Why do economists claim the stimulus worked?

Bryan Caplan points to a piece by Justin Wolfers.

Let’s start with Obama’s stimulus. The standard Republican talking point is that it failed, meaning it didn’t reduce unemployment. Yet in a survey of leading economists conducted by the University of Chicago’s Booth School of Business, 92 percent agreed that the stimulus succeeded in reducing the jobless rate. On the harder question of whether the benefit exceeded the cost, more than half thought it did, one in three was uncertain, and fewer than one in six disagreed.

We here at the Cat are the 8 percent.

Let’s look at Caplan’s critique first.

Wolfers says that the panel is “ideologically diverse.” When I asked Kashyap, however, he said that there’s no public data on panel members’ political views. If you casually peruse the list, its members seem to lean heavily Democratic. Dan Klein’s systematic empirics say that the economics profession has Democrat to Republican ratio of 3:1. None of this would be a problem if becoming an economist caused people to join the Democratic party. In my experience, though, most economists picked their party long before they started studying economics.

Okay – so most academic economists are part of the highly educated elite and have political views consistent with that status. Not surprising – both Hayek and Schumpeter have theories of why intellectuals are likely to have left-wing views. Caplan goes on to talk about the stimulus.

My complaint: These results are basically what you’d expect from a non-expert panel with two Democrats for every Republican. What’s the value-added of the IGM’s economic expertise on this question? Hard to see.

Partisan bias seems particularly troubling when the IGM deals with policies that have recently been in the news. When economists analyze events decades in the past, it’s relatively easy to put politics aside and coolly apply abstract economics to concrete cases. When they analyze events they recently lived through, however, objectivity is harder to achieve. This is especially true when they’re personally close to the administrations that adopted the policies they’re now asked to judge.

I’m not convinced – the evidence is in. I’m happy to believe that people could be wrong ex ante, but ex post? Not so much.

Here is an earlier version of a very famous graph.

A model was used to generate two series of estimates in that graph. First the unemployment figures without a stimulus and then the unemployment figures with the stimulus. The red dots reveal what actually happened. The red dots invalidate the model. If you believe – as do 92 percent of leading US economists in the sample believe – that “the stimulus succeeded in reducing the jobless rate” then you must also believe that the stock standard Keynesian model that generated both sets of forecasts in the graph is wrong too. Now some argue that the stimulus was too small, but why weren’t those 92 percent of economists saying so at the time? Of course, that simply raises the question; how did they know it was too small at the time? Where is their model and its predictions?