Friday, February 09, 2007

From the ridiculous to the sub-prime

Mortgage lending stocks got pummelled yesterday, particularly companies involved heavily in the sub-prime market. It all came on the heals of a report from HSBC that they may have to set aside an additional $1.8 billion to cover sub-prime loans going bad. If you are wondering how HSBC became such a player in the U.S. sub-prime market, remember that they bought Household Bank a couple years ago. I had to laugh when I read in HSBC's release that fraud incidents were higher than they expected. According to Bear Stearns, 49% of sub-prime loans made for home purchases where of the so-called "limited doc" type, which means the borrower could not document his/her income. So borrowers with poor credit to begin with are coming to a bank to borrow money, but claim they cannot prove their income. And HSBC is surprised there is a fair amount of fraud there?

What does this mean to the bond market? Bonds backed by sub-prime loans have been getting killed. The BBB ABX-HE 06 index (which tracks BBB-rated home-equity loan pools originated in 2006) was trading above $101 as recently as December, just closed at $95.45. The next shoe to drop will be ABS CDOs.

A CDO (collateralized debt obligation) is a derivative structure where investors can buy different slices of credit risk on a large portfolio. The most junior tranche takes all losses until it is completely wiped out, then the next most junior tranche gets hit, and so on. At the very bottom of the structure is the equity holders, the actual owners of the portfolio. CDOs are created backed by various types of credits: from bank loans to CDS contracts. But ABS-backed deals tend to be the most levered: the equity is commonly only 1-2% of the entire deal size in ABS deals. So the equity holder has something on the order of 100x leverage.

These deals are structured to produce large returns for the equity holder if defaults are relatively low and spread out over time. But we can imagine that the nationwide housing downturn coupled with overly aggressive lending standards will lead to a larger number of defaults concentrated in a small period of time. This is not something a lot of ABS-backed CDO deals can handle. Not only will the equity holders lose all their principal, but if losses are severe enough, even the most senior tranches will start to suffer losses.

Unfortunately, this could wind up tainting the entire CDO market. Which is too bad, because CDOs are a great way for investors to express a specific view on credits. With a CDO structure, an investor can buy into a diversified portfolio of high-yield bank-loans, but instead of having a series of B and BB-rated bonds, they can buy an A-rated tranche, taking comfort in the fact that if the portfolio experiences some losses, there are probably 2-3 tranches junior to the A-rated piece.

CDOs went through this before. In 2001-2002 many deals backed by high-yield bonds imploded. The market recovered nicely to where it is today, but now deals backed by entirely high-yield bonds are extremely rare. The same fate could befall the sub-prime ABS market, and will have a long-term impact on spreads in that market.


Anonymous said...

great article. i really enjoy reading the blog and getting smarter on the credit markets. keep up the good work!

Anonymous said...

Thanks for the explanation. I'm not that conversant in credit markets and appreciate the explanations. MA had an adverse reaction in the market today -do you think the sub-prime problem has people worried about credit in general? The IYR property managers got whacked as well, and the homebuilders aren't looking good. Assuming that the sub-prime problems scare lenders into cleaning up their books, will we see some builders and/or property managers suffer?
As I said, I'm not conversant in the credit markets. I suspect that the damage will be contained to some sub-prime lenders and to the credit instruments you mentioned in the post, but is there the potential for this to spread further?

Accrued Interest said...

I think that home builders with strong balance sheets will survive without trouble. I'm long RYL bonds for example. Can't comment on the stocks. Not my bag baby.

I don't think problems in the ABS market will hurt credit generally, unless the CDO market dries up entirely. CDOs have been a strong support of credit spreads generally. As long as investors view problems with ABS CDOs as isolated (which I think they will), credit spreads will remain around where they are in high-grade.

High yield is priced for perfection and I think is pretty risky here.

Deborah said...

I really liked this article. It is using a lot of terms and items that I am unfamiliar with, so for example, I don't really understand how the ABS thing ends up being 100x leveraged. But, I do understand the high risk of owning something with this kind of leverage...

Anonymous said...

makes you wonder how HSBC became such a huge bank in the first place

Accrued Interest said...

The 100x leverage works like this:

CDO Equity investors put up $5 million.

The CDO Trust then sells $495 million in debt. The trust then buys a $500 million portfolio of bonds.

Typically there would be debt pieces rated AAA, AA, A, BBB and maybe BB. Each piece with a higher rating is senior to the pieces with lower ratings. Only once the pieces with the lower ratings are wiped out do losses flow through to investors with higher ratings.

The equity buyer gets what's left after all the debt pieces have been paid.

But notice the leverage. $5 million in equity to own a portfolio of $500 million in bonds!

This high level of leverage is only typical among deals where the collateral portfolio is highly rated. In many deals the collateral portfolio is junk-rated bank loans. In those cases the leverage is much lower.

Anonymous said...

good points on these instruments. but the real issue is who is holding the equity portions and when the chips fall who will have to pick up the losses.

history shows us traders inside the industry will/have probably offloaded a good percentage of non-performing equity onto retail investors via a range of synthetics.

hsbc may get the red face but i am pretty sure investors in their funds will pick up the tab. bonuses all round.

Anonymous said...

Wondering where the equity goes? Straight into a deep black hole on the balance sheet of the originating bank. The fees on issuing the highly rated tranches of these deals make it possible for the investment bank to completely write off the equity value of the deal and keep it on the balance sheet. They get a nice earnings pop if everything works out and end up making their fees if it doesn't. Nothing like a free option eh?

Accrued Interest said...

Anon is right in a lot of cases. Particularly with deals involving bank products, like loans.

If the bank can charge 20bps to run the CDO, then on a $500 million deal, they get $1 million/year. So even if they get a paltry return on their equity piece, they still did alright on the fees. Plus, they don't have anywhere near the capital commitment they had when it was a huge portfolio of loans.

So it may sound like the bank gets off easy for taking all this risk, but really they've spread the risk out. Which makes our system better, not worse.