Nassim Taleb and Mark Blyth have a great article in the new issue of Foreign Affairs that makes the case against government efforts to suppress volatility in the political and economic worlds. Their basic argument, as I read it, works along two lines. The first is pretty common: squeezing volatility out of big slices of any complex system inevitably means shifting it elsewhere, in potentially ugly ways. Banks, for example, found ways of getting steady returns 95% of the time out of what should have been a turbulent housing market – but the price was far more catastrophic consequences when the other 5% hit.
The second line of argument strikes me as more interesting. People are bad at preparing for low probability risks, even if they entail high consequences. But they’re actually decent at preparing for higher probability, lower consequence dangers. Moreover, in the process of protecting themselves against these more likely problems, they implicitly protect themselves against the lower likelihood, higher consequence dangers too. Eliminate most but not all of the volatility in the system, and you get a society that’s woefully under-prepared when things get really bad. If you let the volatility show itself more often, things may be a bit more unpleasant most of the time, but society will be much better prepared when the extremes hit.
The authors apply this mostly to Middle East politics (and to a lesser extent to economic management), but it’s a great way to think about oil prices. Five dollar gasoline is unlikely but highly consequential. Four dollar gasoline is more likely but somewhat less problematic. If the U.S. government prevented four dollar gasoline (say by managing prices using the strategic petroleum reserve), it would reduce incentives for fuel economy and other behavioral change, as well as for private stockpiling and hedging. If five dollar gasoline actually hit, people and firms would be more exposed, and the consequences would be more extreme. Of course, if the USG felt confident that it could prevent five dollar gasoline too, then this strategy might make sense. But unless the extreme bits of volatility can be completely eliminated, it’s risky to eliminate the milder ones too.
Obviously, there are limits to this. One might make the case, for example, that while exposure to some gasoline price volatility is generally useful, it’s a bad thing right now, while the economy is weak. That could be the foundation for an argument in favor of suppressing volatility right now, even if it might leave the economy marginally more exposed later. But this is a slippery slope. It is also a dangerous one, since there is no guarantee that the bigger risks will wait until the economy is healthier to show up. As a general rule, if you can’t banish volatility entirely, it’s risky to try and suppress it only some of the time.