210 110 1MB
English Pages VIII, 83 [85] Year 2020
SPRINGER BRIEFS IN ECONOMICS
Jeffrey James
The Impact of Smart Feature Phones on Development Internet, Literacy and Digital Skills 123
SpringerBriefs in Economics
SpringerBriefs present concise summaries of cutting-edge research and practical applications across a wide spectrum of fields. Featuring compact volumes of 50 to 125 pages, the series covers a range of content from professional to academic. Typical topics might include: • A timely report of state-of-the art analytical techniques • A bridge between new research results, as published in journal articles, and a contextual literature review • A snapshot of a hot or emerging topic • An in-depth case study or clinical example • A presentation of core concepts that students must understand in order to make independent contributions SpringerBriefs in Economics showcase emerging theory, empirical research, and practical application in microeconomics, macroeconomics, economic policy, public finance, econometrics, regional science, and related fields, from a global author community. Briefs are characterized by fast, global electronic dissemination, standard publishing contracts, standardized manuscript preparation and formatting guidelines, and expedited production schedules.
More information about this series at http://www.springer.com/series/8876
Jeffrey James
The Impact of Smart Feature Phones on Development Internet, Literacy and Digital Skills
123
Jeffrey James Tilburg University Tilburg, The Netherlands
ISSN 2191-5504 ISSN 2191-5512 (electronic) SpringerBriefs in Economics ISBN 978-3-030-62211-4 ISBN 978-3-030-62212-1 (eBook) https://doi.org/10.1007/978-3-030-62212-1 © The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Acknowledgements
I am deeply indebted to Josette Janssen for the typing and editing of this short book. She performed these functions at an alarmingly fast rate, with great precision and an eagle eye for spotting the many mistakes. I count myself extremely fortunate that she was willing to work with me on this venture. Niko Chtouris at Springer was unfailingly encouraging and helpful. He is the best editor I have dealt with at that publisher. Peter Richardson and Varun Meta at Counterpoint Research provided me with helpful comments as did Tim Metz and Yang Wang at KaiOS Technologies. In no particular order, I am grateful to those who have granted me permission to use their copyright material. The International Telecommunication Union (ITU) for Tables 2.5, 2.6, 5.6, 5.7; the GSMA Intelligence for Tables 2.4, 2.7, 6.1, 6.2, 6.3; and Tables 5.1 and 5.2 are from the World Bank, the little data book on information and communication technology, under https://openknowledge.worldbank.org/handle/ 10986/31967. I am also grateful to UNESCO for permission to use Tables 5.3, 5.4, 5.5 and 5.9 (all from the UNESCO Institute of Statistics); Statista for permission to use Table 3.1; PEW Research Center for the use of Table 2.1; from an article by Silver and Johnson (2018); KaiOS Technologies for permission to use Table 2.3; World Economic Forum for permission to use Table 4.4 under the Creative Commons Attribution─Non-Commercial─No Derivatives 4.0 International Public License; and Counterpoint Research for permission to use Table 2.2; Chap. 2 is reproduced from the Information Society, 36(4), under the Creative Commons CC BY license of Taylor & Francis.
v
Contents
. . . . .
. . . . .
. . . . .
. . . . .
1 1 4 6 7
2 The Smart Feature Phone Revolution in Developing Countries: Bringing the Internet to the Bottom of the Pyramid . . . . . . . . . . 2.1 The Genesis of the Smart Feature Phone . . . . . . . . . . . . . . . . 2.2 The Diffusion of Smart Feature Phones . . . . . . . . . . . . . . . . . 2.3 The Impact of Smart Feature Phones . . . . . . . . . . . . . . . . . . . 2.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
11 12 16 20 24 25
. . . . .
. . . . .
29 30 32 33 35
.. .. ..
38 40 41
to Sub-Saharan Africa . . . . . . . . . . . . . . .
45 45 46 46
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Smart Feature Phones . . . . . . . . . . . . . . . . . . . 1.2 Literacy, the Second Divide and User Choices . Appendix: A Word on Method . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Part I
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
. . . . .
Smart Feature Phones and Development
3 Smart Feature Phones and Welfare in Poor Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Economic Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Smart Feature Phones and Other Dimensions of Poverty . . . . . . 3.3 Localisation and Relevance of Internet Content . . . . . . . . . . . . 3.4 The Linguistic Divide, Smart Feature Phones and the Poor . . . . 3.5 The Lack of Digital Skills, Countervailing Policies and the Poor in Developing Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Extending the Experience 4.1 Introduction . . . . . . . 4.2 Affordability . . . . . . . 4.2.1 Device Price .
................................. ................................. .................................
vii
viii
Contents
4.2.2 Data Affordability . . 4.2.3 Income Affordability 4.3 Local Content . . . . . . . . . . . 4.4 Digital Skills . . . . . . . . . . . 4.5 Conclusions . . . . . . . . . . . . References . . . . . . . . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
46 47 48 50 52 53
5 Measuring the Second Digital Divide: Education and Skills . . . . 5.1 Measuring the Original Conception of the Digital Divide . . . . 5.2 Recognition and Measurement of the Second Digital Divide . . 5.3 The Crisis in Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Measuring the Digital Skills Component of the Second Digital Divide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5 The Second Digital Divide and the Sustainable Development Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . .
. . . .
. . . .
57 58 60 63
...
64
... ... ...
66 68 69
6 Anti-development Bias in the Use of the Internet in Developing Countries. What Underlies It? . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 The Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Information Imperfections and Patterns of Internet Use . . . . . . 6.4 Causes of Scarce Digital Knowledge in Developing Countries 6.5 Knowledge Imperfections and Preferences for Internet Use . . . 6.6 Policy Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.7 Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . .
71 72 75 77 78 78 80 82 83
Part II
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
Digital Skills and Digital Paradoxes
. . . . . . . . .
. . . . . . . . .
Chapter 1
Introduction
Abstract This introductory chapter describes not only the contents of the others, but also, and importantly, how they fit together. The main subject of the book is how smart feature phones can bring the Internet to large numbers of poor people in developing countries. As shown for the case of India, this can occur on a vast scale under the right conditions (which include an effectively zero price for the handset due to heavy subsidization by one of the project partners, Reliance Industries). In Sub-Saharan Africa, however, where KaiOS phones have recently been introduced, the price is around US$20, which although still much cheaper than the average smartphone, is far more expensive than the Jio. It is still too early, however, to gauge whether KaiOS smart feature phones in Sub-Sahara have been successful even at the higher price just noted. Much will also depend, I argue, on non-economic factors such as local content, local languages, the presence of the Google Assistant (which allows illiterate users to communicate in voice rather than text) and digital skills, which have been relatively neglected over recent years in most poor developing countries. Keywords KaiOS · JioPhone · India · Poverty
1.1 Smart Feature Phones This book’s point of departure is that not very long ago, much of the developing world had little or no access to the Internet and the benefits it potentially affords the user. In fact, in an article written in 2016, the World Economic Forum pointed out that More than 4 billion people, mostly in developing countries, still don’t have access to the internet. This means that over half of the world’s population is missing out on the lifechanging benefits of connectivity, from financial services to health and education, being brought about by the increasing pace of innovation known as the Fourth Industrial Revolution. (World Economic Forum 2016, n.p.)
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 J. James, The Impact of Smart Feature Phones on Development, SpringerBriefs in Economics, https://doi.org/10.1007/978-3-030-62212-1_1
1
2
1 Introduction
This was the case, moreover, in spite of the fact that ‘Universal, affordable internet access is part of the UN’s Sustainable Development Goals (SDGs), and governments, companies, local and international organizations, and members of civil society are working to get more people online’ (World Economic Forum, 2016, n.p.). The problem was then primarily that the only means of gaining access to the Internet in poor countries was through expensive smartphones, designed in and for the developed countries (including their average income levels) and out of reach for the vast majority of the poor in developing countries, especially those living in rural areas (ironically, because such persons tend to be most in need of what the Internet can offer, such as communications with friends and family, which do not require difficult, long-distance and expensive travel, measured in what the poor can afford). At least in one of the world’s largest developing countries, however, this problem changed dramatically in 2017 with the introduction in India of the Jio smart feature phone, running the KaiOS OS and selling for virtually nothing due to the heavy subsidisation by Reliance Jio, a partner in the project and one of India’s largest conglomerates. It is hardly surprising in these circumstances that sales of the new phone were extremely rapid, not only because it was effectively free, but also because of the wide range of attractive smartphone-like applications that came with the product (such as Facebook, YouTube and Google Assistant). In fact, ‘Out of more than 100 million subscribers that Reliance Jio added since the Jio Phone launch in late 2017, the KaiOS powered 4G smart feature phone contributes close to half of those net additions. Boasting of smartphone-like features, access to YouTube, WhatsApp and Facebook, and 4G network, Jio Phone has already [by 2019] captured 38% of the feature phone market in India.’ (Livemint 2019, n.p.). In two years, that is to say, 50 million such phones had been sold. The genesis of this remarkable achievement is described in Chap. 2, where I stress that the avowed mission of both KaiOS and Reliance Industries was to reduce and even eliminate the digital divide between rich and poor countries with regard to Internet access and use. Much of this was to be achieved moreover by the design of a product that was shorn of all characteristics except those that bore on the basic functioning of the phone. In this respect the innovation has much to do with the appropriate technology movement that began in 1973 with Schumacher’s book on intermediate technology1 and continued well into the nineteen eighties. For, as noted below, the smart feature phone lies somewhere between the basic mobile phone (without connectivity) and the smartphone with its complicated applications, frills and relatively high price. Chapter 2 also describes how much the historical origins of the JioPhone and its successors owe to the early Firefox OS, which culminated in the KaiOS OS.
1 Schumacher
(1973).
1.1 Smart Feature Phones
3
Chapter 3 introduces a key aspect of the book, which in spite of its salience, has not attracted the attention it deserves. I am referring here not to the access that is available to smart feature phones in poor countries, but rather the benefits that are actually derived from them (though, of course, access is a necessary precursor to use). In this chapter, I contend that the benefits available from the smartphone depend not just on familiar economic mechanisms, but also those that have to do with cultural, social and institutional factors. Such non-economic factors range from issues of relevance and localisation, to linguistics, and the form of communication (whether by text or voice). I do not deal at this stage with literacy and digital skills, not because I regard them as relatively unimportant, but rather because they are so pressing a problem as to occupy the entire second part of the book. In fact, some authors argue that such skills are the primary determinant of what the poor are actually able to derive from the Internet when interaction takes place through KaiOS powered smart feature phones in poor (rural) regions of developing countries. Can they, for example, navigate the Internet or send e-mail attachments? Chapter 3 then deals mainly with the influence exerted by issues of relevance, local content and local language on the gains available from KaiOS-based smart feature phones. The Indian JioPhone is of course the best known of these, but variations on it have recently been widely introduced in Sub-Sahara, in combination with two of the largest telecom operators in the region, MTN and Orange. In fact, Chap. 4 is devoted to the interesting and crucial policy question of whether and to what extent the Indian Jio case can be successfully replicated in Sub-Saharan Africa and other poor countries. Unfortunately, the data required to make this assessment in Sub-Saharan Africa, are not yet available, though initial sales data should soon become available for countries where MTN and Orange already have a substantial presence. Even without such data, it is nonetheless possible, as shown in Chap. 4, to say something on this important question. Much depends, for example, on issues of income and affordability, in which India holds a clear advantage. This advantage is especially acute in regard to the price of handsets and data. For, whereas the JioPhone is effectively given away for free and data are exceptionally affordable, in Sub-Saharan Africa most KaiOS smartphones sell for around US$20 and data prices are typically much higher. And as regards per capita income, while most Sub-Saharan countries belong to the low-income category, India is classified as a low middle-income country (though, of course, there are exceptions where some of the former countries enjoy higher incomes than the Indian average). I was at pains to emphasise, however, that major digital innovations are not unheard of in Sub-Saharan Africa. On the contrary, the region has become a world leader in mobile money (GSMA 2019). More specifically, ‘With 290 live services in 95 countries and 372 million active accounts, mobile money is entering the mainstream and becoming the path to financial inclusion in most low-income countries’ (GSMA 2019, p. 1). Mobile phones effectively bring the unbanked into the formal and more favourable banking system, by allowing ‘users to send, receive and store money without needing a bank account. Funds are stored
4
1 Introduction
in a secure electronic account attached to a mobile phone number’ (Leung 2019, n.p.). However, because mobile money requires Internet use, smart feature phones allow this at a fraction of the cost of smartphones. MTN and Orange, for example, fill a major vacuum here in that they provide an Internet connection without having to purchase expensive smartphones.
1.2 Literacy, the Second Divide and User Choices The second part of the book, as noted above, is concerned with the recognition that what is actually derived from the smart feature phone depends on certain complementary inputs, such as literacy and digital skills. These form part of what is known as the second digital divide, itself a relatively neglected aspect of the literature. In fact, Chap. 5 includes an original discussion on how this divide (which is concerned with differences in the use and benefit from digital technology), can be measured and compared with the first divide, which concerns itself with differences in the adoption of such technology. The former divide is measured with respect to human attainments in reading and mathematics on the one hand and digital skills on the other. What I find is that even the acute divide in the Internet between rich and poor countries (the first divide), is exceeded by the data representing the second divide in literacy and mathematics. There appear to be two possible explanations for this somewhat anomalous outcome. The first has to do with the possibility that some digital skills can be acquired without adequate prior functionings in literacy and numeracy. The more compelling reason, though, is likely to involve the so-called ‘crisis in learning’. That is, the acute crisis in basic learning that besets so many schools in poor countries, especially those in Africa. For, while enrollments in that continent have generally gone up sharply, scores in literacy and mathematics have dwindled.2 The final chapter (Chap. 6) begins with the common observation that users of smart feature phones tend to make shallower and less-demanding choices of available Internet uses. The majority of them, that is to say, tend to prefer entertainment and sport, to development-oriented uses such as for health-care, job-search and e-government (Wang 2020; The Economist 2019). One view, espoused by The Economist (2019), rests on the traditional notion in economics of ‘consumer sovereignty’, which as its name suggests, regards preferences as the sole preserve of the individual him or herself and are not subject to being tampered with, even if they conflict with some of the Sustainable Development Goals. My own view, by contrast, has an affinity with the model propounded by Harsanyi (1953) in which information imperfections cast doubt over the sanctity of revealed preferences.
2 See
UNESCO (2013).
1.2 Literacy, the Second Divide and User Choices
5
The specific problem, as I see it, is that with respect to digital skills in poor developing groups, a lack of knowledge is rife, because there are few places where reliable information can be acquired by first-time users of the Internet. Such persons are then more often than not led to choose relatively undemanding use categories.3 According to Wang’s study of Nigeria, for example, ‘New users tend to learn about the internet from members of their local communities in an offline setting, which often leads to a narrow view of what the internet has to offer. For example, they might only hear about a few specific apps and have misconceptions about both the internet and how to access it’ (Wang 2020, n.p.). Even more telling is his observation that ‘Mobile internet can provide support in areas like health, education and business. However, these uses are the least popular because new users are unaware of their potential value’ (Wang 2020, emphasis added). Note, finally, that ‘Lack of knowledge causes Nigerians to use their internet-enabled phones in a limited way. Without exposure to activities like downloading apps, setting up online accounts, using web browsers, or making video calls, new users, never learn all the internet has to offer them’ (Wang 2020, n.p.). A more specific formulation of the argument, which includes the role of the poor, can be expressed in the following two propositions. The first is that developmental uses of the Internet tend to be skill-intensive and the second is that it is precisely the least advantaged members of developing countries that suffer most from the lack of such skills. Stated thus, the argument has less to do with preferences and choices, as it does about the constraints facing different groups that determine those ‘choices’. It is not my intention here to review the numerous ways in which digital skills could be improved, but one of them warrants particular attention since it deals specifically with smart feature phones created by KaiOS Technologies, a major actor in most of the chapters below. It was indeed the same firm that announced in 2019 the introduction of its ‘in-house’ educational application called ‘Life’ in Sub-Saharan Africa. The goals of this programme have been well-described by Meta (2020) in the following terms: Life serves as a trusted educational resource for first-time internet users, and provides curated content in six categories, including Digital Skills, Education, Health, Gender Equality, Agriculture, and Financial Education. These resources help users navigate their internet experience and are optimized for the non-touch screens of KaiOS powered devices; as well as gain access to valuable information that can improve their lives. … The most accessed section overall is Digital Skills, which was developed by GSMA’s Mobile Internet Skills Training Toolkit (MISTT), a resource for organizations who want to introduce the mobile internet to first time mobile internet users. (Meta 2020, n.p., emphasis in original)
Whether this popular application exerts a countervailing influence on the tendency described above, for users to favour non-developmental uses on the Internet, remains to be seen. Ultimately, though, I cannot escape the conclusion that the success of the smart feature phone outside India, depends heavily on what more can be done in such regions to reduce the affordability constraint. For, whereas the smart feature phone in India and Sub-Saharan Africa has done a great deal to reduce the price of 3 Note
that the more developmental options may not always be available.
6
1 Introduction
getting online, a large number of those living in poor countries still cannot afford the price that is being asked in parts of Sub-Saharan Africa. This is not to say that the region will be unable to record a serious number of smart feature phones. The point is rather that what is achieved will not generally be as spectacular as the Indian case. (Though I did refer to the successful case of mobile money in much of the region on the basis of smart feature phones.)4
Appendix: A Word on Method There are several features of this Brief that distinguish it from the “typical” academic piece in the social sciences which is relatively technical and narrow in focus (apart of course from the relative brevity of this work in comparison to the length of most academic books). One such difference is that there are more than the usual number of places with insufficient data to confirm or deny the hypotheses that are then being advanced. Much of this is due as I see it to the fact that the issues under consideration are too recent to have attracted academic attention (the subject of this book, for example, the smart feature phone, was introduced in India in 2017). As a consequence, I have had to rely to a greater extent than is true of much academic research, on industry reports, financial newspapers and international institutions. A related, but distinct, reason for the departure from the usual academic method, with its emphasis on multivariate techniques, has to do with differences in research goals. In particular, what is presented below may be described as perspective research, which, in spite of its limitations, has been described as playing ‘an important role in the academic research portfolio. They [perspective pieces] stimulate further interest about presented topics within the reader audience. They are different from other types of articles because they present a different take on an existing issue, tackle new and trending issues, or emphasize topics that are important, but have been neglected, in the scholarly literature. In some scientific fields … they bring new issues and ideas to the forefront. In general, their role is to enlighten a general audience about important issues’ (Enago Academy 2019, nd, emphasis added). As illustrations of these distinctive features of the perspective approach, I offer three examples, outside India the inexpensive smart feature phone with Internet connectivity is almost entirely unheard of, in spite of the success it has enjoyed in the country and the potential it affords other poor countries. In that regard, moreover, there is no discussion I am aware of, that is addressed to the question of whether and to what extent the Indian model can be generalized to the developing countries, especially the poorest of them, where the need for the benefits of information is 4 Outside of Sub-Saharan Africa, in Indonesia, there is the case of the ‘Wiz Phone’, which is similar
to the Jio model and sells for only US$7. It is said to be based on a new business model, but the details of this are scant, even on the home-page of the product (Leung 2018).
Appendix: A Word on Method
7
most acute. Finally, to the best of my knowledge, no research has yet addressed the apparent paradox of why poor people use the Internet for the purposes of leisure rather than labour.
References Enago Academy (2019) How to share opinion on research articles. Available http://www.enago. com/academy/perspective-how-to-share-opinion-on-research-articles/. Accessed 10 June 2020 GSMA (2019) State of the industry report on mobile money. Available https://www.gsma.com/ sotir/wp-content/uploads/2020/03/GSMA-State-of-the-Industry-Report-on-Mobile-Money2019-Summary.pdf. Accessed 3 Nov 2019 Harsanyi JC (1953) The welfare of economics of variable tastes. Rev Econ Stud 21(3):204–213 Leung E (2018) The first KaiOS-powered smart feature phone arrives in Indonesia. Available https:// www.kaiostech.com/the-first-kaios-powered-smart-feature-phone-arrives-in-indonesia-a-newpartnership-model-combining-retail-and-banking-to-advance-financial-inclusion/. Accessed 3 Feb 2019 Leung E (2019) How smart feature phones help the unbanked. Industry insights, KaiOS Technology. Available https://www.kaiostech.com/how-smart-feature-phones-help-the-unbanked/. Accessed 31 Dec 2019 Livemint (2019) Reliance Jio sold 5 crore smart feature phones in less than 2 years: Report. Available https://www.livemint.com/technology/tech-news/reliance-jio-sold-5-croresmart-feature-phones-in-less-than-2-years-report-1550635450416.html. Accessed 25 Feb 2020 Meta A (2020) How the Life app is changing lives around the world. Available via https://www.kai ostech.com/author/axel-meta/. Accessed 10 June 2020 Schumacher EF (1973) Small is beautiful. Blond & Briggs, London The Economist (2019) How the pursuit of leisure drives internet use, June 8. Available https://www. economist.com/briefing/2019/06/08/how-the-pursuit-of-leisure-drives-internet-use. Accessed 10 Dec 2019 UNESCO (2013) The global learning crisis. Why every child deserves a quality education. Available https://unesdoc.unesco.org/ark:/48223/pf0000223826. Accessed 3 July 2017 Wang Y (2020) First-time internet users in Nigeria use the internet in a unique way. Available https://www.kaiostech.com/author/yangwang/. Accessed 14 Apr 2020 World Economic Forum (2016) 4 billion people still don’t have internet access. Here’s how to connect them. Available https://www.weforum.org/agenda/2016/05/4-billion-people-still-don-thave-internet-access-here-s-how-to-connect-them/. Accessed 9 Sept 2019
Part I
Smart Feature Phones and Development
Chapter 2
The Smart Feature Phone Revolution in Developing Countries: Bringing the Internet to the Bottom of the Pyramid
Abstract Until recently, the only way for the population of developing countries to access the Internet was through expensive smartphones, designed in and for developed countries. In the past few years, however, a major new innovation has emerged, the smart feature phone with Internet connectivity, which was specifically designed for those with low incomes in developing countries. This chapter explains the development process for the smart feature phone, how this has influenced the nature and extent of adoption, and its use by low-income groups, including their demonstrated preference for uses related to entertainment rather than more traditional ‘workrelated’ goals. The focus is on the case of India, where the JioPhone has already reached millions of people with low incomes. Keywords Poverty · India · The Internet · Mobile phone · JioPhone In 2017, only 3 categories were needed to specify the ownership and non-ownership of mobile phones in developing countries–owners of smartphones with Internet connectivity, owners of basic models with few features and no Internet connectivity, people who owned no mobile phones at all (see Table 2.1). Now, the picture would not be complete without a fourth category–owners of smart feature phone (roughly, a hybrid between a simple phone and a smartphone). Featured most prominently in India, in the form of the JioPhone, smart feature phones saw a 252% growth in demand in 2018 and are predicted to exhibit rapid growth in the next few years (it has just been introduced in many Sub-Saharan African countries). Estimates provided by Counterpoint Research (2019), moreover, suggest that nearly 370 million smart feature phones are expected to be sold, across the world between 2019 and 2021.
This chapter originally appeared as “The smart feature phone revolution in developing countries” in The Information Society, 36(4). © The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 J. James, The Impact of Smart Feature Phones on Development, SpringerBriefs in Economics, https://doi.org/10.1007/978-3-030-62212-1_2
11
12
2 The Smart Feature Phone Revolution in Developing Countries …
Table 2.1 Categories of ownership of mobile phones in select African countries, 2017 % of sample reporting ownership Smartphone
Basic phone
No phone
South Africa
51
40
9
Ghana
35
45
20
Senegal
34
46
21
Nigeria
32
48
20
Kenya
30
50
20
Tanzania
13
62
25
Source Silver and Johnson (2018, p. 13)
The questions to be answered are: What shaped the development of the new technology? What has been the extent and nature of the adoption? What impact it has had on those with relatively low incomes? These questions are best addressed, in my view, with a sequential analytical framework, running from the development of smart feature phones and continuing through their diffusion to the impact they might have had. This framework is based on the recognition that since the various phases influence one another, they need to be considered together. I begin, accordingly, with the critical first phase that has to do with the way in which the new technology was developed.
2.1 The Genesis of the Smart Feature Phone I take note in the first place that the location of an innovation tends to bear on its subsequent rate of diffusion in different markets. This is the case mainly because technologies developed in and for a particular location, tend to fit in with the circumstances prevailing there, where they will to this extent be more readily adopted. For many decades, however, this mechanism worked largely in one direction, because of a widespread lack of innovative capabilities in most developing countries, especially those that were particularly backward in terms of their available technological capabilities.1 Thus it was that most of the world’s innovations were developed in the rich countries and reflected, more or less closely, the socio-economic characteristics of those countries. Technologies designed in and for the problems of low and middle income groups in poor countries were rare and their needs were largely ignored. It was for this reason that there were calls in the nineteen seventies and eighties for “intermediate” and “appropriate” technologies2 in developing countries, especially those with relatively low incomes and small markets for technologies developed in 1 Singer 2 By,
(1970) referred to this as international technological dualism. most notably, Schumacher (1973) and Stewart (1977).
2.1 The Genesis of the Smart Feature Phone
13
the rich world. And while it is true that there were many attempts to supply those technologies,3 they were generally on a small scale and rarely scaled up to anything like the national level4 (though there are some notable exceptions such as the Rural Access Roads Project in Kenya5 and the “Nirma” brand of detergents sold in India6 ). The broad picture of the global innovation system I have just painted, however, has changed quite markedly in recent decades, especially perhaps in the largest developing countries, such as India. In that country, for example, R&D grew rapidly over the period 2000–2015, leading one observer to remark that it was rapidly moving towards becoming an innovation hub in Asia (Financial Express 2019b). On the demand side, too, there have been significant changes. Average incomes of developing countries as a whole, for example, have risen and the numbers of those living in extreme poverty have declined in many poor countries (World Bank 2018a, b). To this extent, the effective demand for technologies that meet developing countries’ needs, will also have risen. More specific reasons are needed, however, to account for the features of the JioPhone and its subsequent popularity in India. One of them has to do with the fact that the two major actors–Reliance Industries and KaiOS–were both driven by the same overriding goal of filling the formidable gap between basic mobiles and smartphones. In the former case, one has to look to the visionary qualities of the CEO, Mr. Mukesh Ambani, who was not content with making marginal change to the telecommunications ecosystem that existed prior to the introduction of the Jio smart feature phone. Rather, he was committed to a vision of massive change to be wrought by smart feature phones that would alter the lives of tens of millions of less advantaged Indians, by connecting them at low cost to the Internet through smart feature phones.7 For its part, KaiOS Technologies, the software developer, expressed the goal it was pursuing in the following terms: “KaiOS and our partners have been driving forces in bringing the smart feature phone segment to life. We developed our platform with emerging regions and niche user groups in mind, from those priced out of the existing market, to others looking for an alternative to smartphones. There’s a huge demand for reliable affordable technology in regions like Asia, Africa and Latin America” (Wang 2019). KaiOS, as Leung (2018) has pointed out, seemed to come entirely out of the blue. Towards the end of 2017, it lacked even a website. A few months later, the firm’s operating system was running on more than 30 million phones in India and North America. Roughly a year later, KaiOS was being offered in no fewer than 100 countries across the globe.
3 For
a review see James (1989). the general problems of scaling up, see Hartmann and Linn (2008). 5 See the ILO’s (International Labour Organization) own description of the program (ILO 1992). 6 For a description of the case, see Singh and Pandey (2005). 7 See for example the data in Table 2.1. 4 On
14
2 The Smart Feature Phone Revolution in Developing Countries …
In essence, the story underlying this remarkable experience turns on open-source software. ‘KaiOS, though, is not Firefox OS.’ (Leung 2018, p. 1) Rather, the platform ‘is based on the original Mozilla project’ (Leung 2018, p. 1). In fact, KaiOS has been developed into a project that is not only more robust but also more expansive than the original Firefox OS. In this, the relationship between KaiOS and a major Indian firm called Reliance Industries, played a dominant role. For example, after it had adopted the KaiOS platform, Reliance acquired tens of millions of 2G feature phone users and transferred them to its 4G network. Moreover, the Indian firm subsidized the JioPhone so heavily that it was sold almost for free (The Economic Times, July 21 2017). The fact that the two firms in question had convergent goals, does not, however, explain how they came to collaborate or the various ways in which they each contributed to the project as a whole. My analysis of the second question I should note, focuses disproportionately on the history of KaiOS because of the dominant role it played in the crucial design stage. More specifically, the genesis of the JioPhone cannot properly be understood without reference to its association with the Firefox operating system, which is based on open-source software. It is convenient to begin this story at the end of 2015, when Mozilla announced that it was abandoning the Firefox operating system as a platform for smartphones. For many in the industry, this seemed to put an end to the idea of a mobile OS that was built around the open web (Summers 2019). Yet, only a few years later, through a series of more or less coincidental events, Firefox reappeared, in much modified form, as the operating system for a smart feature phone (known, as noted above, as the JioPhone). Much had to do with the fact that KaiOS had begun in the United States with an Alcatel brand called “Go Flip,” which AT&T, Sprint, and T-Mobile were persuaded to stock, because of prior work connections between the parties. These events, in turn, allowed KaiOS to win a contract with Reliance Industries in India. Out of this collaboration, the JioPhone was ultimately born, owing much to the original Firefox OS, including the telling fact that KaiOS Technologies hired 30 former Mozilla employees to develop KaiOS (Summers 2019). Note, though, that KaiOS has been “developed into something much more robust and expanded than the original Firefox OS” (Dey 2019). A major difference lies in the range of popular applications that KaiOS Technologies managed to secure from major technology firms such as Google. These included Facebook, YouTube, WhatsApp and Google Assistant. Some of these were especially important because they bore on the ability of poor, illiterate users to use the Internet and to do so effectively. Consider first, in this regard, the case of WhatsApp. According to a recent article in The Economist (2019, n.p.), for example: video is easier to post to your peers than writing is. And speech beats typing–as can be seen from the use of WhatsApp to send voice messages rather than texts. Though usually associated with pricey first-world gadgets such as the Amazon Echo, voice-input systems have found enthusiasts in the poor world, too. New Internet users in India routinely use voice commands to operate their phones, including for such tasks as making calls.
2.1 The Genesis of the Smart Feature Phone
15
Or, to take another pertinent example from India, “an illiterate cab driver in Mumbai, uses Uber’s ride-hailing app through a combination of voice input and audio direction. When he has to send messages, he speaks into a voice-to-text app, copies what turns up on the screen onto a messaging app and sends it to his waiting passenger-to-be, hoping it makes sense” (The Economist 2019, n.p.). The Google Assistant too, reduces the limitations of illiteracy in the use of smart feature phones such as JioPhone, because it allows the substitution of voice for text communications. For example, questions posed to the Assistant in voice are answered in the same mode. And “In places where people are coming online for the first time, millions are discovering that voice is a more natural way to interact with technology, overcoming technological hurdles that previously seemed out of reach” (Bronstein 2019). Other aspects of the KaiOS phone design, moreover, warrant attention because they bear on the suitability of the product for these with low to middle incomes in poor countries (and hence the extent of subsequent diffusion amongst these groups). It is well to recall, in this regard, the basic KaiOS philosophy; that is, that whereas “Most companies are trying to make Internet-connected devices ever more powerful and capable–KaiOS went the other way. It rethought everything to keep the essential capabilities of the smartphones but strip out costs and preserve battery life for people who likely have spotty access to electricity” (Ovide 2019). Note too that the specific design changes to which I now refer, apply also to the phones recently introduced into many African countries by KaiOS Technologies in partnership with MTN and Orange, which sell at roughly US$20 apiece (which I shall discuss further in the next section). In particular, the body of a KaiOS phone is as basic as it gets. There’s no touchscreen, which tends to be the priciest smartphone component and a battery hog. The models that Orange sells … have a screen that’s less than half the size of the latest iPhones and controlled with an old-school keypad. The keys are made from the least expensive plastic possible … To save money, KaiOS also shrank the memory to about one-quarter or less that of the cheapest Android smartphone. That means the phones can handle only one task at a time … For some KaiOS models, Qualcomm Inc. refashioned an old version of its processor, the phone’s brain, at an estimated cost of about $3, compared with the roughly $50 version found in top-end smartphones. In total, KaiOS-powered phones are made from about $15 worth of parts–Apple Inc.’s top of the line iPhone has $390 worth of stuff. (Ovide 2019)
Nor, one should add, cost reductions were the only way in which KaiOS attempted to increase the appeal of its product to those seeking Internet connectivity but are unable to afford a smartphone. Another way, for instance, was to increase the relevance of the Internet to prospective buyers. In February 2019, for example, the company introduced a new pre-installed program called “Life”, which was intended to assist inexperienced users. To this end, what Life offers is a directory of curated content such as on woman’s empowerment, health, education, and agriculture. It also incorporates a Digital Skills app to develop skills such as Internet navigation, privacy, and security (Leung 2019a). It basically helps inexperienced users get started with the build-up of digital skills as they go.
16
2 The Smart Feature Phone Revolution in Developing Countries …
Here it is important to note the difference in price between the KaiOS phones in Africa and the JioPhone in India. For it is a difference that is likely to bear quite heavily on the patterns of diffusion in the two regions. In particular, while I have already noted that the MTN and Orange phones in Africa are sold for around US$20, I should also point out that the JioPhone is effectively given away by Reliance for nothing–this extreme behavior seems to motivated by Reliance’s desire to build up a base of support for the long run, while, at the same time, destroying a lot of the competition. Indeed, since the introduction of JioPhone, 6 competing firms have gone out of business (The Economist 2019). The larger point, as noted by Peter Richardson, Director of Counterpoint Research (2019a), is that Reliance is far from being a typical telecoms company in that its strategy hinges on developing a large base of users, which, in turn, calls for a vast amount of subsidization. “For more conventional telcos,” on the other hand, “The economics of serving marginal communities is not easy to make work” (Richardson 2019b). That is why the KaiOSbased phones in Africa may be more difficult to promote than the experience with the JioPhone in India.
2.2 The Diffusion of Smart Feature Phones While many of the design features of the JioPhone may have contributed to its subsequent success in India, few doubt the role in this of the effectively zero price. According to one Indian source, for example, “In the first quarter of 2018, Reliance JioPhone … held a 15% share of the global feature-phone market … The brand managed this in less than a year of its launch in July 2017, mostly on the back of its ‘effectively free’ 4G feature phones’ (Bhattacharya 2018, n.p., italics added). In fact, by the first quarter of 2018, the JioPhone had, with a 15% share of the world’s feature phone market, reached the top place globally among feature phone brands, as Table 2.2 shows. Moreover, projections for the years 2019–2021 show continuing rapid growth (see Table 2.3). Table 2.2 Market share of major feature phone brands (first quarter 2018) Brand
% share
Reliance Jio
15
Nokia HMD
14
ITEL
13
Samsung
6
Tecno
6
Others
46
Source Counterpoint Research (2018)
2.2 The Diffusion of Smart Feature Phones
17
Table 2.3 Projected global smart feature phone shipments Year
Shipments in millions of units
2017
20.8
2018
73.6
2019
97.5
2020
121.0
2021
152.0
Source Leung (2019b)
It needs to be recognized, though, that for all its phenomenal success thus far, the JioPhone is not the only vehicle through which the smart feature phone will be delivered to poor countries. One reason is that other telecommunication companies have also chosen to rely on the KaiOS OS, especially, but not only, in Sub-Saharan Africa. MTN and Orange, for example, two of the largest telecommunications firms in Africa, have recently formed partnerships with KaiOS to sell their smart feature phones in many countries in the region. At a price of around US$20, the two similar African brands offer users the chance to “leapfrog” smartphones and move directly to the Internet at a much lower price. As yet, however, no sales figures for these brands are available (they were only introduced in Africa in early 2019). What has been reported, though, is that MTN plans to sell 10 million of its new KaiOS-based phones in the next three years (Gilbert 2018). The degree to which the two African brands are successful after their recent introduction, will provide an important test of the viability of the smart feature phone concept, in contexts other than the artificial situation in India, where, as noted above, the JioPhone is effectively given away free because of massive subsidization by Reliance Industries. Few other firms, in India or Africa, however, command the level of resources that are required to emulate this feat. Much therefore remains to be seen in Africa where phones similar to the JioPhone are sold for around US$20. As already noted, it is premature to answer this question, though there are some who expect products such as Orange’s “Sanza phone” to do much to bring the Internet to poor African countries, where, up until now, this technology has been confined to only a very small segment of the population.8 According to GMSA, as per African Wireless’ (2019, n.p.) reporting, “the $100–200 price tag of a smartphone is preventing 64% of people in Africa from upgrading their phones to 3G/4G devices that can access the Internet.” On the other hand, research conducted by GMSA also suggests that the price threshold for making the transition from 2G to 3G/4G phones is US$34. “Below that point, even those in the lowest income groups are said to be capable of upgrading to data-enabled devices” (African Wireless 2019). According to this 8 Other
important ones, for example, are the availability of infrastructure and digital skills.
18
2 The Smart Feature Phone Revolution in Developing Countries …
calculation, therefore, the phones sold by MTN and Orange, should be affordable to many of the poor in Africa (though, of course, price is not the only determinant of adoption). Unfortunately, no research that I am aware of, provides detailed information on the status (income or education) of those who adopt smart feature phones in India or Sub-Saharan Africa. As such, it is difficult at this stage to describe these phones precisely in terms of appropriate technology, which, as noted above, has the potential to alleviate poverty. It could theoretically be the case, for example, that buyers are drawn from among the most affluent groups, seeking to possess a second mobile phone (apart that is from a smartphone). Or, equally, it might be that those who adopt smart feature phones are more from those in middle-income groups, with more education and better digital skills. In that case, it would be incorrect to describe the outcome in terms of Prahalad’s (2006) notion of “bottom of the pyramid.” Clearly, for both conceptual and policy reasons, there is a pressing need to conduct survey research that is capable of throwing some light on the socio-economic status of those who own the new smart feature phones. Certainly, from a thorough review of the academic and non-academic (corporate) literatures, and discussions with some industry workers, I was able to find no data on the socio-economic status of JioPhone users in India. What seems to be called-for instead are micro-based surveys of samples of users. There is, however, some information that strongly suggests involvement by the poor. Some of it pertains to the location of the JioPhone in India. In particular, survey research shows that in percentage terms, the share of the rural areas in the overall subscriber base of Reliance Industries grew from 4.25 two years ago, to over 32 in September of 2018 (Financial Express 2019a). This is a remarkable increase in the percentage share of the JioPhone in areas that are often known to be reluctant to accept new technology in general and information technology in particular (James 1989). In fact, according to a recent estimate (Financial Express 2019a), there are over 62 million subscribers to the JioPhone in the rural areas of India. Since it is in these same areas that the poor are concentrated, it is well-nigh impossible to imagine that the poor will have been entirely bypassed by the new technology. Moreover, since the majority of JioPhone buyers appear to be first time users of the Internet (Newsroom Post 2019), they must, prior to that, have been unable to afford a smartphone and as such, were unlikely to be among those with relatively high incomes. Consider in this regard, the actual case (reported by the Wall Street Journal) of a sidewalk vegetable seller in New Delhi, who earns roughly US$80 a month (putting him on a daily basis at around US$2.6 and not far above the World Bank poverty line. When he decided to replace his basic mobile device, he was unable to afford even the cheapest, bare-bones smartphones that cost around US$100. So, he paid about US$20 for JioPhone. With his new smart feature phone “he listens to Bollywood music on the job, using Google’s built in voice assistant to search for Hindi-language tunes on YouTube. At night his family crowds around the device to watch movies” (Purnell 2019, n.p.).
2.2 The Diffusion of Smart Feature Phones
19
Table 2.4 Activities undertaken on mobile Internet, based on selected usage in developing and developed countries (%) Use
Social networking
Play games
Order or purchase goods online
Access health information
Access government services
Developing countries
76
53
21
18
14
Developed countries
75
50
45
31
26
Source GSMA (2019, p. 34)
This example points to an important aspect of Internet use in developing countries, which is quite different than that in developed countries. Whereas talk about the use of the Internet in developing countries has tended to focus on staple development issues such as farmers seeking information on grain prices, women searching for information on maternity, and students eagerly signing up for online education, the reality has confounded these expectations (The Economist 2019). That is, “where people planning development strategies imagined, metaphorically at least, Blackberries providing new efficiencies and productivity, consumers wanted the chat, apps and games of the iPhone. Worthier uses tend to follow, but they are the cart not the horse” (The Economist 2019, n.p.). It is useful to examine these tendencies in terms of the most recent data provided by the GSMA (2019), as set out in Table 2.4. Whereas those living in developing countries show similar percentages as inhabitants of rich countries with regard to social networking and playing games, this is not the case with uses such as ordering online, accessing information about health, and accessing government services (or what may be called development activities). In a strict “consumer sovereignty” framework, these differing preferences of poor countries need, of course, to be respected. But one also needs to bear in mind that the relatively low use of development services such as education or e-government, may sometimes be less about preferences and more about the availability of such resources, especially in rural areas of poor countries. In any case, there is nothing sacrosanct about the conditions in which preferences are formed: whether, for example, they provide undue support to one use over another. Then, there is the need for recognition of a calculation made by the GSMA (2017a, p. 30) that “For those living below $2 per day, a $100 handset accounts for 14% or more of annual income. A phone in the range of $15–35 would be closer to the affordability threshold for this group”. This estimate is so important because KaiOS-based feature phones tend to be available for a figure of around US$20, that lies well within the said range of affordability for those with low incomes in developing countries (GSMA 2017a, p. 32). Consider, finally, that prior to the recent emergence of the smart feature phone, Internet connections were being made by relatively educated and
20
2 The Smart Feature Phone Revolution in Developing Countries …
high-income individuals in developing countries, using smartphones (PEW 2018). When the smart feature phone was introduced, such individuals would, in the main, have been reluctant to trade in their smart phones for an alternative that was better suited to meet the needs of those with much lower incomes. I conclude this section with an observation about smartphone diffusion around the world. Thus, “smartphone uptake across and within regions and markets is not balanced, risking leaving large populations without the means to come online” (GSMA 2017a, p. 2). Written in 2017, before the take-off of smart feature phones, this GSMA report goes on to provide useful data concerning the geographical pattern of imbalance in the distribution of smartphones. In particular, “Eastern Africa and South Asia are the regions lagging behind the most, with smartphone adoption levels as of mid-2017 at 25% and 30% respectively–much lower compared to the global average of over 50%. A major contributing factor to this inequality is the high rate of poverty. South Asia and Sub-Saharan Africa are home to the majority of the world’s poor people–India is a clear example of this, where–an average priced smartphone can cost up to 16% of income for poor and low-income groups” (GSMA 2017a, p. 2). Put another way, the point is that the opportunities for smart feature phones tend to be greatest in precisely these Internet-deprived regions, where the inhabitants most need it,9 but are least able to be connected. And, as we saw earlier, it is in these same areas (India and Sub-Saharan Africa) where the growth of smart feature phones tends to be concentrated.
2.3 The Impact of Smart Feature Phones What has been discussed so far falls into the category of the “first” digital divide, i.e. the differential extent to which Internet access is available in developed and developing countries. The case of India indicates that with the widespread adoption of smart feature phones, the Internet access gap will have narrowed and will narrow even more in the coming years (though these phones are also adopted to a limited degree in developed countries such as the USA) (Chokkattu 2018). The impact of Internet diffusion in India and other poor countries, however, does not depend only on the reduction of the “first” digital divide. For, much also depends on what is referred to as the “second” digital divide: that is to say, the differential degree to which analog complements are also available (World Bank 2016). I am referring here, for example, to literacy, numeracy, and digital skills. What matters, for welfare, that is to say, depends not only on access to technology, but also on what is done with it (Sen 1985).10
9 This 10 The
was also the case with basic phones as shown empirically by Waverman et al. (2005). idea has also been applied to education by Michaelowa (2001).
2.3 The Impact of Smart Feature Phones
21
Table 2.5 Types of digital skills Type of skill
Description
Basic
“Basic digital skills enable us to function at a minimum level in society–Basic skills cover hardware (for example, using a keyboard and operating touch-screen technology), software (for example word processing, managing files on laptops–and online operations (for example, e-mail, search, or completing an online form)”
Intermediate
“These are effectively job-ready skills since they encompass those skills needed to perform work-related functions such as desktop publishing, digital graphic design and digital marketing. For the most part, these skills are generic, meaning their mastery prepares individuals for a wide range of digital tasks needed to participate as engaged citizens and productive workers”
Advanced
“Advanced skills are those needed by specialists in ICT professions such as computer programming and network management–These include artificial intelligence (AI), big data, coding, cybersecurity, Internet of Things (IoT) and mobile app development”
Source ITU (2018, p. 6)
It is true, as discussed earlier, that KaiOS-based feature phones in India and Africa are designed specifically to overcome certain problems that make it difficult for poor, uneducated users to engage effectively with the Internet (such as the use of voice rather than text communications). But whether and to what extent these design features actually promote digital skills is not at all clear at this stage and the need for research is all too obvious. Consider, for the sake of argument, the definition of these skills as provided by the ITU (2018). As shown in Table 2.5, a distinction is drawn between three levels, ranging from basic to advanced, with standard in between. As far as I can judge, the 2 most advanced categories will be largely out of reach for many, if not most, of the rural, uneducated groups, that lack the contextual background and required skills set to perform, what for them, seem to be relatively complex tasks (recall, in this regard, the surveys noted above, which show the perceived inability to operate the Internet as a major source of non-use in several developing countries).11 Even in the case of basic skills, there appears to be a substantial gap between the developed and the least developed countries, as is evident from Table 2.6. Still more telling is the fact that as much as 80% of those in the least developed countries, lack the basic skills needed to operate the Internet (such as managing files, e-mail, search or filling in an online form). It has to be recognized, though, that in countries where KaiOS smart feature phones have spread widely, such as India, the number just cited will be lower, because the phones are themselves designed to reduce the constraints imposed by illiteracy and inadequate digital skills. Unfortunately, little or no research has been conducted on this important issue.
11 According
to the World Bank (2017), for example, “More than 80 percent of the entire working age population in Ghana and more than 60 percent in Kenya cannot infer simple information from relatively easy texts.”
22
2 The Smart Feature Phone Revolution in Developing Countries …
Table 2.6 The divide in digital skills (basic and standard skills) between developed and developing countries (%) in 2017 Type of digital skill
Developed countries (average proportion of individuals)
Developing countries (average proportion of individuals)
Basic
65
46
Standard
49
29
Source ITU (2018, p. 32)
Yet, a large-scale survey conducted by the GSMA (2017b), of “mobile user engagement” in 56 countries throws some indirect light on what one can expect. The basic idea, as expressed by the GSMA, is to calculate 2 scores for each country surveyed, namely, usage score (“the average number of mobile phone use cases adult phone owners engage in”), and frequency score (“how often they engage in the use case on average”) (GSMA 2017b, p. 4). The higher are the scores, the more engaged is the user. In fact, GSMA has constructed an index based on these scores, which, in a rough way, can be considered as indicators of digital skills. And since I am primarily interested in countries that are most lacking in digital skills, I list in Table 2.7 the bottom 10 countries on the mobile user engagement score. The scores associated with each country should be interpreted as follows. At the one extreme, a zero score indicates that, on average, the participants in a particular country never use mobiles for any of the 29 use possibilities mentioned in the survey. Conversely, a score of 10 means that, on average, users in a country participate in all such possibilities every day. Note that 7 of the bottom 10 countries are drawn from Sub-Saharan Africa, and one, India, was the subject of our previous discussion on JioPhone. These countries, Table 2.7 Countries with lowest engagement scores Country
Engagement score
Ivory Coast
1.4
Cameroon
1.3
Nigeria
1.3
Egypt
1.1
India
1.1
Uzbekistan
1.0
Sierra Leone
0.9
Myanmar
0.8
Pakistan
0.8
DRC
0.6
Ethiopia
0.5
Source GSMA (2017b)
2.3 The Impact of Smart Feature Phones
23
it seems, are especially likely to be plagued by problems of digital skills, when their inhabitants come to use smart feature phones. And one other country in the table, Myanmar, warrants special mention in this regard, because although smartphone ownership is relatively high, users have gained only very limited benefits. More specifically, “In Myanmar, >50% of unique subscribers own a smartphone but their user engagement pattern is below the developing world average. Just over 60% of smartphone users in the country claim that their usage is prevented by the fact that they have trouble understanding how mobile Internet applications, websites or e-mail work on a mobile phone” (GSMA 2017b, p. 11). What is probably lacking in this case of a divergence between the “first” and “second” digital divides, is a severe shortage of literacy and digital skills among the population. Recall, in this regard the substantial gap in the availability of such skills between the developed and least developed countries (which include Myanmar and many African countries), shown in Table 2.7. A major question then arises as to whether and to what extent KaiOS-based smart feature phones in effect lessen the digital skills problem in the group of countries with very low incomes. For, as I have noted at several points in the preceding text, such phones are designed to simplify and enhance the access to and use of the Internet, by, for example, including a wider range of languages. But exactly how far these features of the new phones will help to narrow the gap in the demand and supply of digital skills, is as yet, anything but clear (and requires more and more detailed research). What does seem clear though, is that, some degree of digital skills will be needed, especially in the poorest developing countries, since the features embodied in the KaiOS-based phones are not perfect substitutes for human skills (certainly not across the board). And I would venture to suggest, moreover, that digital skills will need to be promoted to a far greater degree than they have been in most developing countries up until now. This is mainly because, as shown above, most poor developing countries are especially disadvantaged with regard to digital skills, which are which are becoming increasingly important for welfare of the individual and also the society as a whole (ITU 2015). Moreover, there is some suggestion that it is especially rural areas that bear the brunt of the lack of these analog complements. Note, too, and relatedly, that the absence of digital capabilities will tend to be most acutely felt by first-time users of the Internet, who will not yet have had the chance to learn by doing. Yet, for all the benefits that are imparted by the acquisition of digital skills, governments in developing countries have, by and large, done precious little to promote them. In part, this is due to the way in which, at the implementation stage, there is often still a focus on access, rather than analog complements (i.e., on the “first” rather than the “second” digital divide). It may also be a part of the same “crisis in learning,” which has caused literacy and numeracy to stagnate over the past few decades (World Bank 2018c). In any event, the situation has prompted the World Wide Web Foundation (2017) to argue that “it’s time to prioritize digital literacy.” Such a view is underscored by the most recent data from the GSMA (2019), which
24
2 The Smart Feature Phone Revolution in Developing Countries …
indicate that for four major regions of the developing world, a lack of literacy and digital skills, formed the main barrier to mobile Internet use. One step towards redressing this situation has been taken by KaiOS itself, based on the recognition (noted above) that the lack of skills is felt most acutely by first-time users of the Internet. First-time Internet users will probably find the Life application on digital skills to be especially useful because it offers them the chance to learn the rudimentary aspects of the Internet. In this way, too, the design of the KaiOS-based feature phone itself contributes to a reduction in the “second” digital divide between developed and developing countries.
2.4 Conclusions Just a few years ago, there were three categories of mobile phone ownership in developing countries, namely, owners of basic devices (without Internet connectivity), owners of smartphones, and those who owned no mobiles at all (who might nevertheless have some means of going online through kiosks, telecentres, and other shared facilities). Not long, thereafter, however, a new category was needed to account for an innovative alternative–the smart feature phone. In this paper, I have sought to address the questions: What shaped the development of the new technology? What has been the extent and nature of the adoption? What impact it has had on those with relatively low incomes? What I find in relation to the development phase of the new phone is that absolutely no effort was spared in the effort to reduce costs, without losing any essential functionality. (This was reflected partly in the use of a modified form of open-source software.) In this sense, and given the vast numbers of low-income people it has apparently already reached, the new smart feature phone represents one of the most successful cases yet known, of what is known as appropriate technology (the more so, indeed, because of the range of benefits that the Internet offers).12 But the KaiOSbased version of the technology also seeks to make phones easier to use, by, for example, enabling speech to replace text as a means of communication and by adding applications that help (first time) users to find relevant and helpful information (as is the case, for example, with a KaiOS application called Life). Finally, the said phone is or has already been introduced in regions most in need of it, namely, India and Sub-Saharan Africa. Thus it was that the place of origination of the technology influenced both the pattern and extent of adoption and this, in turn, bore on the benefits that were derived. I was at pains to emphasize, however, that the actual benefits derived from the Internet by low-income individuals, depend heavily on the digital skills they possess, a function, among other things, of the programs that are set in place. 12 Note, though, the caveat noted above to the effect that it is still not known what percentage of the poor actually adopt the new phones. More generally, the benefits of the Internet in developing countries are described in numerous sources such as Deloitte (2014).
2.4 Conclusions
25
To conclude, I refer to various points in the text, where note has been taken of what we do not know about the development, diffusion, and impact of the new technology. It is not of course my intention here to repeat all such gaps in our knowledge, but rather to revisit only a few of the most important of them. It is not known, for example, with any certainty, who the users of the KaiOS based feature phones really are: whether they have high or low-incomes; where they live (in rural or urban areas), and how much education they have received (though some indirect evidence is available). As such, it is difficult to gauge with any degree of precision, the impact of the new technology on poverty and inequality in developing countries. Then too, little is known about the extent to which users of the new technology actually derive the gains that are available to them from the Internet. Recent data suggest that low-income users have a preference for entertainment rather than more traditional development goals. I have also shown, for example, that there is a very large gap in digital skills between the rich and the least developed countries. Will this gap result in differences in the extent to which gains are realized by those living in the two groups of countries? To what extent do the new smart feature phones reduce the need for minimum competencies in literacy and numeracy?13 Or, for that matter, will digital skills themselves be reduced? All these questions require far more survey research than they have hitherto received.
References African Wireless (2019) New ‘smart feature phones’ aim to address affordability issues for mobile users in Africa. Available http://www.africanwirelesscomms.com/news-details?itemid= 2508. Accessed 2 May 2019 Bhattacharya A (2018) In just 10 months, Reliance Jio’s become the world leader in feature phone. Quartz India, May 25. Available https://qz.com/india/1288920/reliance-jio-has-becomethe-world-leader-in-feature-phones-in-just-10-months/. Accessed 17 Jan 2019 Bronstein M (2019) Building the Google Assistant on phones for everyone, everywhere. Available https://www.blog.google/products/assistant/building-google-assistant-phoneseveryone-everywhere./. Accessed 2 June 2019 Chokkattu J (2018) A flip phone with Google Maps? KaiOS is making dumb phones smarter. Digital Trends, March 2. Available https://www.digitaltrends.com/mobile/kaios-smarter-featurephones/. Accessed 4 Nov 2018 Counterpoint Research (2018) Global Handset Tracker. Available https://www.counterpointres earch.com/jiophone-takes-top-spot-global-feature-phone-market-q1-2018/ Counterpoint Research (2019) Smart feature phones to create US$28 billion revenue opportunity. Available https://www.counterpointresearch.com/smart-feature-phones-create-us28-bil lion-revenue-opportunity/ Deloitte (2014) Value of connectivity: Economic and social benefits of expanding Internet access. Available https://www2.deloitte.com/content/dam/Deloitte/ie/Documents/TechnologyMedia Communications/2014_uk_tmt_value_of_connectivity_deloitte_ireland.pdf. Accessed 23 Mar 2020
13 I
showed that the text to speech application on the JioPhone reduces the need for literacy among the poor who are unable to reach minimum levels of this capability.
26
2 The Smart Feature Phone Revolution in Developing Countries …
Dey R (2019) KaiOS everything you need to know about this operating system. Available https://www.blogpoke.com/kaios-everything-you-need-to-know-about-this-operatingsystem/. Accessed 16 May 2019 Financial Express (2019a) JioPhone takes Reliance Jio to hinterland JiO India. Available via https://www.financialexpress.com/industry/making-inroads-jiophone-takes-reliance-jioto-hinterland/1728398/. Accessed 23 May 2020 Financial Express (2019b) Indian R&D is pumping iron, must make policy choices carefully. Available https://www.financialexpress.com/opinion/indian-rd-is-pumping-iron-must-make-pol icy-choices-carefully/1580688/. Accessed 3 June 2019 Gilbert P (2018) MTN to sell ‘smart feature phones’ early next year. Available https://www.itweb. co.za/content/4r1lyMRoVp4qpmda. Accessed 4 June 2019 GSMA (2017a) Accelerating affordable smartphone ownership in emerging markets. Available https://www.gsma.com/mobilefordevelopment/wp-content/uploads/2018/08/Acceleratingaffordable-smartphone-ownership-in-emerging-markets-2017_we.pdf. Accessed 22 March 2020 GSMA (2017b) GMEI 2017: Global mobile engagement index. Available https://www.gsmaintel ligence.com/research/?file=e4549aeda553ac832ff9126c7d6c0861&download. Accessed 5 Aug 2018 GSMA (2019) The mobile economy: Sub-Saharan Africa. Available https://www.gsma.com/mob ileeconomy/sub-saharan-africa/. Accessed 4 Feb 2020 Hartmann A, Linn J (2008) Scaling up: A framework and lessons for development effectiveness from literature and practice. Brookings Institution, Washington, DC International Labour Organization (ILO) (1992) The impact of world employment research on technology. Geneva International Telecommunication Union (ITU) (2015) Measuring the information society report International Telecommunication Union (ITU) (2018) Digital skills toolkit James J (1989) Improving traditional rural technologies. Macmillan, London Leung E (2018) A short history of KaiOS. Available https://www.kaiostech.com/short-historykaios/. Accessed 4 Feb 2020 Leung E (2019a) Life, a new initiative by KaiOS to help first-time Internet users make the most of mobile Internet access (press release). Available https://www.kaiostech.com/life-a-new-initiativeby-kaios-to-help-first-time-internet-users-make-the-most-of-mobile-internet-access/. Accessed 1 June 2019 Leung E (2019b) The birth of the smart feature phone revolution. Available https://www.kaiostech. com/the-birth-of-the-smart-feature-phone-revolution/ Michaelowa K (2001) Primary education quality in Francophone Sub-Saharan Africa: Determinants of learning achievement and efficiency considerations. World Dev 29(10):1699–1716 Newsroom Post (2019) Jio launches digital literacy program ‘Digital Udaan’ for first time internet users. Available https://www.newsroompost.com/?s=Jio+launches+digital+literacy+program+% 60Digital+udaan%E2%80%99+for+first+time+internet+users. Accessed 5 Nov 2019 Ovide S (2019) The next big phones could bring a billion people online. Bloomberg Businessweek, June 7. Available https://www.bloomberg.com/news/features/2019-06-07/the-next-big-phonescould-bring-a-billion-people-online. Accessed 7 July 2019 PEW (2018) Social media use continues to rise in developing countries but plateaus across developed ones. Available www.pewresearch.org/global/2018/06/19/social-media-use-continues-to-rise-indeveloping-countries-but-plateaus-across-developed-ones/. Accessed 23 Mar 2020 Prahalad CK (2006) The fortune at the bottom of the pyramid. Pearson, London Purnell N (2019) The hottest phones for the next billion users aren’t smartphones. The Wall Street Journal, July 23. Available https://www.wsj.com/articles/the-hottest-phones-for-the-next-billionusers-arent-smartphones-11563879608. Accessed 23 Mar 2020 Richardson P (2019a) Counterpoint Research, personal communication. May 21, 2019 Richardson P (2019b) Counterpoint Research, personal communication. May 22, 2019 Schumacher E (1973) Small is beautiful: a study of economics as if people mattered. Blond and Briggs, London
References
27
Sen A (1985) Commodities and capabilities. North-Holland, Amsterdam Silver L, Johnson C (2018) Majorities in Sub-Saharan Africa own mobile phone, but smartphone adoption is modest. PEW Global Attitudes and Trends. Available https://www.pewresearch.org/ global/2018/10/09/majorities-in-sub-saharan-africa-own-mobile-phones-but-smartphone-ado ption-is-modest/. Accessed 19 July 2019 Singer H (1970) Dualism revisited: a new approach to the problems of the dual society in developing countries. J Dev Stud 7(1):60–75 Singh A, Pandey S (2005) Rural marketing: Indian perspective. New Age International, New Delhi Stewart F (1977) Technology and underdevelopment. Macmillan, London Summers N (2019) How KaiOS claimed the third-place mobile crown. Engadget, February 26. Available https://www.engadget.com/2019/02/26/kaios-third-mobile-operating-sys tem/. Accessed 26 Feb 2019 The Economic Times (2017) ‘Zero’ price JioPhone to hurt telcos, cable operators alike. Available https://economictimes.indiatimes.com/markets/stocks/news/zero-price-jiophone-to-hurt-tel cos-cable-operators-alike/articleshow/59700013.cms. Accessed 19 Oct 2019 The Economist (2019) A global timepass economy. The Economist, June 8. Available https://www. economist.com/briefing/2019/06/08/how-the-pursuit-of-leisure-drives-internet-use. Accessed 23 Mar 2020 Wang Y (2019) The birth of the smart feature phone in India, KaiOS. Available https://www.kaiost ech.com/the-birth-of-the-smart-feature-phone-revolution/. Accessed 21 May 2019 Waverman L, Meschi L, Fuss M (2005) The impact of telecoms on economic growth in developing countries. Vodafone Policy Papers Series 2:10–24 World Bank (2016) World development report, Washington, DC World Bank (2017) Skills development. Available https://www.worldbank.org/en/topic/skillsdev elopment Accessed 22 Mar 2020 World Bank (2018a) Ending extreme poverty and sharing prosperity: progress and policies, Washington, DC World Bank (2018b) World development report. World Bank, Washington, DC World Bank (2018c) World Bank education overview: skills. Available http://documents.worldbank. org/curated/en/806751541081039061/World-Bank-Education-Overview-Skills. Accessed 19 Jan 2019 World Wide Web Foundation (2017) Why it’s time to prioritise digital literacy. Available https:// webfoundation.org/2017/09/why-its-time-to-prioritise-digital-literacy/. Accessed 15 July 2018
Chapter 3
Smart Feature Phones and Welfare in Poor Developing Countries
Abstract Introduced first in India a few years ago, a major innovation known as the smart feature phone has already brought the Internet to tens of million people in that country. It is affordable even to some of the poor in poor countries, such as those living in Africa, where it sells for around US$20. This price was made possible by a series of design changes, such as the use of open-source software, which distinguish the product from the much more expensive smartphones. Yet, in spite of the vast potential that the new technology brings to developing countries, the actual benefits that accrue to the poor citizens of those countries depend (especially in rural areas) not only on economic mechanisms, but also those of a cultural, social, as well as institutional kind. Such factors range from issues of relevance and localization, to linguistics, the form of communication (whether text or speech), and the extent of literacy and digital skills. What I find is that there are indeed cases where smart feature phones help to promote the degree to which the benefits from the Internet are realized by the poor. Because it substitutes voice for text, for example, Google Assistant allows even the illiterate to participate online and it also provides the highest number of local languages, in India, than any other competing voice-assistant. Then, there is the encouragement given to local developers to submit what one hopes will be locally relevant applications and content that includes live livestock tracking and mobile payments. In other areas, however, such as the severe lack of digital skills to operate the Internet effectively, far too little is being done by governments or private actors, though there are a few notable exceptions. Keywords Information technology · Poverty · Digital skills · India · Geography This chapter is about the smart feature phone, a recent innovation that bridges the gap between basic mobile phones and smartphones. The phone was explicitly designed to bring the Internet to the majority of those living in poor countries, especially in rural areas. As such it portends a sea change in the role of information technology in development. In fact it has already been adopted by tens of millions of those residing in India and has just been introduced in Africa as well, where it sells at a price of around US$20. However, in order to derive the maximum benefits from this new technology, I argue below that issues of relevance, local content, literacy and digital skills, and © The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 J. James, The Impact of Smart Feature Phones on Development, SpringerBriefs in Economics, https://doi.org/10.1007/978-3-030-62212-1_3
29
30
3 Smart Feature Phones and Welfare in Poor Developing Countries
the availability of local languages also need to be considered. As such the chapter poses the question of how a major new innovation impinges on the welfare of the poor in developing countries, with particular reference to India. I will be discussing more specifically KaiOS-type smart feature phones (which are based on open-source software), and which have already achieved such spectacular rates of adoption in one large developing country, India (the new technology also looks likely to succeed in Sub-Saharan Africa, where, until recently, the only means of connection to the Internet has been the smartphone). I have not yet, however, addressed the key question of how smart feature phones of the kind noted above, are likely to impinge on the welfare of those living in poverty in poor countries (improved access, one should note, provides only partial help in answering this question, for it is only a necessary condition for a positive welfare effect). Not only, I suggest, does one need to consider relevant economic mechanisms, but also, a range of other factors—cultural, social as well as institutional—that determine what the poor actually gain from the new technology. Such non-economic factors range from issues of relevance and localisation, to linguistics, the form of communication (whether it is text or speech) and digital literacy. Then there is a major geographical issue: the large gap that exists in mobile Internet access between rural and urban areas. In particular, the former areas are 40% less likely to use mobile Internet than the latter (GSMA 2019). To begin with, then, I deal briefly with some of the economic arguments, before moving subsequently to the other, relatively neglected factors that have just been mentioned. More generally, though it is important to note that this chapter belongs to the genre of an essay, in the sense that it comprises a short form of literary composition, with a main focus of bringing awareness to the reader, of a particular and important topic, which has this far attracted little or no academic attention.
3.1 Economic Mechanisms There can be little doubt that the economic impact of the smart feature phone on poverty will depend on the extent to which it precipitates an increase in growth, whether this is due, for example, to an increase in productivity, a reduction in transactions costs or improved information and communications. Many of these mechanisms can be thought of in terms of a reduction in market imperfections in developing countries, especially those that have to do with information. For, although these imperfections also exist in developed countries, they are generally much less acute than in developing countries. This is mainly because information in the latter countries is far less freely available than in the former, in markets for labour, finance, inputs or outputs. The result, inevitably, is inefficiency and a reduction in growth. Cases of ‘information asymmetry’, for example, where one party to a transaction possesses much more information than the other, may not take place at all (as occurs, classically, in the market for used automobiles) (Akerlof 1970). Such an outcome,
3.1 Economic Mechanisms
31
we should note, is especially likely in cases where trust is in especially short supply. As is perhaps obvious, the Internet, by providing hitherto missing information, has the potential to redress the market imperfections that are so rampant in developing countries and especially those lacking in adequate infrastructure. However, the effect on poverty also depends on the equality of income distribution in a particular country or region. The more equal this is, the greater will be the effect of a given increase in output on the poor and vice versa.1 (This mechanism, one should note, assumes particular importance in light of the fact that there are substantial differences in the degree of equality across developing countries.2 ) What deserves particular emphasis in the context of the economic mechanisms noted above, is that there are reasons to think they will be especially propitious in the context of rural areas in developing countries, where the poor tend to be concentrated. One of them has to do with the geographical isolation of remote areas, which means, for example, that they will generally be particularly responsive to the distance – shrinking effects of the Internet. Thus, the effect of extending Internet access could be particularly important for rural communities. Constraints on the flow of information have limited these communities access to wider markets and to a variety of employment opportunities. Access to mobile and Internet-based applications can extend the range of business services that become available to these communities. Internet access is also valuable to rural development-oriented organisations that act as local communication conduits or intermediaries. (Deloitte 2014, p. 8)
Thus it is, that the Internet holds out particular economic promise in the very same regions where the poor are mostly to be found.3 As far as countries go, moreover, smart feature phones are being introduced mainly in those with especially low per capita incomes, such as India and Sub-Saharan Africa. The latter region, we should note, contains no fewer than 50% of the world’s extremely poor people (IFC 2017). Insofar as it is in these countries (and especially their rural areas) where transactions costs are most severe, information the most scarce and productivity the lowest, that the provision of mobile phones will be most deeply felt. For example, the more onerous is the travel needed to gather information (because, say, of a lack of public transport), the greater will be the gains afforded by the Internet. It is no doubt partly because of this same logic that Waverman et al. (2005) found that mobile phones have double the impact on growth in developing, as opposed to developed countries. Or, for that matter, why survey evidence in developing countries often finds that the gains from mobile telephony accrue more to the poor than the rich (James 2016; Bayes et al. 1999).
1 See
Roemer and Gugerty (1997). the latest estimates on the Gini-coefficient provided by the World Bank data series. 3 Bird et al. (2010) focus on the geographical aspects of communication and they find a strong correlation between isolation and poverty. 2 See
32
3 Smart Feature Phones and Welfare in Poor Developing Countries
It is well to note too that there is an element of leapfrogging in the way that these and other developing countries are bypassing the smartphone and computers as a means of gaining access to the Internet. They are moving instead to smart feature phones that are affordable to a majority of the population, in a similar way to the earlier period when fixed-line phones were eschewed in favour of basic mobile phones (see more below on the particular case of KaiOS smart feature phones). What is also worth noting is that small-scale firms and farms may often play an especially important role as regards the economic impact of the Internet on poverty. According to Deloitte, for example, small and medium-sized enterprises (SMEs) in developing countries are amongst the biggest winners from receiving access to the Internet. By reducing transactions costs and the constraints of distance, throughout the world, the Internet has reduced barriers to market entry and allowed SMEs to innovate and reach a broader market. In countries as diverse as Mexico, Malaysia and Morocco, SMEs with Internet access have been found to have experienced an average 11 productivity gain (Deloitte 2014, p. 3). Indeed, according to many other sources, the growth of SMEs in poor countries contributes to poverty reduction.4 And given the equally broad consensus that the lack of finance is a crucial constraint on the adoption and use of mobile phones, the availability of mobile money on KaiOS-based smart feature phones, also serves mainly the well-being of the poor, who often cannot gain access to the formal banking system, with all its acute market imperfections.
3.2 Smart Feature Phones and Other Dimensions of Poverty What matters to the poor in developing countries is not just about income, but also the usefulness and relevance of the content on the Internet (whether, for example, it is local or global); whether it is available in local languages; the relevance of the form (text vs. voice) and the availability of digital skills, which all determine how much of the Internet can actually be enjoyed by the user. Indeed, especially in its early form, the Internet exhibited all the features of global technological dualism,5 with the gains of the technology accruing mainly to the rich countries, while the needs of less affluent countries went largely neglected. In such a context, Internet content designed for developed countries, tends to exhibit limited relevance (for the poor) in less developed ones; there is quite a marked degree of reliance on English, a neglect of popular languages in poor countries, and a high degree of dependence 4 Survey
results reported by Samuel et al. (2005, p. 51) show that ‘In South Africa, mobile phones were the only source of communication for a large number of small businesses run by black individuals’. 5 This concept was explored first by Singer (1970).
3.2 Smart Feature Phones and Other Dimensions of Poverty
33
on digital skills, which are relatively abundant in the rich countries but not the poor (see below). In what follows, therefore, I deal with the respects in which smart feature phones help to overcome the problems for the poor, associated with technological dualism, that have just been described. Note, at this stage, that the CEO of KaiOS has gone on the record to state that the firm’s goal is to connect to the ‘next billion users’ (Codeville 2019). In India, this firm partnered with a very large local firm, Reliance Industries, to produce the well-known JioPhone. More generally, the KaiOS smart phone has already gone some way to challenge the extent of international technological dualism.
3.3 Localisation and Relevance of Internet Content At a time when the Internet was far too expensive for individual ownership by the poor in developing countries, one of the popular solutions then on offer made use of communal institutions such as kiosks, telecentres and community radio stations. In some such cases, the relevance and local permeation of the Internet were made possible by the presence of an intermediary who was familiar with both the technology and the needs of the local community. The radio ‘browsing’ programmes at Kothmale Community Radio in Sri Lanka provide probably the most interesting example of how local Internet relevance was achieved in a communal setting where radio programmes and their guests acted as intermediaries between the Internet and the listeners. More specifically, the community radio station broadcasts a daily ‘Radio Browsing the Internet’ programme, and in this programme, the broadcasters, supported by resource personnel, browse the Internet on-air together with their listeners and discuss and contextualise information in local language. The radio programme thus contributes to raise awareness about the Internet in a participatory manner. The listeners request the broadcasters to surf the Web on their behalf and the programme transmits information in response to their requests. This information is explained and contextualized with the help of the studio guests, for example: a local doctor may explain data on a health website. (UNESCO Project Documents) (Pringle and David 2002, p. 2)
Recently, however, the advent of affordable smart feature phones allows even the poor to access local information directly rather than indirectly via intermediaries (subject of course to the availability of digital skills, which are discussed below). In the case, specifically, of KaiOS-based phones, this localisation of information occurs mainly through the applications that come attached to the devices, or, are available through the Jio or KaiOS application stores. ‘News’, for example, is an application from the Kai Store which covers 44 different regions and languages and allows users to personalize their choice of subject from a range of 12 possibilities (Leung 2019). These stores, in fact, may be instrumental in providing more local content to the information usually imparted by the Internet, in that they encourage the submission of work by local developers. Thus, ‘with a new vehicle to reach a previously untapped market, the door opens in terms of generating more downloads and engagement,
34
3 Smart Feature Phones and Welfare in Poor Developing Countries
and offering hyper-tailored services that meet the needs of local communities. In Africa, for example, localized content includes live livestock tracking and mobile payments have been highly impactful, and there is room for even more solutions’ (Wang 2019). More generally, it is worth noting that representatives of KaiOS and MTN have explicitly stated the need for local content in their smart feature phones. The former company, for example, has declared its dedication ‘to working with independent app developers to offer localized content that meets the unique needs of the African region’ (Codeville 2019). Furthermore, KaiOS offers a developer portal which ‘includes information, development tools and best practices, to make the development process as easy as possible. The company also partners with local coding schools to provide workshops and training focused on local app development’ (Codeville 2019). It is important to note, however, that it is far from certain whether local developers will produce applications that are suitable for the poorer, more isolated members of developing countries, even though many such persons now form part of the market for locally-oriented products (by virtue of owning ultra low-cost smart feature phones). It is entirely possible, for example, that local developers will produce applications which attempt to merely mimic the globally-oriented material from developed countries. On the other hand, though, one might reasonably suppose that these developers would have some sort of comparative advantage in products that are locally-oriented. Clearly, this is an area which requires a good deal of research, since I know of no other work on the topic. Then, as far as the JioPhone is concerned, one example consists of an application called Jio Express News, which promotes local interests in that it is available in ten Indian languages and allows users to select news material from a very wide range of options according to their own local interests (note, in this regard, that I deal with linguistic issues under a special heading below). Finally, I take note of another application from Reliance Jio, which is directed at providing information on a major Hindu religious event, the ‘Kumbh Mela’ festival.6 It is local in the sense that it is peculiar to India and directed at a particular segment of the Indian population, though again, it is not clear to what extent this group actually includes the poor. What is better known is that the application provides much useful information about the festival, which most likely raises the wellbeing of those who use it. What I conclude, therefore, is that although KaiOS and its partners in India and Africa, are given to making statements about the need to make their smart feature phones relevant to local circumstances, they provide only a few examples of where this has actually occurred (as far, at least, as I was able to discover).
6 See
the official website at www.kumbh.gov.in/en.
3.4 The Linguistic Divide, Smart Feature Phones and the Poor
35
3.4 The Linguistic Divide, Smart Feature Phones and the Poor The gains from the Internet depend heavily on the languages in which Web content is available, in relation to the languages that are most commonly spoken by the poor. Table 3.1 shows the 10 languages that are most commonly used on websites. The fact that English occupies the leading position in the table is certainly not one of its most striking features. For, historically, that language has been much more dominant on the Internet, than it is now. According to one estimate (UNCTAD 2019), for example, in the mid 1990s, English made up no less than 80% of the Internet content. As Table 3.1 shows, that percentage has since then shrunk to 25.9% (though the top ten languages still account for almost 77% of the world total). Note, however, that the beneficiaries of the reduced share of English are not necessarily drawn from low-income countries. In fact, the languages shown in Table 3.1 are very largely associated with high or middle-income, rather than poor countries and especially the rural areas of those residing in the last-mentioned category. SubSaharan Africa, for example, contains an especially high proportion of countries whose languages are rarely mentioned on the Internet. As a result, the benefits derived from this technology are reduced. Even in Gabon, a middle-income country in Sub-Saharan Africa, daily communications still take place in the vernacular languages in rural areas, ‘which leads to a phenomenon of Internet underutilization by rural households. This fact remains even though the Internet and ICT in general provide a platform to communicate, Table 3.1 Most common languages used on the Internet as of January 2020, by share of Internet users (descending order) Language
% of websites using the language
English
25.9
Chinese
19.4
Spanish
7.9
Arabic
5.2
Indonesian/Malaysian
4.3
Portuguese
3.7
French
3.3
Japanese
2.6
Russian
2.5
German
2.0
Top 10
76.8
Rest of languages Total Source Statista (2020)
23.1 100
36
3 Smart Feature Phones and Welfare in Poor Developing Countries
learn, manage and disseminate knowledge, especially in the areas of agriculture, education and health in order to promote sustainable socio-economic development’ (World Bank 2014, n.p., emphasis added). Another study reported in the same year, drew attention to the particular need for local languages in the rural areas of developing countries (D’Monte 2014). While the emphasis is on the effect of such languages on the adoption of the Internet, it is of course true that the benefits from use of this technology, depend on access to it (in the first instance). And in relation to rural India, the study reports that ‘43% of the non-users of the Internet said they would adopt the medium if the content was provided in local language. In urban areas, 13.5% of the non-users mentioned that they would use the Internet if content is provided in local languages’ (D’Monte 2014, n.p.). Indeed, ‘The report identifies local language as the single largest driver of Internet growth in rural areas’ (D’Monte 2014, n.p.). All of this points to a severe linguistic divide on the Internet, which reduces the amount of the Web that is available to people in poor as opposed to rich countries, and to this extent reduces the benefits that the technology potentially offers to persons living in the former. The question I now seek to address is whether and to what extent KaiOS-based smart feature phones bring more linguistic diversity to the Internet than would otherwise have been the case, i.e. when the only means of individual access to the Internet was through expensive smartphones. Recall, to begin with, the point made earlier about how KaiOS Technology and some of its partners, have expressed a keen interest in providing smart feature phones to a wider swathe of the population in developing countries, including those parts that are isolated and marginalized. Thus, if we look at the JioPhone in India as an example, and its Google Assistant in particular, we find that this application is available in at least 9 Indian languages (including Hindi, Bengali, Tamil, Gujarati and Urdu).7 In so doing, Google has extended the reach of this application to millions of users of the JioPhone in India.8 By contrast, well-known competitors in voice recognition software, such as Microsoft’s Cortana, Amazon’s Alexa, Apple’s Siri and Samsung’s Bixby, concentrate (with the partial exception of Siri) on a limited range of mostly rich country languages. These, after all, are the languages spoken in their main markets. Table 3.2, accordingly, compares the languages that are available on the voice assistants that have just been mentioned. Thus, with the exception of China, a high middle-income country, the three assistants other than ‘Siri’, cover a very limited number of (almost entirely) developed countries (which also overlap to a marked degree between the assistants in question). ‘Siri’, by contrast, offers a much longer list of countries, including a few that would be described as developing. But it does not include any from the category of
7 The
nine languages are given by Anwer (2019).
8 Note that many of those who speak only regional languages are likely to be drawn from the poorest
(uneducated) groups in India, who will, to this extent, benefit especially heavily from the Google Assistant.
3.4 The Linguistic Divide, Smart Feature Phones and the Poor
37
Table 3.2 Languages on selected voice assistants Amazon Alexa
Apple Siri
Microsoft Cortana
Samsung Bixby
English
Arabic
Chinese (simplified)
English
French
Chinese
English
Chinese
German
Danish
French
German
Italian
Dutch
German
French
Japanese
English
Italian
Italian
Portuguese (Brazilian)
Finnish
Japanese
Korean
Spanish
French
Portuguese (Brazilian)
Spanish
Hindi
German
Spanish
Hebrew Italian Japanese Korean Malay Norwegian Portuguese Russian Spanish Swedish Thai Turkish Source Based on Wiggers (2019)
low-income developing countries where the majority of the world’s poor tend to be concentrated (such as, for example, those residing in Sub-Saharan Africa). That the results shown in Table 3.2 differ quite markedly from what is available from the Google Assistant on the JioPhone has much to do with the fact that while the voice assistants shown there belong to smartphones, the Jio is a smart feature phone designed for those with relatively low incomes (most notably in Sub-Saharan Africa and India). Smartphones, on the other hand, are much more expensive and cater mainly to those in the developed countries (as shown in the entries for Table 3.2). For its part, WhatsApp goes even further than the Google Assistant in accommodating Indian languages on the JioPhone. In fact, the former supports no fewer than 24 of India’s indigenous languages and was again specifically designed for use on the Jio. It is expected that by the end of 2019, most KaiOS phones will be automatically equipped with the WhatsApp technology. It is of course difficult to say how many more Indians were able to use the Internet on the JioPhone as a result of the attention its applications paid to local languages, but what can be said is that the inclusion of Hindi alone would have covered more than 45% of the population (some of whom, however, could also have spoken English as
38
3 Smart Feature Phones and Welfare in Poor Developing Countries
a second or third language).9 Recall, too, in this regard the problems encountered by non-English users of the Internet in India, with regard to the English keyboard. And note that the JioPhone offers a way round these difficulties by allowing a choice of language for the operation of the keyboard. I turn finally to the many Anglophone and Francophone countries in Sub-Saharan Africa, in which MTN and Orange have recently introduced KaiOS smart feature phones. As far as I am aware, however, the Google Assistant available on these phones, contains no indigenous African languages. This, in my view, probably has to do partly with the relatively small markets in the countries concerned and partly also with the fact that in many of them, English or French (or both) are widely spoken. In the case of Cameroon, for example, 57.6% of the population speak French and 25.2% speak English (World Atlas 2017). A reasonable summary of this discussion might thus be that while the Google Assistant on the JioPhone in India has gone some way towards redressing the acute language divide that was found to exist, much remains to be done in other poor countries. It is to be hoped, for example, that indigenous African languages will be added to the smart feature phones that have recently been introduced (as noted above) by MTN and Orange, two very large telecommunications companies. There is also much that is unknown in this area, which requires extensive survey research. To what extent, for example, do the poor in developing countries manage to use the Google Assistant even when it does not offer their native languages? To what extent does this group benefit from voice as opposed to textual communications? To what extent do local developers produce applications that are locally relevant?
3.5 The Lack of Digital Skills, Countervailing Policies and the Poor in Developing Countries The final issue that determines what, if anything, the poor gain from the smart feature phones discussed above, is the availability of digital skills among this group. For, although, there are activities on the Internet that are undemanding in terms of such skills, there are many others which require at least basic levels of them (where basic refers to activities such as operating a keyboard and touch-screen technology, word processing, managing files, e-mail, search and completing an online form) (ITU 2018). Unfortunately, however, what few data are available for India and Sub-Saharan Africa, do not paint a very promising picture in this regard. For example, according to the World Bank (2019), digital literacy is taught in only 50% of schools in the latter region. Furthermore, in relation to basic digital skills, ITU data across countries 9 According
to one source, English is the second language of 83 million Indians and the third language of roughly 46 million people (Livemint 2019).
3.5 The Lack of Digital Skills, Countervailing Policies …
39
reveal a marked divide between developed and developing parts of the world. To be precise, whereas the percentage of individuals who possess such skills is 65% in the former regions, the percentage for the latter is only 46% (a gap that would be considerably higher if only the poorest developing countries were included in the comparison).10 As regards smart feature phones in particular, moreover, it is worth noting that for many buyers this would have been their first contact with the Internet and the digital skills acquired from learning by doing will not yet have accrued to them. All the more acute, therefore, is the need for policies designed to redress the scarcity of such skills in India, Sub-Saharan Africa and other relatively poor regions. (For a general discussion see van Deursen and van Dijk 2014.) In what follows, I discuss only some aspects of this very large issue in relation to both India and Sub-Saharan Africa. While, for example, governments in both regions have generally not done enough to promote digital literacy,11 some impressive efforts have been made to ameliorate the problem at the national level. One of them, known as the Prime Minister’s Rural Digital Literacy Campaign in India, seeks to bring digital skills to all public primary schools. Begun in 2016, the programme was designed to reach 60 million rural households by early 2019 (The Hindu 2017). For that purpose, approximately US$350 million was appropriated for the scheme, which is certainly one of the world’s largest and for that reason alone research is needed to determine the extent to which it was successful in achieving its goal. In Kenya too, a far-reaching government programme has provided training in digital skills to over 91,000 primary school teachers (Kenya Broadcasting Corporation 2018). Here too, however, the outcomes need to be studied. One should not forget, however, the initiatives taken by private telecommunications firms to promote digital literacy. It is not surprising, for example, that Reliance Jio, with its commitment to spreading use of the Internet more widely across India, is involved in one such venture, called ‘Digital Udaan’. With Facebook as a partner, ‘this is a digital literacy programme that will help those ‘with Internet but are not savvy enough to use the various apps and features available online… The scale of this project is expected to be quite large.’ (Mahadevan 2019). Interestingly, ‘Digital Udaan’ is directed at first time users of the Internet, who, as noted above, have not had the benefit of learning by doing with this technology. What is also notable, from the point of view of linguistic diversity, is that the programme is available in ten regional languages. KaiOS, too, has developed a digital literacy application for first-time Internet users, as part of a broader programme called ‘Life’, to give such persons in Africa and South East Asia, the chance to ‘better leverage their Internet access and make the most of the digital resources at their disposal’ (Metz 2019). The digital skills application is meant to ‘help people learn the basics of the Internet, with topics 10 The
ITU (2018) provides data on the divide in digital skills between developed and developing countries. 11 According to the IFC (2018, p. 9), for example, while many countries in Sub-Saharan Africa recognize the centrality of digital skills, ‘fewer have translated this into a clear-cut agenda’.
40
3 Smart Feature Phones and Welfare in Poor Developing Countries
such as Internet navigation, privacy, security, social media, and more’ (Metz 2019). Particular policy attention, one should stress, needs to be based on what the poor actually do with the Internet, once they have adopted it. There is unfortunately not a great deal of evidence on this, but what of it there is, tends to suggest that ‘Socialising and play, not work and self-improvement, are the draw… Messaging apps help friends stay in touch… People entertain their friends -and strangers- on social media.’ (The Economist 2019). Moreover, ‘Cheap data plans and thumb drives bring pirated films to millions who may never have been to a cinema.’ (The Economist 2019). To this extent, digital literacy training for poor, first-time users of the Internet needs to focus not only on imparting basic skills but also on instructions in the use of popular socially-oriented applications, that are closely related to the well-being of the poor. In this respect as well, there are likely to be important differences between rural and urban areas of developing countries. I refer, finally, to the problem of illiteracy, inasmuch that it constrains the task of achieving some forms of digital literacy. It is not my task here to enter the welltrodden general debate on how best to overcome this problem, but rather to focus on how the problem can be alleviated by KaiOS-based smart feature phones and in particular the Google Assistant that they incorporate. For, what this application affords is the chance for illiterate users to communicate in voice rather than text, and thereby to benefit from the Internet to a much greater extent than would otherwise have been possible. The Economist (2019), for example, provides the case of an illiterate taxi driver in India, who uses the Google Assistant to communicate with a prospective customer. And given the tens of millions who are still illiterate and poor in developing countries, this component of the smart feature phone, has the chance to effect a major change in welfare for the persons concerned.
3.6 Conclusions In the form of a case study this chapter has shown first what can happen when innovating firms turn their attention to the needs of poor rather than rich countries, and more specifically to those with low incomes in the former. It has also argued that the impact on the poor of smart feature phones, based on the KaiOS operating system, is far more than merely an economic one. In particular, I have suggested that what poor individuals actually gain from the Internet depends also on a variety of more socio-cultural factors having to do with the relevance of the content that is made available; the availability of local languages; the form of communication (whether voice or text) and the availability of (at least basic) digital skills. What I find for the devices in question is that the Google Assistant plays an especially important role in making the benefits of the Internet available to the poor. Not only, for example, does it enable the illiterate among this group (and they comprise
3.6 Conclusions
41
a high proportion)12 to communicate in voice rather than text, but it also provides the largest number of local languages (in India) than any other competing voice assistant.13 Special note should also be taken of the role played by Reliance Jio and KaiOS Technology, the two partners involved in the creation and implementation of the JioPhone, both of which are privately owned. With respect to the socio-cultural determinants of what individuals actually gain from the Internet, for example, I refer to the encouragement given to local developers to submit new applications to the Jio Store. For, although there is no evidence yet available on the issue, I expect that it is likely to make available a larger number of locally relevant applications, than is usually the case. Then, with regard to digital literacy, I took note of the programmes provided by both KaiOS and Reliance Jio, for first-time users of the Internet, many of whom, it seems to me, will have relatively low incomes14 and particular needs regarding digital skills. On this last point, however, and indeed many other suppositions in the chapter, it should be emphasized that there is a dearth of evidence on what the poor actually get out of their (smart feature phone) Internet use. And there is consequently a glaring need for further research to fill these gaps in many cases with surveys of low-income users of smart feature phones, not only in India, but also, increasingly, in Sub-Saharan Africa as well. Perhaps most crucial, is the need for training in digital skills to follow what users actually do with their phones, rather than what outsiders assume them to do (e.g. socialising and play versus work and self-improvement).
References Akerlof G (1970) The market for ‘lemons’: quality uncertainty and the market mechanism. Q J Econ 84(3):488–500 Anwer J (2019) Hindi mein bolo! Google Assistant can now talk to users in Hindi, 8 other Indian languages. India Today, September 19 Bayes A, von Braun J, Akhter R (1999) Village pay phones and poverty reduction: insights from a Grameen Bank initiative in Bangladesh, University of Bonn, Center for Development Research Bird K, McKay A, Shinyekwa I (2010) Isolation and poverty: the relationship between spatially differentiated access to goods and services and poverty, Overseas Develoment Institute (ODI), London. Available http://www.chronicpoverty.org/uploads/publication_files/WP162% 20Bird-McKay-Shinyekwa.pdf
12 Thus,
according to one source, ‘Many of the countries who have been reported as having very low literacy rates are also among the poorest in the world’ (World Atlas 2017). 13 By competing voice assistants, I am referring to those included in Table 3.2. 14 The idea is that those who were unable to go online before the introduction of the smart feature phones, were unable to afford even low-end smartphones and were thus likely to be drawn from among those with relatively low incomes.
42
3 Smart Feature Phones and Welfare in Poor Developing Countries
Codeville S (2019) Orange and KaiOS Technologies reinforce their partnership to close Africa’s digital divide through Series B Funding. Available https://www.kaiostech.com/orange-andkaios-technologies-reinforce-their-partnership-to-close-africas-digital-divide-through-series-bfunding/ D’Monte L (2014) Need more local language content for internet to bloom in India. Available https://www.livemint.com/Opinion/zCFoFUXebEbxBOVq8a2UUJ/Need-more-locallanguage-content-for-Internet-to-bloom-in-In.html Deloitte (2014) Value of Connectivity: Economic and social benefits of expanding Internet access. Available https://www2.deloitte.com/content/dam/Deloitte/ie/Documents/TechnologyMedia Communications/2014_uk_tmt_value_of_connectivity_deloitte_ireland.pdf. Accessed 23 Mar 2020 GSMA (2019) The state of mobile internet connectivity report. Available www.gsma.com/resour ces/the-state-of-mobile-internet-connectivity-report-2019. Retrieved 5 Jan 2019 International Finance Corporation (IFC) (2017) Annual report International Finance Corporation (IFC) (2018) Digital skills in Sub-Saharan Africa International Telecommunication Union (ITU) (2018) Measuring the information society report, Geneva James J (2016) The impact of mobile phones on poverty and inequality in developing countries. Springer, Heidelberg Kenya Broadcasting Corporation (KBC) (2018) Digital literacy project transforming lives in Kenya. Available https://www.kbc.co.ke/digital-literacy-project-transforming-education-kenya/ Leung E (2019) News–New app bringing breaking-news to KaiOS-powered devices. Available https://www.kaiostech.com/kainews-new-app-bringing-breaking-news-to-kaios-powereddevices/ Livemint (2019) In India, who speaks in English, and where? Available https://www.livemint.com/ news/india/in-india-who-speaks-in-english-and-where-1557814101428.html Mahadevan S (2019) Jio launches ‘Digital Udaan’ program to train first-time internet users, The News Minute. Available https://www.thenewsminute.com/article/jio-launches-digital-udaan-pro gram-train-first-time-internet-users-104804 Metz T (2019) Life, a new initiative by KaiOS to help first-time Internet users make the most of mobile Internet access. Available https://www.kaiostech.com/life-a-new-initiative-by-kaios-tohelp-first-time-internet-users-make-the-most-of-mobile-internet-access/ Pringle I, David M (2002) Rural community ICT applications: the Kothmale model. Electron J Inf Syst Dev Countries 8(4):1–14 Roemer M, Gugerty K (1997) Does economic growth reduce poverty? CAER discussion paper no. 5, Harvard Institute for International Development Samuel J, Shah N, Hadingham W (2005) Mobile communications in South Africa, Tanzania and Egypt: results from community and business surveys. Vodafone Policy Paper Series 2:44–52 Singer H (1970) Dualism revisited: a new approach to the problems of the dual society in developing countries. J Dev Stud 7(1):60–75 Statista (2020) Most common languages used on the internet as of January 2020, by share of internet users. Available https://www.statista.com/statistics/262946/share-of-the-most-commonlanguages-on-the-internet/ The Economist (2019) The second half of the Internet, June 8, printed version The Hindu (2017) Cabinet nod for rural digital literacy programme. Available https://www.the hindu.com/business/Cabinet-nod-for-rural-digital-literacy-programme/article17264034.ece UNCTAD (2019) The digital economy report. Available https://unctad.org/en/PublicationsLibrary/ der2019_en.pdf van Deursen AJAM, van Dijk JAGM (2014) Digital skills: unlocking the information society. Palgrave Macmillan Wang Y (2019) The birth of the smart feature phone in India, KaiOS. Available https://www.kaiost ech.com/the-birth-of-the-smart-feature-phone-revolution/. Accessed 21 May 2019
References
43
Waverman L, Meschi L, Fuss M (2005) The impact of telecoms on economic growth in developing countries. Vodafone Policy Papers Series 2:10–24 Wiggers K (2019) Which voice assistant speaks the most languages, and why? Available https:// venturebeat.com/2019/02/02/which-voice-assistant-speaks-the-most-languages-and-why/ World Atlas (2017) Languages most commonly used on the web. Available https://www.worldatlas. com/articles/languages-most-commonly-used-by-the-web.html World Bank (2014) Internet access yes, but in my mother language. Available https://www.worldb ank.org/en/news/feature/2014/07/03/internet-access-yes-but-in-my-mother-language World Bank (2019) Africa’s future is bright—and digital. Available https://blogs.worldbank.org/ digital-development/africas-future-bright-and-digital
Chapter 4
Extending the Experience to Sub-Saharan Africa
Abstract The two previous chapters have been concerned with various aspects of a revolutionary new technology, that lies somewhere between a basic mobile phone and a relatively expensive smartphone. Before the advent of the new smart feature phone in India, users were required to buy smartphones in order to access the Internet. Now they are able to do so at a much lower cost. This chapter, however, deals with a different issue, namely, of whether and to what extent, the Indian experience can be replicated in other developing regions and Sub-Saharan Africa in particular. On the basis mainly of available data, I argue that the biggest obstacle to replication lies in affordability: in the price of handsets, data and incomes, Sub-Saharan Africa is at a disadvantage and with respect to device costs, a severe disadvantage. Other dimensions of replicability, however, were less clear-cut, partly because of a paucity of data. What is clear, though, is that associations between KaiOS Technologies and MTN and Orange have resulted in a strong African demand for mobile money through smart feature phones. Keywords Replication · India · Smart feature phones
4.1 Introduction The two previous chapters have focused on the experience with the JioPhone in India, which undoubtedly has been a very successful one. In fact, ‘Telecom company Reliance Jio topped the ‘Change the World’ list released by Forbes in 2018. The list was released for companies which through their services have changed the world positively’ (Manglik 2020, p. 485). Yet, for all the success thus realised by the Jio in India it was still the case in 2017 that access to the Internet for low-income countries was 16% (as opposed to around 85% in rich countries) (World Bank data 2018). The actual achievement of developing countries in this respect differs, one should note, quite sharply from what Goal 9 of the Sustainable Development Goals (2015) had envisaged, namely, ‘to significantly increase access to ICT and strive to provide universal and affordable access to the Internet in least developed countries by 2020’ (Digital Watch n.d.).
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 J. James, The Impact of Smart Feature Phones on Development, SpringerBriefs in Economics, https://doi.org/10.1007/978-3-030-62212-1_4
45
46
4 Extending the Experience to Sub-Saharan Africa
It follows from this recognition and the extent of the gap between supply and demand of digital skills in most developing countries, that questions arise as to the general ability of the Jio/KaiOS experience to replicate itself in other developing countries. Sub-Saharan Africa is often mentioned in this context, especially, Kenya, Ethiopia and Tanzania (Counterpoint Research 2019). The same source, furthermore, points out that 9 of the 20 countries with the most potential for smart feature phones are in Africa (Mishra 2019). Let us then first examine the scope for such phones from the point of view of affordability. Subsequently, I deal with the comparative ability of the two regions to generate local content on their smart feature phones because it is such an influential determinant of adoption and use. Then, finally, an attempt is made to compare the digital skills between India and Sub-Sahara, because these also bear keenly on the adoption and use of the KaiOS-based smart feature phones in these areas.
4.2 Affordability This variable can usefully be divided into the device price, the price of data and the incomes of prospective African users, as they relate to the Indian case of the JioPhone.
4.2.1 Device Price It was made clear in Chap. 2 on the JioPhone, that largely because of heavy subsidisation by Reliance Industries, one of the partners, this phone is effectively free to Indian users seeking something between an unconnected mobile phone and a smartphone. The KaiOS smart feature phone in much of Sub-Saharan Africa, by contrast, typically sells at around US$20, which, though considerably less expensive than most smartphones, is a good deal higher than the JioPhone in India. Two of the most prominent ventures of KaiOS in Africa involve partnerships with the well-known telecom operators, MTN and Orange. More recently, ‘TECNO, the leading mobile phone brand in Africa’, announced the introduction of the ‘latest 3G smart feature phone T901, the first TECNO device running on KaiOS, the leading mobile operating system for smart feature phones’ (Leung 2019a, n.p.).
4.2.2 Data Affordability The outcome of this measure of affordability could hardly be more pronounced between the two regions. As regards India, for example, an extensive comparison across 230 countries by a UK-based price comparison website, found that this country
4.2 Affordability
47
Table 4.1 Per capita incomes by region Region/country
Per capita income (US$)
India
2104.1
Sub-Saharan Africa (excluding high income countries)
1573.0
Lower middle-income
2176.6
Source World Bank data (2019), license CC BY-3.0-IGO
turned out to be the cheapest, with a cost per gigabyte equal to only US$0.26, due largely to intense competition.1 At the other extreme, a Web Foundation survey found that broad-band is by far the most expensive in Africa, where one gigabyte on average costs 9.3% of income. From all these points of view, therefore, Africa would find it difficult to even come close to the Indian experience with JioPhones, an expectation that is only made more likely by income differences between the two regions. This is a subject to which I next turn.
4.2.3 Income Affordability As shown by Table 4.1, this last affordability gap only serves to widen the advantage enjoyed by India over Africa, as a location for smart feature phones with an OS run by KaiOS. The table shows that India, as a low-middle income country, is unsurprisingly more wealthy on average, than the low-income countries of Africa, though, of course, there are some countries in the region with much higher per capita incomes than India (such as South Africa and Mauritius).2 And for those with lower per capita incomes, the distribution of income may be such as to allow some of the poorest groups in Africa to be better off than those in India. Bear in mind, moreover, that even if Sub-Sahara is disadvantaged in comparison with India in the above data, the attractive price of the smartphones there may still enable many of those who were previously confined to the use of basic mobile phones, to become connected to the Internet. The point here being that in order to be considered successful, adoption of smart feature phones does not need to occur at the same spectacular rate as in India.3 Unfortunately, there are, as yet, very few data which indicate how widespread the adoption and use of smart feature phones have been in Sub-Saharan Africa, largely because they were only introduced there relatively recently.
1 As
cited in McCarthy (2019). World Bank data (2019) on per capita income. 3 A point sometimes forgotten in the literature. 2 See
48
4 Extending the Experience to Sub-Saharan Africa
What should be noted in this regard, however, is that KaiOS partnerships with major African telecom operators, with their extensive presence on the continent, means that the products are introduced in a wide variety of countries and operator outlets. Vodacom, for example, is a partner whose presence in 32 African countries, aids substantially in bringing its product to local markets. Much the same is true of the KaiOS partnerships with MTN and Orange, two of Africa’s largest telecommunication companies, with the former oriented to English-speaking countries and the latter to the Franco-phone region. One of the most recent collaborations between Vodacom and KaiOS (as well as Azumi and Media Tek) resulted in the introduction of the Smart Kitochi phone in Tanzania, where 5000 were sold in the first four days, prompting an increase in production capacity, which may or may not serve as an augury of future sales (Gilbert 2019).
4.3 Local Content When considering the extent of local content associated with the smart feature phone in India and Sub-Saharan Africa, one common element is the pervasive role played by KaiOS Technologies. I have already taken note of the nodal role of the OS adopted by this firm in India, but it is also an active player in Sub-Saharan Africa, partnered as it is by some of the largest telecom operators in the latter region (such as Vodacom, MTN and Orange). It is hoped that partnerships with these firms, with their presumptive familiarity with local eco-systems, will help to impart a degree of local content to their products. Leung (2019b, n.p.), for example, has observed that, App developers, content creators, and commerce players will also reap the rewards of smart feature phones. With a new vehicle to reach a previously untapped market, the door opens in terms of generating more downloads and engagement, and offering hyper-tailored services that meet the needs of local communities. In Africa, for example, localized content like livestock tracking and mobile payment platforms have been highly impactful, and there’s room for even more solutions as more of the world’s population is able to access the Internet. Additionally, advertisers can now access this once untapped market with their ads, while users receive free data in return of the ad content.
Smart digital livestock collars are one useful example of a localised digital innovation in Africa, because they are used to promote the neglected and generally impoverished rural sector on that continent. The key general mechanism involved is that ‘When an animal wears a smart collar, the farmer can track its location and health in real time’ (MTN 2019, n.p.). More specifically, in South Africa, such collars help not only to reduce the theft of livestock, but also to identify an animal if it wanders over to a neighbouring farm (MTN 2019, p. 1). Or, ‘if a predator enters the territory of farm animals, the heart rate of the farm animal will increase and the farmer will be alerted about the distress immediately’ (MTN 2019, p. 1).
4.3 Local Content
49
By far the most influential example of a local digital innovation in Sub-Saharan Africa, however, is known as ‘mobile money’.4 It has achieved such unparalelled recognition in the region because so many Africans, including many of those described as poor, have been brought by smart feature phones into the formal banking sector. In fact, the total value of the 23.8 billion transactions carried out in 2019 exceeds US$456 billion, representing a nearly 28% jump compared to 2018 and proving that Sub-Saharan Africa beats all records in the sector, as this figure is 3.5 times the value of transactions recorded in South Asia, the second-highest ranked region in terms of mobile-money services’ (The Africa Report 2020, n.p.). Even numbers such as these, however, tend to understate the extent of what mobile money has meant for those who were formerly unbanked (individuals, who, for the most part, are drawn from the lower income groups that are so prevalent in rural areas of Sub-Sahara). For one thing, loans in the formal banking sector are very hard to come by for those without the required collateral and they are therefore forced to use informal money-lenders who usually charge exorbitant rates of interest and otherwise unfavourable terms. And although the situation has changed significantly since then, in 2014 it was reported that two-thirds of the adult population in Sub-Saharan Africa belonged to the unbanked5 section of society (Leung 2019b, n.p.). Attempts to move into the formal banking sector, moreover, are often costly and sometimes fruitless. They are costly because those in the unbanked sector may have to travel a long distance to find a bank, incurring not inconsiderable expenses, only to find that the bank does not accept those with unstable rural incomes (Leung 2019b, n.p.). Smart feature phones surmount these obstacles by allowing users to benefit from formal banking by relying on mobile telephony rather than actual banks. More specifically, ‘Mobile money … allows users to send, receive, and store money without needing a bank account. Funds are stored in a secure electronic account attached to a mobile phone number. However, because mobile money requires an internet connection, customers need more than a basic cell phone to take advantage of it’ (Leung 2019b, n.p.). KaiOS is involved in this activity through the relations it has forged with local telecommunications companies in parts of Sub-Sahara. I am thinking here especially of MTN and Orange, two of the three largest mobile network operators on the continent. ‘This means that two of the three largest mobile network operators on the continent have already placed their bet that the [KaiOS] operating system will take off’ (Jalakas 2019).
4 This 5 This
model began in Kenya under the name M-Pesa. term refers to those without any ties to a banking institution.
50
4 Extending the Experience to Sub-Saharan Africa
Table 4.2 Internet use by country, selected sample, 2018 Country
Internet use as percentage of population
India
34
Sub-Saharan Africa
25
Kenya
18
Burundi Chad
3 6
South Africa
56
Côte d’Ivoire
47
Malawi
14
Source World Bank Data (2018), license CC BY-3.0-IGO
4.4 Digital Skills According to the GSMA, the lack of digital skills represents the major impediment to Internet adoption in Sub-Saharan Africa. Thus, in a rough way, it seems reasonable to suppose that regions with more Internet use tend to have a more widespread use of digital skills. Consider, in this regard, the entries in Table 4.2 which show Internet use by country for India and selected African countries. The first two entries in the table show the aggregated comparison between India and Sub-Saharan Africa. At this level, India has almost a ten percentage point advantage over Sub-Saharan Africa, although some countries in the latter region perform substantially better than the average for the former. The outliers in question are South Africa at 56% and Côte d’Ivoire at 47%. Evidently, Internet use is not uniformly higher in India than in Sub-Sahara. Nor, I suggest are digital skills. However, while I am suggesting that Internet use implies some degree of such skills, however meagre, the comparison being made says nothing about the actual level of digital skills that are being used in each particular case.6 One alternative is to pick a single component of digital skills and compare how India and Sub-Saharan Africa perform on the basis of it (below I cover a comprehensive measure based on data collected by the World Economic Forum). The importance of foundational skills such as literacy has been well described by the IFC as follows: ‘Skills acquired early in life cement the foundation for learning that occurs into adulthood: skills beget skills. People who do not have a strong grasp of foundational skills, including literacy, digital fluency and numeracy, struggle to attain new or more advanced skills, later in life’ (IFC 2019, p. 31). A comparison of literacy for 2018 between Sub-Saharan Africa and India is contained in Table 4.3.
6 See
more on this below.
4.4 Digital Skills
51
Table 4.3 Literacy rate, adult total in 2018 Region
Rate (%)
Sub-Sahara
66
India
74
Source World Bank Data (2018), license CC BY-3.0-IGO
Table 4.4 Digital skills by country and region (7 is highest score) for 2019 (selected sample), 2019 Country/region
Score
India
4.43
Sub-Saharan Africa
3.64
African countries with score close to or above India Kenya
4.55
Seychelles
4.59
Senegal
4.21
Ghana
4.21
Mauritius
4.34
Source World Economic Forum (2017–18); Global Competitiveness Index data set
As with a number of comparisons already made in this section, Sub-Saharan Africa lags behind India but not by a very wide margin. The final comparison is based on a data-set compiled by the World Economic Forum, which provides a compound measure of digital skills for over a 100 countries. Unfortunately, this data-set does not provide an aggregated measure for Sub-Saharan Africa as a whole, so I have calculated it myself and compared it with India as shown in Table 4.4, where I also list African countries with a score that is close to or higher than that for the Asian nation. The results shown in the table are broadly consistent with those reported earlier in the chapter, namely that digital skills are greater in India than Sub-Saharan Africa (albeit generally not by a very large margin) and that there are parts of the latter which perform at an equal or higher level than the former. One such country, moreover, has been heavily involved in one of the continent’s most important applications of smart feature phones. I am referring here to Kenya and the relationship there between KaiOS and MTN on a mobile-money project. It was also in Kenya, we should note, that the original M-Pesa project was founded.
52
4 Extending the Experience to Sub-Saharan Africa
4.5 Conclusions In a recent article on replicating India’s success with the JioPhone, Mishra argues that ‘Partnerships with operators will help KaiOS make further inroads in African countries. … Clearly, Africa will be the next most important market for KaiOS after India. However, it might be a bit more challenging for KaiOS to replicate the scale and success it gained in India’ (Mishra 2019, n.p.), emphasis added). Based on the evidence I have already adduced in this chapter, however, my view is that this grossly understates the true extent of the challenge in replicating the Indian experience with smart feature phones in Sub-Saharan Africa (or, for that matter, even in coming at all close to it). The major problem may lie in what was described earlier as affordability. In fact, for every dimension that was discussed under this heading, conditions were more favourable to the adoption and use of the KaiOS-based smart feature phone in India than in Sub-Saharan Africa. In terms of the price of handsets and the cost of data, for example, the former region has a discernible advantage, which is only exacerbated by average income differences between them. Most telling, in my view, is the disparity between the costs of handsets. For, although the KaiOS designed smart feature phones in Sub-Saharan Africa are generally on sale for around US$20–a substantial reduction compared to most smartphones designed in the West–the JioPhone in India was effectively given away for free. Whether and to what extent Africans will nevertheless be attracted to the US$20 phones is a question that cannot be answered at this stage, because of a lack of data (though the early experience in Tanzania seems encouraging). What may be helpful in this regard is that ‘starting in 2020, KaiOS-powered smart feature phones will allow users to receive money, contribute to their personal savings, and even manage business finances–all with mobile money services’ (Leung 2019b, n.p.). Recall here that KaiOS works with the two largest telecom operators on the continent, MTN and Orange, both of which have their own extensive mobile money applications. Given the popularity of these, what may follow is an increase in the demand for the smart feature phone designed by KaiOS to include the use of mobile money. By way of conclusion, I return to an observation made earlier in the chapter, that although it is very doubtful that Sub-Sahara will achieve the degree of success enjoyed by the smart feature phone in India, it is more than probable that millions of Africans will nevertheless benefit from it. Part of my conviction in this view stems from some of the findings in this chapter, that on some measures, the disparity in scores between the two regions is not that sizeable. I am thinking here, for example, of the difference in per capita incomes and digital skills. Gains will tend not to occur, however, among the poorest groups in rural areas, whose incomes are too meagre to be able to afford the cost of the handset. A valuable research project would be to determine the incomes and location of those who purchase smart feature phones in India as well as Sub-Saharan Africa (though as noted below, for the Indian case, there
4.5 Conclusions
53
is indirect evidence that the JioPhone is being purchased by some of those living in poverty). And the same would be true, finally, of Indonesia where a partnership between KaiOS, a mobile device company and a franchise convenience store has led to a product called the WizPhone, that is available for only US$7. According to Leung (2018), ‘it is the result of a new business model that couples retail and banking incentives to make the device more affordable’ (Leung 2018, n.p.). It is not clear, however, what the new model actually comprises.
References Counterpoint Research (2019) More than a billion feature phones to be sold over next three years. Available https://www.counterpointresearch.com/more-than-a-billion-feature-phones-tobe-sold-over-next-three-years/. Accessed 3 Jan 2020 Digital Watch (n.d.) Sustainable development goals. Available https://dig.watch/processes/sustai nable-development-goals. Accessed 23 Mar 2020 Gilbert P (2019) More smart feature phones drop at AfricaCom, Connecting Africa. Available http:// www.connectingafrica.com/mobile/author.asp?section_id=761&doc_id=755663. Accessed 12 Feb 2020 International Finance Corporation (IFC) (2019) Digital skills in Sub-Saharan Africa: spotlight on Ghana. Available https://www.ifc.org/wps/wcm/connect/ed6362b3-aa34-42ac-ae9f-c73990495 1b1/Digital+Skills_Final_WEB_5-7-19.pdf?MOD=AJPERES. Accessed 3 Mar 2020 Jalakas W (2019) After years of rapid growth in Africa we’re about to enter the age of mobile money 2.0. Available https://qz.com/africa/1721818/africa-mobile-money-industry-is-enteringits-next-stage-of-growth/. Accessed 10 Jan 2020 Leung E (2018) The first KaiOS-powered smart feature phone arrives in Indonesia. Available https:// www.kaiostech.com/the-first-kaios-powered-smart-feature-phone-arrives-in-indonesia-a-newpartnership-model-combining-retail-and-banking-to-advance-financial-inclusion/. Accessed 3 Feb 2019 Leung E (2019a) The first TECNO device running KaiOS is here: Meet the T901. Available https:// www.kaiostech.com/the_first_tecno_device_running_kaios_is_here_meet_the_t901/ Leung E (2019b) How smart feature phones help the unbanked. Industry insights, KaiOS Technology. Available https://www.kaiostech.com/how-smart-feature-phones-help-the-unbanked/. Accessed 31 Dec 2019 Manglik R (2020) XAT (Xavier Aptitude Test). Edu Gorilla, India McCarthy N (2019) The cost of mobile Internet around the world, Forbes. Available https://www. forbes.com/sites/niallmccarthy/2019/03/05/the-cost-of-mobile-internet-around-the-world-inf ographic/#74220e09226e Mishra V (2019) KaiOS eyes global expansion but replicating India’s success would be a challenge. Available https://www.counterpointresearch.com/kaios-eyes-global-expansion-replic ating-indias-success-challenge/. Accessed 15 July 2020 MTN (2019) Livestock farmers can benefit from rural connectivity. Available https://www. mtn.com/our-story/spotlights/livestock-farmers-can-benefit-from-rural-connectivity/#:~:text= In%20countries%20like%20South%20Africa,over%20to%20a%20neighbouring%20farm.& text=With%20more%20than%201000%20rural,to%20use%20smart%20collar%20solutions. Accessed 13 April 2020 The Africa Report (2020) Scalabrini Institute for Human Mobility in Africa (SIHMA). Available http://www.sihma.org.za/Blog-on-the-move/the-africa-report-2020. Issue January– February–March 2020, No. 110
54
4 Extending the Experience to Sub-Saharan Africa
World Bank Data (2018) Individuals using the Internet (% of population). Available https://data. worldbank.org/indicator/IT.NET.USER.ZS. Accessed 31 Oct 2019 World Bank Data (2019) World Economic Forum (2017–2018) The global competitiveness report. Available https://www. weforum.org/reports/the-global-competitiveness-report-2017-2018. Accessed 27 May 2019
Part II
Digital Skills and Digital Paradoxes
Chapter 5
Measuring the Second Digital Divide: Education and Skills
Abstract As it is usually conceived, the digital divide between rich and poor countries refers to differences in access to digital technologies. The second digital divide on the other hand, is concerned with the factors that determine whether and how the adopted technologies are used and the extent of the benefits that are derived from them (one such factor, for example, is digital skills). The original feature of this chapter is that it seeks to measure the second digital divide between rich and poor countries and to compare the outcome with the more familiar divide in digital access (and the internet in particular). The former is measured with reference to two key technology complements, namely, attainments in reading and mathematics on the one hand and digital skills on the other. What I find is that even the acute divide in the Internet is surpassed by the figures representing the second divide in literacy and mathematics. This finding is attributed mainly to the acute crisis in learning that besets many schools in poor countries, especially those in Sub-Saharan Africa. Somewhat oddly, though, the divide in digital skills is less acute than for learning achievements. Apparently, certain digital skills can be acquired without prior competencies in basic education. Keywords Digital divide · Literacy · Mathematics · Digital skills Discussions of the digital divide between rich and poor countries have long been dominated by differences in access to the technologies themselves. From time to time, however, critics of this way of conceiving of the divide have pointed out that access alone does not determine the benefits that are derived from digital technologies. On the contrary, the critics are wont to argue that such benefits also depend on a range of other factors, such as skills and infrastructure. Indeed, some authors refer to the differential degree to which these other factors are available to rich and poor countries as a second digital divide.1 An OECD report in 2010, for example, observed that ‘the digital divide in education goes beyond the issue of access to technology. A second digital divide separates those with the competencies and skills to benefit from computer use from those without’ (OECD 2010, p. 2).
1 See,
for example, Attewell (2001) and Selwyn (2003).
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 J. James, The Impact of Smart Feature Phones on Development, SpringerBriefs in Economics, https://doi.org/10.1007/978-3-030-62212-1_5
57
58
5 Measuring the Second Digital Divide: Education and Skills
Yet, perhaps because of the sheer complexity of the issue and the fragmentation of the available data, very few attempts have been made to actually measure this important concept. The attempt at measurement that is made below in no way purports to be comprehensive. But it does cover what are arguably the most salient variables involved in the exercise, namely, the Internet on the one hand, and educational achievements and digital skills on the other (see more on this below). In particular, following a brief introductory section, I enquire as to whether a second digital divide actually exists and if so, how large it is and how it affects the Sustainable Development Goals. The penultimate section compares this divide with the original one and concludes that on the basis of the selected variables, the divide in learning exceeds that in Internet access. Apparently, those who have emphasized the theoretical importance of the second divide can also claim some considerable empirical support as well (subject of course to the limitations of the method I have used, as noted below).2 And from a policy point of view, a concomitant shift in attention towards this second divide needs to occur.3
5.1 Measuring the Original Conception of the Digital Divide As noted above, early conceptions of the digital divide were very much about differences in access to information and communications technologies. Thus, according to a common view, the divide was described as ‘the gap between people who do and do not have access to forms of information and communication technology’ (van Dijk 2017, p. 1). And the ‘access’ view still persists in some circles even in much more recent times. The ‘new sustainable development agenda’ of the UN in 2015, for example, ‘calls for universal and affordable access to the Internet by 2020’ (Garnett 2017) Moreover, in achieving this goal, the emphasis is not laid on policy to bolster the non-technological components of the problem, but rather on a need for the ‘development community …. To prioritize Internet connectivity and access by integrating it into their core missions. Internet access helps drive broader development goals’ (Garnett 2017, emphasis added). Even in 2018, moreover, one can read an observation by the ITU to the effect that ‘actual implementations often focus on the expansion of infrastructure and access(ibility)–e.g. ensuring broadband Internet access. This is true for schools both in developed and developing countries, where the main policy so far has been to equip students or schools with a preset number of computers and Internet connections’ (ITU 2018, p. 46).
2 Although it does not specifically refer to the second digital divide as such, the World Development
Report of 2016, is in fact about this divide and its consequences (World Bank 2016). relevant in this regard, is the 2018 version of the World Development Report, which deals in large measure with the ‘crisis in learning’ in developing countries, and the countervailing policies that need to be made in order to ameliorate the problem.
3 Also
5.1 Measuring the Original Conception of the Digital Divide
59
Table 5.1 The digital divide in the Internet (by income group) (2016) Country group
Individuals using the Internet (by income group) (%)
High-income
82.0
Upper-middle income
55.7
Lower-middle income
29.9
Low income
12.5
Source World Bank, The little data book on information and communication technology (2018a), available http://hdl.handle.net/10986/31967
Table 5.2 The digital divide in the Internet (by regional groups) (2016) Region
Individuals using the Internet (%)
World
45.9
East Asia and Pacific
52.8
Europe and Central Asia
78.5
Latin America and Caribbean
26.5
Sub-Saharan Africa
20.5
Source World Bank, The little data book on information and communication technology (2018a), available https://openknowledge.worldbank.org/handle/10986/31967
Nonetheless, even conceived thus in a technologically determinist way, the first digital divide says something important about the relationship between rich and poor countries in the world economy. As exhibited in Table 5.1 for the divide in the Internet, it says that there is a much greater potential for this technology to impart benefits to the populations of rich versus poor countries. I have chosen the Internet for this comparison, because among digital technologies it offers the widest array and most extensive benefits for society in rich and poor countries alike (Deloitte 2014). The former have already realized many such benefits, whereas for the latter, these gains are to a greater extent still only potential. The table indicates a clear positive group association between Internet use and country group incomes. This is hardly surprising because it reflects the way in which new digital technologies tend to be generated. In particular, such technologies are generally designed in and for the rich countries and their characteristics, in terms of skills, incomes, infrastructure and so on. What, in effect is a rich country technological system,4 is then generally ill adapted to the systems prevailing in the poorest countries, where, as shown in the table, Internet use is very much lower (i.e. 82% vs. 12.5% for the poorest group of countries). Indeed, as argued below, sizeable differences in the second digital divide serve as indicators of the mismatch between technological systems in rich versus poor countries. Table 5.2 provides a geographical supplement to the information shown in the previous table on differences in country income groupings. It shows, for example, 4 On
the concept of a technological system see James and Khan (1998) and Stewart (1977).
60
5 Measuring the Second Digital Divide: Education and Skills
that the lowest amount of Internet use occurs in Sub-Saharan Africa and South Asia and that such use in Latin America is only fractionally lower than the region with the highest use, namely, Europe and Central Asia.5
5.2 Recognition and Measurement of the Second Digital Divide Although it is not cast in the same terms, the World Development Report of 2016, makes the most thorough and compelling case for the need to focus more conceptual and policy attention on the second digital divide. Consider, for example, its view that those who comprise the ‘bottom billion in the world ‘are capturing only a modest share of the digital dividends’ (World Bank 2016, p. 16). For, among other reasons, many of the poor ‘lack the basic literacy and numeracy skills needed to use the Internet. In Mali and Uganda, about three-quarters of third-grade children cannot read. In Afghanistan and Niger, 7 of 10 adults are illiterate’ (World Bank 2016, p. 16). Or again, ‘For digital technologies to benefit everyone everywhere requires closing the remaining digital divide, especially in Internet access. But greater digital adoption will not be enough. To get the most out of the digital revolution, countries also need to work on the ‘analog complements’ (World Bank 2016, abstract). Yet, for all such recognition of the second digital divide, few and far between are the attempts to measure it across rich and poor countries (and thus to compare it in size with the access (or first) divide. It is true that a few academic attempts to measure digital skills have been made in the past, but largely in relation to differences within developed countries, and the United States in particular. Hargittai (2002), for example, with reference to that country, argues that, It is increasingly important to look at not only who uses the Internet, but also to distinguish varying levels of online skills among individuals. Skill, in this context, is defined as the ability to efficiently and effectively find information on the Web. By exploring the differences in how people use the Web for information retrieval, we can discern if there is a ‘second-level digital divide’ in the making as the Web spreads to the majority of the American population. (Hargittai 2002, n.p., emphasis in original)
More specifically, the author applies a novel method of measuring digital skills in relation to the Internet. ‘First, the binary success/failure rate shows what portion of the respondents was able to complete a certain task. Second, the time to completion of each task is measured in seconds to show the gradual differences in how long people take to find information on the Web’ (Hargittai 2002, n.p.). Because Hargittai finds a striking degree of variation in the time taken by respondents (who are differentiated by age, sex, education and so on), she concludes that ‘Although providing Internet access may help alleviate some of the problems of the digital divide, information presented in this paper demonstrates that a second-level digital divide exists relative 5 Quite
why Latin America performs so well is not clear to me.
5.2 Recognition and Measurement of the Second Digital Divide
61
Table 5.3 Children and adolescents (primary and lower secondary school age) not achieving MPLs in reading by country income, 2015 Country grouping by income level
School-age population who will not achieve MPLs in reading (%)
Share of world total of children not learning (%)
High income
13
2
Upper middle
31
15
Lower middle
75
60
Low income
90
23
World
58
100
Source UNESCO, Institute of Statistics (2017, p. 15)
to specific abilities to effectively use the medium’ (Hargittai 2002, n.p., emphasis added). My attempt to measure a second digital divide between, rather than within developed or developing countries, also uses the Internet as the mechanism of choice, because of the superior range and degree of benefits that it offers compared to other digital technologies. Unlike the study just described, however, the exercise conducted here does not rely on given tasks to find specific information online. Rather, it relies partly on relatively widely accepted definitions of minimum competency levels in numeracy and literacy across countries, because the extent to which such competencies are achieved, influences both the second divide and the SDGs.6 In the same vein, digital skills themselves are measured by the complexity of different tasks, as defined, for example, by the ITU (the UN organisation that is specialized in matters pertaining to information technology). Having already discussed the size of the gap in access to the Internet in developed and developing countries (or, more accurately, between high and low-income nations), I turn next to the first component of the second divide, namely, the differences in mathematical and literacy achievements between the same two groups of countries. Tables 5.3 and 5.4 deal with reading and mathematical achievements respectively and the relevant comparison for our purposes is again between high and low-income countries. Thus, among the school-age population in the former, only 13% fail to reach the minimum proficiency level in reading, as opposed to the figure of 90% of the school-age population in poor countries. The last column of Table 5.3 also shows the percentage contribution of each income group to the total number of those who do not meet minimum proficiency levels in reading.7 So, while the high-income countries contribute only 2% to the total, those in the low-income category contribute more than 10 times as much.
6 On 7 For
the latter see below. a discussion of minimum proficiency levels see UNESCO (2017).
62
5 Measuring the Second Digital Divide: Education and Skills
Table 5.4 Children and adolescents (primary and lower secondary school age) not achieving MPLs in mathematics by country income, 2015 Country grouping by income level
School-age population who will not achieve MPLs in mathematics (%)
Share of world total of children not learning (%)
High income
13
2
Upper middle
32
16
Lower middle
73
59
Low income
87
22
World
56
100
Source UNESCO, Institute of Statistics (2017, p. 15)
The figures shown in Table 5.4, for mathematics, are remarkably similar to those shown for reading in the previous table. So too, therefore, is the size of the (second) digital divide in these 2 dimensions of learning. This can best be seen by converting the data in the 2 tables from the percentage of those who do not achieve minimum proficiency levels into the percentage of those who do. Thus, in Table 5.3 for example, the first entry of the first column changes from 13 to 87%, and for low-income countries the change is from 90 to 10%. The reasons for so deep a divide in learning achievements are examined in the next section. Before that, however, it is instructive to consider the regional dimension of the divide in learning achievements, as shown in Table 5.5. What has been referred to as a ‘crisis in learning’ is most apparent in Sub-Saharan Africa, where 88% of children and adolescents will be unable to reach MPLs in reading and 84% in mathematics, ‘by the time they are of age to complete primary and lower secondary education’ (UNESCO 2017, p. 7). Indeed, ‘if current trends continue, this crisis will affect about 202 million children and adolescents, including Table 5.5 Proportion of children and adolescents (primary and lower secondary school age) not achieving MPLs in mathematics and reading by SDG income, 2015) SDG region
Mathematics (%)
Reading (%)
World
56
58
Sub-Saharan Africa
84
88
Central and Southern Asia
76
81
Western Asia and Northern Africa
57
57
Latin America and the Caribbean
52
36
Eastern Asia and South-Eastern Asia
28
31
Oceania
22
22
Northern America and Europa
14
14
Source UNESCO, Institute of Statistics (2017, p. 7)
5.2 Recognition and Measurement of the Second Digital Divide
63
138 million of primary school age and 63 million of lower secondary school age’ (UNESCO 2017, p. 7). At the other extreme, predictably, is the situation in North America and Europe, where only 14% of the students in question will fail to reach MLPs in mathematics and reading. Underlying so pronounced a divide in learning achievements is what, as noted above, is sometimes called a ‘crisis in learning’, especially among the poorest countries. It is to this aspect of the second digital divide that I next turn.
5.3 The Crisis in Learning Of such concern is the stagnancy of learning achievements in poor countries, that the World Development Report of 2018 is mostly devoted to the topic. The report deals not only with the current situation (summarized in Tables 5.3, 5.4 and 5.5) but also with the inferences that it draws about the future. In particular, Students often learn little from year to year, but early learning deficits are magnified over time. Students who stay in school should be rewarded with steady progress in learning, whatever disadvantages they have in the beginning. And yet, in Andra Pradhesh, India, in 2010, low-performing students in grade 5 were no more likely to answer a grade 1 question correctly than those in grade 2. Even the average student in grade 5 had about a 50 percent chance of answering a grade 1 question correctly. (World Bank 2018b, p. 8)
What is more, it is often the case that poorly-performing students are drawn from among the most disadvantaged groups and vice versa. Left to itself, therefore, the system tends to exacerbate inequalities over time, as Myrdal pointed out many years ago (Myrdal 1968). Unfortunately, moreover, it is fair to say in general that countervailing forces only rarely occur in poor countries.8 Indeed, what are often found instead are policies and circumstances that tend to exacerbate rather than ameliorate the crisis in learning. The World Bank (2018b), for example, points to four such factors. ‘First, children often arrive in school unprepared to learn–if they arrive at all. Malnutrition, illness, low parental investments and the harsh environments associated with poverty undermine early childhood learning’ (World Bank 2018b, p. 10). The second reason is that ‘teachers often lack the skills or motivation to be effective’ (World Bank 2018b, p. 10). This may seem obvious but it derives particular significance from the view that ‘Teachers are the most important factor affecting learning in schools’. Third, as the Bank sees it, inputs often fail to reach classrooms or to affect learning when they do. Public discourse often equates problems of education quality with input ‘gaps’. Yet, there is evidence that schools with roughly similar amounts of inputs sometimes produce vastly different educational outcomes. Finally, ‘poor management and governance often undermine schooling quality …. Across eight countries that have been studied, a 1.00 standard
8 With
respect, say, to income inequality.
64
5 Measuring the Second Digital Divide: Education and Skills
deviation in an index of management capacity-based on the adoption of 20 management practices–is associated with a 0.23–0.43 standard deviation increase in student outcomes’ (World Bank 2018b, p. 11). These problems will all need to be addressed in order to ameliorate the learning crises and narrow the gap in school achievements between rich and poor countries. And indeed many and varied are the proposals that have been espoused for this purpose.9 Yet, given the cumulative nature of the education process described above, it does seem to be well worth stressing that countervailing policy needs to begin at the very outset of the schooling process. Indeed, ‘getting learners to school ready and motivated to learn is a first step to better learning. Without it, other policies and programs will have a minimal effect’ (World Bank 2018b, p. 21, emphasis added). Note, finally, that while the attainment of basic literacy and numeracy is a good thing in its own right–enabling among other things, accounts to be checked and books to be enjoyed–it also contributes towards digital skills. Thus, to the extent that such skills contribute to learning in schools, the two components of the second digital divide are inter-dependent rather than independent of one another.
5.4 Measuring the Digital Skills Component of the Second Digital Divide It is difficult to find a concise definition of digital skills because, as shown below, there are a variety of them (skills), which need to be arranged on a continuum of difficulty and complexity. One relatively simple definition, however, is that these skills concern ‘the ability to effectively and critically navigate, evaluate, and create information using a range of digital devices and technologies’ (Deloitte 2014, p. 41). For what follows, however, a more concrete definition is required, one that takes into account the varying complexities of different digital tasks involved in operating the Internet. Fortunately, the ITU (2018) has recently provided just such a classification, which divides digital skills into three categories, namely, basic, intermediate and advanced. Each of these is described above in Table 2.5. On the basis of these more specific definitions, it becomes possible to measure the other component of the second digital divide between rich and poor countries, as shown in Table 2.6 above. Note, though, that the absence of advanced skills in that table tends to understate the true extent of the digital skills divide between rich and poor countries, because of the comparative rarity of this type of skill in poor countries. Even so, the size of the digital skills gap in Table 2.6 might seem unexpectedly low, with only a 20 point 9 See,
for example, the papers produced by the Brookings Institution, such as, by Watkins (2013). Moreover, the World Bank and UNESCO frequently publish material on this topic.
5.4 Measuring the Digital Skills Component of the Second …
65
Table 5.6 The digital skills divide between developed and least-developed countries, 2017 Region
Basic skills (%)
Standard skills (%)
Developed countries
65–70
45–50
Least developed countries
±20
10
Source ITU (2018, p. 14 based on Chart 2.5)
difference in the percentages of basic skills and 29 points in the case of standard skills. I suggest that the anomaly can be resolved by acknowledging the difference between developing and low-income countries. In particular, the point being that the former category includes many countries that are too developed to be considered as being part of the latter category. Tables 5.6 and 5.7 contain 2 proxies for the absence of digital skills data in lowincome countries. The former uses the category of least-developed countries, while the latter refers to countries in Sub-Saharan Africa. Both tables indicate an appreciable increase in the size of the digital skills gap compared to Table 2.6, thus confirming the inadequacy of the concept of developing countries as an indicator of low-income countries. For, whereas the two proxy indicators show a difference of around 40 percentage points, the measure based on developing countries is roughly only half of that. All in all, our findings on the measurement of the second digital divide, as regards education and digital skills, can usefully be summarized as in Table 5.8. In order to facilitate a comparison between the various measures, the last row shows the percentage gap (i.e. the divide) in each case, between high and low-income Table 5.7 The digital skills divide between developed and Sub-Saharan African countries Region
Basic skills (%)
Standard skills (%)
Developed countries
65–70
45–50
Sub-Saharan Africa
±25
10–15
Source ITU (2018, p. 14 based on Chart 2.6)
Table 5.8 A summary of the divides: differences between high and low-income countries Internet access (%)
Reading % achieving MPL
Mathematics % achieving MPL
Digital skills (basic) (%)
Digital skills (standard) (%)
High income
82.0
87
87
65–70
45–50
Low income
12.5
10
13
20a
10b
Gap
69.5
77
74
45–50
35–40
Source Tables above in this chapter a Refers to least-developed country proxy b Refers to Sub-Saharan proxy
66
5 Measuring the Second Digital Divide: Education and Skills
countries. Thus juxtaposed, the entries show, firstly, that even the acute Internet divide of 70%, is surpassed by the figures representing the second digital divide in literacy and mathematics. Much of the reason for this result, as I see it, has to do with the severity of the ‘learning crisis’ discussed above, which has so severely inhibited progress in learning achievements in low-income countries. The main practical import of the comparison that has just been described is that the basic educational achievements of the low-income countries are not sufficient even to cope with their limited Internet access. Or, put another way, the point is that on average, such countries derive fewer (and in many cases, much fewer) benefits from the Internet than are actually available (and generally captured in high-income countries). Oddly, though, Table 5.8 shows that the digital divide in skills is quite a bit lower than it is in numeracy and literacy. Apparently, there are digital skills that can be acquired without minimum competencies in literacy and numeracy. Van Deursen and van Dijk hint at such a possibility when they argue that ‘digital skills can be taught even to complete and functional illiterates. For, to adopt the opposite conclusion would mean that there was no possibility of reaching such skills in developing countries (van Deursen and van Dijk 2014). What also needs to be mentioned in this regard, moreover, is Mitra’s (2003) wellknown experiments with children from slums in India. In particular, his ‘hole-in-thewall’ experiments consisted of setting up a computer with an Internet connection in a Delhi slum and observing the result. What occurred was basically that the hardware attracted the attention of a group of illiterate, poverty-stricken children, who, within a single day, were able to surf the Internet. Much more research, however, is required to understand how digital skills are acquired in low-income countries without reaching the prior levels of competency in reading and mathematics.
5.5 The Second Digital Divide and the Sustainable Development Goals The UNESCO Institute of Statistics (UIS) makes it clear that the crisis in learning described above, has a potentially marked negative influence on the Sustainable Development Goals. Thus, ‘The …… data signal a tremendous waste of human potential that could threaten progress towards the Sustainable Development Goals (SDGs)’ (UNESCO 2017, p. 1). In this last section of the chapter, I enquire more specifically as to how the findings in the sections above, impinge on the SDGs. The most direct connection seems to involve the lack of learning in poor countries and Goal 4, the progress toward which, according to the UIS, influences many of the other goals as well. The goal in question is essentially about ensuring an equitable and quality primary and secondary education for boys and girls. Unfortunately, however, the reality is that educational attainments are generally neither equitable nor of high quality. The lack of quality can be inferred from the vast gulf in these attainments between high and low-income countries, which, as
5.5 The Second Digital Divide and the Sustainable Development Goals
67
Table 5.9 Urban and rural literacy rates (total in %), selected sample Country
Urban
Rural
Date
Brazil
92.5
76.5
2008
Burkina Faso
62.9
19.5
2007
Chad
43.7
13.1
2004
India
78.2
53.7
2001
Niger
52.0
23.4
2005
Pakistan
71.1
46.3
2008
Senegal
69.1
33.2
2009
Sierra Leone
55.7
21.5
2004
Source UNESCO, Institute for Statistics (2011)
shown above, exceeds even the sizeable gap in Internet access. Moreover, there are also acute gaps in learning within the low-income countries themselves. Consider, for example, the divide in rural and urban literacy rates for a selected sample of developing countries, shown in Table 5.9. The figures in the table show that the rural-urban divide tends to be most acute in the African countries, which comprise the majority of the sample. This tendency also has negative implications for inequality in as much that the poor tend to be concentrated in the rural areas of developing countries. Especially in these areas, moreover, the learning crisis also tends to have negative effects on the other SDGs (such as when, for example, the illiterate are unable to read the instructions on medicines and are thereby thwarted in their ability to maintain good health (Goal 3). Or, when illiteracy frustrates the ability of children to learn adequately about climate change and promote sustainable development (Goal 13). Finally, one needs to point to the learning crisis and a possible reason why Goal 9, to provide universal and affordable access to the Internet in the least developed countries by 2020, has failed so conspicuously. I have also taken note of the retarding influence of the learning crisis on digital skills and hence the ability to use and benefit from the Internet (though, regrettably, the attainment of such skills is not mentioned as a goal in the SDGs). Yet, there are at least 3 surveys of Internet use in developing countries, which show that a lack of knowledge is the main reason why prospective users of this technology fail to access it. According to one such study, for example, ‘the most-cited reason for why Africans don’t access the Internet is that they haven’t developed the skills to do so. A 2014 survey … of Chinese consumers—found that approximately 60% of the offline population cited a lack of knowledge of how to use a computer as a primary reason for not accessing the Internet’ (Deloitte 2014, pp. 41–2). Ideally, one would like to separate out the effects of stunted learning and the lack of digital skills, as causes of these outcomes but in practice that would prove very difficult, if not entirely impossible, since these deficiencies interact with each other.
68
5 Measuring the Second Digital Divide: Education and Skills
5.6 Conclusions This chapter has addressed itself to the measurement of a second digital divide, that is, a divide that is concerned not with the differential access to ICTs by rich and poor countries (the conventional measure of the digital divide), but rather with the differences between those countries with regard to educational achievements and digital skills, variables that form part of what the World Bank (2016) refers to as ‘analog complements’ to digital technologies. One such technology, the Internet, was selected for the purposes of this chapter, mainly because of the range and depth of the benefits that it offers to those living in poor countries. More specifically, the questions I have posed above, have to do with whether a second divide exists; how pronounced it is; whether it is larger or smaller than the divide which focuses on access to digital technologies; and how the answers to these questions bear on the Sustainable Development Goals (SDGs). The most telling finding of the analysis that is devoted to these questions, is that not only does the second digital divide exist, but also that as regards school learning achievements, it actually exceeds the digital divide as conventionally measured (in terms of Internet access). Much the most important reason for this outcome, I argue, is what some observers refer to as the crisis in learning, which is particularly severe in Sub-Saharan Africa, where learning achievements are lagging behind even the dismally low rates of Internet access. Somewhat surprisingly, though, the divide in digital skills is less acute than it is for literacy and numeracy. Apparently, certain digital skills can be gained without the prior acquisition of basic learning achievements (van Deursen and van Dijk 2014). Strangely, though, this important policy issue has been largely ignored in the literature and much more research on it is clearly required. Finally, the research reported here is most directly connected to Goal 4 of the SDGs, the one that is concerned with achieving an equitable and quality education for all school-children. For what we have found instead is that school education in lowincome countries has generally been neither equitable nor of anything approaching high quality. Unfortunate too, is the fact that Goal 9 of the SDGs to provide universal and affordable access to the Internet by 2020 has fallen so wide of the mark, despite the enormous potential that they offer, especially to the rural poor. And while there is by now a vast literature on ways to improve levels of literacy and numeracy in developing countries, the same cannot be said of digital skills, which have been relatively neglected especially among the poorest developing countries. (Though, of course, there are exceptions to this general picture, such as the digital skills application on some of the new smart-feature phones that allow ultra low-cost Internet access, as well as voice applications that substitute for gaps in literacy.)10
10 I
am referring here principally to the ‘JioPhone’ in India which has already brought low-cost Internet to millions of Indians.
5.6 Conclusions
69
Then too there are some very poor countries which have made the acquisition of digital skills into a policy priority. For example, Bangladesh’s Vision 2021, ‘aims to ‘catapult’ Bangladesh into a strong middle-income country with a focus on digital skills. … The country’s digitalization program has trained some 180,000 teachers in how to teach digital skills … Also, they have launched programs to train girls and young women in ICT skills, especially in rural areas’ (ITU News 2017, n.p.).
References Attewell P (2001) The first and second digital divides. Soc Educ 74(3):252–259 Deloitte (2014) Value of connectivity. Available https://www2.deloitte.com/content/dam/Del oitte/ie/Documents/TechnologyMediaCommunications/2014_uk_tmt_value_of_connectivity_ deloitte_ireland.pdf. Accessed 29 Oct 2016 Garnett P (2017) Bringing the other half of the world online. Microsoft. Available https://blogs.mic rosoft.com/on-the-issues/2017/02/22/bringing-half-world-online/. Accessed 14 Aug 2018 Hargittai E (2002) Second-level digital divide: differences in people’s online skills. First Monday 7(4). Available https://firstmonday.org/article/view/942/864 International Telecommunication Union (ITU) (2018) Measuring the information society report, Geneva ITU News (2017) Digital skills for a new economy: What’s needed now? Available http://digitalsk illsforaneweconomy-what’sneedednow.html. Accessed 13 May 2019 James J, Khan H (1998) Technological systems and development. Macmillan, Basingstoke Mitra S (2003) Minimally invasive education: a progress report on the ‘hole-in-the-wall’ experiments. Br J Educ Technol 34(3):367–371 Myrdal G (1968) Asian drama: an inquiry into the poverty of nations. Random House, New York Organisation for Economic Co-operation and Development (OECD) (2010) Are the new millenium learners making the grade? Paris. Available http://www.oecd.org/education/ceri/45053490.pdf. Accessed Mar 3 2018 Selwyn N (2003) Apart from technology: understanding people’s non-use of information and communication technologies in everyday life. Technol Soc 25(1):99–116 Stewart F (1977) Technology and underdevelopment. Macmillan, London UNESCO (2011) Urban and rural literacy rates. Available https://uil.unesco.org/fileadmin/keydoc uments/Literacy/LIFE/Mid-termPackage/8_statistical_data_on_Literacy/4UIS_LIFE_urban_ rural_graph_2011.pdf UNESCO (2017) More than one-half of children and adolescents are not learning worldwide UNESCO Institute for Statistics, Fact sheet No. 46. Available http://uis.unesco.org/sites/default/ files/documents/fs46-more-than-half-children-not-learning-en-2017.pdf. Accessed 3 May 2018 van Deursen AJAM, van Dijk JAGM (2014) Digital skills: unlocking the information society. Palgrave Macmillan van Dijk J (2017) Digital divide: impact of access. The international encyclopedia of media effects. Wiley. Available https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118783764.wbieme0043 Watkins K (2013) Too little access, not enough learning: Africa’s twin deficit in education. Brookings Institution. Available https://www.brookings.edu/opinions/too-little-access-not-enough-lea rning-africas-twin-deficit-in-education/ World Bank (2016) World Development Report, Washington, DC World Bank (2018a) The little data book on information and communication technology World Bank (2018b) World Development Report, Washington, DC
Chapter 6
Anti-development Bias in the Use of the Internet in Developing Countries. What Underlies It?
Abstract Recent years have seen a shift away from issues related to access to new technologies in general and the internet in particular. There is talk, in some policy circles, of extending ‘beyond access’, to examine how such technologies are used and what benefits they actually yield to users. In fact, numerous field studies have now examined this question in the context of developing countries and come to the unexpected conclusion that the selection of uses on the internet tends to favour leisure over work. After reviewing the evidence that underlies this conclusion, this final chapter is concerned to examine its implications. One response is to accept and even commend the result, by suggesting that it reduces the leisure divide between rich and poor countries. An alternative response, one to which I adhere, is to examine the causes that underlie the anti-development bias before deciding to accept it for policy purposes. What I suggest is that the choice of internet uses reflects major gaps in knowledge about the internet and a severe lack of digital skills on the part of first time users of the technology (especially among uneducated and low-income individuals). Policy should attempt to redress these limitations rather than accept the outcomes to which they give rise. Keywords Internet users · Preferences · Digital skills · Internet benefits In quite recent years attention has begun to swing from physical access to information technologies such as mobile phones and the Internet, to the topic of how such technologies are actually used after access to them has been gained (Correa 2016). It is this, after all, that determines the gains in welfare from information technologies in general and the Internet in particular. My concern below is with one form of Internet, the mobile Internet, which has grown rapidly in recent years. Indeed, ‘in 2018 almost 300 million people connected to mobile Internet for the first time, bringing the total connected population to more than 3.5 billion people globally. For many of these individuals, mobile is the only method of accessing the Internet, so growth in mobile Internet adoption also drives digital inclusion, especially in low and middle income countries (LMICs)’ (GSMA 2019). Only recently, however, have attempts been made to bring empirical evidence to bear on the ways in which the Internet is actually used in (rich and) poor countries and to derive the implications thereof. The main finding of this body of research–much of © The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 J. James, The Impact of Smart Feature Phones on Development, SpringerBriefs in Economics, https://doi.org/10.1007/978-3-030-62212-1_6
71
72
6 Anti-development Bias in the Use of the Internet in Developing …
which was conducted by the GSMA–is that use preferences in developing countries reveal an anti-development bias (or more specifically, an unexpected preference for uses that favour leisure over work). One response to this finding is based on the traditional economics notion of consumer sovereignty, which suggests that the user reveals a preference for his or her choices, which are not to be tampered with, except under certain exceptional conditions. The Economist (2019), for example, sees no reason to question the use pattern of those in the developing world who, in using the internet, eschew the work over leisure paradigm. Indeed, this publication celebrates the diminution of digital inequality in the pursuit of leisure between rich and poor countries. In what follows, by contrast, I argue that reliance on unfettered use patterns is unlikely to yield the gains that are expected of them. One has to consider, for example, that preferences are rarely well-informed, especially among the poor; that the skills required to execute them are often absent; and that the selection of certain uses may reflect merely the absence of others. In general, my belief is that Internet use should be governed by an active support framework which addresses these and other issues.1 To begin with, I review the evidence that supports the notion of an antidevelopment bias in the use of the Internet in developing countries. The section thereafter questions whether this bias reflects the ‘true’ preferences of users, or, whether instead in many cases there are too many imperfections for this to be the case. The final section then discusses the need for countervailing policies to redress the factors that impinge on the effective use of the mobile Internet in developing countries (especially among first-time users of it).
6.1 The Evidence The need to better understand the use of mobile Internet technology stems in part from the rapid rate at which these items are now growing and are projected to grow in developing countries.2 In Greater China, for example, smartphone adoption is predicted to grow from 72% in 2019 to 89% in 2025 (GSMA 2020). Beginning from a lower base of 45% in the former year, Sub-Saharan Africa is expected to exhibit especially rapid growth in the same technology to 67% in 2025 (GSMA 2020). For the purpose of discerning the existence of an anti-development bias in the use of this technology, I rely mainly on the cross-country evidence contained in Table 6.1 and also from a study of Africa (RIA 2019), and a number of country case studies. The data contained in Table 6.1 show both the relatively intensive use of activities that can be considered non-developmental (such as video calls and social media) 1 James
and Stewart (1981) have argued that developing countries tend to ignore issues related to products and suggest that they should have an active products policy. 2 The increase in the number of mobile Internet phones has been greatly boosted in India by the invention of the smart feature phone, which was specifically designed for developing country conditions, see Chap. 1.
6.1 The Evidence
73
Table 6.1 Use of the mobile Internet by poor countries, most and least popular uses Most
Activity (%)
Instant messaging
85
Social networking
76
Make or receive calls online
75
Make or receive video calls
64
Least
Activity (%)
Access information for education
26
Order or purchase goods online
21
Transfer money
19
Access information to improve health
18
Access government services
14
Source GSMA, The State of Mobile Internet Connectivity Report (2019)
and the under-use of activities that are developmental, in the conventional sense. The paradoxical element of this result derives from the expectation that relatively poor people would favour developmental uses with a direct and positive bearing on their physically deprived levels of well-being. I am thinking here for example of job information that might lead to increased income to purchase essential commodities or services, such as health care. Or again, there are relatively unused and plentiful opportunities to benefit from e-government in various ways that are designed specifically for the poor. Yet, for all their counter-intuitiveness the results just described bear a clear affinity to what was found in the other evidence referred to above. The study of African countries, conducted by the Research Institute Africa (RIA), for example, showed that Internet use is driven by the high demand for social media. ‘The 2017 After Access survey shows that of the 77 million Internet users in the surveyed countries, 70 million are on social media (Facebook, WhatsApp or Twitter), which represents 90% of Internet users. More than half (55%) of people using the Internet in the surveyed countries, excluding Rwanda, spend most of their time on social media. Only 21% of users access educational content on the Internet, while 15% use the Internet for work’3 (RIA 2019, p. 22). Let us next consider briefly the case studies of Tanzania, Côte-d’Ivoire and Nigeria. Because the first two countries were covered in the same study, I shall deal with them together (GMSA 2018). It is possible to be brief because the results comport closely with what has already been reported as an anti-development bias in the use of the mobile Internet. In fact, the entries in Table 6.2 reveal some of the most and least popular use cases in the two countries. They contribute to an overall impression that new users ‘have made the leap into mobile Internet use, but usage is 3 The
countries covered were 10 from Africa. Without referring to it, the Report provides evidence of cumulative causation. Thus, ‘without complementary policies, new digital technologies and Internet-based services simply amplify existing inequalities’ (RIA 2019, p. 1).
74
6 Anti-development Bias in the Use of the Internet in Developing …
Table 6.2 Percentage of mobile Internet users who use the following services on mobile at least once a month Use
Côte d’Ivoire
Tanzania
(a) Most popular Use social meda
77
64
Use instant messaging
71
58
Play games
63
46
Browse the Internet
61
48
Use map/transport apps
11
12
Order/purchase goods online
10
6
Pay for on-demand TV/movie
5
1
(b) Least popular
Source GSMA (2018)
often shallow and restricted to a few key applications’ (GSMA 2018, p. 36, emphasis added). Notice, to begin with, that (with one minor exception) the most and least popular use cases are the same in both countries (which, one should also note, are very different, not least because they are located in the diverse East and West parts of the continent). More salient, however, is the extent to which the results comport with those that have already been reported in this section and summarized by the (apparent) preference for leisure and entertainment over work and labour. According to the GSMA (2018, p. 40), for example, the two main uses of the mobile Internet are: ‘(1) Communicating with friends and family (both local and those further afield) is the greatest need fulfilled … and the best understood… WhatsApp and Facebook (including Messenger) are the dominant platforms. (2) Entertainment: Using mobile Internet to find and consume music and video content … particularly for certain demographic groups (urban, educated, youth)’ (GSMA 2018, p. 40). Enough has probably been said to convince the reader of a quite distinct pattern in the use of the mobile Internet in developing countries, at least among first-time users. For the sake of completeness, however, in a study by Wang (2020) of KaiOS, 819 newly connected respondents were surveyed from seven states in that country, accurately depicting its overall locational make-up. The conclusion is by now a familiar one, namely, that ‘As in most other markets around the world, one found that new users generally use phones for communication and entertainment. The most impactful uses — career development, personal health management, and business applications — are the least popular’ (Wang 2020, n.p.).
6.2 Discussion
75
6.2 Discussion Having hopefully established an anti-development bias in the results of the previous section, the key question becomes one of deciding how it should be interpreted. One extreme possibility, noted above, evolves out of the key notion in economics that the user is sovereign, that his or her preferences are revealed and not to be tampered with by those who purport to know better. Thus applied to the results above, one is bound to accept the finding that first-time users favour leisure over uses that would seem to confer more developmental value (as per, for example, The Economist cited above). At the heart of a competing view, however, is the distinction between informed and uninformed preferences. Not only, that is to say, might there be a striking distinction between these two types of preferences, ‘but also the need, in such cases, to reject one’s actual preferences in favour of one’s (hypothetical) informed preferences’ (James 1993). Harsanyi, for instance, has suggested that, a person’s utility function should be defined in terms of his hypothetical informed preferences rather than in terms of his actual preferences because some of the latter may be badly mistaken–this approach–permits us to bypass his possibly mistaken actual preferences which are contrary to his own, more fundamental, informed preferences, which can be considered as his actual preferences, as freed from the distorting effects of his factual errors. (Harsanyi 1992, pp. 6–7)
The same author asks us to consider in this context an example, of a patient ‘who has a bad case of pneumonia, for which A is the best medication. But he erroneously thinks that B is the best medication, even though the latter would be quite ineffective against the type of pneumonia he has’ (Harsanyi 1992, p. 6). In what follows, I argue that writ large, the problem of imperfect information contributes much to the antidevelopment bias described above, though the imperfections are broader and more complex than in the simple example just cited.4 Thus, by information here, I am not just alluding to what is needed in the traditional case of comparing 2 products, but also to knowing about which use cases exist on the Internet, and what they offer, but also knowledge about how to use them effectively (which includes knowledge about the language of the Internet) and what amount of digital literacy (and levels of literacy more generally) are required and available in each case. Note too that the first-time users, with whom I am mainly concerned, are likely to be especially disadvantaged in all these areas, since they will not have derived the many advantages of learning by doing, as regards knowledge acquisition in an unfamiliar setting. By extension, there is almost certainly a divide in these respects between rich and poor countries and between rural and urban locations in the latter.
4 Harsanyi argues ‘a person’s utility function should be defined in terms of his hypothetical informed preferences rather than in terms of his actual preferences because some of the latter may be badly mistaken’ (Harsanyi 1992, pp. 6–7, emphasis added).
76
6 Anti-development Bias in the Use of the Internet in Developing …
Table 6.3 Rural gap in mobile Internet use in low and middle-income countries Region
Gapa (2018) (%)
Global
40
GDP per capita
East Asiab and Pacific
23
7822.9
Europe and Central Asia (excl. high-income)
26
25,107
Latin America and Carribean
29
9044.2
Middle East and North Africa
36
8042.8
South Asia
45
1902.8
Sub-Saharan Africa (excl. high-income)
58
1585.8
Source GSMA (2019); World Bank data (2019) (GDP per capita current US dollars), license CC BY-3.0-IGO a The rural gap refers to how much less likely a person living in a rural area is to use mobile Internet than a person living in an urban area b Including high-income countries
I begin then with information about the Internet: what it offers, in which language(s) and where to find it. Problems with finding answers to all these questions derive ultimately from the fact that this technology is fundamentally Western, generated at the outset, as now, by and for advanced developed countries.5 This means, among other things, that the mobile Internet is designed for educated, urban-based, digitally skilled and English speaking, individuals, whose language dominates the technology. Translated into the context of developing countries, this observation leads one to expect a digital gap between rural and urban areas, as is confirmed by the findings in Table 2.4. It also implies that the rural-urban divide (or rural gap as it is called in Table 6.3) will tend to be most pronounced in the poorest regions, where income, education, skills and so on, differ most sharply from conditions in the rich countries. This expectation too, is broadly confirmed in Table 6.3, where, for example, the two poorest regions also suffer from the most pronounced divides. What is more relevant for my purposes, though, is not so much the relationship between income and the extent of mobile Internet in these countries, as it is between income and the pattern of mobile usage shown above in Table 6.3. For, what I contend is that the influence of income (and location) work closely through differential degrees of information about the Internet (see below).
5 With the notable exception of the smart feature phone noted in footnote 2, and discussed in Chap. 2.
6.3 Information Imperfections and Patterns of Internet Use
77
6.3 Information Imperfections and Patterns of Internet Use Some evidence on this issue can be found in a recent survey of Nigeria by Wang (2020). He notes, for example, the difference in knowledge between urban and rural inhabitants in that country, as it relates to patterns of Internet usage. Thus, ‘Urban residents tend to use the internet for a wider range of activities because they have more exposure to mobile internet through friends, family, school and work.’ Rural residents, on the other hand, do not have the same level of access, which means they have fewer real world examples of how the internet can be used in their communities (Wang, 2020, n.p.). Thus it is that difference in knowledge about the technology, impinge on the way it is used. What seem like differences in preferences (as suggested by The Economist 2019), are at least partly due to disparities in knowledge about how the Internet works and what it offers. Wang also points to the particular problems of information deficiencies faced by first-time users and the way these are reflected in different use patterns. In particular, he argues that ‘Nigerians have heard good things about the internet, but have yet to discover how to take advantage of all the benefits. As in most other markets around the world, we found that new users generally use phones for communication and entertainment. The most impactful uses–career development, personal health management, and business applications–are the least popular. Nigerians in rural areas are even less likely to understand how mobile internet can benefit them personally’ (Wang 2020, n.p.). In a brief but revealing passage, the author goes on to examine the role of social groups in determining the way in which the Internet is used in poor countries, especially their rural areas. Specifically, he observes that ‘Experienced internet users in developed markets are quick to Google how to do something, but new users turn to people they know personally and feel they can trust. Unfortunately, this means new users only learn what their communities can teach them. … Lack of knowledge causes Nigerians to use their internet-enabled phones in a limited way. Without exposure to activities like downloading apps, setting up online accounts, using web browsers, or making video calls, new users never learn all the internet has to offer them’ (Wang 2020, n.p.). More is required, however, than mere exposure to these activities. What are needed in addition, are the required levels of digital skills (basic, intermediate or advanced).6 Indeed, according to a survey of low and middle income countries conducted by the GSMA (2019), ‘A lack of literacy and digital skills is the top reason preventing consumers in LMICS from using mobile internet’ (GSMA 2019, p. 32). Such scarcities can be viewed as a lack of knowledge of one form or another. In the case of the Internet, it is largely technical knowledge about the operation of this technology, which can be divided into different degrees of complexity (ITU 2018). As noted in Chap. 2, moreover, even with regard to basic digital skills there is a substantial divide between the developed and least developed countries. 6 See
the ITU (2018) for a discussion of the differences between these skill levels.
78
6 Anti-development Bias in the Use of the Internet in Developing …
6.4 Causes of Scarce Digital Knowledge in Developing Countries Many and varied are the reasons for the scarcity of digital knowledge in developing countries and it is beyond the purview of this chapter to attempt a survey of all of them. In part, of course, scarcities of valuable resources are a defining feature of underdevelopment itself. The following citation, however, from the World Wide Web Foundation is more relevant in the present context, because it emphasizes those parts of the problem that are amenable to countervailing policy. As the Foundation puts it Not knowing how to use the internet continues to be a significant barrier to digital inclusion, particularly for women and girls. Among the urban poor, women are 1.6 times more likely than men to cite lack of know-how as a barrier to their internet access and use… Progress on providing internet access and digital literacy training in public schools has been painfully slow. (World Wide Web Foundation 2017, n.p.)
It is true that this excerpt was written in 2017, but the situation since then does not appear to have subsequently changed dramatically. According to the ITU, it is still the case, for example, that a lack of digital skills remains a major constraint on further use of the mobile Internet (ITU 2019). ‘In 40 out of 84 countries for which data are available, less than half the population possesses basic computer skills such as copying a file or sending an e-mail with an attachment’ (ITU 2019, p. 10, emphasis added).
6.5 Knowledge Imperfections and Preferences for Internet Use The purpose of this synthesis section is to bring the above discussion to bear on the main argument of the chapter, namely, that the preferences for different uses of the Internet are not so much anti-developmental or frivolous, as they are made in a situation of highly uninformed choices. And rather than accept the conclusion that users harbour a preference for leisure over work, one should seek rather to redress the underlying situation of highly uninformed preferences (which, in the discussion above were not considered legitimate). So what appears to The Economist as an admirable choice of entertainment over work in various consumer surveys of Internet use, on the part of low-income consumers, reappears instead as an acute shortage of digital skills, with which to undertake more developmental activities (which are apt to appear at or near the bottom of the list of preferred options). The poor consumer, that is to say, generally tends to choose options on the basis of the ease with which they can be mastered, rather than some perverse anti-developmental set of preferences. I should stress, however, that the proposed relationship is meant to be interpreted as a tendency, rather than an iron law.
6.5 Knowledge Imperfections and Preferences for Internet Use
79
Left to itself, moreover, this tendency is likely to exhibit certain cumulative tendencies which were described many years ago as ‘cumulative causation’ by Myrdal (1957). His idea was basically that the market mechanism does not tend to rectify initial inequalities in the social system, but rather to exacerbate them. In his Economic Theory and Under-Developed Regions, for example, he proposes that in general, ‘there is no such tendency towards automatic self-stabilisation. … The system is by itself not moving towards any sort of balance between forces, but is constantly on the move away from such a situation’ (Myrdal 1957, p. 155). Thus, with regard to digital skills, those who are initially well-endowed with them, will tend to benefit cumulatively as they employ their knowledge to use education and other projects to acquire even more skills. Those with only limited skills, on the other hand, will tend to remain in a poverty trap, with stagnant digital abilities. Such persons, as noted above, will tend to emanate from the rural sector, surrounded by social groups which are versed largely in the simpler uses of the Internet and do not always hold a positive view of this technology. So far, I have drawn attention to possible differences in user preferences between rural and urban areas. However, there appear to be other differences with regard to users who exhibit anti-development bias and those who do not. Such other differences, I should emphasise, may be important for the design of policies to shift preferences in favour of more developed-oriented uses on the Internet. Some of the literature on user differences has been reviewed in an article by Correa (2016), which also includes a case-study of Internet uses in Chile. I shall not, however, attempt here to cover the entire article. My purpose is rather to focus on those parts of it that throw light on the choice between uses by different individuals or groups. Such research, which falls in the tradition of what was referred to earlier as the second digital divide, has found, for example, ‘socioeconomic differences in frequency and breadth of actions that require involvement and technological skills, such as content creation and educational, economic, or political Internet activities, even among young people. Those with lower levels of education and income, tend to engage in less skilful activities’ (Correa 2016, n.p.). Research has also revealed that ‘expressive, informational, and mobilizing uses of social media are associated with more meaningful outcomes, including increased social capital and political and civic participation, than consumptive and entertaining uses’ (Correa 2016, n.p.). Perhaps most telling, however, is the finding that ‘the relationship between sociodemographics, skills- and Facebook use depends upon types of Facebook use. More educated and skillful individuals tend to use Facebook in more expressive and potentially beneficial ways than lower educated and less skilful people. Specifically, it was found that people from more privileged backgrounds who have developed greater digital skills use Facebook for informational and mobilizing purposes’ (Correa 2016, n.p.). Taken together, these findings lend a measure of support to the hypothesis advanced above, that the choice of less developmental uses of the Internet by the relatively disadvantaged groups, has more to do with a lack of digital skills than seemingly perverse preferences. For, what the research seems to show, is on the one hand, that developmental uses of the Internet are more demanding in terms of
80
6 Anti-development Bias in the Use of the Internet in Developing …
digital skills than entertainment uses, and on the other hand, that the less priviledged groups are especially disadvantaged in terms of possession of these skills. Without policy intervention, moreover, the process will tend to cumulate, as the benefits from advanced digital skills enable the privileged members of society to acquire yet more skills and so on and so on. Countervailing policy should focus, to a much greater extent than it now does, on providing lessons in digital literacy in (especially rural) schools and for uneducated adults on television and the Internet itself.
6.6 Policy Implications Evidence presented in the first part of the chapter, strongly suggests that Internet users in developing countries favour entertainment over more developmental uses. One influential observer has suggested that this finding is indicative of a revealed preference for these activities and goes so far as to commend the observed pattern. I argue, by contrast, that there is a need to investigate what lies behind the ‘preferences’ of users rather than simply to accept them as given. So doing, I suggest, leads to the finding that what drives the observed pattern is a serious lack of knowledge about the Internet and how to use it, especially in rural areas of poor countries (and especially among first-time users of the technology). Gaming, for example, is an extremely popular form of use in many developing countries, but it should be accompanied, at a minimum, by warnings over the dangers of addiction, to which it may sometimes give rise. More generally, as the GSMA (2018) points out, ‘New users have made the leap into mobile internet use, but usage is often shallow and restricted to a few key applications’ (GSMA 2018, p. 36).7 Certainly, there is a perception among major UN agencies that digital skills are not getting the attention they deserve. According to the ITU, for example, ‘we know that investments in building digital skills are falling far short of needs. Even in the world’s wealthy nations, millions of students still aren’t getting access to technology in the classroom and at home. In the developing world, access is even more limited. Without digital skills, these young people are being left behind in a world that grows even more digital by the day’ (ITU 2019, n.p.).8 The main problem is arguably the lack of countervailing policy on the part of governments in developing countries. And this in turn may be due in large part to the legacy of a focus on only the technological aspects of the digital divide, i.e. on gaining access to the technologies themselves, without regard to the skills needed to operate them effectively (that is, on the first rather than the second digital divide).9 7 It
bears emphasis that the anti-developmental bias documented above, does little to enhance the Sustainable Development Goals. 8 See also the Digital Skills Toolkit by the ITU (2018). 9 As the IFC (2019, p. 10) sees it, for example, though many countries in Sub-Saharan are cognizant of the centrality of digital skills, ‘fewer have translated this into a clear-cut agenda’.
6.6 Policy Implications
81
There have, however, been some major exceptions to the general picture of governmental inactivity in this area. One of the most prominent of them is the Indian government’s attempt to bring digital literacy training to 60 million rural households. Initiated by the Prime Minster in 2016, this endeavour is ambitious not only in the number of its target households, but also the fact that these units reside in rural areas, where conditions are especially unamenable to improving digital skills, as noted above. In Kenya, as well, training in digital skills has been provided to over 91,000 primary school teachers. As far as I am aware, however, neither endeavour has been subject to rigorous evaluation and much more work in this area consequently needs to be carried out. More promising efforts to impart digital skills at the school level, however, may be the initiatives undertaken by private technology operators, because the latter are generally better informed about the habits and propensities of Internet users. From this point of view, a good example is the programme called ‘LIFE’ designed by KaiOS Technologies for use with its smart feature phones, the best known of which is the JioPhone in India.10 From the point of view of this chapter, what is striking about LIFE is that it promotes digital literacy from a distinctly developmental perspective. In particular, ‘it serves as a trusted educational resource for first-time internet users, and provides curated content in six categories, including Digital Skills, Education, Health, Gender Equality, Agriculture and Financial Education. These resources help users navigate their internet experience, and are optimized for the non-touch screens of KaiOS-powered devices; as well as gain access to valuable information that can improve their lives’ (Meta 2019, n.p.). Whether users of LIFE do indeed adopt a more developmental pattern of Internet use than the one described above, is a question that remains thus far unanswered, but it is one that would surely repay survey research. Another private initiative to promote digital literacy has recently begun in India. Known as “Digital Udaan’, and promoted by Reliance Jio, it has been launched in roughly 200 locations in the country, covering 13 states (Livemint 2019). The programme is designed for first-time users, who, as noted above, are particularly susceptible to having underdeveloped digital skills. The goal of the programme is to ‘empower’ such users ‘with digital literacy and understanding of the Internet’ (Indian Express 2019). Towards this end, Reliance Jio will provide workshops at its Jio Stores and Jio Centres each week. It is notable that instruction will be given in 10 regional languages. To conclude the chapter, I redraw attention to an assumption that was been made throughout, which needs to be seriously examined. It is that the Internet uses described as developmental, were actually eschewed by first-time users of the Internet. That is to say, that such uses were actually available for selection. For, if this is not the case, then no preference ordering can be inferred from the choices that are made. And it is far from true in reality that information about telemedicine and government services are universally available in developing countries and in 10 See
the study by James (2020) in the Information Society. This article appears as Chap. 2 above.
82
6 Anti-development Bias in the Use of the Internet in Developing …
local languages. For this reason, too, therefore, data which appear to reveal an anti-development bias among Internet users may actually not do so.
6.7 Summary and Conclusions In this chapter I have challenged the legitimacy of interpreting the conclusion of several recent papers that find an anti-development bias in the way the Internet is used in developing countries. More specifically, I interpret the apparent bias as reflecting a major knowledge gap on the part of first-time users of this technology, especially those in rural areas. Put yet another way, the point is that the observed preferences are highly questionable from a welfare point of view, because they are based on imperfect knowledge and a lack of digital skills on the one hand and a possible inadequacy of the supply of more developmental options (as when say e-medicine is unavailable in certain locations) on the other. Arguably, the most binding constraint on the choice of developmental uses is a severe lack of digital skills, especially on the part of first-time users of the Internet. Much of this situation in turn may be due to the continuing focus of developing country governments on gaining access to the technology, rather than a concern with the factors that determine how it is used and the benefits that are derived therefrom. In any case, however, governments have paid relatively little attention to enhancing digital skills among first-time users in poor countries, as suggested by the discussion above. Left largely to their own devices, therefore, such users, are wont to choose a range of relatively unsophisticated uses that are relatively easy to master and result in what some authors refer to as a ‘shallow’ range of choices. Dominated in many cases by entertainment. What needs to be borne in mind in redressing this situation is first of all that the process of acquisition of digital skills tends to be a cumulative one, that is, that initial inequalities tend, without countervailing policy, to intensify over time (as described above in relation to the theory of cumulative causation). Countervailing policy is thus best aimed at reducing initial inequalities, before the cumulative process has set in too corrosively. Then, too, if my analysis about the role of information in shaping preferences is correct, particular attention needs to be paid to imparting the digital skills needed specifically to appreciate and master developmental uses of the Internet, as described above in relation to two private initiatives in developing countries. For, whereas a great deal of information (including advertising) is usually devoted to entertainment and other non-developmental uses, there is a relative dearth of attention devoted to the more developmental activities. These, moreover, need to be cognisant of the wide gender gap11 referred to above and the specific problems faced by women in the acquisition of skills that are oriented to such activities.
11 According
to the GSMA (2019), women are 23% less likely than men to use the mobile internet.
References
83
References Correa T (2016) Digital skills and social media use: how internet skills are related to different types of Facebook use among ‘digital natives’. Inf Commun Soc 19(8):1095–1107. Available https:// www.tandfonline.com/doi/full/10.1080/1369118X.2015.1084023 GSMA (2018) Triggering mobile internet use in Côte d’Ivoire and Tanzania. Available https://www.gsma.com/mobilefordevelopment/wp-content/uploads/2018/07/GSMA-Trigge ring-Mobile-Internet-Use-in-Cote-DIvoire-and-Tanzania.pdf. Retrieved 8 July 2019 GSMA (2019) The state of mobile internet connectivity report. Available www.gsma.com/resour ces/the-state-of-mobile-internet-connectivity-report-2019. Retrieved 5 Jan 2019 GSMA (2020) The mobile economy. Available https://www.gsma.com/mobileeconomy/wp-con tent/uploads/2020/03/GSMA_MobileEconomy2020_Global.pdf. Retrieved 4 Feb 2020 Harsanyi JC (1992) Utilities, preferences and substantive goods. UNU Wider working papers, no. 101. Available https://www.wider.unu.edu/sites/default/files/WP101.pdf. Accessed 5 Oct 2000 International Finance Corporation (IFC) (2019) Digital skills in Sub-Saharan Africa: spotlight on Ghana. Available https://www.ifc.org/wps/wcm/connect/ed6362b3-aa34-42ac-ae9f-c73990495 1b1/Digital+Skills_Final_WEB_5-7-19.pdf?MOD=AJPERES International Telecommunication Union (ITU) (2018) The digital skills toolkit. Available https:// www.itu.int/en/ITU-D/Digital-Inclusion/Documents/ITU%20Digital%20Skills%20Toolkit.pdf, Geneva International Telecommunication Union (ITU) (2019) Measuring digital improvements. Available https://www.itu.int/en/ITU-D/Statistics/Documents/facts/FactsFigures2019.pdf, Geneva James J (1993) Consumption and development. Macmillan James J (2020) The smart feature phone revolution in developing countries: bringing the internet to the bottom of the pyramid. Inf Soc 36(4):226–335 James J, Stewart F (1981) New products: a discussion of the welfare effects of the introduction of new products in developing countries. Oxford Econ Pap 33(1):81–108 Livemint (2019) ‘Digital Udaan’: Reliance Jio ties up with Facebook to teach how to use internet to first timers. Available https://www.livemint.com/technology/tech-news/-digital-udaan-reliancejio-ties-up-with-facebook-to-teach-how-to-use-internet-to-first-timers-1562211608820.html. Accessed 23 Aug 2019 Meta A (2019) Life, the KaiOS in-house app that bridges the gap to educational resources, is now live in Africa. Available https://www.kaiostech.com/life-the-kaios-in-house-app-that-bri dges-the-gap-to-educational-resources-is-now-live-in-africa/. Accessed 3 Feb 2020 Myrdal G (1957) Economic theory and under-developed regions. G. Duckworth, London Research ICT Africa (RIA) (2019) After access: Africa comparative report. Available https://resear chictafrica.net/2019/05/16/after-access-africa-comparative-report/. Accessed 10 Jan 2020 The Economist (2019) A global timepass economy. Available https://www.economist.com/briefing/ 2019/06/08/how-the-pursuit-of-leisure-drives-internet-use. Accessed 18 Dec 2019 The Indian Express (2019) Reliance Jio’s Digital Udaan initiative wants to teach first-time users benefits of internet. Available https://indianexpress.com/article/technology/tech-news-techno logy/reliance-jios-digital-udaan-initiative-wants-to-teach-first-time-users-benefits-of-internet5813599/. Accessed 5 Jan 2020 Wang Y (2020) First-time internet users in Nigeria use the internet in a unique way. Available https://www.kaiostech.com/author/yangwang/. Accessed 14 April 2020 World Bank Data (2019) World Wide Web Foundation (2017) Why it’s time to prioritise digital literacy. Available https:// webfoundation.org/2017/09/why-its-time-to-prioritise-digital-literacy/. Accessed 18 Sept 2019