Posted by Mitch Townsend on January 26th, 2008 (All posts by Mitch Townsend)
Before this is through, nearly everyone on the planet will have expressed an opinion on the sub-prime mortgage crisis. It’s a little late, but I thought I should get mine in. Here are some points about the issue that I don’t think have been given much discussion.
Mortgage-backed derivatives are not new. Some 20 years ago, FNMA introduced the REMIC (Real Estate Mortgage Investment Conduit). These were pools of mortgages that were split into various tranches or classes of maturity and quality, which were then sold separately, similar to the way today’s collateralized debt obligations (CDOs) are sold.
There were some important differences between this first generation and its descendants. I would like to point out some of the differences, since they may highlight the reasons behind the collapse.
When REMICs were issued,
- The mortgages in the pool had to conform to the agency’s standards, or they could not be used. The agencies had minimum debt payment to income ratios, minimum down payments, maximum loan amounts, and other requirements designed to make sure the borrower could repay the loan.
- FNMA used to use only “seasoned” mortgages for the pool. This meant that the homeowner had been paying the mortgage for a couple of years. Any obvious payment problems, flipping, or serial refinancing had been largely taken out of the pool. Prepayments and defaults were reasonably predictable.
- FNMA had a much-disputed soft guarantee from the federal government. None of its debt was issued with the “full faith and credit” of the government, unlike a Treasury bond, but it was generally believed that the government would step in if there was a question of default or bankruptcy.
- You only owned what you bought. It was always possible to buy on margin, but the buying and borrowing were separate events. Too much leverage would draw scrutiny from your lenders.
By contrast, here is how CDOs, including bundled sub-prime mortgages, were handled:
- Sub-prime mortgages were almost all originated by mortgage brokers. Low initial “teaser” interest rates permitted loans to be made to people who could never afford the payments at a market rate. While there were there were supposed to be requirements to ensure creditworthiness, fraud and misrepresentation were widespread. It was always understood that a sub-prime mortgage was risky; soon it emerged that some of the mortgages did not even meet the lax standards supposed to be in place.
- Sub-prime mortgages and other loans were bundled and securitized almost immediately, not seasoned. The volatility of the first few years of a loan was incorporated into the securitized asset. We now hear that the problem became obvious when the adjustable sub-prime mortgage rates reset after two years.
- Non-governmental CDOs automatically have counterparty risk, the risk that you cannot collect from the other party to the transaction. If you follow the process from the original mortgages, to the bundling and securitization, credit enhancement, and credit insurance, there are quite a few parties involved. If something goes wrong, the mortgage broker is likely judgment-proof; the bank doing the bundling and securitization is liable only for gross negligence or actual knowledge of fraud by the broker. MBIA and AMBAC, the companies issuing the credit enhancement (essentially a guarantee), are themselves in financial difficulty and may not be able to fulfill their obligations. Credit insurance, in the form of credit default swaps, is only as good as the counterparty to the swap. ACA, one of the largest issuers of credit default swaps, is unable to make good. Its creditors, including Merrill Lynch, have already reserved for losses from ACA. (MBIA and AMBAC are also credit default swap dealers.)
- Credit default swaps began as a way of reducing risk. In a credit default swap, one party buys insurance against default on one of its debt holdings. The other party (the one issuing the insurance) receives a regular stream of payments while the contract is in force. If a specified “credit event” (bankruptcy or missing a payment) happens, the insurer must either buy the defaulted bond at par or make up the difference in the bond’s value. The volume of the credit default swaps market, in notional value, is now greater than the par value of the underlying bonds. This indicates that there is more speculation than insurance in credit default swap activity. Hedge funds are significant buyers and sellers of credit default swaps. In theory, the buy and sell side are balanced. In practice, if one party cannot meet its obligations, the others who cannot collect may not have the liquidity to meet their own obligations.
The holders of CDOs are suddenly coming to realize that they are pretty much on their own. The insurance and the guarantees they had relied upon are very much in doubt, and they have no idea what exactly they own and whether it is any good. Until they figure that out, they cannot sell it, and no one will buy it. Their holdings are illiquid, and if they can be valued at all, there is a huge liquidity discount on top of a newly-discovered risk discount. Banks and other holders have had to write down the values of their holdings and take the losses against their earnings.
The thing to watch in the next few weeks is the hedge funds. They do not have the same stringent valuation and reporting requirements as banks or other public holders. The only hedge funds we have heard from are those associated with banks. Let’s hear what the others have to say.