When Should the Government Intervene?

When should the government intervene? In introductory economics classes, we introduce market failures. In intermediate economics classes, we discuss them in greater detail. In advanced and field courses, we get into the nuts and bolts of different ways pure exchange doesn’t get the job done and how, theoretically, a government could improve things.

Theoretically. The assumptions most textbooks make about governments, their capacities, and their motivations. A perfectly benevolent, omnipotent, all-knowing government would set a tax equal to the marginal external cost upon observing an externality. That’s a pretty heroic assumption, and it’s important not to mistake the model for reality and its assumptions for plausible descriptions of the world as it is.

To this end, a presentation from the philosopher Jason Brennan inspired a paper I’m contributing to a symposium in the Journal of Private Enterprise on the contributions of my late friend and Liberty Fund contributor Steven G. Horwitz. Brennan and his coauthor Christopher Freiman explained how all the arguments for regulating consumer choices apply equally to political choices. Sure, we’re irrational and weak-willed at the supermarket. The problem is worse in the voting booth because our incentives are even worse. 

Brennan and Freiman explain that there are, therefore, a few conditions the government has to meet before it can override people’s choices. I summarize them here. First, we should ask whether or not the problem we want the government to solve is an unintended consequence of another government policy. Are you worried about the environment and insufficient urban density? Let’s examine how government policies discourage urbanization and density (single-family zoning, for example).

Second, the private sector has already fixed a lot of market failures. Are you worried about externalities from secondhand smoke in restaurants and bars? That’s already capitalized into wages and prices–and one unintended consequence of urban smoking bans was that people drove to the suburbs to drink–and then killed people driving home.

Third, it’s not always clear the governments we actually have run by the people we actually elect who face the incentives they actually face will improve things. “Affordable housing” could be fixed tomorrow if we got rid of rent control and the layers of red tape preventing new construction. Telling people you support policies that reduce their property values or challenge their most cherished beliefs is a pretty nifty way to ensure you don’t get reelected.

Fourth and fifth, it’s incumbent upon us to ask whether or not the policy passes a cost-benefit analysis and how it changes people’s incentives. The Transportation Security Administration, for example, is a monumental waste of resources with a cost per life vastly above the cost per life saved by other policies and initiatives. The TSA also made flying less convenient and induced more driving–which meant more highway deaths.

Finally, even if policies pass all these tests, people might have rights that trump the proposed policies. Even if eugenic forced sterilization policies passed all the other tests–which they almost surely wouldn’t–they violate people’s bodily autonomy and are therefore disqualified.

How should we decide when and where to intervene? As Steven Horwitz and I argued a long time ago, market failures are necessary but not sufficient conditions. Before trying to cure this or that ill, we must make sure the cure isn’t worse than the disease.