A partial marvel

Jul 16th 2009
From The Economist print edition

Microcredit may not work wonders but it does help the entrepreneurial poor


Illustration by Jac Depczyk

MICROCREDIT looks like a miracle. It involves providing unsecured small loans to poor people in
developing countries whom most banks would turn away. Yet these small borrowers almost always repay
their loans (and the fairly steep interest charges) on time, which suggests that they find
productive uses for the money. The industry’s backers make some big claims as a result: Mohammad
Yunus, the founder of Grameen Bank in Bangladesh and the father of microfinance, reckons that 5% of
Grameen Bank’s clients exit poverty each year. Yet economists point out that there are surprisingly
few credible estimates of the extent to which microcredit actually reduces poverty.

This would not matter too much if all microfinance funding were raised via the market (as an
increasing proportion is). As long as investors were satisfied with their returns, there would be
no cause for concern. Yet despite growing interest from private investors, 53% of the $11.7 billion
that was committed to the microfinance industry in 2008 still came at below-market rates from aid
agencies, multilateral banks and other donors. Given that there are other things that aid money
could be spent on, and that the rationale for subsidising microcredit is its effectiveness as an
anti-poverty tool, it is important for donors to know whether it has the advertised effects.

Measuring the impact of microcredit is complicated by the fact that the counterfactual—what would
have happened to a person who borrowed from a microlender if he had not done so—cannot easily be
tested. Many early studies compared borrowers with non-borrowers. But if borrowers are in any case
more entrepreneurial than those who do not borrow, such comparisons are likely to overstate hugely
the effect of microcredit.

Worries of this nature are not mere nitpicking. One study surveyed 1,800 families in rural
Bangladesh and found that an impressive 62% of school-age sons of those who borrowed from Grameen
Bank were in school, compared with 34% of the sons of non-borrowers. Advocates argued that this
showed that microcredit helped increase school enrolment. But a comparison with people of similar
backgrounds in villages without access to microcredit showed that the difference was because people
who were already more likely to have children in school were also those who signed up for
microcredit. Even comparisons between areas with and without microcredit may be misleading, because
microlenders naturally choose to work in areas where their prospects of success are the greatest.

The pervasiveness of these self-selection issues has led researchers to devise experiments that
allow them to ensure that participation in a programme is determined essentially by chance. Two new
papers* apply this idea to measure the effect of access to microcredit. Researchers from the
Poverty Action Lab at the Massachusetts Institute of Technology (MIT) worked with an Indian
microfinance firm to ensure that 52 randomly chosen slums in the city of Hyderabad were given
access to microfinance, while 52 other slums, which were equally suitable and where the lender was
also keen to expand, were denied it. This allowed the researchers to see clearly the effect of
microcredit on an entire community. Dean Karlan of Yale University and Jonathan Zinman of Dartmouth
College carried out a similar exercise in the Philippines, this time at the level of the individual
borrower. They tweaked the credit-scoring software of a microfinance firm so that only a random
subset of people with marginal credit histories were accepted as clients. These clients could then
be compared with those who sought credit but were denied it.

Broadly speaking, neither study found that microcredit reduced poverty. There was no effect on
average household consumption, at least within a year to 18 months of the experiment. The study in
the Philippines also measured the probability of being under the poverty line and the quality of
food that people ate, and again found no effects. Microcredit may not even be the most useful
financial service for the majority of poor people. Only one in five loans in the Hyderabad study
actually led to the creation of a new business. Providing people with safe places to store their
(small) savings may help them more in the long run.

Small and perfectly formed?

That said, microcredit did have discernible effects. In India, people in the slums that had access
to microcredit were more likely to cut down on things like tobacco and alcohol in favour of durable
goods (particularly items such as pushcarts or cooking pans that are used heavily by traders and
food-stall owners). One reason average consumption failed to increase may therefore be that more
people were diverting some of their own income into starting or expanding their businesses.
Microcredit clearly allowed more people to overcome the barrier posed by start-up costs. The MIT
researchers found that as many as one-third more businesses had opened in slums which had a
microcredit branch. This may mean that even though there was no measurable impact on poverty during
the study period, there may well be some over a longer time-frame as these businesses prosper.

Tiny loans are unlikely to be enough to allow these businesses to grow to an efficient scale, of
course. But the role of microcredit in allowing people to signal their creditworthiness is
valuable, especially if their success makes banks more willing to lend them larger sums and leads
to even more economic activity. By being willing to take a risk on entrepreneurial sorts who lack
any other way to start a business, microcredit may help reduce poverty in the long run, even if its
short-run effects are negligible.