In the middle of a generally apposite article, Steve Lohr of the New York Times says, “During the 1930s, the unemployment rate fell somewhat under Roosevelt, but remained stubbornly high, averaging more than 17 percent for the decade.” Dean Baker caught this, and so does Cosma Shalizi via the email: you get an average 17% if you count people who had jobs as unemployed.1 Which, Baker notes, is not what you’d call “in keeping with current methodology”. Or, you know, helpful in trying to make sense of the situation.
Baker goes on: what the New Deal achieved was “still far from full employment, but it is less than half the 23 percent rate that Roosevelt faced when he took office.” And in saying that perhaps the New Deal could have done better, Lohr’s overall point is sound: bailing out the banks wasn’t good enough, you had also to get creditworthy people lining up to borrow. In the 1930s, this was taken—by, e.g., Keynes—to mean you needed some fiscal stimulus: “it is no use creating a demand for credit, if there is no supply. But an increased supply will not by itself create an adequate demand.”
One might also point out that in the 1930s, the massive purchase of bank equity came after auditing the banks’ books to see what assets they had, and shuttering the ones that couldn’t survive.
1Also, it seems to me Lohr must count the decade as running 1929-1939 inclusive; this—using the old Lebergott data—gives you an average rate of 17.05%; if you use the current HSUS data for the same period, you get 13.16%. If you look at Roosevelt’s first two terms—1933-1940—you get 18.7% (Lebergott) as against 13.0% (HSUS current).