Monday, July 18, 2005

Policy and Uncertainty

Economists have great faith in markets, that the result of the market process will be good more often than not. Less publicized is that many, even most, economists also have great faith in government, at least implicitly. The social planner is ubiquitous in economic models, even more so than perfect competition and rational expectations. While imperfect competition and myopic agents have memed their way into mainstream models, one always solves for the optimal policy where it is appropriate. But how many times has optimal policy been implemented?

Case in point: the Bush administration. Tax cuts may have been an excellent stimulus in the recession of 2001, but the Bush tax cuts were not nearly as effective as they could have been. A prescription drug benefit for Medicaid may have been a great idea to relieve elderly poverty but it turned out to be largely a boondoggle for pharmaceutical companies. Even if regime change in Iraq was justified in some sense (though I don't think it was), mismanagement has ruined any good that may have been accomplished. Free trade is generally good, but when agreements like CAFTA are loaded with unnecessary intellectual property provisions, its effect becomes more ambiguous. Social Security privatization could turn out to be a net good if done correctly. But the chances that that will happen are small.

Corruption is not rampant in the U.S. government. Yet ideal policies rarely, if ever, get carried out (even with Democrats in power.) This is true to a much greater degree in poor countries. Moreover, bungled policy implementation leads to increased opposition to similar, but much better policies, as in the free trade example. Perhaps we should focus more on developing policies that don't turn on a few key parameters and instead will be robust to many different political climates.

What does that imply? Consider a Pigouvian tax on pollution. Taxing activities that pollute is good, but only up to a point. No one knows with certainty where that point is. Overshooting that point by too much has a negative effect. A too-zealous liberal government may do just that. Or, consider a tax cut to stimulate demand. Too much tax cut and deficits will rise too far and squeeze out private investment leading to lower long-run growth.

In both of these situations it may behoove economists to perform some sort of Bayesian calculus as to the wedge between the policy proposal and the policy implementation. Let's be more realistic about what the social planner actually does and more risk-averse with respect to our policy recommendations.


Post a Comment

<< Home