1. The Question
The Kenyan economy has been referred to as a dichotomy of a rapidly growing, technologically enabled formal economy (banking, telecommunications, and exports) on top of an extensive informal economy comprising micro and small firms. Formal employment in Kenya has not been able to catch up with informal employment over decades; by the late 2010s, about 83% of the total employment was in the informal economy from less than 10% during the mid-1970s. These firms have mostly been unregistered, untaxed and unlinked to the formal credit market, which is the reason why development economists have referred to the firm size distribution in Sub-Saharan Africa as having a “missing middle,” with many micro firms, a few large firms and few firms in between. The M-Pesa system that was introduced by Safaricom in 2007 is often seen as part of the solution. By 2009, it had become part of around two-thirds of households, and by 2021, mobile money adoption amongst adults had increased to about 81%, from 28% in 2009. What this paper seeks to find out is limited to a small, testable hypothesis: Does the availability of M-Pesa really push informal SMEs towards formalization in Kenya, or does it just make informality more sustainable? This is important because formality is much more than a mere designation. Formally registered firms have access to bank credit, they can enforce their contracts through courts, they can recruit large numbers of people without getting entangled with the law on a continual basis, and above all, they count in the tax base that finances public goods.
2. The Microeconomic Mechanism
Standard theory (inspired by de Soto’s analysis of informality and credit market theory with asymmetric information) considers the decision to become formal as a cost stack comparison between two alternatives:
Costs of remaining informal: limited access to bank financing and trade financing (since banks and other financiers cannot verify income and contract enforcement), no ability to bid on formal contracts, risks of harassment/ extortion by inspectors, and incapacity to grow beyond the size where informality is evident.
Costs of becoming formal: costs associated with KRA registration and compliance, taxation, minimum wage and labor laws, and fixed costs of accounting/reporting.
A firm becomes formal only if the marginal gain from being formal outweighs these fixed compliance costs. Mobile money seems to affect only the benefits side of this calculation via the following two routes:
Transaction history that is easy to verify. The logs from M-Pesa transactions serve as a record of the company’s cash flow, which may serve as collateral or credit history for the informal firm. Mobile money allows creating digital financial instruments where loan underwriting takes place via transactions history rather than land title or certified accounts. 2. Cost savings on transaction and security. Using digital payments lowers the cost and risk of transactions involving cash between suppliers and retailers, relieving liquidity constraints forcing firms to trade below capacity.
Most importantly, neither channel affects the cost of formalization itself. In other words, a firm can benefit from improved cash flow information and low transactions cost by remaining informal. If the tax and regulatory burden is indeed the constraint on formalization, mobile money should increase financial inclusion and liquidity without much effect on the level of formalization.
3. Evidence
There have been remarkable improvements in financial inclusion. The evidence from the FinAccess survey indicates that the share of adults having a formal bank account in Kenya increased from approximately 14% in 2006 to around 44% in 2021, while the proportion of mobile money usage grew from 28% in 2009 to 81% in 2021. In other words, this is the most robust and documented effect of M-Pesa; M-Pesa turned Kenya from a country with low levels of formal banking access to a country with almost universal digital financial access.
Mobile money seems to help firms to overcome their liquidity and credit-adjacent constraints. In particular, with the use of the FinAccess Business enterprise survey, Beck et al. have found a positive and significant correlation between using mobile money to pay suppliers and access to trade credit and estimated that adoption of mobile money has increased Kenya's average annual GDP growth rate during 2006 to 2013 period by 0.33-0.47 percentage points. Moreover, according to the famous study by Suri and Jack, access to mobile money enabled 2% of Kenyan households (mainly female-headed households) to escape poverty due to occupational change (e.g., from agriculture to business).
Formalization has followed a different trend. With the adoption rate of mobile money exceeding 80%, the proportion of informal employment in Kenya has been very high and, according to a recent World Bank study on informality in cities of Sub-Saharan Africa, exhibits very little systematic reduction throughout the decade, which is nearly contemporaneous with the development of M-Pesa. In a 2021 study carried out in collaboration with the ILO, it has similarly been noted that the number of informal jobs created in Kenya significantly exceeded those created in the formal sector (about 768,000 informal jobs compared to 78,400 formal jobs in 2019 alone). The recent qualitative and survey literature presents this as a paradox: while the use of formal financial mechanisms through digital means has risen, informal financial transactions such as the formation of chamas (saving circles) has also increased.
Formal Enterprise Data
The sampling framework of formal enterprise data, in itself, highlights the issue. According to the World Bank’s Enterprise Surveys, firms surveyed are those that are registered with the KRA and have more than five employees; these criteria, by design, exclude precisely the population of interest for formalization purposes, informal enterprises. In order to be able to reach such a population, separate Informal Enterprise Surveys are necessary.
4. Argument
When we put together the mechanism and the empirical evidence, what becomes clear is that we cannot say “M-Pesa formalizes SMEs”, rather, what seems to happen is that M-Pesa alleviates the constraint that exists on productivity and survival of informal SMEs because of the constraints of liquidity and information asymmetry. However, it does not affect the regulation costs that are at stake in the decision to register formally. With mobile money, SMEs derive much of the benefits of financial technologies (verifiable flows, payments from suppliers, savings) without bearing much of the cost of formalization (being taxable). Mobile money may thus make informality “better,” not necessarily lead to formalization.
The missing middle question gets reframed: if the constraint on formalization is compliance cost and not credit access, then digital innovations such as M-Pesa, such as digital ID and e-tax filing and making it easier for micro-enterprises to register with the KRA, will be a better way forward than mobile money lending products that largely target firms that have opted to remain informal.
5. Implications
- For policy: It looks like Kenya’s “missing middle” may just be a problem of compliance costs rather than access to credit. The intuition here is simple, low-fixed-cost registration tiers (such as presumptive taxation systems with low cost of compliance) are a better solution than the expansion of fintech credit if one seeks formalization specifically.
- For fintech and development economics: The real success of mobile money lies in financial inclusion and poverty alleviation at the household level. These are positive outcomes in themselves and should not be confused with formalization, an outcome which is not necessarily supported by the evidence.
- For future research: A good test would be to examine formalization rates of matched pairs of informal firms that either have or do not have access to mobile money, controlling for industry, geography and education of the owner, what would be closer to a natural experiment (for instance, relying on geographic variation in the timing of M-Pesa agent roll-out, which was the key element of Jack and Suri’s identification strategy in the case of households).
6. Limitations
This study represents a secondary synthesis of survey and academic evidence, and there are a few key limitations associated with it. First, the measure of formalization adopted for the purpose of this paper (the informal employment share) captures this concept in a roundabout way and not by using a firm-level panel of firms’ formalization status compared to their M-Pesa adoption over time. Second, reverse causality could be a problem since firms whose growth potential would push them towards formalizing could be exactly those that adopt M-Pesa sooner, leading to an upwards-biased cross-sectional relationship. Finally, there is inconsistency in the definition of informality across the various sources used in Kenya (2016 MSME Survey, KNBS, and FinAccess use different cut-offs). The best possible way to formally test the hypothesis of this paper is using a firm-level panel data from the combination of FinAccess Business waves and KRA registration data; such data is not publicly available, to the best of our knowledge.
References (selected)
- Jack, W. & Suri, T. (2011). Mobile Money: The Economics of M-PESA. NBER Working Paper No. 16721.
- Suri, T. & Jack, W. (2016). The long-run poverty and gender effects of mobile money. Science, 354(6317).
- Mbiti, I. & Weil, D. (2011). Mobile Banking: The Impact of M-Pesa in Kenya. NBER Working Paper.
- Beck, T., Pamuk, H., Ramrattan, R., & Uras, B. R. (2015). Mobile money, trade credit, and economic development: theory and evidence.
- FSD Kenya / Central Bank of Kenya / KNBS. FinAccess Household and FinAccess Business Surveys, various years (2006-2021).
- World Bank Group. Enterprise Surveys and Informal Enterprise Surveys, Kenya country profiles.
- World Bank. (2024). Urban Informality in Sub-Saharan Africa. Policy Research Working Paper 10703.
- International Labour Organization. (2021). The Informal Economy in Kenya.
- Understanding Informality and Mutuality in Kenya’s Digital Finance Landscape (2024/2025 study using FinAccess 2006-2021 and FGD/ethnographic data).
Prepared as part of the East Africa Microeconomics Research Series. This is a student working paper that synthesises existing research and published data rather than presenting original fieldwork; feedback and data corrections are welcome via the contact page.