By ,Ìý andÌýNicholas Loubere

Over the past decade the expansion of digital-financial inclusion through innovations in financial technology (fin-tech) has been identified by the World Bank, the G20, USAID, the Bill & Melinda Gates Foundation, and other major international institutions, as a key way to promote development and alleviate poverty in the Global South (; ; ). Perhaps the most influential and widely reported publication pushing forward this narrative is an article examining M-Pesa written by US-based economists Tavneet Suri and William Jack—and published in the prestigious journal Science—entitled ‘The Long-run Poverty and Gender Impacts of Mobile Money’. M-Pesa is a mobile phone, agent-assisted platform for transferring money from one person to another. It was originally developed with funding from DFID and has quickly become a darling of the digital-financial inclusion movement. In this particular article, the authors make the far-reaching claim that ‘access to the Kenyan mobile money system M-PESA increased per capita consumption levels and lifted 194,000 households, or 2% of Kenyan households, out of poverty’ ().

Suri and Jack’s article in Science has sent ripples through the global development community and has served—as perhaps was intended—to solidify support for upping the promotion of digital-financial inclusion initiatives across the Global South. Importantly, the article’s claims of unprecedented poverty reduction have been uncritically picked up by all of the international development agencies and microcredit advocacy organisations, as well as by many mainstream economists, so-called ‘social entrepreneurs’, tech investors, and media outlets. Much like microcredit in the 1980s, fin-tech and digital-financial inclusion is now very widely seen as a °ì±ð²â—if not the key—to reducing global poverty and promoting local development.

In this post we summarise our recent article entitled ‘Is Fin-tech the New Panacea for Poverty Alleviation and Local Development?’ (), which challenges Suri and Jack’s findings, and urges the global development community to take a second, more critical look at their study. We argue that the article contains a worrying number of omissions, errors, inconsistencies, and that it also employs flawed methodologies. Unfortunately, their inevitably flawed conclusions have served to legitimise and strengthen a false narrative of the role that fin-tech can play in poverty alleviation and development, with potentially devastating consequences for the global poor.

Challenging Suri and Jack’s Claims

Suri and Jack’s study is based on a household panel survey from 2008 to 2014 that compares households in different areas based on access to M-Pesa. The article argues that simply having increased access to M-PesaÌýservices helps the poor become more resilient. In their words: ‘basic financial services such as the ability to safely store, send, and transact money – taken for granted in most advanced economies, and which in the form of mobile money have reached millions of Kenyans at unprecedented speed over the past decade – appear to have the potential to directly boost economic well-being’ (). This understanding of development being facilitated through incorporation into the formal financial system rests on logical leaps of faith and the omission of important considerations.

Firstly, the key premise of their argument is that M-Pesa facilitates the move from livelihoods based on subsistence agriculture to more profitable informal microenterprises, mainly taking the form of tiny trading units. However, they do not account for the fact that running a successful microenterprise is extremely difficult in Kenya (as everywhere in the Global South) because local demand in poor communities is by definition very weak. This demand deficit accounts for why—according to the Kenya Bureau of Statistics—almost 46% of new enterprises are shuttered within a year of opening (). This means that we must assume that almost half of the new businesses facilitated by M-Pesa ultimately ended in some sort of failure scenario, likely resulting in individual and household over-indebtedness and having potentially catastrophic consequences for the lives of the poor people who started them (for example, through the loss of collateral and other invested assets, the trashing of one’s social reputation, and so on).

Secondly, and directly linked to the issue of low local demand and high levels of microenterprise failure, the article did not take into account the ways in which new microenterprises entering the market would impact existing micro-businesses and local economies more broadly. The study rests on the erroneous assumption—known as ‘Say’s Law’—that supply creates its own demand. In reality, however, demand in impoverished local economies is not elastic, and so new businesses inevitably enter into competition with incumbent local enterprises, thus displacing their jobs and incomes. As such, some of the purported gains made by M-Pesa clients inevitably come at the expense of existing microenterprises, which will contract, lose income, and possibly close. Moreover, the influx of new microenterprises has undoubtedly had negative knock-on effects for entire local economies, driving down already tiny profit margins and exacerbating hypercompetitive markets dominated by the poor acting as producers.

Thirdly, Suri and Jack are surprisingly selective in what they measure and attribute to M-Pesa. Specifically, while M-Pesa is credited with facilitating a rise in client incomes and savings, increases in levels of client debt are completely ignored in the study. This presents a wholly incomplete picture of the role that M-Pesa plays in the lives of Kenya’s poor. While M-Pesa is primarily known for facilitating peer-to-peer money transfers, lending is also a major part of the business model of the platform’s parent company—Safaricom. Through the Safaricom lending application M-Shwari, Kenyans can now get instant digital loans ranging from five to five hundred US dollars. These borrowed funds can then be transferred through M-Pesa or directly used for online gambling. As a result of this easy access to digital loans through mobile phones, Kenya is now facing a looming crisis of over-indebtedness (). This situation cannot be discounted when attempting to ascertain the impact that M-Pesa has on poor households as part of a wider suite of digital-financial inclusion applications offered by Safaricom.

Fourthly, Suri and Jack fail to critically examine the wider international political economy and legacies of colonial extraction that M-Pesa is embedded within. While Safaricom is 35% owned by the Kenyan government, the majority shareholder is the UK multinational corporation Vodafone, including through its South African subsidiary, which holds a controlling 40% of the shares. An additional 25% of the shares are mainly held by wealthy foreign investors. Safaricom is hugely profitable, and in 2019 reported annual profits of $US620 million, almost all of which were directed into large dividend payments for investors. Not surprisingly, from 2013 to 2018 the share price grew six fold. As such, however, almost all of the value being created by M-Pesa is extracted and deposited into the coffers of wealthy corporations and individuals in the Global North. This represents a new form of ‘digital mining’ facilitated by the micro-transactions made by the poor through M-Pesa. Far from helping the poor in Kenya, as in previous times where mining involved physical resources such as gold, diamonds, and ivory, this extraction process gradually further impoverishes the local communities in which poor Kenyans live.

Fifthly, the article depicts M-Pesa as efficiently facilitating the transfer of resources within a network to those most in need. However, they do not take into account the fact that not all networks are created equal. Certainly, for those people in networks with wealthy members, M-Pesa could very well serve to provide support in times of need or allow people to quickly take advantage of business opportunities. However, for those in networks comprising people with few resources, there are far fewer opportunities to benefit from the platform. In this sense, M-Pesa simply serves to more efficiently mobilise resources in wealthier networks, ultimately feeding into the rising levels of inequality we see in Kenya and, indeed, all across Africa

Sixthly, the impact evaluation itself was based on a seriously flawed methodological approach. For one, there is no control group, making it impossible to attribute the supposedly beneficial impacts to M-Pesa alone. Second, the chosen analytical approach does not fully account for selection bias. Third, the sample size appears to be small and, in the absence of power calculations, we will never know whether it was appropriate. Fourth, the high levels of attrition are worrying and, finally, the study uses M-Pesa agent density as a proxy for access, which is particularly problematic. In other words, the article finds that places with more M-Pesa agents have become wealthier, but, in our view, it is not access to M-Pesa agents that explains (cause) wealth creation but the presence of wealthier clients that explains (causes) the higher density of M-Pesa agents. M-Pesa agents are simply drawn to areas with more potential for wealth creation.

Conclusion

It is surprising that this deeply-flawed study was accepted for publication in Science—one of the most prestigious scientific publications and infamous forÌýits rigorous vetting of articles. Unfortunately, the end result of this is that a largely inaccurate picture of M-Pesa’s positive role in development and poverty reduction has now emerged and been solidified in global development discourse. M-Pesa has become the new development darling, and digital-financial inclusion initiatives are being vigorously pushed as part of development strategies across the Global South. Moreover, individual investors and wealthy investment bodies around the world have been made aware that if they ‘get into fin-tech’ they can make huge returns quite irrespective of the longer-term damage that will likely be inflicted on the poor communities in which they hope to work.

This adverse situation has eerie echoes of the rise of the global microcredit industry, which was legitimised through a headline-grabbing study by then World Bank economists Mark Pitt and Shahidur Khandker () also using a suspiciously flawed approach to claim that microcredit programmes benefitted poor women. Like Suri and Jack today, two decades ago Pitt and Khandker helped catalyse into existence a misleading narrative about microcredit, ultimately playing a key role in its rapid adoption in the Global South. This ushered in a new era of neoliberal development that resulted in widespread over-indebtedness, wasted resources, and individual suffering. While Pitt and °­³ó²¹²Ô»å°ì±ð°ù’s impact evaluation was later definitively shown to be faulty (, ; ), the damage, however, had already been done.

We are calling for more serious scrutiny of the claims made in Suri and Jack’s study, and hope to prompt a more critical assessment of digital-financial inclusion strategies across the board, before it is too late. We worry that digital-financial inclusion through innovations in fin-tech have simply replaced microcredit in development strategies without addressing any of the fundamental issues that made microcredit such a developmental disaster for the global poor. We also fear that fin-tech has the potential to exacerbate the worst forms of extraction perpetrated by microcredit. ¹ó¾±²Ô-³Ù±ð³¦³ó’s obvious potential for doing good, such as when deployed by community-owned financial institutions that aim to serve the poor and not push them into un–repayable levels of debt, has been almost entirely overlooked in the investment community’s rush to get rich and to locate the next fin-tech ‘unicorn’.

Rather than being lauded as the new panacea for development and poverty reduction, digital-financial inclusion initiatives like M-Pesa should, therefore, be rigorously scrutinised. If this does not happen, we can look forward to two more decades of failed development policy with hugely negative outcomes for poor households around the world.

References

Bateman, Milford, Maren Duvendack, and Nicholas Loubere. 2019. . Review of African Political Economy: 1–16. Doi: 10.1080/03056244.2019.1614552.

Duvendack, Maren, and Richard Palmer-Jones. 2012a. . Journal of Development Studies 48(12): 1864–1880.

Duvendack, Maren, and Richard Palmer-Jones. 2012b. . Journal of Development Studies 48 (12): 1892–1897.

GPFI (Global Partnership for Financial Inclusion). 2016. . Global Partnership for Financial Inclusion.

±áä°ù¾±²Ô²µ, Norbert. 2017. . Money and More.

Muhatia, Abel. (2019) . The Star.

Omondi, Dominic. 2016. . Standard Media.

Pitt, Mark, and Shahidur Khandker. 1998. ? Journal of Political Economy 106 (5): 958–996.

Roodman, David, and Jonathan Morduch. 2014. The Impact of Microcredit on the Poor in Bangladesh: Revisiting the Evidence. Journal of Development Studies 50 (4): 583–604.

Suri, Tavneet, and William Jack. 2016. . Science 354 (6317): 1288-92.

World Bank. 2014. . Washington D.C.: The World Bank.

Wright, Graham. 2017. Digital Credit – ? Microsave Blog.

Ìýis a Visiting Professor of Economics at the Department of Tourism and Economics, Juraj Dobrila University of Pula in Croatia.

ÌýisÌýSenior Lecturer in Development Economics at the University of East Anglia, UK. .

Nicholas LoubereÌýis Associate Senior Lecturer at the Centre for East and South-East Asian Studies, Lund University, Sweden.Ìý.

This blog post is a part of the blog seriesÌýInclusiveÌýor Exclusive Global Development? Scrutinizing Financial Inclusion,Ìýin which a new perspective on financial inclusion is published onÌýs in the spring of 2019.