CH. 4 THE CHOICE OF UNITS 39

change;i.e. he is covertly introducing changes in value.Moreover, he is unable to devise any satisfactory tor-mula1 to evaluate new equipment against old when,owing to changes in technique, the two are not iden-tical. I believe that the concept at which Professor Pigouis aiming is the right and appropriate concept for eco- nomic analysis. But, until a satisfactory system of unitshas been adopted, its precise definition is an impossibletask. The problem of comparing one real output withanother and of then calculating net output by settingoff new items of equipment against the wastage of olditems presents conundrums which permit, one can con-fidently say, of no solution.

(iii) Thirdly, the well-known, but unavoidable, ele-ment of vagueness which admittedly attends the con-cept of the general price-level makes this term veryunsatisfactory for the purposes of a causal analysis,which ought to be exact.

Nevertheless these difficulties are rightly regardedas“conundrums.” They are“purely theoretical” inthe sense that they never perplex, or indeed enter inany way into, business decisions and have no relevanceto the causal sequence of economic events, which areclear-cut and determinate in spite of the quantitativeindeterminacy of these concepts. It is natural, there-fore, to conclude that they not only lack precision butare unnecessary. Obviously our quantitative analysismust be expressed without using any quantitativelyvague expressions. And, indeed, as soon as one makesthe attempt, it becomes clear, as I hope to show, thatone can get on much better without them.

The fact that two incommensurable collections ofmiscellaneous objects cannot in themselves provide thematerial for a quantitative analysis need not, of course,prevent us from making approximate statistical com-parisons, depending on some broad element of judg-ment rather than of strict calculation, which may

1 Cf. Prof. Hayek’s criticisms, Economica, Aug. 1935, p. 247.