Saturday, October 31, 2009

The Real Bank Pay Scandal

Federal Reserve chairman Ben Bernanke announced last week that all federally regulated banks will be subject to regulatory scrutiny of compensation practices “to ensure that they do not encourage excessive risk-taking.” Bernanke justified this plan by asserting that “flawed compensation practices at financial institutions” helped cause the financial crisis.

It would be an understatement to say that this assertion has not been proven. Bernanke’s heavily footnoted speech provides dozens of statements from various regulatory bodies that simply assert the Bernanke thesis as if it were self-evidently true. The real scandal of bankers’ pay contributing to the financial crisis is that nobody, literally nobody, has offered a shred of evidence that it did contribute to the crisis. Yet public opinion, the financial press, and the regulators have decided that it did, and drastic new regulatory policies are being constructed on that basis.
Moreover, when we consult what evidence is known, it seems that bankers, far from deliberately courting disaster so as to boost their pay, unintentionally took risks that jeopardized their own wealth.

We can see some of the results in the huge hits taken by key bank executives, such as Richard Fuld of Lehman Brothers, who lost $1 billion by retaining his Lehman holdings until the bitter end. Sanford Weill of Citigroup lost half that amount. Ohio State economists RĂ¼diger Fahlenbrach and RenĂ© Stulz have shown that banks whose CEOs held a lot of their banks’ stock fared worse in the crisis than banks with CEOs who held less stock. The CEOs who held stock in the riskier banks must have been ignorant of the risk, or they would have sold their shares.

Ignorance, not a perverse pay structure, also seems to have been at work at Bear Stearns. William D. Cohan’s "House of Cards" suggests that before losing about $900 million in stock when the bank imploded, Bear Stearns CEO James Cayne had no idea that anything was amiss. In fact, he was obsessed with the construction of a huge new headquarters building that would testify to the permanence of Bear Stearns, formerly the upstart of investment banks. It isn’t plausible that he would have let anything jeopardize this legacy—if he knew about the risk. All the while, two Bear Stearns hedge funds were buying billions of dollars worth of mortgage-backed securities that ended up sinking the parent bank. Did the employees who ran these funds, Ralph Cioffi and Matthew Tannin, understand the risks they were taking? That is part of what’s at issue in their trial for securities fraud. But we already know that they had millions of dollars of their own invested in their hedge funds, and emails between them suggest that until nearly the end, they felt secure in the knowledge that the funds’ holdings were rated AAA.

Indeed, 93 percent of the mortgage-backed securities purchased by banks either were rated AAA or—better yet—were insured by Fannie Mae or Freddie Mac, with implicit federal guarantees. It turned out to be a mistake to trust the AAA ratings, but in making this mistake, bankers again demonstrated that they were not seeking short-term profits to boost their annual bonuses. Had that been the bankers’ aim, they would not have bought AAA-rated mortgage-backed bonds, which, being “safe,” paid a lower interest rate than AA- and lower-rated bonds. Bankers who were indifferent to risk because they were seeking higher return should have bought the more-lucrative, riskier bonds, but they did so only 7 percent of the time.

The theory that the financial crisis was caused by pay incentives that encouraged risk-taking simply does not fit the facts. Moreover, if we ask why bankers invested so heavily in mortgage-backed bonds instead of other securities, the answer seems to be a previous regulatory attempt to steer banks away from risk. In 2001, the Fed, the FDIC, the Comptroller of the Currency, and the Office of Thrift Supervision promulgated the Recourse Rule, which encouraged commercial banks to trade individual mortgages, which the regulators deemed risky, for mortgage-backed securities, which the regulators deemed safe—as long as they were guaranteed by Fannie or Freddie, or were rated AA or AAA. The bankers’ repudiation of the higher-yielding AA bonds shows that they were not deliberately courting risk. But the regulators’ judgment about the risklessness of AAA mortgage-backed bonds proved to be disastrously wrong—and this judgment, in the form of the Recourse Rule, made those bonds much more profitable for banks than they otherwise would have been. As a result, banks ended up buying about half of the world’s entire supply of mortgage-backed securities, and when housing prices began to drop, the uncertain value of these securities caused the financial panic of September 2008.

One lesson of the financial crisis, then, is that regulators, like bankers, are human. Nobody can flawlessly predict the future, but regulators—like bankers and other businesspeople—have no choice but to try.

There is a difference, however. Competing businesses may pursue a variety of strategies based on divergent risk/reward assessments. Citigroup jumped into mortgage-backed bonds with both feet, even opening its own securitizing unit to transform mortgages into mortgage-backed securities. J. P. Morgan thought the risk was too great and went in the opposite direction, even though that meant passing up the extra rewards offered by the Recourse Rule. Such heterogeneous behavior is perhaps the one saving grace of capitalism. By behaving heterogeneously in order to compete with each other, capitalists spread society’s bets among a variety of ideas about the location of risk and reward.

Capitalists as a group aren’t smarter or better at predicting the future than regulators are, and when capitalists fall victim to the herd mentality, they may homogeneously converge on a mistaken prediction. This is a danger, but regulations aggravate it. Of necessity, a regulation homogenizes capitalists’ behavior. Prudential bank regulation, for instance, imposes one idea about risk on the whole banking system, skewing the various risk/reward calculations of all bankers in a direction the regulators deem safe. If the regulators turn out to be mistaken, as they were in 2001, they put the whole system at risk.

The Bernanke plan is intended to reduce the risk taken by banks, but by homogenizing all banks’ compensation practices, it may actually aggravate systemic risk. To take such an action with no evidentiary basis flies in the face of reason.

--Jeffrey Friedman and Wladimir Kraus

7 comments:

Greg Hill said...

I can’t quibble with the claim that the future of the human affair is unpredictable, and I’m willing to accept the proposition that regulators are no better at this impossible game than bankers. But these premises are not sufficient to generate the conclusion that the compensation of bankers should be unregulated.

Risk-taking by bankers, whether it is too much or too little, wouldn’t be a matter for public concern if bankers, their shareholders and bondholders, suffered all of the misfortunes, and enjoyed all of the fruits, of their risk-taking. But they don’t.

Suppose banks enter into contracts in which one party agrees to pay the other $100,000,000 if event A occurs, and the other party agrees to pay $100,000,000 if event A does not occur. Such contracts might be attractive because the parties have different beliefs about the probability of event A occurring, or because event A is correlated with other events, favorable or unfavorable, which makes it possible for banks to hedge their risks by entering into these, “derivative,” contracts.

Once the progress of time produces an outcome, event A will have occurred or not, and, in the ideal market of our imagination, all of the contractually obligated payments based on the occurrence, or non-occurrence, of event A will be forthcoming.

Of course, it doesn’t always turn out this way. Some parties to contractual agreements may not be able to make good on their obligations, and some parties who fail to receive the payments to which they are legally entitled may, in turn, default on their obligations, giving rise to a self-reinforcing cascade of defaults and related misfortunes.

It’s the cascading ramifications of illiquidity, insolvency, and default that draw our attention to the risk-taking of banks and other institutions that have become “too big to fail.”

We can justifiably invoke the concept of “externality” to characterize these kinds of cascades because a substantial part of the economic cost created by such widespread failures “to make payment” is borne by people who were one or more transactions removed from the original “trade” (or “bet” on the occurrence of event A).

The presence of externalities – costs and benefits that don’t figure in a bank’s decision making – constitutes a prima facie case for regulation (or for some other departure from laissez faire).

Now, the general question raised by Friedman and Kraus is whether a regulatory regime designed to reduce the likelihood and extent of such adverse cascades necessarily presupposes that government officials are better at risk assessment than bank CEOs?

I don’t believe this premise is, in fact, essential to the case for regulation. A Bank’s CEO aims to advance the interests of her shareholders, and she judges risk and reward with this end in view. By contrast, a regulator of the banking system aims to forestall a cascade of defaults that could push the economy into a self-reinforcing tailspin.

The argument for regulation is not that the civil servant has a better grasp of the risks facing a particular bank than its CEO, but that the CEO is preoccupied with the particular, rather than the general, interest (as Rousseau, or Hegel, might have put it).

Greg Hill

Wladimir Kraus said...

Thanks, Greg!

I believe there are two points that you’re trying to make.

1. The banking industry, like no other, is a source of potentially devastating systemic externalities. It is like a highly flammable or radioactive substance that needs to be intelligently fenced so as not to endanger innocent bystanders. Leaving the dangerous substance to itself (laissez-faire economic policy) would be highly irresponsible because there is no guarantee that if not constrained it will behave benignly. The financial crisis that triggered the present depression appears to confirm this story.

Undeniably, the financial system was indeed very fragile. But why? What does fragility mean here? It was fragile because (a) of a very high degree of leverage, which means that even a slight drop in the value of its assets wiped out the capital of a bank. And (b) the problem of highly overpriced assets such as ABS of different categories. Basically, the combination of the two factors, whose mutual effects tended to reinforce each other, has produced a highly unstable financial environment.

But who or what made the bankers to borrow more and invest into assets of very dubious value?

The explanation and the solution of the problem offered focus exclusively on the issue of incentives to borrow more or less (regulation of lending and borrowing practices), to buy more or less *risky* assets (capital requirements + compensation practices), to engage in more or less *risky* operations (Glass-Steagal). Fix the incentives so that no systemic externalities are emitted and we will solve the problem.

From where I stand, there are precisely as many unknowns in this formula as there are elements to be fixed. We don’t know what *risky* behavior is, what *risky* operations are, what *risky* assets are because we don’t know what is *risky* or not risky EX ANTE. Yes, it turned out to be very *risky*, EX POST, to leverage up to the degree banks did, to buy assets in quantities and qualities that they were bought. But ex post *knowledge* that something bad happened is not a substitute for the theoretical knowledge of how the system works. Much less is it a *prima facie* case for trying to fix something through regulation.

What we need is intelligent regulation, which means on the basis of the knowledge of what works and what does not work.


Prima facie is our need for a better understanding of how the financial markets function and the nature of the link to the “real” economy. We all are aware of the dismal failure of the economics profession to predict and account for the crisis. Quite simply, the profession was completely oblivious/ignorant of the mechanics of the underlying processes. Greg calls for “regulation” but on basis of what knowledge?


2. Greg objects to our point that the competition as a mechanism of knowing and reacting to changing circumstances is even relevant to the problem at hand. The intention of a “regulator of the banking system aims to forestall a cascade of defaults that could push the economy into a self-reinforcing tailspin” is not in dispute here. What is in dispute here is the level of regulator’s knowledge. Also, there are the unintended but still homogenizing effects of *harmful* regulations that impose caps on flexibility and actions of market participants. Of course, there is no a priori case that regulations are necessarily harmful or that they are necessarily worse than competition with its discovery process. What is sauce for the goose is sauce for the gander. Dispersed knowledge and evolutionary processes are not a magical mojo.

So, our purpose in the article was, first, to show why the compensation story does not make sense and how specific regulations (not *ideal* and perfect ones in abstract) might have contributed to the problems in the banking sector. Another point was the importance of competition in choosing between different ideas and actions.

Charlie Tapp said...

Thomas Cooley, Dean of NYU's Stern School has also shown quite clearly that much of the policy debate on this front is wrong-headed.

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1493019

Even if it weren't, after taking public choice considerations into account, there's no reason to believe that government would get it right more often than the private sector.

Greg Hill said...

Wladimir, thanks for your reply.

I’ll address the “knowledge problem” momentarily, but first I want to stress the general point that, insofar as externalities are present, the profit-maximizing decisions of firms will not, except in special cases, produce efficient outcomes.

The planner aims to minimize total costs, whereas the CEO aims to minimize private costs. And the “knowledge issue” when externalities are present is not whether the planner is a better judge of what minimizes a firm’s private costs than its CEO, for this is not the planner’s objective. Nor is the “knowledge problem” whether the planner is a better judge of what minimizes total costs than the CEO, for this is not the CEO’s objective.

That said, let’s turn to the banks. Suppose there are two alternatives: 1) let banks make as many loans of whatever kind they wish to; or 2) impose some limits on the quantity and quality of the loans banks can make.

The first alternative is fine if banks are small and bank failures don’t produce “default cascades.” If, however, there are large banks, then I wouldn’t bet the economy on laissez-faire banking. But the reason I would reject this bet is not because I believe CEOs are ignorant when it comes to minimizing their private costs, but rather because their private costs are not the end of the matter.

Granted, the second alternative requires a good understanding of the banking system, and judgments must be made about what poses unacceptable systemic risk. And, of course, there will be mistakes, and not everybody’s risk preferences can be accommodated.

But this departure from laissez-faire banking cannot be swept aside by the Hayekian argument that each bank’s CEO knows more about her banking operations, risk exposure, etc., than the central planner. This may well be true, but there’s no reason to believe that competition among banks concerned only with their private costs won’t produce a financial crisis.

Greg Hill said...

Jeff and Wladimir,

I owe you an example of plausible macro risk regulation. See the plan below, which, granted, does assume some knowledge:

http://www.voxeu.org/index.php?q=node/4173

anne said...

People who are educated at the very best business schools in the country and are paid millions of dollars for their business acumen have opted out of the "regulation made me do it" argument.

Yes the feds propped up the bankers' business model in ways that have been extremely damaging. But bankers really should be held accountable for their "profit-maximizing" ways, if such ways crash their businesses and the economy at large.

QUALITY STOCKS UNDER 5 DOLLARS said...

Arrest the banksters.