The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria December 2020 The GSMA represents the interests of mobile operators The World Bank Group is one of the world’s largest worldwide, uniting more than 750 operators with almost sources of funding and knowledge for developing 400 companies in the broader mobile ecosystem, including countries. It consists of five institutions with a shared handset and device makers, software companies, equipment commitment to reducing poverty, increasing shared providers and internet companies, as well as organisations prosperity, and promoting sustainable growth and in adjacent industry sectors. development. International Bank for Reconstruction and Development (IBRD) lends to governments of The GSMA also produces the industry-leading MWC events middle-income and creditworthy low-income countries. held annually in Barcelona, Los Angeles and Shanghai, as International Development Association (IDA) provides well as the Mobile 360 Series of regional conferences. financing on highly concessional terms to governments of For more information, please visit the GSMA corporate the poorest countries. International Finance Corporation website at www.gsma.com. (IFC) provides loans, equity, and advisory services to stimulate private sector investment in developing Follow the GSMA on Twitter: @GSMA countries. Multilateral Investment Guarantee Agency (MIGA) provides political risk insurance and credit enhancement to investors and lenders to facilitate foreign direct investment in emerging economies. International Centre for Settlement of Investment Disputes (ICSID) provides international facilities for conciliation and GSMA Intelligence is the definitive source of global mobile arbitration of investment disputes. operator data, analysis and forecasts, and publisher of authoritative industry reports and research. Our data covers every operator group, network and MVNO in every GSMA Connected Society country worldwide – from Afghanistan to Zimbabwe. It is the most accurate and complete set of industry metrics The Connected Society programme works with the mobile available, comprising tens of millions of individual data industry, technology companies, the development community points, updated daily. and governments to increase access to and adoption of GSMA Intelligence is relied on by leading operators, mobile internet, focusing on underserved population groups vendors, regulators, financial institutions and third-party in developing markets. Key activities include: industry players, to support strategic decision-making Generating and disseminating insights and learnings •  and long-term investment planning. The data is used as on the mobile internet coverage and usage gap. an industry reference point and is frequently cited by the media and by the industry itself. Supporting mobile operators to extend coverage and •  drive usage. Our team of analysts and experts produce regular thought-leading research reports across a range of Undertaking advocacy and policy work to ensure •  industry topics. that mobile operators’ efforts to achieve greater digital inclusion are being effectively supported by www.gsmaintelligence.com governments, the international community and other info@gsmaintelligence.com stakeholders. For more information, please visit www.gsma.com/connected-society This material has been funded by UK aid from the UK government; however the views expressed do not necessarily reflect the UK government’s official policies. Acknowledgements This brief was prepared by a team co-led by World Bank authors’ workshop for the Africa Digital Transformation and GSMA staff. The core team included Kalvin Bahia Flagship of the Africa Region of the World Bank, and the (GSMA Intelligence), Pau Castells (GSMA Intelligence), United Nations Development Programme TRAP (Topics for Genaro Cruz (GSMA), Takaaki Masaki (World Bank), Xavier Research and Acceleration Policies) virtual seminar. The Pedros (GSMA Intelligence), Tobias Pfutze (World Bank team would also like to thank mobile network operators in and Florida International University), Carlos Rodríguez- Nigeria for sharing their mobile network infrastructure data. Castelan (World Bank) and Hernan Winkler (World Bank). This study received funding from the Digital Development In addition, Siobhan Murray (World Bank) contributed Partnership and from UK aid from the UK government; to data integration, Aibek Baibagysh (World Bank) and however, the views expressed do not necessarily reflect the Abul Azad (World Bank) provided advice on building the UK government’s official policies. consumption aggregate for the Nigeria household survey, Elizabeth Howton (World Bank) advised on communication This work is a product of the staff of The World Bank with and messaging, and Karem Edwards (World Bank) provided external contributions. The findings, interpretations, and administrative support. conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive The team would like to thank Andrew Dabalen, Mark Dutz, Directors, or the governments they represent. The World Rogelio Granguillhome, Jonas Hjort, Samantha Lach, Bank does not guarantee the accuracy of the data included David McKenzie, Fausto Patino, Guido Porto, Chandni Raja, in this work. The boundaries, colors, denominations, Viviane Sanfelice, Emmanuel Skoufias, Daniel Valderrama and other information shown on any map in this work and Tara Vishwanath for their comments and suggestions. do not imply any judgment on the part of The World The authors would also like to thank participants of the Bank concerning the legal status of any territory or the World Bank seminar organised by the Poverty and Equity endorsement or acceptance of such boundaries. and Digital Development Global Practices, the virtual Attribution World Bank, GSMA. 2020. “The Poverty Reduction Effects of Mobile Broadband in Africa: Evidence from Nigeria.” 3 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria 4 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria Contents Executive summary................................................................................................... 6 1 The role of mobile broadband in poverty reduction ............................ 8 2 Nigeria case study.............................................................................................. 10 3 Policy implications............................................................................................. 15 Appendix: Methodology....................................................................................... 17 References ................................................................................................................. 18 5 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria Executive summary The COVID-19 pandemic has brought home the To answer this question, the GSMA and the World salience of digital technologies as vital elements of Bank developed an innovative, ground-breaking today’s economic and social systems. Technology study to subject to empirical scrutiny the impact plays an integral role in spurring economic growth, of mobile broadband on welfare and poverty helping to create new job opportunities, promote reduction, using data from Nigeria2 – the largest greater efficiency and generate innovation. mobile market and economy in Africa. The analysis shows that having at least one year of mobile Mobile technology, in particular, constitutes the broadband coverage increases total consumption primary – and often only – method of digital by about 6%. This estimate reaches 8% after two access for people around the world, with almost years of coverage. Similarly, the proportion of 4 billion mobile broadband subscribers globally.1 households below the extreme poverty line ($1.90 It plays a particularly important role in Africa, per day) drops by about 4 percentage points after where a large majority of people access the internet one year of gaining mobile broadband coverage through mobile phones rather than through fixed and by about 7 percentage points after two or more broadband. That said, Africa also lags behind in the years of coverage. This corresponds to moving rollout of mobile broadband infrastructure. Almost a approximately 2.5 million people out of extreme quarter of the continent’s population is not covered poverty. The study also shows that rural households by a mobile broadband network and almost three in particular stand to gain more than urban quarters do not use mobile internet, meaning they households from such benefits in relative terms. are unable to benefit from the digital revolution that has swept across the world over the past Mobile technology – and mobile broadband in two decades. particular – can play a significant role in poverty reduction. While it remains to be seen whether Against this background, however, there is little mobile broadband has had equivalent effects in evidence that speaks to the question of how other African countries, it is important to accelerate access to mobile broadband can affect the welfare the rollout of mobile broadband infrastructure in of people in Africa – particularly those living in those countries lagging behind. poverty. This has become increasingly important as there are signs of widening digital gaps in access to the internet across countries as well as along demographic and socioeconomic lines, which could exacerbate existing inequalities. 1. GSMA, The State of Mobile Internet Connectivity Report 2020 2. This short paper presents the main findings of the GSMA and World Bank's study. For detailed results and methodology download the full study 6 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria These findings call for policymakers, international organisations and other decision makers to put the development of digital infrastructure at the heart of strategies to reduce poverty and promote inclusive growth in Africa. Three policy lessons can be derived from this work:  xpanding broadband coverage to rural and remote areas can reduce extreme poverty. Globally, there are still 1. E almost 600 million people that do not live within reach of either a 3G or 4G network – around half of them in Africa. Most of them live in rural and remote areas. Bridging the digital divide between rural and urban areas and providing universal mobile broadband coverage can be an effective instrument to alleviate inequality and reduce poverty.  olicies to address the usage gap across socioeconomic groups are necessary to ensure everyone 2. P benefits from digital technologies. In Nigeria, as in many parts of the world, inequalities persist in access to mobile broadband services by gender, location and education, as well as between people with and without disabilities. For example, in 2019 the gender gap in mobile internet usage in Nigeria was 29% (meaning that women were 29% less likely than men to use mobile internet). These underserved groups face a multitude of barriers to fully reaping the benefits of mobile internet, including limited purchasing power, affordability of services and devices, relevance of content, lack of digital skills, and online safety and security.3  omplementary policies are needed to realise the social and economic benefits of digital connectivity 3. C for all. Our study suggests that increased labour force participation is one of the key mechanisms through which mobile broadband infrastructure enhances the welfare of individuals and households. Expansion of digital infrastructure and access to the internet can not only help create more jobs, but can also reduce transaction costs for people finding jobs and can improve productivity.4 Well-functioning markets, strong institutions and complementary investments including road infrastructure, competition in input and output markets, and high-quality public services, help the deployment of infrastructure in rural and remote areas and need to interact with mobile broadband access to maximise its potential welfare effects. Nigeria mobile market: Impact of mobile broadband coverage: Household total consumption increases by: 103m mobile subscribers 6% after 1 year of 3G/4G coverage and by 8% after 2 years 68m mobile internet subscribers The proportion of households in extreme Effects comparable with moving 2.5m poverty decreases by: 3G coverage: 4 percentage points after 1 year of coverage people 7 and by out of extreme 2010 2015 2020 percentage points poverty between after 2 years 2010 – 2015 21% 57% 78% Welfare effects are more pronounced 3G/4G coverage improves welfare among rural through labour force households participation. 3. See GSMA, State of Mobile Internet Connectivity, 2020, for detailed data and analysis on these barriers. 4. Other studies also show that the internet can help create new job opportunities and increase labour productivity. See Paunov and Rollo 2014; Fernandes et al. 2019; Chun and Tang 2018; Hjort and Poulsen 2019; Viollaz and Winkler 2020. 7 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria 1  The role of mobile broadband in poverty reduction Over the past three decades, the world has made by a mobile broadband network10 and almost three considerable progress in reducing extreme poverty, quarters do not use mobile internet, meaning they are with the share of people living on less than $1.90 per unable to benefit from the digital revolution that has day (in 2011 purchasing power parity, PPP) falling from swept across the world over the past two decades. 36% in 1990 to roughly 10% in 2015, or approximately 736 million people (World Bank, 2018). Despite this In this context, the objective of this brief is to illustrate impressive progress, poverty remains ubiquitous the potential role mobile broadband can play in across much of Sub-Saharan Africa, which hosts 27 accelerating the pace of poverty reduction in Africa, of the 28 poorest countries in the world and was based on evidence from a case study of Nigeria. The home to 41% of people living in extreme poverty COVID-19 pandemic has brought home the salience in 2015. Also, as of 2015 about one in eight of the of digital technologies as vital elements of today’s world’s extreme poor resided in Nigeria (World Bank, economic and social systems. From teleworking and 2018). Under a business-as-usual scenario, the region e-commerce to telemedicine and remote learning, will likely fall short of achieving the United Nations digital technologies have supported continued Sustainable Development Goal of eradicating poverty access to education, healthcare, essential goods and by 2030.5 On top of this, the World Bank recently services, and family and friends. Businesses have estimated that the current COVID-19 pandemic and also had to accelerate digitisation, reshape the way its associated economic crisis could push between they are run and how they interact with the rest of 71 and 100 million people into extreme poverty.6 the economy. Therefore, expanding the evidence base on the importance of mobile broadband to With more than 5 billion subscribers of mobile services improving livelihoods is a priority policy issue. and almost 4 billion subscribers of mobile internet globally,7 mobile technology has been the primary – A rapidly growing body of research shows that and often only – method of digital access for people the internet – particularly mobile broadband – has around the world, but especially in low- and middle- widespread positive macroeconomic impacts, income countries. According to the International improving the utilisation of labour and capital, and Telecommunications Union (ITU), in 2019, mobile increasing productivity. For example, recent studies accounted for 87% of broadband connections in have shown that a 10% increase in broadband developing countries.8 It plays a particularly important penetration raised annual per-capita growth by role in Africa, where a large majority of people access 0.9–1.5% in OECD countries, and provide a 1.5–2.5% the internet through mobile phones rather than through increase in GDP in developing countries.11 A study fixed broadband.9 That said, Africa also lags behind in by GSMA Intelligence shows that the positive the rollout of mobile broadband infrastructure. Almost economic impact of mobile technology increases a quarter of the continent’s population is not covered as it is upgraded from 2G to 3G12 and 4G.13 5. Beegle and Christiaensen (2019) 6. Mahler et al. (2020) 7. GSMA, The State of Mobile Internet Connectivity Report 2020 8.  Source: ITU. Developing countries in this analysis are categorised according to UN classifications. 9.  The number of active mobile broadband subscriptions per 100 inhabitants in Africa in 2017 was 34.0, while the corresponding rate for fixed broadband subscriptions was 0.4 (ITU 2019). 10. Mobile broadband is defined as the provision of 3G or 4G coverage, which enables high-speed access to the internet, and excludes 2G coverage as it only provides for limited internet browsing and applications. 11. Czenich et al., 2011 for OECD countries; ITU, Katz and Callorda (2018) and ITU (2019) 12. 2G technologies enable voice, SMS, and limited internet access, while 3G technologies or above enable more rapid internet browsing and data downloading. 13. Bahia, Castells and Pedros (2020). 8 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria However, there is little evidence that speaks to the of consumption than those that do not, and mobile question of how access to mobile broadband can operators are more likely to deploy networks in higher- affect the welfare of individuals and households income areas that have higher expected returns. at the micro level, particularly among the extreme poor. In fact, only a few studies subject the impact of Furthermore, the benefits to be gained from the mobile broadband on welfare to rigorous empirical internet may not necessarily accrue to people testing.14 One reason is a lack of high-quality and uniformly; there can be significant gaps in terms of sufficiently granular data available over time, both in who stands to benefit. For instance, some evidence terms of access to mobile broadband and individual suggests that the internet may benefit more measures of welfare for the same individuals or educated individuals who are better able to realise households. A second reason is that the direction the full value of digital technologies.15 This could of impact can work both ways. Individuals that use exacerbate digital divides and widen inequalities in mobile broadband may already have higher levels other socioeconomic indicators, such as education.16 14. Of the few studies that have assessed the causal impact of digital technology, especially in low- and middle-income countries, the focus has generally been on mobile coverage (including 2G) and mobile money. While these represent important advances, there is a lack of evidence that considers the causal impact of mobile internet in a rigorous manner. For a detailed list of references on this topic, see Bahia et al. (2020). 15. See World Bank (2016) and Hjort and Poulsen (2019) 16. See Autor and Acemoglu (2011) for a comprehensive review of the literature on the impacts of information and communication technologies on the labour market. 9 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria 2 Nigeria case study In a recent joint GSMA and World Bank study, we rapid expansion in mobile broadband penetration analyse the welfare impact of mobile broadband based over the past decade, which mirrors the overall trend on data from Nigeria, the largest mobile market and for the entire Africa region (Figure 1). At the end economy in Africa.17 The country offers an interesting of 2019, there were more than 170 million mobile case study to test a potential causal linkage between connections, 60% of which used 3G or 4G technology. mobile broadband and welfare. Nigeria has witnessed Figure 1 Mobile broadband penetration, based on connections, 2010–2019 80% 70% World 60% Africa Mobile Broadband Penetration Nigeria 50% 40% 30% 20% 10% 0% 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Source: GSMA Intelligence. Penetration is calculated by dividing the total number of 3G/4G connections by total population. A mobile connection is a unique SIM card (or phone number, where SIM cards are not used) that has been registered on a mobile network. Connections differ from subscribers as a unique subscriber can have multiple connections. Within Nigeria, mobile broadband coverage expanded 2%, reaching just over 5% in 2016. Mobile technology rapidly from a few main cities initially to a large has therefore been the primary platform for users in number of intermediate cities and rural areas between Nigeria to access broadband services. Based on these 2010 and 2015 (see Figure 2). This represented an facts, the goal of this study was to identify whether increase in population coverage from 21% to 57%.18 households covered by 3G have experienced any gains By contrast, during the same period, penetration in their overall welfare (or consumption) and whether of fixed broadband in the country averaged around the rollout of 3G has resulted in a reduction in poverty. 17. GSMA and World Bank, The Welfare Effects of Mobile Broadband Internet: Evidence from Nigeria (2020) 18. Note that coverage is distinct from usage. The latter refers to when an individual has a SIM card that can be used in a mobile phone to access the internet. Coverage refers to when an individual lives within range of a 3G/4G cellular signal, irrespective of whether or not they use mobile. 10 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria Figure 2 Changes in 3G coverage area within Nigeria, 2010–2015 A) 2010 B) 2015 Source: GSMA Mobile Coverage Maps (www.mobilecoveragemaps.com) Blue = areas covered by 3G. The main issue when identifying the effects of mobile and jobs as well as other types of services such as broadband on household welfare is separating the education and healthcare, which would explain better impact solely attributable to internet access from observable outcomes for households in urban areas. other confounding factors that may simultaneously affect internet access and welfare outcomes. For To alleviate this concern, the study utilised longitudinal instance, urban areas tend to have greater access to data of household-level consumption and poverty 3G compared to rural areas. When simply comparing collected over three waves of the General Household households with and without 3G coverage, one Survey (GHS) of Nigeria (2010/11, 2012/13 and 2015/16). would expect to see those covered by 3G enjoying a One of the key advantages of the data is that the higher level of welfare and lower poverty compared same households were tracked across the three waves, to those not covered by 3G. But this does not mean allowing us to observe how the welfare level of each that 3G coverage per se has caused the former household changed over time. Next, we merged to be wealthier than the latter. Urban areas also this data with the high-resolution coverage maps provide better access to economic opportunities of mobile broadband networks provided by mobile To produce 3G/4G coverage maps for Nigeria over time, we collected network infrastructure data directly from three of the four largest mobile network 19.  operators (MNOs) in Nigeria and calculated mobile broadband coverage based on a radio propagation model – a tool used by MNOs to plan the deployment and coverage of their infrastructure. 20. The data is matched with the maximum offset of 45 metres. 11 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria operators19 based on the (almost) exact locations faster growth in welfare.21 Furthermore, we studied of households.20 Using this unique dataset, we can several sub-samples to explore heterogeneous compare those households that gained coverage of effects by: gender; initial level of consumption of each mobile broadband technologies over time (or during household; region; age; area (urban/rural) of residence; the period of study) and those that had not yet education of the household head or respondent; and gained coverage; and test if the former experienced the presence of children under the age of three. Main results Our analysis shows that the deployment of mobile broadband infrastructure has played a significant role in increasing household welfare and reducing poverty in Nigeria. The welfare effects are different depending on how long households have been exposed to coverage: Total consumption of households is expected to increase by around 6% for at least one year of 3G/4G •   coverage and reaches 8% after two years of coverage (Figure 3a). The proportion of households in extreme poverty declined by about 4 percentage points for at least one •   year of coverage and about 7 percentage points for at least two or more years of coverage (Figure 3b). The result corresponds to moving approximately 2.5 million people out of extreme poverty. The welfare effects of 3G/4G are particularly pronounced among rural households (Figure 4). •  One of the key mechanisms through which 3G/4G coverage affects welfare is its impact on labour force •   participation. 21. Even then, there is still a potential concern with our difference-in-difference strategy. For instance, it is plausible that the rollout of mobile broadband is not random over time, since operators may target areas that are more prosperous and growing faster, which in turn are likely to have faster growth in consumption levels. However, our results turn out to be robust to alternative approaches that strongly suggest a causal effect. Indeed, the consumption and poverty trends for both households that eventually are and are not covered by mobile broadband are similar before the rollout of mobile broadband; and the results also hold when looking at the impact on households that receive mobile broadband coverage unintentionally due to surrounding terrain features – which should not be directly correlated with welfare. 12 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria Figure 3a Figure 3b Impact of 3G/4G access on total Impact of 3G/4G access on poverty ($1.90) consumption (point estimates) (point estimates) 16% Poverty $1.90 (percentage point change) Total consumption (percentage change) 14% 0% 12% -2% 10% -4% 8% -6% 6% -8% 4% 2% -10% 0% -12% One or more Two or more Three or more One or more Two or more Three or more Years of 3G/4G coverage Years of 3G/4G coverage Point estimate Confidence interval (95%) Point estimate Confidence interval (95%) Source: Bahia et al. 2020 Notes: Point estimates at 5% confidence interval. Difference-in-difference estimators across total consumption and poverty. Standard errors are clustered by Local Government Authorities (LGAs). Additional controls include access to electricity, ownership of dwelling, household size and a wealth index, although the coefficients for these variables are not reported to save space. Poverty measure is equal to 1 if the household per-capita consumption is less than the international extreme poverty line of $1.90 per day (measured in purchasing power parity). 13 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria Figure 4 Impact of three or more years of 3G/4G access on poverty for different sub-samples ($1.90) (point estimates) 10% 5% Poverty $1.90 (percentage 0% point change) -5% -10% -15% -20% No Yes Low High South North <50 >50 Rural Urban Primary Less than No Yes or more primary Female Total household Geographic Age of Household Education level Households with headed consumption region of household head locality type of household children 3 years household household head old or younger Three or more years of 3G/4G coverage Point estimate Confidence interval (95%) Source: Bahia et al. 2020 Notes: Point estimates at with 95% confidence intervals. Difference-in-difference estimators across poverty by sub-sample. Standard errors are clustered by Local Government Authorities (LGAs). Additional controls include access to electricity, ownership of dwelling, household size and a wealth index, although the coefficients for these variables are not reported to save space. Poverty measure is equal to 1 if the household per-capita consumption is less than the international extreme poverty line of $1.90 per day (measured in purchasing power parity). The study measured the impact across all households mobile broadband coverage in this period already had living in areas that received mobile broadband access to basic mobile coverage (voice and text), the coverage in Nigeria, independently of whether results of the analysis reflect mostly the impact from they adopted the technology or not. It is therefore upgrading mobile coverage from 2G to 3G/4G, and to plausible that the impact might be larger for those a lesser extent capture potential effects from providing that adopted the technology than for those that did mobile coverage to previously uncovered areas. not. In addition, since most households that received 14 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria 3 Policy implications It is important to continue expanding the evidence coverage globally could be substantial, particularly base on the impacts of mobile broadband on if policymakers assign a greater social value to the household welfare in Sub-Saharan Africa to examine potential welfare gains of those at the bottom of how far our findings extend beyond Nigeria. the income distribution and those that belong to However, this study represents a solid foundation to excluded socioeconomic groups. show the significant role that expanding access to mobile broadband can have on improving the lives of the extreme poor in Africa.  olicies to address the usage gap across 2. P socioeconomic groups are necessary to ensure According to the results of our study, three main everyone benefits from digital technologies pathways should be considered by decision makers, international organisations and other stakeholders In Nigeria, as in many other parts of the world, to enhance the potential of mobile broadband to certain socioeconomic and demographic groups reduce poverty: can face significant barriers to using mobile internet, which may prevent them from reaping the  xpanding broadband coverage to rural and 1. E benefits of broadband coverage.23 For example, remote areas can reduce extreme poverty in 2019 the gender gap in mobile internet usage in Nigeria was 29% (meaning women were 29%  lobally, there are still almost 600 million people G less likely than men to use mobile internet), while that do not live within reach of either a 3G rural populations were 47% less likely to use mobile or 4G network, around half of them in Africa. internet than urban residents.24 Most live in rural and remote areas. Expanding mobile broadband coverage will remain difficult; Strategies to address the usage gap of those new approaches to financing coverage as well individuals living within reach of a mobile as innovations and technical solutions will be broadband network but not using it will remain required.22 However, this study provides a good important policy priorities to realise the full benefits rationale of why such investments are worth of digital technologies. Globally, the usage gap consideration since these can provide large includes 3.4 billion people, with more than 600 economic and social returns for those in greater million living in Africa. Key barriers to bridge the need. In the case of Nigeria, extreme poverty in usage gap include the affordability of services and rural areas fell by 8 percentage points two years devices, relevance of content, lack of digital skills, after being covered by mobile broadband and by and safety/security online. The incidence of these 10 percentage points three years after (see Figure factors is especially large for underserved groups, 4). If these effects were of a similar magnitude including women, rural populations, individuals elsewhere, the potential gains of extending with less education and people with disabilities. 22. See for example GSMA, Closing the Coverage Gap, 2019 23. GSMA, Mobile Gender Gap Report 2018 24. GSMA, Mobile Gender Gap Report 2020 and GSMA, The State of Mobile Internet Connectivity Report 2020 15 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria  omplementary policies are needed to realise 3. C recruit qualified staff. It can also provide individuals the social and economic benefits of digital with better access to information on health, connectivity for all nutrition and education, which not only improves livelihoods and well-being but can also contribute In Nigeria, we found strong evidence that to employment and higher labour productivity.25 increased labour force participation was one of the main benefits from mobile connectivity. Of course, providing access to mobile broadband Expanding access to digital connectivity can also does not reduce poverty nor improve people’s expand labour demand by means of enabling the livelihoods by itself. Well-functioning markets, productive capacity of rural and non-rural firms strong institutions and complementary (especially small and medium-sized enterprises) investments including road infrastructure, to reach more customers in non-local markets. In competition in input and output markets, and combination with higher labour force participation, high-quality public services, help the deployment this could allow them to expand and create new of infrastructure in rural and remote areas and jobs for local communities. need to interact with mobile broadband access to maximise its potential welfare effects. Moreover, mobile broadband can make it easier for individuals to find jobs and for businesses to Other studies also show that the internet can help create new job opportunities and increase labour productivity. See Paunov and Rollo 2014; Fernandes et al. 25.  2019; Chun and Tang 2018; Hjort and Poulsen 2019; Viollaz and Winkler 2020. 16 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria Appendix: Methodology Calculating welfare and poverty-reduction effects precision in the coverage data means that the study of mobile broadband in Nigeria is not affected by the type of measurement error that occurs when relying on other proxy measures, The findings presented in this report are based on such as distance to the nearest mobile site. The data a study that evaluates the welfare and poverty- allows us to match the location of each household reduction effects of mobile broadband in Nigeria with the coverage provided by each mobile site, to during 2010–2016.26 Jointly developed by a team of determine with precision the time when individual researchers from the GSMA and the World Bank, the households began receiving coverage. study makes use of an innovative dataset that is built from two primary sources of data: Second, the granularity of the data allows us to identify households that received coverage by  elfare and poverty – We used the World Bank’s • W accident, in that they were not specifically targeted General Household Survey (GHS) of Nigeria – a by operators but received coverage because of nationally representative survey that tracks around geographical factors outside of their control (terrain 4,000 households and 12,000 individuals over in particular). This allows us to consider two groups time (in 2010/11, 2012/13 and 2015/16). The GHS is of households that had no broadband coverage and a dedicated living standards survey that provides similar levels of poverty and consumption in 2010 detailed information on households’ consumption as and then compare them in 2016, when one group well as other key demographic and socioeconomic received coverage (even though operators did not indicators of their members including education, intend for them to be covered) and one did not. labour market outcomes and access to the internet.27 The poverty status of households is Finally, by using coverage data from operators rather calculated based on the international poverty line of than data reported on mobile broadband use by $1.90 per day (measured in 2011 PPP).28 households, the analysis captures not only the direct impacts of individuals accessing the internet but also • Network coverage – We gathered data on the spillover effects.31 The latter includes, for example, location of each mobile site from three of the four internet users sharing information with non-users largest mobile operators in Nigeria29 as part of the and non-users benefitting from job creation and GSMA Mobile Coverage Maps tool.30 This enabled productivity gains among local firms. us to determine with precision the time when individual households began receiving coverage. To estimate the number of people lifted out of Coverage of mobile broadband is defined as the extreme poverty for the period of study (2010–2015), provision of 3G or 4G coverage, which enables we first calculate the total number of households high-speed access to the internet, and excludes that obtained coverage for each treatment over 2G coverage as it only provides for limited internet this period as implied by the sampling weights. We browsing and applications. then multiply that number by the weighted point estimates for each treatment to obtain estimates By combining these two rich datasets, the analysis on the number of households lifted out of poverty. has several advantages over previous studies, We multiply this number by the average size of enabling us to isolate the impact of mobile households receiving coverage, which results in the broadband on household and individual welfare and total number of people lifted out of extreme poverty establish a clear direction of the effect. First, the due to 3G/4G coverage. 26. GSMA and World Bank, The Welfare Effects of Mobile Broadband Internet: Evidence from Nigeria (2020) 27. More Information on the GHS including design, survey instruments, datasets and methodology can be found via this link: http://surveys.worldbank.org/ lsms/programs/integrated-surveys-agriculture-ISA/nigeria 28. For further details on international poverty lines, see World Bank (2018). 29. We carried out a number of checks using third-party data sources to confirm that the fourth operator did not have coverage in areas outside of the other three. 30. GSMA Mobile Coverage Maps 31. Coverage is distinct from usage, which is when an individual has a SIM card that can be used in a mobile phone to access the internet. 17 The poverty reduction effects of mobile broadband in Africa: Evidence from Nigeria References Autor, D. and Acemoglu, D. 2011. “Skills, Tasks and Technologies: Implications for Employment and Earnings.” In Handbook of Labor Economics, Vol. 4B, edited by Orley Ashenfelter and David Card, 1043–1171. Amsterdam: Elsevier. Bahia, K., Castells, P. and Pedros, X. 2020. “Mobile technology: two decades driving economic growth.” GSMA Intelligence Economic Research Working Paper Bahia, K., Castells, P., Cruz, G., Masaki, T., Pedrós, X., Pfutze, T., Rodríguez-Castelán, C., and Winkler, H. 2020. “The Welfare Effects of Mobile Broadband Internet: Evidence from Nigeria.” World Bank Group Policy Research Working Paper No. 9230. Beegle, K and Christiaensen, L. 2019. “Accelerating Poverty Reduction in Africa.” Washington, DC: World Bank. Chun, N. and Tang, H. 2018. Do Information and Communication Technologies Empower Female Workers? Firm- Level Evidence from Viet Nam (May). ADBI Working Paper 545. Czernich, N., Falck, O., Kretschmer, T., and Woessmann, L. 2011. Broadband infrastructure and economic growth. The Economic Journal, 121(552), 505-532. Fernandes, A. M., Mattoo, A., Nguyen, H and Schiffbauer, M. 2019. “The internet and Chinese exports in the pre-ali baba era.” Journal of Development Economics 138: 57-76. GSMA. 2018. The Mobile Gender Gap Report 2018. GSMA. 2019. Closing the Coverage Gap. GSMA. 2019. The State of Mobile Internet Connectivity 2019. GSMA. 2020. The State of Mobile Internet Connectivity 2020. Hjort, J. and Poulsen, J. 2019. “The Arrival of Fast Internet and Employment in Africa.” American Economic Review 109 (3): 1032–79. International Telecommunications Union. 2019. “Economic impact of broadband in LDCs, LLDCs and SIDS: An empirical study”. Geneva: International Telecommunications Union. Katz, R. and Callorda, F. 2018. “The economic contribution of broadband, digitization and ICT regulation” Geneva: International Telecommunications Union. Mahler, D., Lakner, C., Castaneda Aguilar, R., and Wu, H. “Updated estimates of the impact of COVID-19 on global poverty” Data Blog (blog), June 8 2020, https://blogs.worldbank.org/opendata/updated-estimates-impact-covid- 19-global-poverty Paunov, C. and Rollo, V, 2015. Overcoming Obstacles: The Internet's Contribution to Firm Development, The World Bank Economic Review, Volume 29, Issue suppl_1, Pages S192–S204. Viollaz, M. and Winkler, H. 2020. “Does the Internet Reduce Gender Gaps?: The Case of Jordan." The Case of Jordan (March 13, 2020). World Bank Policy Research Working Paper 9183. World Bank. 2016. World Development Report 2016: Digital Dividends. Washington, DC: World Bank. World Bank. 2018. Poverty and Shared Prosperity 2018: Piecing Together the Poverty Puzzle. Washington, DC: World Bank. 18 www.gsma.com www.worldbank.org For more information, visit www.gsma.com/connected-society www.worldbank.org GSMA HEAD OFFICE THE WORLD BANK HEAD OFFICE Floor 2 1818 H Street NW The Walbrook Building Washington, DC 20433 25 Walbrook USA London EC4N 8AF Tel : +1 (202) 473-1000 United Kingdom Tel: +44 (0)20 7356 0600 Fax: +44 (0)20 7356 0601