The Market is Dead! Long Live the Market!

Sudhir Venkatesh‘s book “Gang Leader for a Day,” originally published last January, is now out in paperback. You can read reviews of it here, here, and here; and The Economist named it a book of the year.

Robert Rosenkranz has proposed means of financial market regulation. His Wall Street Journal op-ed offers redress for the abysmal behavior of credit agencies, those once-heralded assessors of risk whose reputation now lies somewhere underneath a common garden snake. If you listen to public and media criticism, these agencies are more than partly responsible for mistakenly valuing the “worthiness” of various structured financial products that saturated the bond market in the past decade.

With some hubris, and a touch of brazenness, Rosenkranz says that we don’t need these agencies at all: get rid of all independent ratings, and let the market assess risk.

The credit worthiness of a bond rating is a letter grade of sorts — “AAA” being the most secure, “C” signaling substantial risk, and “D” being in default. Seven ratings agencies, recognized by the Nationally Recognized Statistical Ratings Organization and overseen by the SEC, grade various forms of corporate and public sector debt issuances, the most prominent being Moody’s, Standard and Poor’s, and Fitch. These firms have taken quite a hit recently for failing to measure risk responsibly and thereby providing investors effective counsel.

Rosenkranz says the rating agencies need not exist at all for investors to understand the risks of betting on various financial products. In his proposal, the market could function as an advertiser of risk by simply publicizing bond spreads: as the spreads widen, the risk grows. Q.E.D. And as the risks grow, so too does the obligation of the bond holder to shore up the note. Q.E.D. again.

He writes: “The amount of capital required to hold a fixed-income security should be determined not by a rating but by its yield, expressed as a spread over Treasurys. The higher the spread, the riskier the market has determined the asset to be, and more capital should be required to hold it.”

Whatever we think of the proposal, Rosencranz points to a structural imperfection in the ratings market — one that precludes true competition.

Think of it this way: why don’t more companies seek out the business of rating debt? The payouts for ratings agencies are not chicken feed. The market has grown enormously since the 1990’s. Where once ratings firms primarily counseled subscribers about the risks of fairly straightforward corporate debt, they now help issuers of debt attain AAA ratings. I use “help” advisedly, since we have learned how they coach their clients into presenting the products in such a way that “AAA” is almost a certainty. See, for example, Gretchen Morgenson‘s pithy assessment: “The woefully inaccurate ratings that have cost investors billions were not, mind you, a result of issuers paying ratings agencies handsomely for their rosy opinions …” Her writings in The Times have been some of the best.

The usual answer is that the top firms have built up decades of trust in the financial community. In non-crisis periods, it would be difficult for a newcomer to enter the ratings market because corporate boards seek out the top firms to legitimize their issues, pension fund managers look to these firms for seals of approval, and so on. But this could change, no? Couldn’t a newcomer build up trust over time by demonstrated performance? We don’t have a limited supply of trusted lawyers or economists, for example. So, I’m not sure why the advice on bonds must be so precious that the market works like a cartel.

But we are obviously in a crisis period. One wonders whether the future will see new companies entering the fray by promising more honest, truly objective valuations. Notwithstanding the complexity in evaluating the risk of innovative structured financial products, it seems to me that a few recently laid-off physicists and economists might come together to design better ratings formulas. Otherwise, we will undoubtedly see the same folks running the show. This may be fine in a world where classic corporate bonds predominate, but I’m not expecting structured finance to go the way of the dinosaur.

So we return to Rosenkranz’s proposal for market self-regulation. Can spreads on yields adequately measure risk?

One problem may be the potential for manipulation of the bond market — much like the way Soros and others moved currency markets in their favor. What will stop an investor from influencing the spread on a bond in such a way that tremendous pressures are placed on the holder to find capital quickly (not so easy in these illiquid times)? As long as this isn’t declared illegal, I suppose nothing is wrong. (No one accused L.T.C.M. of being unseemly.) But we could see such high volatility as trust becomes further eroded and lending becomes unpredictable and riskier.

I’ve never believed in the free market myth, but faced with the choice of letting the usual-suspect ratings agencies continue misfiring, I’m open to letting the markets rule themselves.


First off, I agree with Gang Leader for a Day. Bought it. Read it. Loved it.

And being libertarian, I agree with this article... Thanks Sudhir!

Imad Qureshi

I disagree. This article ignores a lot of other facts on why these rating agencies are needed. Yes they failed in rating mortgage back securities but what about everything else they have done so far.


Fantastic idea! People rely way too much on uninformative indexes and telltales when making investment decisions.

Now that we've abolished those particular anchoring heuristics, let's move on to the DJIA, the S&P 500, the FTSE, the DAX... oh, wait, you mean people demand and pay good money for others to obscure and repackage data into easily-digestible but non-informative snack-nuggets?

Well, it was a great theory anyway.


I could be wrong, but I thought that Moody's and the rest of the rating companies were, in fact, themselves products of the free market. After all, they did not just pop out of a vacuum, and they were not imposed on markets by government.

The ratings companies evolved naturally, as private-enterprise solutions to the problem of assessing market risk. The rest of the markets trusted them, a trust that evolved over decades, and collectively looked to them for guidance.

If the ratings companies -- and the trust the markets had in them -- were not products of the markets themselves, I don't know what would have been. Can we be sure any subsequent solution the markets come up with, will prove to have been any more valid?

Mike B

I never understood why such a critical market function was left in the hands of profit seeking individuals. Where profits are involved impartiality will simply fly out the window.

Credit rating needs to reside within the domain of a Federal Agency or a Federally Chartered non-profit corporation. They don't need a statutory monopoly like the Post Office, but for various regulatory requirements a sufficient rating from the impartial judge would be required. The fee schedule for being rated would be fixed and transparent and there would be a single transaction type, a debt issuer pays for a rating and they get a rating in return.

The other possible solution would be to have NIST or whomever issue and maintain a set of standard rating guidelines and algorithms. These would be open to public inspection and competitive modification so that anybody with the required information and a personal computer could plug in their own numbers and get a NIST standard rating. In the Information Age, the information that once took rating's agencies expertise to dig up and warehouse is often available to everyone at low cost. With a set pf open ratings criteria everyone with access to all available public information should be able to generate their own ratings. Ratings would be comparable across several sources and any discrepancies could be attributed to information asymmetries that could then be investigated by the market.



I have to agree with Psi corp. They are products of the free market in the credit space just as Morningstar is a product of the free market for equities. Being able to outsource analysis seems like free market capitalism to me..


As anybody who has Dan Ariely's "Predicitbly Irrational" knows, human beings are not particularly good at making judgemets regarding risk nor are they very rational.

Free markets only work properly if participants have perfect information and make rational decisions. In the real world, however, people rarely have perfect information and are highly irrational, something which proponents of free markets tend to overlook.

The ratings agencies aren't great, but at least they give people more information than they would have otherwise.


Lower than a garden snake? Garden snakes eat rats and mice, thereby serving a useful function. Even the venomous snakes perform vermin control, the very action the credit agencies no longer participate in.

Justin James

I am constantly amazed at this inane belief that the problem with everything is a lack of "free markets". PUH-LEEZE. The "free market" has done a really LOUSY job of assessing risk. Anyone with half a brain could tell that credit default swaps, derivatives, etc. were extraordinarily risky ideas. Anyone who has read more than 2 pages in a textbook knows that reward is directly related to risk, so when someone offers a 50% yield in an area where the average is 5%, you have to know that the risk is through the roof.

As #4 points out, these ratings agencies are indeed the products of the free market. The free market made it more profitable to deliver bogus results than accurate results. The fact that so many people were involved, and that we have been hearing about issues with the credit agencies for about a decade, tells me that anyone who actually beleived the credit agencies may be interested in a bridge I have to sell at a very reasonable rate.

No, the problem here, is that the financial community is essentially in unspoken collusion with itself. We pretend that the credit ratings are accurate (even though we know they are not), we pretend our investments are not risky (even though we know they are), we pretend that bubbles can grow forever (even though we know they can't), and to justify it, we'll hire some really smart folks with really powerful computers to juggle it all.

Remember: if you can't understand it, it is a risky investment!




The rating companies were products of a free market until the market was cartelized by the SEC: Only a select few were granted recognition as NRSROs.


@ #7 mfw13: Unfortunately, the information the agencies gave investors was worse than outright lies. They gave information investors *thought* they understood. In some cases, it was worse than that. The ratings agencies gave ratings based on information they *thought* they understood.

@ # 9 Justin James: Actually, CDS and derivatives, properly used, are instruments designed to *reduce* risk. Unfortunately, the people underwriting them were poorly incentivized, to wit: they only cared about writing the contract and never analyzed its potential consequences if/when it was called. Multiply this by thousands of transactions and trillions of dollars and you have "the credit crunch."

Gavin Andresen

Responding to psiCop's comment and expanding on lukas' :

Rating companies created by the free market are well and good; the real problem is that government regulations were tied to the ratings. Which created perverse incentives for both the ratings agencies and the banks (and maybe a false sense of security on the part of investors, but that's more debatable).


I heard a talk recently here at IU about this exact issue and was surprised to learn that credit ratings used to be paid for by the subscribers/investors, not the companies being rated (as it is now). I personally think more federal regulation is probably required, plus a reorientation of who pays for the information, since that certainly has a distorting effect.


Ratings agencies are the sole culpable party in the mortgage crisis. Everyone else was doing their job, and the system broke down because these guys gave investors a false sense of security.

Blaming lenders who borrowed more than they can afford is ridiculous, because who in their right minds wouldn't take free money?

Blaming mortgage originators for "predatory lending" is ridiculous, because if ratings agencies had properly assessed the risk of these securities, originators that originated risky debt would not have been able to resell it on the secondary market and soon gone under.

Blaming the securitization groups at the investment banks is ridiculous, because they weren't telling investors how much to pay for the securities - these prices were dictated by market demand, and if investors had been properly forewarned by the ratings agencies, then the ibanks wouldn't have had nearly as much business.

I could maybe see an argument for investors being idiots and falling for a false sense of security that was provided by the ratings agencies.

Ultimately, ratings agencies artificially inflated demand by failing to do their jobs. The incentives are completely out of whack (and I haven't heard of any serious proposal to change the incentive structure). The ratings industry is a sham and I can't believe it isn't under more scrutiny right now.

I think allowing market forces to dictate ratings is an excellent idea, with a couple of reservations:

1. The extremely high bid/ask spreads on CDS right now indicate that in times of stress, market forces may not be great at calculating. Also, my understanding is that CDS spreads are primarily used to hedge against counterparty risk - so the final price may include a risk premium for companies that have clients with high counterparty risk, and not reflect the risk neutral view of default.
2. There needs to be more regulation around CDS - what good is it if the seller goes belly up? Needs to be some kind of reserve requirement to make the insurance contracts exactly that, regardless of the underwriter.


Bobby G


Anyone? Anyone..?


I don't agree with some of the ideas that are spewed out of WSJ. The idea that Market can assess risk correctly and we can let the market assess the risk, really makes me to ask the questions - What's your cool aid..

There is seems to be no reason the rating agencies should be broken up and ensured that their business models is attuned to getting correct advise to the investors and not to the companies. The critical problem was the issue of 'Conflict of Interest' and knowledge about assessing risk for MBS.

A better way may be to ensure that Rating Agencies have two different divisions -
a) Retail
b) Corporate

The retail investing arm would be able to give accurate information on the performance of bonds AAA, AA+, AA- to the retail investors who seek this information. They would sell this as a monthly subscription fee with upside/downside from individual investors.

The Corporate Arm can consult, create and provide specialized ratings for Companies, Asset Management companies.

The two revenue streams would be different and hopefully there would less conflict of interest.

Obviously, given the current state of the Market and its predictabilty, volatility , such indexes, does provide indication of the direction the market is going it hardly tells the investors how hard it is recovery is going to be.



How 'bout the guver'mint creates a Rating Agency Rating Agency. They would compare each agency's past ratings to the actual performance of the bonds and post that info on The agency would be called RA-RA, and they cheer on the companies that do a good job.


The markets decided to rely on ratings agencies to assess risk. Investment firms thought ratings agencies did a decent job. Markets made the mistake of trusting ratings agencies in the first place, why should we trust them again?

Stephen McNamee

There are a whole bunch of problems with ratings agencies as defacto regulators but the system worked until now. The major problem that I have observed in reading the SEC act is that broker dealers with more then 5 billion dollars in assets were allowed to regulate themselves. In 2004 the securities exchange commission made changes to the long standing net capital rule. The net captial rule was established in 1975 and was the legislation that changed the buisness model for ratings agencies from independant firms into defacto regulators. The net capital rule also established a maximum numerical limit on leverage. Broker dealers were not able to leverage more then 15 times their assets.

In 2004 an exception to this maximum leverage clause was made for broker dealers with more then 5 billion dollars in market cap. The five firms large enough were Goldman Sachs, Merril, Lehman, Morgan, and Bear Sterns.These broker dealers quickly increased their leverage until the system collapsed. Merril had leveraged itself 40-1 and bear was 33-1 when they went under. This is what really allowed the huge bubble in these toxic, poorly rated assets. This exception must be removed quickly. It is the least that could be done to make regulation more effective.

I look forward to the new Presidents suggestions on how to change the regulatory structure.


Imad Qureshi

@Archana #13: I heard a talk recently here at IU about this exact issue and was surprised to learn that credit ratings used to be paid for by the subscribers/investors, not the companies being rated (as it is now).

Archana you don't pay Transunion, Experian and Equifax to rate you. When you apply for credit and bank pulls your credit report they pay these companies to see your credit. Why should it be otherwise in case of companies?