Returns to Information and Returns to Capital

One of the benefits of maintaining this blog is that it gives me the opportunity to think aloud, expressing half-formed ideas in the hope that the feedback will help me sort through some interesting questions. My last post on double taxation attracted a number of thoughtful (and in some cases skeptical) comments for which I am grateful.

What I was trying to do in that post was to evaluate two incompatible statements: Warren Buffet's declaration that he pays a substantially lower tax rate at 18% than any of his office staff, and Mitt Romney's conflicting claim that his effective tax rate is close to 50%, the sum of the corporate tax rate and the rate on long-term capital gains. I argued that since the corporate tax is capitalized into prices at both the time of purchase and the time of sale, it ought not to be simply added to the capital gains tax to determine an effective rate.

The point may be expressed as follows. Over the past couple of years Romney seems to have paid about 3 million dollars in taxes on income of about 20 million annually, a rate of about 15%. If his effective tax rate is 50% then his "effective" gross income is about twice his current after-tax income, or approximately 34 million. What he is claiming, in effect, is that in the absence of the corporate tax, and with no change in the nature of his economic activities, he would have been able to secure a capital gain of 34 million annually. This does not seem plausible to me. Elimination of the corporate tax would certainly result in a one-time gain to any currently held long positions, but I don't see how it could allow him to generate an extra 14 million, which is 70% greater than his current gross income, on an ongoing basis every year.

Whatever the merits of this argument, I think that most commenters on my earlier post agree with me on two things:

1. The adding-up approach to effective tax rates does not work for short sales and related derivative positions, since it would lead to the absurd conclusion that short sellers were paying a negative effective tax rate on capital gains.
2. Elimination of the corporate tax would result in a sharp rise in equity prices and a windfall gain to current long investors, but would have more modest and uncertain effects on the returns to future investors who enter positions after the lower rate has been capitalized into prices. 

In particular, the following comment from Richard Serlin got me thinking about the nature of capital gains:

With regard to short selling, when the corporate tax first hits (or becomes known to hit), they'll get a windfall, but then their expected returns (of the short sales people actually choose to take) will adjust to the new norm for their risk. It's not like short selling opportunities that pay a fair market risk adjusted return always exist, anyway. When they do, it's largely not a reward for the capital, but for the information that the stock is an overpriced bad deal.

It is certainly true, as Richard points out, that profits to short positions are rewards for information, broadly interpreted to include the processing and analysis of information. They are not returns to capital in any meaningful sense, although one requires capital to enter a short position. But the same is true for at least some portion of the profits to long positions. In fact, the essence of Buffet's investment strategy is to identify underpriced companies in which to take long (and long-term) positions on which capital gains are then realized.

If capital gains are viewed largely as a return to capital, then the double taxation argument makes some sense. But viewed as a return to information and analysis, it is not clear why capital gains should be given preferential tax treatment relative to the income generated, for instance, by doctors or teachers.

I suspect that Warren Buffet views his income as being generated largely by information and judgment, and does not believe that his opportunities for ongoing capital gain would be substantially increased if the corporate tax were eliminated. He does not therefore see the tax as a significant burden, and does not consider his effective gross income to be substantially greater than that which he declares on his tax returns. Whether Romney himself feels the same way is impossible to know, since political expediency currently compels him to take a very different position. 

Double Taxation

The release of Mitt Romney's tax returns has drawn attention yet again to the disparity between the rates paid on ordinary income and those paid on capital gains. It is being argued in some quarters that the 15% rate on capital gains vastly underestimates the effective tax rate paid by those whose income comes largely from financial investments, on the grounds that corporations pay a rate of 35% on profits. Were it not for this tax, it is argued, dividends and capital gains would be higher, and so would the after-tax receipts of those who derive the bulk of their income from such sources.

Romney himself has made this argument recently, claiming that his effective tax rate is closer to 50%:

One of the reasons why we have a lower tax rate on capital gains is because capital gains are also being taxed at the corporate level. So as businesses earn profits, that's taxed at 35 percent. Then as they distribute those profits in dividends, that's taxed at 15 percent more. So all total, the tax rate is really closer to 45 or 50 percent.

The absurdity of this claim is clearly revealed if one considers capital gains that accrue to short sellers, who pay rather than receive dividends while their positions are open. Following the logic of the argument, one would be forced to conclude that short sellers are taxed at an effective rate of negative 20%, thereby receiving a significant subsidy due to the existence of the corporate tax. The flaw in this reasoning is apparent when one recognizes that asset prices are lower (relative to the zero corporate tax benchmark) not only when a short position is covered, but also when it is entered.

There is no doubt that the presence of the corporate tax depresses the price of equities, but it does so both at the time of purchase and at the time of sale. If there were no corporate tax, dividends and capital gains per share would certainly be higher, but an investor would have paid substantially more per share to acquire his assets in the first place. As a result he would be holding fewer shares for any given initial outlay, and his after-tax income (holding constant the rate paid on capital gains) would not be substantially different.

To see why, it is useful to think about what determines the price of equities. Three factors are especially important: the current earnings of a firm (after payment of interest and taxes), the rate at which these earnings are expected to grow, and the riskiness of the security, which itself is linked to the degree to which the firm's earnings are correlated with broader market movements. Securities that are riskier in this latter sense tend to appreciate faster on average because investors would otherwise avoid them, depressing their prices and raising their expected returns until such returns are viewed as adequate compensation for the greater risk of holding them. This risk is routinely expressed as a market capitalization rate, interpreted as the expected return that investors require in order to hold the security. Airline and automobile stocks, for instance, have higher market capitalization rates than do shares in utilities.

The manner in which these factors interact to influence prices may be illustrated by considering the simplest possible case of a firm with constant expected earnings growth and a fixed dividend payout ratio. In this case, for reasons discussed in any introductory finance textbook, the fundamental value of the security is given by the simple formula D/(k-g), where D is the current dividend forecast (a constant share of the earnings forecast), g is its expected rate of growth, and k is the market capitalization rate. Shares in a debt-free firm that pays 20% of its earnings as dividends, is currently earning $10 per share annually, is expected to grow at 10%, and has a market capitalization rate of 12% would then have a share price of $100. After a year (assuming no change in these parameters) the share price would be $110 and the dividend payout $2. An investor would have made $12 on a $100 investment, a percentage return precisely equal to the market capitalization rate. All this is with no corporate tax.

Now suppose that a 35% corporate tax is in place, so after-tax earnings per share are $6.50 instead, with no change in other specifications. Dividends are then $1.30 per share and the initial share price is $65. After a year this rises to $71.50. Adding dividends and capital gains, an investor makes $7.80 for each share purchased at $65, again earning precisely 12%. Each share results in lower revenues to the investor, but since more shares can be purchased at the outset, aggregate income is no different.

None of this should be in the least bit surprising. Note, however, that if the corporate tax were to be eliminated today, there would be a sharp rise in the price of equities and current asset holders would enjoy a windfall gain. Similar issues arise with respect to the mortgage interest deduction: eliminating this would result in an immediate decline in home values, severely punishing those who purchased recently at prices that reflected the anticipated tax savings over the duration of the mortgage.

This does not necessarily mean that eliminating the corporate tax while simultaneously raising the rate on capital gains is necessarily a bad idea, or that elimination of the mortgage interest deduction is necessarily bad policy. A case could be made for both initiatives. The corporate tax is not uniformly applied due to the broad range of loopholes and exemptions, and the mortgage deduction is regressive and inhibits both neighborhood integration and labor mobility. But any such changes will have major distributional effects that must be taken into account in any comprehensive evaluation of the policy. Doing so properly requires a clear distinction between stocks and flows, and an analysis that goes a little deeper than simple arithmetic.

---

Update: Follow-up post here.

Self-Fulfilling Prophecies and the Iowa Caucus

A few days ago Nate Silver made the following intriguing comments on the Iowa Caucus (emphasis added):

There are extremely strong incentives for supporters of Mrs. Bachmann, Mr. Santorum and Mr. Perry to behave tactically, throwing their weight behind whichever one appears to have the best chance of finishing in the top two. What that means is that if any of these candidates appear to have any momentum at all during the final week of the campaign, their support could grow quite quickly as other voters jump on the bandwagon.

This is also a case in which the polling may actually influence voter behavior. In particular, if one of these candidates does well in the highly influential Des Moines Register poll that should be published on New Year’s Eve or thereabouts, that candidate might be a pretty good bet to overperform polling as voters use that as a cue on caucus night to determine which one is most viable...

I’m not sure that this theory actually makes any sense... But it may not matter if the theory is true. If voters are looking for anything to break the logjam between these candidates, mere speculation that one of them has momentum could prove to be a self-fulfilling prophecy.

What's most interesting about this is the possibility that even a methodologically flawed or misleading poll, provided that it is given credence, could coordinate expectations on one of these three candidates and result in a surge of support.

In fact, this seems to be precisely what has happened. A CNN/Time poll covering the period December 21-27 revealed Santorum to be in third place with 16% of the vote. This was an outlier at the time, and was sharply criticized by Tom Jensen of PPP and by Nate himself for surveying only registered Republicans:

What’s wrong with using a list of Republican voters for a Republican caucus poll? The answer is that it’s extremely easy for independent and Democratic voters to register or re-register as Republicans at the caucus site. Historically, a fair number of independent voters do this.

According to entrance polls in Iowa in 2008, for instance, about 15 percent of participants in the Republican caucus identified themselves as independents or Democrats on the way into the caucus site... Most other pollsters are making some attempt to account for these voters. They are anticipating that the fraction of independents and Democrats will be at least as high as it was in 2008 if not a little higher, which would make sense since Republicans do not have a competitive Democratic caucus to compete with this year.

The recent Public Policy Polling survey, for instance, estimated that 24 percent of Iowa caucus participants are currently registered as independents or Democrats and will re-register as Republicans at the caucuses. This month’s Washington Post/ABC News poll put the fraction at 18 percent. There is room to debate what the right number is but it will certainly not be zero, as the CNN poll assumes.

Since few independents and Democrats are inclined to vote for Santorum, the CNN/Time poll very likely exaggerated the level of support he enjoyed at the time. But despite this, it contributed to expectations of a surge which seem to have become self-fulfilling. The Des Moines Register poll released last night confirms this, with Santorum rising sharply from 10% on the 27th all the way to 22% four days later. This survey, conducted by the highly regarded Ann Selzer, has historically been among the most reliable of Iowa polls.

Did a misleading poll based on an unsound sample shift expectations in such a manner as to fulfill it's own flawed forecast? Tom Jensen certainly appears to think so:

Selzer had Santorum at 9% Tu-W. We had him at 10% M-Tu. Surge quite possibly generated by CNN poll that was quite possibly wrong... If CNN had shown Perry at 15% and he got all the momentum stories, the buzz in Iowa might be all about him this weekend.

The CNN/Time poll may also have given Romney an expectational boost at the expense of Paul by excluding independents from the survey. As Tom Jensen noted in his response, Romney was ahead of Paul in the restricted sample of the PPP poll, but quite clearly behind overall on December 27. It's an interesting example of how a seemingly innocuous methodological decision on a single primary poll could end up having major ramifications for the direction of the country.

The mechanisms at work here have some broader implications. They reveal the potential value to candidates (or their supporters) of manipulating prices in prediction markets such as Intrade, which have come to be closely monitored indicators of candidate viability. And they appear in all sorts of other contexts, from sovereign debt crises to speculative currency attacks.

In fact, any borrower who has financed long-dated assets with short term liabilities needs to periodically roll over debt, and the willingness of investors to facilitate this depends on their beliefs about whether other investors will continue to facilitate it in the future. These expectations are subject to capricious change, often as a result of small and seemingly unimportant triggers. The Iowa caucus illustrates the phenomenon, and the Eurozone debt crisis demonstrates its broader relevance.

Price Coherence on Intrade

A couple of days ago, Richard Thaler tweeted this:

Intrade prices seem incoherent. How can Newt nomination price soar but Obama win stay at 50%?

Here's what Thaler is talking about. Over the past couple of weeks, the price of a contract that pays $10 in the event that Gingrich is nominated has risen sharply from about a dollar to above $3.50:

Over the same period a contract that pays $10 if Obama is reelected has remained within a narrow window, trading within a ten cent band a shade above $5:

Thaler considers this pattern to be incoherent because Gingrich is widely believed to be a weaker general election candidate than Romney. For instance, in head-to-head poll averages Obama currently leads Gingrich by 5.7%, but leads Romney by the much smaller margin of 1.5%.

But even if Gingrich really is the weaker candidate against Obama under any set of conditions that might prevail on election day, it does not follow (as a point of logic) that a rise in the Gingrich nomination price must be associated with a rise in the Obama reelection price. For instance, a belief among voters that Obama is more vulnerable would ordinarily result in a decline in his likelihood of reelection, but this could be offset if the same belief also leads to the nomination by the GOP of a more conservative but less electable candidate.

This reasoning is consistent with the so-called Buckley Rule, which urges a vote for the most conservative candidate who is also electable. As perceptions about the electability of the incumbent shift, so does the perceived viability of more ideologically extreme members of the opposition. These countervailing effects can dampen fluctuations in the electability of the incumbent. Hence the market data alone cannot decisively settle the question of price coherence. 

Notes on a Worldly Philosopher

The very first book on economics that I remember reading was Robert Heilbroner's majesterial history of thought The Worldly Philosophers. I'm sure that I'm not the only person who was drawn to the study of economics by that wonderfully lucid work. Heilbroner managed to convey the complexity of the subject matter, the depth of the great ideas, and the enormous social value that the discipline at its best is capable of generating.

I was reminded of Heilbroner's book by Robert Solow's review of Sylvia Nasar's Grand Pursuit: The Story of Economic Genius. Solow begins by arguing that the book does not quite deliver on the promise of its subtitle, and then goes on to fill the gap by providing his own encapsulated history of ideas. Like Heilbroner before him, he manages to convey with great lucidity the essence of some pathbreaking contributions. I was especially struck by the following passages on Keynes:

He was not without antecedents, of course, but he provided the first workable intellectual apparatus for thinking about what determines the level of “output as a whole.” A generation of economists found his ideas the only available handle with which to grasp the events of the Great Depression of the time... Back then, serious thinking about the general state of the economy was dominated by the notion that prices moved, market by market, to make supply equal to demand. Every act of production, anywhere, generates income and potential demand somewhere, and the price system would sort it all out so that supply and demand for every good would balance. Make no mistake: this is a very deep and valuable idea. Many excellent minds have worked to refine it. Much of the time it gives a good account of economic life. But Keynes saw that there would be occasions, in a complicated industrial capitalist economy, when this account of how things work would break down.

The breakdown might come merely because prices in some important markets are too inflexible to do their job adequately; that thought had already occurred to others. It seemed a little implausible that the Great Depression of the 1930s should be explicable along those lines. Or the reason might be more fundamental, and apparently less fixable. To take the most important example: we all know that families (and other institutions) set aside part of their incomes as saving. They do not buy any currently produced goods or services with that part. Something, then, has to replace that missing demand. There is in fact a natural counterpart: saving today presumably implies some intention to spend in the future, so the “missing” demand should come from real capital investment, the building of new productive capacity to satisfy that future spending. But Keynes pointed out that there is no market or other mechanism to express when that future spending will come or what form it will take... The prospect of uncertain demand at some unknown time may not be an adequately powerful incentive for businesses to make risky investments today. It is asking too much of the skittery capital market. Keynes was quite aware that occasionally a wave of unbridled optimism might actually be too powerful an incentive, but anyone in 1936 would take the opposite case to be more likely.

So a modern economy can find itself in a situation in which it is held back from full employment and prosperity not by its limited capacity to produce, but by a lack of willing buyers for what it could in fact produce. The result is unemployment and idle factories. Falling prices may not help, because falling prices mean falling incomes and still weaker demand, which is not an atmosphere likely to revive private investment. There are some forces tending to push the economy back to full utilization, but they may sometimes be too weak to do the job in a tolerable interval of time. But if the shortfall of aggregate private demand persists, the government can replace it through direct public spending, or can try to stimulate additional private spending through tax reduction or lower interest rates. (The recipe can be reversed if private demand is excessive, as in wartime.) This was Keynes’s case for conscious corrective fiscal and monetary policy. Its relevance for today should be obvious. It is a vulgar error to characterize Keynes as an advocate of “big government” and a chronic budget deficit. His goal was to stabilize the private economy at a generally prosperous level of activity.

This is as clear and concise a description of the fundamental contribution of the General Theory that I have ever read. And it reveals just how far from the original vision of Keynes the so-called Keynesian economics of our textbooks has come. The downward inflexibility of wages and prices is viewed in many quarters today to be the hallmark of the Keynesian theory, and yet the opposite is closer to the truth. The key problem for Keynes is the mutual inconsistency of individual plans: the inability of those who defer consumption to communicate their demand for future goods and services to those who would invest in the means to produce them.

The place where this idea gets buried in modern models is in the hypothesis of "rational expectations." A generation of graduate students has come to equate this hypothesis with the much more innocent claim that individual behavior is "forward looking." But the rational expectations hypothesis is considerably more stringent than that: it requires that the subjective probability distributions on the basis of which individual decisions are made correspond to the objective distributions that these decisions then give rise to. It is an equilibrium hypothesis, and not a behavioral one. And it amounts to assuming that the plans made by millions of individuals in a decentralized economy are mutually consistent. As Duncan Foley recognized a long time ago, this is nothing more than "a disguised form of the assumption of the existence of complete futures and contingencies markets."

It is gratifying, therefore, to see increasing attention being focused on developing models that take expectation revision and calculation seriously. A conference at Columbia earlier this year was devoted entirely to such lines of work. And here is Mike Woodford on the INET blog, making a case for this research agenda:

This postulate of “rational expectations,” as it is commonly though rather misleadingly known... is often presented as if it were a simple consequence of an aspiration to internal consistency in one’s model and/or explanation of people’s choices in terms of individual rationality, but in fact it is not a necessary implication of these methodological commitments. It does not follow from the fact that one believes in the validity of one’s own model and that one believes that people can be assumed to make rational choices that they must be assumed to make the choices that would be seen to be correct by someone who (like the economist) believes in the validity of the predictions of that model. Still less would it follow, if the economist herself accepts the necessity of entertaining the possibility of a variety of possible models, that the only models that she should consider are ones in each of which everyone in the economy is assumed to understand the correctness of that particular model, rather than entertaining beliefs that might (for example) be consistent with one of the other models in the set that she herself regards as possibly correct...

The macroeconomics of the future, I believe, will still make use of general-equilibrium models in which the behavior of households and firms is derived from considerations of intertemporal optimality, but in which the optimization is relative to the evolving beliefs of those actors about the future, which need not perfectly coincide with the predictions of the economist’s model. It will therefore build upon the modeling advances of the past several decades, rather than declaring them to have been a mistaken detour. But it will have to go beyond conventional late-twentieth-century methodology as well, by making the formation and revision of expectations an object of analysis in its own right, rather than treating this as something that should already be uniquely determined once the other elements of an economic model (specifications of preferences, technology, market structure, and government policies) have been settled.

I think that the vigorous pursuit of this research agenda could lead to a revival of interest in theories of economic fluctuations that have long been neglected because they could not be reformulated in ways that were methodologically acceptable to the professional mainstream. I am thinking, in particular, of nonlinear models of business cycles such as those of Kaldor, Goodwin, Tobin and Foley, which do not depend on exogenous shocks to account for departures from steady growth. This would be an interesting, ironic, and welcome twist in the tangled history of the worldly philosophy.

David Levey on the Ratings Downgrade

David Levey (Managing Director, Sovereign Ratings, Moody's Investors Service, 1985-2004) sent out the following statement yesterday to a number of publications, including the New York Times, Wall Street Journal, Financial Times, and Bloomberg. Since I haven't seen it published anywhere and he has granted permission to freely reproduce it, I'm posting it here (I thank Sam Bowles for forwarding the statement to me):

The recent S&P downgrade of the credit rating of US Treasury bonds is unwarranted for the following reasons: 

1. The US dollar remains the dominant global currency and no viable competitor is on the horizon. The euro is heading into dangerous and uncharted waters while deep and difficult political, economic and financial reforms will be required before the renminbi could become fully convertible for capital flows and Chinese government bonds a safe reserve asset. 
2. US Treasury bills and bonds, along with government-guaranteed bonds and highly-rated corporates, will for the foreseeable future remain the assets of choice for global investors seeking a "safe haven", due to the unparalleled institutional strength, depth and liquidity of the market. Although there are several advanced Aaa-rated OECD countries with lower debt ratios and better fiscal outlooks than the US, their markets are generally too small to play that role. Since ratings are intended to function as a market signal, it makes little sense to implicitly suggest to investors seeking "risk-free" reserve assets that they reallocate their portfolios toward these relatively illiquid markets. 
3. Despite the above positive factors for the US, it is certainly the case that the US long-term debt outlook is deteriorating under the pressure of rising entitlement costs and an inefficient, distortionary tax system. Failure to reverse that trajectory would eventually make a downgrade unavoidable. But the recent discussions signal to me that -- finally -- public awareness of the fiscal crisis is growing and beginning to influence Washington. There is still a window of time -- perhaps as much as a decade -- within which structural reforms to spending programs and the tax system could reverse the negative debt trajectory.
4. The bottom line is that the global role of the dollar and the central position of US bond markets make somewhat elevated debt ratios more compatible with a Aaa rating than is the case for other countries, another version of the US's "exorbitant privilege". But that extra leeway is finite and serious reforms to entitlement programs, particularly Medicare, must be made in a reasonable time horizon. If not, global investors will eventually conclude that our political system is incapable of making the needed changes and turn away from US assets, regardless of the institutional strengths of US markets.

This is consistent with Warren Buffet's view of the downgrade.

Even more interesting than Levey's statement was his preamble, in which he states that he has "no connection with Moody's nor any non-public knowledge of what its analysts think about the rating or what they intend to do" and then adds the following: 

As I see our current situation, the Federal Reserve, with its too-tight monetary stance since the summer of 2008, has allowed nominal GDP to fall far below trend, causing a collapse of output and employment -- as described by the monetary bloggers Scott Sumner, David Beckworth, Bill Woolsey, and David Glasner. Had the Fed acted properly (by, for example, setting a nominal GDP level target) the recession would have been much shallower and fiscal stimulus might not have been undertaken. As it was, the collapse of nominal GDP drove the "fiscal multiplier" to zero, leaving us with more debt and nothing to show for it.

Whether or not the Fed had the capacity and the commitment to have substantially mitigated the recession in the absence of fiscal policy, I'm not qualified to judge. But I remain skeptical that the rating agencies have the ability to evaluate credit risk with greater accuracy than the market itself would do in their absence. Were it not for the fact that capital requirements for financial institutions are set on the basis of their ratings, I doubt that there would be much of a market for their services, or that they would have such visibility and influence. And as far as sovereign debt is concerned, I'm not sure that they provide us with any useful information or guidance.

Rating the Agencies

It's being argued that yesterday's downgrade of the credit rating of the United States government by Standard and Poor's could increase borrowing costs throughout the economy, worsen the burden of debt, retard a recovery that already appears to be faltering, affect political brinkmanship in future negotiations, and further tarnish our national reputation.

Unless, of course, we chose to collectively ignore it, as Dan Alpert recommends:

Effectively – the S&P pronouncement last evening amounted to not much more than a guest in your house telling your children to clean up their rooms “or else.” I don’t know about you, but in my case, at least, I would ask such a guest to apologize or leave. 

But it's difficult to ignore events on which everyone else is lavishing such great attention, and this seems like an appropriate time to examine how these agencies managed to gain such visibility and influence. As Ross Levine notes in his recent autopsy of the financial crisis, this is where we stood forty years ago:

Until the 1970s, credit rating agencies were comparatively insignificant, moribund institutions that sold their assessments of credit risk to subscribers. Given the poor predictive performance of these agencies, the demand for their services was limited for much of the twentieth century (Partnoy, 1999). Indeed, academic researchers found that credit rating agencies produce little additional information about the firms they rate; rather, their ratings lag stock price movements by about 18 months (Pinches and Singleton, 1978).

But then a policy shift occurred that continues to have major ramifications to this day. The SEC provided a special designation to a class of rating agencies and then proceeded to use their opinions as a basis for setting capital requirements. The selected agencies suddenly found themselves endowed with vastly increased market power and a very lucrative business model:

In 1975, the SEC created the Nationally Recognized Statistical Rating Organization (NRSRO) designation, which it granted to the largest credit rating agencies. The SEC then relied on the NRSRO's credit risk assessment in establishing capital requirements on SEC-regulated financial institutions.

The creation of – and reliance on – NRSROs by the SEC triggered a cascade of regulatory decisions that increased the demand for their credit ratings. Bank regulators, insurance regulators, federal, state, and local agencies, foundations, endowments, and numerous entities around the world all started using NRSRO ratings to establish capital adequacy and portfolio guidelines. Furthermore, given the reliance by prominent regulatory agencies on NRSRO ratings, private endowments, foundations, and mutual funds also used their ratings in setting asset allocation guidelines for their investment managers. NRSRO ratings shaped the investment opportunities, capital requirements, and hence the profits of insurance companies, mutual funds, pension funds, and a dizzying array of other financial institutions.

Unsurprisingly, NRSROs shifted from selling their credit ratings to subscribers to selling their ratings to the issuers of securities. Since regulators, official agencies, and private institutions around the world relied on NRSRO ratings, virtually every issuer of securities was compelled to purchase an NRSRO rating if it wanted a large market for its securities. Indeed, Partnoy (1999) argues that NRSROs essentially sell licenses to issue securities; they do not primarily provide assessments of credit risk.

This shift in business model by the selected agencies raised some rather obvious conflicts of interest, since their customers were now issuers of debt who stood to gain from overly optimistic assessments of their credit risk. As is common in such cases, the counterargument was made that the need to preserve one's reputation for accuracy would provide adequate incentives for objective ratings:

There are clear conflicts of interest associated with credit rating agencies selling their ratings to the issuers of securities. Issuers have an interest in paying rating agencies more for higher ratings since those ratings influence the demand for and hence the pricing of securities. And, rating agencies can promote repeat business by providing high ratings...

Nevertheless, credit rating agencies convinced regulators that reputational capital reduces the pernicious incentive to sell better ratings. If a rating agency does not provide sound, objective assessments of a security, the agency will experience damage to its reputation with consequential ramifications on its long-run profits. Purchasers of securities will reduce their reliance on this agency, which will reduce demand for all securities rated by the agency. As a result, issuers will reduce their demand for the services provide by that agency, reducing the agency's future profits. From this perspective, reputational capital is vital for the long-run profitability of credit rating agencies and will therefore contain any short-run conflicts of interest associated with “selling” a superior rating on any particular security.

I have previously discussed some of the limitations of this argument in a different context, and such limitations were clearly evident in the case of the agencies:

Reputational capital will reduce conflicts of interest, however, only under particular conditions. First, the demand for securities must respond to poor rating agency performance, so that decision makers at rating agencies are punished for issuing bloated ratings on even a few securities. Second, decision makers at rating agencies must have a sufficiently long-run profit horizon, so that the long-run costs to the decision maker from harming the agencies reputation outweigh the short-run benefits from selling a bloated rating.

These conditions do not hold, however... regulations weaken the degree to which a decline in the reputation of a credit rating agency reduces demand for its services. Specifically, regulations induce the vast majority of the buyers of securities to use NRSRO rating in selecting assets. These regulations hold regardless of NRSRO performance, which moderates the degree to which poor ratings performance reduces the demand for NRSRO services. Such regulations mitigate the positive relation between rating agency performance and profitability.

This brings us to the role of the agencies in the financial crisis. The rapid growth of structured products provided the agencies with a substantial new source of demand, as well as the problem of assessing credit risk for securities of much greater complexity. Minor changes in modeling assumptions could lead to significantly different ratings for such assets. Nevertheless, there were strong incentives in place for the agencies to act as if they could make competent assessments of credit risk:

The explosive growth of securitized and structured financial products from the late 1990s onward materially intensified the conflicts of interest problem. Securitization and structuring involved the packaging and rating of trillions of dollars worth of new financial instruments. Huge fees associated with processing these securities flowed to banks and NRSROs. Impediments to this securitization and structuring process, such as the issuance of low credit rating on the securities, would gum-up the system, reducing rating agency profits.

In fact, the NRSROs started selling ancillary consulting services to facilitate the processing of securitized instruments, increasing NRSRO incentives to exaggerate ratings on structured products. Besides purchasing ratings from the NRSROs, the banks associated with creating structured financial products would first pay the rating agencies for guidance on how to package the securities to get high ratings and then pay the rating agencies to rate the resultant products.

Other evidence also indicates that rating agencies adjusted their behavior to capture the profits made available by securitization and the design of new structured financial products. Lowenstein's (2008) excellent description of the rating of a MBS by Moody's demonstrates the speed with which complex products had to be rated, the poor assumptions on which these ratings were based, and the profits generated by rating structured products... Indeed, internal e-mails indicate that the rating agencies lowered their rating standards to expand the business and boost revenues... A collection of documents released by the US Senate suggests that NRSROs consciously adjusted their ratings to maintain clients and attract new ones.

The short-run profits from these activities were mind bogglingly large and made the future losses from the inevitable loss of reputational capital irrelevant. For example, the operating margin at Moody's between 2000 and 2007 averaged 53 percent. This compares to operating margins of 36 and 30 percent at Microsoft and Google, or 17 percent at Exxon... Thus, rating agencies faced little market discipline, had no significant regulatory oversight, were protected from competition by regulators and legislators, and enjoyed a burgeoning market for their services... It was good to be an NRSRO.

Levine's bottom line is this:

While the crisis does not have a single cause, the behavior of the credit rating agencies is a defining characteristic. It is impossible to imagine the current crisis without the activities of the NRSROs. And, it is difficult to imagine the behavior of the NRSROs without the regulations that permitted, protected, and encouraged their activities.

Perhaps the time has come to consider a complete overhaul of this dysfunctional system. Withdraw the special designation accorded to the major agencies, so that they compete on a level playing field with new entrants. If they really do have the expertise to make assessments of credit risk that are more accurate than the market, let them build reputation and find clients willing to pay for their pronouncements. Make capital requirements for financial institutions independent of ratings, thus stripping the agencies of their monopoly power and guaranteed sources of income. And in the meantime, greet their pronouncements on sovereign debt not with an anxious wringing of hands, but with a collective yawn.

---

Update (8/10). Andrew Gelman follows up:

Another way to look at this is: Given all the above, those S&P dudes must really really think the U.S. is at risk of defaulting. Keeping the AAA rating would’ve been the safe default choice. Deciding to downgrade—that’s political dynamite, with a risk of losing their lucrative quasimonopoly. That’s a decision you’d only make for a really good reason. Or maybe they’re just overcompensating for all those bad AAA ratings they gave out a few years ago?

I certainly see his point. But I don't think that the agencies are in much danger of losing their quasimonopoly, which makes the decision a bit harder to interpret as a bold act driven by conviction.