Reactions to New York Times Mortgage Market Story

Yesterday, we had a link and some commentary on a front-page New York Times article, “Crisis Looms in Mortgages,” by Gretchen Morgenson.

Morgenson likes a take-no-prisoners style of writing, and she tends to be controversial due to her forceful articles about CEO pay. I will confess to having read past it (I am inured to her tone), but at the beginning of her piece, she discussed the failure of a subprime mortgage originator, New Century Financial, and then made a comparison to the dot com era.

The problem is that comparison between the mortgages and dot coms. She does allude to a distinction between subprimes and the entire mortgage market (“as it involves the nation’s $6.5 trillion mortgage market”) but it is awfully easy to miss or misconstrue. It’s too easy to read her set up as saying the entire mortgage market is imperiled by subprimes, and the prospects are as bad as they were for dot coms. A comparison between the overall mortgage market and dot coms would be barmy. After all, there is a lot of fundamental value in housing (duh) while a ton of dot coms went poof.

So my take is that Morgenson is guilty of sloppy writing in not inserting the word “subprime” often enough before the word “mortgage” in her piece. Maybe I am being too charitable. But if you read in context, it’s pretty clear that the vast majority of her uses of the word “mortgage” in the story refers to subprimes.

But that sloppiness in drafting, which is extreme, particularly in a page one story, doesn’t fully explain the vitriol of the reactions to her piece. I detect a considerable resentment to the fact that she writes in a swashbuckling fashion and gets away with it because she usually has the goods. If you write like her, it is incumbent upon you to be pitch perfect.

Now to the substance. It is completely fair to say that there is a mess in subprimes, that alt-As (the next tier of credit) are also in pretty bad shape, that subprimes alone constituted 20% of new mortgages in 2005 and 2006, and the loss of these buyers (and the return of some of their houses to the market as foreclosures) will slow, maybe reverse the housing recovery.

It is also fair to note that most subprime borrowers were buying lower priced housing, so their share of the market in dollar terms, as opposed to number of mortgages, is smaller, so it’s likely the damage to the housing market will be localized. On the flip side, banks tend to be lemmings, so the fact that they screwed up on weak borrowers means they are likely to be more cautious with better borrowers. This too would have a depressing effect on demand for housing. And that could have much greater ramifications, since consumers are overextended (consumers spent more than they earned in 2005 and 2006) and continued weakness in housing could finally induce them to curtail spending.

The other weak link is potentially a repudiation of risk in the credit markets (a levelheaded piece in the Financial Times last weekend explored this idea). Risky credits of all sorts have been trading at very tight spreads, meaning the interest rates are way too low in relationship to the underlying quality of the borrower. Hedge funds have been particularly heavy buyers of these risky assets, and they would be likely to flee them if there were serious signs of trouble. So there is the potential here for broad scale economic damage, but several dominoes have to fall before that happens. The subprime problem alone won’t do it.

That is a long-winded way of saying that I believe Morgenson is directionally correct, although she overstated and overdramatized her case. And she isn’t the only one making dire forecasts. Nouriel Roubini, in his RGE Monitor, has, for example, “From the current subprime credit crunch to a generalized credit crunch: the incoming financial train wreck.” He is normally data driven and bearish by temperament, but he too has gotten a bit histrionic, which means either things are about to get wretched or he has developed a large ego investment in his point of view. We’ll know soon enough which is true.

There were some indirect and direct responses to her piece. The Wall Street Journal ran a page two story on Monday, “Subprime Fallout May Not Infect Broader Market.”

As more financially stretched homeowners renege on their debts, and mortgage lenders go under by the dozen, economists are surprisingly sanguine about the broader economy’s ability to weather the storm. But they add a big caveat: Much depends on how investors react to an increasing wave of worrying news, and how much some homeowners’ difficulties aggravate the nation’s deep housing slump….

“No doubt some of the worst practices of the housing boom are going to yield some payback,” says Steve Wieting, senior U.S. economist at Citigroup in New York. “But it’s not large enough to derail an otherwise healthy economy.”

He expects inflation-adjusted gross domestic product, a broad measure of the nation’s economic activity, to expand 2.6% this year, slower than normal but well short of a recession.

The main reason for economists’ equanimity: Those who took out subprime loans tend to be less-affluent consumers who make up a relatively small share of consumer spending, the most important driver of the U.S. economy. Labor Department data show that the fifth of U.S. households with the lowest incomes account for about 8% of all consumer outlays, while the most-affluent fifth accounts for nearly 40% of spending.

Meanwhile, the unemployment rate remains relatively low and incomes have been rising, suggesting poorer people have some resources to spend, even if they can’t afford their homes and can’t borrow money.

That said, the subprime mess has added some risks. The possibility economists fret about most is that investors and lenders will react to rising defaults by pulling back from all kinds of borrowers, good and bad — the sort of “credit crunch” that has triggered recessions in the past….

Still, the pullback in credit for subprime-mortgage borrowers could have a meaningful effect on its own. As some potential home buyers find it harder to get money and more bad loans beget more foreclosures, the decreased demand and increased supply of homes could depress prices, deepening the housing slump….

“In some of these regions you could have a pretty tough environment, in which a bad local economy, tightening credit and weakening home prices all kind of reinforce each other,” says Mr. Harris.

A direct response comes from Felix Salmon (note his post was also picked up on Mark Thoma’s blog, Economist’s View), “Is there a looming crisis in the mortgage market?” who was very critical of Morgenson’s piece. However, he went off the deep end a bit himself:

….she doesn’t seem to understand the difference between two entirely different types of investment: equity in subprime mortgage originators, on the one hand, and debt backed by pools of subprime mortgages, on the other. It’s certainly true that originating subprime mortgages does not seem to have been a very good business to invest in over the past year or so. But Morgenson never connects the dots and explains why that means that the market in subprime MBSs is likely to implode.

Morgenson also talks at great length about the enormity of the market in MBSs, but never stops to point out that the vast majority of that market is in bonds issued by Fannie Mae and Freddie Mac, and that no one has any worries whatsoever about those securities crashing.

Reading the piece again, more closely, I don’t think his charge is fair. In context, it is clear she is talking about subprimes (although there is a chart about the entire mortgage market that doesn’t help matters). But as we said earlier, the headline and body are poorly written in not saying often enough that the mortgages they are talking about are subprimes, so a casual or unknowledgeable reader could easily get the wrong impression (and that is made more likely by the article’s placement on page one, rather than in the business section). That, in combination with her characteristic strident tone might indeed make her seem hysterical.

I have a sneaking suspicion this was a regular weekly piece of hers that the powers that be decided to run as a general interest article, but it was not correctly repurposed. Another bad sign about the state of editing at the Times.

Salmon next takes on Morgenson’s argument that investor losses related to the subprimes haven’t been recognized in many cases because the instruments are held by investors that don’t have to mark them to market:

Morgenson is missing two crucial points here. The first is that here simply isn’t a market in most MBSs tranches – that’s why so much of the recent activity has concentrated on MBS indices rather than the underlying securities. The liquid, mark-to-market activity that Costello is talking about is entirely in Fannie and Freddie bonds, not in individual tranches of securitized subprime mortgages. You can’t mark subprime MBS tranches to market daily for the very good reason that most such tranches simply don’t trade on a daily basis.

And the second point is that if you actually look at the prices for those subprime MBS tranches when they do trade, guess what? They haven’t actually fallen much in price at all. If investors were marking to market, it really wouldn’t make much difference. As Josh Rosner told me, the problem is not that existing MBSs are likely to default or drop in price. A default is much like a prepayment, from an investor’s point of view, so investors only really care about default rates when they start approaching prepayment rates. And they’re nowhere near those levels.

Morgenson wasn’t on the mark here, but Salmon isn’t either.

These subprime mortgages could wind up one of three places: in a regular mortgage backed security (that is, in a pool with other mortgages of similar type, likely by geography and credit), in an investment bank warehouse waiting to be packaged, or in a collateralized debt obligation. These are much more complicated instruments, often involving derivatives, Most subprimes wind up in CDOs, not MBS. It’s not clear if Morgenson appreciated this distinction (likely not, but one can’t say so conclusively); I noticed when I read the piece that she finessed the point by saying the subprimes went into “pools,” which could be either MBS or CDOs.

As an aside, Salmon mentions in passing that subprimes issued by Bank of America had an 82% loan to value ratio, as if that was a good number. That’s not a good number, particularly for no doc borrowers. The approach when banks are lending on a normal basis, as opposed to frothy times like the last couple of years,is to have much more collateral in loans to dubious borrowers. If you start with an 82% loan to value ratio, throw in 1-2% costs of making the loan, another 1-2% in costs to foreclose, 6% brokerage fee to sell the house you now own, 10+% fair value of money for that credit, 1 year of lost interest income between foreclosure and sale time, some costs in fixing up the house for sale, which might be more than cosmetic if the angry owner trashed it before eviction, which happens, and you are at breakeven or a loss. And housing prices are likely not going up in the markets that have a lot of subprime loans. So loss is a real possibility here. And the BofA loan to value ratios are likely to be as good as they get. A May 2006 WSJ story reported that 29% of the mortgages in 2005 had no down payment. Ahem, that means no equity.

Back to CDOs. They are levered, hence losses are multiplied They are often bought by hedge funds, which in turn are levered. The Financial Times in a January article on CDOs pointed out how little equity is underneath these puppies.

And despite Salmon’s assertion that a default is like a prepayment, hence no big deal, in fact the yields on CDO-related bonds are up 4%. So the markets are saying the value of these bonds is impaired. That equates to a serious downward move in the bond markets (remember higher yields translates into lower prices). And dealers have cut back on lending to the usual buyers of these securities. That means they’ve marked down the value of the collateral on existing margin loaned to these buyers (ie, the CDOs and whatnot they already own) to the point where they won’t extend new credit. That is a pretty serious reduction in value.

And his argument that an instrument not trading means you don’t mark it to market is also wrong. Securities firms mark all their inventory to market. Any investor who had to mark a security to market could find a dealer who would give him a price (certainly the lead manager of an MBS underwriting or the packager of a CDO; they stand ready to buy and sell the securities they originated). The price would be extrapolated from the prices of comparable, liquid securities.

Here is a Bloomberg article that in passing discusses the relationship between subprimes and CDO, and the fall in CDO prices, “Subprime Mortgage Bonds Extend Drop as Dealers Cut CDO Funding.”

Subprime mortgage bonds are falling in part because Wall Street dealers are lending less money to managers of collateralized debt obligations that buy the securities, according to investors.

“The dealers were caught off guard by the speed of the collapse and were still trying to aggressively” encourage the creation of new CDOs before pulling back, said Peter Schendel, who helps manage $100 million in asset- and mortgage-backed bonds at hedge fund Good Hill Partners LP in Westport, Connecticut.

The drop in demand caused yields on subprime mortgage bonds with the lowest investment-grade ratings to double to 8 percentage points over benchmark rates last month, according to New York-based Lehman Brothers Holdings Inc. The rates consumers with bad credit pay on home loans fluctuate with the yield that bond investors demand to own securities backed by the mortgages.

CDOs repackage loans, bonds and derivatives as new securities. Strategists at firms including JPMorgan Chase & Co. say spreads on mortgage CDOs will widen further. That suggests subprime bonds held in “warehouses” waiting to become part of a CDO won’t be able to be repackaged at a profit. Subprime loans are made to homebuyers with poor or limited credit histories.

Securities firms often either share or take all of the risk of CDO securities not being sold for enough to create a profit, or of bonds expected to be put into a CDO needing to be sold at loss instead, taking interest payments in the interim in return.

`Real People’

About $173 billion of CDOs backed mainly by U.S. subprime mortgage bonds and related derivatives were created last year, according to JPMorgan.

“Real people didn’t buy” subprime bonds at the spreads being offered, said Scott Simon, head of mortgage- and asset- backed bond investments at Newport Beach, California-based Pacific Investment Management Co., manager of the world’s largest bond fund. “CDOs bought the bonds.”….

Subprime mortgages made up about a fifth of new mortgages last year, according to the Washington-based Mortgage Bankers Association. About 10.12 percent of the loans in securities were delinquent by at least 90 days, in foreclosure or already turned into seized property on Dec. 31, up from 5.37 percent in May 2005 and the most in at least seven years, according to a March 2 Friedman Billings Ramsey Group Inc. report…. rates.

Lower demand from CDOs is encouraging subprime mortgage bond issuers to hold more of the securities they create, “at least until the market retains its footing,” Credit Suisse Group analysts led by Rod Dubitsky said in a March 2 report.

TABX Contracts

Pressure on the spreads has also come from a new type of derivative introduced on Feb. 14, called TABX contracts.

The contracts re-slice the risks of credit-default swaps on low-rated securities, just as CDOs do with the actual bonds. TABX contracts that look similar to high-rated bonds from CDOs have carried premiums more than 10 times the spreads on CDOs, leading some investors to want more yield on the CDO securities.

“Either it’s absurdly cheap” to buy the contracts or too expensive to buy CDOs, said Howard Hill, a managing director in Springfield, Massachusetts, at Babson Capital Management LLC. The unit of MassMutual Financial Group manages about $30 billion of mortgage- and asset-backed bonds….

The drop in CDO demand may be temporary as rising yields on subprime mortgage bonds make it possible to offer higher yields on CDO securities.

“Maybe yesterday’s bonds aren’t going to make it, but going forward it still might work if you bought new” bonds at wider spreads, said Jeremy Shor, a portfolio manager at Brown Brothers Harriman & Co. in New York, who oversees about $3 billion in asset-backed securities.

Print Friendly, PDF & Email

One comment

  1. Felix

    Wow, thanks for such a wonderful response!

    We can fairly differ, I think, on whether GM was talking only about subprimes or about mortgages in general. The fact that she talks about the mortgage market as a whole, and things like the Costello quote, make me think the latter. But if it’s the former, then I think a lot of the scaremongering and worries about systemic risk go out the window, no? Subprime can’t drive the financial system into chaos on its own without the problems spilling over into mortgages in general and other credit product in particular.

    As for loss being a real possibility on subprime loans, well, yes, of course it is. Subprime loans are risk products, after all. But remember that most debt is unsecured, so from that point of view secured debt is much more attractive, even if it’s impaired. Look at scratch & dent prices in the 80s.

    But I’d like to ask you about this distinction you’re drawing between CDOs and MBSs. Why is the distinction important? I understand that CDOs are levered, and so if they can’t lever up then they’re not going to be the same source of demand for mortgages that they used to be. But I have a mental model of the mortgage market whereby subprime mortgages are bundled into MBSs, which in turn are sold to CDOs, among other investors. The demand for MBSs from CDOs has gone down — but they way I see it this can all be modelled as lower demand for MBSs in general, and you don’t really need to worry too much about where the demand is or isn’t coming from. Now it might well be the case that some CDOs bought mortgages directly rather than MBS tranches, but does that really make any practical difference?

    In general, I see equity investors in CDOs as precisely the sort of people who can and should take lots of pain as a result of the problems in the subprime market.

Comments are closed.