Color-Blind Merrill in a Sea of Red Flags

Posted by archive 
May 16, 2008

Would you invest money — at a very low interest rate — to finance mortgage loans made to risky borrowers who put no money down? What if you knew the company that made most of the loans had gone bankrupt because so many of its loans had turned bad almost immediately?

Now, no one would do that. But it was just a year ago that Merrill Lynch was wrapping up a securitization that met just those criteria. The securities were snapped up by buyers like the Bond Fund of America, one of the largest mutual funds.

Now, that mortgage securitization is a candidate for the title of worst ever. This week, the Merrill securitization had its first anniversary, an event that brought no celebrations, and Moody’s forecast that by the time it is wrapped up, so many of the mortgages will have gone bad that 60 percent of the money lent will not be paid back. Already some public investors are seeing their money vanish, and it seems certain that more will follow.

It turns out that so-called piggyback loans are highly vulnerable to homeowners who walk away when property values fall. In a typical deal, a buyer took two mortgage loans, one for 80 percent of the purchase price, and the other to provide the remaining 20 percent.

It is the latter loans that this securitization, and others like it, now own.

“In light of the pressure on home prices and limited or negative borrower equity in their homes, many second liens were simply written off” after several months of payments were missed, Moody’s said.

With these loans, it turns out, foreclosure is seldom worth the effort, since all the money would go to the first mortgage holder.

Making the situation worse is the nature of many of the mortgages in the Merrill securitization. Fewer than 30 percent of the loans were made to borrowers who provided full documentation of their income and assets. Many of the other borrowers probably lied about their income. Nearly all had borrowed the full appraised value of the home, either for the purchase or for refinancing, and it is possible that some appraisals were unreasonably high even before home prices began to fall.

The loans in this group did not go to solid credits. Although market interest rates were low when these mortgages were written, the mortgages had rates averaging 11.2 percent. Yet investors who put up most of the money were willing to accept a floating rate of just 30 basis points — three-tenths of one percentage point — over the London interbank offered rate. At the moment, that gives them a yield of 3.2 percent.

If Moody’s is right, those investors will eventually suffer capital losses. That is quite a combo for a security: low yield and high risk.

Moody’s could be wrong, of course. A year ago, it thought those senior securities were Aaa, the highest rating possible. Now it rates them as Baa3, one notch above junk. But if it is wrong, it could turn out to still be too optimistic. Standard & Poor’s cut its rating of that security from AAA to B, a low level of junk. (On Thursday night, Moody's cut its rating to B3, comparable to the S.& P. rating.)

S.& P. is even more negative than Moody’s about the likely recoveries from securities backed by piggyback loans, and wants nothing more to do with such deals. Citing an “unprecedented level of loan performance deterioration,” it announced this month that it would no longer rate such securities. At the moment, there is no chance that any will be issued, so S.& P. is not passing up any potential revenue with that decision.

As with other mortgage-backed securities, the ones sold to finance this package of loans stand in line to take losses as they occur. After one security is wiped out, the losses start accruing to the next one on the totem pole. The lowest publicly held security has already lost most of its capital, and others are virtually certain to follow.

A majority of the loans backing this securitization were issued by Ownit Mortgage Solutions, a California-based firm that went bankrupt in late 2006 when it was unable to repurchase loans that went bad almost immediately.

In the prospectus for the securitization, Merrill disclosed the bankruptcy but did not mention that it had been Ownit’s largest creditor. Nor did it point to the rapid default of previous Ownit loans. But those facts had been in the newspapers, and you might think supposedly sophisticated institutional investors would have noticed. Neither Merrill nor the managers of the Bond Fund of America were willing to comment this week.

It does not come as a surprise that the defaults seem to have been worst in California, which provided 40 percent of the original loans and 54 percent of the ones that have run into serious difficulty so far. Florida and Nevada are also overrepresented in the problem loan group. Home prices have fallen in these states, and California and Florida loans were particularly likely to have been made without normal documentation.

The Merrill deal is not unique. Moody’s lists three others that are likely to have comparable losses. All are from 2006, including one underwritten by Goldman Sachs. But none of them were sold after the original lender had gone broke amid evidence that it had a penchant for making bad loans.

All this seems like a distant memory now. Then, the investors and the ratings agencies seemed to have assumed that the housing boom would continue, and that even risky borrowers were unlikely to bring big losses. They did not understand how much underwriting standards had deteriorated, or how likely owners were to walk away from second mortgages if home prices plunged.

Perhaps some of those owners have received effective reductions in the principal they owe, something that homeowner advocates have been calling for and that lenders have resisted. With the second-lien holder taking its losses and deciding that foreclosure would be pointless, those owners are left with a first mortgage for 80 percent of the price they paid. Perhaps they will be able to meet those payments, and keep the house they probably never should have purchased to begin with. Unfortunately for them, the fact that the lender stopped trying to collect does not wipe the lien off the books, and it would resurface if the homeowner ever sought to sell the home.

Floyd Norris comments on finance and economics in his blog at norris.blogs.nytimes.com.