The development of mortgage-backed securities, deeply implicated in the crisis, is a telling instance. A traditional bank keeps loans on its own books and has to worry about the creditworthiness of its borrowers. Regulations require it to hold some safe, low-yielding assets as a reserve against loans going bad. But the packaging of loans into tradable securities let some of the risk be moved to non-bank financial entities. The loans could then be shuffled into baskets of varying risk, and traded globally: the safest securities, as judged by credit-rating agencies, would be bought by conservative investors, and the riskiest by the more adventurous. The system as a whole would then be safer, it was believed, because the risk would be spread to the parts best able to bear it.
One result of the development of mortgage-backed securities, among other innovations, was higher leverage in the financial system as a whole. In the 1950s, bank capital ratios of 30 percent were common. Ratios eventually fell to 10 percent or less before the crisis struck, and regulators saw no problem with the reduction, since the banks were carrying less risk. Meanwhile, the investment banks that were creating and sometimes holding mortgage-backed securities got by with much less in reserve—sometimes as little as 3 percent. When the crash came, reserves of capital throughout the financial system proved to be much too thin. But again, few people believed that was the case before the crash; more sophisticated management of risk was thought to have made the system safer.
Like most financial innovations, mortgage-backed securities married regulatory avoidance, undue optimism (new means better), an inadvertent narrowing of commercial oversight (because of sheer complexity), and, yes, good-faith striving for financial and economic efficiency. Complex financial models were used to bundle and rebundle the loans, and to price and rate the different securities. Innovations in information technology and the application of advanced mathematics drove the process, and obscured the risks from regulators and investors alike.
Financial innovation serves a useful economic purpose, which has made many governments ambivalent about their regulatory wars. Seen from one angle, the invention of mortgage-backed securities, which to a large degree enabled the expansion of lending to subprime borrowers, was a way for financial firms to squirm around the rules and get a piece of some very profitable action. Seen from another angle, it was a way to widen homeownership and empower the working class—to get loans to people who had previously been financially disenfranchised, despite being able, in the great majority of cases, to afford their loans. Many observers viewed subprime lending as a good thing. Mostly, it was a good thing. It was not a problem until it became a problem.
At any rate, once information technology and instant telecommunications upped the pace of financial innovation, the battle tilted dramatically toward the innovators. If the executives in charge cannot understand what their repurposed physicists and computer scientists are doing, how likely is it that regulators can stay on top? As long as taxpayers are unwilling to buy these officials multimillion-dollar Manhattan apartments and Ferraris to get to the office, there will be a disparity of intellectual resources. In the struggle between regulators and innovators, most of the firepower will stay on one side.
Lacking the will as well as the means to resist, governments went with the flow, coping as best they could, choosing to let finance, up to a point, regulate itself. Governments would insist on disclosure and openness, so banks and non-banks could keep an eye on each other; private, profit-making monitors—the rating agencies, among others—would play a large role as well. But the basic idea was that the professionals could partly be left to take care of themselves, knowing that if they behaved recklessly and things went wrong, they would bear the losses.
Well, as I say, that was the idea.
This decades-long compromise—an evolving and uneasy accommodation with the forces of financial innovation—has been smashed by the crash of 2008. But what comes next is far from obvious. The popular view that the current mess is all the fault of “deregulation” is misleading, at best. The implication that all we need do is return to an earlier era of stricter supervision is an illusion. To repeat, regulators did not choose to retreat; they were forced to. In the United States, Democratic administrations, rarely inclined to see deregulation as an end in itself, ceded at least as much ground as Republican ones. The main forces that spurred the retreat—the incentives to evade close supervision, and the technological opportunities to do so—continue to grow more powerful. So what do we do?
First, the financially sophisticated cannot be trusted to monitor themselves and each other—not to the extent that they have been lately, at any rate. They have made a hash of it, and the rest of us are now paying the price. Even if it means suppressing innovation, at significant cost to the rest of the economy, top-down supervision will have to be tightened. The scale of the present crisis will force a new balance to be struck.
Second, the idea that banks can be neatly segregated for regulatory purposes from other kinds of financial firms must be ditched. When firms that are not ordinary deposit-taking banks act in many other respects like banks—transforming short-term money into long-term money—they face the risk of collapses in confidence and bank-like runs. Moreover, they may be so big, or so interconnected with the rest of the financial system, that when they go bust they cause as much collateral damage as would big conventional bank that fails.
A new approach to financial regulation is already taking shape. The Treasury began to dissolve the distinction between banks and non-banks, for instance, when it widened access to emergency borrowing from the Federal Reserve—a privilege once held only by deposit-takers. Having extended the assistance they are willing to offer, the authorities must extend their supervision as well. The logic is the same as with deposit insurance: if you socialize the costs of failure, you have to regulate against recklessness. The broadening of intervention and supervision will need to go further.