Roger Lowenstein on the Subprime Blowup

Roger Lowenstein almost makes me sorry for Moody's.

First: God! What a lousy lead!

Moody's - Credit Rating - Mortgages - Investments - Subprime Mortgages: In 1996, Thomas Friedman, the New York Times columnist, remarked on “The NewsHour With Jim Lehrer” that there were two superpowers in the world — the United States and Moody’s bond-rating service — and it was sometimes unclear which was more powerful. Moody’s was then a private company that rated corporate bonds, but it was, already, spreading its wings into the exotic business of rating securities backed by pools of residential mortgages...

I find it hard to imagine how anybody could think that this is an informative way to begin an article.

The body of Lowenstein's article is pretty good:

[S]uppose the security had a rating. If it were rated triple-A by a firm like Moody’s, then the investor... wouldn’t need to know what properties were in the pool, only that the pool was triple-A — it was just as safe, in theory, as other triple-A securities. Over the last decade, Moody’s and its two principal competitors, Standard & Poor’s and Fitch, played this game to perfection — putting what amounted to gold seals on mortgage securities that investors swept up with increasing élan.... By providing the mortgage industry with an entree to Wall Street, the agencies also transformed what had been among the sleepiest corners of finance. No longer did mortgage banks have to wait 10 or 20 or 30 years to get their money back from homeowners. Now they sold their loans into securitized pools and — their capital thus replenished — wrote new loans at a much quicker pace....

But who was evaluating these securities? Who was passing judgment on the quality of the mortgages, on the equity behind them and on myriad other investment considerations? Certainly not the investors. They relied on a credit rating. Thus the agencies became the de facto watchdog over the mortgage industry....

Moody’s and S&P... they reject the notion that they should have been more vigilant. Instead, they lay the blame on the mortgage holders who turned out to be deadbeats, many of whom lied to obtain their loans.... Moody’s recently was willing to walk me through an actual mortgage-backed security step by step. I was led down a carpeted hallway to a well-appointed conference room to meet with three specialists in mortgage-backed paper... the case they showed me, which they masked with the name “Subprime XYZ,” was a pool of.... All the mortgages in the pool were subprime.... In an earlier era, such people would have been restricted from borrowing more than 75 percent or so of the value of their homes, but during the great bubble, no such limits applied.

Moody’s did not have access to the individual loan files, much less did it communicate with the borrowers or try to verify the information they provided.... “We aren’t loan officers,” Claire Robinson, a 20-year veteran who is in charge of asset-backed finance for Moody’s, told me. “Our expertise is as statisticians on an aggregate basis. We want to know, of 1,000 individuals, based on historical performance, what percent will pay their loans?” The loans in Subprime XYZ were issued in early spring 2006.... The investment bank provided an enormous spreadsheet chock with data on the borrowers’ credit histories and much else that might, at very least, have given Moody’s pause. Three-quarters of the borrowers had adjustable-rate mortgages, or ARMs... 43 percent... did not provide written verification of their incomes... 12 percent of the mortgages were for properties in Southern California, including a half-percent in a single ZIP code, in Riverside.... 94 percent of those borrowers with adjustable-rate loans said their mortgages were for primary residences... all of the ARM loans in the pool were first mortgages (as distinct from, say, home-equity loans). Nearly half of the borrowers, however, took out a simultaneous second loan. Most often, their two loans added up to all of their property’s presumed resale value....

[T]he Moody’s analyst had only a single day to process the credit data from the bank. The analyst wasn’t evaluating the mortgages but, rather, the bonds issued by the investment vehicle created to house them.... [T]he S.P.V. would float 12 classes of bonds, from triple-A to a lowly Ba1.... It was this segregation of payments that protected the bonds at the top of the structure and enabled Moody’s to classify them as triple-A....

However elegant its models, forecasting the behavior of 2,393 mortgage holders is an uncertain business. “Everyone assumed the credit agencies knew what they were doing,” says Joseph Mason, a credit expert at Drexel University.... Mortgage-backed securities like those in Subprime XYZ were not the terminus... were, in fact, building blocks for even more esoteric vehicles known as collateralized debt obligations, or C.D.O.’s.... Miscalculations that were damaging at the level of Subprime XYZ were devastating at the C.D.O. level....

When a bank proposes a rating structure on a pool of debt, the rating agency will insist on a cushion of extra capital, known as an “enhancement.” The bank inevitably lobbies for a thin cushion (the thinner the capitalization, the fatter the bank’s profits).... [I]n structured finance, the agencies face pressures that did not exist when John Moody was rating railroads. On the traditional side of the business, Moody’s has thousands of clients (virtually every corporation and municipality that sells bonds). No one of them has much clout. But in structured finance, a handful of banks return again and again, paying much bigger fees. A deal the size of XYZ can bring Moody’s $200,000 and more for complicated deals. And the banks pay only if Moody’s delivers the desired rating. Tom McGuire, the Jesuit theologian who ran Moody’s through the mid-’90s, says this arrangement is unhealthy. If Moody’s and a client bank don’t see eye to eye, the bank can either tweak the numbers or try its luck with a competitor like S.&P., a process known as “ratings shopping.”...

[T]he analyst forecast losses for XYZ at only 4.9 percent of the underlying mortgage pool. Since even the lowest-rated bonds in XYZ would be covered up to a loss level of 7.25 percent, the bonds seemed safe. XYZ now became the responsibility of a Moody’s team that monitors securities and changes the ratings if need be.... [A] sliver of folks in XYZ fell behind within 90 days of signing their papers. After six months, an alarming 6 percent of the mortgages were seriously delinquent.... By the spring of 2007, 13 percent of Subprime XYZ was delinquent — and it was worsening by the month.... Poring over the data, Moody’s discovered that the size of people’s first mortgages was no longer a good predictor of whether they would default; rather, it was the size of their first and second loans — that is, their total debt — combined... credit scores, long a mainstay of its analyses, had not proved to be a “strong predictor” of defaults this time.... Amy Tobey, leader of the team that monitored XYZ, told me, “It seems there was a shift in mentality; people are treating homes as investment assets.” Indeed. And homeowners without equity were making what economists call a rational choice; they were abandoning properties rather than make payments on them.... By early this year, when I met with Moody’s, an astonishing 27 percent of the mortgage holders in Subprime XYZ were delinquent. Losses on the pool were now estimated at 14 percent to 16 percent — three times the original estimate. Seemingly high-quality bonds rated A3 by Moody’s had been downgraded five notches to Ba2....

Many of the lower-rated bonds issued by XYZ, and by mortgage pools like it, were purchased by C.D.O.’s, the second-order mortgage vehicles, which were eager to buy lower-rated mortgage paper because it paid a higher yield.... In 2006 and 2007, the banks created more than $200 billion of C.D.O.’s backed by lower-rated mortgage paper. Moody’s assigned a different team to rate C.D.O.’s. This team knew far less about the underlying mortgages than did the committee that evaluated Subprime XYZ.... [A]gencies rate pools with assets that are perpetually shifting. They base their ratings on an extensive set of guidelines or covenants that limit the C.D.O. manager’s discretion.

Late in 2006, Moody’s rated a C.D.O. with $750 million worth of securities. The covenants, which act as a template, restricted the C.D.O. to, at most, an 80 percent exposure to subprime assets, and many other such conditions.... Moody’s rated three-quarters of this C.D.O.’s bonds triple-A.... Moody’s had underestimated the extent to which underwriting standards had weakened everywhere. When one mortgage bond failed, the odds were that others would, too. Moody’s estimated that this C.D.O. could potentially incur losses of 2 percent. It has since revised its estimate to 27 percent.... A triple-A layer of bonds has been downgraded 16 notches, all the way to B....

However, the conclusion is very, very weak indeed:

[I]f the Fed or other regulators wanted to restrict what sorts of bonds could be owned by banks, or by pension funds or by anyone else in need of protection, they would have to do it themselves — not farm the job out to Moody’s.... Moody’s itself... like S&P, embraces the idea that investors should not “rely” on ratings for buy-and-sell decisions. This leaves an awkward question, with respect to insanely complex structured securities: What can they rely on? The agencies seem utterly too involved to serve as a neutral arbiter, and the banks are sure to invent new and equally hard-to-assess vehicles in the future. Vickie Tillman, the executive vice president of S&P, told Congress last fall that in addition to the housing slump, “ahistorical behavorial modes” by homeowners were to blame for the wave of downgrades. She cited S&P’s data going back to the 1970s, as if consumers were at fault for not living up to the past. The real problem is that the agencies’ mathematical formulas look backward while life is lived forward. That is unlikely to change.

The problem is not that the rating agencies' formulas look backward while life is lived forward. Rating agencies' business is to look backward: to say that this bond looks, historically, like that class of bonds which have in the past defaulted at such-and-such a rate. The proper business of investment is to take the rating agencies' work as a guide and ask two of the three standard questions:

• What is there in the situation that could make things different this time--that could generate "ahistorical behavioral modes"?
• How much of the risk that things are different this time are we willing to bear?

And then there is the third standard question:

• If this is such a good deal for me to buy it at this price, why is the seller selling--why isn't it a equally good deal for the seller to hold on to it?

Without satisfactory answers to these three questions, you are not investing: you are gambling in Las Vegas and you are not the house--you are, indeed, on a blind date with destiny and she has just ordered the lobster.

Lowenstein's lament in his last paragraph is implicitly a demand that investors not only not have to have answers to these three standard questions, but that they not even be expected to ask them--that investors find somebody else to "rely" on in the cases of "insanely complex structured securities" and of "new and equally hard-to-assess vehicles in the future" that "banks are sure to invent," somebody else who can predict the future. That seems completely wrongheaded. If you truly do not want to ask the three standard questions and evaluate credit risk you should be in U.S. Treasuries (and even there you have to assess inflation risk and, unless you are planning to hold to maturity, monetary policy risk). And if you want higher yields than Treasuries offer--well, then you are back to asking your three standard questions again.

In my view, there seem to be six problems here, rather different than one would think from reading Lowenstein's last paragraphs:

• First, the problem of private organizations where the higher-ups let their CFOs and their subordinates game them by letting them say "it's triple A" when they didn't have answers to any of the three standard questions. This seems to be a problem for shareholders and trustees to deal with.
• Second, the problem of local government officials who were reaching for yield and saying "it's all in triple-A." This seems to be a problem for voters to deal with, or perhaps it's time to require that local governments use the Treasury as some form of fiscal agent.
• Third, the problem of highly-leveraged large financial firms where the traders and portfolio managers asked the three standard questions with respect to CDOs and went ahead otherwise--either because they misjudged the risks, because they correctly concluded that the risks were acceptable but this time got caught, or because they correctly assessed the risks but indulged in moral hazard while thinking that they would have collected their bonuses and moved on by the time the thing blew up. This is a much more serious problem--but it has little to do with the rating agencies.
• Fourth, the problem of the systemic risk--the fact that millions of jobs are now at risk--because of the fallout from problem three. The Fed and the Treasury are now trying to deal with this problem.
• Fifth, the problem that a good many people who could afford to pay their mortgages and stay in their homes will not be able to do so because of crisis risk and default premia loaded onto their mortgage payments--a problem that Frank and Dodd are trying to deal with, that Obama and Clinton would deal with, and that McCain is trying to ignore.

And:

• Sixth, the problem that bond salesmen throughout the world have been deflecting customers' attempts to ask for answers to the three standard questions by saying, "it's triple A! And it offers you a higher yield than other triple A!" and that the rating agencies have been too eager to increase their business by making the fine print in their descriptions of what they do even finer.

Lowenstein's article is written as if problem six is in some sense the problem. And this seems to me to be badly misleading.

Comments

[Patrick Burke]

Not only has destiny ordered the lobster, but her cell phone is ringing and it's her "friend" Paulo on the phone and he's wondering if she's free tonight.

[Neal]

There was no benefit to the rating agencies to look harder--they were paid for their services in a competition with other raters. The rater who provided the highest rating got paid for the rating. Close inspection or rudimentary analysis were detrimental to the raters income.

Besides, what did they have to lose anyway (other than reputation).

Whodathunk that total debt would be a good predictor of how likely it was that a borrower would default? After all, doesn't everyone have a separate money tree for those pesky other debts.

A real surprise to everyone was that borrowers were less likely to let their credit cards default than their home mortgage.

The default rates, even at 13% would not be fatal, except for that pesky concept of leverage--25 times leverage meant a 4% default dropped the value to zero.

But whodathunk that house prices at 5+ times anuual income were unaffordable.

The main problem was that almost all of the parties in the chain of construction, selling, loan origination, rating, packaging and bond selling got their money up front. Who gave a damn about two months from now? There was no-one in that line who had any intention of sweating out the possibility that the borrower might default.

The ultimate dummies were the bond holders, deceived by the rating agencies that were shocked(!!) that total debt was a predictor of default.

[Emma Anne]

I am very uncomfortable with this analysis. I suspect that investors did ask your three questions and thought they knew the answers. If they had been willing and able to do the kind of analysis that would have flushed out the real answers, then they would have been doing far more than the ratings agencies, whose *job* it was. Here's what the investor saw at the time.

(Q1) What is there in the situation that could make things different this time--that could generate "ahistorical behavioral modes"?

(A1 - investor) I don't see anything in particular. Everyone is telling me that FICO scores are a good predictor of whether loans will be paid back. And house prices never go down all over the country, so pooling mortgages should cover whatever risks I am not seeing

(A1 - ratings agency) I look at a lot of these things and I know that a lot more loans have no money down than traditionally. Also FICO scores have proved reliable for consumer debt - they aren't really tested against mortgages. Furthermore, the huge increases in house prices all over the country increases the risk that they will, in fact drop all over the country.

(Q2) How much of the risk that things are different this time are we willing to bear?

(A2 - investors) Moody's tell me again, these things are pooled to reduce risk right? And they are rated AAA? Sounds like it has been accounted for.

(A2 - ratings agency) Hmm, we are in completely uncharted waters here. Mutliple houses are being bought by the same person with no money down. Flipping is rampant. Maybe this is riskier than we are telling people.

(Q3) If this is such a good deal for me to buy it at this price, why is the seller selling--why isn't it a equally good deal for the seller to hold on to it?

(A - both) This one is easy, and Brad answered right in his post. Banks sell these so that they have more capital to make more loans. That is what banks do. Doesn't seem at all surprising to me.

Seriously, if investors are left to their own devices I don't think that is going to benefit anyone. Their has to be transparency to make the system work, just like buying stocks. We don't tell a stock buyer that she personally has to go take inventory at a warehouse. She feels that she can rely on the reports she gets and the analysis the experts provide regarding the details that she clearly can't confirm herself. How can we tell a municipality that they have to investigate individual loans?

[Ponzi Q. Globalization]

The proper taxonomy would place this 'blowup' under "Swindlilus Pyramidus". Worthless paper begetting more worthless paper.

In these types of swindles, lotsa money is being made upfront, no one is getting hurt, and so few complain. This lack of complaint doesn't mean that the securities were not known to be essentially worthless by the guys and gals who got in at the beginning and made the really big bucks (and who some say have been punished by being let go with tens of millions of dollars in retirement packages -- what a pathetic joke). It means that some made money in full knowledge that many would be hurt when the swindle was inevitably exposed.

Economists and their close kin, Homo Sapiens, should really try to see reality clearly and stop pretending that Wall Street is not run by a bunch of serial swindlers. Swindlers who now have far too much influence in the the way the world is run.

[Anon]

Actually, there's a fourth question:

4) If this is AAA debt, why does it pay more interest than government securities? If this doesn't pay more interest than government securities, why shouldn't I buy them instead?

[Matt]

Exlpain, not being sarcastic, how it is that a PHD in economics can go to work at a monopoly ratings industry and not expect things to blow up? It points to the idea that environment is everything, or, that we know more information than than our acts would suggest.

[Cranky Observer]

Can't say I agree with your police work there Professor DeLong. In a different field of human experience, the ASME Boiler & Pressure Vessel Code and what was once one of the world's most reputable entities The Hartford Steam Boiler Inspection Company exist exactly so that every widget factory owner, apartment building super, and shopping mall manager does NOT have to be an expert in whether or not the boiler on his property will blow up. Because experience from 1850-1900 showed that when such onus was placed on people who needed a steam boiler but where that boiler was not the primary purpose of the facility that it was only a matter of how many would be killed when, not if, that boiler exploded. Heck, I worked for an entity that was packed from top to bottom with steam boiler experts and we were still required by internal procedure, the financial markets, and common sense to call in HSBI to audit our assessments and procedures.

It simply isn't possible for any organization to operate on the assumption that it is filled with The Smartest Guys in the Room. And that is doubly true on Wall Street where the other entity's Smart Guys may be deliberately trying to deceive you. If the standard of care required for any financial transaction is complete reanalysis from first principles, and no auditing agency can be trusted to be disinterested or close to correct, then perhaps it is time for some very heavy handed State regulation and control of those markets and some vastly reduced salaries and bonuses in that industry.

Cranky

[Alex]

Strange, if you read The Smartest Guys in the Room, you'll know that the rating agencies never actually added up all the Enron fancydan debt when rating the company itself - even though they certainly knew it was there, because...they rated it.

[Alex]

Strange, if you read The Smartest Guys in the Room, you'll know that the rating agencies never actually added up all the Enron fancydan debt when rating the company itself - even though they certainly knew it was there, because...they rated it.

[richard]

The whole idea of ratings agencies is that you don't have to do your own diligence. The idea seems flawed.

Lots of people made money by realizing the rating agencies didn't have a clue.

[Ian Whitchurch]

I know that some suckers, erm I mean *buyers* just went 'It's not like that subprime junk. It's AAA rated'.

I know this because I spoke to one of my girlfriend's relatives about this at a party ... he is the finance officer at the council of one of NSW's coastal cities, and that was his line.

I was too polite to say 'Wow man, you actually *believe* that shit from Moody's and S and Pees ? Where were you when Enron went down ?'

[MattF]

Another problem is just plain old dishonesty from just about all parties in recent years. Imaginary assets, imaginary incomes, imaginary debts... about a trillion or so dollars worth.

[pickledshark]

And this is why we are where we are. From the point of view of the bond purchasers ask yourself if the ratings agencies models were wrong about securitized mortgages maybe they are wrong about securitized student loans or securitized commercial loans. Why buy now if the risk is underestimated? In my opinion some kind of shakeup of the ratings agencies is in order, if I was a bondholder or one of the monoline insurers facing bankruptcy on account of wildly innacurate ratings i would be calling for the head of moodys and fitch and S&P. Lowensteins a good writer, i think Brads criticism is unwarranted.

[Bernard Yomtov]

"Sixth, the problem that bond salesmen throughout the world have been deflecting customers' attempts to ask for answers to the three standard questions by saying, "it's triple A! And it offers you a higher yield than other triple A!" "

An investor who accepts this answer would not know what to make of the answers to the three questions if he got them.

[Cranky Observer]

> I know that some suckers, erm I mean
> *buyers* just went 'It's not like that
> subprime junk. It's AAA rated'.

If you are seriously saying that municipal investment managers cannot and should not count on an investment-quality rating from a leading Wall Street rating agency having some resemblance to reality, then I am not sure how you think the financial markets as we have known them since 1970 will survive. Certainly such false information blows away every fundamental theory of microeconomics and finance as they claim to apply to the real world.

Cranky

[John Drake]

After Orange County went bankrupt investors should have been suspicious of the ratings agencies. After all they gave OC a top rating when their portfolio was loaded with reverse repros. A bomb just waiting for detonate when interest rates went up. Moreover everyone was duly warned by John Moorlach who ran against Robert Citroen for county treasurer. Unfortunately the voters and pretty much everyone else didn't take Moorlach's intelligent and timely warnings seriously and reelected Citroen. After the debacle the ratings agencies said something to the effect: "We rate credit risk, not market risk." Lame excuse. These agencies have a big conflict of interest and no one has done anything about it.

[John Drake]

"If this is such a good deal for me to buy it at this price, why is the seller selling--why isn't it a equally good deal for the seller to hold on to it?"

Exactly. I bring this question up frequently when someone offers me a "deal" like some supposedly wonderful real estate investment. Of course there are reasonable answers. The seller might need the money for some reason, to meet a margin call for example. Or the seller might have access to a superior alternative investment not available to the buyer. The seller might be re balancing his portfolio, or has just decided to take on more or less risk for some reason. Nevertheless it's a great question that always needs to get asked.

[Sean Peters]

Apparently, in Brad DeLong-world, I and millions of other non-finance experts are doomed to invest in nothing but Treasuries, because we're not, you know, finance experts. I (and the aforementioned millions of others) have no way of answering the "three standard questions". So given the choice between investing in nothing but Treasuries and some federal agency ensuring that I get reliable information about the risks/rewards of various securities, guess which one I'm going to pick.

[Matthew Austern]

I'll go along with the claim that the job of a rating industry is "to say that this bond looks, historically, like that class of bonds which have in the past defaulted at such-and-such a rate." And in retrospect it's clear that the rating agencies were not doing their job. They should have said: this bond does not look like anything we have historical data on, so we are unable to estimate the rate at which bonds like it will default.

[Ian Whitchurch]

Cranky,

In a word, yes.

In fact, I'll go harder. You have a duty of care not to trust what they guys the vendor paid to assess the property said it was worth.

And if you have that much money on the table, then you hire *your* expert to assess them.

If you wouldnt buy a house on the vendor's assessed value, then dont buy debt that the vendor paid a third party to assess.

Ian Whitchurch

[Steve J.]

HERE'S ANOTHER PART OF THE PUZZLE

Inquiry Focuses on Withholding of Data on Loans
By VIKAS BAJAJ and JENNY ANDERSON
Published: January 12, 2008
NY Times

To vet mortgages, Wall Street underwriters hired outside due diligence firms to scrutinize loan documents for exceptions, errors and violations of lending laws. But Jay H. Meadows, the chief executive of Rapid Reporting, a firm based in Fort Worth that verifies borrowers’ incomes for mortgage companies, said lenders and investment banks routinely ignored concerns raised by these consultants.
“Common sense was sacrificed on the altar of materialism,” Mr. Meadows said. “We stopped checking.”

And as mortgage lending boomed, many due diligence firms scaled back their checks at Wall Street’s behest. By 2005 , the firms were evaluating as few as 5 percent of loans in mortgage pools they were buying, down from as much as 30 percent at the start of the decade, according to Kathleen Tillwitz, a senior vice president at DBRS, a credit-rating firm that has not been subpoenaed. These firms charged Wall Street banks about $350 to evaluate a loan, so sampling fewer loans cost less.

[Ginger Yellow]

Apparently, in Brad DeLong-world, I and millions of other non-finance experts are doomed to invest in nothing but Treasuries, because we're not, you know, finance experts. I (and the aforementioned millions of others) have no way of answering the "three standard questions". So given the choice between investing in nothing but Treasuries and some federal agency ensuring that I get reliable information about the risks/rewards of various securities, guess which one I'm going to pick.

That's quite a false dichotomy you've got going there, but even if you accept it, your argument doesn't hold up. The whole point is that there was no reliable information, because these new products (and laxer underwriting on existing products) didn't perform anything like their predecessors. So the choice (in the false dichotomy) is investing in nothing but Treasuries and having no reliable information at all.

I think Brad's analysis is more or less right, and I'll try to explain why.

"How can we tell a municipality that they have to investigate individual loans? "

I talk to investors all the time, and the better ones, who have avoided the worst of this financial crisis, did exactly that. They demanded loan tapes from originators/arrangers and refused to invest if they didn't receive them. When they received them, they did large amounts of number crunching and scenario analyses to see what could happen to the cashflows under different assumptions. That's what you have to do with structured products. If you can't do that, invest in corporate bonds or something else. It's really very simple - if you don't know how your investment is likely to perform, don't invest in it. Even in an ideal world, a one-dimensional credit rating doesn't give you anything like the information you need to make an investment decision in a structured product. It tells you nothing about liquidity, price volatility, extension risk, prepayment risk, or default timing.

This isn't to say that ratings don't have a place for less sophisticated investors or in the regulatory framework, or that the agencies didn't screw up badly. But at the end of the day, structured products perform very differently to corporate bonds and require more analysis than credit risk alone, and investment decisions need to take that into account. There's a reason retail investors aren't allowed to buy structured products in the UK.

" In my opinion some kind of shakeup of the ratings agencies is in order, if I was a bondholder or one of the monoline insurers facing bankruptcy on account of wildly innacurate ratings i would be calling for the head of moodys and fitch and S&P"

Monolines do their own intensive credit work, or at least they claim to. They have only themselves to blame. People seem to forget that pretty much everyone called subprime wrong - obviously there were a few noble exceptions. Plenty of people who weren't relying on the agencies made bad assumptions of their own and paid the price.

[James Wimberley]

"[I]n structured finance, the agencies face pressures that did not exist when John Moody was rating railroads."
Of course nice Mr. Fitch and honest Mr. Jay and upright Mr. Vanderbilt would never have dreamt of trying to pull the wool over the first Mr. Moody's eyes.
Part of the problem is surely a loss of guts, bottle, nerve.

[lgm]

"I know that some suckers, erm I mean *buyers* just went 'It's not like that subprime junk. It's AAA rated'."

People have raised the question: why did buyers accept suspicious ratings?

Part of the answer is competitive pressure among institutional investors. A private investor chooses a money market fund based on small return differentials. The mutual fund manager who refused to touch subprime AAA instruments (and their slightly higher returns) would have lost his/her customers.

[lgm]

"I know that some suckers, erm I mean *buyers* just went 'It's not like that subprime junk. It's AAA rated'."

People have raised the question: why did buyers accept suspicious ratings?

Part of the answer is competitive pressure among institutional investors. A private investor chooses a money market fund based on small return differentials. The mutual fund manager who refused to touch subprime AAA instruments (and their slightly higher returns) would have lost his/her customers.

[polo]

Interesting to me is that two quotes from Moodys people suggest they were falling down on the job:

1:“Our expertise is as statisticians on an aggregate basis. We want to know, of 1,000 individuals, based on historical performance, what percent will pay their loans?”

2: "It seems there was a shift in mentality; people are treating homes as investment assets.”

For the first, isn't it their job to look beyond the immediate experience and assess broader market conditions over a long period? Decades ago there were few or no FICA scores but there were real estate bubbles. It seems like Moodys stuck to "historical performance" that had no relation to the overall macroeconomic situation: extremely low interest rates, high personal leveraging along with higher home "ownership", and all sorts of other characteristics of the "new economy" that would have required a closer look.

On the second, by 2006 it was pretty clear, I would say extremely clear, that "people (we)re treating homes as investment assets". Isn't Moodys, in assessing default risk, supposed to know the market and examine overall risk, beyond a mathematical model of default risks that is not updated as the market hyperventilates?

It seems to me like Moodys got away with saying, "There hasn't been a bubble bursting in the last five years, so we don't have to factor that risk in."

[Dave]

I work for a local government that is required to hold nothing but AAA.

We are not required to audit the ratings agencies.

Still, when it became obvious that the ratings for mortgage-backed securities were fishy, we sold all of them: Good ones and bad ones alike. Now, people are trying to blame us for panic selling, because so many small investors got trampled in our stampede.

But if the ratings agencies had been honest or competent (either one would do), we could have been more selective. But when the trail boss keeps pouring A-1 steak sauce on your back, sooner or later your stampede starts to seem like a pretty rational decision.

[Doctor Jay]

Well, I can have a good credit rating on Monday and file for bankruptcy on Thursday. I have the good rating because I pay my bills on time, and I file bankruptcy because I incur some unexpected debt, like medical bills or something.

I think that's the real crack in this system. We're in this problem as deep as we are because the market moved in a way it hasn't moved in 20-30 years, at least, maybe 50. So all the quants calculate their data, and estimate variability of prices using data from the last 50 years and don't see it coming.

This is the same criticism of rating agencies, sellers and buyers. We haven't seen home prices go down in 50 years, so it was assigned a very low probability.

This is the problem I have with all the Black-Scholes-based quantitative analysis. Why should I believe that a backward looking estimate of the variance should continue to hold in the future? We aren't talking about physical phenomena here, we're talking about financial markets.

Not that I think quant investing is all bad. I just wish that I saw some respect for the unknown, some sense that even with all the rocket science, there's still stuff we don't know, and shit that can happen. That doesn't mean bury your money in the backyard, just leave a bigger reserve.

If Moody's had simply decided to increase the required cash component by 10 percent to get that AAA on a securitized mortgage derivative, how much better off would we be? And the 10 percent would be there just because we don't have years of experience with these sorts of vehicles.

However, it seems these days that if you don't have a formula, nobody pays attention. You're just an old fart, with stone-age thinking. And you're in the way. That's where the nerve comes in. And where being paid by the people you are rating becomes a big problem.

It also demonstrates the difference between the mentality of "be number one" and "live long and prosper".

[Barry]

Kevin Drum has some comments here (http://www.washingtonmonthly.com/archives/individual/2008_04/013609.php,
and the next post). The nutgraph:
"There are a very limited number of extremely smart people in Wall Street firms creating all these complex new financial instruments, and they're very highly motivated to make them as impenetrable as possible. The average pension fund manager or county treasurer is simply never going to be able to analyze them in any serious kind of way. It's easy to say they should, but that's like saying that our schools would be better off if every schoolteacher had a PhD. Given the current state of the art in human nature, it's just not going to happen.

So in reality, there's a large class of investors who have little choice but to trust the rating agencies. And if rating agencies have a fundamental financial interest in colluding with their clients, then that collusion is almost certain to happen."

This is what I agree with, and I propose Barry's Rule: if a party has a conflict of interest, and acts in a compromised fashion, then that is the first explanation.

[Ginger Yellow]

"The average pension fund manager or county treasurer is simply never going to be able to analyze them in any serious kind of way. It's easy to say they should, but that's like saying that our schools would be better off if every schoolteacher had a PhD. Given the current state of the art in human nature, it's just not going to happen.

So in reality, there's a large class of investors who have little choice but to trust the rating agencies."

No, they do have a choice. If they don't understand the product, they don't have to invest. It's really that simple. If bankers can't find buyers for your "impenetrable" product, they have no incentive to create it.

[low-tech cyclist]

"Apparently, in Brad DeLong-world, I and millions of other non-finance experts are doomed to invest in nothing but Treasuries, because we're not, you know, finance experts."

Index funds are also good, Sean.

Seriously, you CAN invest other things (from individual stocks and bonds to much more complex and risky derivatives), but chances are excellent that the other people playing the game know more than you do, and they'll take your money home with them.

At least with plain old stocks and bonds, the risks are pretty clear. You understand the game itself, and you have at least a decent chance of putting your money on the right player.

It's when you don't even understand **the game itself** that you should absolutely, positively keep your money in your wallet. If the nature of the game is beyond your comprehension, then you're crapshooting in the dark: you may not even know you've rolled snake-eyes until it's too late, because you may not know what snake-eyes even looks like. You can't judge the risks.

[zeno2vonnegut]

Two systemic problems are ignored. There's the old amusement called "Gossip" where a chain of people whispers to their right hand neighbor what they heard from their left hand neighbor. Information is steadily lost. Second, moral hazard is introduced at each stage of the transmutation of a home mortgage into a high level derivative. Our ancestors financed through a highly regulated S&L which either held the mortgage or sold it to a highly regulated government corporation. The saying, "safe as houses" was a truism. There has been no or minimal benefit and large detriment to society by replacing this system with one that presented profits to the richest sliver of society. Isn't the simple answer to return to the system proven safe and to forbid the harmful system?

[Maynard Handley]

Just to add to those complaining about Brad's analysis, how is this different from the claim, justly panned by Brad, that "medical consumers" should shop around for the best deals (which of course implicitly means taking off a huge chunk of time to learn vast swathes of modern medicine)?

A large part of the efficiency of modern society is specialization, but that specialization isn't going to function if nitwit economists' answer to every agency problem is that the humans involved have to look out for themselves. Who, exactly, is going to be doing any work if I and everyone else is spending all my time learning finance (so that investment banks won't cheat me) and medicine (so that hospitals won't cheat me) and auto repair (so mechanics won't cheat me) and food science (so that I can trust what I buy in the supermarket) etc etc etc.

This libertarian view of things whereby we rid ourselves of the FDA and similar oversight bodies is, quit simply, insane, and it's insane of Brad to support the idea in this little space of financial oversight.

[Bram Cohen]

You're being too charitable to the ratings agencies. The question is not whether they made their fine print finer, but whether their ratings ever had any meaning at all. The underlying mortgage is usually mostly rated based on the credit rating of the buyer, and the idea that credit ratings have some scientific basis is more than a little bit absurd.

Your comments about there being other problems are correct, but even without understanding the underlying theoretical problems, even a journalist can notice that the layer upon layer of resellers are each losing and obfuscating a bunch of data about the underlying asset, and then selling it upstream for more, as if their information lossage somehow produced value.

Seriously, some of the structural aspects border on farce. Why were only three-quarters ARM? Why weren't all the ARMs bundled into something separate from the fixed rates? Why weren't first and second mortgages on the same property bundled together?

The purpose of that lead, by the way, is to set up a narrative structure. The author is telling a story, not writing a class paper on finance.

[Cranky Observer]

> No, they do have a choice. If they don't understand
> the product, they don't have to invest. It's really
> that simple. If bankers can't find buyers for your
> "impenetrable" product, they have no incentive to create it.

I am sure this line of thought is in no way related to the chart in Prof. DeLong's next post. No, there couldn't possibly be any cause-and-effect there...

Cranky

Again, if you live in a region with hot water heat are you qualified to inspect your boiler and furnace? Have you run the design calculations on your boiler through from scratch and performed assays on test coupons in your own metallurgical lab using reagents you purified yourself? Why not?

[Ginger Yellow]

"Who, exactly, is going to be doing any work if I and everyone else is spending all my time learning finance (so that investment banks won't cheat me) and medicine (so that hospitals won't cheat me) and auto repair (so mechanics won't cheat me) and food science (so that I can trust what I buy in the supermarket) etc etc etc."

Again, is it really too much to ask that professional investors responsible for other people's money spend their time learning finance?

[Ginger Yellow]

"Again, if you live in a region with hot water heat are you qualified to inspect your boiler and furnace?"

No, but if I'm a professional plumber, then I am. That's kind of the point.

[Achilleas]

Excellent comments about the rating agencies. After all they are called NRSRO organisations, to indicate that they perform statistical work in order to assign ratings and not predictive (otherwise they would be called NRPRO). But seriously, if the rating agencies told investors to jump out of a window, would they do it? After all they even say that the ratings are just a factor to help investment decisions and not a suggestion to purchase a specific security.

If the job of asset managers is to look at 3-4 letters assigned to a security and also the yield on this security and make an investment decision then why are they paid so well? Even a high school drop out could do that job for minimum wage. The excuse "they (the rating agencies) made me buy this security, its not my fault" is not credible for "professional" investors.

In any case the rating agencies made the mistake to assume that relationships between economic variables will hold for many years and that any shifts will be gradual and that they will have time to adjust their methodologies accordingly (mostly based on the feedback from their rating surveillance teams for each asset class). However, slowly but steadily some of the economic variables started to change from exogenous to endogenous threatening the statistical power of their models. Also, i know that in europe they get loan level data, but i think that in the US they haven't done that yet.

In any case, whatever the weekness of the rating agencies this cannot explain the entire bubble. I think the main reason for the mess that the US economy is in today is the lack of appropriate regulations for the main banking institutions. With earnings approaching hedge fund returns it looked that there was something wrong with the banking system over the last couple of years and now we all know what was going on.

[Ginger Yellow]

"Also, i know that in europe they get loan level data..."

This is far from universally true, although the situation has improved a lot in recent years. In Germany for instance, banking laws prohibit disclosure of loan level data to third parties.

[Cranky Observer]

> No, but if I'm a professional plumber, then I am.
> That's kind of the point.

Um, _you_ are the fiduciary party. And oddly enough you are depending on certified experts at one or two[1] removes, not doing the work yourself. _That's_ kind of the point.

Cranky

[1] Very few plumbers are certified HSBICo inspectors - those inspections are done at the factory.

[Ginger Yellow]

What in God's name are you talking about, Cranky? Seriously, I don't understand your point. Retail investors are not allowed to buy structured products directly, precisely because they're too complicated for the likes of us to grasp all the risks. Only qualified institutional investors are allowed to buy them. As Achilleas says, these people and institutions are paid to do this sort of analysis.

[Bram Cohen]

No money down mortgages, contrary to what some commenters here seem to imply, are not particularly exotic instruments, and bundling them together isn't particularly hard to analyze. If housing prices go down, they completely go into the tank. (Some of the details actually are hard to analyze, but its overall shape is hard to miss).

What we had before was a situation where no distinction was made between a 40% down mortgage for a person with bad credit but no second mortgage and a 10% down mortgage for a person with great credit who also got a second mortgage. For the former it's completely reasonable to treat the likelihood of default as an individual problem and reduce risk by bundling. For the latter all bundling does is obfuscate the problem. You need to either hedge against housing prices or accept that you're going long on real estate.

The question 'How much would your asset go down if the housing market dropped by X%?' should be front and center the big question that any mortgage-backed securities vendor or customer should have on their minds. The fact that this is considered an obscure and technical line of reasoning exposes an astounding pervasive lack of competence.

[Ginger Yellow]

"What we had before was a situation where no distinction was made between a 40% down mortgage for a person with bad credit but no second mortgage and a 10% down mortgage for a person with great credit who also got a second mortgage."

This isn't the case. The second mortgage bit seems to have been true, but the loan to value ratio of a mortgage is second only to borrower quality as a determinant of the credit enhancement assigned to it in the rating process.

If you want to understand the assumptions that led the rating agencies astray, read this passage from Moody's rating methodology for RMBS from the late 90s:

Borrower equity is an important buffer against default risk and a cushion against loss where a default occurs. In a distress situation (death, divorce, or unemployment), a homeowner with a large equity stake will typically sell the home (and pay off the loan) rather than face loss through a foreclosure proceeding. But as we
discuss below, negative homeowner equity may not automatically portend default where borrower capacity and credit measures indicate a willingness to preserve a strong credit record. The loan-to-value (LTV) ratio
has less impact than it did in our previous approach, but it is an indicator of remaining equity and contributes to estimates of default frequency and default severity.
At best, the LTV ratio captures the initial effective borrower equity in a property. But equity can grow through time as a result of scheduled amortization, partial prepayment of principal, and any appreciation in home
price. Nevertheless, an equity cushion helps protect lenders from housing market declines and extensive carrying costs during foreclosure. Among the components of such costs are past due principal and interest,
legal fees, taxes, and selling costs.
The continued popularity of second mortgages highlights the need to consider all mortgage related debt when calculating an LTV ratio. The combined loan-to-value (CLTV) ratio, calculated as the sum of all mortgage debt divided by home value, gives a better measure of total borrower equity and is the preferred indicator of default frequency. Unfortunately, lenders normally cannot restrict subsequent second mortgages, making it difficult to track CLTV on a loan-by-loan basis and reducing its predictive power.
In some cases, such as a refinancing, the reported CLTV may not reflect the true resale value of the pledged property if the current home price was not independently established. Even if there is an appraisal, standards
vary considerably among originators. In the worst cases, appraisals are made on a “drive by” basis. Each of these factors has led us to reduce the weight we place on CLTV as a risk indicator.
Our view of the role of housing plays a part in our decision to downplay the importance of LTV as a risk indicator. An LTV-dominated model implicitly treats a home as a speculative asset. The borrower has an “option” to walk away from the home if its value falls below a “strike price,” represented by the balance of the mortgage loan. In our revised approach, we view a home as providing a “housing service” to the homeowner and his/her family. The cost of housing service depends on many variables, including the opportunity cost of renting. As such, we now place less reliance on LTV as the primary determinant of credit risk.

As you can see, they massively overestimated the value to borrowers of staying in a negative equity home, and hugely overestimated the extent to which borrowers viewed the purchased property as providing a housing service. It was the failure to update these assumptions in line with the changing marketplace that caused them to massively misrate subprime RMBS, not the idea that they didn't expect house prices to fall. Believe it or not, rating agencies do stress deals for double-digit house price declines. What they didn't stress them for was double digit house price declines coupled with delinquencies and defaults way above the historical precedents (because, of course, there were no direct precedents for these products).

"You need to either hedge against housing prices or accept that you're going long on real estate."

I don't know a single RMBS investor who doesn't accept they're going long on real estate (apart from the ones who are shorting, of course). That's the whole point of investing in the asset class. If you don't want to go long real estate, buy something else like auto loan ABS or corporate bonds.

"Why were only three-quarters ARM? Why weren't all the ARMs bundled into something separate from the fixed rates? Why weren't first and second mortgages on the same property bundled together?"

Your first question makes sense from an investor perspective - homogeneous deals are easier to analyse. But the simple answer is that there was no real price incentive for sellers to separate out the loans - investors scarcely penalised mixed pools at all - and combining them together allowed them to get more funding. Second mortgages weren't bundled together because they were usually provided by someone else.

[Ginger Yellow]

Bram: "If someone has an ARM and the price of their underlying asset goes down and interest rates go up, they frequently can't make their payments, period. This business of viewing credit scores as an ultimate judge of a person's moral character, and ignoring, you know, the fundamentals of how much the person makes and how much their underlying collateral is worth, is just nuts."

Again, it's always worth checking what the methodologies actually said/say. 95 times out of 100, when somebody says "I can't believe the agencies didn't think this would happen", they did, but they didn't give it the appropriate weight or considered the risk sufficiently remote at a given rating level. From the same methodology (written well before ARMs became widespread):
"As mentioned above, our traditional classifications by loan type (fixed-rate, ARM, etc.) appeared to overlap measures of borrower quality because of the types of programs historically offered. However, we recognize that adjustable-rate mortgage borrowers may face significantly higher monthly payments under certain interest rate scenarios. This exposure to payment shock could lead to a higher default frequency for ARM borrowers than has been observed, but to date, there have been only limited periods of interest rate increases since ARMs have emerged as a product type. Moreover, some ARM borrowers are qualified at so-called teaser rates, reaching two or more percentage points below the stated fully indexed rate (the contract index rate plus a margin), exposing the borrower to payment shock in a stable or rising interest rate environment. Raising the ARM rate on a $100,000, 30-year mortgage from 6% to 8% boosts the monthly payment from $599 to $734. For ARMs, we adjust upward the default frequency estimate in order to compensate for the added risk of payment shock."

Bram: "If credit scores were computed seriously using back-running of models, there would be a moody's credit score of individuals rather than an experian. But we don't, what we have instead is a number treated reverentially with no questioning of the highly dubious method of calculating it."

There's a really interesting feature buried in this point, but I'm not sure I'm the person to ask it. Here in the UK we don't have a FICO equivalent, so assessment of borrower quality is done based on a subjective analysis of data like county court judgements, the borrower's past record with the same lender (if any), employment status and so on. The agencies publish the weightings they ascribe to different factors, and investors can judge the models based on that. There's no enshrining of a single number as an infallible metric. I wonder if that played a part in ensuring (so far) that the UK has experienced a relatively soft landing to its housing bubble, despite having the biggest subprime mortgage market outside the US and suffering the same mortgage funding problems. Obviously the fact that the mortgage origination chain is much more tightly regulated has played a part too, but it's an interesting thought nonetheless.

[Ginger Yellow]

I should also add that Moody's updated its methodology for hybrid ARMs in 2005 to take into account product characteristics and borrower behaviour, although it obviously didn't work. I can dig through the other agencies' methodologies if you like.

[Bram Cohen]

Thanks for the interesting info Ginger. No need to dig up further information for me, I'm just an curious homeowner.

This section really stands out -

"we recognize that adjustable-rate mortgage borrowers may face significantly higher monthly payments under certain interest rate scenarios. This exposure to payment shock could lead to a higher default frequency for ARM borrowers than has been observed, but to date, there have been only limited periods of interest rate increases since ARMs have emerged as a product type."

This goes directly to Brad's question #1 - if you're worried that something without historical precedent might happen, and your ratings agency says that's a possibility in so many words, that would seem to be a bit of a red flag.