Yesterday, I cited Granger and Sims for trying to convert a fallacy into a principle by introducing a purportedly more sophisticated version of the fallacy.
Today, I accuse Granger and Newbold of abusing the English language, quoting myself from a 1993 essay:
“Only in the early phase of [Granger’s] empirical innovation was it made clear that [his] tests are based upon a newly stipulated definition of the word “cause.” Stripped of all its subtle and difficult philosophical content and of its etymological link with reason, the word “cause” is used to describe observed temporal patterns in time-series data. In the judgment of Clive Granger and Paul Newbold, “A better term might be temporally related, but since cause is such a simple term we shall continue to use it.” [!] It is interesting to note that, though this usage is defended on the basis of simplicity of expression, economists who employ empirical techniques developed by Granger use the decidedly unsimple and unaesthetic term “Granger-cause,” as in: Falling interest rates Granger-cause wage rates to rise....
“The linguistic technique introduced by Granger is nothing short of a scandal. (A better term might be career enhancing innovation, but since scandal is such a simple term I shall continue to use it.) Publishers and editors are not likely to be interested in research that yields limp conclusions about the temporal relationships in the movements of economic variables; they are interested in research that demonstrates that one thing causes another.”
I can’t help but note that yesterday’s announcement of the Nobel Prize winners demonstrates just how “career enhancing” Granger’s abuse of the English language was!
