Saturday, 16 January 2016

Stigler and the Unbalanced Keynes.

"His statistical diagrams [W.S. Jevons'] were so precisely prepared that I have been able to recover his data in several cases with sufficient precision to reproduce his averages to two significant figures. It is rare to be able to do that, even with modern computer drawn graphics. Indeed, it is rare today to be able to reproduce averages even if the data are published!"
Prof. Stephen M. Stigler is Ernest DeWitt Burton Distinguished Service Professor at the Department of Statistics of the University of Chicago. He's also the son of that Stigler.

During a long academic career, Stigler the Younger has written extensively on the history of statistics, with occasional stopovers on allied subjects. In his 2002, in equal measures interesting and entertaining short paper (Statisticians and the History of Economics, Journal of the History of Economic Thought, 24(2), pp.155-164. paywalled) Stigler evidences subtle humour, wit, and irony while commenting on a set of 39 authors selected for their contributions to both statistical theory and economics.

Stigler puts those authors in subsets (e.g. Precursors, Masters, the Great Middle Class, among others) and provides passing or at most brief references to most authors (like the Middle Class Jevons in the opening quote). His exposition centres on Francis Ysidro Edgeworth -- one of the Masters -- whom Stigler contrasts with another prominent figure:
"There is one major figure on this list that I could be accused of slighting, and I want to explain why. I refer to the Unbalanced John Maynard Keynes. Possibly someone from Chicago needs no excuse for referring to Keynes as unbalanced, but I offer one anyway."
For Stigler, Keynes' "immense reputation in economics was so far out of kilter with his reputation in statistics". Keynes' 1921 "Treatise on Probability" enjoys "a reputation principally among philosophers … that seems to me to far outstrip its influence or merits".

Stigler, following Merton, sees in that undeserved reputation an instance of the so-called Matthew Effect: "To him that hath shall be given". Bluntly: because of Keynes' holy cow status in economics (at least for his fans), the Treatise is presumed without study to be another proof of his genius.

Regardless, although generously remarking that "to a degree [Ronald] Fisher's criticism could be judged unnecessarily harsh", Stigler however seems to embrace each and every single one of those harsh criticisms. More importantly,
"Edgeworth himself reviewed Keynes's book, and his review was essentially as scathing as Fisher's, and on much the same points. But Edgeworth was so polite in his expression that a reader could come away with a quite contrary impression. Keynes would have understood, but the soft words required no response. Keynes did not see the need to reply."
However, unlike Fisher and Edgeworth, Stigler the Younger is not uncharitable to Keynes the Younger:
"Keynes removed himself from probability after the book was published, a wise investment in human capital, in my view."
Let's not, for once, quibble with the "human capital" thing.

No comments:

Post a Comment