MICROCAPITAL PAPER WRAP-UP: The Miracle of Microfinance? Evidence From a Randomized Evaluation, by Abhijit Banerjee, Esther Duflo, Rachel Glennerster, and Cynthia Kinnan

“The Miracle of Microfinance? Evidence From a Randomized Evaluation”, by Abhijit Banerjee, Esther Duflo, Rachel Glennerster, and Cynthia Kinnan, Published by CGAP, May 2009. Available at: http://collab2.cgap.org//gm/document-1.9.34827/The%20miracle%20of%20micro%C2%85nance.pdf

This study, claimed by the authors to be the first randomized evaluation on the effects of microfinance, comes to the pessimistic conclusion that microfinance has only a small effect on the fortunes and almost no effect on the lifestyles of poor people after the opening of a microfinance institution (MFI) in their neighborhood. In this case, the term randomized refers to the random selection of 52 of 104 similar poor areas as locations for new branches of the MFI Spandana in the Indian city of Hyderabad. Dubbing the randomly selected areas with microfinance the “treatment” group and the areas without microfinance the “comparison” group, the authors distributed a survey amongst an average of 65 households in each area fifteen to eighteen months after the openings of the branches. The authors ultimately find that there is some difference, albeit not a major difference, in business profit and consumption patterns between households in treatment and comparison areas.

Characteristics of treatment and comparison areas were similar, save the presence of a microfinance branch. After the opening of the Spandana branch, the probability to receive a microloan was 27 percent in treatment areas and 18.7 percent in comparison areas. The baseline study conducted in the areas before the introduction of a Spandana branch showed that 69 percent of households had at least one outstanding loan of about USD 400 with an interest rate of 3.85 percent per month. 34 percent of the households had a savings account, 26 percent had life insurance, and a negligible amount had health insurance.

The authors follow three groups of households which may alter consumption patterns with the advent of microfinance. The first group, consisting of households who will use loans to start businesses, reduce consumption to contribute to the fixed cost of the project. The second group, consisting of households who do not use their loans for starting businesses, increases consumption due to the extra funds provided by the loan. The third group, consisting of households who use loans to improve existing businesses, increases consumption because the extra investment leads to immediate profits. The first and third groups were typically the most prosperous, with an average consumption per capita of USD 5 per day prior to microfinance. The third group borrowed 8.5 percent more from an MFI in treatment areas versus comparison areas while the first and second groups borrowed 9.6 percent more in treatment areas.

As for comparisons between treatment and comparison areas, 7 percent of households in treatment areas opened businesses versus 5.3 percent in comparison areas. Additionally, business owners in treatment areas reported higher profits of an average of about USD 200. Although about 90 percent of this was nondurable expenditure, households in treatment areas spent about 50 cents more per adult on durable goods versus comparison households. Treatment households also spent 25 cents per adult on durable goods used in the household business versus 10 cents in comparison households. The authors also correlate changes in consumption with households’ propensity to start a business. The households most likely to start a business reduced nondurable spending as well as consumption of temptation goods, including alcohol and cigarettes, by 17 percent.

No significant changes in lifestyles or culture were observed. Although Spandana is an MFI that aims to impact women in particular, women in the treatment areas reported that they were not any more likely to make important household financial or education decisions than women in comparison areas. Furthermore, treatment and comparison households spent equal amounts on medicine and sanitation, were equally likely to have enrolled children in school, and spent equal amounts on tuition and school fees.

Rather than evaluate success on absolute terms across all businesses, the authors examine the change in profits in businesses of different sizes. In the 50th percentile of business profit, the effect of microfinance on profits is zero. However, the effect on the least and most successful businesses is more significant, with businesses with profit at the 95th percentile reported average increased profit of about USD 130. The authors conclude that in the 15 to 18 month period after the opening of the Spandana branches, there are changes in household expenditure and profits in some businesses although there is little change in aspects of lifestyle including healthcare and education.

Spandana’s loans are typical microfinance loans and follow the Grameen model, popularized by Nobel Peace Prize winner Mohammed Yunus. Spandana granted group loans to 6 to 10 women and grouping 25 to 45 groups in a center. With a first loan amount of about USD 200 and an interest rate of 12 percent, the loan takes 50 weeks to repay. All clients must be women aged 18 to 59 who have lived in the same area for a minimum of 1 year. Distribution of loans was as follows: 30 percent of loans were granted to clients for starting new businesses, 22 percent to buy stock for an existing business 30 percent to repay an existing loan, 15 percent to buy a durable good for the household, and 15 percent to smooth household consumption. Clients were permitted to use loans for multiple purposes and were not obligated to use the loan for businesses.

Comments on the validity of this study are available at the Private Sector Development Blog of the World Bank.

By Goda Thangada, Research Assistant

Similar Posts: