Wednesday, November 16, 2011

On Crises

. Wednesday, November 16, 2011

Was on a panel this evening with Karl Smith and a couple other UNC folks that focused on the EU debt crisis. The event was hosted by the UNC economics club (I think). I am happy to report that we solved the crisis, or at least figured out what to do to prevent the next one.

Karl's contributions got me wondering about some things, so I wandered over to Modeled Behavior to get a better sense of his point of view. He is rather critical of the ECB, by the way, because he thinks it should be playing a LOR role and is very annoyed that it refuses to do so. He seemed unsure about whether the ECB's refusal is sincere (willing to blow up the system to avoid higher inflation in the short run and moral hazard problem in the long run) or strategic (credible commitment to no LOR in order to force reform in the PIIGS, then step in as LOR). I tried to convince him it was sincere. He seemed to remain hopeful that it was strategic. Obviously, if it is strategic, the ECB would want us to think it sincere, so it makes sense that we are unsure.

As I was scrolling through Modeled Behavior, however, I got distracted by this quote from Larry Summers that I had not seen before:
“In four years of reflection and rather intense involvement with this financial crisis, not a single aspect of dynamic stochastic general equilibrium has seemed worth even a passing thought,” Summers said, adding: “I think the profession is not entirely innocent.” Still, Summers said, the complaint that economists should have seen the crisis coming represents “a confusion” on the part of critics. Identifying financial bubbles and knowing when they will burst, he claimed, “is to ask more of the profession than it can reasonably expect to discover.”
Karl comments:
The problem here is that we have had hundreds of years of market evolution under a capitalist system where bubbles can and did occur. No solution has emerged. This is despite the fact that evolutionary systems can evolve solutions to problems that no one understands or indeed is even aware exist. This tells me that there is at least a reasonable chance that bubbles are the result not of mistaken expectations but of some combination of market and government failure. Its not crazy to suggest that we will be able to see this.

All of this makes me wonder: how many hundreds of years must elapse before we conclude that "bubbles" and the resulting market corrections are a part of the underlying distribution of asset price movements? It seems that everyone who looks (from Benoit Mandelbrot on) finds that (asset) price movements are power law distributed; most are rather small, some are extremely large (see also Didier Sornette). Thus, large market corrections are not market failures, because they are an expected outcome of typical market behavior. We don't need to develop special explanations for them because they result from the same processes that cause smaller movements. We can't see this because there isn't anything to see. The dynamic process is what Per Bak dubbed "self organized criticality": a steady input drives the system to a critical state, and in this critical state the system generates outputs of varying magnitude. Consequently, most of the time markets don't crash. Sometimes they do. End of story.

There is an associated story here that I have been puzzling over for a while: humanity's need to make sense of big events in concrete and personal terms. Those damn bankers and their crazy complex mortgage backed securities. Those crazy corrupt Greeks. That Allen Greenspan with his low interest rates. Thus we substitute description for explanation and don't seem to care much about (or even recognize) the distinction. We also seem to have this compelling need to figure out who is to blame (in fact, the last question on the list at this evening's panel was precisely that--who is to blame for the EU debt crisis?). Explanations based on the underlying distribution fail entirely to satisfy us (though we do seem to accept them when it comes to earthquakes).

I don't know if this reflects some cognitive bias (as individuals we only believe what we can observe) or some social science thing (social scientists prefer actor-centered explanations) or some manifestation of how negative shocks enter into the political system, shape legislative hearings, and thus enter the public consciousness through media reports of the political reaction which can only be about concrete things like Greenspan and greedy bankers. Or maybe it is a product of our belief that we should be able to control our world and the reluctance to accept that sometimes we cannot separate things we want (financial systems) from things we don't want (market crashes).

Because I don't really have a conclusion, let me exit on the following. Rather than waste time trying to explain something that doesn't need explanation, we could focus on how to respond to these events when they occur. Given that we will confront them because we cannot prevent them, what can we do to minimize their negative consequences for the financial sector and real economies? California and Japan employ earthquake resistant construction techniques rather than try to prevent large earthquakes. Maybe that strategy could be usefully applied to thinking about financial crises. 

Which I think brings me back to Karl's point about the utter failure of the ECB to respond effectively to the EU crisis. The system needs someone to be LOR; if the ECB refuses to play this role, perhaps the Federal Reserve must.


On Crises




Add to Technorati Favorites