While digital credit broadens market access and reduces frictions in developing countries, default rates are often high. In Mexico, reducing loan speed—by doubling delivery time—decreased the likelihood of default significantly. Such waiting periods used to selectively slow down credit could help improve lender profitability.
Editor’s note: For a broader synthesis of themes covered in this article, check out Issue 2 of our VoxDevLit on Mobile Money.
Over the past decade, digital credit has transformed access to finance, offering fast, short-term loans to millions in low- and middle-income countries. With just a few clicks on a phone, borrowers can receive funds quickly and without need for collateral—making digital lending a powerful tool for financial inclusion. Digital loans are typically designed for instant access, with approvals and transfers happening within seconds (as credit provided through mobile money operators in African and Asian markets) or several hours (as in Mexico). These loans play a critical role for borrowers in need of quick liquidity for emergency situations or time-sensitive needs.
But convenience comes with risks. Default rates are often high, and there are concerns that credit access may be fuelling overborrowing among the already vulnerable (Melzer 2011). Several studies estimate default rates ranging from 7% in Kenya (Bharadwaj et al. 2019), 15% in Malawi (Brailovskaya et al. 2024), and 27% in Mexico (Burlando et al. 2025). These default rates are high in comparison to microfinance or standard consumer credit. These patterns raise the possibility that the unique characteristics of digital credit, such as its rapid disbursement, minimal verification processes, and low ability to enforce repayments, may contribute to default. It is therefore important to better understand the role of the innovative features of digital credit, and whether there is a trade-off between convenience and responsible credit use.
Our research with a digital credit provider in Mexico focuses on one such feature: credit speed. We study whether loan speed affects repayment rates. This work reveals a surprising insight: unexpected slowdowns in loan disbursement, even just a few hours, significantly reduces default rates. Specifically, we found that doubling the disbursement time from 10 to 20 hours reduced default rates by 21%. This challenges the idea that faster access to credit—reduced friction—is always beneficial, and raises an important question for lenders and policymakers alike: should the speed of digital credit be regulated?
Slowing down to speed up repayment: Evidence from a Mexican digital lender
We work with the administrative loan records of a digital lender in Mexico that operated a website through which clients requested loans. We obtained three key pieces of information about each loan: (1) the time of the client’s loan submission request, (2) the time when the loan was deposited into the client’s bank account following approval and processing by the lender, and (3) whether the loan was ultimately repaid. Using the former two pieces of information, we calculate the amount of time it takes for a loan to be disbursed, i.e. the loan delay.
To estimate the causal effect of longer delays on repayment rates, we use a regression discontinuity strategy. We take advantage of the fact that the lender processed loans continuously throughout the working day, but only disbursed loans a few times a day. Loans are thus delivered in batches. To be included in a given batch, a loan must be submitted before a certain time—referred to as the batch delivery time; submissions after this time are processed in the next batch. In our analysis, we compare the repayment rate of loans that were submitted around the same time, but where the slightly later loan is submitted after the batch delivery time and is therefore delivered hours later.
Batch processing lead to long delays in loan disbursement
Figure 1 shows the result of this process. In Panel A, we plot the amount of time it takes to disburse a loan (vertical axis) against this four-hour time span (horizontal axis). We also show the trend line before and after the batch delivery time (which is located in the band between the two vertical lines). In all three graphs, the trend line ‘jumps’ for loans submitted after the batch delivery window. The jump shows that the lender’s batch processing increases the loan’s delay. On average, when considering the entire sample of loans, missing a batch cut-off extends a loan’s delay by 9.84 hours, roughly doubling the delay a borrower experiences.
Figure 1: Regression discontinuity plots around batch delivery times
Delayed loans are paid back more frequently
Panel B plots the repayment rate of those same loans against the four-hour batch delivery times. Again, the trend lines jump for loans submitted after the window. This jump highlights the impact of the extra delay on repayment: in the entire sample of loans, borrowers who faced extra delays due to missing batch cutoffs had an 8% higher repayment rate (alternatively, a 21% lower default rate).
Delayed loans are more profitable
Perhaps unsurprisingly, delayed disbursements also have longer-term financial benefits for lenders. Borrowers who successfully repay loans become eligible for larger credit amounts, driving repeat borrowing and long-term customer retention. Our findings suggest that increased repayment rates from delayed loans generate sustained revenue flows, as borrowers who clear their debts are more likely to take out future loans and repay them. Thus, while immediate loan disbursement may seem like a competitive advantage, implementing structured waiting periods could enhance lender profitability while also reducing borrower defaults.
Why delays in loan disbursement improve repayment rates
There are several reasons that may explain why a delay in disbursement leads to better repayment outcomes. We are able to exclude a few of these explanations based on available evidence, but others remain possibilities that will require additional research.
- Delayed loans may be rejected by the borrower and returned to the lender. Our administrative records include information on whether a loan is rejected and returned to the lender. We can rule this possibility out; borrowers rarely reject approved loans.
- Delays may cause borrowers to rethink how to use their loans. Borrowers who experience a short delay may reconsider their loan use, shifting away from impulsive spending, such as gambling, towards more financially prudent decisions. In this explanation, delays play the same role as a mandatory ‘waiting period’, which is common in certain markets, in that they reduce impulsive decision-making and improve repayment discipline. We do not currently have evidence of shifts in borrower behaviour, so this remains a possible explanation.
- Delays may force borrowers to change how to use their loans. Related to the preceding explanation, the delay may force a change in borrowers’ time sensitive plans if the loan arrives too late. We find some patterns in our data that indicate this may be the case, at least for the context of betting on football matches.
- Delays may force borrowers into bargaining with their intimate partner over the use of the loan. A borrower whose loan request is delivered late might have to confront their partner over the use of the funds, potentially leading to more prudent financial choices. Consistent with this, we find strong effects of delays on both male and female married borrowers, particularly when the loan is delivered the following day.
- Delays may prompt further borrowing from other sources. Recipients facing too long of a delay may seek and obtain other sources of credit. Once all loans are received, the extra liquidity may be used for repaying the delayed loan. We cannot rule out this channel.
Lessons for digital lenders and policymakers
Our research has implications for several alternative interventions:
- Smart loan design: Digital lenders may benefit from implementing selective waiting periods, particularly for high-risk borrowers, to reduce default rates without discouraging loan uptake.
- Consumer protection measures: Regulators could explore policies that incorporate brief cooling-off periods in digital credit markets, similar to those used in payday lending and high-cost credit sectors (e.g. mandatory cooling-off periods in the UK payday loan market (FCA 2013) and restrictions on same-day lending in some US states (CFPB 2024)).
- Behavioural nudges: Financial literacy programmes and loan interface designs can be structured to encourage borrowers to reflect on their financial decisions before disbursement.
Whether the above suggestions will lead to improved well-being of borrowers is unclear. If delays cause borrowers to reflect on their spending and make smarter choices that lead to higher repayment, then slowing down the speed of digital loans may be beneficial. On the other hand, for borrowers using digital credit for important time-sensitive needs or emergencies, delays could be harmful. Moreover, our research explored delays that were unexpected; implementing a systematic waiting period would impact expectations and lead to less predictable outcomes.
As digital credit markets evolve, the industry must continuously balance accessibility with responsible lending practices. Novel features, such as the immediacy of credit, must be thoroughly evaluated for benefits as well as harms. The insights from this research suggest the possibility that slower may be better—that a minor delay in loan disbursement may enhance borrower outcomes without limiting financial inclusion or hurting the lenders’ bottom line.
References
Bharadwaj, P, W Jack, and T Suri (2019), “Fintech and household resilience to shocks: Evidence from digital loans in Kenya,” Unpublished manuscript.
Brailovskaya, V, P Dupas, and J Robinson (2024), “Is digital credit filling a hole or digging a hole? Evidence from Malawi,” The Economic Journal, 134(658): 457–484.
Burlando, A, M A Kuhn, and S Prina (2025), “Too fast, too furious? Digital credit delivery speed and repayment rates,” Journal of Development Economics, 174: 103427.
CFPB (Consumer Financial Protection Bureau) (2024), “New protections for payday and installment loans slated to take effect next year.”
FCA (Financial Conduct Authority) (2013), “The FCA sets out in detail how it will regulate consumer credit, including payday lending, when it takes over responsibility in April 2014.”
Melzer, B (2011), “The real costs of credit access: Evidence from the payday lending market,” Quarterly Journal of Economics, 126(1): 517–555.