Case closed, right? Not so fast. Now let’s look at how conservatives fail to appreciate the way that people respond to incentives. Samuel Bowles divides economists into priests and engineers. The engineers are the ones who are getting their hands dirty with how real world people and economies actually work. Priests are the ones who sit down in ivory towers and reason about how economies work with abstruse mathematics or deductions from first principles. Conservatives sometimes go wrong by being priests.
Let's continue to use unemployment insurance. The conservative priest says "Unemployment insurance rewards shirking and punishes work. Therefore we'll get more shirking and less work. Therefore unemployment insurance is bad."
Problem One: The Size of the Effect Matters
There are two problems with this analysis. The first is that while we know that unemployment insurance creates an incentive to be lazy, we don’t know by how much. The strength of the effect matters. If unemployment insurance causes the unemployment rate to skyrocket then we might have a problem. But if it only causes a tiny increase then we probably don’t. Priests can't tell us that unemployment insurance is bad. We need an engineer.
That’s what Lawrence Summers and Kim Clark (1979) did. They went out and got the data and found that unemployment insurance doubles the number of cases where people are unemployed for three months or more. It also increases the unemployment rate by a half percent, such as from 5.5% to 6%. So the effect is actually pretty small. For every ten people who really need help, there is at most one person who is unemployed who would otherwise be working.
But it is the second problem that really gets conservative priests (not all economic conservatives) into trouble. We'll cover that in the next post.