138 KEPOET— 1889. 



Suppose there were a set of outsiders, too young or too old, too wise,' 

 or who thought themselves too grand to take part in the game, but do 

 not refuse to accept a pension from the bounty of the players — a pension 

 enjoyed in sugar-plums but paid in counters. For this class, certainly, 

 it would be most convenient that the appreciation should be based npon 

 sugar-plums, so that pensions should afford a constant value in use. 

 But in the interest of the actual players, puzzled and damnified by the 

 change of denomination, ought not the measurement to be grounded on 

 the pieces ? It is tenable that the ratio in which the number of counters 

 should be increased is (the reciprocal of) the ratio of the total value in 

 counters of the whole number of pieces moved, in a game at the first 

 period at the rate then prevailing, and the total value of the same pieces 

 at the rate prevailing in the second period. 



Suppose that the counters consisted of two sorts, one metal and the 

 other paper, exchanging indiscriminately with each other, and with 

 pieces and sugar-plums, yet demarcated by some material differences. In 

 particular the metal symbols are more under the control of the managers ; 

 while the paper tickets are provided by the players themselves. Should the 

 appreciation be expressed by the deficiency in metal counters, or in 

 counters generally ? Prima facie surely in terms of the latter, though, as 

 indicating the duty of the managers, the result may sometimes be 

 expressed as a deficiency in metal ; in terms of dose rather than 

 diagnosis. 



Upon the whole, it appears that the Currency Standard deserves more 

 attention than it has received. The stone unaccountably set aside by 

 former builders of Index-numbers may become the corner-stone of future 

 constructions. 



It is not to be thought because the proposed method is likely not to 

 be so revolutionary in practice as it is distinctive in speculation,^ that 

 therefore it is unbefitting a separate and high place here. For we are 

 concerned here with distinctions of method rather than differences of 

 result. There is attempted here — to illustrate small things by great — 

 for a particular province of industry, the sort of analysis which an 

 eminent member of our Committee has performed upon the ' Methods ' of 

 conduct in general. In the sphere of Finance, as well as Ethics, theo- 

 retical distinctions arc important, although they may not correspond in 

 practice to such marked discrepancies as might have been expected. 



Nor is it a fatal objection to the scheme that it would be impossible to 

 ascertain with precision the proportions in which each commodity 

 absorbs, or exercises a pull upon, the currency ; that here the number of 

 resales, and there the exceptional use of credit, would defy calculation. 

 For, regarding the proposed Index-number as a Weighted Mean of nume- 

 rous given variations of price, we see that the objection amounts to saying 

 that the weights are liable to a considerable error. But, as shown in a 

 former Memorandum, and to be insisted on again in the present one, the 

 erroneousness of the weights is likely to produce much less error in the 

 computed mean than might have been expected. 



' The most direct application of the Consumption Standard is in the interest of 

 annuitants, fellows of colleges, and those whom Mill calls ' idle ' landlords. 



^ Thus the example which we have Imagined is probably an extreme one ; yet it 

 presents a difference between the compared Index-numbers of only seven per cent, ; 

 which, in view of the 'probable error,' say two or three per cent., to which a»jy Index- 

 number is liable, cannot be considered as colossal. 



