After the Fall

As we would expect from Richard Posner, the distinguished and polymathic writer, law professor, and circuit court judge, his book on the financial crisis is thoughtful, rigorous, and balanced.

It has also been hailed as “an event” by some on the left, such as Nobel laureate economist Robert Solow, in whose eyes it marks the conversion of Posner, a former free-marketeer (albeit no ideologue), to the tenets of progressive common sense. For this volume claims that, since 2008, we have been experiencing “a failure of capitalism,” and Posner’s takeaway point is that “we need a more active and intelligent government to keep our model of capitalism from running off the rails.”

Strictly speaking, Posner is right: We would need an active and intelligent government to keep the model of capitalism used by Posner from running off the rails. But in truth, Posner’s model tells us little about the real world factors that produced the financial crisis. And once we take account of facts that Posner overlooks, it seems that the cause of the crisis was not the “laissez-faire economic regime” that Posner imagines might have been responsible, but the legal regulations that actually shaped the behavior of our banks.

Many different economic models can explain what might have caused the crisis. But readers will want to know what actually did cause the crisis. Posner tells us much about the economics of depressions in the abstract, the economics of housing in the abstract, the economics of banking in the abstract, and the economics of corporate compensation in the abstract. All of this economic theory is valuable; but in principle, most of it could have been written in 1999, 1989, or 1939–any time after Keynes’s General Theory appeared in 1936. (Posner is a recent convert to Keynes’s macroeconomics.) What A Failure of Capitalism lacks is evidence showing that any of these theories explains the crisis of 2008.

The heart of Posner’s case against “capitalism” is the following theory, which has been embraced by no less than the president of the United States: Perverse incentives, created by banks’ executive-compensation systems, caused the crisis. As Posner puts it, bank executives’ pay was structured so that bankers would think to themselves,

Moreover, according to Posner, subordinate employees had essentially the same incentives as top executives. Subordinates received bonuses for making money but were not penalized for losing it.

The theory is perfectly logical, and it might explain the crisis, but Posner does not show that it actually does explain the crisis.

For one thing, he doesn’t show that all banks used the same compensation system and paid the same bonuses for risk-taking. There are, in fact, differences among banks’ compensation systems, so Posner might have been able to test his theory by seeing if the banks that took more risks were the ones that provided bigger golden parachutes or paid higher bonuses. But Posner treats “banks” as a homogeneous lump. This makes it difficult for him to check his theory against reality.

Moreover, despite having written the bible of the “law and economics” movement–his 1973 treatise Economic Analysis of the Law–Posner tells us too little about the many laws that regulate real world capitalism, which surely must have affected bankers’ behavior. For instance, some of the investment banks that avoided mortgage-backed securities, such as Brown Brothers Harriman, are structured as partnerships; this encourages prudence because each partner has a lot at stake if the firm goes under. As the huge law firms demonstrate, partnerships need not be small–there can be hundreds or thousands of partners. But Richard Rahn has pointed out that the tax code–not capitalism–discourages partnerships in banking and other industries.

A Failure of Capitalism contains a devastating rebuttal of widely popular “irrational exuberance” explanations of the crisis. This leaves Posner to solve the puzzle of why rationally self-interested bankers seemed to ignore risk. But in the real world of contemporary capitalism, rational self-interest does not conform to the patterns it would follow under “a laissez-faire economic regime.” Instead, rational self-interest follows the tens of thousands of pages of the tax code; it follows the millions of pages of the regulatory code. And these tortuous legal pathways are largely overlooked by Posner.

Thus, he argues that the most important risky behavior prompted by the banks’ compensation structures was that bankers increased their leverage ratios. But banks’ leverage ratios are regulated by law, and this law, unmentioned by Posner, was probably the main cause of the crisis.

A bank’s leverage ratio consists of its capital divided by its assets, and its assets include its loans, such as mortgages. We usually think of loans as debits because most of us are borrowers. But to a lender, a mortgage (for instance) is an asset because it is supposed to be paid back. All assets are risky in an uncertain world, but a loan is especially risky, since any number of factors might cause a borrower not to pay it back. By increasing the ratio of mortgages and other loans to its capital, a bank is taking on more risk, even if the mortgages themselves aren’t riskier–and subprime mortgages were riskier.

“Banks wanted to make risky mortgage loans,” Posner writes, but this seemingly irrational behavior was mainly due, he explains, to the bankers’ rational-self interest: They were being paid to ignore risk. But since Posner homogenizes “banks” into an undifferentiated mass, he cannot tell us which bankers are supposed to have known they were taking excessive risks. And they would have had to know that if they were, as the executive-compensation theory maintains, deliberately ignoring risk in pursuit of a bigger bonus.

Nor can Posner tell us whether all banks “leveraged up” to the same degree (they didn’t) or made subprime loans to the same degree (they didn’t). If all bankers had essentially the same incentives because of the way they were paid, what could explain their actually heterogeneous behavior?

Meanwhile, Posner does not discuss the legal rules that govern banks’ leverage ratios in the real, far-from-laissez faire world. These regulations go under the name of the “Basel accords” after the Swiss town where, in 1988, the developed world’s central bankers agreed to them. The Basel accords set a ceiling on banks’ leverage by regulating the amount of capital banks must hold–and, crucially, the type of assets they may hold.

The Basel accords required a minimum level of 8 percent capital for lending banks (as opposed to investment banks) yielding a 12.5-to-1 ratio of assets to capital–once assets were adjusted for the riskiness that the Basel regulators saw in different types of assets. For instance, they saw zero risk in cash but 50 percent risk in mortgages, so a bank needed to hold no capital against cash but 4 percent capital against mortgages. To the Basel accords, each country’s regulators were free to add their own fillips.

Ten years after the Basel accords were implemented in the United States, the American regulators amended them. Under the “Recourse Rule,” adopted in 2001, mortgage-backed securities were risk-weighted at 20 percent, requiring 60 percent less capital than actual mortgages required. The only qualification was that the mortgage-backed securities had to be rated AA or AAA by one of the three “rating agencies,” Moody’s, Standard and Poor’s, or Fitch.

These three private companies had a legally protected oligopoly. The oligopoly found a way to give AA and AAA ratings to slices of mortgage-backed securities that consisted entirely of subprime mortgages. Thus, the Recourse Rule created an incentive for lending banks to “leverage up” by originating as many mortgages as possible, selling them for securitization to an investment bank such as Bear Stearns, and buying them back as part of an AA- or AAA-rated security. Thus could a bank increase its lending power–hence its potential profitability–by 60 percent per transaction.

Would this have been “normal business activity in a laissez-faire economic regime,” as Posner contends? No. But it was consistent with the rational self-interest of bankers under the amended Basel accords. The Recourse Rule did not force anyone to leverage up; but it richly rewarded those bankers who–like the regulators themselves–saw little risk in leveraging up by buying highly rated mortgage-backed securities.

The Recourse Rule, however, gave banks the same ability to increase their leverage whether they bought AA-rated or AAA-rated securities. If the reason that “banks” were leveraging up in the first place is, as Posner maintains, that they cared only about profits and ignored possible losses (because their executives and employees were compensated for short-term profits, regardless of the long-term risks), then banks should have bought AA securities every time: The AAs paid more than the AAAs–precisely because they were riskier.

But in fact, only 19 percent of the mortgage-backed securities held by the banks were rated AA or lower. Eighty-one percent were rated AAA, yielding less short-term profit because they carried less risk. This one fact may refute the executive-compensation theory all by itself.

Another inconvenient fact: If banks were seeking to maximize their leverage because they were heedless of the risk, then they should have driven their capital holdings down to the minimum allowed by law: 8 percent for “adequately capitalized” banks; 10 percent for “well-capitalized” banks, to which American regulators give privileges that most banks need. But in December 2007, as the crisis was getting underway, the average capital level of all American banks combined was roughly 13 percent–30 percent higher than the legal minimum.

This level had indeed declined from previous levels, so it is true that, in the aggregate, “banks” had leveraged up, just as the Recourse Rule would have led us to expect. The banks’ greater leveraging, however, cannot have been caused by the general indifference to risk blamed by Posner since, if there were any such indifference, the banks’ average capital ratio would have been 30 percent lower than it actually was.

These facts dovetail with a recent study by René Stulz and Rüdiger Fahlenbrach showing that banks with CEOs who held a lot of stock in the bank did worse than banks with CEOs who held less stock. Whatever mistakes they made, the CEOs were not making them deliberately, contrary to the executive-compensation theory.

What seems to have happened, then, is not that banks ignored risk. Rather, to the extent that generalizations can be made, banks tried to avoid “excessive” risk–but different bankers had different ideas about what was excessively risky and what wasn’t. Some bankers, such as those at Citigroup, saw little risk in leveraging up: Its capital level at the end of 2007 was 10.7 percent, barely above the legal minimum. Others, such as those at JPMorgan Chase, saw greater risk: Its capital level was 12.57 percent, and it avoided subprime securities despite the incentives offered by the Recourse Rule, because even those incentives were not large enough to compensate for the risk perceived by the Morgan bankers.

“Banks” did not homogeneously leverage up by buying mortgage-backed securities, heedless of the risk; their willingness to seize the rewards offered by the Recourse Rule varied according to their differing perceptions of the risk involved.

This thesis is borne out by two sensationalized but fact-stuffed books about Bear Stearns and JPMorgan: William D. Cohan’s House of Cards and Gillian Tett’s Fool’s Gold. From Cohan we learn that neither the Bear Stearns executives nor the subordinates whose actions brought down the bank had any idea that they were taking “excessive” risks. From Tett (whose book appeared too late for Posner to consider) we learn that the conservative risk perceptions of JPMorgan president Jamie Dimon and his subordinates counteracted the very real temptation to leverage up.

Was Dimon less rationally self-interested than Bear Stearns president Jimmy Cayne? No, but Cayne and his subordinates didn’t see the same risks that Dimon and his subordinates saw, or thought they saw, in AAA-rated mortgage-backed securities.

Under any version of capitalism, laissez faire or regulated, the rational pursuit of self-interest characterizes the successful companies, like JPMorgan. But it also characterizes the failures, like Bear Stearns and Citigroup. Therefore, a realistic “model” of capitalism has to contain more than rational self-interest if it is going to explain capitalists’ mistakes–and in the financial crisis of 2008, there were plenty of those.

If we are going to understand these errors, we have to bear in mind that capitalists have different ideas about how to pursue their self-interest–including different ideas about how to avoid undue risk. Unless capitalists’ ideas about these matters were different, there’d be no economic case for competitive capitalism. Competition is the only way to sort out the good ideas from the bad ones. The good ideas help a company survive and prosper; the bad ones cause losses or bankruptcy.

If anybody really knew in advance which ideas were good and which were bad, there’d be no point testing the ideas against each other through competition. But the hidden premise of banking regulations such as the Basel rules is that regulators can, indeed, know such things in advance. This premise puts the regulators in the position of trying to be omniscient judges of what constitutes “prudent” behavior. In 2001, the American regulators had decided that it would be more prudent for banks to hold AA or AAA rated mortgage-backed securities than to hold actual mortgages, so banks that made this switch were rewarded with 60 percent more potential profits.

Among fallible human beings, of course, what constitutes prudence is a matter of legitimate dispute. But unlike capitalists’ ideas about prudence, regulators’ ideas cannot compete against each other to sort out the bad from the good: Only one regulation at a time is the law of the land. So if we see a high proportion of capitalist enterprises making the same mistakes, as we do when we look back at the run-up to the crisis, we might suspect that a homogenizing force–such as the incentives imposed on all banks by mistaken regulations–were at work.

If one seeks the cause of a systemic problem, a logical place to look is among the laws that govern the system as a whole. Individual capitalists, of course, make mistakes all the time; we discover this when they go broke. And being human, they are as susceptible as anyone to herding around the conventional wisdom of their time and place, which is so often wrong. Thus, a systemic “failure of capitalism” is possible: In a “laissez-faire economic regime” capitalists could all make the same mistake. This is what Posner proves, and proves well.

But in the real world of 2008, the systemic tendency toward mistakes seems to have been caused not by any risk-insensitivity inherent to capitalism or to banking, nor by the banks’ executive-compensation systems–which, of course, are also subject to competition–but by the skewed incentives produced by particular regulations imposed on capitalism. The American amendments to the Basel rules created incentives for capitalists to buy mortgage-backed securities, tipping the risk-benefit calculations of many bankers toward what turned out to be disastrously imprudent behavior.

The regulators were human, and it turned out that their ideas about prudent behavior were wrong. Now the world is paying for their mistakes.

Jeffrey Friedman, a political scientist at the University of Texas, is the editor of Critical Review.