Credit Agencies
I’ve heard it argued that we could get along without things like government inspections of food safety. After all, consumers would still want to know that their can of black beans contains black beans rather than poison. So bean sellers, in order to bolster consumer confidence in their brand, might pay firms that specialize in food certification services to inspect their plants and proclaim them okay or not. Wouldn’t the certification firms have a huge conflict of interest? Well, no, the theory goes. Certification would only be worthwhile to the bean seller if the certification firm had a very strong brand. And a certification firm could only maintain its strong brand by doing the certification job with integrity. Thus we march along the libertopia without all dying of botulism.
Of course, things don’t work like that. Except, it turns out that this basically is the system we use to rate the quality of bonds. But as a system it’s failed miserably:
Since the subprime mortgage troubles exploded into a full-blown financial crisis last year, the three top credit-rating agencies — Moody’s, Standard & Poor’s and Fitch Ratings — have faced a firestorm of criticism about whether their rosy ratings of mortgage securities generated billions of dollars in losses to investors who relied on them.
The agencies are supposed to help investors evaluate the risk of what they are buying. But some former employees and many investors say the agencies, which were paid far more to rate complicated mortgage-related securities than to assess more traditional debt, either underestimated the risk of mortgage debt or simply overlooked its danger so they could rake in large profits during the housing boom.
Relative to the hypothetical bean scenario in which this sort of thing works out, we seem to have had a few issues here. One, for this scheme to work, the rating agencies need to place a higher value on the long-term viability of their brand than on short-term profit opportunities. But of course we know that people are often short-sighted, and often heavily discount the future relative to the present. Relatedly, for the scheme to work we need the firms to be primarily concern with the long-term interests of the firms rather than the interests of the managers. But even if Moody’s, qua company, winds up taking a giant hit over this, it’s still not clear that Moody’s top executives won’t have come out ahead.
Last of course there’s the simple issue of limited options. If two of the ratings agencies had stayed on track but one had gone in for a lot of bad actions, then the one firm might have profited in the short-run but would be dead now. But instead all three went chasing the money, so nobody gains a competitive advantage.
All of which is a long-winded way of saying that it would make a lot of sense to try to develop a public agency that rates credit instruments. Wouldn’t stop anyone from relying on private sector ratings if they wanted to. Nor would it guarantee that the public agency would always get things right. But it would provide a check on some of the distortions that the current system produces.
88 Responses to “Credit Agencies”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
December 9th, 2008 at 8:54 am
We have an economic system predicated on the sage observation that almost everyone is a self-dealer almost all of the time, and yet we expect executives to put the interests of their company above their own interests, we expect ratings agencies to put the interests of an abstraction called “the integrity of ratings” above their own interests, etc. Well, we don’t 100% expect this and therefore we invent a combination of institutional controls and purely wishful thinking about how self-dealing will be controlled by long time horizons (”the success of our ratings agency depends upon the integrity of our ratings!”), but there’s all the evidence in the world that this doesn’t work. Or, more fairly put, it works sometimes and not other times, and the failures are catastrophic. To me it’s obvious that we need much stronger institutional controls, because self-dealing is too firmly embedded in our nature, and after all we have a system that’s based on harnessing it.
December 9th, 2008 at 8:56 am
“Of course, things don’t work like that.”
One could say the same thing about occupational licensing of plumbers.
December 9th, 2008 at 9:02 am
http://www.democracyjournal.org/news.php?ID=33
December 9th, 2008 at 9:09 am
If two of the ratings agencies had stayed on track but one had gone in for a lot of bad actions, then the one firm might have profited in the short-run but would be dead now.
Part of the problem is that the rating agencies rely on the cooperation of the issuers. If two of the agencies had gone in for a lot of bad actions but one had stayed on track, that one would have lost access, rated fewer securities, lost influence, perhaps gone bankrupt or been bought out by a buyer willing to go in for a lot of bad actions.
December 9th, 2008 at 9:11 am
But some former employees and many investors say the agencies …. either underestimated the risk of mortgage debt or simply overlooked its danger so they could rake in large profits during the housing boom.
Shocking. Disgruntled people come up with two options, and Matt picks the one that fits his pre-existing narrative, involving a fetish for more government.
So, we have one example here of private agencies making a mistake and that justifies a new government agency that is not guaranteed to get things right (your words), and you go, “Yep! That’s change I can believe in!” Delusional. Do you also think you can fly when you wear your Superman underoos?
December 9th, 2008 at 9:20 am
Mike,
Here’s the situation: these rating agencies have one, and only one function: to rate securities. They failed at this function, in so epic a fashion as to have endangered the entire world financial system. Do you think that we should do nothing about this? And if you do think something should be done, but don’t want “more government” (God forbid!), then what exactly do you propose?
December 9th, 2008 at 9:20 am
Heck, just look at all those crappy products we import from China that are toxic. Toxic leaded toys, toxic cat food, basically everything that comes to the US that isn’t regulated is toxic and kills people and pets. Free market self regulation?
December 9th, 2008 at 9:21 am
Mike: Matt’s explanation has the advantage of being true, while your explanation is bullshit you made up so that you can indulge in your fantasies of the evil statists. Everybody on Wall Street knew that the ratings agencies had become corrupt, but other than the people doing the corrupting no one could point to a smoking gun.
December 9th, 2008 at 9:28 am
On the other hand, testing for the presence or absence of pathogens in canned goods is many orders of magnitude simpler than measuring the risk associated with a given bond. Microbes are either there at measurable levels or the aren’t–the health/future prospects of a firm are contingent on many hard to measure factors.
December 9th, 2008 at 9:33 am
We already have the issue of regulatory capture when dealing with government agencies. The issue becomes infinitely worse when the regulators are getting their salaries from those they regulate.
There are some nice smoking gun type emails out there showing how the rating agencies were wrapped around the finger of the banks. Something about ‘…we’d rate this if it was put together by cows…’.
December 9th, 2008 at 9:42 am
I’m not saying the government is evil. I’m just saying that if there’s a subjective job, there will be mistakes no matter who performs it, so saying “the private sector messed up so we should nationalize that function” is not compelling. Would you apply the same standard to the government? When the government has an epic fail, is it fair to call for that function to be privatized? If somebody dies from fecal-contaminated produce imported from Central America, do we say “government messed up so now we’ll privatize all food inspection?” No, we say “Inspect more or otherwise improve your process.”
Maybe there’s a good argument to be made that the government should perform the ratings, but Matt didn’t make it in this post.
Walt – Oh yeah? Well, my argument is true and yours is bullshit. How do you like that? No givesies backsies!
December 9th, 2008 at 9:46 am
This isn’t the only recent case of this kind of failure of free market certification. Anyone remember Enron and its books being certified by Arthur Andersen? Or how about all of those dot-com stocks rated “BUY” when the people doing the rating knew they were crap?
Gov’t is clearly not perfect, but having a different set of eyes with different incentives take a look at these things is not a bad idea. We already have lots of complex banking supervision which (mostly) works well. If these complex financial products are going to be allowed onto the books of FDIC-insured banks (or “too-big-to-fail” institutions of other sorts) then the taxpayer has the right to a second opinion on the value of these assets.
December 9th, 2008 at 9:56 am
The key to solving the ratings problem to break the practice of having the firm being rated being the one who picks and pays ratings agency doing the work.
Firms will never pay to get a bad rating.
The best way to solve this would be to have the firm being rated apply to a third party and have that party randomly decide which ratings firm actually does the rating.
This way the ratings firm is not working for the company being rated but instead will work for the assigning agency.
Accounting rules should be changed to only recognize ratings created in this manner.
In addition to this, as Matt says, the assigning agency should keep track of the long term record of each rating firm and disqualify those with poor long term records.
The assigning agency should also encourage a substantial number of new firms to become ratings firms.
December 9th, 2008 at 9:57 am
it really doesn’t make sense for that entity to be anyone other than the government.
Said that guy with the government fetish.
On a serious note, is Hank Paulson free of undue influence? Where do you suppose a government agency might find employees with the necessary experience to rate complex financial instruments? It will be the exact same people doing it now, and as long as they treat the banks right, they’ll have a nice VP spot open when they leave government service. The same way so many other government positions work.
December 9th, 2008 at 10:01 am
In the absence of sane and comprehensive regulation, a fly-by-night canner can simply stuff mouse turds in brown goop and disappear after selling to markets.
This is a pretty good approximation of what happened with unregulated mortgage brokers.
December 9th, 2008 at 10:01 am
Oh come on. Anyone that would use recent outbreaks of illness from spinach, tomatoes or beef as a way to say that FDA monitoring doesn’t work would be laughed out of the room here. Same goes with people that could use examples of the FDA messing up on drugs. Consumer Reports and the Better Business Bureau are highly trusted, private concerns that are less corrupt. There are ways for both the government and private concerns to be corrupted.
Do you really think that a credit rating agency run by the government would have correctly rated those bonds? The government agency would likely be understaffed and would not have the level of talent that the private concerns have. The better solution would be to eliminate the conflictd of interest, not kill CRAs.
December 9th, 2008 at 10:02 am
Hey! I have an idea! Let’s get DISTRACTED by accusing MY of having a fetish for government!!
C’mon, it’ll be fun!!!
December 9th, 2008 at 10:03 am
Matt, thanks for drawing this useful connection from credit ratings agency failure to the general weakness of independent, private certification mechanisms. Your explication of the particular weaknesses in play (the managerial agency problem, the short-term/long-term problem, and the oligopoly versus perfect competition scenario) are apt.
As someone who wanders in the wilderness between libertarianism and liberalism, this was clarifying. I smell a link from Tyler Cowen coming…
December 9th, 2008 at 10:05 am
Seems to me that the way to address the problem of investors relying on dodgy ratings is not to create another source of potentially dodgy ratings (nobody can guess credit risk perfectly, and a state controlled body has its own conflicts of interest), but to make sure investors don’t rely on dodgy ratings. It’s not too much to ask, surely, that institutional investors buying credit products should be able and required to assess credit risk. What else are they being paid for?
December 9th, 2008 at 10:06 am
Whee… Let’s use a poor implementation to generalize to failure of the implementation itself!!
Understaffed, poorly talented competition with private concerns is only a failure of willpower, and in no way is an indictment of the concept.
December 9th, 2008 at 10:07 am
“…it would make a lot of sense to try to develop a public agency that rates credit instruments….”
It might make sense if the people running the agency weren’t the stooges of politicians bought off with “campaign contributions” or other legal bribes. But when has that ever been true in D.C.?
December 9th, 2008 at 10:12 am
Mike: The difference is that you don’t know anything. You have your little rule about the world works, and that’s it. The ratings agencies didn’t “make a mistake”, they were corrupted. I know people who have suspected for several years that the investment banks were corrupting the ratings agencies by leaning on the fact that the investment banks throw repeat business their way.
December 9th, 2008 at 10:14 am
What always got me was how many people insist on using economic theory to explain not why empirical findings occurred, but rather why empirical findings are somehow wrong or won’t happen in the future even if they occurred in the recent past. China has incentives to not let poison contaminate their exports, the credit agencies have incentives to properly evaluate risk, and business executives have incentives not to run their businesses into the ground. And so long as incentives (theoretically) exist, then no regulation is ever needed no matter how obviously flawed the current outcomes are.
December 9th, 2008 at 10:15 am
Dodgy ratings are one thing. Complete blindness to a systematic meltdown of the fundamentals of the market are quite another. I’m not sure that a government entity wouldn’t have been as blind… However, it seems to me that, in the absence of valid information, a government entity might be better positioned to say, with a straight face, ‘we have no valid information.’
I think Matt is coming at this a little sideways. I don’t think a straight up ‘ratings agency’ that simply mirrors whats done in the private sector is all that efficacious. I do think that the SEC ought to be funded better and that the SEC ought to publish compliance guidelines. Fr’instance, when the SEC allowed the 5 largest investment banks (none of which, it ought to be noted, now exist in the same form…) exceptions in 2005 to the 12:1 leverage rule, they ought to widely publish this fact and attach a ‘compliance’ rating to this. Lehman went from 12:1 to something like 60:1 and this is the reason they’re dead. Had the SEC noted this more widely, Lehman might have been reigned in, or others might have been given pause.
December 9th, 2008 at 10:16 am
> Of course, things don’t work like that.
Well, there is also the issue of professional integrity. Quite a bit of inspection and certification in the world of aviation is in fact done that way (although admittedly the FAA conducts random re-inspections and audits and levies fairly brutal punishments for transgressors), both in manufacturing and training. It is very rare that any person or organization sets out to subvert that system even in the worst economic times. You can come up with ECON101 reasons why this system maintains itself, but there does seem to be a lot of personal integrity involved.
December 9th, 2008 at 10:30 am
DTM,
I liked davenj’s idea. He said:
Another way would be to have a small fee (think fractions of a cent) added to all debt and equity transactions (who are the true beneficiaries of credit ratings). You could then decide which credit agency(ies) you disburse your fees to and have those CRAs receive the funds. Like mutual funds, you could have the CRAs publish their results for different credit ratings to provide a historical batting average. So if you think Moody’s is really good at industrials, you have all your industrials fees go there, but you think S&P is the est at financials, so you have those fees go to S&P. It would provide a pay for performance vehicle.
December 9th, 2008 at 10:35 am
Possibly, although the IBs don’t exactly have a lot of respect for the employees of the rating agencies. Still, the problem of regulators being able to move out of government and into private jobs at the companies they regulate is a seperate issue, and one that needs to be dealt with.
The immediate issue is that the incentives at rating agencies are set up so that the agencies have an incentive to give the best rating they possibly can in order to get repeat business. This is the same issue that came up with Enron. The accounting firms would do whatever they could on the accounting side in order to keep their customers happy because then they would make money off of consulting business with the same companies.
December 9th, 2008 at 10:51 am
An excellent idea, Matt.
December 9th, 2008 at 10:54 am
What about government bond rating? This seems at least a much mor elogical first step. Public entities (water districts, counties, states) pay to have their debt rated by companies, and then insured (poorly).
If you are going to start a program like this, start with a system based on government debt ratings.
I don’t think there is a simple way around the payment problem. Having the government pay will mean that we subsidize repeat borrowers. Or small dollar borrowers. Thisdoesn’t seem right.
Some part of the system for any private rating should involve insurance (ie, that the rating agency also insures). That way, the ratings agency also participates on up and downside of rating correctly.
Hmm, this seems like a reasonable approach on both sides–for private and for government systems. Any idea why this wouldn’t be right. After all, insurance companeis profit by pricing risk accurately. That is exactly what a rating agency should do.
bp
December 9th, 2008 at 10:59 am
Maybe there’s a good argument to be made that the government should perform the ratings, but Matt didn’t make it in this post.
Yes he did. He listed a number of unstated assumptions needed to make the idea of independent agencies work: focus on long-term reputation-building rather than quarter-to-quarter profits, focus on the welfare of the individuals rather than the company as a whole, and the same incentives leading to everyone making the same mistakes and therefore getting no competitive advantage from being right.
If you had read more carefully, you might have noticed that all of those problems would be reduced or nonexistent if there were a public rating agency in addition to the existing private ones. (The PRA wouldn’t be seeking profits. It could probably depend on a known amount of support within a presidential term or at the very least a Congressional session, either of which is a longer time period than a financial quarter. And it’s not like government bureaucrats are paragons of selflessness or anything, but surely you agree that they’re less likely to have a profit-oriented attitude than people in business.)
Also, it’s hard to take your comment very seriously when you start out by accusing Yglesias of fetishizing government. As DTM said in the second comment, Matt tends to be more libertarianish than most liberals. FBOFW.
December 9th, 2008 at 11:04 am
davenj’s idea only turns them into a public agency if the third party is a public agency. I’m not too wild about the randomly, idea. I would prefer to have the intermediary actively choose the credit agency based on performance.
The reason you would have to have the ratings work on the resale is that credit ratings have value well after initial issuance. The transaction costs would only go up by the size of the fee minus the cost of a CRA that’s already baked into the issuance. It’s possible to figure out what the normal fee would be for an issuance (take average trading volume over the lifetime of the issuance, adjust for time horizon, create a multiple for the complexity of the vehicle and back into the fee). There would be probably be a slight increase overall in transaction costs, but that would probably significantly reduce the deadweight loss from credit rating becoming a government function.
December 9th, 2008 at 11:26 am
OK, but how exactly? The conflict of interest is caused by the CRAs being paid by the issuers. So, to eliminate the conflict of interest we need to find someone else to pay the CRAs.
That doesn’t eliminate all the conflicts, by any means. Pretty much anyone who pays the agency is going to generate a conflict. Moreover, there is another inherent conflict regardless of how the agencies are paid. New asset classes represent new revenue streams, so agencies will always be incentivised to grant high ratings to emerging asset classes which may not have a performance history or which may be too complex to assign a rating with any confidence. If you ask me, this had far, far more of a corrupting influence on the rating agencies than the issuer-pay business model.
And this leads to a natural division of labor: it would be redundant and wasteful for every institutional investor in the world to be performing independent evaluations of every possible investment, instead of each institutional investor focusing more on understanding their particular clients.
I think this is arguably (though by no means definitely) true for things like corporate bonds where there is generally a consensus on the risks involved. Credit opinions on companies rarely deviate by more than a few notches (or the equivalent). But it’s not true at all for most structured securities – there are simply too many variables for which you have to make assumptions, and different investors will have different views on them. You can’t possibly outsource your evaluation of a structured security or portfolio without also outsourcing your opinion on future interest rate movements, prepayment rates, house price movements, default rates, recovery rates and so on, let alone non-credit risks like liquidity and duration.
Some part of the system for any private rating should involve insurance (ie, that the rating agency also insures). That way, the ratings agency also participates on up and downside of rating correctly.
The obvious problem with this is that it would restrict credit ratings only to highly capitalised companies, making it even more of an oligopoly than it currently is.
December 9th, 2008 at 11:34 am
Why should a CRA insure a CCC rated bond? If you’re dumb enough to invest in it, it’s your problem, not theirs.
DTM,
You could reduce those costs by consolidating the decision-making process to happen monthly/quarterly/annually rather than right when the transaction occurs.
December 9th, 2008 at 11:34 am
But even if Moody’s, qua company, winds up taking a giant hit over this, it’s still not clear that Moody’s top executives won’t have come out ahead.
The problem, of course, is that – even if the United States takes a hit, Bush and his cronies are going to come out ahead.
Public inspectors are apt to be every bit as prone to corruption and indolence as private.
December 9th, 2008 at 11:54 am
Another point against having investors do their own credit ratings is that CRAs have inside information not available to the public to make appropriate ratings.
Also, another conflict of interest for a public agency doing credit ratings, that hasn’t been discussed is sovereign and municipal debt. Having State pressure the credit rating agency because the Russians don’t like their credit rating and use it as leverage during negotiations over the placement of interceptor missiles will complicate factors further. Also, what happens if some Congressperson or Senator doesn’t like the credit rating for a municipality in their district or state (or even a company in their district or state*). Making it a government agency is going to eliminate conflict of interests and will exacerbate others.
* Say Dingell and one of the Big 3.
December 9th, 2008 at 12:31 pm
There’s no need for the CRA’s to be public agencies. The body to which issuers apply for certification (paying a fee) could be a public agency, and that agency could randomly assign issues to one of several private rating services (which the public agency would also oversee/regulate and license in some sense). Analogies exist for some safety certification processes overseen but not entirely administered or fulfilled by the FAA.
December 9th, 2008 at 12:45 pm
“For example, Mo and I are talking about ways to get the buyers to pay for ratings. In theory I don’t see much potential for conflict there: buyers should want the best possible ratings of what they are buying.”
On the contrary – most investors would want the lowest possible rating for a given bond, regardless of its actual credit quality, so that there would be much more chance of an upgrade than a downgrade. Similarly an investor-pays model means there would be even more pressure not to downgrade underperforming deals. I’m not saying the investor-pays approach wouldn’t be better, just that it doesn’t remove conflicts of interest totally. There will always be conflicts when people are paid to analyse things (witness the absurd kerfuffle recently about Goldman Sachs and California). The question is how to best manage those conflicts and minimise the potential for corruption.
Your premise is only true if the agencies are being paid on the basis of a percentage of transaction revenues. Accordingly, this is a potential point against Mo’s proposal, but not necessarily against a public agency.
True enough. I don’t have a problem with a public agency as a “second opinion” on all deals, but I’m skeptical that the market would pay it much attention. I speak to a lot of investors and none of them – not a single one – wants the government to take on credit ratings. Investors tend to be pretty staunchly anti-government types.
But for none of those things is a given institutional investor uniquely situated to provide evaluations.
Nor are the credit rating agencies, which is why nobody should rely on them as more than just one opinion. Even if investors outsource those evaluations, they have to decide who to outsource them to. Like I said before, credit ratings address only credit risk (and for that matter only one aspect of credit risk), and they’re only one institution’s assessment of that risk. How can you possibly serve your clients well if you don’t have an opinion on when they’ll get their money back, or what will happen to it if interest rates fall by a certain amount?
December 9th, 2008 at 1:29 pm
Someone discovered that the last three bussiness cycles were 8 years long, which may mean that Presidents became somewhat adept in postponing the date when economic shit hits the fan.
What happened in the last cycle is that it started very anemically, and then the only branch of economy that was expanding in a way to lift the rest was residential real estate and, to a lesser extend, consumer finance. There were many actions of regulators and Fed that helped to extend it. Relaxing the capital requirement ratios and of the supervision of loan quality was not the accident but a decision to extend the duration of the bussiness cycle.
December 9th, 2008 at 1:30 pm
DTM, actually there is a fairly clear, fairly wide, and fairly bright line between what I described and public agencies all the way down. I am, however, grateful for your generous assessment of the consistency of my suggestion with the spirit of Matt’s initial inquiry and proposal.
December 9th, 2008 at 1:58 pm
DTM — think banks (depository institutions), vis a vis the federal agencies that regulate (and insure, which may be irrelevant, or maybe not) them.
I think it’s an overstatement to say that the CRA’s would be clients of the government. The allocation of issues to CRAs would be a pass through involving the goverment, and fees would largely be fixed (depending on complexity of the issue) but the clients/stakeholders of the CRAs would be the investing community. The firms wouldn’t be as razzle dazzle as a hedge fund, but they wouldn’t be all that different from the CRAs of not so long ago either.
December 9th, 2008 at 2:08 pm
Well, the problem with your line of thought is that the issuers were forced to buy the services of one of the few credit rating agencies. That´s why they didn´t lost all their clients after the tech buble.It was not that investors or issuers believed on the ratings. They didn´t. Nobody did.
It is weird that the whole “deregulation caused this crisis” crowd seems to not undersdand this aspect of the problem. The credit rating agencies were a product of regulation. They failed miserably time and time again, but, regardless, remained in bussiness. Because they had nothing to do with free markets and all to do with statist coercion.
December 9th, 2008 at 3:15 pm
Vlad, the level of ire that has been directed at the rating agencies (both publicly now and in private (now publicly available) messages and correspondence over the past few years) suggests that many people and firms relied on agency information and regarded it as something that should be reliable, at least in a relative sense.
December 9th, 2008 at 3:32 pm
The investors are the ones anticipating profit from the right investment decisions, and shouldn’t be subsidized by anyone else.
IMO the best answer is for individual investors to pay for ratings from a firm that does not get compensation from anyone else. One example – TheStreet.com Rating Service (formerly Weiss Ratings) has been around for years and will sell you individual ratings as well as annual comprehensive access packages. I don’t know whether there are other rating services like this, and I can’t speak for its performance, but this is the one business model that absolutely depends on getting the ratings right from the investors’ viewpoint.
It doesn’t cover bonds (but somebody else could develop such a service) or exotic derivatives (no loss there). It does rate banks, HMOs, and insurance companies.
December 9th, 2008 at 5:09 pm
How about throwing the book at them for fraud and racketeering and taking all their money? Then reinstate Glass-Steagall (sp?) and whatever else has contributed to this fiasco. This crisis didn’t happen in a vacuum.
December 9th, 2008 at 5:27 pm
Except the CRAs would not be a government agency. The SEC is a government agency so conceptual headflips can be done to say “the investing community” is the SEC’s “client.” But since the SEC is in fact a government agency your analogy is empty. Perhaps you have one to the behavior of some actual “de facto government agency” that you might have in mind?
Might the CRAs be held, with some sort of firewall, as part of some larger holding company or entity? sure, why not? shareholders? sure, why not? managers as agents of shareholders? sure, why not?
The key is uncoupling the transaction between issuer and CRA, which could be done by random allocation of issue needing rating to CRA, with set fees (which might pass directly to the CRA, or might pass through the government agency). This puts the CRA in the position of running a cost minimization operations, subject to quality standards set and monitored by the goverment agency.
The questions you are raising are not idle ones, but they are not particularly troubling, surprising or unusual ones either. Given the revenue structure, these CRAs would probably be organized differently than at present. Would the organization be more like what prevailed several decades ago, when issuers did not purchase ratings? Perhaps.
December 9th, 2008 at 8:05 pm
DTM — The government role in what I describe is primarily to distribute issues to CRAs for rating, so that a significant avenue for CRA self dealing would be stymied. An automated random distribution device would work as well. Whether individual CRAs would develop reputations for unique approaches or behaviors, like judges, remains to be seen. No doubt this would be a sleepy sort of business, compared to many in the financial sector. But there’s no reason it couldn’t be run as a profit seeking sleepy private business. There are sleepy community banks and (until recently) sleepy municipal bond insurers. Whether the problem Ginger Yellow identifies is insurmountable would have to be thought through. I don’t know that it is that big a deal. You have laid out sources of complexity and imperfection in the world, and no doubt what I propose is not perfect — investors would still have to use their own noggins a bit, and risk would not disappear. That’s a standard I can live with.
I think your Justice example is a red herring — I’ll let it lie there and stink up the place, although perhaps vlad would discuss it with you. Maybe you have an example from one of those “de facto government agencies” you mentioned, and that would help clarify your concerns?
December 9th, 2008 at 8:19 pm
Well, sure, and if I am a ratings agency, I just rate the issue high. The person paying me then is happy, and you, the purchaser, are out of luck. Oh wait, that _is_ what happened. To some extent, I agree, I should be able to take risks on my own. If we do that, the credit market will be alot more frozen than it has been in recent history. That doesn’t seem positive.
Maye the answer is to give up on ratings (and ratings agencies) all together. Those that want them (pension funds) can do their own rating. That is a fair alternative, but would, I suspect, lead to a vastly worse credit market for midsize firms. That might be OK, as banks would then work with small companies, and the bond market would only work for the very large companies. Is that a good or bad answer for the economy?
If the bond agency has to insure losses at some rate, they then have an incentive to rate as low as possible. They still get paid by the entity selling the bonds to rate it, so they have an incentive to rank high.
The bond agency pays out (insurance) if they underprice the risk, and they pay (by not getting work) if they overprice the risk, and because they have to place collateral against the insured portfolio. Seems to me that they then have an incentive to project accurately. They could even bid a suggested interest rate to the companies. Seems like it should work, and it is a farily reasonable system.
I am missing something, since what I described is a combination CDS clearinghouse/bond rating agency, but maybe that would make better sense of the CDS market, as well.
December 9th, 2008 at 8:21 pm
Well, except that there are barriers to entry in becoming a credit rating agency. Credit rating is not a competitive business. I am not sure what effect that has, but it surely means that it was not a ‘free market.’ But maybe we could never have one anyway, so I am not arguing against regulations.
There is another source of information for these securities-the bond yields themselves, which were higher than riskless bonds. In fact, the magic was you could get a high(ish) yield and a AAA rating too! So people knew these things were risky. I have seen no reports on the actual return distributions relative to the risk. Were people really paying too much if you compute the returns? Who knows.
And finally, bond ratings are to really help the buyers cover their asses if things go wrong, not to provide information. It’s working too. No one seem to blame the buyers of the securities. I wonder why.
December 9th, 2008 at 11:24 pm
You guys do know the SEC already regulates and has the power to regulate processes regarding record-keeping and deal with conflicts of interest. It’s called the Credit Rating Agency Reform Act of 2006 and predates the crisis.
You’re astonishingly ignorant. I’m guessing you’ve never taken a course on bonds, read about bonds outside of newspaper articles or taken a finance class beyond Finance I. The historical default rate by credit rating is a pretty basic and well known table. The table I linked to above is relatively old, the one in my bond valuation course is more up to date, but I don’t own a scanner. It has similar, though more granular info.
Rating these exotic derivatives AAA was amazing dumb, simply because there was no past history to base these ratings on. However, vanilla bond ratings are pretty well understood and those rating generally turn out pretty well. BTW, if you don’t like S&P’s rating (for a vanilla bond) you can’t just dump them for Moody’s. Most investors require at least two ratings from a NRSRO.
December 10th, 2008 at 8:27 am
Amtrak decisionmaking is subject to all manner of stupid political constraints from members of Congress keeping hugely unprofitable routes open, and as a rail service faces big problems because it owns or controls relatively few of the rails it operates on — luckily it does have right-of-way in one of the few US markets where rail makes some sense — the Boston-NY-PHL-Washington corridor — and where the trains, both Acelas and regional lines, tend to be pretty full, in my experience. I’ll miss seeing Joe Biden from time to time.
Silly example, except perhaps as a set of warnings for organizing my proposal.
December 11th, 2008 at 10:20 am
FYI, the European Fund and Asset Management Asssociation just put out industry guidelines to address overreliance on ratings. Foremost among them: “Where asset managers do not have the necessary competences in relation to any type of [securitisation] they should refrain from buying and managing such investments (subject to any specific instructions).”
February 17th, 2009 at 7:28 pm
Hey very nice blog!! Man .. Beautiful .. Amazing .. I will bookmark your blog and take the feeds also…
February 17th, 2009 at 7:36 pm
I keep listening to the news speak about getting free trial of acai berry so I have been looking around for the best site to get one.
March 4th, 2009 at 7:23 am
Wow I know more about botullisim than I ever wanted to know.
March 4th, 2009 at 8:02 am
Holy shit number 12 is incredible!
March 5th, 2009 at 3:14 pm
amazing stuff thanx
March 11th, 2009 at 4:55 am
Very interesting site. Hope it will always be alive!
March 17th, 2009 at 2:31 am
Very interesting site. Hope it will always be alive!
tramadol
March 22nd, 2009 at 6:25 am
tramadol
I bookmarked this site. Thank you for good job!
April 2nd, 2009 at 5:33 am
Excellent site. It was pleasant to me.
buy cheap viagra
April 3rd, 2009 at 4:23 am
Excellent site. It was pleasant to me.
cheap brand pfizer viagra
April 9th, 2009 at 7:03 am
Thanks for the review! viagra