Tuesday, October 23, 2007

Microfinance Bonds! - Liberal Goals through Conservative Means!

Following the mortgage backed securities which ended up as the underlying payment stream against whose stream of payments commercial paper was issued – prior to the credit crunch – there is a generic aversion today to getting deeply involved in that whole rigmarole again. But, innovation and profits are the life-blood and élan-vital of modern finance. So, bonds backed by microfinance lending are here to stay. Famous CLOs backed by loans to microfinance institutions who in turn lend to thousands of really poor individuals. For eg., see BlueOrchard. The exact fate of such CLOs, in wake of the credit crises, is unclear. Nevertheless, I suspect, this is an instrument that is here to stay.

A key issue here is what are the risks involved in such a setup. i.e., when does repayment risk amongst the really poor kick in. By and large, those individuals who qualify for microfinance are those whose income patterns are only marginally affected by business cycle fluctuations, unlike the middle class. In contrast, the real risk comes from weather extremes, failure in governmental subsidy programs etc., So the crash of the NASDAQ or the Sensex barely affects the villagers in Guntur, Andhra Pradesh – but the increase in rationed rice and sugar prices due to governmental policies will affect their income and repayment abilities. So, the bonds/CLOs issued by Wall Street can be a useful market mechanism tool with governmental subsidies to alleviate poverty!

However, one interesting observation is that many of the borrowers borrow from various lenders. i.e., they might borrow from the microfinance initiative offered by ICICI Bank, by World Bank, by Bill & Melinda Gates Foundation etc. Naturally, this raises the issue of what are the likelihood of payments being made out. One my friend from college, Karuna Krishnaswamy, has written a pretty interesting paper on this phenomenon. He argues that those who end up borrowing from various sources – actually end up having a better payment record as opposed to the single source borrowers. The exact reason is conjectural, and I suspect, untraceable to a single factor.

For those who will eventually model these things, the level of complexity has been increased by an order of magnitude. Typical prepayment structures in the Mortgage world grapple with the issue of a single stream being pre-paid. In contrast, in the MFI backed bonds – any single borrower can decide to prepay one stream over the other!

Awesomely technical challenge!

No comments: