Wagner's 'law of increasing state activity' has recently been subjected to
intensive use of recondite econometric testing procedures. An examination o
f the relevant articles reveals ignorance both of Wagner's definition of 's
tate activity' and of his insistence that he was not engaged in prediction.
Consequently, the omission of public enterprises contained in Wagner's def
inition leads to a model misspecification and the object of testing his hyp
othesiss remains obscure.