« Sentences to ponder | Main | Concise Encyclopedia of Economics »
How did the credit rating agencies misfire?
A second view is that because the methodologies used for rating CDOs are complex, arbitrary, and opaque, they create opportunities for parties to create a ratings “arbitrage” opportunity without adding any actual value. It is difficult to test this view, too, although there are reasons to find it persuasive. Essentially, the argument is that once the rating agencies fix a given set of formulas and variables for rating CDOs, financial market participants will be able to find a set of fixed income assets that, when run through the relevant models, generate a CDO whose tranches are more valuable than the underlying assets. Such a result might be due to errors in rating the assets themselves (that is, the assets are cheap relative to their ratings), errors in calculating the relationship between those assets and the tranche payouts (that is, the correlation and expected payout of the assets appear to be higher and therefore support higher ratings of tranches), or errors in rating the individual CDO tranches (that is, the tranches receive a higher rating than they deserve, given the ratings of the underlying assets). These arguments are complex and subtle...
That is from a very interesting paper by Frank Partnoy. The paper is not always easy reading but so far it is the best piece on its topic I have found. This was another good section:
If the mathematical models have serious limitations, how could they support a $5 trillion market? Some experts have suggested that CDO structurers manipulate models and the underlying portfolio in order to generate the most attractive ratings profile for a CDO. For example, parties included the bonds of General Motors and Ford in CDOs before they were downgraded because they were cheap relative to their (then high) ratings.67 The primary reason that the downgrades of those companies had an unexpectedly large market impact was that they were held by so many CDOs.
Thus, with respect to structured finance, credit rating agencies have been functioning more like “gate openers” rather than gatekeepers. The agencies are engaged in a business, the rating of CDOs, which is radically different from the core business of other gatekeepers. No other gatekeeper has created a dysfunctional multi-trillion dollar market, built on its own errors and limitations.
There is also a good discussion of how the ratings agencies have claimed First Amendment protection for their activities, more or less successfully. p.96 offers some good policy conclusions.
Posted by Tyler Cowen on October 3, 2008 at 05:50 AM in Economics | Permalink
Comments
Almost all financial crises have resulted because of model error, and usually the major error involves not modeling rigorously enough the change in correlations as asset prices move from a rising market to a falling one (correlations increase dramatically in a falling market and while most new models correct for this, the ones that fail don't allow for enough of it).
Posted by: nelsonal at Oct 3, 2008 7:44:27 AM
The fault is in the models, assuming independence of failures. You can't multiply the probabilities down to zero the way they do.
Posted by: rhhardin at Oct 3, 2008 8:41:02 AM
Yup. Tulip mania was caused by faulty models. Panic of 1907; faulty models. South China Sea; faulty models.
Faulty models have contributed to most of the financial crises of the last 30 years, but are not the primary cause. That was, is and always will be human emotion. Crises are born out of manias. Manias are born out of the human desire to get rich quickly and easily. All the models have done is provide more ways for manias to grow.
Posted by: joe at Oct 3, 2008 9:55:56 AM
I plead almost complete ignorance on what exactly the ratings agencies do, but why can't purchasers perform their own due diligence before buying these CDOs? Is it that there is too much volume to do so without requiring a huge staff - i.e. is this just outsourcing the task?
This meltdown is a result of many technical issues (e.g. models not working, as pointed out), but I think it's more a testament of people following their own financial incentives:
1) CEOs pursuing win-win risky strategies: hitting a 'home-run' pays out handsomely and failure results in golden parachutes
2) Banks granting questionable loans as the repackaging and the servicing provide streams of income regardless of the loan performance
3) Mortgage brokers pushing applications of risky borrowers to the banks in pursuit of maximum commissions
4) Borrowers taking on insane mortgages in terms of loan amounts and terms under the false assumption that "real estate prices always go up" (at the prodding of Greenspan, banks, mortgage brokers, the NAR, and talking heads)
5) Ratings agencies issuing unwarranted favorable ratings as unfavorable ratings result in less patronage
and I'm sure there's more.
Posted by: meter at Oct 3, 2008 10:01:44 AM
That first paragraph is really strange. The last two thirds of it is correct: the problem wasn't that the models were "complex, arbitrary, and opaque", but that they were simplistic and open. No model will ever be perfect, but if you don't publish your model, it may be good enough. Once you make your model available to people who can make lots of money exacerbating its imperfections, your model is going to get a lot worse.
Posted by: dWj at Oct 3, 2008 10:53:50 AM
meter,
Purchasers usually do perform their due own dilligence both on the collateral side (prepays and defaults) and the security side (structure), but depending on the institution, might need a specific rating in order to meet their investment tolerances. Analysis is usually done for one particular bond in a deal (tranche), but since CDOs use cash flows from different tranches of different deals, it can get complicated. Just because it is complicated does not mean it can't be done, but I think that there is no way the ratings on the CDOs were justified.
Posted by: Patinator at Oct 3, 2008 11:12:02 AM
So part of the subprime crisis was that mortage-related securities underestimated the probability of a GM
default because they thought that a near-investment-grade rating was justified for a company with a net
book value several times the opposite of its market value?
LOL!!!! This just gets funnier and funnier! Gonna have to blog that. (My blog is getting lonely...)
Posted by: Person at Oct 3, 2008 11:21:07 AM
I don't understand what information is allowing the bond rating agencies to somehow rate these companies better than the market at large would without them. Could they be replaced by propositional betting markets on future failure?
Posted by: Nfate at Oct 3, 2008 11:22:07 AM
Patinator, agree - the ratings were flawed. So even though as you say the calculations are complicated (and I think I already knew that), is it ridiculous to suggest that institutions perform these calcs and weigh associated risks in-house?
I know there are charters that require the purchasing of securities that exceed designated rating floors - as you alluded to - but that system seems useless if the ratings can't be trusted.
Posted by: meter at Oct 3, 2008 11:30:12 AM
meter, no it is not ridiculous at all to suggest that. I am sure that the first wave of CDOs were scrutinized very heavily, but there is no way that the default assumptions, since they had a limited backwards looking data set of normal defaults, could have factored in what actually happened.
Everyone, rating agencies included, used some form of the same model and some assemblence of the same assumptions. If everyone is thinking the same way and that thinking is wrong, none of it works. The old garbage in, garbage out.
Posted by: Patinator at Oct 3, 2008 11:48:20 AM
step 1 : get anointed a Nationally Recognized Statistical Rating Agency to decide what bonds are investment grade and can be held by banks and other portfolios in satisfaction of covenants and capital ratios
step 2: find ways to game the ratings.
step 3: profit!
the only shock is how blatantly they did it. and that anyone is surprised that for-profit regulators get captured.
(not just CDO ratings, but especially bond insurer ratings. Bond insurers were of course huge customers and enablers of the CDO ratings scam. They were allowed to keep AAA ratings despite insane contingent liabilities that would never have been allowed if they were structured as actual liabilities supported by a corresponding asset in an identically risky portfolio )
Posted by: curmudgeonly troll at Oct 3, 2008 12:52:33 PM
1) rating agencies are a government-chartered oligopoly, and contracts (never mind models) are fixed upon these ratings edicts, regardless of integrity
2) how much integrity can a rating agency have it is paid solely by ISSUERS (and not investors)?
Posted by: caveat bettor at Oct 3, 2008 5:19:26 PM
I find it amusing that the author contests the "issuer financed" ratings and at the same time dislikes "unsolicited ratings". I agree that the agencies may be double-dipping but if you want the buyers to fund the analysis then the ratings will have to be unsolicited in order to be objective. I also doubt that the solicited issues have a significant amount of insider information taken into consideration.
As for the market not believing that the opinions of ratings agencies are just that, opinions, that is the market's fault. If law says those are just opinions; and the reality is that the issuers are funding those opinions, then the market (and, by extension the government) is being ignorant in relying so heavily on those opinions when making investment decisions. Its not like you HAVE to buy issues that meet these ratings but rather you cannot buy issues that fail to meet those ratings. In the end the solutions are not the complicated to enact or devise - its more a matter of political will.
Posted by: David J at Oct 5, 2008 1:26:53 PM
Guess I should have read further before my prior statement...I rehash some of what was explicitly said toward the end and in the addendum.
Anyway, two other points that I found interesting in the addendum: the role of the agencies in rating new issuers and new issues. I do see added value here and thus the agencies would still operate status-quo in those areas regardless of any move toward a market-based system.
Second, the regulatory approach adjustment would be, more clearly stated, to require issuers of regulatory affecting information (investment grade for buy/sell decisions) to explicitly give up their First Amendment protections for those ratings. And, ideally, require those ratings to be unsolicited; or alternatively generated from an accredited model that is available to all market participants and not subject to assumption changes but only to variable inputs coming from the market. Let the agencies publish written addendum to the ratings to add whatever value-added information they feel is warranted - and possibly offer an unofficial rating that cannot be used for institutional purposes but would still be of value to sophisticated investors.
Posted by: David J at Oct 5, 2008 2:24:40 PM
There is a risk of stripping first amendment protections from the credit rating agencies. The ratings agencies also rate municipal and government debt. Suppose that they decide to downgrade a municipality where a powerful Senator lives, or worse yet, downgrade the US credit rating from AAA to say, AA+? Pretty similar to a political reporter uncovering malfeasance at the DOD, no? After all, a credit rating is a reflection on the government's financial acumen. What happens then if the government threatens Moody's or S&P with endless subpoena to hand over documents and intimidates them in this way. Chilling effects and all.
As for payment and being prodded to issue ratings by the issuers of the securities, isn't this sort of like a gadget maker sending a scrubbed, pristine version of the gadget to CNET or NYT or PC World to get them to review it and give it a good rating? The gadget makers then also buy advertising space. Why don't we also start banning those practices? They happen every single day.
I don't think that the credit ratings agencies are pure and innocent here. I take serious issue with how they operate. But to strip them of these protections, I think, would be a mistake. If the NYT gets to claim privilege, so should Moody's.
Posted by: Gene2 at Oct 6, 2008 11:41:42 AM
I was so concerned about the ratings agencies in 2005 that I made them the subject of my admin law exam. I also blogged about the First Amendment issues here:
http://www.concurringopinions.com/archives/2007/08/from_first_amen.html
As I've said in my work on search engines (another black box rating system), this type of absolute First Amendment protection is quite dangerous--and threatens to become a new form of Lochnerism for the digital age.
Posted by: Frank Pasquale at Oct 6, 2008 10:36:58 PM
