In 2008, Ben Bernanke’s and Paul Krugman’s star former doctoral student at Princeton, the economist Gauti B. Eggertsson, published an article in the American Economic Review, the field’s top journal, on the manifest excellence of the New Deal of the 1930s. One of the claims: “1933-1937 registered the strongest output growth (39 percent) of any four-year period in US history outside of wartime.”
The cited source for this statement was the Office of Management and Budget, which does not maintain statistics on “output growth,” or GDP growth, from before 1929. It is not possible to use this source for a claim about “any four-year period in US history.”
As for databases that do include the whole run of U.S. history, those at measuringworth.com show an increase in national output of 43% from 1878 to 1882, a peacetime period.
Either 1933-37 was not the greatest peacetime four-year run of growth in American economic history, or the issue is unsettled, for lack of canonical sources. The untenable position was that presented in the AER. Not only is the 1933-37 claim probably wrong as a matter of fact; attached to the claim was a faulty apparatus of verification.
Chalk it up to yet another elision in the academic journals. But once these things are out there, they take on a life of their own, re-cited and re-enunciated as they are by readers and then their own readers and listeners in turn. I have certainly heard historians who teach considerable number of students approvingly quote Eggertsson’s claim about the supremacy of 1933-37 output growth.
In scholarship, often what happens is that stuff is put out there, it’s erroneous, it spreads, a mess results, and the mess has to be cleaned up. We seem now to be in mess stage about the macroeconomic quality of the New Deal.