CGAP has just published a study that Jessica Schicks and I have written on over-indebtedness among microborrowers. The paper is not exactly bedtime reading: I’m slightly embarrassed to say that it runs to 43 pages. So some readers may feel like the third grade student whose book review began, “This book told me more about whales than I wanted to know.”
It’s been an interesting exercise to go back over the paper’s detailed discussion and cull out a few conclusions that feel important enough to highlight in a blog. The first big message has to do with the question in this blog’s title. After many years of enviable repayment performance, serious default problems have broken out in recent years in Morocco, Nicaragua, Bosnia/Herzogovina, Pakistan, and Andhra Pradesh in India. Some of this default is probably “strategic”—borrowers who could easily repay take advantage of political or other situations to stop making payments. But much of the default appears to be due to over-indebtedness. (Without going into the complications of defining “over-indebtedness,” let’s just say for now that it refers to borrowers who can’t repay their loans without serious difficulty. Note that this is a client-centered definition, and that it implies that borrowers can still be over-indebted even if they manage to pay off their loans.) Read the rest of this page »
The study referred to in my post yesterday, defines over-indebtedness as the inability to repay all debts fully and on time. Over-indebtedness only occurs if this situation occurs chronically, i.e. several periods in a row and against the borrowers will. I agree with Rich Rosenberg and Jessica Schicks who argue that over-indebtedness can start before a default is reached.
Still one of the keys to addressing over-indebtedness is analyzing the ability of the borrower to repay. A weakness of many group lending models is that no cash flow analysis is done, hence the debt service burden cannot be compared with net disposable income and the risk of over-indebting a client is not appropriately addressed. Some MFI’s that offer group loans have introduced cash flow analysis in the group lending methodology. Read the rest of this page »
Like most microfinance practitioners, particularly in the sub-continent, I have thought a lot about the issue of over-indebtedness in the last few months. It’s not that it was any less of an issue before, but the Andhra Pradesh meltdown really made me question seriously whether we in Bangladesh have also been cavalier to the point of being reckless about the amount of credit we have sold to our clients.
Milford Bateman in an earlier post in this series seemed to suggest that it was the ‘commercialization’ of microfinance and alleged tendency of senior managers to enrich themselves that are to blame for borrower over-indebtedness. Maybe he is right with regard to some other countries, but this is not the case as far as Bangladesh is concerned. Sure, MFIs in Bangladesh have “commercialized” insofar as they have moved away from donor funded microfinance in the 1990s to a self-sustaining microfinance model in the 2000s. But almost all microfinance players here are non-profits that re-invest surpluses back into their portfolio and there is little or no evidence of abnormally high salaries or private enrichment of senior managers.
Yet, though due to the absence of any credit profile I have no way to know the exact percentage of borrowers who fall in the over-indebted category, I have a hunch that the percentage is high enough, particularly in certain regions of the country, for us to be concerned about it. While I don’t agree with Mr. Bateman that it is commercialization that is solely to blame, I do agree that almost all the microfinance players bought into the idea propagated by some advocates that microcredit is the silver bullet to eradicate poverty and all that the poor people need is more and more microcredit.
This belief prompted MFIs to seek fantastic growth in the numbers of borrowers and sizes of portfolio over a 15 year period in which they, or we, did not stop to assess whether more and more credit was causing some people to become over-indebted. (I strongly believe though that the vast majority of microfinance borrowers in Bangladesh have economically benefitted from having access to institutional financial services.)
I completely agree with Richard Rosenberg that every credit program will leave some borrowers worse off. This might be because of pure bad luck, as Mr. Rosenberg has demonstrated with a great example in an earlier post. In Bangladesh, many borrowers become unable to repay their loans midway through the loan cycle because a sudden health shock in the family or a natural calamity wipes out their assets and income source.
In these cases as well, one cannot blame the loan officer or the borrower for taking a loan that ultimately made her worse off. What matters in these types of situations is what action the MFIs take once the borrower starts to default, as that can further deepen vulnerability or help build resilience. That, however, is a separate discussion.
But what about those instances where a borrower, at the very outset, takes on more debt than she can possibly repay, or will become economically worse off as a result of paying back? I have heard some microfinance practitioners argue that the MFIs are not to blame for this because they are not forcing anyone to take loans. The problem, they say, is created by the borrowers themselves because they don’t know when to stop. There is of course some truth to this. However, I belong in the camp that thinks that the MFIs here are, at least in part, to blame for borrower over-indebtedness, whatever the extent of it may be.
It is true that often the borrowers take on too much debt, wittingly or unwittingly, and the loan officers lack the capacity, the information, or both, to recognize this and pull back on disbursements. But just as often if not more so, it is the loan officers of MFIs who seduce their clients into taking on too much debt because they are incentivised on hitting sales targets that, at least for some time, were over ambitious. This was made worse by the fact that the microfinance industry, though chasing fantastic growth on one hand, had stopped innovating on the other and was offering much the same product to every type of borrower.
Whether it is the borrowers who took on more debt than they could handle or the MFIs that pushed too much debt on to borrowers or whether both have been complicit in this, I believe that MFIs now have a responsibility to intervene on both the demand and supply sides to counter the problem of over-indebtedness.
On the demand side, the MFIs can work to increase the financial literacy of their clients so that borrowers can take better financial decisions. Providing financial literacy training to millions of borrowers will, of course, drive up costs. But perhaps the time has come for governments and donors to subsidize these kinds of activities of MFIs rather than the loan fund itself. On the supply side, the MFIs can work to ensure that too much credit is not pushed on to borrowers through a number of steps.
First, the MFIs can set more realistic and sensible growth targets that are desegregated over space and time. Second, MFIs can re-train their frontline staff and assess their performance not only in terms of sales targets and portfolio quality but also in terms of other non-financial indicators of borrower wellbeing. Third, MFIs can work towards a greater sharing of information among themselves through the setting up of a microcredit bureau so that decisions are not based on partial information in a country where multiple loans are an endemic feature of microfinance. And lastly, there is tremendous scope for innovation and product development in the microfinance space.
I do see signs of progress on most of the issues in Bangladesh, particularly on the supply side. The large MFIs seem to be revising down their growth targets, there are ongoing experiments with new products and there is some hope that a microcredit bureau, at least on a limited scale, will see the light of day within a year or so. These are positive signs as far as I am concerned and indicate that MFIs have begun to counter the problem of over-indebtedness before it becomes a major issue in this country.
This post is the next in the blog series on over-indebtedness. In the coming weeks we’ll be featuring a variety of voices from across the globe on this topic. We welcome your participation in this discussion through comments.
Some months ago I asked CGAP senior advisor Rich Rosenberg what he thought was the most pressing challenge for microfinance today, and his response–understanding over-indebtedness–was passionate. You can watch a video clip of Rich talking here .
The issue of over-indebtedness, and whether microcredit does more harm than good to clients living in poverty is all over the news as the issue has bubbled up in the Indian state of Andhra Pradesh and hit the global headlines over the past year from Bosnia to Nicaragua. But how much do we really know about levels of over-indebtedness, or even what is a reasonable level of debt for a poor household to carry? Over the coming weeks, CGAP will be running a series from leading experts on financial inclusion with their thoughts on the issue, and what the global microfinance community should do to address it.
Part of Rich’s frustration comes from the lack of robust research tools for understanding whether there is an over-indebtedness problem. “I am more concerned about getting a handle on the issue of over-indebtedness,” he told me, “than about fine tuning other areas of customer protection. Yes, fair disclosure of interest rates is important. Yes, avoiding abusive collections methods is important. Yes, protecting client privacy is important. But for me, the basics, basics, of customer protection is to avoid over-indebtedness.”
Tomorrow, Rich will share his latest thinking, and over the coming weeks he’ll be joined by other experts, from regulators who have been dealing with the issue, to researchers who have been gathering data. Join us and share your insights through comments to help move the conversation forward.
—–Jeanette Thomas
This post kicks-off a blog series on over-indebtedness. Over the coming weeks we’ll be featuring a variety of voices from across the globe on this topic. We welcome your participation in this discussion through comments.
In August 2010, several internet-based media in Russia, including a number of very popularones, published information about a company called AktivDen’gi (Active Money). The company, established in May 2009, claimed to work in nine cities in Russia through over 70 outlets located primarily in supermarkets and trade centers, disbursing loans for 5 to 15 days at 2% daily – 730% p.a. In their view, they were “creatively developing the ideas of Muhammad Yunus” and identified themselves with microfinance. ActiveMoney also made claims to efficiency and transparency – saying they could disburse loans in 10-15 minutes, requiring minimum documentation, and fully disclosing the interest rate. They also claimed to have half a million clients after six months of operations.
Microfinance market players in Russia were somewhat taken aback at the information, and wanted to find out more about ActiveMoney. It turned out, however, that the company had very few tiny outlets, and the level of their lending activities was minimal – it was unlikely that they served as many customers as they claimed. But how come the company that was so tiny, if at all real, had garnered so much publicity? One theory was that it was in fact a campaign initiated by one of Russia’s retail banks to discredit microfinance – just several weeks after the President approved the Federal Law “On Microfinance Activity and Microfinance Organizations.” Feeling threatened by competition from MFIs, this theory goes, the bank was combatting the competitive inroads Russian MFIs are making into the market and scaring off investors potentially interested in investing in microfinance organizations.
Now the commotion has died down, there are still a number of open questions left after the incident: What makes microfinance different, given that high interest rates have always been part of the picture? At what level do interest rates become “too high” (the complexity of this issue is discussed in CGAP’s Occasional Paper The New Moneylenders: Are the Poor Being Exploited by High Microcredit Interest Rates? by Richard Rosenberg, Adrian Gonzalez, and Sushma Narain)? And how should the microfinance community in Russia react to such an attempt at discrediting what they do?
Many of us have thought for a long time that microcredit clients are not very sensitive to interest rates. But is that actually true? Does the interest rate impact clients’ decision to take the loan, or are most borrowers so eager for reliable credit (or in need of funds right now), that they do not weigh the interest rate of the loan that heavily in their decision to borrow?
A recent CGAP-commissioned study by Innovations for Poverty Action measured borrowers’ sensitivity to interest rates at 132 branches of Mexican MFI Compartamos. A randomized trial offered some potential borrowers a lower interest rate relative to another group of potential borrowers, offering a range of interest rates from 3 to 4.5% a month. The results were interesting:
1. Loan uptake was 22.3% in low-interest rate clusters, versus 15% in high interest-rate clusters. Evidently, interest rates mattered to a substantial numbers of clients.
2. Over an 18-month period, lower interest rate groups took more loans and bigger loans. Take-up rates for the lower interest rate groups were 22.3%, versus 15% amongst the higher interest rate groups. Similarly, average loan size disbursed grew by 21% in the lower interest rate groups post-treatment, resulting in an average loan size that was 4% larger than in the higher interest rate groups post-treatment.
3. Existing clients were particularly influenced by the change in interest rates, with a 17% increase in loans disbursed, versus an 11% increase amongst new clients (defined as their first loan disbursed within the current month). Maybe the existing clients were more sensitive to rate differentials because they had a baseline (the rate on their prior loans) to compare the new rates with?
4. The business case will depend on the individual MFI. In some cases, MFIs might improve profits by lowering their rates, as the revenue gained from doing more lending might exceed the revenue lost due to the lower interest rate. In Compartamos’ case, however, the revenue gains from additional lending did not fully offset the revenue losses from the lower rates. But the balance might tip the other way if other factors are considered—for instance, gains from other services that the new customers eventually buy, or reputational value.
Now that we have some strong evidence that interest rates do matter to many of Compartamos’ borrowers, I think we could benefit from further field trials to answer two additional questions on this topic:
1. Were the new borrowers attracted by the lower interest rates people who would not have borrowed otherwise? In other words, do lower rates produce an increase in aggregate credit use, or do they just move borrowers back and forth among competing MFIs?
2. Beyond the immediate gains and losses from reduced interest rates, can we quantify some of the other elements in the equation? To do so we will need studies of total client profitability that go beyond revenue gains in the first year or two of the client’s relationship with the MFI. Glenn Westley’s recent paper on the business case for small savers looked at marginal rather than average costs, and the profitability of other products eventually sold to the client, and came away with some promising results on long-term client profitability, even if short-term profits weren’t present. Also, does the larger average loan size produced by lower rates bring with it lower operating cost ratios? A $400 loan produces twice the revenue of a $200 loan, but the administrative cost of the two loans might be pretty much the same.
Stay tuned for more work in this area, as CGAP is currently funding two other Innovations for Poverty Action studies of interest rate sensitivity, and we look forward to sharing their results in the future.
The previous post in this series ticked off some reasons why we ought to be looking more closely at overindebtedness in microfinance. But what do we mean when we use that term? This question, like most others about overindebtedness, gets annoyingly complicated.
There is no uniform definition of overindebtedness. The term is sometimes associated with loans that don’t get repaid, or with loans whose repayment ties up more than X percent of household income, or with loans whose repayment becomes stressful for the borrower, or with clients taking loans from too many sources at once. Can we find a unifying concept at some deeper level?
CGAP has just published a study that Jessica Schicks and I have written on over-indebtedness among microborrowers. The paper is not exactly bedtime reading: I’m slightly embarrassed to say that it runs to 43 pages. So some readers may feel like the third grade student whose book review began, “This book told me more about whales than I wanted to know.”
The study referred to in my 
Like most microfinance practitioners, particularly in the sub-continent, I have thought a lot about the issue of over-indebtedness in the last few months. It’s not that it was any less of an issue before, but the Andhra Pradesh meltdown really made me question seriously whether we in Bangladesh have also been cavalier to the point of being reckless about the amount of credit we have sold to our clients.
