Rethinking Multiple Borrowing

By Daniel Rozas,

Microfinance Focus, September 15, 2011: Some time ago, I had a conversation with a microfinance investor.  What is the greatest challenge facing the sector? – I asked.  His answer:  multiple borrowing – multiple borrowing getting people into too much debt; multiple borrowing transforming micro-enterprise lending into consumer finance; multiple borrowing rewriting the traditional relationship between MFIs and their clients.

Of course, multiple borrowingis present in all of these cases.  But thinking about multiple borrowing along these lines misunderstands the basic situation. Multiple borrowing isn’t a reflection of some recent or extreme developments to be ascribed to runaway growth, greed, or willing ignorance.  Nor is it some foreign element to be excised from microfinance.  No, multiple borrowing is an intrinsic part of the practice, one that has been with us for years.  Nor, despite press articles to the contrary, is it a result of heavy market penetration, or even saturation.

This is a realization I came upon during a recent trip to Haiti.  Haiti, you see, is a relatively unpenetrated market, with peak numbers on MIX showing about 120,000 borrowers, or about 1.2% of the population.  And yet the feedback from loan officers on the ground was that multiple borrowing is widespread, some suggesting that as many as 50% of their clients have loans with other MFIs.  And this wasn’t a case of saturation in one place and nothing elsewhere – I heard similar stories both in the capital and in the regions.

Haiti is full of informal markets and businesses that make up the heart of a traditional microfinance portfolio.  Demand is clearly there.  But given the spread of multiple borrowing, the true penetration numbers suggest less than 1% of the population – far below the numbers of other (stable) microfinance markets.  This level of penetration is very low indeed.

Now, a disclaimer – nothing I heard warrants outsize concern about the potential ill-effects of multiple borrowing.  Haiti faces a host of problems, but a credit bubble in its microfinance sector isn’t one of them.  Still, the spread of multiple borrowing took me by surprise.

But really, why should I have been surprised?  In theory, microfinance should feature more, not less, multiple borrowing. After all, traditionally the sector targets individuals without documented credit histories, and this is still largely the case in most countries.  For MFIs, the most costly loans are for new clients, whose repayment histories aren’t known.  Such “first cycle” loans tend to feature especially low balances (which are costly to serve), and their repayment rates tend to be lower than those of more mature clients.

Clearly, if one could get some evidence of timely repayment, even from an existing loan, the returns for that in terms of lower risk and lower cost will nearly always outweigh the added risk of the additional debt burden on the client.  And this is even more true for the near term – the financial effects to the MFI of over-burdened borrowers tend to take time to manifest themselves, and often don’t appear until the borrower is very deeply in debt or when the entire market hits a crisis level.

Yet in the long-run, such an approach is a losing proposition to the industry, hurting both the clients and the MFIs.  My recent study of MFIs in crisis – Weathering the Storm – featured two MFIs (“Hestia” in Pakistan and “Phaethon” in Morocco) where multiple borrowing was a key component of the institutions’ uncontrolled growth path that ended in their eventual collapse.  There are, of course, many other such cases (Bosnia, Nicaragua, Andhra Pradesh), and these are now well-understood by the industry – hence the focus on preventing over-indebtedness and the runaway success of the Smart Campaign in bringing this issue to the forefront of the sector’s consciousness.  But this recognition doesn’t appear to have come with the realization that multiple borrowing is an industry norm.  Some continue to reject the practice altogether.  One large MFI in Azerbaijan has been promoting a one-loan, one-lender policy.

These are understandable and even laudable positions.  Unfortunately, they run against the inherent dynamics of microfinance.  The most sustainable answer isn’t to avoid multiple borrowing, but to recognize it, understand it, and ultimately, manage it better.

Living with multiple borrowing


The first and most common response in the sector has been two words:  credit bureau. And it’s of course the right one, since this can be one of the key tools in assessing a client’s existing debt burden.  But credit bureaus only provide the information – they can’t by themselves change lending practices or perceptions.  And they will do nothing to help with the loss-making economics of first-time borrowers.  If anything, one can expect multiple borrowing to increase with the introduction of credit bureau data, which improves MFIs’ ability to target clients with existing repayment histories, that is to say, with existing loans.

In truth, the most appropriate response is the one the Smart Campaign has identified – preventing over-indebtedness.  Yet it’s also a more difficult response than setting up a credit bureau.  The problem, as is so often the case in microfinance, is that it’s not a response that can be effectively implemented by any one MFI.  The most obvious example of this is the case of the “bad actor.”

Consider two MFIs:  Jerry Community Bank and Tom Loans.  Jerry diligently evaluates the repayment capacities of its clients.  In fact, it’s so good at evaluating clients that its portfolio becomes perfect pickings for Tom, an MFI on the lookout for clients with good histories.  That’s the main reason Jerry has been reluctant to report its data to the new Credit Bureau, since it fears that Jerry would simply use the information to steal its clients.

But besides keeping Tom away from its clients, Jerry has another problem.  It has committed to prevent its clients from becoming over-indebted – a principle it believes in strongly.  But it has increasingly been finding itself at a crossroads.  When its clients seek to renew their loans, Jerry learns that they have also borrowed from Tom, and for some, renewing the loans would put their total debt burden above the level deemed acceptable to Jerry.  So what’s Jerry to do?  Turn these clients away?  Lecture them against borrowing from Tom?  Or avert its gaze, and renew the loans?

Jerry’s dilemma has no good options.  What makes it particularly insidious is that staying true to its principles and refusing to relend to clients who have over-borrowed from Tom would throw Jerry into financial turmoil, as it would now be in a position of absorbing the high costs of new client evaluation, only to lose many of them to competitors facing no such costs.  Averting its eyes and continuing to lend is risky too, but that risk is more remote.  Most clients will continue to pay, at least for the foreseeable future.

Reality is rarely as stark as the struggle between Jerry and Tom, but elements of this dynamic can readily be seen in microfinance markets the world over.  And given the options above, it’s no surprise that most Jerrys of the microfinance world, when pressed, put aside their principles and choose institutional survival, namely averting their eyes to over-indebtedness and continuing to lend.  And notice that this situation develops regardless of whether a credit bureau exists or not.

Escaping the vortex

The only sustainable means of avoiding such a dynamic is to seek an outside force that balances out the market, making Tom’s strategy too costly to pursue.  One such force is regulation, specifically client protection regulation that requires lenders to demonstrate that they have taken reasonable steps to verify that the client is able to repay the debt within accepted norms.  Some countries have this, but most do not.  I suggest the latter may want to study the success of the former.

There is also another way:  make Tom’s funding so costly that it will no longer have an economic incentive to take such an approach.  This is now happening.  Many foreign investors have signed on to the Smart Campaign’s client protection principles, including prevention of over-indebtedness.  The key to this is vigilance – every MFI would be required to demonstrate that it has evaluated clients’ repayment capacity according to specified norms and taken related steps (such as not overemphasizing new client acquisition in its incentive scheme).  Those whose lending methodology cannot be thus certified would lose access to a large part of microfinance funding, and would become limited to local markets only.  Exceptions could be made for young/small lenders, whose impact on local market dynamics is in any case minimal.

Microfinance is a sector where such self-enforcement might just work.  The Smart Campaign is working to develop a process for certifying an institution’s client protection practices.  The standard would then become a natural filter for funds, and funds’ own reporting on compliance rates would become a filter for investor capital.  The result could be a powerful force for shaping lending practices.  After all, most foreign microfinance investors expect to see positive social returns, and one can reasonably expect that most would use such certification as a key criterion in their investment decisions.

There would still be holes left.  Institutions with little dependence on foreign capital would still be able to operate outside these lending standards.   In some cases, such institutions may be large enough to dominate local markets.  It such situations, the sole answer is appropriate regulation.  To that end, concerned MFIs, investors, and industry advocates should be actively involved in lobbying for local regulator support.

Such efforts could well yield fruit, and in the end, perhaps Jerry Community Bank will find itself in a more sustainable market dynamic than having its clients chased endlessly by Tom Loans.


Note: This article was first published on the Financial Access Initiative: Link


Add a Comment

Your email address will not be published. Required fields are marked *