17 May 2012

Hysteresis and the Pretense of Knowledge

Imagine having a fever so bad that it permanently raised your body temperature. Now imagine a period of unemployment so bad that it permanently reduces our economy's ability to produce things and employ people. That's hysteresis -- the long-term scarring of our economy from periods of short-term unemployment. I've discussed this before, and I think the evidence is very convincing it is a major issue. Hysteresis is part of the engine in the recent Brad Delong/Larry Summers paper arguing for self-sustaining stimulus.
Crucially, hysteresis is an intellectual challenge to the so-called structuralists who would argue that we should ignore the short-term economy and just focus on the long-run health of the economy. Beyond us all being dead in the long run, the long run is just a series of short runs right after each other. And hysteresis shows that short-run problems can perpetuate themselves and become embedded in the long-run economy.
The fatal flaw in this analysis is that it presupposes that the future is knowable to a high degree of certainty and, more importantly, that the effect of theoretical tradeoffs are knowable and calculable.  This is, of course, nothing more than the pretense of knowledge, writ large.  There is no way to tell what an economy would “naturally” look like in the absence of a recession, nor is there any way of knowing whether a particular recession was avoidable or whether it would have been possible to reduce the severity of a given recession.

More specifically, it’s impossible to know what theoretical production looks like in the future in the absence or presence of given policies. Production may permanently decrease because, say, demand for the product may decrease.  Or maybe the new regulation that pops up to deal with a recession increase marginal costs.  Maybe there are structural problems.  There’s simply no way to tell, and playing “what-if” games does nothing to identify the problem, let alone determine a possible solution.

In short, the hysterics over hysteresis are completely unnecessary, and are likely a by-product of a few theoreticians perceived self-importance.  Thus, it’s easy to get worked up over potential problems when you think that a) you can accurately identify them and b) actually fix them.  If this isn’t arrogance, I don’t know what is.

Additional reading: My initial post on the subject.


  1. Hysteresis. No duh! Which is why bad banks should fail, bad loans should be defaulted on, assets should be marked to market rates, and CFOs should hurl themselves out of high office windows. Hysteresis is why we should NOT maintain illusions, NOT monetize public debt, and NOT blow bubbles in various sectors of the economy. Liquidate, liquidate, liquidate. It's as good a proposition today as it was in 1930. Instead we extend and pretend and wring our hands about hysteresis.

  2. @Steve N.- It is ironic, possibly hypocritical, that those who spend all their time theorizing about hysteresis use their theory to defend that actions that have made the problem worse, if not caused it outright in some cases.