> Posted by V. McIntyre, Freelance Writer for the Harvard Kennedy School

Often, we hold out hope that innovation will happen through the great leap forward, the stroke of luck, the miracle cure – and when one candidate fails, we go off in search of another.

There is justifiable concern that this yes-or-no approach hampers international development. A recent article in the New Republic listed “big ideas” in international development that failed – not because they were bad, but because they were big. The article describes a $15 million-plus project to install thousands of water pumps attached to merry-go-rounds in sub-Saharan Africa, as well as Jeffrey Sachs’s Millennium Villages which sought to overhaul entire villages by building housing, schools, clinics, roads, and other key infrastructure. In these and the article’s other cases, with expectations high and money and attention flowing in, the projects sank, often because they outgrew the scale at which they had proven to work. Yet some of a project’s apparent lack of success may simply come down to the measurement you’re using. Many of the world’s most successful development efforts – deworming campaigns, for example – only improve the average life in tiny increments.

Microfinance has also suffered from the boom and bust phenomenon. Over-enthusiasm led to massive expansion, which has been followed by disillusion among many who expected greater miracles. Now, rigorous studies in several countries suggest that the average effect of the availability of microcredit on poverty alleviation is modest. But wholesale rejection may represent a missed opportunity to find the aspects that do work.

Rohini Pande is an economist at Harvard’s Kennedy School and the faculty co-chair of “Rethinking Financial Inclusion: Smart Practice for Policy and Practice,” a weeklong executive education program coming up in May. She thinks that to continue to innovate in the financial inclusion space we should seek out the small changes that emerge from the data and inspect all of microfinance’s varied effects – not just the average impacts on poverty.

Pande and her colleagues recently researched whether changing some features of a loan contract can alter its impact for the women who choose to take them up. Certain aspects of the classic microfinance loan contract, developed by the Grameen Bank in Bangladesh in the 1970s and still prevalent today in South Asia, seem more conducive to some business activities than others. For instance, borrowers with these loans generally start paying installments immediately after receiving the loan. They often keep back part of the money to make the first payments rather than using it to invest in assets or inventory. Wouldn’t that discourage a bigger investment – say, in a sewing machine for a seamstress or refrigerator for a shopkeeper– which would take longer to show a return?

Working with a West Bengal microfinance institution, Pande and colleagues Erica Field, Natalia Rigol, and John Papp separated clients into two groups: the first got a microfinance contract with immediate repayments, and the second got one that featured a two-month grace period before the first repayment installment. The researchers surveyed clients at the time of the loans and revisited them three years later to check on their businesses.

Clients who received the grace period invested 6 percent more in their businesses and were twice as likely to start new ones. Three years later, their weekly business profits were 41 percent greater and monthly household income about 20 percent greater. They also reported roughly 80 percent more business capital. However, there was also a cost, which points to the reason many MFIs have gone in for contracts that start repayment immediately: grace period clients were three times as likely to default.

This study suggested that for microfinance to best foster female entrepreneurship, it should take into account the real needs of small, fragile businesses and start thinking of a broader array of financial products to meet these needs. Perhaps grace period contracts should be accompanied by insurance and financial literacy programs to help address default risk. Or alternatively, perhaps governments in low-income countries should be more willing to consider interest-rate subsidies for poor borrowers, as the United States has.

More broadly, this research supports the notion of carefully identifying the varied effects of microfinance – which may be quite subtle – and not just focusing on the average impacts on poverty. Pande and Field are conducting a series of such studies to understand the nuanced effects of microfinance, not only on entrepreneurship, but on women’s social networks, their levels of financial stress, the number of children they have, and other measures of empowerment and wellbeing. Interesting results, even modest ones, could be the seeds for future studies.

This process, which Pande calls “smart policy design,” could counteract the “big idea” tendency in development. In a chapter from a book on financial inclusion forthcoming from the Upjohn Institute for Employment Research, Pande, Field, and their coauthor Abraham Holland point out that even the best example of a “miracle cure,” penicillin, was in fact the product of much trial and error. The many years between the initial discovery in 1928 and when the public began to reap the benefits in 1945 is typical and to be expected in policy as well as medicine. They write, “This intervening period is filled with years of iterations, attempts, failures, intermediate successes, and a little serendipity. The penicillin strain ultimately found to have the best properties for commercial production came from a moldy cantaloupe in an Illinois fruit market.”

Pande and her Co-Director at Harvard’s Evidence for Policy Design, Asim Khwaja, teach smart policy design as a framework for policy research, both to students and to practitioners. The process involves collaborating to identify problems that can be approached through research, then working to diagnose the underlying causes. The results of evaluations are used to refine the policy and ask new questions.

Last March, Pande and Khwaja introduced this process to leaders in the microfinance sector – as well as regulators and representatives from NGOs – in the “Rethinking Financial Inclusion” program. In daily afternoon sessions over the five-day course, participants, working in groups, attempted to design an approach to a real-world financial inclusion problem one member brought with them. These included how to draw shop owners away from illegal and possibly dangerous lenders in marketplaces in Peru, how to improve financial literacy and decrease over-indebtedness among borrowers in Mexico, and how to extend insurance to rural farmers in Indonesia.

The practice of taking methods from medicine and applying them to the social sciences has caused a revolution in development economics: randomized trials now give policymakers far better information to use in their decision-making. To advance microfinance farther, perhaps the financial inclusion community should take another lesson from medicine, and prepare for the long process of finding a miracle cure.

Rohini Pande and Asim Khwaja of Evidence for Policy Design (EPoD) at Harvard Kennedy School will offer the “Rethinking Financial Inclusion: Smart Design for Policy and Practice” program again May 10-15, 2015. Visit the HKS Executive Education website for details.

V. McIntyre is a freelance social science writer for the Harvard Kennedy School.

Have you read?

Smart Design for Policy and Practice: Consider Bandwidth – Insights from the Harvard Kennedy School’s ‘Rethinking Financial Inclusion’

Measuring the Impact of Microcredit – Six New Studies

Beyond Design, Behavioral Science for the Pilot and Scale of Product Innovations