The Economist is hosting a debate on financial system regulation between no less than two Nobel Laureates, Myron Scholes and Joseph Stiglitz. (Be sure to read the opening statements before the rebuttals, or it may not make sense.) The debate is less over specifics than over the general question of how much regulation there should be. They may be lying low right now, but there will surely be legions of executives and economists arguing that we actually need less regulation in order to foster financial innovation.

The Economist recruited Scholes to defend this view, but unfortunately he puts on a rather tepid defense. I read his arguments three times and I think they boil down to this: Crises stem from too much leverage, and therefore bank capital requirements should be increased. (He also says, however, that “Determining the amount of leverage to be used by financial institutions is a business decision.”) If banks need additional capital in a crisis, it should be provided by the government and priced accurately. In his rebuttal he also proposes a new accounting framework, potentially implemented by a regulator, that provides a more accurate assessment of the risk faced by a financial institution. So, as far as I can tell, it boils down to: (a) higher capital requirements; (b) government capital in times of crisis; and (c) better accounting. For the rest, we can count on existing laws against things like fraud. Unfortunately, the only evidence he provides for the thesis that “more regulation is bad” is that economic growth was lower from the 1930s to the 1970s, which he calls an era of regulation, than since the 1970s, an era of deregulation. (Like everything in history, economic growth levels are overdetermined, meaning that you can find a dozen different explanations of any given historical phenomenon.)

Stiglitz doesn’t do such a great job proving the “more regulation is good” thesis, either; his evidence is that countries with “strong regulatory frameworks” are less likely to have financial crises. But Stiglitz gets at the basic question: is unbridled financial innovation good or bad? Does it really lower the cost of capital enough to compensate for the costs of crises like the current ones? Which innovations are good and which are bad? Can we get the good ones without the bad ones?

If we just take the case of mortgages and mortgage-backed securities (this is me and not Stiglitz), which innovations contributed to increasing home ownership in a sustainable way, and which put people in mortgages they had no chance of affording? I think most people would say that mortgage securitization was a good thing because it lowered borrowing costs without compromising underwriting standards. At the other extreme, most people would say that option ARM mortgages that reset to high interest rates and had prepayment penalties were not a good thing for either party. I don’t see how higher capital requirements would have solved the latter problem; better accounting might have helped if it discouraged banks from keeping the things on their books (but many were already doing their best to sell them instead of keeping them on their books).

Steve Randy Waldman has a list of what he thinks were good innovations and bad innovations. I don’t agree with everything on the lists, but at least he tried.

In Stiglitz’s view, we need regulation to compensate for the information asymmetries that enable the sophisticated to prey on the unsophisticated, and to compensate for the screwed-up incentive structures that are too common within companies. He doesn’t go into great detail on the actual mechanisms, although he does favor a “financial products safety commission” (also recommended by Crotty and Epstein).

Stiglitz also mentions but does not solve the tough problem of regulatory capture – what happens when regulatory bodies become advocates for the industry they are supposed to be regulating, often because the regulators themselves are drawn from that industry and intend to go back to it after their stint in Washington. What do you do if you have regulators who don’t want to regulate?

Anyway, this is where I think the debate needs to start.

• Commentary