In January 2019, co-blogger Bryan Caplan wrote:

The theory of market failure is a reproach to the free-market economy.  Unless you have perfect competition, perfect information, perfect rationality, and no externalities, you can’t show that individual self-interest leads to social efficiency.*  And this anti-market interpretation is largely apt.  You can’t legitimately infer that markets are socially optimal merely because every market exchange is voluntary.

Contrary to popular belief, however, market failure theory is alsoa reproach to every existing government.  How so?  Because market failure theory recommends specific government policies – and actually-existing governments rarely adopt anything like them.

What we also often see and, depressingly, even usually see, is that economists who are pro-government intervention to fix market failures have a much lower standard for the government than they have for the market. So the odds are that avoiding the specific government policy being proposed would get us closer to the optimum than implementing the government policy.

A case in point is Paul Krugman and his views on the recent $1.9 trillion spending bill. In Benjamin Wallace-Wells, “Larry Summers versus the Stimulus,” March 18, 2021, Wallace-Wells makes that point, although I’m not sure that that’s his intention.

Wallace-Wells, describing a recent debate between Krugman and Larry Summers about the Biden spending plan, writes:

Krugman asked, rhetorically, which elements of the package Summers would cut. Not the public goods, like vaccination and funds for school reopening, and surely not the needed income support. What was left was the part that members of Congress had most vociferously demanded: the aid to state and local governments (which Krugman agreed probably exceeded the fiscal need) and the checks to people who had not much suffered. Krugman said, “The checks, which are the least-justifiable piece in terms of standard economics, are also by far the most popular, and I don’t think we can entirely disregard that.”

Put aside the fact that the funds for school reopening are almost certainly not justified because the risks to students and teachers are so low. Notice what even Krugman admits. First, that the aid to state and local governments is too much, even by his standards. Second, the checks to people who hadn’t suffered much, which are a huge part of the package, are the “least-justifiable piece in terms of standard economics.” And what’s Krugman’s justification for those payments? That they are “by far the most popular” and, for that reason, we can’t “entirely disregard that.”

In short, in order to get hundreds of billions in spending that Krugman thinks are justified, he is willing to have the government spend other hundreds of billions for things that are not justified. Such is the nature of many, perhaps most, economists’ advocacy of government policy.

Note: The picture above is of a Rube Goldberg machine, which is what I think a lot of government policy is like. There is one difference. The Rube Goldberg machine always worked.