Allocating capital to microentrepreneurs through community surveys in India

Our research demonstrates that community members in low-income countries can be a great source of entrepreneurial assessment information of their microentrepreneur neighbours, providing details traditional methods often miss. But how do lenders and grant distributors know those same community members aren’t exaggerating or even lying to favour their friends when real money is on the table? 

Microentrepreneurship and the allocation of microcredit and grants

Microentrepreneurs in low-income countries, often in family-run businesses like small grocery stores and tailor shops, tend to have high returns to capital that are above standard microcredit interest rates (Fafchamps et al. 2014, McKenzie et al. 2008). Economic theory suggests that expanding access to credit should be a key factor for faster growth; yet recent research shows the rapid spread of microlending has had limited effect on entrepreneurship and income growth (Banerjee et al. 2015). This may be because lenders often treat microentrepreneurs as a single homogeneous group. 

This is where peers (community members running their own small businesses) can help lenders or grant distributors with assessment, such as which shop is the most popular among several competitors. Other details can include: The variety of goods on hand, cleanliness, customer service, and the diligence and ambition of the entrepreneur. 

In a densely populated city in the developing world, where businesses often aren’t known outside their neighbourhood, a survey of peers may help outside lenders or grant distributors, like non-governmental organisations (NGOs), draw distinctions in ways traditional metrics methods can’t (Besley and Ghatak 2005, Varian 1990). 

The study: Ranking microentrepreneurs using peer information for the allocation of capital

We surveyed microentrepreneurs in a section of the city of Amravati in Maharashtra, India, concentrating on an area that’s home to about 550,000 people. Our research demonstrates that rankings of microentrepreneurs by peers in their own community gives a far better entrepreneurial assessment than merely relying on characteristics of a business, such as its industry, age, detailed measures of size and profitability, and numerous others. 

We focused on nine neighbourhoods on the outskirts of Amravati. After going door-to-door, we identified about 1,300 microentrepreneurs for our study. Businesses in our study typically earn about US$2.50 a day, 60% of them are run by men, and the owner is, on average, 40 years old. These small enterprises include manufacturing (usually a tailor or seamstress), food stalls, and hair salons. 

The findings: Peer evaluation optimises potential microenterprise growth

To reach our initial findings, we conducted several rounds of ranking exercises within groups of four to six neighbours. The microentrepreneurs ranked one another on their household income, business profits, and ability to grow their business using a cash grant, in addition to several other outcomes. 

After the rankings, we randomly distributed US$100 grants to one-third of the microentrepreneurs in our sample. This allowed us to measure their ability to grow their business using the grant and also to judge the quality of community information in predicting business growth. 

The results, based on follow-up surveys, were striking: If we had distributed grants using information from the community reports instead of giving them out randomly, we would have more than tripled the resulting business growth.

The problem of bias in ranking surveys

But what if the evaluating peer group knew their reports would influence the distribution of grants? Would they lie to make sure friends or family had an edge in receiving a grant? It turns out the answer is yes. 

Here’s how we found out: Half of our respondents were told that their rankings would influence which entrepreneurs received a cash grant. The other half were told that their reports were purely for research (in both cases, we controlled the probability of receiving a grant to assure that our statements were true). The respondents whose reports could influence the distribution of grants tended to favour their family and friends more than those who had no control over the distribution of grants.

Incentivising accurate peer information

So how do you avoid such bias? It is, after all, human nature to want to help your friends.

We decided to pay the survey takers based on their accuracy to see if a cash reward would eliminate bias. Accordingly, we split the survey takers into two groups; half received cash incentives for accuracy and half did not. Respondents in the incentivised group were instructed to be honest in their answers to get the reward.

We found that even with cash on the table, paying these peer evaluators virtually eliminated the temptation to lie to help their favourite microentrepreneur.

Conclusion: A tool for spurring microenterprise growth

Microentrepreneurs in Amravati, Maharashtra have a wealth of information about one another. They were able to reliably assess which of their peers had profitable investment opportunities and which did not, far more effectively than methods based on business characteristics.

While we find that community members are prone to distort their reports when they know they can influence who gets the cash, simple monetary incentives for accuracy nearly fully realign their incentives to report their beliefs truthfully. 

We are cautiously optimistic that these tools could be flexibly incorporated into the standard practices of microfinance institutions and NGOs with the intent of spurring microenterprise growth.  

Editors' note: This column is based on research under the PEDL projects here and here. 

References

Banerjee, A, Karlan, D and Zinman, J (2015). “Six randomized evaluations of microcredit: Introduction and further steps”, American Economic Journal: Applied Economics, 7 (1), 1–21.

Besley, T and Ghatak, M (2005). “Competition and incentives with motivated agents”, American Economic Review, 95 (3), 616–636.

Fafchamps, M, McKenzie, D, Quinn, S and Woodruff, C (2014). “Microenterprise growth and the flypaper effect: Evidence from a randomized experiment in Ghana”, Journal of Development Economics, 106, 211–226

McKenzie, D, de Mel, S and Woodruff, C (2008). “Returns to capital: Results from a randomized experiment”, Quarterly Journal of Economics, 123 (4), 1329–72.

Varian, H R (1990). “Monitoring agents with other agents”, Journal of Institutional and Theoretical Economics (JITE)/Zeitschrift Für Die Gesamte Staatswissenschaft, 153–174.