Chicago school economists are often maligned  for their supposed blind faith in markets. And it is true that some  of the theories associated with Chicago have a certain Panglossian feel to them; they give the impression that markets everywhere and always yield the best possible results.
But as Milton Friedman noted in an interview for A Modern Guide to Macroeconomics (p. 174 of the first ed. ), one need not have blind faith in markets to think that government intervention will make matters worse:
I believe that what really distinguishes economists is not whether they recognize market failure, but how much importance they attach to government failure, especially when government seeks to remedy what are said to be market failures.…Speaking for myself, I do not believe that I have more faith in the equilibrating tendencies of market forces than most Keynesians, but I have far less faith than most economists, whether Keynesians or monetarists, in the ability of government to offset market failure without making matters worse.
With this, the founder of the Chicago school was articulating a notion more closely associated with the Virginia School of Political Economy  than Chicago. The Virginia School emphasizes the inherent biases of public policy and the ways these biases can make government intervention fall far short of the imagined ideal:
- The incentives of political action  are such that concentrated, established special interests have a significant advantage over diffused, general interests.
- The probabilities of voting are such that individual voters have little incentive to gather information .
- Nor do they have an incentive to rationally process the information they do gather .
- The incentives of simple majority voting  are such that unconstrained majorities are able to impose costs on minorities (and the costs borne by the losers outweigh the gains of the winners).
- The powers of agenda setters  are such that chairmen of congressional committees are able to steer outcomes toward their most-preferred policies by determining the order in which votes are taken.
- The incentives of a bureaucracy  are such that bureaucrats are able to act as monopolists providing low-quality service at greater-than-necessary cost.
At every step of the political process, perverse incentives ensure that economic-policy-in-reality is a far cry from economic-policy-as-it-is-ideally-envisioned. So even if there is a rationale for government correction of market failure, it is irresponsible to ignore the very real possibility that government “correction” of market failure often makes matters worse.
In a post last month, I used this line of argument to critique  an article by Dylan Matthews  on fiscal stimulus. I noted that if the macroeconomics of stimulus look murky, the public choice of stimulus look downright bad: the incentives of democratic politics do not encourage voters, politicians, political aids, or bureaucrats to implement stimulus as Keynesian theory says it ought to be implemented.
Yesterday, Matthews helped me make my case. In a superb post  he pointed to the results of a recent survey which found that fully 15 percent of Ohio Republicans are willing to give Gov. Romney credit for killing Osama Bin Laden. To help explain this bizarre result, Matthews cited research  showing ideological beliefs tend to affect voters’ assessment of objective facts. I’d note that this doesn’t mean people are dumb. It just means that the political process–in contrast with the market process–does not reward information gathering or information processing .
This should make one more skeptical of stimulus, and other government solutions.