Impact Evaluation of Compartamos Released

The long-awaited impact study of Compartamos, led by Manuela Angelucci of the University of Michigan and Dean Karlan and Johnathan Zinman of IPA, has finally been published. The research team used a randomized trial to test the impact of loans offered at 110% APR by Compartamos, the largest microlender in Mexico. After three years of data collection on a variety of factors, the results were generally positive with no evidence that the loans caused harm or significant negative effects.  Researchers found that loan recipients grew their business revenues and expenses, were happier, more trusting, had greater household decision power, and were better able to manage liquidity and risk.  However, there was little evidence that loans had an impact on building wealth like household income, business profits, or consumption.

One of the more interesting conclusions from the paper is as follows:

In all, our study adds to the mounting evidence that microcredit is generally beneficial on average, but not necessarily transformative in the ways often advertised by practitioners, policymakers, and donors. The most consistent impacts come out in the more subjective, qualitative wellbeing outcomes, rather than the more traditional economic outcomes such as income and consumption. We also provide new evidence on various distributional impacts suggesting that expanded credit access has multifaceted, complex, and heterogeneous effects on businesses and households. Better understanding of these distributional impacts may hold a key for making progress on the modeling, application, and evaluation of credit market innovations and interventions.

While this study looked at the household impact of borrowing, the IPA team has also published the results of complementary research on long-run price elasticities of demand for credit in Mexico. In this study, some randomly assigned branches lowered the interest rate of loans offered. Consistent with other studies, this increased demand. These additional borrowers covered the profit impact of the lower interest rates. In other words, Compartamos was able to lower interest rates while maintaining the same total profitability.  There was no evidence of crowd-out and competitors did not respond by reducing rates.  One of the key take-aways is authors’ note that Compartamos has a range of price options that enables it to still maximize profits, allowing for greater flexibility in fulfilling the “social” side of its social enterprise mission, with lower prices being a simple option to expand credit access. 

Both papers provide support for a more nuanced understanding of how microcredit actually “works” for poor households.  To this end, Angelucci et al. make recommendations for further areas of research such as how the typical borrower’s profile and experience with credit determines “success,” how lenders can incent individuals to borrow, the effectiveness of financial literacy, and why credit elasticities grow over time.