The “Rebound Effect” Falls Flat

Prompted in part by a recent article in the New Yorker, there’s been a lot of attention to the rebound effect lately.  The theory is that increased energy efficiency in effect makes energy cheaper (as measured in cost per unit of benefit), so people actually consumer more energy.  The empirical evidence is that this is a relatively small effect, far outweighed by the energy savings from efficiency, as Greenwire reports.  But some people argue that the rebound effect actually outweighs the energy savings.

This  version of the rebound effect  is reminiscent of the Laffer Curve, which said that the way to increase government revenue was to cut taxes.  If that’s true, then “small government” advocates like the tea party should be lobbying for tax increases in order to starve government!  Similarly, if the rebound effect were as big as advocates say, we should ban hybrids and pay a bounty to keep old gas guzzlers on the road longer, while also giving a tax credit to people for removing insulation from their houses.

This also reminds me of the argument by a prominent University of Chicago economist that the best way to improve auto safety would be to install a large spike in the middle of the driver’s wheel, which would create a large incentive to drive safely.

There’s a sort of appealingly counter-intuitive and bold nature to all claims of this type, which essentially say that moving in one direction is the best way to make progress in the opposite direction: lower taxes to raise government revenue, lay off government workers to raise employment, give more people guns to cut gun violence, decrease energy efficiency to save energy. Or, on the liberal side, reduce drug abuse by legalizing drugs, or increase employment by raising wages and making it more expensive for employers to hire.

There is a heavy presumption against all of these claims, which should require strong evidence to overcome.  Some of these claims may be true, but they should never be accepted on faith (or on the basis of an unvalidated model, which amounts to the same thing).

Most actions produce a host of indirect effects, some contrary to the direct effect. It’s always possible that the indirect effects of an act will cause a large enough boomerang to outweigh the direct effects, but the odds are against it — indirect effects usually aren’t that large and uni-directional.  Usually, the best way to get from one place to another is to move in the direction of your goal rather than the opposite direction.

, ,