When Pilate saw that he could prevail nothing, but that rather a tumult was made, he took water, and washed his hands before the multitude, saying, I am innocent of the blood of this just person: see ye to it.
Matthew 27:24
It no doubt sounds extreme to compare the SEC’s proposed changes in regulation of rating agencies to Pontius Pilate’s decision as judge in the trial of Jesus to distance himself from any responsibility. However, the SEC has similarly decided to avoid the thorny question of the rating agencies’ culpability in the credit crisis (who knows who such an inquiry might implicate) and take as little action as possible (the expression, washing one’s hands of a situation, originates with the passage from Matthew above).
Rather than address the rating agency conundrum frontally, whether through an inquiry or public hearings (this would be necessary, even if the SEC had a new regime in mind, to create the appearance of having been fair and solicited all views), the agency has decided to take the path of least resistance. As we discuss below, its remedies, even if they can be unduly dignified with that name, are cosmetic. The SEC has revived a 2003 measure that failed to win support and does not address the issues at hand; its other proposal is sure to be a non-starter.
However, the SEC has tried to couch its measures, such that they are, as an effort to reduce the importance of ratings. I consider that thrust to be a mistake.
Ratings have proven to an effective system for 25 years, from roughly the mid 1970s till the early 2000s. Yes, they were not ideal; the scorekeepers were known to be somewhat slow to downgrade deteriorating corporate credits. But as Myron Scholes and others have pointed out, ratings do perform a legitimate function. Having an expert assess default risk means that thousand of investors do not have to make their own examination (or at least, not as detailed a one as they would have otherwise). They are particularly valuable for retail investors, who in many cases lack the time, skill, and resources to perform credit analysis.
The real problem with ratings agencies was that they were given a protected franchise (it was well-nigh impossible for prospective entrants to get licensed as a National Recognized Statistical Ratings Organization) that played an central role in the capital markets, yet were subject to no regulatory supervision. Power corrupts, and while the rating agencies had far from absolute power, they had enough unchecked authority to lead them into plenty of trouble.
So it took two and a half decades of utter neglect before the rating agencies, which had performed a useful function, drove themselves and the credit markets off the cliff with their flawed structured credit ratings. And the roots of those problems are well known, starting with conflicts of interest and lack of rating agency liability for their decisions.
But like Pilate, the SEC has decided to take the most expedient course in a politically charged situation. In the unlikely event that their proposals were implemented, weakening the importance of ratings has the potential to produce all sorts of unexpected side effects, but they wouldn’t be evident until well after Cox has left office. And the Bushies have a knee-jerk preference for cosmetic measures and deregulation; this program draws on both predilections. And as noted earlier, it sidesteps the need for potentially embarrassing inquiries.
From the Wall Street Journal:
The Securities and Exchange Commission plans to propose rules that may diminish the longstanding importance of credit ratings across various markets, including the $3.4 trillion money-market industry, in the latest blow to the rating business stemming from the credit crunch.
The most significant portion of the rules, to be proposed Wednesday, would make it possible for U.S. money-market funds to invest in short-term debt without regard to ratings put on those securities by firms such as Moody’s Investors Service and Standard & Poor’s, people familiar with the matter said. Currently, SEC rules generally require that money-market funds purchase only short-term debt with high investment-grade ratings. The new rule would put more discretion in the hands of money managers to determine whether the debt is investment grade.
The SEC also will propose rules that may diminish the importance of credit ratings in determining the amount of capital that investment banks are required to hold. In all, the proposal will put about a dozen changes on the table that could touch on the role of credit ratings for investors and banks.
Even though the last measure, reducing the role of ratings in investment bank capital requirements, may not sound that bad after the fiasco of super-senior AAA paper that went south at a rapid clip, it is a mistake to weaken the structure of capital requirements without putting anything in its place, particularly when the thrust of financial services regulators around the world is to get firms to carry considerably more equity relative to their asset bases.
More important, the Fed, along with other US and international bank regulators, is trying to come up with more consistent capital requirements across banks and securities firms that have access to central banks’ emergency facilities. Ratings are currently a fixture in Basel II bank regulations, although they play a greater role for small banks than large ones, who have more latitude in the approaches they can use (note that is in the process of being rethought). Thus the SEC move on capital rules is premature and out of step with the larger initiatives underway. Cynically, I assume it’s diversionary but I welcome other views.
In all likelihood, these recommendations will go nowhere. The money market plan was floated in 2003 and died on the vine. And the overarching idea, gutting the importance of ratings, will stir opposition. Again from the Journal:
“My initial reaction is, what’s the alternative?” to using rating firms for the rules, said Hal Scott, a Harvard University law professor specializing in capital-markets regulation. “What we need to do is have more assurance that these ratings will be accurate.”
As we have discussed elsewhere, ratings are so deeply enshrined in a various regulations and practices that simply getting rid of them, or de-emphasizing them, is vastly easier said than done.
If you think reforming rating agencies is difficult, a short list of tough-minded proposals would make a great deal of difference. These suggestions from Josh Rosner, interviewed by Institutional Risk Analytics, have merit:
The IRA: So let’s switch gears and talk about the ratings industry. What are you going to be saying at AEI tomorrow?
Rosner: The proposed solutions that we’ve seen from the global securities regulators are soft and senseless approaches. While I don’t question the sincerity of the efforts or their intentions, I have to question to their understanding of the problem. It is very clear to me that the solutions do not require hundreds of pages of analysis. There are some very clear, simple and elegant solutions, but nobody is talking about them in the regulatory community. The failure to do so is ultimately going to force the political leadership in the US and Europe to reconsider the role of the rating agencies. We will see assets other than RMBS and CDOs having troubles and it will become increasingly clear that the rating problems are spreading across most structured assets classes.
The IRA: Do you see the EU moving to set up their own quasi-governmental rating agency?
Rosner: It is unclear at this point. What is clear, though, is that unlike the US legislature, the European Commission recognizes that there may be no way to do a global solution. The UK is a holdout, for example and is still pushing for a global solution.
The IRA: Well, the UK was the chief culprit in constructing most of the rancid CDOs that have caused the crisis, so their abstention is no surprise. Nobody has focused on the fact that most of the private label securitizations that have caused hundreds of billions in losses to global financial institutions came out of the City of London, not New York. And look at the breakdown of the “global” process around Basel II.
Rosner: Right. And since the US SEC and UK Financial Services Authority have different approaches to regulation, I’m not sure how you could accomplish a global approach even if there were the political will to do so. We are rules based and they are principles based so what they could interpret as rules from the principles in the IOSCO Code only become suggestions here in the US.
The IRA: Agreed. There is a huge political backlash building in the US against the ratings agencies and Wall Street that should come to a boil around Election Day. We got a really scary call from a reporter in Wisconsin yesterday. It seems that several of the largest school districts in the state decided to borrow short-term and buy CDOs to help finance future health care liabilities – obligations that the state school system cannot fund. This is financial engineering, zaitech pure and simple. Now that the ratings on these CDOs have fallen, along with the market value, the cost of the debt has gone up thanks to an “innovative” trigger provision crafted by a certain large ibank we won’t mention. Why these school districts were even allowed to borrow to fund interest rate speculation is unclear. The WI AG reportedly is investigating both the members of the school district boards and the dealers involved….
So on that happy note, give us your list of sensible ratings reforms.
Rosner: First, starting from the bottom of the list, it …. there should be a cooling off period for ratings analysts before they can go to work for an investment banking team that structures deals, for a period of one or two years.
The IRA: We’d be happy to say two years….. What’s next on your list?
Rosner: Second, I would say that banks, mutual and pension funds can only invest in rated, exchange traded structured assets. This proposal would address the issues of transparency and liquidity by narrowing the bid-offer spread on these securities. The fact of being exchange traded would require SEC registration. This public disclosure would help investors in structured assets be less dependent upon ratings for making investment decisions. Could you have an exception to the exchange-traded rule for structured assets? Yes, but the bank or fund must be able to demonstrate how their internal credit risk models supports that investment. The funds would have to prove to their regulators that they understood and could model the structures independent of the rating agencies or dealers. But as a practical matter, we are talking about moving most of structured finance onto the exchanges.
.
Yves here. That would require standardization of structures, which we think is a good thing. Back to the interview.
The IRA: So you agree with our view that we fix the problem with the ratings agencies by fixing the market structure issue?
Rosner: Yes. I don’t have a real problem with how MCO or S&P rate corporates. It is only in the conflicted world of structured finance where a problem exists, where the company being rated exists only after it has been rated.
The IRA: What else?
Rosner: Next is the issue of liability. Where a rating agency knowingly rates a security which they helped to structure and were aware of — or should have been aware — of potential fraud in the collateral pool, then they should lose their current regulatory liability exemptions.
The IRA: Haven’t the courts in the Southern District of New York already done that?
Rosner: No, they have only ruled on the claim of journalistic privilege in the context of discovery. The courts have ruled that the ratings agencies cannot hide behind a claim of journalistic privilege for advisory activities. In the American Savings Banks v. UBS litigation, the Court in May of 2003 ruled that the rating agency personnel were not acting as journalists but instead were acting as advisors and therefore could not assert their journalistic privilege. The exemption from liability has not yet been defeated but it really hasn’t yet been tested in the wake of the structured securities scandals.
The IRA: No, but the stage has clearly been set by the findings of fact in the American Savings Banks v. UBS ruling.
Rosner: Correct. When the ratings agency acts as advisor, then they should have a fiduciary role to the investors who buy the paper.
The IRA: Yup. If they act as investment bankers and get paid like investment bankers, then they have a duty to both the issuer and to the end-investor.
Rosner: Next, when a rating agency changes its model, it must be required to go back and, in a timely manner, re-rate the securities that were rated in the primary market using that model.
The IRA: Haven’t the ratings agencies been forced to do this with the disclosures of model errors?
Rosner: No, they have not. When the ratings agencies change the model they typically only apply it on a prospective basis. They wait until the problem is obvious with a given transaction, only then do they tend to apply the new model to current ratings. They do not change the rating approach that was applied to all of the deals priced with that model in the primary market.
The IRA: That is incredible. We can hear the dinner bell ringing for the trial lawyers now. Just imagine the fun the plaintiffs will have with discovery on deals where a modeling error was admitted but no change was made to the secondary market assessment.
Yves again. The failure to re-rate outstanding issues even when the models changed was discussed at length in a paper by Rosner and Joseph Mason released more than a year ago, yet there was surprisingly little notice of this issue.
Rosner: Lastly, in the secondary market, the ratings agencies should be required to automate that they use their originally assumed loss curves and original at issuance assumptions tied to the monthly remittance data, to re-draw the loss curve. This would cause a natural “truing up” of the loss curve and the rating, reducing volatility and market risk. Now here’s the problem: The SEC, like the other securities regulators say that they do not want to tell the rating agencies what models to use or how they should operate. My response to that is reminiscent of a comment in a presidential debate that is very simple: It’s the models, stupid.
The IRA: Sounds like a title. Chris Cox will be out in a couple of months and we may have a new commission appointed by Barrack Obama.
Rosner: One of the first issues that the SEC needs to consider after the November election is whether the “hands off” approach by the SEC to modeling should be reconsidered.
The IRA: Well, duh. If the model is used for pricing a security, rating a security and, in a larger sense, for all of the asset allocation decisions made by Buy Side investors, then the SEC has a statutory responsibility to supervise the use of models. This just shows, yet again, that the SEC does not fully understand the markets that they are required by law to supervise.
Rosner: Now, to preempt the objections from our conservative friends who think we are all socialists for asking the SEC to extend regulation to models, consider this. If we are going to remove the statutory requirement that investors, especially banks, insurance companies and pension funds, use ratings to guide asset allocation decisions, then not only should there not be regulation of models but investors really should treat ratings as just journalistic opinions, like stories they read in newspapers. But if you view the ratings agencies as exercising authority outsourced by the SEC and other regulators, then the companies should be well regulated.
The IRA: Of course. It is clear that the ratings agencies are part of the infrastructure of the investment world and are essentially standards setting bodies via the ratings process.
Rosner: If we legislate what regulators can do then we should be able to legislate what the ratings agencies do when they act, in essence, as agents of the government in terms of setting standards and the basic rules for the investment process.
19 comments
Comments are closed.
A couple of comments.
“We got a really scary call from a reporter in Wisconsin yesterday. It seems that several of the largest school districts in the state decided to borrow short-term and buy CDOs to help finance future health care liabilities – obligations that the state school system cannot fund. This is financial engineering, zaitech pure and simple. Now that the ratings on these CDOs have fallen, along with the market value, the cost of the debt has gone up thanks to an “innovative” trigger provision crafted by a certain large ibank we won’t mention. Why these school districts were even allowed to borrow to fund interest rate speculation is unclear.”
See, this is a perfect example of how the big problem was dumb investors. School districts (!) should not be investing on margin, for God’s sake, no matter what they’re investing in. Let alone with an asset/liability mismatch. It doesn’t take a genius to work that out.
“The SEC, like the other securities regulators say that they do not want to tell the rating agencies what models to use or how they should operate. My response to that is reminiscent of a comment in a presidential debate that is very simple: It’s the models, stupid.”
It’s not a question of whether they want to. They don’t have the legal authority. The Credit Rating Agency Reform Act of 2006 explicitly states: “Notwithstanding any other provision of law, neither the Commission nor any State (or political subdivision thereof) may regulate the substance of credit ratings or the procedures and methodologies by which any nationally recognized statistical rating organization determines credit ratings.”
If we want the SEC to regulate the substance of ratings, Congress will have to legislate it.
Finally, many, but by no means all, of the concerns raised in that interview are addressed (how effectively we’ll just have to see) by the measures already announced by the SEC. For instance, the SEC has banned agencies from rating a product if they participate in its structuring.
There is a pretty simple solution to this problem: stop both a) protecting the monopoly of these ratings agencies and b) having the companies they rate pay for their ratings. Let these companies duke it out in full on competition, selling their research to people who are considering buying their rated products. I think the MBIA Ambac requests of Fitch have proved that you can’t have companies paying for their own ratings, and I still don’t understand why people think government sponsored monopolies have ever worked.
I don’t agree that his suggestions will solve the problem, so long as you allow the construction of instruments whose values depend on models which use opaque parameters. The SEC can try to regulate models all it wants; it’s just not going to be able to do it. IBs have trouble regulating their own models, and they’re the ones who came up with them.
Here’s an example of the problem. Regulators demand stress tests. So companies would create stress tests with a movement of 20% in the underlying, and then create products which showed 0 risk for a movement of 20% but explode for a movement of 21%. All models have defects, and all models can be taken advantage of.
IMHO the original sin is the attitude, “Everything which is not prohibited is allowed.” This has to become, “Everything which is not allowed is prohibited.” Once that becomes the rule, then you can start working on what is allowed. But without it, regulators can be as diligent as they want to be about what is allowed; they’re just not going to be able to keep up with the innovation of the IBs.
“IMHO the original sin is the attitude, “Everything which is not prohibited is allowed.” This has to become, “Everything which is not allowed is prohibited.” “
That’s the very nature of a common law system, though. To be flippant, if you want “everything which is not allowed is prohibited”, move to France.
a: “Here’s an example of the problem. Regulators demand stress tests. So companies would create stress tests with a movement of 20% in the underlying, and then create products which showed 0 risk for a movement of 20% but explode for a movement of 21%. All models have defects, and all models can be taken advantage of.”
Let’s be frank. The banks have marketed these products to instititional and retail investors in the form of now infamous portfolio insurance and more recently structured capital at risk products. Since regulators permit the stuff to be sold to investors, there’s no reason to think they wouldn’t allow bankers to stuff their own bank with them if they run out of customers.
I have blasted the rating agencies and the SEC’s “regulation” of them for months. Joshua Rosner is one of the few people out there who understands what’s going on. The rating agencies have the same problem as the CPA firms: lack of user accountability. Nothing the SEC has proposed will help matters.
Danny said “…stop both a) protecting the monopoly of these ratings agencies and b) having the companies they rate pay for their ratings…”
Danny, I think this would create an environment where the companies (ratings service buyers) would “opinion shop” for the ratings they desire (which may have been a significant part of the problem over the past few years). I think a better idea would be to have investors pay for the service. That seems how it was done in the old days. This would be more consistent with your statement “Let these companies duke it out in full on competition, selling their research to people who are considering buying their rated products.” This is not a new concept. The PCAOB was created because of the audit industry’s failure to regulate itself. If you accept th premise that ratings agencies must be a key component of the financial system, as Yves, Rosner and IRA have ably argued –then- it they will have to be more heavily regulated.
The problem would be how to set up a viable system where the investor pays. The easy answer would be a de-minimis transaction tax on all rated securities. The hard part would be structuring the collection and payment system so that the tax collection and distribution system would not impede nor have undo influence upon market operations.
I find it unclear how greater competition would help matters. In my experiece – and I work for a rating agency – the much maligned conflicts of interest play a much smaller, if any, role in the way the RA’s arrive at the ratings. On the other hand, structured finance departments at rating agencies are characterised by (1) creaking technology; (2) high turnover; (3) lack of experience – the majority of analysts are in their late 20’s or early 30’s, and last but not least, (4) basic lack of enough people. It is unclear how competition would adress any of these – it may in fact have a counterproductive effect insofar as stretched resources are concerned.
Take as an example the Moody’s CPDO situation. Before we get to the cover up (which I doubt actually occurred), the real issue is why there was a bug in the first place and how a highly aggressive rating methodology was signed off by Moody’s credit policy. The basic answer is that the ratings and the methodology involved a very small number of analysts, working under extreme time pressure and hassled by the bankers. Something has to give and in this, and in many other cases, what gave was (a) quality assurance; and (b) ability of the analyst to step back and take a look at the bigger picture – they simply don’t have the time to do so.
In this regard, the agencies are not that different from your typical underresourced regulator…
The Cox/SEC proposals are consistent with the ‘free market’ ideology seeking to return all human activity to the Hobbesian state of ‘freedom’.
From one perspective, yes, this is doing the minimum. But, really, it is not the minimum. It’s the maximum in pursuit of a radical ideology that would eliminate government as a player in favor of contract as the only law of the land — and a judiciary whose only concern with contract is caveat emptor.
Anon at 8:52am:
I presume that failures of the consulting work done on the front end and analytical work done on the back end by the RA’s on all of those structured/securitized financial instruments and entities that received investment grade ratings and then subsequently saw their ratings reduced to junk, were the solely result of an inexperienced and overworked staff and creaking technology? I guess then that none of the folks working at the RA’s were incentivized to bring in additional business and maintain customer relationships.
been there,
“Let these companies duke it out in full on competition, selling their research to people who are considering buying their rated products.”
That was the gist of my argument, although it wasn’t the focus of my post. I agree with your statement to let the people who want to buy rated products pay for the research. But I don’t need the government telling me which companies do and don’t do a good job.
There is a reason the SEC puts rules proposals out as proposals, it is so their relative merits can be discussed, and discussing what has gone wrong is certainly a part of that, so the inquiry that Yves is asking for is already here.
The SEC has rules it has to follow and a process that it has to adhere to in order to ensure appropriate transparency and feedback. They can’t just do whatever they want to do and tell people later; I mean, we’re not talking about the Federal Reserve here.
Ginger Yellow,
But wouldn’t a financial products regulator make sense. Every time a bank comes up with a new product they have to submit it to the regulator which will determine legal standards for accounting and tax reporting of the product before approving it. I think this would solve a lot of the gaming of the system that’s been going on. And while I’m sure the “if the regulator wasn’t established in the Constitution, it’s clearly unconstitutional” libertarians will fuss, I don’t how anyone could argue that this would violate common law standards.
Here’s a link to the Wisconsin School District CDO disaster. Sigh…
http://www.jsonline.com/story/index.aspx?id=763770
It’s an idea, but it’s got a ton of problems. One, it would obviously stifle innovation – now a lot of people will say that might not be a bad thing, but you’ve got to take it into account. On a similar note, you can be sure that New York would no longer be the dominant financial centre in the world. Again, the benefits may outweigh the disadvantages, but you’d have to do the weighing.
Two, we already have bodies who set tax and accounting rules – you’d have an awful lot of overlap and potentially conflict.
Three, the burden on the regulator would be huge. You’d need an enormous, highly paid staff and sophisticated IT to validate the hundreds or thousands of products proposed every month.
Four, just because a particular product is approved by the regulator, it doesn’t mean that investors will understand it or make wise investment decisions.
Five, on what basis would the regulator approve or disallow products? Simply that the product does what it claims to? Some measure of risk or leverage? I find it hard to believe that such a regulator would not have approved subprime RMBS, say, when it was first introduced.
As I’ve said before, it seems to me the best solution is to regulate the buy side, alongside tough penalties for misselling and fraud.
Jesus, that Wisconsin story is horrible. They didn’t even hedge the interest rate risk. This is elementary stuff, to the extent that there may have been misselling. It’s also possible, however, that the district was offered a swap but decided not to spend the money.
GY: Thank you for the answer. I don’t think a regulator could solve all the problems — you’d definitely still need to regulate the buy side and penalize mis-selling, but I think that just having to pass a product by a regulator first would help discourage a bunch of the products that are designed to exploit loopholes in the laws/regulations.
Subprime MBS would definitely have gotten a pass, but synthetics might have been slowed down dramatically.
The States are going to have to dictate what investments can and cannot be purchased with tax payer money and the courts are going to have to prosecute malpractice by the “investment professionals”.
The Federal government is sick and may not recover.
Been there, I am not sure what that means. Maintaining customer relations is a normal part of any business, rating agencies included. Not so long ago you would have people complaining that rating agencies are too aloof, don’t pick up the phone etc. So, there is a balance there.
Individual analysts don’t actually have much by way of an immediate incentive to assign aggressive ratings on any individual transaction (note by way that this would involve a group of analysts colluding as all decisions are made by a committee). The incentives come from the broader perspective of senior analysts holding stock in the rating agency and the general culture of ‘getting the deal done’. This is pretty debatable.
The under-resourcing point is as far as I can see not one that anyone is arguing against (i.e. it happens to be true…). As annyoing as it is, the flaws of rating agencies are subtler than the pretty superficial ‘conflict’ argument. One has to establish how the theoretical conflict actually gets converted into a specific rating and what the process is. I have been thinking about it, and this is not obvious at all. Take the consulting/rating point: the reality is the iterative nature of rating agencies’ work in structured finance is not really consulting: since the structures are set up from scratch, the ra’s provide a useful service in providing feedback to the issuers/bankers whether a particular structure would get the targetted rating or not. There is actually a strict prohibition on analysts providign structuring advice to the issuers. So, not so easy to establish this leads to bad ratings.
Have you ever seen bad regulation put in place? Same reason: not enough people, high turnover, lack of time, lack of experiece, pressure/capturing from/by the industry.