ i mages/ r esumes/ Z . Annual Report 2015: The Future is Now Zain (2016), . 2 015_ E nglish_ 2 eport_ p df . Annual_ R Mobile phones, internet, and gender in Myanmar , LIRNEasia and Zainudeen, A. and Galpaya, H. (2015), http:// l irneasia. n et/ w p- GSMA. ontent/ u ploads/ 2 016/ 0 4/ G SMA_ M yanmar_ G ender_ R 2_ S preads. p df c Measuring the Information Society Report 2016 221

238

239 Annex 1 Annex 1. ICT Development Index (IDI) methodology VoIP, it refers to subscriptions that offer the ability This annex outlines the methodology used to to place and receive calls at any time and do not compute the IDI, and provides additional details require a computer. VoIP is also known as voice- on various elements and steps involved, such over-broadband (VoB), and includes subscriptions as the indicators included in the index and their through fixed-wireless, DSL, cable, fibre-optic and definition, the imputation of missing data, the other fixed-broadband platforms that provide fixed normalization procedure, the weights applied to telephony using IP. the indicators and sub-indices, and the results of the sensitivity analysis. 2. Mobile-cellular telephone subscriptions per 100 inhabitants 1. Indicators included in the IDI Mobile-cellular telephone subscriptions refers The selection of indicators was based on certain to the number of subscriptions to a public criteria, including relevance for the index t mobile - elephone service providing access to objectives, data availability and the results of the public switched telephone network (PSTN) various statistical analyses such as the principal using cellular technology. It includes both the 1 The following 11 component analysis (PCA). number of postpaid subscriptions and the number indicators are included in the IDI (grouped by the of active prepaid accounts (i.e. that have been three sub-indices: access, use and skills). active during the past three months). It includes all mobile-cellular subscriptions that offer voice communications. It excludes subscriptions via data a) ICT infrastructure and access indicators cards or USB modems, subscriptions to public mobile data services, private trunked mobile radio, Indicators included in this group provide an telepoint, radio paging and telemetry services. indication of the available ICT infrastructure and individuals’ access to basic ICTs. Data for all these 2 3. International Internet bandwidth (bit/s) per indicators are collected by ITU. Internet user 1. Fixed-telephone subscriptions per 100 refers to the International Internet bandwidth inhabitants total used capacity of international Internet bandwidth, in megabits per second (Mbit/s). Used refers to the sum Fixed-telephone subscriptions international Internet bandwidth refers to the of active analogue fixed-telephone lines, voice- average traffic load of international fibre-optic over-IP (VoIP) subscriptions, fixed wireless local cables and radio links for carrying Internet traffic. loop (WLL) subscriptions, ISDN voice-channel The average is calculated over the 12-month equivalents and fixed public payphones. It includes period of the reference year, and takes into all accesses over fixed infrastructure supporting consideration the traffic of all international voice telephony using copper wire, voice services Internet links. If the traffic is asymmetric, i.e. if using Internet Protocol (IP) delivered over fixed there is more incoming (downlink) than outgoing (wired)-broadband infrastructure (e.g. DSL, fibre (uplink) traffic, the average incoming (downlink) optic), and voice services provided over coaxial- traffic load is used. The combined average traffic cable television networks (cable modem). It load of different international Internet links can also includes fixed WLL connections, defined as be reported as the sum of the average traffic services provided by licensed fixed-line telephone loads of the individual links. International Internet operators that provide last-mile access to the bandwidth (bit/s) per Internet user is calculated by subscriber using radio technology, where the call converting to bits per second and dividing by the is then routed over a fixed-line telephone network total number of Internet users. (not a mobile-cellular network). In the case of Measuring the Information Society Report 2016 223

240 4. Percentage of households with a computer b) ICT use indicators refers to a desktop computer, laptop Computer The indicators included in this group capture ICT (portable) computer, tablet or similar handheld intensity and usage. Data for all these indicators 5 computer. It does not include equipment with are collected by ITU. some embedded computing abilities, such as smart TV sets, or devices with telephony as a main 1. Percentage of individuals using the Internet function, such as mobile phones or smartphones. Individuals using the Internet refers to people who Household with a computer means that the used the Internet from any location and for any computer is available for use by all members of the purpose, irrespective of the device and network household at any time. The computer may or may used, in the last three months. Usage can be via a not be owned by the household, but should be computer (i.e. desktop computer, laptop computer, 3 considered a household asset. tablet or similar handheld computer), mobile phone, games machine, digital TV, etc.). Access can Data are obtained by countries through national be via a fixed or mobile network. household surveys and are either provided directly to ITU by national statistical offices (NSOs) or Data are obtained by countries through national obtained by ITU through its own research, for household surveys and are either provided directly example from NSO websites. There are certain to ITU by NSOs or obtained by ITU through its own data-related limits to this indicator, insofar research, for example from NSO websites. There as estimates have to be calculated for many are certain data-related limits to this indicator, developing countries which do not yet collect insofar as estimates have to be calculated for ICT household statistics. Over time, as more data many developing countries which do not yet become available, the quality of the indicator will collect ICT household statistics. Over time, as improve. more data become available, the quality of the indicator will improve. 5. Percentage of households with Internet access 2. Fixed-broadband subscriptions per 100 The Internet is a worldwide public computer inhabitants network. It provides access to a number of communication services, including the World Wide Fixed-broadband subscriptions refers to fixed Web, and carries e-mail, news, entertainment subscriptions for high-speed access to the public and data files, irrespective of the device used (not Internet (a TCP/IP connection) at downstream assumed to be only a computer; it may also be a speeds equal to or greater than 256 kbit/s. This mobile telephone, tablet, PDA, games machine, includes cable modem, DSL, fibre-to-the-home/ digital TV, and so on). Access can be via a fixed or building, other fixed-broadband subscriptions, mobile network. Household with Internet access satellite broadband and terrestrial fixed wireless means that the Internet is available for use by all broadband. The total is measured irrespective of 4 members of the household at any time. the method of payment. It excludes subscriptions that have access to data communications Data are obtained by countries through national (including the Internet) via mobile-cellular household surveys and are either provided directly networks. It includes fixed WiMAX and any other to ITU by NSOs or obtained by ITU through its own fixed wireless technologies, and both residential research, for example from NSO websites. There subscriptions and subscriptions for organizations. are certain data-related limits to this indicator, insofar as estimates have to be calculated for 3. Active mobile-broadband subscriptions per 100 many developing countries which do not yet inhabitants collect ICT household statistics. Over time, as more data become available, the quality of the Active mobile-broadband subscriptions refers indicator will improve. to the sum of standard mobile-broadband subscriptions and dedicated mobile-broadband subscriptions. The subscriptions can be used Measuring the Information Society Report 2016 224

241 Annex 1 sed or computer-based (USB/ ba - through handset 2. Gross enrolment ratio (secondary and tertiary dongle) devices. It covers actual subscribers, not level) potential subscribers, even though the latter may According to UIS, the gross enrolment ratio is “the have broadband-enabled handsets. total enrolment in a specific level of education, Standard mobile-broadband subscriptions regardless of age, expressed as a percentage • refers to active mobile-cellular subscriptions of the eligible official school-age population with advertised data speeds of 256 kbit/s corresponding to the same level of education in a or higher that allow access to the greater given school-year.” Internet via HTTP and which have been used to set up an Internet data connection using 2. Imputation of missing data IP in the past three months. Standard SMS and MMS messaging do not count as active A critical step in the construction of the index is to Internet data connection, even if messages create a complete data set, without missing values. are delivered via IP. A number of imputation techniques can be applied 7 Each of the imputation to estimate missing data. Dedicated mobile-broadband data • techniques, like any other method employed in the subscriptions refers to subscriptions to process, has its own strengths and weaknesses. dedicated data services (over a mobile The most important consideration is to ensure that network) that allow access to the greater the imputed data will reflect a country’s actual Internet and are purchased separately from level of ICT access, usage and skills. voice services, either as a stand-alone service (e.g. using a data card such as a USB modem/ Given that ICT access and usage are both dongle) or as an add-on data package to correlated with national income, hot-deck voice services which requires an additional imputation was chosen as the method for imputing subscription. All dedicated mobile-broadband the missing data where previous year data are subscriptions with recurring subscription fees not available to calculate growth rates. Hot-deck are included regardless of actual use. Prepaid imputation uses data from countries with “similar” mobile-broadband plans require use of the characteristics, such as gross national income monthly data allowance where there is no (GNI) per capita and geographical location. For monthly subscription. This indicator could also example, missing data for a given country A were include mobile WiMAX subscriptions. estimated for a certain indicator by first identifying countries in the same region with similar levels of GNI per capita and similar levels for an indicator c) ICT skills indicators that has a known relationship to the indicator to be estimated. For instance, Internet use data for Data on mean years of schooling and gross country A was estimated by using Internet use data secondary and tertiary enrolment ratios are for country B from the same region with a similar collected by the UNESCO Institute for Statistics GNI per capita and similar level of fixed Internet (UIS). bro and wireless - adband subscriptions. The same approach was applied to estimate missing data for 1. Mean years of schooling all indicators included in the index. is the average number Mean years of schooling of completed years of education of a country’s 3. Normalization of data population, excluding years spent repeating individual grades. It is estimated by UIS using the Normalization of data is necessary before any distribution of the population by age group and aggregation can take place, in order to ensure that the highest level of education attained in a given the data set uses the same unit of measurement. year, and time series data on the official duration Regarding the indicators selected to construct 6 of each level of education. the IDI, the values must be converted into the same unit of measurement, since some values are expressed as a percentage of the population/total Measuring the Information Society Report 2016 225

242 households, where the maximum value is 100, a value derived by examining the distribution while other indicators can have values exceeding of countries based on their value for mobile- 100, such as mobile-cellular and active mobile- cellular subscriptions per 100 inhabitants in 2013. For countries where postpaid is the broadband penetration or international Internet predominant mode of subscription, 120 is the bandwidth (expressed as bit/s per user). maximum value attained, while in countries Certain particularities need to be taken into where prepaid is dominant (57 per cent of all consideration in selecting the normalization countries included in the IDI have more than 80 per cent prepaid subscriptions), 120 is also method for the IDI. For example, in order to the maximum value attained in a majority of identify the digital divide, it is important to measure the performance of countries relative countries. It was therefore concluded that (i.e. the divide among countries). Secondly, the 120 is the ideal value that a country could normalization procedure should produce index attain, irrespective of the predominant type of results that allow countries to track progress in mobile subscription. Although the distribution their evolution towards an information society of 2015 values may differ slightly from that of 2013 values, the ideal value of 120 was used over time. to calculate this year’s IDI, in the interests of consistency with the value used in previous A further important criterion in selecting the years. normalization method is replicability by countries, as some countries have shown a strong interest in - applying the index methodology at the national or Fixed-telephone subscriptions per 100 regional level. Certain methods therefore cannot inhabitants, which ranged from zero to 128.1 be applied, for example those that rely on the in 2015. The reference value was calculated by values of other countries, which might not be adding two standard deviations to the mean, available to users. resulting in a rounded value of 60 per 100 inhabitants. distance to a reference measure For the IDI, the was used as the normalization method. The Fixed-broadband subscriptions per 100 - inhabitants. Values ranged from zero to 47.5 reference measure is the ideal value that could be per 100 inhabitants in 2015. In line with fixed- reached for each variable (similar to a “goalpost”). telephone subscriptions, the ideal value was For all the indicators chosen, this will be 100, defined as 60 per 100 inhabitants. except in regard to the following five indicators: International Internet bandwidth per Internet Mean years of schooling. Values ranged - - user, which in 2015 ranged from 28 (bit/s/ from 1.4 to 13.8 in 2015. The ideal value of user) to almost 7 186 378. Values for this 15 is used for this indicator, which refers to indicator vary significantly between countries. the projected maximum number of years of 8 schooling by 2025. To diminish the effect of the enormous dispersion of values, the data were first After normalizing the data, the individual series converted to a logarithmic (log) scale. Outliers were all rescaled to identical ranges, from 1 to 10. were then identified using a cut-off value This was necessary in order to compare the values calculated by adding two standard deviations of the indicators and the sub-indices. to the mean of the rescaled values, resulting in a log value of 5.99. 4. Weighting and aggregation Mobile-cellular subscriptions, which in 2015 - ranged from 23.9 to 324.4 per 100 inhabitants. The indicators and sub-indices included in the The reference value for mobile-cellular IDI were weighted on the basis of the PCA results subscriptions was reviewed in the previous 9 obtained when the index was first computed. edition of the index and was lowered to 120, Measuring the Information Society Report 2016 226

243 Annex 1 We ights used for indicators and sub-indices included in the IDI Annex Box 1.1: Weights Weights (indicators) (sub - indices) 0.40 ICT access Fixed-telephone subscriptions per 100 inhabitants 0.20 Mobile-cellular telephone subscriptions per 100 inhabitants 0.20 International Internet bandwidth per Internet user 0.20 Percentage of households with a computer 0.20 Percentage of households with Internet access 0.20 0.40 ICT use Percentage of individuals using the Internet 0.33 Fixed-broadband Internet subscriptions per 100 inhabitants 0.33 Active mobile-broadband subscriptions per 100 inhabitants 0.33 0.20 ICT skills Mean years of schooling 0.33 Secondary gross enrolment ratio 0.33 0.33 Tertiary gross enrolment ratio Source: ITU. • ICT skills are approximated by mean years of 5. Calculating the IDI schooling, secondary gross enrolment ratio and tertiary gross enrolment ratio. Sub-indices were computed by summation of the weighted values of the indicators included in the The values of the sub-indices were calculated respective subgroup. first by normalizing the indicators included in each sub index in order to obtain the same unit - is measured by fixed-telephone ICT access • of measurement. The applied in reference values subscriptions per 100 inhabitants, mobile- the normalization process were discussed above. cellular subscriptions per 100 inhabitants, The sub-index value was calculated by taking the international Internet bandwidth per Internet simple average (using equal weighting) of the user, the percentage of households with a normalized indicator values. computer and the percentage of households with Internet access. For computation of the final index, the ICT access and ICT use sub-indices were each given ICT use is measured by the percentage of • a 40 per cent weighting, and the skills sub-index individuals using the Internet, fixed-broadband (because it is based on proxy indicators) a 20 per Internet subscriptions per 100 inhabitants and cent weighting. The final index value was then active mobile-broadband subscriptions per computed by summation of the weighted sub- 100 inhabitants. indices. Annex Box 1.2 illustrates the process of computing the IDI for the Republic of Korea (which tops the IDI 2016). Measuring the Information Society Report 2016 227

244 Annex Box 1.2: ample of how to calculate the IDI value Ex Korea (Rep.) Indicators ICT access Ideal value* 60 58.1 Fixed-telephone subscriptions per 100 inhabitants a b Mobile-cellular telephone subscriptions per 100 inhabitants 118.5 120 c International Internet bandwidth per Internet user** 976,696 46,764 Percentage of households with a computer 100 77.1 d Percentage of households with Internet access e 98.8 100 ICT use f 100 89.9 Percentage of individuals using the Internet Fixed-broadband Internet subscriptions per 100 inhabitants g 40.2 60 h Active mobile-broadband subscriptions per 100 inhabitants 100 109.7 ICT skills i Mean years of schooling 11.9 15 Secondary gross enrolment ratio 97.7 j 100 Tertiary gross enrolment ratio 100 k 95.3 Normalized values Formula Weight ICT access z1 Fixed-telephone subscriptions per 100 inhabitants a/60 0.20 0.97 z2 Mobile-cellular telephone subscriptions per 100 inhabitants 0.20 0.99 b/120 z3 International Internet bandwidth per Internet user log(c)/5.99 0.20 0.79 z4 Percentage of households with a computer d/100 0.20 0.77 z5 Percentage of households with Internet access e/100 0.20 0.99 ICT use z6 f/100 0.90 0.33 Percentage of individuals using the Internet z7 Fixed-broadband Internet subscriptions per 100 inhabitants 0.33 0.67 g/60 h/100 0.33 z8 Active mobile-broadband subscriptions per 100 inhabitants 1.00 ICT skills z9 Mean years of schooling i/15 0.33 0.79 z10 Secondary gross enrolment ratio j/100 0.33 0.98 z11 Tertiary gross enrolment ratio 0.33 0.95 k/100 Sub-indices Formula Weight ICT access sub-index (L) y1+y2+y3+y4+y5 0.40 0.90 y1 Fixed-telephone subsriptions per 100 inhabitants z1*.20 0.19 y2 Mobile-cellular telephone subscriptions per 100 inhabitants z2*.20 0.20 y3 International Internet bandwidth per Internet user z3*.20 0.16 y4 Percentage of households with a computer 0.15 z4*.20 y5 Percentage of households with Internet access z5*.20 0.20 y6+y7+y8 0.40 0.86 ICT use sub-index (M) z6*.33 0.30 y6 Percentage of individuals using the Internet y7 Fixed-broadband Internet subscriptions per 100 inhabitants z7*.33 0.22 y8 Active mobile-broadband subscriptions per 100 inhabitants z8*.33 0.33 ICT skills sub-index (N) 0.20 0.91 y9+y10+y11 y9 Mean years of schooling z9*.33 0.26 y10 Secondary gross enrolment ratio z10*.33 0.33 y11 Tertiary gross enrolment ratio z11*.33 0.32 8.84 IDI ICT Development Index 2016 ((L*.40)+(M*.40)+(N*.20))*10 Note: *The ideal value for indicators a, b, c and g was computed by adding two standard deviations to the mean value of the indicator. **To diminish the effect of the large number of outliers at the high end of the value scale, the data were first transformed to a logarithmic (log) scale. The ideal value of 976'696 bit/s per Internet user is equivalent to 5.99 if transformed to a log scale. Source: ITU. Measuring the Information Society Report 2016 228 Measuring the Information Society Report 2015 8

245 Annex 1 country rankings for different combinations of 6. Sensitivity analysis the processes mentioned above. Results show that, while the computed index values change, Sensitivity analysis was carried out to investigate the message remains the same. The IDI was found the robustness of the index results in terms of to be extremely robust with regard to different the relative position in the overall ranking, using methodologies, with the exception of certain different combinations of methods and techniques countries including in particular those in the to compute the index. “high” group. Potential sources of variation or uncertainty can The relative position of countries included in be attributed to different processes employed the “high” group (see Chapter 1) can change in the computation of the index, including the depending on the methodology used. Caution selection of individual indicators, the imputation should therefore be exercised in drawing of missing values and the normalization, weighting conclusions based on these countries’ rankings. and aggregation of the data. However, the relative position of countries included in the “low” group is in no way affected Each of the processes or combination of processes by the methods or techniques used, and the affects the IDI value. A number of tests were countries in this group ranked low in all index carried out to examine the robustness of the computations using different methodologies. This IDI results (rather than the actual values). The confirms the results conveyed by the IDI. tests computed the possible index values and Measuring the Information Society Report 2016 229

246 Endnotes 1 PCA was used to examine the underlying nature of the data. A more detailed description of the analysis is available in Annex 1 to the 2009 report “Measuring the Information Society. The ICT Development Index” (ITU, 2009). 2 More information about the indicators is available in the ITU “Handbook for the collection of administrative data on telecommunications/ICT” 2011, (ITU 2011) and the ITU “Manual for Measuring ICT Access and Use by Households and Individuals” (ITU 2014). 3 This definition reflects the revisions agreed by the ITU Expert Group on ICT Household Indicators (EGH) at its meeting in ww w. itu. int/ en/ ITU- D/ Sta tistics/ Documents/ Sao Paulo, Brazil, 4-6 June 2013. See http:// ev ents/ brazil2013/ Final_ report_ EGH. pdf). 4 See footnote 3. 5 See footnote 2. 6 http:// www. uis. unesco. org See Educ ation/ Documents/ Mean- year s- schooling- indica tor- methodology - en. pdf . / 7 See OECD and European Commission (2008). 8 hdr. undp. org / sites/ def ault/ See Human Development Index (HDI), Technical Notes, available at http:// hdr2015_ files/ technical_ notes. pdf. 9 For more details, see Annex 1 to ITU (2009). Measuring the Information Society Report 2016 230

247 Annex 2 . ICT price data methodology Annex 2 Price data collection and sources The mobile-cellular sub-basket The mobile-cellular sub-basket refers to the The price data presented in this report were price of a standard basket of mobile monthly collected in the fourth quarter of 2015. With usage for 30 outgoing calls per month (on-net/ the exception of the data on mobile-broadband prices, which were collected by ITU directly from off-net to a fixed line and for peak and off-peak 1 operators’ websites times) in predetermined ratios, plus 100 SMS , all data were collected 3 messages ICT Price Basket Questionnaire, . It is calculated as a percentage of a through the ITU country’s average monthly GNI p.c. and is also which was sent to the administrations and presented in USD and PPP$. The mobile-cellular statistics contacts of all 193 ITU Member States 2 Through the questionnaire, sub-basket is based on prepaid prices, although in October 2015. contacts were requested to provide 2015 data postpaid prices are used for countries where for fixed-telephone, mobile-cellular and fixed- prepaid subscriptions make up less than two per cent of all mobile-cellular subscriptions. broadband prices; the 2013 and 2014 prices were included for reference, where available. The mobile-cellular sub-basket is largely based For those countries that did not reply to the ITU on, but does not entirely follow, the 2009 ICT Price Basket Questionnaire and for mobile- methodology of the OECD low-user basket, which broadband services, price data were collected is the entry-level basket with the smallest number directly from operators’ websites and/or through of calls included (OECD, 2010). Unlike the 2009 direct correspondence. Price data were collected OECD methodology, which is based on the prices from the operator with the largest market share, of the two largest mobile operators, the ITU as measured by the number of subscriptions. mobile sub-basket uses only the largest mobile Insofar as, for many countries, it is not clear operator’s prices. Nor does the ITU mobile-cellular which Internet service provider (ISP) has the sub-basket take account of calls to voicemail dominant market share, preference was given per cent (which in the OECD basket represent four to prices offered by the (former) incumbent of all calls) or non-recurring charges, such as the telecommunication operator. In some cases, one-time charge for a SIM card. The basket gives especially where prices were not clearly advertised the price of a standard basket of mobile monthly or were indicated only in the local language, and usage in USD determined by OECD for 30 outgoing where operators did not respond to queries, calls per month in predetermined ratios, plus 100 alternative operators were chosen. All prices were 4 The cost of national SMS is the SMS messages. converted into United States dollars using IMF’s charge to the consumer for sending a single SMS average annual rate of exchange for 2015, and text message. Both on-net and off-net SMS prices into PPP$ using World Bank conversion factors are taken into account. The basket considers for 2014 (as published in February 2016). Prices on-net and off-net calls as well as calls to a fixed are also presented as a percentage of countries’ 5 and, since the price of a call often telephone monthly gross national income per capita (GNI p.c.) depends on the time of day or week it is made, using GNI p.c. values from the World Bank (Atlas peak, off-peak and weekend periods are also taken method) for 2014 (as published in February 2016) into consideration. The call distribution is outlined or the latest available year adjusted in accordance in Annex Table 2.1. with international inflation rates. Price data for 2008, 2009, 2010, 2011, 2012, 2013 and 2014, Prepaid prices were chosen because they are often which are also shown and used in this chapter, the only payment method available to low-income were collected in previous years (always during users, who might not have a regular income and the second half of the respective year), in national will thus not qualify for a postpaid subscription. currencies, and converted using the average Rather than reflecting the cheapest option annual rates of exchange. available, the mobile-cellular sub-basket therefore corresponds to a basic, representative (low-usage) Measuring the Information Society Report 2016 231

248 Annex Table 2.1: OE CD mobile-cellular low-user call distribution (2009 methodology) Call Off-net TOTA L To fixed distribution by On-net time of day (%) 56.0 17.0 26.0 100.0 100.0 Call distribution (%) 5.2 16.9 7.9 30.0 Calls 3.6 7.8 2.4 46.0 Peak 13.8 29.0 Off-peak 1.5 4.9 2.3 8.7 25.0 Weekend 4.2 2.0 7.5 1.3 Duration (minutes per call) 2.0 1.6 1.7 Duration (total minutes of calls) 27.0 13.4 50.9 N/A 10.4 Peak 4.8 12.4 6.2 23.4 46.0 Off-peak 7.8 3.0 3.9 14.8 29.0 Weekend 2.6 3.4 12.7 25.0 6.8 Calls 30 calls per month 100 SMSs per month (50 on-net, 50 off-net) SMS Source: ITU, based on OECD (2010). package available to all customers. In countries if applicable) of a postpaid subscription is added to where no prepaid offers are available, the monthly the basket. To make prices comparable, a number of rules are applied (see Annex Box 2.1). fixed cost (minus the free minutes of calls included, Annex Box 2.1: Ru les applied in collecting mobile-cellular prices 1. e prices of the operator with the largest market share (measured by the number of Th subscriptions) are used. If prices vary between different regions of the country, prices refer to those applied in the largest city (in terms of population) or in the capital city. ice data should be collected in the currency in which the prices are advertised, including 2. Pr taxes. If prices are not advertised in local currency, a note should be added specifying the currency. Pr ices refer to prepaid plans. Where the operator offers different packages with a certain 3. number of calls and/or SMS messages included, the cheapest one on the basis of 30 calls and 100 SMSs should be selected. If, instead of a pay-per-use plan, a package is selected for the whole basket (e.g. a bundle including 100 SMSs, 60 minutes and 100MB) or for some of its elements (e.g. a package including 100 SMSs), this should be indicated in the notes. In countries where prepaid subscriptions account for less than 2 per cent of the total subscription base, postpaid prices may be used. In this case, the monthly subscription fee, plus any free minutes, will be taken into consideration for the calculation of the mobile- cellular sub-basket. er-minute prices are only advertised in internal units rather than in national currency, If p 4. the price of the top-up/refill charge is used to convert internal units into national currency. If there are different refill prices, then the “cheapest/smallest” refill card is used. If different refill charges exist depending on the validity period, the 30-day validity period (or that closest to 30 days) is used. Measuring the Information Society Report 2016 232

249 Annex 2 5. Prices refer to a regular (non-promotional) plan and exclude special or promotional offers, imited discounts or options such as special prices to certain numbers or restricted to new l customers, or plans where calls can only be made during a limited number of (or on specific) days during the month. ubscribers can chose “favourite” numbers (for family, friends, etc.) with a special price, If s 6. this special price will not be taken into consideration, irrespective of the quantity of numbers involved. Pr ices refer to outgoing local calls. If different rates apply for local and national calls, then 7. the local rate is used. If different charges apply depending on the mobile operator called, the price of calls to the operator with the second largest market share (measured by the number of subscriptions) should be used, indicating in the notes the rates for calling to other mobile operators. If charges apply to incoming calls, these are not taken into consideration. If pr ices vary between minutes (1st minute = price A, 2nd minute = price B, 3rd minute = 8. price C), the sum of the different prices is divided by the number of different prices (e.g. price per minute = (A+B+C)/3). If pr ices vary beyond three minutes, the average price per minute is calculated based on the 9. first three minutes. If t here is a connection cost per call, then this is taken into consideration in the formula for 10. the mobile-cellular sub-basket, based on 30 calls. 11. If t here are different off-peak prices, then the one that is the cheapest before midnight is used. If the only off-peak period is after midnight, then this is not used. Instead, the peak price is used. here are different peak prices, the most expensive one during the daytime is used. If t 12. 13. If t here are different weekend prices, the price that applies to Sundays during the daytime is used (or the equivalent day in countries where weekends are not on Sundays). If there is no weekend price, the average peak and off-peak price that is valid during the 14. week is used. 15. If peak and off-peak SMS prices exist, the average of both is used for on-net and off-net SMSs. 16. If c alls are charged by call or by hour (and not by the minute), the mobile-cellular sub-basket formula will be calculated on the basis of 30 calls or 50.9 minutes. Similarly, if calls are charged by call or by number of minutes for a specific network/time of the day, this will be taken into account for that particular network/time of the day. 17. If mo nthly, recurring charges exist, they are added to the sub-basket. Source: ITU. Measuring the Information Society Report 2016 233

250 data volume. Where providers set a limit of less The fixed-broadband sub-basket GB on the amount of data that can be than 1 transferred within a month, then the price per The fixed-broadband sub-basket refers to the additional byte is added to the monthly price in price of a monthly subscription to an entry- order to calculate the cost of 1 GB of data per level fixed-broadband plan. It is calculated as a month. Preference is given to the most widely percentage of a country’s average monthly GNI used fixed (wired)-broadband technology (DSL, p.c., and is also presented in USD and PPP$. For fibre, cable, etc.). The sub-basket does not comparability reasons, the fixed-broadband include the installation charges, modem prices or sub-basket is based on a monthly data usage telephone-line rentals that are often required for of (a minimum of) 1 GB. For plans that limit the a DSL service. The price represents the broadband monthly amount of data transferred by including entry plan in terms of the minimum speed of 256 data volume caps below 1 GB, the cost for the kbit/s, but does not take into account special offers additional bytes is added to the sub-basket. The that are limited in time or to specific geographical minimum speed of a broadband connection is areas. The plan does not necessarily represent the 256 kbit/s. fastest or most cost-effective connection since the price for a higher-speed plan is often cheaper in Where several offers are available, preference is relative terms (i.e. in terms of the price per Mbit/s) given to the cheapest available connection that (see Annex Box 2.2). offers a speed of at least 256 kbit/s and 1 GB of Annex Box 2.2: Ru les applied in collecting fixed-broadband Internet price data 1. e prices of the operator with the largest market share (measured by the number of fixed- Th broadband subscriptions) should be used. 2. ice data should be collected in the currency in which the prices are advertised, including Pr taxes. If prices are not advertised in local currency, a note should be added specifying the currency. On ly residential, single-user price data should be collected. If prices vary between different 3. regions of the country, prices applying to the largest city (in terms of population) should be provided. If that information is not available, prices applying to the capital city should be reported. The selected city should be mentioned in a note in the monthly subscription indicator. Fr om all fixed-broadband plans meeting the above-mentioned criteria, the cheapest plan on 4. B monthly usage and an advertised download speed of at least 256 kbit/s G the basis of a 1 should be selected. If there is a price distinction between residential and business tariffs, the residential tariff should be used. 5. If t he plan selected places no limit on the monthly data usage, the cap should be set at 0 and a note added to that indicator specifying “unlimited”. 6. If o perators propose different commitment periods, the 12-month plan (or the one closest to this commitment period) should be used. If the plan selected requires a longer commitment (i.e. over 12 months), it should be indicated in the note regarding the monthly subscription. Furthermore, if different prices apply (e.g. a discount price for the first year, and a higher th month), then the price after the discount period should be selected (e.g. price as of the 13 th the price as of the 13 month). The discount price charged during the initial period should be indicated in a note regarding the monthly subscription charge. This is because the initial price paid is considered a limited/discount price, whereas the price subsequently charged is the regular price. Measuring the Information Society Report 2016 234

251 Annex 2 7. Price data should be collected for the fixed (wired)-broadband technology with the greatest umber of subscriptions in the country (DSL, cable, etc.). n e same price plan should be used for collecting all the data specified. For example, if Plan Th 8. A is selected for the fixed-broadband service, according to the criteria mentioned above, the elements in Plan A shall be taken into account in regard to the monthly subscription, the excess-charge price, the volume of data that can be downloaded, etc. Pr ice data should be collected for regular (non-promotional) plans and should not include 9. promotional offers or limited or restricted discounts (e.g. for students only, for existing customers, etc.). Wi th convergence, operators are increasingly providing multiple (bundled) services such as 10. voice telephony, Internet access and television reception over their networks. They often bundle these offers into a single subscription. This can present a challenge for price data collection, since it may not be possible to isolate the prices for one service. It is preferable to use prices for a specific service (i.e. unbundled); if this is not possible, then the additional services that are included in the price plan should be specified in a note. The cost of a fixed-telephone line should be excluded if it can be used for other services as well. If a monthly rental for the physical line is not required (e.g. naked DSL), this should be mentioned in a note. If a monthly rental of a fixed-telephone line is required, this should also be explained in a note. Source: ITU. For plans that were limited in terms of validity Mobile-broadband prices (less than 30 days), the price of the additional days was calculated and added to the base package in ITU has been collecting mobile-broadband order to obtain the final price. Two possibilities price data through its annual ICT Price Basket exist, depending on the operator, for extending Questionnaire since 2012. The collection of a plan that is limited in terms of data allowance mobile-broadband price data from ITU Member (or validity). The customer either (i) continues to States was agreed upon by the Expert Group use the service and pays an excess usage charge 6 in on Telecommunication/ICT Indicators (EGTI) 7 or (ii) purchases an additional for additional data, 2012, and revised by EGTI in 2013 in the light of (add-on) package. Thus, for some countries, prices the lessons learned from the first data collection presented in this chapter reflect the price of the exercise. The revised methodology was endorsed base package plus an excess-usage charge (e.g. a by the eleventh World Telecommunication/ICT base package including 400 MB plus the price for Indicators Symposium held in December 2013 100 MB of excess usage for a monthly usage of in Mexico City, and was applied in the 2014 data MB), or a multiplication of the base package 500 collection. MB plan for a price (e.g. twice the price of a 250 MB). monthly usage of 500 To capture the prices of different data packages, covering both prepaid and postpaid services The plans selected represent the least expensive and support by different devices (handset and offers that include the minimum amount of data computer), mobile-broadband price data were for each respective mobile-broadband plan. The collected for two different data thresholds, based guiding principle is to base each plan on what Box 2.3). on a set of rules (see Annex customers could and would purchase given the data allowance and validity of each plan. Measuring the Information Society Report 2016 235

252 8 Annex Box 2.3: Rules applied in collecting mobile-broadband prices Price data should be collected based on one of the following technologies: UMTS, HSDPA+/ 1. HS DPA, CDMA2000 and IEEE 802.16e. Prices applying to WiFi or hotspots should be excluded. ice data should be collected in the currency in which they are advertised, including taxes. If 2. Pr prices are not advertised in local currency, a note should be added specifying the currency. On ly residential, single-user prices should be collected. If prices vary between different 3. regions of the country, prices applying to the largest city (in terms of population) or to the capital city should be provided. Pr ice data should be collected for both: a) handset-based mobile-broadband subscriptions 4. and b) c omputer-based mobile-broadband subscriptions. le-broadband price data should be collected from the operator with the largest market 5. Mobi share measured by the number of mobile-broadband subscriptions. If this information is not available, mobile-broadband price data should be collected from the mobile-cellular operator with the largest market share (measured by the number of mobile-cellular subscriptions) in the country. fferent operators can be chosen for different mobile-broadband services if: a) there are Di 6. different market leaders for specific segments (postpaid, prepaid, computer-based, handset- based); b) there is no offer available for a specific sub-basket. Pr ice data should be collected for prepaid and postpaid services, for both handset and 7. computer-based plans. If there are several plans, the plan satisfying the indicated data volume requirement should be used. ere operators propose different commitment periods for postpaid mobile-broadband Wh 8. plans, the 12-month plan (or the plan closest to this commitment period) should be selected. A note should be added if only longer commitment periods are offered. 9. Pr ice data should be collected for the cheapest plan, with a data volume allowance of a minimum of: B for USB/dongle (computer-based) subscription 1G i. ii. 50 0MB for the handset-based subscription The selected plan should not necessarily be the one with the cap closest to 500 MB or 1 GB, but include a minimum of 500 MB/1 GB. This means, for example, that if an operator offers a 300 MB and an 800 MB plan, the 800 MB plan or twice the 300 MB plan (if the pack age can be purchased twice for a monthly capacity of 600 MB) should be selected for the 500 MB et. The cheapest option should be selected. sub-sub-bask Data volumes should refer to both upload and download data volumes. If prices are linked to “hours of use” and not to data volumes, this information should be added in a separate note (ITU will not be able to include these cases in a comparison). Measuring the Information Society Report 2016 236

253 Annex 2 10. The validity period considered for the basket is 30 days or four weeks. If a plan with a alidity of 15 days is selected, it will be taken into consideration twice to cover the whole v period. Likewise, if a plan with a validity of a day or a week is selected, it will be taken into consideration as many times as necessary to cover a period of four weeks. The cheapest plan on the basis of a validity period of 30 days or four weeks should be selected. Pr eference should be given to packages (including a certain data volume). Pay-as-you-go 11. offers should be used when they are the cheapest option for a given basket or the only option available. If operators charge different pay-as-you-go rates depending on the time of the day (peak/off-peak), then the average of both should be recorded. Night-time data allowances will not be considered. Ev en if the plan is advertised as “unlimited”, the fine print should be read carefully since the 12. data volumes are usually limited, either by throttling (limiting the speed) or by cutting off the service. Da ta on non-recurrent fees, such as installation/set-up fees, are not collected. 13. Pr 14. eference should be given to the cheapest available package even if this is bundled with other services (e.g. with voice services). If the plan chosen includes other services besides mobile-broadband access, these should be specified in a note. Pr ices refer to a regular (non-promotional) plan and exclude promotional offers and limited 15. discounts or special user groups (e.g. existing clients). Special prices applying to a certain type of phone (iPhone/Blackberry, iPad) should be excluded. Night-time allowances are not included. Source: ITU. Measuring the Information Society Report 2016 237

254 Endnotes 1 Price data for mobile-broadband services were collected by ITU, in collaboration with Teligen/Strategy Analytics. 2 Data for fixed-telephone, mobile-cellular and fixed-broadband have been collected since 2008 through the ITU ICT Price Basket Questionnaire, which is sent out annually to all ITU Member States/national statistics contacts. 3 On-net refers to a call made to the same mobile network, while off-net and fixed-line refer to calls made to other (competing) mobile networks and to a fixed-telephone line, respectively. 4 See OECD (2010). 5 See footnote 3. 6 EGTI was created in May 2009 with the mandate to revise the list of ITU supply-side indicators (i.e. data collected from operators), as well as to discuss outstanding methodological issues and new indicators. EGTI is open to all ITU members http:// www. itu. and experts in the field of ICT statistics and data collection. It works through an online discussion forum ( int/ ITU- D/ ict/ ExpertGroup/ def ault. ) and face-to-face meetings. EGTI reports to the World Telecommunication/ICT asp Indicators Symposium (WTIS). 7 Some operators throttle speeds after the data allowance included in the base package has been reached. Customers can then pay an excess-usage charge in order to continue to have full-speed connections. In some cases, even throttled kbit/s, according to ITU’s definition). speeds are still considered to be broadband (i.e. equal to or greater than 256 Measuring the Information Society Report 2016 238

255 Annex 3 Annex 3. Statistical tables of indicators used to compute the IDI Measuring the Information Society Report 2016 239

256 Access indicators Fixed-telephone Percentage of Percentage of Mobile-cellular International Internet subscriptions per households households subscriptions bandwidth Economy with Internet 100 inhabitants with computer per 100 inhabitants Bit/s per Internet user 2015 2014 2015 2014 2014 2015 2014 2015 2015 2014 1 0.3 10’213 0.3 58.8 6’850 2.9 3.0 61.6 3.9 1 Afghanistan 2.7 30’660 26’117 25.7 106.4 26.6 105.5 35.5 7.1 23.5 2 Albania 7.4 1 1 37.0 108.4 24’669 7.8 8.0 32.0 113.0 19.3 31.9 Algeria 3 30’119 82’857 58’543 88.1 82.7 82.6 81.6 48.0 84.8 82.6 Andorra 47.7 4 6’518 6’236 1.2 1.3 10.4 11.1 8.8 10.2 63.5 5 60.8 Angola Antigua & 21.9 6 132.1 57.6 137.2 54.0 13.1 56.3 71’825 73’997 56.1 Barbuda 2 2 146.5 143.9 48’065 23.5 46’145 62.1 65.1 52.0 55.5 7 Argentina 24.0 67’871 18.4 57.0 115.1 64.7 56.2 52.4 37’749 115.9 19.2 Armenia 8 2 3 1 3 38.9 132.8 80.4 75’569 38.0 81’564 81.2 9 84.3 85.9 Australia 131.2 3 4 151.9 90’501 157.4 81.5 42.2 82.1 82.4 81.0 79’636 38.2 10 Austria 4 111.3 26’205 18.7 35’127 110.9 62.4 76.2 76.7 18.9 11 Azerbaijan 60.3 5 2 61.1 225’877 78’227 69.2 80.3 58.3 31.2 67.7 82.3 32.8 Bahamas 12 173.3 81.0 185.3 41’103 13 47’205 94.6 94.8 Bahrain 88.7 21.2 20.5 1 4 5 6’181 6.9 0.5 8.2 4’583 9.9 83.4 11.0 80.0 0.6 Bangladesh 14 106.8 15 116.5 126’089 Barbados 247’474 69.0 70.8 61.5 62.9 52.9 54.6 59.1 139’374 122.5 59.9 63.1 123.6 49.0 57.1 142’536 48.5 16 Belarus 82.0 115.7 221’688 40.7 241’805 114.3 82.1 82.8 81.8 Belgium 17 40.1 32’703 48.9 31.0 50.7 32.0 6.0 35’970 25.6 21.0 6.7 Belize 18 3 1.8 81.7 5.4 85.6 2’508 19 3’002 4.8 5.1 3.5 Benin 1.8 6 4 5 2 6 87.1 81.6 12’933 24.6 21.9 24.0 2.8 31.7 11’220 Bhutan 3.1 20 8.0 96.3 21 92.2 17’483 Bolivia 19’673 27.5 33.1 14.3 23.8 8.1 Bosnia and 22 22.2 56’331 90.2 45.0 91.3 47.1 43’003 50.0 20.2 53.6 Herzegovina 7 167.3 Botswana 169.0 16’437 7.8 11’379 14.8 16.0 14.5 19.6 23 8.3 54.5 43’634 21.4 53.5 43’553 49.6 126.6 50.5 139.0 24 Brazil 21.8 Brunei 5 106.8 52’914 108.1 25 11.4 63’090 92.0 93.4 79.2 81.7 9.0 Darussalam 6 7 6 1 25.3 26 Bulgaria 57.9 59.0 135’113 56.7 129.3 59.1 132.4 23.3 145’170 2’862 0.7 80.6 2’860 Burkina Faso 0.4 4.6 5.2 8.3 12.5 27 71.7 6’913 5’702 1.0 30.5 1.2 0.2 1.0 46.2 4.0 Burundi 0.2 28 8 1 Cambodia 133.0 10’484 2.3 17’792 10.6 16.0 1.6 15.0 21.0 132.7 29 9 7 6.7 992 11.8 1’219 12.7 71.8 4.5 75.7 8.6 Cameroon 30 4.6 8 85.1 44.3 129’244 Canada 135’496 84.3 81.9 84.9 86.6 31 46.2 81.0 27.0 17’149 121.8 32.2 24.8 34.2 127.2 11.5 12’330 11.6 Cape Verde 32 0.1 39.8 3.5 40.2 733 33 2’575 2.9 Chad 2.7 3.1 0.2 86’548 133.2 129’825 59.7 60.3 129.5 63.6 19.2 57.4 19.3 34 Chile 46.7 93.2 5’141 17.9 6’530 92.3 49.6 47.4 54.2 China 35 16.5 9 10 105’050 113.1 44.5 115.7 45.5 14.4 38.0 104’991 41.8 Colombia 36 14.7 Congo (Dem. 2.0 2.3 1.9 53.0 384 37 369 2.4 53.5 0.0 0.0 Rep.) 11 7 2 8 Costa Rica 17.8 38 60.2 52.3 150.7 55.1 61’746 53.2 17.2 50’359 142.2 1.3 106.2 39 119.3 5’163 Côte d'Ivoire 5’194 7.2 8.8 12.2 17.2 1.2 10 58’034 72’381 104.4 76.7 70.1 103.8 76.8 34.7 68.4 36.7 Croatia 40 22.5 4.1 29.7 519 Cuba 572 12.9 13.0 41 5.6 11.2 11.5 96.3 89’791 27.8 70.9 95.4 71.5 71.2 68.6 75’055 28.4 Cyprus 42 Czech 12 9 43 129.5 129.2 79.0 18.6 110’965 18.1 119’841 78.5 78.9 78.0 Republic Denmark 44 33.2 29.9 92.7 128.3 92.3 93.1 293’498 127.0 91.7 328’018 2.5 2.6 32.4 45 34.7 8’955 Djibouti 10’255 18.0 19.1 7.1 8.1 Measuring the Information Society Report 2016 240

257 Annex 3 Percentage of Fixed-telephone Percentage of Mobile-cellular International Internet subscriptions per households households subscriptions bandwidth Economy with computer with Internet 100 inhabitants per 100 inhabitants Bit/s per Internet user 2014 2015 2014 2015 2014 2015 2015 2015 2014 2014 106.3 46.8 193’358 102.1 120’204 58.4 20.8 46 50.0 Dominica 22.7 48.5 Dominican 8 13 11 82.6 12.3 47 36’155 26.2 30.1 21.1 23.6 78.9 11.6 24’903 Rep. 14 12 10 15.3 48 Ecuador 34’796 32.4 79.4 32.8 103.9 15.5 37.5 40.8 56’561 111.0 8’700 114.3 11’318 47.3 7.4 39.2 41.8 7.6 Egypt 49 50.8 61’959 25.2 28.1 145.3 13.9 144.0 15.0 14.7 60’342 50 El Salvador 14.9 Equatorial 8.9 1’320 18.0 1.4 8.5 1’499 51 1.9 66.7 66.4 19.3 Guinea 9 11 31.7 52 Estonia 87.9 82.9 82.5 87.7 30’924 28’665 148.7 30.3 160.7 12 9.8 42.8 1’959 3.0 3.5 6.9 0.9 53 Ethiopia 0.8 1’884 31.6 10 8.1 36.7 108.2 39.2 31.3 29.0 19’769 98.8 27’399 8.5 Fiji 54 11 13 89.9 11.7 189’910 9.8 208’526 89.0 89.3 89.8 Finland 139.7 55 135.5 59.9 102.6 81.6 129’973 81.5 119’203 83.0 101.2 82.6 60.0 France 56 57 168.9 9’642 1.1 8’505 12.5 13.7 13.3 18.0 1.1 171.4 Gabon 13.3 8.3 8.9 10’305 12.6 131.3 13’342 119.6 2.3 Gambia 58 2.9 41.0 124.9 22.1 101’468 45.8 49.5 98’432 44.8 59 Georgia 25.4 129.0 13 15 12 14 117’540 90.6 116.7 91.0 120.4 89.5 54.9 90.3 103’877 56.9 Germany 60 1.0 129.7 3’602 Ghana 2’841 39.9 43.5 29.0 34.1 61 114.8 1.0 16 15 100’861 114.0 68.6 110.3 65.6 46.5 68.1 78’189 62.7 62 Greece 46.9 14 13 112.3 182’308 25.3 191’597 110.2 48.1 39.8 42.8 63 25.7 Grenada 45.0 15 14 17.4 24’676 27’471 22.2 111.5 15.0 106.6 20.9 10.6 10.8 64 Guatemala 72.1 2’365 87.2 65 Guinea 930 2.3 2.6 1.5 3.7 0.0 0.0 16 2’923 2.5 0.0 2.7 2’674 1.9 69.3 2.1 63.5 0.0 Guinea-Bissau 66 70.5 67 67.2 9’994 Guyana 25’607 26.9 29.1 24.2 26.1 19.9 19.1 22.8 23’617 93.5 21.6 19.6 23.0 95.5 5.9 21’765 6.4 Honduras 68 Hong Kong, 59.2 233.6 3’487’142 228.8 69 60.9 4’155’651 81.3 80.4 78.7 79.0 China 70 Hungary 30.3 74.0 75.0 75.1 37’027 75.6 118.9 31.2 118.1 55’410 49.9 458’708 111.1 725’806 98.1 98.5 96.5 96.5 Iceland 71 51.5 114.0 15 16 17 17 13.0 4’982 14.1 74.5 17.0 5’725 20.0 2.0 78.8 2.1 India 72 16 128.8 73 6’225 8.8 6’584 17.3 18.7 28.7 38.4 Indonesia 10.4 132.3 1 1 87.8 52.5 6’056 38.3 93.4 44.7 8’502 52.2 53.4 39.1 Iran (I.R.) 74 17 18 105.1 Ireland 103.7 155’337 43.2 155’521 83.6 83.5 82.2 84.9 40.9 75 17 18 19 81.3 89’638 83.5 90’321 72.1 133.5 76.0 121.5 43.1 43.8 76 Israel 19 20 77 154.3 76’640 33.7 77’322 71.8 72.5 72.6 75.4 151.3 33.1 Italy 13’261 9.0 111.5 32.3 34.3 14’244 26.4 107.4 30.3 9.1 78 Jamaica 18 20 21 18 96.4 120.2 50.1 62’618 79.3 80.0 49’150 96.5 Japan 50.2 79 125.1 2 27’524 4.8 47.0 147.8 47.0 18’285 69.0 75.9 179.4 5.0 Jordan 80 24.7 172.2 187.2 81 42’821 Kazakhstan 69’615 70.0 73.8 82.0 82.2 26.2 22 0.2 80.7 12.3 40’067 13.1 25’200 16.9 73.8 19.6 Kenya 82 0.4 Kiribati 38.8 11’781 83 2’916 6.1 6.7 5.6 6.3 1.9 1.4 28.9 98.8 78.3 77.1 43’358 98.5 118.5 46’764 115.7 58.1 84 59.5 Korea (Rep.) 218.4 Kuwait 50’096 13.4 48’619 87.8 89.0 75.4 80.5 14.2 85 231.8 6’219 19.5 7.1 13.8 7’357 16.5 132.8 134.5 17.6 86 Kyrgyzstan 7.9 23 19 3 19 5.2 53.1 2’848 13.4 16’795 67.0 10.5 11.4 13.7 11.4 87 Lao P.D.R. 24 21 76.0 116.8 111’881 73.5 127.0 76.1 19.5 73.4 93’683 19.6 Latvia 88 Lebanon 89 87.1 24’551 88.3 27’275 80.7 81.0 68.4 69.0 19.4 19.2 3’862 105.5 6.9 2.1 7.5 4’321 6.5 102.0 11.5 Lesotho 90 2.4 0.2 2.2 0.2 81.1 6’306 Liberia 7’522 91 2.4 2.5 2.7 73.4 125’454 139.5 66.8 67.6 141.9 66.0 18.7 68.3 158’030 Lithuania 19.5 92 20 25 Luxembourg 149.5 93 148.5 6’887’708 50.5 7’186’378 94.8 95.3 95.6 96.8 51.0 Measuring the Information Society Report 2016 241

258 Fixed-telephone Percentage of Percentage of Mobile-cellular International Internet households households subscriptions per subscriptions bandwidth Economy with Internet 100 inhabitants with computer per 100 inhabitants Bit/s per Internet user 2015 2014 2015 2014 2015 2014 2015 2015 2014 2014 26 21 81.1 322.6 88’921 25.0 84.3 111’931 86.3 324.4 26.7 94 Macao, China 79.0 95 46.0 8’026 Madagascar 12’420 4.5 5.3 4.7 5.8 1.1 1.0 41.2 22 5.2 33.5 4’237 5.8 6.2 2’429 9.1 35.3 0.3 0.4 Malawi 96 97 143.9 29’932 Malaysia 34’119 63.3 67.6 64.1 70.1 14.6 14.3 148.8 4 88’008 206.7 65.9 189.4 68.5 69’077 44.5 6.1 49.6 6.4 Maldives 98 22 99 1’879 139.6 1’279 2.7 3.3 6.7 8.2 1.0 1.0 149.1 Mali 1’220’570 53.4 80.7 127.0 81.1 1’178’759 80.7 81.9 129.3 53.6 Malta 100 20 23 6.2 89.3 101 1’454 Mauritania 1’451 4.4 4.7 1.3 15.6 1.3 94.2 24 33’896 53.1 30.3 57.0 30’513 51.9 140.6 60.0 132.2 29.8 Mauritius 102 25 103 85.3 Mexico 20’926 15.5 20’855 38.3 44.9 34.4 39.2 15.9 84.7 108.0 194’898 152’362 42.2 108.0 46.4 35.0 47.5 49.0 35.2 104 Moldova 2 133.0 88.8 56’862 128.1 64’287 73.5 74.9 74.7 76.6 105 Monaco 88.5 159’595 32.1 121’819 42.6 105.0 20.8 105.1 24.5 8.7 7.9 Mongolia 106 26.5 162.2 77’016 Montenegro 102’166 54.7 56.4 56.6 61.1 24.8 107 163.0 21 10’768 126.9 52.3 131.7 54.8 6.5 50.2 18’316 66.5 Morocco 108 7.4 6.1 69.8 9’157 Mozambique 6’145 5.6 74.2 10.6 13.2 0.3 0.3 109 1.0 3’676 6.6 76.7 14.0 54.0 7.0 5’226 15.0 1.0 Myanmar 110 7.6 113.8 111 102.1 34’531 Namibia 22’546 16.5 17.7 17.3 24.5 7.8 23 27 22 26 96.7 3.0 81.9 8.9 3’109 5.6 8.2 6.3 2’700 3.0 112 Nepal 24 23 96.2 116.4 229’961 41.3 242’326 95.7 123.5 95.8 96.0 Netherlands 41.3 113 24 82.8 108’506 112.1 79.8 121.8 82.3 40.2 79.8 95’081 40.6 New Zealand 114 27 28 25 25 5.7 116.1 115 5.5 114.6 21’090 11.1 11.8 11.6 14.0 Nicaragua 23’025 2.2 2’490 46.5 2.4 44.4 2.7 0.6 2’688 2.6 0.6 Niger 116 26 77.8 0.1 117 Nigeria 82.2 3’150 0.1 2’986 9.1 9.8 8.5 11.4 29 28 20.0 113.6 95.4 220’937 96.5 203’935 93.1 116.1 96.6 118 Norway 21.2 Oman 9.6 159.9 33’724 157.8 59’784 84.0 87.5 80.7 84.0 119 10.5 27 30 26 29 5 3 1 11’907 66.9 1.6 19.0 73.3 21.0 5’684 24.0 15.9 Pakistan 120 2.6 72.1 63.1 77.6 14’700 121 13’399 Palestine* 66.7 48.3 52.4 9.1 8.9 28 3 31 27 2 30 75’906 38.2 174.2 39.6 158.1 41.6 15.6 52.7 72’678 Panama 122 15.0 3 4 31.9 13’935 Paraguay 17’922 5.5 5.4 34.1 24.6 105.4 27.4 123 105.6 28 36’381 109.9 30.6 103.6 32.4 9.3 23.5 23.2 43’154 9.9 124 Peru Philippines 3.1 118.1 27’688 111.2 37’409 24.3 27.0 26.9 28.3 125 3.0 29 29 86’573 11.1 148.7 77.7 77.9 80’535 74.8 148.9 75.8 Poland 126 12.6 30 31 112.1 127 110.4 Portugal 202’825 43.2 44.1 69.4 71.1 64.9 70.2 232’080 30 31 145.8 71’566 18.2 88.0 153.6 88.3 95.8 95.8 67’473 Qatar 128 18.4 3 32 32 6 Romania 107.1 19.8 129 21.1 146’012 105.9 63.8 68.7 60.5 67.7 117’320 Russian 26.8 130 26’377 71.0 25.7 72.5 160.0 69.9 26’845 72.1 155.1 Federation 32 64.0 0.1 70.5 8’946 131 5’661 3.4 4.0 3.8 6.7 Rwanda 0.4 33 33 7 7’842 58.5 55.5 21.1 22.6 6’676 21.9 5.6 25.5 Samoa 132 6.1 4 5 179.6 12.5 69’556 12.3 88’669 80.0 Saudi Arabia 67.0 94.0 133 94.0 176.6 8’349 11.6 99.9 12.9 98.8 12.6 15.7 2.0 6’931 134 Senegal 2.1 4 36.5 122.1 Serbia 120.5 17’475 135 20’478 63.3 64.4 62.8 63.8 37.3 8 28’945 158.1 162.2 61.8 22.8 66.8 38’395 55.0 59.4 Seychelles 22.7 136 33 34 146.9 89.5 146.1 137 677’114 Singapore 737’006 85.7 87.5 87.2 36.2 36.0 15.9 17’240 78.8 80.5 122.3 85.2 116.9 78.4 14’901 16.8 138 Slovakia 37.1 36.2 112.1 77.6 113.2 121’137 139 154’627 77.1 77.8 76.8 Slovenia Solomon 1.3 140 72.7 6.1 4’277 6.7 4’277 5.6 65.8 6.3 1.3 Islands 6.9 7.7 149.2 149’542 159.3 141 South Africa 147’630 20.8 23.4 44.2 50.6 34 31 27 10.0 23.9 11.7 24.5 9.6 0.0 11.2 28 0.0 142 South Sudan Measuring the Information Society Report 2016 242

259 Annex 3 Percentage of Fixed-telephone Percentage of Mobile-cellular International Internet subscriptions per households households subscriptions bandwidth Economy with computer 100 inhabitants with Internet per 100 inhabitants Bit/s per Internet user 2015 2014 2015 2014 2015 2015 2014 2014 2015 2014 34 92’025 107.9 105’006 74.0 107.9 74.4 78.7 40.6 41.2 Spain 143 75.9 13’886 22.4 24.2 112.8 15.3 103.2 18.1 12.0 12’651 144 Sri Lanka 12.6 St. Kitts and 32 35 131.8 134’205 70.3 131’203 145 70.8 61.0 70.5 35.7 118.6 35.4 Nevis 36 33 17.9 146 St. Lucia 41.1 130’720 36.6 18.9 39.7 134’277 101.5 38.4 102.6 St. Vincent 58.2 103.7 176’691 105.2 194’268 22.7 61.8 44.3 49.4 147 21.9 and the Grenadines 35 148 Sudan 1.1 70.5 32.2 72.2 33.5 0.3 16.6 2’189 2’499 17.9 6 5 50’458 149 51’164 Suriname 15.6 45.2 36.1 15.5 44.2 170.6 42.8 180.7 2’053 3.3 73.2 17.0 19.8 1’717 18.4 72.3 22.3 3.5 Swaziland 150 36.7 392’780 39.2 421’237 90.1 88.3 89.6 91.0 151 127.8 Sweden 130.4 35 37 34 36 87.6 136.7 266’480 50.3 82.6 275’957 84.7 142.0 88.4 53.6 Switzerland 152 Syria 62.4 3’724 16.5 3’146 47.6 49.9 40.9 42.3 18.3 153 63.9 4.5 4’107 4’054 4.0 75.9 4.1 62.8 3.8 0.3 Tanzania 154 0.3 TFYR 105.5 155 105.4 41’812 18.2 53’890 67.6 68.4 68.3 69.4 17.7 Macedonia 6 4 Thailand 8.5 156 125.8 29.5 34.7 54’788 52.2 64’907 7.9 144.4 33.9 119.4 18.9 117.4 3’072 157 2’546 17.1 Timor-Leste 18.2 21.7 0.3 0.2 64.9 3.2 64.6 8’790 3.4 3.3 0.7 6.2 7’310 0.8 Togo 158 37 38 36 35 9 11.3 65.6 159 11’817 12.4 Tonga 64.3 34.0 37.1 35.7 39.5 14’623 Trinidad & 160 21.5 122’703 64.0 67.9 157.7 58.0 147.3 65.0 20.1 48’903 Tobago Tunisia 129.9 25’972 128.5 33’812 33.2 38.7 29.5 36.1 8.5 161 8.4 59’034 55.6 42’911 60.2 96.0 69.5 94.8 15.0 56.0 16.5 Turkey 162 39 37 10 38 36 0.8 4’002 Uganda 4’633 0.8 5.8 6.7 6.2 7.2 163 50.4 52.4 7 7 38’208 144.0 56.1 144.1 59.2 21.6 47.7 45’743 51.1 Ukraine 24.6 164 United Arab 37 5 11 8 79’641 165 23.1 178.1 87.9 89.3 90.1 22.3 95.4 187.3 107’904 Emirates United 52.4 166 374’554 125.8 89.9 52.6 89.9 123.6 91.3 89.0 361’057 Kingdom 38 39 9 8 40 38 87.3 117.6 39.8 99’017 85.1 110.2 84’931 79.9 167 82.2 37.5 United States 39 41 12 6 60’676 160.2 67.4 160.8 68.0 32.3 57.4 59.7 73’151 31.7 168 Uruguay Uzbekistan 169 73.3 1’581 73.8 2’075 36.9 43.2 44.7 52.6 8.6 8.4 8’477 1.8 66.2 22.0 24.8 8’237 28.8 60.4 34.5 Vanuatu 2.2 170 40 40 43.7 99.0 171 14’398 24.9 16’310 93.0 46.9 34.2 34.7 Venezuela 25.3 41 41 20’749 24’374 147.1 20.5 6.3 22.0 130.6 18.6 24.1 Viet Nam 6.0 172 2’496 4.7 68.0 173 68.5 4.7 6.1 6.5 5.1 5.5 Yemen 2’487 3’434 10.1 74.5 6.6 67.3 7.4 0.7 3’187 12.7 Zambia 0.8 174 2.2 6’380 175 80.8 Zimbabwe 11.8 84.8 18.0 2.3 18.1 4’806 10.7 Note: Data in italics are ITU estimates. *Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU World Telecommunication/ICT Indicators database. Measuring the Information Society Report 2016 243

260 Use indicators Fixed-broadband Active mobile Percentage of individuals subscriptions broadband subscriptions using the Internet Economy per 100 inhabitants per 100 inhabitants 2015 2014 2014 2015 2014 2015 7.0 3.2 8.3 6.0 1 0.0 Afghanistan 0.0 63.3 2 7.6 30.9 40.6 Albania 60.1 6.5 1 25.0 3 Algeria 40.1 4.0 38.2 20.8 5.6 96.9 35.9 37.9 36.6 42.1 95.9 Andorra 4 1 Angola 10.2 5 19.3 0.7 12.4 0.4 16.4 65.2 33.0 33.8 64.0 6 13.1 11.8 Antigua & Barbuda 1 2 7 Argentina 64.7 67.3 53.6 16.1 69.4 15.6 2 9.6 58.2 9.1 8 34.2 41.3 Armenia 54.6 1 2 1 1 3 27.7 84.6 112.2 112.9 27.9 9 Australia 84.0 2 3 83.9 27.7 28.6 67.2 68.6 Austria 81.0 10 4 3 19.8 77.0 60.9 61.5 19.9 75.0 11 Azerbaijan 3 76.9 78.0 20.9 14.6 21.1 12 Bahamas 20.2 5 4 21.4 93.5 126.2 131.8 18.6 90.5 Bahrain 13 2 4 2.4 14 13.4 13.5 13.9 14.4 2.0 Bangladesh 27.2 76.1 43.7 75.2 54.9 Barbados 15 27.2 5 6 16 28.8 31.3 55.0 61.8 Belarus 62.2 59.0 7 6 36.8 36.0 57.8 85.1 66.6 85.0 Belgium 17 Belize 38.7 41.6 2.9 2.9 10.2 30.2 18 Benin 6.0 19 2.8 0.7 0.4 4.2 6.8 3 5 20 3.6 Bhutan 28.2 56.4 30.3 39.8 3.3 6 2 8 1.6 12.2 34.6 33.8 Bolivia 45.1 1.6 21 60.8 65.1 14.2 16.6 27.8 33.5 22 Bosnia and Herzegovina Botswana 18.5 23 49.7 1.8 67.3 1.6 27.5 9 4 4 3 59.1 11.7 24 12.2 Brazil 54.6 78.2 88.6 4 8.0 7.1 3.8 71.2 4.5 25 Brunei Darussalam 68.8 7 5 5 7 10 Bulgaria 22.4 55.5 66.4 81.3 20.7 26 56.7 0.0 11.4 9.4 15.4 0.0 Burkina Faso 27 9.6 1.4 4.9 0.0 0.0 0.5 7.6 28 Burundi 29 Cambodia 14.0 31.1 42.8 0.4 19.0 0.5 11 16.2 Cameroon 20.7 0.1 0.1 0.3 4.3 30 35.4 88.5 54.3 56.3 36.4 31 Canada 87.1 Cape Verde 32 3.4 3.0 51.3 72.9 40.3 43.0 6 5 6 Chad 33 2.5 0.1 0.1 0.5 1.4 2.7 6 61.1 64.3 14.0 15.2 50.1 34 57.6 Chile 12 8 7 14.4 41.8 50.3 56.0 18.6 China 47.9 35 13 9 7 10.3 Colombia 36 11.2 36.0 55.9 41.0 52.6 0.0 0.0 7.9 3.8 8.5 37 3.0 Congo (Dem. Rep.) 10 38 10.4 Costa Rica 86.9 95.5 53.0 59.8 11.2 0.5 0.6 24.6 21.0 40.4 14.6 Côte d'Ivoire 39 11 14 68.6 23.0 23.2 68.5 75.4 40 Croatia 69.8 15 31.1 0.1 0.0 0.0 0.1 29.1 Cuba 41 16 12 22.4 71.7 42 21.1 Cyprus 42.1 54.8 69.3 8 17 13 8 7 27.9 81.3 27.9 68.8 66.7 79.7 43 Czech Republic 14 9 18 8 42.5 96.3 96.0 44 109.7 116.8 Denmark 41.5 2.3 11.9 3.2 5.6 2.3 10.7 Djibouti 45 Dominica 57.5 67.6 14.9 20.9 29.3 42.2 46 15 47 Dominican Rep. 49.6 6.4 39.6 51.9 5.7 30.1 19 16 10 48.9 48 8.3 Ecuador 9.2 45.6 35.1 30.9 20 35.9 43.5 3.7 50.7 4.5 33.9 49 Egypt Measuring the Information Society Report 2016 244

261 Annex 3 Fixed-broadband Active mobile Percentage of individuals broadband subscriptions subscriptions using the Internet Economy per 100 inhabitants per 100 inhabitants 2015 2014 2015 2014 2015 2014 21 El Salvador 5.5 18.4 19.9 50 26.9 24.8 5.0 9 0.5 0.0 21.3 0.0 0.5 51 Equatorial Guinea 18.9 22 17 28.7 Estonia 114.3 88.4 84.2 28.9 52 117.0 10 9 0.7 0.5 11.9 11.6 7.5 Ethiopia 53 7.7 1.4 42.3 46.3 54 Fiji 41.8 1.4 48.2 18 23 92.4 Finland 55 144.1 92.7 32.3 31.7 138.5 19 24 83.8 40.2 41.3 66.3 74.7 56 France 84.7 15.8 0.6 33.1 0.6 23.5 20.0 57 Gabon 17.1 0.1 0.2 8.0 10.0 58 Gambia 16.5 25 20 11 59 44.0 Georgia 50.4 14.6 13.9 35.0 45.2 26 21 87.6 35.8 37.2 63.6 75.1 86.2 60 Germany 0.3 0.3 59.8 23.5 66.8 Ghana 18.9 61 27 22 63.2 28.4 30.7 41.0 45.6 62 66.8 Greece 8 2.6 53.8 18.5 17.7 28.8 51.6 63 Grenada 9 12 2.8 9.4 64 10.1 Guatemala 23.4 27.1 2.7 0.0 10.8 1.7 13.9 0.0 65 4.7 Guinea 11 10 3.5 0.1 0.1 0.0 0.0 3.3 66 Guinea-Bissau 5.6 6.6 0.2 38.2 0.2 Guyana 67 37.4 10 20.4 2.3 16.3 17.2 68 19.1 Honduras 1.9 28 23 31.9 104.5 84.9 107.0 31.4 79.9 Hong Kong, China 69 29 24 70 Hungary 26.0 76.1 34.0 39.8 72.8 27.4 30 35.9 85.3 98.2 93.4 37.0 Iceland 98.2 71 12 11 11 13 5.5 1.2 26.0 21.0 9.4 1.3 India 72 31 25 1.2 22.0 34.7 17.1 42.0 Indonesia 1.1 73 12 32 39.4 74 10.7 Iran (I.R.) 20.0 44.1 9.5 10.9 33 13 14 26 81.0 80.1 95.0 27.7 26.9 79.7 Ireland 75 12 34 13 14 75.0 52.2 27.4 56.1 27.2 78.9 76 Israel 35 27 23.8 70.6 65.6 82.1 23.5 62.0 77 Italy 36 5.4 5.8 38.8 53.5 78 40.4 43.2 Jamaica 37 13 15 15 29.8 121.4 93.3 126.4 30.5 89.1 79 Japan 38 35.6 53.4 4.7 4.2 19.1 80 Jordan 46.2 28 59.4 13.0 60.0 12.9 72.9 Kazakhstan 81 66.0 45.6 15.5 0.3 9.1 43.4 82 Kenya 0.2 12.3 Kiribati 83 0.3 0.5 13.0 0.1 0.1 29 39 89.9 87.9 40.2 108.6 109.7 84 Korea (Rep.) 38.8 1.4 139.3 139.8 1.4 82.1 78.7 85 Kuwait 16 30.2 3.0 3.7 26.7 31.0 86 Kyrgyzstan 28.3 16 14 17 18.2 0.5 14.2 6.5 0.2 14.3 Lao P.D.R. 87 40 30 Latvia 24.7 25.1 61.2 67.0 88 75.8 79.2 15 22.8 53.4 22.8 74.0 53.5 73.0 Lebanon 89 90 0.1 0.1 25.5 37.7 Lesotho 11.0 16.1 91 5.4 Liberia 20.5 0.2 5.9 0.1 7.6 41 31 16 26.7 74.2 27.8 70.2 72.1 Lithuania 92 71.4 32 17 14 42 36.5 83.3 97.3 34.8 88.9 Luxembourg 94.7 93 43 33 18 17 94 28.1 29.1 322.2 Macao, China 324.4 77.6 69.8 95 0.1 4.2 6.1 3.7 9.0 0.1 Madagascar Malawi 9.3 0.1 0.0 10.9 16.6 5.8 96 18 44 15 34 63.7 Malaysia 97 10.1 71.1 58.3 89.9 9.0 19 19 54.5 5.6 6.5 48.9 98 63.6 Maldives 49.3 20 20 10.3 0.0 7.0 11.3 Mali 18.8 0.0 99 Measuring the Information Society Report 2016 245

262 Active mobile Fixed-broadband Percentage of individuals subscriptions broadband subscriptions using the Internet Economy per 100 inhabitants per 100 inhabitants 2014 2015 2014 2015 2014 2015 45 35 21 37.8 56.6 76.2 35.2 73.2 Malta 100 63.2 0.2 14.4 15.2 23.1 10.7 101 Mauritania 0.2 46 22 36 16 14.6 15.7 37.0 44.8 Mauritius 50.1 31.7 102 37 17 23 47 103 41.5 Mexico 50.4 44.4 57.4 11.6 10.2 46.6 49.8 14.7 15.5 49.4 51.9 104 Moldova 24 21 92.4 105 Monaco 65.2 93.4 46.8 47.5 63.2 48 38 106 6.8 7.1 57.6 Mongolia 19.9 76.0 21.4 18.1 31.0 64.6 43.7 16.7 Montenegro 107 61.0 49 39 57.1 Morocco 3.0 3.4 26.8 39.3 56.8 108 9.0 0.1 9.4 0.1 3.0 5.9 Mozambique 109 11.5 21.8 0.3 0.3 14.9 29.5 110 Myanmar Namibia 14.8 111 34.2 22.3 1.8 1.7 62.1 18 22 15.4 1.1 17.4 112 21.1 17.6 Nepal 0.9 50 40 23 19 41.7 40.8 70.5 93.1 69.2 93.2 113 Netherlands 88.2 31.0 31.5 92.7 114 New Zealand 85.5 114.2 20 18 24 17.6 Nicaragua 115 1.9 7.2 19.7 1.4 1.8 Niger 2.2 0.0 0.1 0.9 1.8 116 2.0 117 42.7 Nigeria 0.0 21.0 47.4 11.7 0.0 51 41 19 25 25 38.8 38.9 96.3 88.8 118 92.8 96.8 Norway Oman 119 73.7 4.5 78.3 74.2 70.2 5.6 42 21 20 26 26 1.0 Pakistan 5.1 1.1 18.0 13.8 120 13.0 52 53.7 6.0 57.4 0.0 121 0.0 Palestine* 5.3 27 53 43 22 21 27 51.2 7.9 Panama 29.5 44.9 32.7 7.9 122 54 22 28 28 2.7 43.0 3.1 44.4 31.0 123 39.2 Paraguay 55 44 29 124 Peru 5.7 6.4 28.5 40.9 36.7 40.2 40.7 28.0 3.4 41.6 2.9 Philippines 125 39.7 45 56 66.6 68.0 18.9 19.5 55.7 60.2 126 Poland 57 46 29.6 44.8 68.6 52.0 25.7 64.6 Portugal 127 47 92.9 128 9.9 10.1 73.0 80.0 Qatar 91.5 29 48 23 23 58 30 18.6 49.3 55.8 63.5 19.8 54.1 129 Romania 59 49 130 73.4 Russian Federation 70.5 18.8 65.8 71.3 17.5 0.0 11.1 18.0 25.9 0.2 10.6 131 Rwanda 24 30 Samoa 25.4 1.1 1.1 6.0 9.6 21.2 132 60 50 10.3 69.6 99.0 64.7 111.7 Saudi Arabia 133 12.0 61 17.7 134 21.7 0.7 0.7 23.7 Senegal 26.4 62 51 16.8 15.6 66.4 65.3 71.8 Serbia 62.1 135 136 Seychelles 54.3 58.1 12.7 14.3 12.7 19.1 63 31 31 25 137 79.0 Singapore 141.7 26.7 142.2 26.5 82.1 64 59.5 85.0 21.8 23.3 138 67.5 Slovakia 80.0 65 52 73.1 27.6 52.0 46.7 26.8 71.6 139 Slovenia 32 9.0 10.0 0.2 0.2 13.0 11.4 140 Solomon Islands 59.5 5.3 South Africa 46.7 141 49.0 51.9 3.2 142 South Sudan 15.9 17.9 0.0 0.0 1.3 1.4 26 66 53 Spain 143 76.2 27.6 82.1 77.3 78.7 28.3 Sri Lanka 30.0 2.6 3.1 13.0 15.8 144 25.8 68.0 St. Kitts and Nevis 145 25.6 75.7 18.6 71.0 29.6 St. Lucia 50.0 52.4 15.3 15.4 27.4 33.6 146 147 St. Vincent and the Grenadines 47.4 15.5 39.0 51.8 13.5 34.4 Sudan 24.6 26.6 0.1 0.1 27.2 29.4 148 149 Suriname 40.1 9.5 71.6 8.5 75.8 42.8 Measuring the Information Society Report 2016 246

263 Annex 3 Fixed-broadband Active mobile Percentage of individuals broadband subscriptions subscriptions using the Internet Economy per 100 inhabitants per 100 inhabitants 2015 2014 2015 2015 2014 2014 30.4 0.4 8.0 17.0 27.1 Swaziland 150 0.5 67 54 92.5 Sweden 151 116.3 36.1 122.1 34.1 90.6 68 32 55 33 24 27 44.8 86.8 88.0 97.6 87.4 152 Switzerland 42.5 2.3 28.1 8.5 10.4 3.1 30.0 153 Syria 5.4 0.2 0.2 3.0 3.2 4.9 Tanzania 154 69 56 68.1 TFYR Macedonia 155 17.2 56.2 70.4 49.5 16.8 70 57 34.9 8.1 9.2 79.9 75.3 Thailand 156 39.3 0.1 37.5 0.1 13.4 31.2 11.3 Timor-Leste 157 158 0.2 0.9 3.5 6.0 Togo 5.7 7.1 34 33 28 Tonga 40.0 159 45.0 29.5 1.9 19.3 1.7 160 65.1 20.7 28.3 32.2 Trinidad & Tobago 69.2 18.4 46.2 161 Tunisia 47.6 62.6 4.3 48.5 4.5 71 58 12.4 53.7 162 11.7 Turkey 42.7 50.9 51.0 35 34 29 25 0.3 14.7 19.2 18.3 0.3 Uganda 163 17.7 72 Ukraine 49.3 9.3 11.8 7.5 8.1 46.2 164 73 26 30 11.6 90.0 92.0 91.2 12.8 United Arab Emirates 165 90.4 74 59 35 36 37.4 37.7 88.8 United Kingdom 91.6 87.8 92.0 166 60 27 31 36 37 31.5 73.0 102.7 30.3 109.2 74.6 167 United States 32 61 28 75 37 38 26.3 Uruguay 60.6 24.6 77.7 64.6 168 61.5 Uzbekistan 2.8 25.0 35.5 28.7 42.8 169 3.6 29 33 38 1.8 170 1.6 Vanuatu 26.2 18.8 41.3 22.4 34 39 61.9 44.0 8.2 43.0 7.8 Venezuela 171 57.0 35 40 48.3 52.7 6.5 8.1 31.0 39.0 Viet Nam 172 1.4 22.6 25.1 4.8 Yemen 5.9 173 1.5 36 30 39 Zambia 0.1 21.0 8.8 0.1 13.8 174 19.0 76 1.0 16.4 39.2 39.0 1.1 Zimbabwe 175 16.3 Note: Data in italics are ITU estimates. *Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: ITU World Telecommunication/ICT Indicators database. Measuring the Information Society Report 2016 247

264 Skills indicators Gross enrolment ratio Mean years of schooling Economy Tertiary Secondary 2014 2015 2015 2015 2014 2014 54.3 Afghanistan 3.2 3.2 3.7 1 54.3 3.7 96.4 96.4 Albania 2 62.7 9.3 9.3 62.7 97.6 34.6 34.6 7.6 7.6 3 Algeria 97.6 130.8 Andorra 4 130.8 10.3 10.3 84.6 84.6 31.5 9.9 9.9 5 4.7 Angola 31.5 4.7 102.3 102.3 Antigua & Barbuda 6 23.5 9.2 23.5 9.2 Argentina 106.3 80.0 80.0 9.8 9.8 7 106.3 95.9 95.9 8 Armenia 10.9 10.9 46.6 46.6 137.6 9 86.6 13.0 13.0 Australia 137.6 86.6 99.3 99.3 10 Austria 80.0 10.8 10.8 80.0 102.8 23.2 23.2 10.6 10.6 Azerbaijan 102.8 11 92.9 12 Bahamas 92.9 57.1 57.1 10.9 10.9 Bahrain 13 95.5 36.8 36.8 9.4 9.4 95.5 14 58.3 58.3 Bangladesh 13.2 13.2 5.1 5.1 109.2 15 60.8 10.5 10.5 109.2 Barbados 60.8 107.0 107.0 Belarus 16 88.9 88.9 12.0 12.0 163.1 163.1 72.3 72.3 17 11.4 Belgium 11.4 Belize 80.2 18 80.2 24.2 10.5 10.5 24.2 54.4 19 15.4 15.4 3.3 3.3 Benin 54.4 Bhutan 84.2 84.2 20 10.9 3.0 3.0 10.9 Bolivia 84.7 37.7 37.7 8.2 8.2 21 84.7 89.3 Bosnia and Herzegovina 22 89.3 37.7 37.7 9.2 9.2 Botswana 83.9 83.9 27.5 27.5 8.9 23 8.9 24 105.8 105.8 Brazil 25.6 7.7 7.7 25.6 Brunei Darussalam 99.1 99.1 25 31.7 8.8 8.8 31.7 26 Bulgaria 100.9 100.9 70.8 70.8 10.6 10.6 27 Burkina Faso 30.3 4.8 4.8 1.4 1.4 30.3 Burundi 37.9 28 37.9 4.4 2.7 4.4 2.7 45.0 4.4 15.8 45.0 4.4 29 15.8 Cambodia 56.4 56.4 30 Cameroon 11.9 6.0 11.9 6.0 Canada 103.4 66.6 66.6 13.0 13.0 31 103.4 Cape Verde 92.6 32 92.6 23.0 23.0 4.7 4.7 Chad 22.8 22.8 3.4 3.4 1.9 1.9 33 34 Chile 100.5 100.5 83.8 83.8 9.8 9.8 96.2 30.2 30.2 7.5 7.5 China 96.2 35 93.0 36 Colombia 93.0 51.3 51.3 7.3 7.3 Congo (Dem. Rep.) 37 43.5 6.6 6.6 6.0 6.0 43.5 38 120.3 120.3 Costa Rica 53.0 8.7 8.7 53.0 Côte d'Ivoire 39 40.1 8.7 8.7 4.3 4.3 40.1 40 Croatia 98.4 98.4 61.6 11.0 11.0 61.6 Cuba 99.7 41.0 41.0 11.5 11.5 41 99.7 99.4 Cyprus 42 99.4 53.1 53.1 11.7 11.7 Czech Republic 104.4 104.4 65.4 65.4 12.3 43 12.3 44 129.8 129.8 Denmark 81.2 12.7 12.7 81.2 Djibouti 47.1 47.1 45 4.9 3.8 3.8 4.9 46 Dominica 96.7 96.7 34.2 34.2 7.9 7.9 47 Dominican Rep. 78.4 47.5 47.5 7.7 7.7 78.4 Ecuador 104.2 48 104.2 40.5 40.5 7.6 7.6 Egypt 86.0 86.0 30.3 30.3 6.6 6.6 49 50 El Salvador 78.1 78.1 29.2 6.5 6.5 29.2 51 Equatorial Guinea 28.2 28.2 3.3 3.3 5.5 5.5 52 108.6 108.6 Estonia 72.9 72.9 12.5 12.5 Measuring the Information Society Report 2016 248

265 Annex 3 Gross enrolment ratio Mean years of schooling Economy Tertiary Secondary 2014 2015 2015 2015 2014 2014 28.9 Ethiopia 2.4 2.4 53 6.3 28.9 6.3 88.3 Fiji 54 88.3 9.9 16.1 9.9 16.1 91.1 10.3 10.3 55 Finland 143.2 91.1 143.2 110.9 France 56 110.9 11.4 62.1 62.1 11.4 53.9 8.5 7.8 7.8 57 Gabon 53.9 8.5 57.5 57.5 Gambia 58 3.4 3.4 2.8 2.8 59 39.2 39.2 12.3 12.3 Georgia 99.4 99.4 102.5 60 Germany 102.5 61.1 13.5 61.1 13.5 15.6 7.0 7.0 71.0 61 Ghana 15.6 71.0 108.2 Greece 62 108.2 10.5 10.5 110.2 110.2 101.1 63 52.8 8.6 8.6 Grenada 101.1 52.8 63.5 63.5 Guatemala 64 18.3 18.3 7.0 7.0 38.8 38.8 10.8 10.8 2.4 2.4 65 Guinea 34.5 66 Guinea-Bissau 34.5 2.6 2.8 2.6 2.8 101.0 Guyana 8.5 8.5 12.9 67 101.0 12.9 68.4 68.4 Honduras 68 21.2 6.2 6.2 21.2 100.6 68.8 68.8 11.2 11.2 69 Hong Kong, China 100.6 108.2 Hungary 70 108.2 11.6 11.6 57.0 57.0 112.0 81.4 81.4 71 10.6 Iceland 112.0 10.6 68.9 68.9 India 72 23.9 5.4 23.9 5.4 Indonesia 82.5 31.3 31.3 7.6 7.6 73 82.5 88.4 88.4 74 Iran (I.R.) 8.2 8.2 66.0 66.0 126.5 75 73.2 12.2 12.2 Ireland 126.5 73.2 101.5 101.5 76 Israel 66.3 12.8 12.8 66.3 102.4 63.5 63.5 10.1 10.1 Italy 102.4 77 83.0 78 Jamaica 83.0 27.4 27.4 9.7 9.7 Japan 101.9 101.9 62.4 62.4 79 11.5 11.5 80 87.8 87.8 Jordan 46.6 46.6 9.9 9.9 109.1 109.1 46.0 11.4 11.4 81 Kazakhstan 46.0 67.0 Kenya 82 67.0 6.3 4.0 6.3 4.0 17.0 83 7.8 7.8 Kiribati 86.4 86.4 17.0 97.7 84 Korea (Rep.) 97.7 95.3 11.9 95.3 11.9 92.5 7.2 27.0 92.5 7.2 85 27.0 Kuwait 90.8 90.8 86 Kyrgyzstan 47.3 10.6 10.6 47.3 Lao P.D.R. 57.2 17.3 17.3 5.0 5.0 87 57.2 110.5 110.5 88 Latvia 11.5 11.5 67.0 67.0 68.2 89 42.8 7.9 7.9 Lebanon 68.2 42.8 52.2 52.2 90 Lesotho 9.8 5.9 5.9 9.8 37.9 11.6 11.6 4.1 4.1 Liberia 37.9 91 105.4 92 Lithuania 105.4 72.0 72.0 12.7 12.7 Luxembourg 102.4 102.4 19.7 19.7 11.7 11.7 93 Macao, China 94 96.1 96.1 69.4 69.4 7.5 7.5 38.4 95 4.2 6.0 6.0 38.4 Madagascar 4.2 39.5 39.5 Malawi 96 0.8 0.8 4.3 4.3 71.1 71.1 38.5 38.5 10.0 10.0 97 Malaysia 72.3 98 Maldives 72.3 13.2 5.8 5.8 13.2 43.5 43.5 7.5 99 2.0 2.0 Mali 7.5 Malta 85.5 85.5 100 45.1 45.1 11.3 11.3 101 29.9 29.9 5.5 5.5 3.8 3.8 Mauritania Mauritius 97.9 102 97.9 38.7 8.5 8.5 38.7 87.0 103 29.2 29.2 8.4 8.4 Mexico 87.0 Moldova 88.3 88.3 104 41.3 11.9 11.9 41.3 Monaco 109.7 54.9 54.9 11.4 11.4 105 109.7 Mongolia 90.7 106 90.7 64.3 64.3 9.3 9.3 Measuring the Information Society Report 2016 249

266 Gross enrolment ratio Mean years of schooling Economy Tertiary Secondary 2014 2015 2015 2015 2014 2014 90.3 Montenegro 11.2 11.2 55.5 107 90.3 55.5 68.9 68.9 Morocco 108 24.6 4.4 4.4 24.6 24.5 6.0 6.0 3.2 3.2 109 Mozambique 24.5 51.3 Myanmar 110 51.3 4.1 4.1 13.4 13.4 64.8 9.3 9.3 111 6.2 Namibia 64.8 6.2 67.2 67.2 Nepal 112 15.8 3.3 15.8 3.3 Netherlands 130.7 77.3 77.3 11.9 11.9 113 130.7 117.2 117.2 114 New Zealand 12.5 12.5 79.7 79.7 68.9 115 17.9 6.0 6.0 Nicaragua 68.9 17.9 18.8 18.8 116 Niger 1.8 1.5 1.5 1.8 43.8 10.4 10.4 5.9 5.9 Nigeria 43.8 117 113.0 118 Norway 113.0 76.1 76.1 12.8 12.8 Oman 93.5 93.5 28.1 119 8.0 8.0 28.1 120 41.6 41.6 Pakistan 10.4 10.4 4.7 4.7 82.2 121 44.0 8.9 8.9 82.2 Palestine* 44.0 75.5 75.5 Panama 122 38.7 38.7 9.3 9.3 69.6 69.6 34.5 34.5 8.1 8.1 123 Paraguay 95.6 124 Peru 95.6 42.6 9.0 9.0 42.6 88.4 88.4 35.8 125 8.9 8.9 Philippines 35.8 Poland 108.7 108.7 126 71.2 71.2 11.9 11.9 127 119.7 119.7 66.2 66.2 8.9 8.9 Portugal Qatar 111.6 128 111.6 15.8 9.8 9.8 15.8 97.9 129 52.2 52.2 10.6 10.6 Romania 97.9 Russian Federation 98.8 98.8 130 78.0 78.0 12.0 12.0 Rwanda 3.7 40.2 7.5 7.5 3.7 131 40.2 Samoa 132 86.9 86.9 7.5 7.5 10.3 10.3 133 Saudi Arabia 108.3 108.3 61.1 61.1 8.7 8.7 134 Senegal 41.0 41.0 7.6 7.6 2.5 2.5 94.3 135 58.1 10.8 10.8 94.3 Serbia 58.1 74.6 74.6 Seychelles 136 6.5 6.5 9.4 9.4 97.2 97.2 43.8 43.8 11.6 11.6 137 Singapore 91.8 138 Slovakia 91.8 54.4 12.1 12.1 54.4 110.9 110.9 85.2 139 12.1 12.1 Slovenia 85.2 Solomon Islands 48.4 48.4 140 16.2 16.2 5.0 5.0 141 98.2 98.2 19.7 19.7 10.3 10.3 South Africa South Sudan 40.7 142 40.7 17.2 5.4 5.4 17.2 131.1 143 87.1 87.1 9.8 9.8 Spain 131.1 Sri Lanka 99.7 99.7 144 20.7 20.7 10.8 10.8 St. Kitts and Nevis 8.4 91.5 79.1 79.1 8.4 145 91.5 St. Lucia 146 86.5 86.5 16.9 16.9 9.3 9.3 147 St. Vincent and the Grenadines 104.7 104.7 18.2 18.2 8.6 8.6 148 Sudan 40.7 40.7 16.9 16.9 3.1 3.1 78.5 12.1 12.1 7.7 7.7 Suriname 78.5 149 63.0 150 Swaziland 63.0 5.3 5.3 7.1 7.1 Sweden 128.5 128.5 151 63.4 12.3 12.3 63.4 152 96.3 96.3 Switzerland 56.3 13.8 13.8 56.3 Syria 50.5 50.5 153 34.5 6.3 6.3 34.5 154 Tanzania 32.3 32.3 3.6 3.6 5.1 5.1 155 TFYR Macedonia 82.8 39.4 39.4 9.3 9.3 82.8 Thailand 86.2 156 86.2 51.4 51.4 7.3 7.3 Timor-Leste 73.1 73.1 17.7 17.7 4.4 4.4 157 158 Togo 54.9 54.9 10.1 4.5 4.5 10.1 159 Tonga 90.6 90.6 6.3 6.3 10.7 10.7 160 85.5 85.5 Trinidad & Tobago 12.0 12.0 10.9 10.9 Measuring the Information Society Report 2016 250

267 Annex 3 Gross enrolment ratio Mean years of schooling Economy Tertiary Secondary 2014 2015 2015 2015 2014 2014 90.1 Tunisia 6.8 6.8 161 34.6 34.6 90.1 114.6 Turkey 162 114.6 7.9 79.0 79.0 7.9 27.6 27.6 9.1 9.1 5.4 5.4 Uganda 163 99.2 99.2 164 Ukraine 82.3 11.3 11.3 82.3 83.6 165 22.0 9.5 9.5 83.6 22.0 United Arab Emirates 124.4 124.4 166 United Kingdom 56.9 13.3 13.3 56.9 United States 95.9 95.9 88.8 88.8 13.6 13.6 167 Uruguay 90.3 90.3 168 63.2 63.2 8.6 8.6 Uzbekistan 105.2 8.9 8.9 10.9 10.9 169 105.2 59.5 Vanuatu 170 59.5 4.7 6.8 6.8 4.7 91.6 91.6 78.1 78.1 8.9 8.9 Venezuela 171 77.2 77.2 Viet Nam 172 30.5 30.5 7.5 7.5 Yemen 173 48.6 10.3 10.3 2.6 2.6 48.6 Zambia 45.5 174 45.5 2.4 2.4 6.6 6.6 Zimbabwe 47.2 47.2 175 5.9 7.3 7.3 5.9 Note: At the time of the calculation of the IDI 2016, UIS data for skills indicators were available for 2014 only. Therefore, 2014 data were used for both years. *Palestine is not an ITU Member State; the status of Palestine in ITU is the subject of Resolution 99 (Rev. Busan, 2014) of the ITU Plenipotentiary Conference. Source: Gross enrolment ratio refer to latest available data from UIS. Mean years of schooling data are from UNDP HDR and UIS. Measuring the Information Society Report 2016 251

268 Notes The notes are presented here as submitted by countries to ITU. Access indicators Fixed-telephone subscriptions per 100 inhabitants, 2014 1) Includes 272'960 WLL subscriptions. 2) Includes PSTN and other fixed-line telephone services. Due to a methodology change in 2014, data reported for 2014 differs from data reported in previous communications reports. In 2014, the total resale (retail services directly connected via another network) and retail services in operation are reported. In previous communications reports, wholesale and retail totals were reported. 3) Incl. ISDN channels measured in ISDN B channels equivalents. 4) Incl. VoIP. 5) Obtained from URCA's Licensees. 6) Bhutan Telecom is the only service provider in Bhutan. 7) December 2014. 8) Total retail access lines. 9) Source: Colombia TIC. 10) Counting voice channel equivalents, 1'500'563 is the number of subscriptions. 11) Incl. IP lines. 12) Incl. public payphones. 13) Including ISDN voice-channel equivalents. Data based on estimates. 14) Providers data. 15) New tax on numbering resources, which has prompted operators to return several numbers, either inactive ones or with low consumption. 16) December 2014. Excluding fixed wireless local loop (WLL) subscriptions, ISDN voice-channel equivalents. 17) Incl. PSTN lines, ISDN paths, FWA subscriptions, public payphones and VOIP subscriptions. 18) The number of fixed public payphones is as of March 2014.(This data is reported by carrier every March). 19) Data from 4 main operators, LTC, BEELINE, UNITEL, ETL. 20) Including digital lines. Without including separate ISDN channels (abonnements au téléphone fixe). 21) Excl. ISDN channels and fixed wireless subscriptions. 22) From ICT Indicators Survey by PPPC. 23) December 2014. Source: January 2015 Management Information System Report. 24) Based on ACM 2014Q4 data. 25) Estimate. 26) Refers to active Fixed Wired/Wireless lines. 27) Figures are as on 31st December, 2014 based on data received from Fixed Line Operators. 28) Preliminary. 29) POTS, ISDN BRA & ISDN PRA. 30) Operators' data. 31) Sudatel terminated its fixed landline, CDMA and GSM Services in South Sudan in Oct 2012. 32) Base on ECTEL's research. 33) Refers to March 2014. 34) Estimates. 35) Preliminary data. 36) December. 37) Telecommunications Regulatory Authority (TRA). 38) FCC trend-based estimate using recent historical data. Fixed-telephone subscriptions per 100 inhabitants, 2015 1) Includes 254132 WLL subscriptions. Source: ARPT/Algérie Télécom. 2) Preliminary. 3) Includes PSTN and other fixed- line telephone services. Due to a methodology change in 2014, data reported here differs from data reported in previous communications reports. In 2014, the total resale (retail services directly connected via another network) and retail services in operation are reported. In previous communications reports, wholesale and retail totals were reported. 4) Incl ISDN channels measured in ISDN B channel equivalents; corrected data for 2014: 3.518.900. 5) December 2015. 6) Bhutan Telecom is the only service provider for fixed-lines in Bhutan. 7) Preliminary. 8) Decrease due to the reduction of FWLL services which contributes to 78% of the entire fixed telephone subscription. 9) Sept. 10) Source: Sistema de Infromación Integral Colombia TIC. 11) Preliminary. 12) Estimates. 13) Incl. IP lines. 14) Incl. public payphones. 15) Including ISDN voice- channel equivalents. Data based on estimates. 16) As at 30/6/2015. 17) December 2015. 18) Incl. PSTN lines, ISDN paths, FWA subscriptions, public payphones and VOIP. 19) Including PRI access lines. 20) Source: AGCOM. 21) December 2015 The number of fixed public payphones is as of March 2015.(This data is reported by carriers every March). 22) The major fixed network provider shut down its fixed wireless network and migrated the subscibers to its GSM network. 23) Data from 4 main operators, LTC, BEELINE, UNITEL, ETL. 24) Data to 1.07.2015, source - Public Utilities Commission. 25) Including digital lines. Without including separate ISDN channels (abonnements au téléphone fixe). 26) Excl. ISDN channels and fixed wireless subscriptions. 27) December 2015. Source: January 2016 Management Information System Report. 28) Estimate. 29) First half 2015. 30) Figure is based on data received from Fixed Line Operators. 31) Estimate. 32) Inactive fixed telephones were disconnected. 33) Figures obtained from Bluesky and Digicel. 34) Q4 (consolidated end 2015 data not available yet). 35) Strong decrease due to the disconnection of inactive subscriptions. 36) Estimates. 37) Preliminary data. 38) December. 39) FCC trend-based estimate using recent historical data. 40) Preliminary. 41) Estimated. Mobile-cellular subscriptions per 100 inhabitants, 2014 1) Internet Activity Survey. 2) Obtained from URCA's Licensees. 3) Break in comparability: from this year, incl. only active prepaid subscriptions. Total: 10.56M. 4) For both Bhutan Telecom and Tashi Cell. 5) Break in comparability: previous year data refer to total number of configured sim instead of active subscriptions. 6) Incl. all mobile-cellular subscriptions that offer voice communications, but excludes mobile data subscriptions (via data cards, USB modems and M2M cards). 7) Preliminary. 8) Validation process of mobile accounts carried out in 2014, resulting in unverified accounts being disactivated. 9) Excl. 3 135 687 prepaid cards that are used to provide Travel SIM/WorldMobile service. 10) Reduction in multiple sim usage per subscriber (of different network operators). 11) Excludes data-only subscriptions. 12) Excl. data-only SIM cards and M2M cards. Data in line with ITU definition. Data before 2010 not in line with present ITU-Definition (it included non-active cards). 13) Providers data -NTRC Grenada. 14) New tax on numbering resources, which has prompted operators to return several numbers, either inactive ones or with low consumption. 15) December 2014. Including fixed wireless local loop Measuring the Information Society Report 2016 252

269 Annex 3 (WLL) subscriptions. 16) MCIT. 17) Estimate. 18) Dec.2014. Including PHS and data cards, undividable. 19) There are Ref. no from 4 main operators, LTC, BEELINE, UNITEL, ETL. 20) Active subscriptions. 21) December. 22) December 2014 Source: NTA Management Information System Report. 23) Q3 data. Excl. M2M and dedicated mobile broadband. 24) Estimate of subscriptions active in last 90 days. 25) Estimate. Incl. inactive. 26) Figures are as on 31st December, 2014 based on data received from Cellular Mobile Operators. 27) Preliminary. 28) From this year excl. data-only subscriptions. 29) Break in comparability: from this year, excl. M2M. 30) Excl. 495 811 M2M subscriptions. 31) Active subscriptions. 32) Includes active (in the last 6 months) prepaid accounts. 33) Decline was due to the regulatory controls on prepaid SIM cards which restricts each end user to hold no more than 3 prepaid cards. 34) 4 operators. 35) As researched by ECTEL. 36) Reduction due to change in accounting method for prepaid subs by major provider. 37) Estimates. 38) Preliminary data. 39) December. 40) Reported CTIA numbers. 41) Incl. dedicated data subscriptions. Mobile-cellular subscriptions per 100 inhabitants, 2015 1) Break in comparability: Active subscriptions. 2) Preliminary. 3) Internet activity survey June 2015. 4) December 2015. 5) Bhutan Telecom and Tashi Cell combined. 6) Preliminary. 7) Sept. 8) Preliminary. 9) Estimates. 10) Operators cleaned inactive lines. 11) Excl. 2 351 881 prepaid cards that are used to provide Travel SIM/WorldMobile service. 12) There was A Telecom Expansion Project (TEP) ongoing which results in about 12 Million new subscriber than the previous year. 13) Excludes data- only subscriptions. 14) Excl. data-only SIM cards and M2M cards. Data in line with ITU definition. Data before 2010 not in line with present ITU-Definition (it included non-active cards). 15) 4 players compete on the local market. The latest player entered the market at the 4 quarter of 2014. Data on the new entrant concern the 9th month of 2015. 16) Les donnes sont relatives aux deux operateurs (Orange et MTN). Le troisième opérateur n'est plus en service (Guinétel). 17) December 2015. 18) Estimate. 19) Source: AGCOM. 20) December 2015 including PHS and data cards, undividable. 21) Data to 1.07.2015, source - Public Utilities Commission. 22) Decline due to a new law regarding the identification of all subscribers. 23) Active subscriptions. 24) Information and Communication Technologies Authority of Mauritius. 25) Preliminary. 26) December 2015 Source: NTA Management Information System Report. 27) Incl. inactive. 28) First half 2015. Figure for 2014 needs to be revised. 29) Figure is reported after biometric re-verification of SIMs in 2015 by all Cellular Mobile Operators. 30) Estimate. 31) Excl. 492.761 M2M subscriptions. 32) Includes active (in the last 6 months) prepaid accounts. 33) Figures obtained from Bluesky and Digicel. 34) Estimated using Dec 2015; data as at end Mar 2016 is not available yet. 35) Estimates. 36) Preliminary data. 37) December. 38) UBS Investment Research Data as of 6/30/15 as reported in the FCC's Eighteenth Mobile Wireless Competition Report. 39) Incl. dedicated data subscriptions. 40) Preliminary. 41) Estimated. International Internet bandwidth Bit/s per Internet user, 2014 1) Total purchased international capacity contracted with operators outside Bulgaria. 2) Break in comparability: from this year, used capacity. 3) Contracted capacity. 4) Break in comparability: from this year used capacity. 5) Including Yahsat & Thuraya. 6) Installed capacity. International Internet bandwidth per Internet user, 2015 1) Dec 2015. 2) 2867.2 Mbit/s for Bhutan Telecom and 600 Mbit/s for TashiCell. 3) Ref. LTC&UNITEL. 4) As a policy more international capacity is acquired by the operators when the traffic load reaches 80% of the provisioned capacity. 5) As per data received from PTCL and TWA. 6) Contracted capacity. 7) Figures obtained from Bluesky and Digicel. 8) Downlink capacity. 9) Tonga Cable Limited. 10) December. 11) Including UAEs Yahsat & Thuraya. 12) Installed capacity. Percentage of households with computer), 2014 1) Country estimate. 2) Preliminary. Number to be revised. 3) A household is considered to have access only when it is available to all household members at any time. 4) ICT market survey. 5) Refers to urban households. 6) Incl. desktop, notebook and tablet, and excl. PDA and smartphone. 7) Excluding households who reside on the territory of temporary occupied Autonomous Republic of Crimea, in the city of Sevastopol and in a part of the zone where anti-terrorist operation is conducted. 8) American Community Survey. Percentage of households with computer, 2015 1) Estimate. 2) As of 2015, inc. tablets. 3) Preliminary estimate based on ICT HH survey. 4) Incl. desktop, notebook and tablet, and excl. PDA and smartphone. Percentage of households with Internet, 2014 1) Country estimate. 2) Inc. through computer or mobile. 3) Incl. access via mobile phones. 4) Break in comparability. A household is considered to have access only when it is available to all household members at any time. 5) ICT market survey. 6) Refers to urban households. 7) Excluding households who reside on the territory of temporary occupied Autonomous Measuring the Information Society Report 2016 253

270 Republic of Crimea, in the city of Sevastopol and in a part of the zone where anti-terrorist operation is conducted. 8) Incl. access to Internet via mobile phone. 9) American Community Survey. Percentage of households with Internet, 2015 1) Preliminary estimate based on ICT HH survey. Use indicators Percentage of individuals using the Internet, 2014 1) Census. 2) All population. 3) Population age 16-74. 4) Population age 7+. 5) Population age 15+. 6) Population age 6+. 7) Population age 16-74. 8) Population age 5+ in the last 3 months. 9) Population age 10+. ) 10) Population age 16-74. 11) Population age 15+. 12) Permanent residents at the age of 6 or above. In the last 6 months. 13) Population age 5+. 14) Population age 16-74. 15) Population age 6+. 16) Population age 16-74. 17) Population age 16-74. 18) Population age 16-74. 19) Population age 5+. 20) Population age 6+. 21) Population age 10+. 22) Population age 16-74. 23) Population age 16-74. 24) Population age 16-74. 25) Survey value is 49.1% for population age 6+ using Internet in the last 12 months. The current value is an ITU estimate provided for comparability reasons with 2015 value, which corresponds to users in the last 3 months. 26) Population age 16-74. 27) Population age 16-74. 28) Population age 10+. 29) Population age 16-74. 30) Population age 16- 74. 31) Population age 5+. 32) Country estimate. 33) Population age 16-74. 34) Population age 20+. 35) Population age 16-74. 36) Population age 14+. 37) Population age 6+. 38) Population age 5+. 39) Population age 16-74. 40) Population age 16-74. 41) Population age 16-74. 42) Population age 16-74. 43) Population age 3+. 44) Population age 15+. 45) Population age 16- 74. 46) Population age 5+. 47) Population age 6+. 48) All population. 49) Population aged 5 to 75 using Internet in the last 3 months. 50) Population age 16-74. 51) Population age 16-74. 52) Population age 10+. 53) Population age 10+. 54) Population age 10+ using internet in the last 3 months. 55) Population age 6+. 56) Population age 16-74. 57) Population age 16-74. 58) Population age 16-74. 59) Population age 15-72 who used the Internet in the last 12 months. 60) Population age 12-65 over total population. 61) Refers to total population. 62) Population age 16-74. 63) All population. 64) Population age 16-74. 65) Population age 16-74. 66) Population age 16-74. 67) Population age 16-74. 68) In the last 6 months. Population age 14+. 69) Population age 16-74 using in the last 3 months. 70) Population age 6+. Slight break in comparability since total population estimates were revised and they are lower than in 2013. 71) Population age 16-74 in the last 12 months. 72) All population. 73) Population age 15-74 in the last 3 months. 74) Population age 16-74. 75) Population age 6+. 76) Population age 3+. Percentage of individuals using the Internet, 2015 1) Population age 15+. 2) Population age 16-74. 3) Population age 7+. 4) Population age 15+. 5) Population age 6+. 6) Population age 16-74. 7) Population age 16-74. 8) Permanent residents at the age of 6 or above. In the last 6 months. 9 Population age 5+. 10) Population age 5+ in the last three months. 11) Population age 16-74. 12) Population age 16-74. 13) Population age 16-74. 14) Population age 16-74. 15) Population age 12+. 16) Population age 5+. 17) Population age 16-74. 18) Population age 16-74. 19) Population age 16-74. 20) Population age 6+. Break in comparability, reference period in the last 3 months. 21) Population age 16-74. 22) Population age 16-74. 23) Population age 10+. 24) Population age 16-74. 25) Population age 5+. 26) Population age 16-74. 27) Population age 16-74. 28) Population age 6-74. 29) Population age 16-74. 30) Population age 16-74. 31) Population age 16-74. 32) Population age 16-74. 33) Population age 3+. 34) Population age 15+. 35) Population age 16-74. 36) Population age 5+. 37) Population age 6+. Break in comparability: as of 2015 the respondent of ICT use questions is a self-respondent randomly selected and the survey is a stand-alone ICT survey. Before the ICT survey was a module attached to a main survey and respondent was an informed person of the household who responded about self and the other members of the household. 38) All population. 39) Population age 5+ using Internet in the last 3 months. 40) Population age 16-74. 41) Population age 16-74. 42) Preliminary estimate based on ICT HH survey, population aged 10+. 43) Population age 10+. 44) Population age 6+. 45) Population age 16-74. 46) Population age 16-74. 47) "Mainstream" population age 15+ living in households. 48) Population age 16-74. 49) Population age 15-72 who used the Internet in the last 12 months. 50) Population age 12-65 over total population. 51) Population age 16-74. 52) Population age 16-74. 53) Population age 16- 74. 54) Population age 16-74. 55) In the last 6 months. Population age 14+. 56) Population age 16-74. 57) Population age 6+. 58) Population age 16-74. 59) Population age 16-74. 60) Population age 3+. 61) Population age 6+. Fixed-broadband subscriptions per 100 inhabitants, 2014 1) Preliminary. 2) Internet Activity Survey, June 2014. 3) Obtained from URCA's Licensees. 4) Estimate. 5) Information provided by 85,4% of all ISPs. 6) ADSL, ADSL+, CDMA. 7) "Excluding satellite broadband users, the ground fixed wireless broadband". 8) Incl. WiFi subscriptions (not WiFi hotspots). 9) Incl. 144 kbit/s to less than 256 kbit/s. Excl. subscriptions with unspecified download capacity. 10) Estimate. 11) Fixed wimax. 12) "December 2014. These are the subscriptions with the minimum download speed of 512 kbps. This is as per the revised definition of Broadband (>= 512kbps) in India " 13) Incl. DSL and cable. 14) There are Ref.no from 4 main operators, LTC, BEELINE, UNITEL, ETL. 15) From this year incl. Internet cable subscriptions Measuring the Information Society Report 2016 254

271 Annex 3 (596'663) that were not included before. In 2014, for the first time MOT issued licences to cable operators to allow them to offer Internet services. 16) The number of subscriptions in 2014 went down because from 2014 WiFi services that are used as ad-on to subscriptions to other internet access services are no longer included. 17) Incl. non-residential customers (ca 25'000). 18) December 2014. Source: Management Information System Report. 19) Q3. 20) Estimate. 21) Figures are as on 31st December, 2014 based on data received from Broadband Operators. 22) Preliminary. 23) Incl. subscriptions at downstream speeds equal to or greater than 144 kbit/s (the number of subscriptions that are included in the 144-256 range is insignificant). 24) Estimates. 25) December. 26) Include 2,878 WiMax subscriptions, and 1850 Satellite subscriptions. 27) Please note that FCC collects information about broadband Internet access subscriptions in service that have downstream bandwidths exceeding 200 kbps, rather than 256 kbps. 28) Incl. ADSL and FTTH + LMDS. 29) Includes xDSL, fixed wireless data subscription and fixed broadband internet subscribers. The figure excludes prepaid wireless internet subscriptions. Increase from 2013 has been due to operators investing in fiber optic cable and landing of the submarine cable in Vanuatu in Jan 2014 has increased capacity and reduced price thus increasing the number of subscribers. 30) ISP subscribers with internet speed of at least 256 kbps. Fixed-broadband subscriptions per 100 inhabitants, 2015 1) Internet activity survey June 2015. 2) December 2015. 3) Number is inclusive of WIMAX, wireless broadband, FTTB, FTTC and FTTH. 4) Estimate. 5) Estimate. 6) ADSL, ADSL+, CDMA. 7) Incl. WiFi subscriptions (not WiFi hotspots). Estimates. 8) Incl. 144 kbit/s to less than 256 kbit/s. Excl. subscriptions with unspecified download capacity. 9) Narrow band dial-up service is abandoned and all have migrated to broadband connection using ADSL, and this has increased the fixed broadband service subscription significantly. 10) Fixed wimax. Includes 861 subscriptions at speeds of 128-255 kbps. 11) December 2015. Subscription with download speeds of at least 256 kbit/s. 12) Dec. 2015 - Inc. DSL and cable. 13) December 2015. 14) Incl. non-residential customers (ca 30'000). 15) Segregation of WiMAX services between mobile broadband and fixed broadband. 16) Information and Communication Technologies Authority of Mauritius. 17) Preliminary. 18) Estimate. 19) First half 2015. 20) Figure is based on data received from Broadband Operators. 21) Estimate. 22) Estimate. 23) Incl. subscriptions at downstream speeds equal to or greater than 144 kbit/s (the number of subscriptions that are included in the 144-256 range is insignificant). 24) Figures obtained from Bluesky and Digicel. 25) Estimated using Dec 2015; data as at end Mar 2016 is not available yet. 26) Q4 (consolidated end-2015 data not yet available). 27) Estimates. 28) Preliminary data. 29) December. 30) Include 2,359 WiMax subscriptions, and 389 Satellite subscriptions. 31) FCC trend-based estimate using recent historical data. 2015 data as of 6/30/15. Please note that FCC collects information about broadband Internet access subscriptions in service that have downstream bandwidths exceeding 200 kbps, rather than 256 kbps. 32) Incl. ADSL and FTTH + LMDS. 33) Includes xDSL, fixed wireless data subscription and fixed broadband internet subscribers. Numbers are believed to have dropped as some subscribers have preferred to switch to mobile broadband alternatives as prices for these services have fallen and quality has increased. 34) Preliminary. 35) Estimated. 36) ISP subscribers with internet speed of at least 256 kbps. Active mobile broadband subscriptions per 100 inhabitants, 2014 1) Internet Activity Survey, June. 2) Break in comparability: from this year excl. GPRS/EDGE only connections. Activity criteria: data communication in the last month. 3) Preliminary. Counting plans that allow mobile-broadband access and are using LTE, WCDMA and CDMA2000 enabled devices. 4) Break in comparability: previous year data refer to total number of configured sim instead of active subscriptions. 5) January 2015. 6) 3G and other more advanced mobile connections of at least 256 Kbit/s. 7) Preliminary. 8) Providers Data -NTRC Grenada. 9) Does not incl. prepaid smartphones. 10) Speeds equal or greater than 1 Mbit/s. 11) December 2014. These are the subscriptions with the minimum download speed of 512 kbps. This is as per the revised definition of Broadband (>= 512 kbps) in India. 12) In 2014, 3G and 4G licenses were awarded to the two largest mobile operators (Hamrahe Avval and IranCell). 13) Users who have made a transaction in the last 90 days via a handset, dongle/USB modem or other mobile Internet device, whereby they accessed advanced data services such as web/Internet content, online multiplayer gaming content, VoD or other equivalent data services (excluding SMS and MMS). 14) Estimate. 15) December 2014. Including standard and dedicated mobile broadband Wimax. 16) Estimate. 17) There are Ref.no from 4 main operators, LTC, BEELINE, UNITEL, ETL. 18) 3G subscriptions (prepaid+postpaid) provided instead, as all 3G subscriptions provide download speeds of at least 256 kbits/s when enabled. Users may disable/enable their mobile- broadband functionality via USSD code, via service hotline or in person. The number of 3G subscribers who have disabled their mobile-broadband functionality is not collected. Internet usage statistics of individual users are not collected either. 19) Equal to dedicated mobile-broadband subs as CAM does not report on standard mobile-broadband pay-as-you-go subscriptions. 20) Includes primarily Orange customers. 21) Lignes ayant réalisé des connections data sur les 3 derniers mois. 22) Source: January 2015 Management Information System Report. 23) Q4 data. 24) Estimate. 25) Subscriptions generating >0.5MB/month + data-only subscriptions. 26) Figures are as on 31st December, 2014 based on data received from Broadband and cellular mobile Operators. 27) Preliminary. 28) Activity period: 6 months. 29) Break in comparability: from this year, incl. handset-based mobile broadband. 30) Includes active subs (in the last 6 months), by 3G and higher technologies. 31) Decline was due to the regulatory controls on prepaid SIM cards which restricts each end user to hold no more than 3 prepaid cards. 32) Estimates. 33) Preliminary data. 34) December. 35) Break in comparability: from this year, excl. M2M subscriptions. 36) Based on data from Ovum. 37) Incl. subscriptions with potential access. 38) Data refer to theoretical ability of subscribers to use broadband speed mobile data services, rather than the number of active users of such services. 39) Break in comparability: blackberry and all mobile broadband subscriptions incl. pay-per-use. Measuring the Information Society Report 2016 255

272 Active mobile broadband subscriptions per 100 inhabitants, 2015 1) Abonnés 3G. Source: ARPT/Opérateurs. 2) Preliminary. 3) Internet activity survey June 2015. 4) December 2015. 5) Combined number for two operators. The increase is due to increased growth of smart phones and increase in 3G network. 6) Activity criteria: data communication in the last month. 7) Preliminary. 8) Estimates. 9) Only postpaid mobile-broadband subscriptions. 10) The Telecom Expansion Project undertaken which includes 2G to 3G migration in all major cities throughout the country has resulted in over 4 million mobile broadband subscriptions than the previous year. 11) Before 2014, Mobitel offered only 2G. In 2015 it received an LTE license and launched the service. 12) Does not incl. prepaid smartphones. 13) December 2015. Subscription with download speeds of at least 256 kbit/s. 14) Users who have made a transaction in the last 90 days via a handset, dongle/USB modem or other mobile Internet device, whereby they accessed advanced data services such as web/Internet content, online multiplayer gaming content, VoD or other equivalent data services (excluding SMS and MMS). 15) Dec.2015 Including standard and dedicated mobile broadband Wimax. 16) There are Ref.no from 4 main operators, LTC, BEELINE, UNITEL, ETL. 17) 3G + LTE subscriptions (prepaid+postpaid) provided instead, as all 3G + LTE subscriptions provide download speeds of at least 256 kbits/s when enabled. Users may disable/enable their mobile- broadband functionality via USSD code, service hotline or in person. The number of 3G & LTE subscribers who have disabled their mobile-broadband functionality is not collected. Internet usage statistics of individual users are not collected either. 18) The increase was due to the attractive price offered in postpaid and prepaid packages; pay per use subscriptions and the introduction of LTE package. 19) Equal to dedicated mobile broadband subs as CAM does not report on standard mobile broadband pay-as-you-go subscriptions. 20) Includes both, Orange and Sotelma customers. 21) Q2 2015. 22) Information and Communication Technologies Authority of Mauritius. 23) Preliminary. 24) Estimation DCE. 25) First half 2015. 26) Figure is based on data received from mobile broadband operators. 27) Estimate. 28) Estimate. Activity period: 6 months. 29) Includes active subs (in the last 6 months), by 3G and higher technologies. 30) Figures obtained from Bluesky Samoa Digicel Samoa and Lesa Telephone Service. 31) Estimated using Dec 2015; data as at end Mar 2016 is not available yet. 32) Slight drop due to upgrades on data services network by the two operators in April, Sept, Dec. 33) Estimates. 34) Preliminary data. 35) December. 36) Excl. M2M subscriptions. 37) Based on data from Ovum as of 6/30/15. 38) Incl. subscriptions with potential access. 39) Preliminary. 40) Estimated. Measuring the Information Society Report 2016 256

273

274 Internati onal Measuring Telecommunicati on Union the Information Place des Nati ons CH-1211 Geneva 20 Switzerland Society Report ISBN: 978-92-61-21431-9 2016 6 4 0 4 3 3 9 6 1 2 1 4 2 1 9 7 9 8 Printed in Switzerland Geneva, 2016

Related documents

City 2018 2019

City 2018 2019

2018–2019 CATALOG Fall 2018, Spring 2019, Summer 2019 1313 Park Blvd., San Diego, CA 92101 619-388-3400 www.sdcity.edu Ricky Shabazz, Ed.D. President San Diego City College is accredited by the Accred...

More info »
catalog e

catalog e

How to Buy. Wright Tool is proud to partner with thousands of distributors world wide to bring our products to tool users where they need them, Visit us at and when they need them. To locate a distrib...

More info »
TRUMPF bending tools catalog EN

TRUMPF bending tools catalog EN

Tool- catalog Edition 2018 TRUMPF LASERdur Bending Tools TRUMPF Angle Measuring System ACB Machine Tools / Power Tools Laser Technology / Elektronics

More info »
CDP water security score category weightings

CDP water security score category weightings

Water Security 2019: GENERAL SCORING METHODOLOGY CATEGORY WEIGHTINGS This 'summary sheet' outlines the 2019 Water Security scoring categories and the weightings that will be applied to these categorie...

More info »
BIG HORN 1   Floor Plan: MAIN

BIG HORN 1 Floor Plan: MAIN

FLIP PLAN________Initials GABLE ENTRY PORCH________Initials 40 FT PORCH________Initials C B A 39' - 11" 20' - 11 1/2" 18' - 11 1/2" 11' - 2" 12' - 9" 7' - 5 1/2" 3' - 7" 4' - 11 1/2" B - XLR-01-001-39...

More info »
IP 2019(7)

IP 2019(7)

Informational Publication 2019(7) Is My Connecticut Withholding Correct? Effective January 1, 2019 through December 31, 2019 Connecticut Income Tax Withholding Requirements for Individuals Tax informa...

More info »
UNSCEAR 2008 Report Vol.I

UNSCEAR 2008 Report Vol.I

This publication contains: VOLUME I: SOURCES SOURCES AND EFFECTS Report of the United Nations Scientific Committee on the Effects of Atomic Radiation to the General Assembly OF IONIZING RADIATION Scie...

More info »
w4

w4

RESIDENT PH-W4 City of Port Huron Income Tax - Employee's Withholding Certificate NONRESIDENT 1. First name and initial Last name Social security number Office, plant or department Employee identifica...

More info »
GM GR03 Study Guide 11.13.18

GM GR03 Study Guide 11.13.18

G e o rgia i nes st o M e l sessment stem As Sy Study/Resource Guide for Students and Parents Grade 3 Study/Resource Guide The Study/Resource Guides are intended to serve as a resource for parents and...

More info »
GM GR05 Study Guide 7.5.18

GM GR05 Study Guide 7.5.18

G e o rgia i nes st o M e l sessment stem As Sy Study/Resource Guide for Students and Parents Grade 5 Study/Resource Guide The Study/Resource Guides are intended to serve as a resource for parents and...

More info »
AC 2018

AC 2018

THE APPEALS BOOK FOR 2017 - 2020 20 0 2 - 7 1 0 www.ussailing.org 2

More info »
Cree 5 mm Round LED: C503C WAS/WAN

Cree 5 mm Round LED: C503C WAS/WAN

CLD-CT1098.011 PRODUCT FAMILY DATA SHEET ® Cree 5mm Round LED C503C-WAS/WAN APPLICATIONS PRODUCT DESCRIPTION FEATURES • Torch • Size (mm): 5 Round LEDs offer superior light output for excellent readab...

More info »
J Link / J Trace User Guide

J Link / J Trace User Guide

J-Link / J-Trace User Guide Document: UM08001 Software Version: 6.44 Revision: 2 Date: April 25, 2019 A product of SEGGER Microcontroller GmbH www.segger.com

More info »
Microsoft Word   Setups USB ME Systems.doc

Microsoft Word Setups USB ME Systems.doc

USB-ME-Systems System Suggestions

More info »
Microsoft Word   19 074, SRB

Microsoft Word 19 074, SRB

Formatted Courtesy of: www.ArmyReenlistment.com MILPER Message Number: 19-074 Proponent: AHRC-EPF-R Title Selective Retention Bonus (SRB) ...Issued: [01 Mar 2019]... https://www.hrc.army.mil/Milper/19...

More info »
W4 Form

W4 Form

LONG BEACH UNIFIED SCHOOL DISTRICT, 1515 Hughes Way, Long Beach, CA 90810 Employer's Federal ID# 95-6001886 EMPLOYEE'S WITHHOLDING ALLOWANCE CERTIFICATE State ID# 800-9069-9 W4 FORM Federal and State ...

More info »
Aid Category to Benefit Plan Crosswalk

Aid Category to Benefit Plan Crosswalk

Last Update: 10/16/17 Note: This document is a working document and modifications may be needed to reflect future changes. Federal Waiver iC Aid Legacy Aid Poverty Federal Inst. iC Benefit Category Le...

More info »
USAIS PAM 350 6 02 JAN 2019

USAIS PAM 350 6 02 JAN 2019

US AIS PAMPHL ET 350 -6 Exp ert Infant ryman Badge 02 JANUARY 2019 Obs olete Editi ons All Prev io us DE PARTMENT OF THE ARMY Ar antry my Inf United States Sch l oo 1 Page

More info »
Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space

Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space

TIONAL ACADEMIES PRESS THE NA This PDF is available at http://nap.edu/24938 SHARE     Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space DET AILS 700 pages | 8.5 ...

More info »
ez pmi revj 2

ez pmi revj 2

® EZ-ZONE PM User’s Manual Integrated Controller Models TA TO L S CU T O M ER N A S A CT O I F TIS arranty W Y 3 ear ISO 9001 Registered Company 1241 Bundy Boulevard., Winona, Minnesota USA 55987 Wino...

More info »