For some reason the debate around the top levels of unemployment in the 1930s has been particularly hot. The question has been whether you take the official data, as John Kenneth Galbraith did in his Great Crash book and the federal government has (“Lebergott”) or whether you also count the part-time make work jobs of the New Deal. Even if you include those you get unemployment rates we would find incomprehensible — over ten percent, lots of the time. It’s hard to see therefore why many observers have therefore spent so much energy on that. Eric Rauchway, whose own book is pretty good, for some reason feels very defensive about all this. Rauchway’s book I’ve listed in my NYU Stern School course. He’s even criticizing Jon Stewart.
I’ve had my hesitations about fooling around with the make-work New Deal jobs. One reason is that they were often jobs for just a few months. We are not talking about jobs teaching school, but rather true project-to-project tasks.
On the question of the reason for the unemployment, as opposed to the precise datum, I recommend Richard Vedder . Vedder early on, in Out of Work explained that unemployment of the 1930s was probably due to unusually high wages. Read: Wagner Act. NRA. Boosterism. Fordism.
But the big issue: was the unemployment under Hoover and FDR truly severe? Maybe it is time for another measure. Lee Ohanian of UCLA looks at the whole problem by different data and finds the answer is truly clear.