Giving up on an international financial regulator

One aspect of the present global crisis is the largely unregulated flow of global capital.  Yglesias just summed up the problem:

It’s not controversial, even among pretty serious free market types, that some kind of effective financial regulation is necessary. Nor is it controversial to observe that any given country could still be subjected to “systemic risk” emanating from a differently-regulated foreign country with which it has economic ties. Nor is it controversial to observe that firms actively seek to engage in “regulatory arbitrage,” i.e. take advantage of opportunities to dodge regulatory burdens by, among other things, taking advantage of nation-to-nation difference in regulation.

One solution might be a global regulator.  Yglesias notes that this is probably impossible but he also notes that this sort of debate is not his specialty.  So now let's go to Dani Rodrik who does this sort of thing for a living at Harvard.

That is absolutely right.  But Ken's preference for a "global financial regulator with real teeth" overlooks three major problems. Global financial regulation is a bad idea because it is neither desirable, nor prudent, nor feasible.  

It is not desirable because countries at different levels of development and with different national preferences with regard to how much risk they want to encourage as the price of financial innovation will want to select quite different national regulatory regimes.  There is a large element of a "local public good" in the financial system, and you need to recognize the heterogeneity of national preferences. 

It is not prudent, because a common global regulator will require global harmonization of rules. What if we converge on the wrong ones?  That was one of the points brought out by Katharina Pistor recently.   

Finally, it is not politically feasible because I just do not see that major countries will surrender national sovereignty to a global regulator with teeth. There is not enough political convergence globally for this to happen.  But the major principled objections are the previous two, rather than this one.

What is the alternative?  Ken is right that the only real alternative is a system of capital controls.  But viewed in the context of the arguments I just made, it is not at all clear that a system that allows and legitimizes capital-account management would not dominate a futile and undesirable effort to set up a global regulator.  In fact, the danger is that we will obsess on getting international regulation right with no plan B, and in doing so will simply prepare the groundwork for the next crisis.

But if we create a world of nationally segmented finance, won't we give up the benefits of global financial integration? Please.   

Capital controls in essence are national policies that control the flow of money in and out of a country.  Here he is going in more detail retorting common arguments against capital controls.   Paul Krugman hasn't actively been advocating for them, but he had reversed himself on the Asian financial crisis by noting that at very least temporary controls helped some economies, such as China's, weather the crisis.  That said, they're no panacea, China still has controls and is definitely suffering along with the rest of us.

I don't really have the expertise to go much more deeper than that, but I thought people might be interested in learning about those who could.  I'm only able to go this far thanks to Prof. Haider Khan at University of Denver whose class on globalization introduced me to many economist I still follow.  Thanks professor!