Thursday, April 10, 2008

Securitizations Will Stay Around

From IBD:

Just like junk bonds in the 1980s, this decade's mortgage-backed securities carry the demon label.

Back then, brainy bankers repackaged high-yield junk into complex securities, just like subprime mortgage debt.

Still, junk bonds rebounded. Most high-yield debt issued by U.S. companies now goes toward general corporate purposes, not buyouts like 25 years ago.

Mortgage-backed securities will bounce back, too, experts say, though reforms are much needed.

"Junk bonds are an interesting analogy," said Robert van Order, professor of finance at the University of Michigan and a former chief economist at Freddie Mac. (FRE)

"It's very much the same idea, that these were risky loans, and deals were set up to go through the bond market," he added. "The key going forward is that loan originators take bigger equity positions in securities, that they keep more skin in the game. And investors will look for more transparency and information in deals."

I am very glad that someone wrote this article.

As my screen name implies, I use to be involved with the bond market. I was a bond broker with several regional firms, although now I am in graduate school studying for an LLM in international taxation. As such, I have some familiarity with the market that is currently being trashed in the press. Here are some points we need to remember.

If memory serves, the first pass-through mortgage deal was done in the late 1970s or early 1980s. It was a simple pass through deal, meaning all that was really involved is an investment bank bought a lot of mortgages with similar coupons and maturities, put them in to a pool, and then sold it to a third party. By today's standards, it was an incredibly simple deal. But, those types of securities are still around.

The process started to get more complicated when investment banks started to "carve cash flows" from the same pools of mortgages. This means that instead of there being only a single pool of mortgages that was sold to a third party, there were now multiple bonds with different maturities and interest rate characteristics. For example, there were now "interest only" bonds or IOs which only paid the interest from the mortgages to the bondholder and its flip-side "principle only" or PO bonds (which worked in a very similar way to zero coupon bonds).

The point here is investment banks learned that fixed-income investors had different needs. For example, an insurance company with a particular anticipated liability stream may not need money until a particular pay window opens up. For example, an insurance company that deals in auto insurance may anticipate they will need to pay out $5,000,000 in three years (and yes, their actuaries are that good and accurate). So they need a zero coupon bond that will mature say a month or two before that liability stream opens up. As such they purchase a zero coupon bond or PO to solve that need. Now, everybody is happy.

These types of bonds are called "collateralized mortgage obligations" or CMOs and they started to get a bad reputation in the late 1980s and early 1990s. Several prominent municipalities (of which I think Orange County California was one) had their portfolios go belly-up or close to bankruptcy because of CMOs. As a result of those stories CMOs got an extremely bad reputation. Frankly, I forget what the exact facts of the situation were -- that is, I forget how the portfolio was structured etc....

Let's fast-forward to now. A lot of the alphabet soup of names you're seeing in the financial press are getting hit really hard. For example, collateralized debt obligations (CDOs) and collateralized loan obligations (CLOs) are now getting hammered just as hard as CMOs use to be. Both of these types of securities are based on the same concepts as CMOs; that you can carve different case flows from a pool of collateral to serve a variety of needs. There is nothing inherently wrong with that idea now, just as there was nothing wrong with it 15 or 20 years ago.

So -- what went wrong?

A lot. The main issue over the last 3-5 years is a complete breakdown of underwriting standards. Mortgage brokers had no incentive to verify income, instead going for fat commissions that riskier loans offered. Investment banks had no incentive to verify collateral's characteristics because the bank was going to sell the collateral anyway. And the ratings agencies gave everything a AAA rating without much thought.

As a result, there is a lot to fix. But that doesn't mean the basic concept is wrong or should be abolished. It shouldn't be because the raw idea is still sound and has a track record of success. But there are a lot of things that need to change.

For more on CMOS - here is a link to the wikipedia entry which is really good.