Thursday, August 6, 2009

Rating Agency Reform ... continued

An excerpt from our colleague Mark Froeba's testimony before the U.S. Senate Committee on Banking, Housing, & Urban Affairs' hearing on Proposals to Enhance the Regulation of Credit Rating Agencies (August 5, 2009):

-----

"First, [the rating agencies] enjoyed an effective monopoly on the sale of credit opinions. Second, and more importantly, they enjoyed the benefit of very substantial government-sanctioned demand for their monopoly product. (A buggy whip monopoly is a lot more valuable if government safety regulations require one in every new car). Third, the agencies enjoyed nearly complete immunity from liability for injuries caused by their monopoly product. Fourth, worried about the monopoly power created by the regulations of one branch of government, another branch encouraged vigorous competition among the rating agencies. This mix of regulatory “carrots” and “sticks” in the period leading up to the subprime melt-down may have contributed to making it worse than it might have been. Thus, a third goal of rating agency reform should be to untangle these conflicting regulatory incentives. Here are some proposals that I believe will help with all three reform goals.

First, put a “fire wall” around ratings analysis. The agencies have already separated their rating and non-rating businesses. This is fine but not enough. The agencies must also separate the rating business from rating analysis. Investors need to believe that rating analysis generates a pure opinion about credit quality, not one even potentially influenced by business goals (like building market share). Even if business goals have never corrupted a single rating, the potential for corruption demands a complete separation of rating analysis from bottom-line analysis. Investors should see that rating analysis is virtually barricaded into an “ivory tower,” and kept safe from interference by any agenda other than getting the answer right. The best reform proposal must exclude business managers from involvement in any aspect of rating analysis and, critically also, from any role in decisions about analyst pay, performance and promotions.

Second, prohibit employee stock ownership and change the way rating analysts are compensated. There’s a reason why we don’t want judges to have a stake in the matters before them and it’s not just to make sure judges are fair. We do this so that litigants have confidence in the system and trust its results. We do this even if some or all judges could decide cases fairly without the rule. The same should be true for ratings. Even if employee stock ownership has never actually affected a single rating, it provokes doubt that ratings are disinterested and undermines investor confidence. Investors should have no cause to question whether the interests of rating agency employees align more closely with agency shareholders than investors. Reform should ban all forms of employee stock ownership (direct and indirect) by anyone involved in rating analysis. These same concerns arise with respect to annual bonus compensation and 401(K) contributions. As long as these forms of compensation are allowed to be based upon how well the company performs (and are not limited to how well the analyst performs), there will always be doubts about how the rating analysts’ interests align.

Third, create a remedy for unreasonably bad ratings. As noted above, the rating agencies have long understood (based upon decisions of the courts) that they will not be held liable for injuries caused by “bad” ratings. Investors know this. Why change the law to create a remedy if bad ratings arguably cause huge losses? The goal is not to give aggrieved investors a cash “windfall.” The goal is to restore confidence — especially in sophisticated investors — that the agencies cannot assign bad ratings with impunity. The current system allows the cost of bad ratings to be shifted to parties other than the agencies (ultimately to taxpayers). Reform must shift the cost of unreasonably bad ratings back to the agencies and their shareholders. If investors believe that the agencies fear the cost of assigning unreasonably bad ratings, then they will trust self interest (even if not integrity) to produce ratings that are reasonably good.

My former Moody’s colleague, Dr. Gary Witt of Temple University, believes that a special system of penalties might also be useful for certain types of rated instruments. Where a governmental body relies upon ratings for regulatory risk assessment of financial institutions — e.g. the SEC (broker-dealers and money funds), the Federal Reserve (banks), the NAIC (insurance companies) and other regulatory organizations within and outside the US — the government has a compelling interest and an affirmative duty to regulate the performance of such ratings. Even if other types of ratings might be protected from lawsuits by the first amendment, these ratings are published specifically for use by the government in assessing risk of regulated financial institutions and should be subject to special oversight, including the measurement of rating accuracy and the imposition of financial penalties for poor performance.

Fourth, change the antitrust laws so agencies can cooperate on standards. When rating agencies compete over rating standards, everybody loses (even them). Eight years ago, one rating agency was compelled to plead guilty to felony obstruction of justice. The criminal conduct at issue there related back to practices (assigning unsolicited ratings) actually worth reconsidering today. Once viewed as anticompetitive, this and other practices, if properly regulated, might help the agencies resist competition over rating standards. Indeed, the rating problems that arose in the subprime crisis are almost inconceivable in an environment where antitrust rules do not interfere with rating agency cooperation over standards. Imagine how different the world would be today if the agencies could have joined forces three years ago to refuse to securitize the worst of the subprime mortgages. Of course, cooperation over rating analysis would not apply to business management which should remain fully subject to all antitrust limitations.

Fifth, create an independent professional organization for rating analysts. Every rating agency employs “rating analysts” but there are no independent standards governing this “profession”: there are no minimum educational requirements, there is no common code of ethical conduct, and there is no continuing education obligation. Even where each agency has its own standards for these things, the standards differ widely from agency to agency. One agency may assign a senior analyst with a PhD in statistics to rate a complex transaction; another might assign a junior analyst with a BA in international relations to the same transaction. The staffing decision might appear to investors as yet another tool to manipulate the rating outcome. Creating one independent professional organization to which rating analysts from all rating agencies must belong will ensure uniform standards — especially ethical standards — across all the rating agencies. It would also provide a forum external to the agencies where rating analysts might bring confidential complaints about ethical concerns. An independent organization could track and report the nature and number of these complaints and alert regulators if there are patterns in the complaints, problems at particular agencies, and even whether there are problems with particular managers at one rating agency. Finally, such an organization should have the power to discipline analysts for unethical behavior."


Mark's complete testimony can be viewed here.

5 comments:

Sane Person said...

#1 and #2 make sense. #3, on the other hand, is laughable.

"Reform must shift the cost of unreasonably bad ratings back to the agencies and their shareholders."

I find it hard to believe that Mr. Froeba hasn't noticed the Moody's stock price drop from the 70s to the teens.

"should be subject to special oversight, including the measurement of rating accuracy and the imposition of financial penalties for poor performance"

It doesn't take a rocket scientist to realize that this concept is entirely unattainable, naive, and silly. How do you measure the performance of a rating? A rating is the reflection of probability, not one future path. Future events do not validate or invalidate a probability. Even looking across the entire asset class, on average, does not work because there are an infinite number of factors in the performance of financial instruments, very many of which are unpredictable by anyone.

If this is such a great idea, perhaps we should create ANOTHER government agency (staffed with bright, responsible government workers) whose job is to watch over this magical new government oversight agency and fine THEM for giving out unfair fines to the agencies.

Yes, yes, the answer is always more government.

GP said...

@Sane Person,

You make a good point insofar as the increased regulation is not always a net positive. At the very least, it's expensive and burdensome.

As it currently stands, however, rating agencies define what AAA means, what "default" means, and they remain unsupervised in their collection and representation of their rating performance.

As Dr. van Deventer points out in his Kamakura blog, Fannie Mae, Freddie Mac and AIG were conveniently excluded as "defaulted" assets by the rating agencies for the purposes of measuring the performance of AAA securities.

Thus S&P continues to show an (exemplary) 0% cumulative 2-year default rate for AAA corporate assets. Taking only FNMA and FHLMC into account, brings this number up to 0.18%, which tarnishes S&P's record.

As our guest blogger noted: "the agencies act as judge, jury and executioner, and are not required to justify their actions to any regulator or other third party." Their opinions therefore, protected from lawsuits by the 1st amendment, remain largely unsupervised and have been given free-reign. It's not surprising, thus, to have seen the decline in rating standards and quality over time.

Thus, we do believe that there should be both:
(1) oversight of a rating agency's measurement of its own performance - not necessarily on the security-by-security level, but more likely overall or within a certain asset class and;
(2) some measure that disincetives rating agencies and/or rating agency analysts from assigning egregiously poor quality ratings. Perhaps you could suggest a preferable way to create that outcome?

GP

Sane Person said...

The measure that should disincentivize rating agencies from assigning egregiously poor quality ratings would be their reputation in the free market.

Now, sure, other factors in the market made this somewhat inconsequential historically: the S&P/Moody's duopoly, the issuer-pays model, etc etc..

Just saying that having an artificial government agency with some fuzzy, pretty much undefinable mandate on punishing rating agencies for poor performance makes virtually no sense in the real world.

Just separating the business unit from the ratings unit in the agencies (a la #1) would go a long way. Rating analysts generally aren't stupid which, in the current structure, cuts both ways when it comes to ratings quality.

loans said...

As it currently stands, however, rating agencies define what AAA means, what "default" means, and they remain unsupervised in their collection and representation of their rating performance.

Anonymous said...

Sane Person,
Your idealized "market" for rating agencies does not exist because of regulatory restrictions around investments by fund managers. What exactly do you propose as an alternative, because the status quo obviously has not worked (since Enron)?

Secondly, why would you need a government agency to monitor ratings quality? It's trivial to examine ratings assigned, versus losses realized on an instrument after assignment of the rating. If losses > [losses implied by rating] the agency/analyst had it wrong. Of course, this raises other problems - such as analysts being unwilling to downgrade etc. etc., but it's definitely preferable to the "publishing company" defense put forth by the agencies so far.