News that megabank JPMorgan Chase lost $2 billion on a trading scheme designed to help hedge risk has predictably resulted in smug told-you-sos from those who favor increased regulation of the financial sector. In his column this morning, for example, Paul Krugman writes that while many business mistakes do not require a government response, "banks are special, because the risks they take are borne, in large part, by taxpayers and the economy as a whole. And what JPMorgan has just demonstrated is that even supposedly smart bankers must be sharply limited in the kinds of risk they're allowed to take on."
But here's the thing: JP Morgan Chase's bad deal might show up as a quarterly loss on the company's balance sheet (although CEO Jamie Dimon is still predicting a small profit), but it is almost certainly not going to result in a direct loss to taxpayers. This is a $2.3 trillion company with a $190 billion capital base. The banking unit made more than $12 billion last year. Gene Kirsch of Seeking Alpha points out that the $800 million quarterly loss "represents approximately 4% of its total net profit for all of 2011, less than 2.7% of its operating income." Even Jared Bernstein, a former White House economic adviser who also thinks the loss proves the case for regulation, agrees that the bank "appears to be handily able to cover the losses." So even with a relatively large loss like this, we're not talking about a serious direct risk to taxpayers. Instead, we're looking at a substantial loss to the individual bank.
That's important, because the Wall Street calamities that shook the economy a few years back weren't a result of isolated mistakes at the individual bank level. They were the result of networked failures, in which multiple market players make the same set of mistakes at the same time, taking up all the give in the system simultaneously.
Would tougher regulation of the financial sector have prevented JPMorgan's loss? That's not at all clear. The Washington Post's Allen Sloan, who favors many stricter financial sector rules, says that because there's no likely loss to taxpayers, the blown deal proves mostly that the bank should be embarrassed. Bernstein argues that Dodd-Frank would have prevented the loss if "properly implemented and enforced."
But "proper" implementation is always harder than it sounds. And I'm not sure we have any more reason to trust that regulators have the wisdom and judgment to prevent such losses any more or better than the bankers themselves. As Bernstein writes, "the fundamental truth here is the one known since Adam (Smith, that is) and amplified by the great financial economist Hy Minsky: humans underprice risk." But that's true of regulators as well. And even the smartest and most capable regulators have several disadvantages compared to their industry counterparts. One is that they lack the kind of intimate knowledge that deal makers have of their own transactions. The second is that they don't have the same sort of financial incentives not to fail. As an institution, JPMorgan Chase lost several billion dollars and took a huge hit to their credibility with this trade. Three executives tied to the blown deal are likely to lose their jobs. CEO Jamie Dimon is facing tremendous embarrassment and pressure from the media, public officials, and, no doubt, his shareholders. Overall, these folks had a lot more riding on the success of these deals than any regulator ever would have. And yet they still made mistakes.
Which is ultimately the nature of the marketplace. Markets don't evolve by preventing mistakes entirely. They learn by making mistakes, by experimenting with new business models, some of which prove unsuccessful, and then further refining the process, and usually making more mistakes along the way. But it's incredibly difficult for anyone — regulator or market player — to know what will fail in advance, and regulations that prevent some failures also typically end up blocking a lot of potential successes. What JPMorgan's blown deal mostly proves is that complex systems sometimes fail, and that it's very hard to know exactly when and how those systems will fail until they do.