Saturday 3 May 2008

What can Indian Microfinance Learn from the Subprime Mortgage Crisis?


“Third-Party Originators/Lack of Accountability. Mortgage brokers, who originate the majority of subprime mortgages, have a strong incentive to close as many loans as possible, but very little reason to consider the loans’ future performance. Lenders shield themselves from the full potential cost of foreclosures by selling their loans to investors through the secondary mortgage market. Together, third-party originations and the risk dispersion made possible through the secondary market help distance loan originators from serious adverse consequences of foreclosures.”

This quote comes from a paper published the Center for Responsible Lending on the causes and effects of the subprime mortgage crisis. The quote comes from a list of the causes of the crisis, though I found this argument the most salient.

Although not a perfect parallel, there are definitely similarities between the subprime mortgage market and the microfinance market. Although subprime mortgages now have a negative connotation there are those who still believe that, if targeted correctly, subprime loans give those with lower income and less credit history the opportunity to purchase houses and develop wealth. Former chairman of the United Stated Fed, Alan Greenspan even lauded subprime mortgages as “the democratization of credit.” This all may sound familiar to those who follow the MF industry. Like subprime loans are characterized by higher interest rates than those for higher income families, so the same is usually true in MF. Just as people have believed that loans to the poor in developing countries are risky, so too for the poor in America. So clearly there must be something that the Indian MF industry can learn from this crisis.

There can be no doubt that the financial markets are becoming exponentially more interconnected and complex. Most people, myself included, simply find the futures market difficult enough to understand. Financial constructs like securitization, collaterized debt obligations (CDOs) and mortage backed securities are even more confusing and sophisticated. The basic idea behind all of these instruments is that now banks and investors can purchase the rights to repayment from a loan. Not only that, but people can even bet on the likelihood that these loans are repaid. Referring to this phenomenon in the housing market, The Economist, in its way with words, calls this the “the baroque superstructure of mortgage-backed derivatives.”

For example, lets say that today an MFI makes a loan of 10,000 rupees to an SHG in Chhattisgarh on 20% annual interest that is to paid in one year. The rights to the repayment on that loan can be sold by that MFI to an investment bank for 11,000 rupees (you will notice the MFI has made 1000 rupees and now has almost no risk). The investment bank may want to do this to diversify their holdings or simply to show their customers that they are involved in “socially responsible investing.”

In turn that investment bank may hedge their risk by selling the right to that repayment on the futures market. Meaning the investment bank may guarantee to another fund that they will sell them the rights to the repayment at 11,200 6 months from now. This is a highly simplified version of how these sales truly work, but that is the gist (usually the investment fund will buy the entire portfolio, so 10,000 rupees was just used for ease of understanding).

What I have just set out is essentially what people are referring to when they talk about the secondary and tertiary markets. What this all means is that the organization who distributed the loan and even the first investment bank that bought the rights to the loan, hold a reducing amount of risk, and therefore their incentives to assure repayment and target their loans well have been reduced. One the flipside, there is the positive that this allows banks to hedge their risks and raise capital, which is particularly important for fast growing MFIs. A strong argument has been made for securitizations when conducted responsibly and by well prepared institutions.

Several MFIs have already engaged in the sale of their portfolios, including BRAC and ProCredit Bank Bulgaria. The SHG-bank linkage model in India is also kind of a variation on securitization. In this model, SHGPIs (self help group promoting institutions), set up SHGs, which are linked to large banks who distribute the loan. But the large banks often rely on the staff of the SHGPI to assure the quality of the SHG and sometimes even to collect repayment. In this case, the SHGPI is comparable to the mortgage broker in the subprime market.

Like the phantom organization I used in the example earlier, the worry is that MFIs that have securitized their portfolios and banks, which have leant money through the SHG-bank linkage program no longer have the incentives to ensure quality. The intention is that these financial instruments are designed with these issues in mind, and they are able to mitigate these problems, but in reality, that is not often true, as we have seen from the subprime mortage crisis.

Expansion of MF and an increase in lending to the poor are certainly worthy goals, and the same is true of giving the poor in America the ability to borrow to purchase homes. The lesson to be learned for the Indian MF industry from the subprime mortgage crisis is that it is imperative to focus on keeping incentives aligned as that growth occurs.

For more info on the subprime mortgage crisis and its global ramification see this BBC news article.

4 comments:

Suyash said...

Dan,

Thanks for raising this important issue. Evidence on the consequences of misalignment between incentives for those who originate the risk and those who bear the costs of default has really started pouring in. Recently came across two more papers that shed light on this. Keys et al have found that the portfolio that was more likely to be securitized defaulted significantly more than a similar risk profile group with a lower probability of securitization. A paper by Mian and Sufi of University of Chicago also points at a similar phenomenon, how allowing originators to get rid of credit risk increased the lending to less creditworthy borrowers. The evidence is screaming, "Moral Hazard!!". It's clear that we need to stick to the basics when it comes to alignment of incentives, let the first loss be on those who assess the borrowers, and let this loss be big enough to be a credible deterrent.

The MF sector may end up paying indirectly for the reputation loss suffered due to the disaster in the securitizsed mortage papers. It's likely that it will become difficult to find takers for securitized papers internationally (in India, priority sector requirements should offset this), and also reverse the so-called democratization of credit (read this caustic article by Michael Lewis).

Given the importance of securitization for raising money for micro finance, we should see if it is possible to insulate micro finance market from the reputation loss suffered by the market for securitized mortgage papers. Because of the usually non-collateralized nature of the loans, the risks associated with speculation on the underlying asset do not exist in micro finance markets in the same way. The micro finance loans are usually small loans made for short or medium term, and with less complexity in design, thus reducing certain risks associated with predicting households' cashflows for distant future and also makes it easier for the borrowers to make a measured choice. This size and tenure of the loans also affects the incentives for the originator, who, in case of default, would get a major reputation hit for a relatively small, short-term gain.

As the sector grows and sophistication increases, we need to ensure that the growth is managed in such a manner that the micro finance sector doesn't suffer from the consequences of excess that the mortgage market has had to suffer. For instance, as one of the papers points out, one of the factors behind the sub-prime crises is that the pricing models put excessive weight on individual credit scores, and ignored essential elements of strategic behavior on the part of lenders. As the use of scores increases in MF, they should not be seen as replacement for the lender's wisdom.

It seems there is a long tightrope walk ahead.

Dan Kopf said...

Suyash,

Thanks for these terrific and insightful comments.

I agree with you that the repercussions for MFIs and mortgage brokers are quite different. For lack of space, this is something I left out of my post and probably should not have. As you pointed out, the big difference is that the MFI will take a major reputation hit if they neglect repayment, and the loans they are distributing are comparatively simple (with no exploding ARMs).

Thank you for making me even more aware of these moral hazard issues.

Dan

Dan Kopf said...

Suyash,

If you read this, please let me know whether you would mind if I posted your comment in the front the blog. I think your thoughts are truly insightful and it would be nice to highlight them.

I could not find your email address.

Dan

shashi singh said...

If you read this, please let me know whether you would mind if I posted your comment in the front the blog. I think your thoughts are truly insightful and it would be nice to highlight them.

http://www.leavetheloan.com/