A full understanding of the crisis has to explain not just the housing and subprime bubbles, but why, when they popped, it should have had such disastrous worldwide effects on the financial system. The problem was that commercial banks had made a huge overinvestment in mortgage‐backed bonds sold by investment banks such as Lehman Brothers.
Commercial banks are familiar to everyone with a checking or savings account. They accept our deposits, against which they issue commercial loans and mortgages. In 1933, the United States created the FDIC to insure commercial banks’ depositors. The aim was to discourage bank runs by depositors who worried that if their bank had made too many risky loans, their accounts, too, might be at risk.
The question of whether deposit insurance was necessary is worth asking, and I will ask it later on. But for now, the key fact is that once deposit insurance took effect, the FDIC feared that it had created what economists call a “moral hazard”: bankers, now insulated from bank runs, might be encouraged to make riskier loans than before. The moral‐hazard theory took hold not only in the United States but in all of the countries in which deposit insurance was instituted. And both here and abroad, the regulators’ solution to this (real or imagined) problem was to institute bank‐capital regulations. According to an array of scholars from around the world — Viral Acharya, Juliusz Jablecki, Wladimir Kraus, Mateusz Machaj, and Matthew Richardson — these regulations helped turn an American housing crisis into the world’s worst recession in 70 years.
WHAT REALLY WENT WRONG
The moral‐hazard theory held that since the FDIC would now pick up the pieces if anything went wrong, bankers left to their own devices would make clearly risky loans and investments. The regulators’ solution, across the entire developed world, was to require banks to hold a minimum capital cushion against a commercial bank’s assets (loans and investments), but the precise level of the capital reserve, and other details, varied from country to country.
In 1988, financial regulators from the G-10 agreed on the Basel (I) Accords. Basel I was an attempt to standardize the world’s bank‐capital regulations, and it succeeded, spreading far beyond the G-10 countries. It differentiated among the risks presented by different types of assets. For instance, a commercial bank did not have to devote any capital to its holdings of government bonds, cash, or gold — the safest assets, in the regulators’ judgment. But it had to allot 4 percent capital to each mortgage that it issued, and 8 percent to commercial loans and corporate bonds.
Each country implemented Basel I on its own schedule and with its own quirks. The United States implemented it in 1991, with several different capital cushions; a 10 percent cushion was required for “well‐capitalized” commercial banks, a designation that carries privileges that most banks want. Ten years later, however, came what proved in retrospect to be the pivotal event. The FDIC, the Fed, the Comptroller of the Currency, and the Office of Thrift Supervision issued an amendment to Basel I, the Recourse Rule, that extended the accord’s risk differentiations to asset‐backed securities (ABS): bonds backed by credit card debt, or car loans — or mortgages — required a mere 2 percent capital cushion, as long as these bonds were rated AA or AAA or were issued by a government‐sponsored enterprise (GSE), such as Fannie or Freddie. Thus, where a well‐capitalized commercial bank needed to devote $10 of capital to $100 worth of commercial loans or corporate bonds, or $5 to $100 worth of mortgages, it needed to spend only $2 of capital on a mortgage‐backed security (MBS) worth $100. A bank interested in reducing its capital cushion — also known as “leveraging up” — would gain a 60 percent benefit from trading its mortgages for MBSs and an 80 percent benefit for trading its commercial loans and corporate securities for MBSs.
Astute readers will smell a connection between the Recourse Rule and the financial crisis. By 2008 approximately 81 percent of all the rated MBSs held by American commercial banks were rated AAA, and 93 percent of all the MBSs that the banks held were either triple‐A rated or were issued by a GSE, thus complying with the Recourse Rule. (Figures for the proportion of double‐A bonds are not yet available.) According to the scholars I mentioned earlier, the lesson is clear: the commercial banks loaded up on MBSs because of the extremely favorable treatment that they received under the Recourse Rule, as long as they were issued by a GSE or were rated AA or AAA.
When subprime mortgages began to default in the summer of 2007, however, those high ratings were cast into doubt. A year later, the doubts turned into a panic. Federally mandated mark‐to‐market accounting — the requirement that assets be valued at the price for which they could be sold right now — translated temporary market sentiment into actual numbers on a bank’s balance sheet, so when the market for MBSs dried up, Lehman Brothers went bankrupt — on paper. Mark‐to‐market accounting applied to commercial banks too. And it was the commercial banks’ worry about their own and their counterparties’ solvency, due to their MBS holdings, that caused the lending freeze and, thus, the Great Recession.
What about the rest of the world? The Recourse Rule did not apply to countries other than the United States, but Basel I included provisions for even more profitable forms of “capital arbitrage” through off‐balance‐sheet entities such as structured investment vehicles, which were heavily used in Europe. Then, in 2006, Basel II began to be implemented outside the United States. It took the Recourse Rule’s approach, encouraging foreign banks to stock up on GSE‐issued or highly rated MBSs.
THE PERFECT STORM?
Given the large number of contributory factors — the Fed’s low interest rates, the Community Reinvestment Act, Fannie and Freddie’s actions, Basel I, the Recourse Rule, and Basel II — it has been said that the financial crisis was a perfect storm of regulatory error. But the factors I have just named do not even begin to complete the list. First, Peter Wallison has noted the prevalence of “no‐recourse” laws in many states, which relieved mortgagors of financial liability if they simply walked away from a house on which they defaulted. This reassured people in financial straits that they could take on a possibly unaffordable mortgage with virtually no risk. Second, Richard Rahn has pointed out that the tax code discourages partnerships in banking (and other industries). Partnerships encourage prudence because each partner has a lot at stake if the firm goes under. Rahn’s point has wider implications, for scholars such as Amar Bhidé and Jonathan Macey have underscored aspects of tax and securities law that encourage publicly held corporations such as commercial banks — as opposed to partnerships or other privately held companies — to encourage their employees to generate the short‐term profits adored by equities investors. One way to generate short‐term profits is to buy into an asset bubble. Third, the Basel Accords treat monies set aside against unexpected loan losses as part of banks’ “Tier 2” capital, which is capped in relation to “Tier 1” capital — equity capital raised by selling shares of stock. But Bert Ely has shown in the Cato Journal that the tax code makes equity capital unnecessarily expensive. Thus banks are doubly discouraged from maintaining the capital cushion that the Basel Accords are trying to make them maintain. This litany is not exhaustive. It is meant
only to convey the welter of regulations that have grown up across different parts of the economy in such immense profusion that nobody can possibly predict how they will interact with each other. We are, all of us, ignorant of the vast bulk of what the government is doing for us, and what those actions might be doing to us. That is the best explanation for how this perfect regulatory storm happened, and for why it might well happen again.
By steering banks’ leverage into mortgage‐backed securities, Basel I, the Recourse Rule, and Basel II encouraged banks to overinvest in housing at a time when an unprecedented nationwide housing bubble was getting underway, due in part to the Recourse Rule itself — which took effect on January 1, 2002: not coincidentally, just at the start of the housing boom. The Rule created a huge artificial demand for mortgage‐backed bonds, each of which required thousands of mortgages as collateral. Commercial banks duly met this demand by lowering their lending standards. When many of the same banks traded their mortgages for mortgage‐backed bonds to gain “capital relief,” they thought they were offloading the riskiest mortgages by buying only triple‐A‐rated slices of the resulting mortgage pools. The bankers appear to have been ignorant of yet another obscure regulation: a 1975 amendment to the SEC’s Net Capital Rule, which turned the three existing rating companies — S&P, Moody’s, and Fitch — into a legally protected oligopoly. The bankers’ ignorance is suggested by e‐mails unearthed during the recent trial of Ralph Cioffi and Matthew Tannin, who ran the two Bear Stearns hedge funds that invested heavily in highly rated subprime mortgage‐backed bonds. The e‐mails show that Tannin was a true believer in the soundness of those ratings; he and his partner were exonerated by the jury on the grounds that the two men were as surprised by the catastrophe as everyone else was. Like everyone else, they trusted S&P, Moody’s, and Fitch. But as we would expect of corporations shielded from market competition, these three “rating agencies” had gotten sloppy. Moody’s did not update its model of the residential mortgage market after 2002, when the boom was barely underway. And Moody’s model, like those of its “competitors,” determined how large they could make the AA and AAA slices of mortgage‐backed securities.
THE REGULATORS’ IGNORANCE OF THE REGULATIONS
The regulators seem to have been as ignorant of the implications of the relevant regulations as the bankers were. The SEC trusted the three rating agencies to continue their reliable performance even after its own 1975 ruling protected them from the market competition that had made their ratings reliable. Nearly everyone, from Alan Greenspan and Ben Bernanke on down, seemed to be ignorant of the various regulations that were pumping up house prices and pushing down lending standards. And the FDIC, the Fed, the Comptroller of the Currency, and the Office of Thrift Supervision, in promulgating one of those regulations, trusted the three rating companies when they decided that these companies’ AA and AAA ratings would be the basis of the immense capital relief that the Recourse Rule conferred on investment‐bank‐issued mortgage‐backed securities. Did the four regulatory bodies that issued the Recourse Rule know that the rating agencies on which they were placing such heavy reliance were an SEC‐created oligopoly, with all that this implies? If you read the Recourse Rule, you will find that the answer is no. Like the Bank for International Settlements (BIS), which later studied whether to extend this American innovation to the rest of the world in the form of Basel II (which it did, in 2006), the Recourse Rule wrongly says that the rating agencies are subject to “market discipline.”
Those who play the blame game can find plenty of targets here: the bankers and the regulators were equally clueless. But should anyone be blamed for not recognizing the implications of regulations that they don’t even know exist?
Omniscience cannot be expected of human beings. One really would have had to be a god to master the millions of pages in the Federal Register — not to mention the pages of the Register’s state, local, and now international counterparts — so one could pick out the specific group of regulations, issued in different fields over the course of decades, that would end up conspiring to create the greatest banking crisis since the Great Depression. This storm may have been perfect, therefore, but it may not prove to be rare. New regulations are bound to interact unexpectedly with old ones if the regulators, being human, are ignorant of the old ones and of their effects.
This is already happening. The SEC’s response to the crisis has not been to repeal its 1975 regulation, but to promise closer regulation of the rating agencies. And instead of repealing Basel I or Basel II, the BIS is busily working on Basel III, which will even more finely tune capital requirements and, of course, increase capital cushions. Yet despite the barriers to equity capital and loan‐loss reserves created by the conjunction of the IRS and the Basel Accords, the aggregate capital cushion of all American banks at the start of 2008 stood at 13 percent — one‐third higher than the American minimum, which in turn was one‐fifth higher than the Basel minimum. Contrary to the regulators’ assumption that bankers need regulators to protect them from their own recklessness, the financial crisis was not caused by too much bank leverage but by the form it took: mortgage‐backed securities. And that was the direct result of the fine tuning done by the Recourse Rule and Basel II.
HOW DID WE GET INTO THIS MESS?
The financial crisis was a convulsion in the corpulent body of social democracy. “Social democracy” is the modern mandate that government solve social problems as they arise. Its body is the mass of laws that grow up over time — seemingly in inverse proportion to the ability of its brain to comprehend the causes of the underlying problems.
When voters demand “action,” and when legislators and regulators provide it, they are all naturally proceeding according to some theory of the cause of the problem they are trying to solve. If their theories are mistaken, the regulations may produce unintended consequences that, later on, in principle, could be recognized as mistakes and rectified. In practice, however, regulations are rarely repealed. Whatever made a mistaken regulation seem sensible to begin with will probably blind people to its unintended effects later on. Thus future regulators will tend to assume that the problem with which they are grappling is a new “excess of capitalism,” not an unintended consequence of an old mistake in the regulation of capitalism.
Take bank‐capital regulations. The theory was (and remains) that without them, bankers protected by deposit insurance would make wild, speculative investments. So deposit insurance begat bank‐capital regulations. Initially these were blunderbuss rules that required banks to spend the same levels of capital on all their investments and loans, regardless of risk. In 1988 the Basel Accords took a more discriminating approach, distinguishing among different categories of asset according to their riskiness — riskiness as perceived by the regulators. The American regulators decided in 2001 that mortgage‐backed bonds were among the least risky assets, so they required much lower levels of capital for these securities than for every alternative investment but Treasury’s. And in 2006, Basel II applied that erroneous judgment to the capital regulations governing most of the rest of the world’s banks. The whole sequence leading to the financial crisis began, in 1933, with deposit insurance. But was deposit insurance really necessary?
The theory behind deposit insurance was (and remains) that banking is inherently prone to bank runs, which had been common in 19th‐century America and had swept the country at the start of the Depression.
But that theory is wrong, according to such economic historians as Kevin Dowd, George Selgin, and Kurt Schuler, who argue that bank panics were almost uniquely American events (there were none in Canada during the Depression — and Canada didn’t have deposit insurance until 1967). According to these scholars, bank runs were caused by 19th‐century regulations that impeded branch banking and bank “clearinghouses.” Thus, deposit insurance, hence capital minima, hence the Basel rules, might all have been a mistake founded on the New Deal legislators’ and regulators’ ignorance of the fact that panics like the ones that had just gripped America were the unintended effects of previous regulations.
What I am calling social democracy is, in its form, very different from socialism. Under social democracy, laws and regulations are issued piecemeal, as flexible responses to the side effects of progress — social and economic problems — as they arise, one by one. (Thus the official name: progressivism.) The case‐by‐case approach is supposed to be the height of pragmatism. But in substance, there is a striking similarity between social democracy and the most utopian socialism. Whether through piecemeal regulation or central planning, both systems share the conceit that modern societies are so legible that the causes of their problems yield easily to inspection. Social democracy rests on the premise that when something goes wrong, somebody — whether the voter, the legislator, or the specialist regulator — will know what to do about it. This is less ambitious than the premise that central planners will know what to do about everything all at once, but it is no different in principle.
This premise would be questionable enough even if we started with a blank legal slate. But we don’t. And there is no conceivable way that we, the people — or our agents in government — can know how to solve the problems of modern societies when our efforts have, in fact, been preceded by generations of previous efforts that have littered the ground with a tangle of rules so thick that we can’t possibly know what they all say, let alone how they might interact to create another perfect storm.