■gw*;..

82 A TREATISE ON MONEY bk. h

working classes. But such changes cannot affectthe price-level “as a whole ” or the value of moneyin itself. By a change in the value of money initself or in the price-level as a whole, they meanthe amount of the uniform residual movement dueto “ changes on the side of money ” after we have“ averaged out ” the chaotic but compensatorymovements in individual prices due to their move-ments relatively to one another and to the price-level. For the purpose of isolating “ changes onthe side of money ” they employ the Doctrine ofAverages based on the Theory of Probability . If wetake enough unbiassed observations of individualprices, their relative movements will, it is argued,cancel out in accordance with the law of error,and we shall be left—subject to a probable errorcalculated in the usual way—with a reasonablysatisfactory index of the residual movement of theprice-level itself which is our quaesitum,}

Three quotations from Jevons , Dr. Bowley 1 2 andEdgeworth respectively will indicate the line ofthought which I am trying to describe.

Jevons, Investigations in Currency and Finance,

p. 181:

The geometric mean seems likely to give in the mostaccurate manner such general change in prices as is due toa change on the part of gold. For any change in gold willaffect all prices in an equal ratio ; and if other disturbingcauses may be considered proportional to the ratio of change

1 This quaesitum, is the same as Edgeworth’s Indefinite Standard inSections VIII. and IX. of his Memorandum prepared for the BritishAssociation in 1887 (reprinted in his Papers relating to Political Economy,vol. i. pp. 233 el seq.). His Section VIII. seeks the “ Determination ofan index irrespective of the quantities of commodities : upon the hypo-thesis that there is a numerous group of articles whose prices vary afterthe manner of a perfect market, with changes affecting the supply of money ”.Section IX. seeks the “ Determination of an index utilising quantities ofcommodities : upon the hypothesis that a common cause has produced ageneral variation of prices ”.

2 Cf. also the first part of Dr. Bowley’s “ Notes on Index Numbers ”( Economic Journal, June 1928).