Pakistani currency
Pakistani currency
Research shows that microfinance clients use credit and savings as commitment devices to accumulate lump sums. Evidence from Pakistan suggests high demand for fixed-repayment contracts, but low demand for commitment add-ons in both credit and savings.
Editor’s note: You can read more aboutmicrofinance contracts in our VoxDevLit onMicrofinance.
Microfinance has expanded access to credit for the poor in developing countries. While repayment rates on loans have been consistently high (Banerjee et al. 2015, Cai et al. 2023), saving among the poor remains stubbornly low (Demirgüç-Kunt et al. 2018). The fact that poor people can repay high-interest rate loans but struggle to save has long puzzled economists and policymakers. Research has reconciled this contradiction by demonstrating that poor people ‘borrow to save’: credit and saving are alternative ways through which the poor accumulate money for lump-sum expenditures (Afzal et al. 2018, Rutherford 2000, Morduch 2010, Collins et al. 2009, Armendáriz and Morduch 2010, Pomeranz and Kast 2022). The rigid repayment schedules of microfinance loans thus serve as a commitment device to save.
Recognising that saving and credit are substitutes for the poor implies that the separate evidence bases on microcredit and microsaving can be leveraged together to design better financial products. Both bodies of research have explored how varying levels of commitment affect loan default and saving behaviour (Battaglia et al. 2024, Field and Pande 2008, Ashraf et al. 2006, Karlan and Linden 2014, John 2020). Less is known on how the demand for flexibility, and soft versus hard commitment varies across the saving and credit domain.
Recent research on microcredit uses repayment flexibility to accommodate for the variable income of microfinance clients, finding mixed impacts on default rates. Although clients value flexibility, evidence suggests that this does not correspond with higher default rates (Barboni and Agarwal 2023). In the saving domain, soft (e.g. reminders) and hard (e.g. penalties) commitment devices have been shown to improve saving rates (Karlan et al. 2016, Dupas and Robinson 2013). Taken together, these findings suggest that the impact of commitment add-ons may differ for credit and saving contracts.
Assessing demand for commitment and flexibility in microfinance contracts
In our paper (Afzal et al. 2024), we test how the take-up of a basic commitment saving or credit contract changes as the level of built-in commitment varies. We partner with a large microfinance institution in Pakistan, the National Rural Support Programme, and offer around 3,200 of their current and past clients an individual-liability product that provides a lump-sum in return for regular weekly instalments. This product guarantees commitment, as missed instalment payments result in contract cancellation. In the first phase of the experiment, we test the demand for credit and savings products by randomly varying the interest charge and whether the lump-sum is disbursed at the beginning or end of the contract. In the second phase, we test how take-up varies with the level of commitment built into the contract.
Specifically, we augment the standard product with add-ons, drawing inspiration from studies on saving and borrowing. In some treatments, we increase commitment either through hard commitment devices, such as a penalty for missing an instalment (sunk treatment), or through soft commitment devices, such as reminders to the client (reminders to self) or their family members (reminders to peers). In other treatments, we decrease commitment by allowing clients to defer one instalment (flex treatment). We also cross-cut add-ons by adding reminders to the sunk and flex treatments. This design allows us to test how demand for commitment add-ons varies when they are included, in isolation or combined, to a credit or saving contract.
There is high demand for fixed repayment, but low demand for commitment add-ons
The study has three key findings. First, we find substantial demand for fixed repayment contracts in both credit and savings domains, indicating that clients value the commitment built into standard contracts. Take-up rates are about 30% and 8% for credit and saving contracts, respectively. This aligns with results from related research on microcredit and commitment saving products.
Second, we observe low average demand for additional contractual features, such as flexibility, penalties, and reminders. This is particularly true for credit contracts, where take-up falls when additional commitment features are introduced (Figure 1). These results suggest that, on average, the commitment built into fixed repayment credit and savings contracts satisfy the needs of our study population.
Figure 1: Average take-up by contractual add-ons
(a) Credit products
Average take-up by contractual add-ons: Credit products
(b) Saving products
Average take-up by contractual add-ons: Saving products
Average take-up by contractual add-ons: Saving products
Notes: The Figures show the average take-up for the basic product in the first bar, followed by average take-up for each of the eight possible add-ons and combinations thereof. Error bars show 90% and 95% confidence intervals for the difference in take-up relative to the basic contract. Stars indicate p-values of 0.01 for ***, 0.05 for ** and 0.1 for *.
Third, specific combinations of contractual add-ons appeal to certain clients, depending on their level of financial discipline. High-discipline clients prefer flexibility in credit contracts when combined with self-reminders, indicating that flexibility exacerbates inattention concerns and that participants see reminders as a way to mitigate these concerns (Figure 2a). In contrast, low-discipline clients value penalties in savings contracts only when paired with reminders, suggesting that they recognise that adding a penalty increases commitment, but fear incurring this penalty due to inattention (Figure 2b).
Figure 2: Average take-up by contractual add-ons among individuals with high and low financial discipline
(a) Credit products
Average take-up by contractual add-ons among individuals with high and low financial discipline: Credit products
Average take-up by contractual add-ons among individuals with high and low financial discipline: Credit products
(b) Saving products
Average take-up by contractual add-ons among individuals with high and low financial discipline: Saving products
Average take-up by contractual add-ons among individuals with high and low financial discipline: Saving products
Notes: The Figures show the average take-up for the basic product in the first bar, followed by average take-up for each of the eight possible add-ons and combinations thereof. Averages for individuals with low and high financial discipline appear as opaque and transparent bars, respectively. Error bars show 90% and 95% confidence intervals for the difference in take-up relative to the basic contract. Stars indicate p-values of 0.01 for ***, 0.05 for ** and 0.1 for *.
Default rates decrease with contractual add-ons and increase with flexibility
Default rates vary with contractual add-ons among clients who take-up these contracts. Since take-up is affected by selection, our evidence on default is not causal. On average, the probability of paying at least one instalment late is 12% and it is higher in savings contracts, as opposed to credit contracts. Default rates generally decrease with contractual add-ons, more so when reminders are combined with the sunk and flex treatments (Figure 3). In contrast, flexibility by itself increases default rates slightly. This result, combined with the low demand for flexibility, makes this type of contract unappealing to microfinance institutions.
Figure 3: Rate of late payment by contractual add-ons
(a) Credit products
Rate of late payment by contractual add-ons: Credit products
(b) Saving products
Rate of late payment by contractual add-ons: Saving products
Rate of late payment by contractual add-ons: Saving products
Notes: The Figures show the rate of late payment for the basic product in the first bar, followed by average take-up for each of the eight possible add-ons and combinations thereof. Error bars show 90% and 95% confidence intervals for the difference in take-up relative to the basic contract. Stars indicate p-values of 0.01 for ***, 0.05 for ** and 0.1 for *.
Policy implications for microfinance contracts
Through studying the take-up of different contracts within the same population, we confirm important patterns emerging in microfinance research. Microfinance clients prefer credit over savings products, a finding attributable to the use of credit contracts as a commitment device to save and borrowing as a response to unanticipated, urgent expenditure needs. This helps explain why microfinance institutions primarily offer loans rather than commitment saving contracts to meet clients’ needs for lump-sum accumulation.
We also show that demand for commitment contracts and add-ons varies with individual characteristics, particularly their level of financial discipline. Even though the level of commitment provided by fixed repayment contracts is correct on average, there is unserved demand for specific contractual add-ons that align with clients’ discipline levels. The increasing availability of data on clients’ payment and take-up histories should facilitate the provision of such targeted contracts.
References
Afzal, U, G d’Adda, M Fafchamps, S Quinn, and F Said (2018), “Two sides of the same rupee? Comparing demand for microcredit and microsaving in a framed field experiment in rural Pakistan,” The Economic Journal, 128(614): 2161–2190.
Afzal, U, G d’Adda, M Fafchamps, S Quinn, and F Said (2024), “Demand for commitment in credit and saving contracts: A field experiment,” The Economic Journal, 134(664): 3063–3095.
Armendáriz, B, and J Morduch (2010), "The economics of microfinance," MIT Press.
Ashraf, N, D Karlan, and W Yin (2006), “Tying Odysseus to the mast: Evidence from a commitment savings product in the Philippines,” The Quarterly Journal of Economics, 121(2): 635–672.
Banerjee, A, D Karlan, and J Zinman (2015), “Six randomized evaluations of microcredit: Introduction and further steps,” American Economic Journal: Applied Economics, 7(1): 1–21.
Barboni, G, and P Agarwal (2023), “How do flexible microfinance contracts improve repayment rates and business outcomes? Experimental evidence from India,” Unpublished manuscript.
Battaglia, M, S Gulesci, and A Madestam (2024), “Repayment flexibility and risk-taking: Experimental evidence from credit contracts,” The Review of Economic Studies, 91(5): 2635–2675.
Cai, J, M Meki, S Quinn, E Field, C Kinnan, J Morduch, J de Quidt, and F Said (2023), “Microfinance,” VoxDevLit, 3(2).
Collins, D, J Morduch, S Rutherford, and O Ruthven (2009), "Portfolios of the poor: How the world’s poor live on $2 a day," Princeton University Press.
Cordaro, F, M Fafchamps, C Mayer, M Meki, S Quinn, and K Roll (2022), “Microequity and mutuality: Experimental evidence on credit with performance-contingent repayment,” Unpublished manuscript.
Demirgüç-Kunt, A, L F Klapper, D Singer, and P Van Oudheusden (2018), "The global findex database 2014: Measuring financial inclusion around the world," World Bank.
Dupas, P, and J Robinson (2013), “Why don’t the poor save more? Evidence from health savings experiments,” American Economic Review, 103(4): 1138–1171.
Field, E, and R Pande (2008), “Repayment frequency and default in microfinance: Evidence from India,” Journal of the European Economic Association, 6(2-3): 501–509.
John, A (2020), “When commitment fails: Evidence from a field experiment,” Management Science, 66(2): 503–529.
Karlan, D, and L L Linden (2014), “Loose knots: Strong versus weak commitments to save for education in Uganda,” Unpublished manuscript.
Karlan, D, M McConnell, S Mullainathan, and J Zinman (2016), “Getting to the top of mind: How reminders increase saving,” Management Science, 62(12): 3393–3411.
Morduch, J (2010), “Borrowing to save,” Journal of Globalization and Development, 1(2).
Pomeranz, D, and F Kast (2022), “Savings accounts to borrow less: Experimental evidence from Chile,” Journal of Human Resources, 0619–10264R3.
Rutherford, S (2000), "The poor and their money," Oxford University Press.
Pakistan microfinance microcredit Commitment device