Assessing the Status of Microcredit
Recently I had the opportunity to take part in a webinar that discussed the status of microcredit, specifically whether or not researchers believe it works. To complement a piece on the webinar that appeared in Vox recently, I thought it would be helpful to provide a short summary of the panel discussion (organized and led by Tim Ogden, with Cynthia Kinnan from Tufts, Gisella Kagy from Vassar College, Karthik Muralidharan from UCSD, and myself), along with some of my own comments that represent a short preview of reflections on microcredit in my book Shrewd Samaritan, that will be out this next summer.
Certainly, there are few poverty interventions in the developing world that have run the gamut of the popularity cycle as thoroughly as microcredit. From late 1980s through about 2010, microcredit was arguably the most popular poverty intervention on the planet. After the famous Microcredit Summit, the number of borrowers spread exponentially across the globe, growing from 13 million in 1997 to well over 205 million by 2010. Both the practitioner and researcher worlds lived in a virtual microcredit lovefest, until research began to show it didn’t help poor people as much as we thought it did.
While there were some early hints that the impact of microcredit wasn’t living up to the anecdotes of entrepreneurial success stories emblazoned on microlender webpages, the 2015 study published in the American Economic Journal: Applied Economics provided compelling results from randomized trials in six countries: Bosnia, Ethiopia, India, Mexico, Morocco, and Mongolia, that gave us the best current evidence on microcredit’s effectiveness or lack thereof.[1]
No one on the panel disputed the basic validity of these studies, which on the positive side show that microcredit increases the size of small businesses, the hours that entrepreneurs devote to their businesses, their freedom to expand them. It also allows households to smooth out shocks to their income. The studies also find little evidence to feed the narrative of the few microcredit critics, who had begun to paint a picture of chronic damage done to borrowers through luring them into a state of perpetual indebtedness. This is all good news. And if the buzz about microcredit had been that it was all about promoting entrepreneurialism, the results of the research would have confirmed the buzz.
But the bad news from these studies is that microcredit is far from the poverty elixir its advocates touted it to be, suggesting that its impact on income and consumption for the average borrower is modest at best, and nothing close to transformative. And a recent meta-study by Rachel Meager seems to confirm the external validity and generalizability of these results.[2]
Many have interpreted the findings from these studies to mean that microcredit was a deception. It is true that there is an element of illusion in the effect of microcredit loans. This illusion occurs because borrowers tend to take loans when economic opportunities present themselves: a new market opens up, or for one reason or another an entrepreneur decides to dedicate more time to her business. To a practitioner or a casual observer making a “before-and-after” comparison, it appears that microcredit works wonders: the microenterprise appears to be booming subsequent to the loan compared to before it. But because the timing of the loan coincides with the new opportunity or business effort, without microcredit the microenterprise might have prospered more anyway.
But in the panel we discussed a couple of reasons to remain cautiously optimistic about microcredit. First, there are hints that the average impact of microcredit may be bigger than the six studies suggest. Within the countries where the randomized trials were carried out, three of them (Mexico, Bosnia-Herzegovina, and Mongolia) were already saturated with microcredit, and in these countries researchers found its impact on income to be essentially zero. But in the three other countries (India, Morocco, and Ethiopia), microcredit was less widespread, and the estimates of income impact were larger.[3] Since impact is likely to be higher among the motivated entrepreneurs who jump at the first microcredit loans than those who only take it up later, some of the studies done in saturated areas probably underestimate its benefits on average. The important thing they do tell us though is that adding more microcredit in already credit-saturated areas is unlikely to have much of an impact.
The second reason for optimism is a study by Cynthia and her coauthor Emily Breza, which looks at the “general equilibrium” effects of microcredit availability, impacts on the economy as a whole. If microcredit causes borrowers to spend more time in their enterprises and less time in the labor market, and they create jobs in microenterprises, tightening in the labor market ought to push up wages for everyone. This is just what happened in 2010 when in response to a crisis of over-indebtedness, the government of the Indian state of Andhra Pradesh shut down the state’s microcredit institutions, so that $1 billion of microcredit activity disappeared overnight. Cynthia and Emily studied the impacts on the economy, and they found that downward pressure in the labor market caused wages to fall by 6%, household income dropped, and there were large reductions in consumption.[4] This super interesting result parallels that of another earlier study on the impact of a government microcredit program rolled out across villages in Thailand, where Joe Kaboski, and Robert Townsend find that village wages increased about 7% in the first two years of a large government-funded microcredit program.[5]
How do you reconcile these seemingly conflicting results? Well, they aren’t fully reconciled yet, which creates some interesting new possibilities for future research, but there are inklings of insight into what is going on here. Tim Ogden pointed out that what we may be seeing is that the cumulative effect of each of these relatively small microfinance lenders may have a much bigger impact than the loans any one of them make to a single individual. He compares it to a recent study by Lauren Bergquist, Marshall Burke and Ted Miguel in which credit was supplied to African farmers.[6] Instead of always having to buy maize when prices are high and sell their crop when crop prices are low (at times when everyone else is buying or selling too), the credit given at harvest allowed the farmers to exploit arbitrage opportunities and do the reverse.
This not only benefited the arbitraging farmers, but everyone else as well, since the arbitraging farmers sell less at the harvest and buy less in the off season. It gives an example of how this puzzle may be reconciled: The general impact of microloans is likely to be significantly bigger than the sum of their parts.
[1] Banerjee, Abhijit, Dean Karlan, and Jonathan Zinman. (2015) “Six Randomized Evaluations of Microcredit: Introduction and Further Steps.” American Economic Journal: Applied Economics, 7(1): 1–21. For an update on the generalizability of these studies see Meager, Rachael (2018) “Understanding the Average Impact of Microcredit Expansions: A Bayesian Hierarchical Analysis of Seven Randomized Experiments” American Economic Journal: Applied Economics, in press.
[2] Meager, Rachael (2018) “Understanding the Average Impact of Microcredit Expansions: A Bayesian Hierarchical Analysis of Seven Randomized Experiments” American Economic Journal: Applied Economics, in press.
[3] Wydick, Bruce (2016) “Microfinance on the Margin: Why Recent Impact Studies May Understate Average Treatment Effects” Journal of Development Effectiveness, 8(2) 257–265.
[4] .Breza, Emily and Cynthia Kinnan (2018) “Measuring the Equilibrium Impacts of Credit: Evidence from the Indian Microfinance Crisis” NBER Working Paper No. 24329.
[5] Kaboski, Joseph and Robert Townsend (2012) “The Impact of Credit on Village Economies,” American Economic Journal: Applied Economics, 4(2), 98-133, which studies the impact of a government microcredit program rolled out across villages in Thailand and found that wages increased about 7% in the median village of the first two years of a large government-funded microcredit program was introduced.
[6] Burke, Marshall, Lauren Falcao Bergquist, and Edward Miguel. (2018) “Sell Low and Buy High: Arbitrage and Local Price Effects in Kenyan Markets.” Working Paper.
- How Trump’s Misunderstanding of Foreign Trade May Help His Bargaining Position
- 3 Reasons the Impact of Microcredit May be Bigger Than We Thought