NextBillion - An Initiative of the William Davidson Institute at the University of Michigan

Earlier this year, following an enthusiastic conversation with Bob Christen from the Boulder Institute of Microfinance, my team and I at EA Consultants set off on a path to think deeply about over-indebtedness and microfinance. At the time, we were working with New York City’s Office of Financial Empowerment to understand the experiences of clients at their free Financial Empowerment Centers. Many of these clients were suffering from severe over-indebtedness, characterized by low credit scores, severe financial stress, and exclusion from both access to credit and (in some cases) housing. This topic seemed both timely and extremely relevant.

We borrowed some of the thinking around post-recession indebtedness from the OECD countries to consider patterns in lesser-developed countries. The first thing we found was at a systemic, countrywide level. Perhaps unsurprisingly, we found that rapid growth in aggregate microfinance loan portfolios countrywide can be a red flag for over-indebtedness. At the institutional level, loan portfolio growth may not be a problem. It can even be a sign of strength. Microfinance institutions (MFIs) might grow because of their relative market advantages vis-à-vis their competition, or because they are able to increasingly lend to untapped markets. Countrywide, however, a spike in the overall loan portfolio growth of all providers of microfinance loans on aggregate might look suspicious, particularly where it is driven by large growth in average loan sizes rather than numbers of clients. It can suggest supply-side factors that contribute to over-indebtedness, such as market saturation, growing competition, multiple borrowing by clients, and increasing loan sizes for borrowers that do not correspond to specific increases in borrowing needs or business growth.

When we analyzed some key indicators from seven “hot” microfinance markets (Guatemala, Mexico, Cambodia, Bosnia, Nicaragua, Morocco and Kenya), we noted a pattern whereby this rapid growth in loan portfolios was followed, though not immediately, by spikes in portfolios at risk. The lag between the sharp growth and spikes in arrears, measured by portfolio at risk greater than 30 days (PAR), varied; but generally took two to three years. So we asked ourselves: If these cycles are symptomatic of over-indebtedness, why would it take over two years for the effects of loan portfolio growth to exhibit itself as poor repayment?

The answer probably lies in a series of qualitative studies on over-indebtedness, including a well-known study from 2011 by Jessica Schicks in Ghana and our own very recent experience in New York City. Client over-indebtedness looks like a lot of things before it looks like a repayment problem. It looks like clients taking on additional work, depleting savings, selling assets, and undergoing severe stress, and at times, depression. It leads to feelings of shame and disempowerment, which is ironic since microfinance in its early stages was meant to empower people who had a limited voice in their societies. One of the cornerstones of microfinance is that borrowers pay back their loans because they want to protect their reputations, often their largest asset. But when protecting their reputation leads to severe sacrifices, we wonder whether the original premise made sense. Perhaps they were better off without the loans.

We did not stop at the MFI and market level – the next step in our thinking was to consider over-indebtedness at the client level. We used a standard measure, the debt-to-income ratio (DTI): monthly household debt as a percentage of total monthly household income. By applying the measure to a dataset from micro-entrepreneurs in the English-speaking Caribbean, we found that businesses with low returns, or those that do not invest loan proceeds into their business, can become over-indebted as easily as those who invest in consumption. For example, only 8 percent of those eligible borrowers in our data set with solid margins would have DTI above 25 percent after a microloan for their business. However, 23 percent of eligible borrowers with lower margins would spiral into DTI levels above 25 percent after a microloan.

Our thinking is still preliminary, and sharing it in these spaces can help us refine our ideas with input from others. Still, we believe that there is space to start acting on our findings. Firstly, data should be tracked and monitored more systematically and spikes in lending on aggregate should be seen with suspicion rather than promise. Regulators can play an important role in monitoring these trends, and tracking other indicators such as multiple loans and, of course, arrears. Investors and other second-tier lenders should be wary of balancing their own need to grow and raise new money with the risk of over-lending to MFIs. MFIs can and should track data more regularly as well, avoiding the trap of increasing loan sizes automatically when borrowers repay, rather than assessing borrowers’ use of funds and potential for growth. Consumer loans in particular have no rationale for becoming progressively larger, as consumption does not generate new income to pay off larger loans. Finally, end consumers need to take responsibility for their actions. Borrowing, while fraught with behavioral complexities in low-income market, is a voluntary choice. Stakeholders in the sector should offer clients the tools and information needed to make good choices for themselves, their businesses and their families.

Over-indebtedness is the yang to microcredit’s yin: There can be no over-indebtedness without credit access. For that reason, we must be cautious about raising a moral red flag about the issue, at the risk of spurring a backlash against irresponsible lending that could exclude rather than include those who are harder to reach. Instead, we are better off rolling up our sleeves and taking many small and difficult steps toward strengthening a system that we now understand can do unintended harm.

Barbara Magnoni is president of EA Consultants.