This Will Likely End Badly

Tyler Cowen and Alex Tabarrok, the two bloggers at Marginal Revolution, are rightly impressed with GiveDirectly. In a post yesterday, Alex points out that four economists started GiveDirectly. That’s figuratively putting their money where their economists’ mouths are because economists tend to believe that the most efficient way to help people is to give them money and let them spend according to their priorities. I basically agree. I think some people will spend it on drugs and alcohol and not on their children but that doesn’t mean that the government’s choices would have been better: I think that a lot of people who get money spend it well, where “well” is vaguely defined as spending it on things that will enhance their families’ lives longer term.

In his post yesterday, Alex Tabarrok reports that the city government of Chicago is getting GiveDirectly to administer a program “that will give $500 a month to each of 5,000 households in Chicago as soon as the end of June.” The city government is getting those funds from the federal government.

I see two problems that could well cause this plan to end badly. I’m leaving aside the issue of whether a Universal Basic Income is a good idea. I think it’s not, for reasons I laid out at length in an article in 2015.

The sense in which I think this could end badly is that the $31.5 million is a large enough number that it could distort how GiveDirectly functions. There will potentially be two levels of oversight from government officials: oversight from federal officials and oversight from Chicago city government officials. Governments tend to like to get their hands in things, dictating how various recipients of aid will act. They could do so in two ways: (1) by regulating how GiveDirectly acts and (2) by regulating how the recipients of the funds act. My bigger concern is (1).

If the government intrusion is large enough, it could turn GiveDirectly into something quite different from the organization that the four founding economists envisioned.