ON VAKIATIONS IN THE VALUE OF THE MONETAE Y STANDARD. 273 



-i-. Or, if we suppose several such dealers, we have the weighted mean 

 p. 



T.jtJlU—l — ^ (assuming that the quantities have not materiallv varied") 

 ^P. + t^Ve +&C. 



between two revisions, and that the ' commission ' of all the dealers may 

 be regarded as the same. 



Consider next residential rent and stipendiary wages. The incomes of 

 certain classes vary directly with these payments ; yet, as tliese incomes 

 are not, like the preceding, equal to a small fraction, but to the entire 

 volume, of the transactions in question, it will not be easy to combine 

 these data with the preceding into a properly weighted mean. 



Again, when we take in ordinary wages and industrial rent, we are 

 met by the fact that, while the income of some classes varies directly with 

 these amounts, the interest of another class, entrepreneurs, varies inversehj 

 — not indeed in exact inverse ratio, but in an opposite direction to the 

 same quantities. Again, the materials of one manufacturer are frequently 

 the finished products of another. Accordingly the price of such articles 

 constitutes a very bad measure of the income of all the parties concerned. 



It follows from these considerations that from an examination of 

 prices we can obtain at most a very rough and precarious indication of 

 the variation of resources. Such a method would be related to the more 

 exact calculation of income very much as our method A B C d was related 

 to ABC D. 



At the same time, when we consider the purpose of our sliding scale to 



mitigate the evil of industrial fluctuations — it may be doubted whether this end is 

 not realised nearly as well by a rough-and-ready method as by the most exact 

 calculation. For a standard based upon the vicissitudes of all cannot well be 

 adapted to the vicissitudes of each. The lit is at best so bad that it is not made 

 much worse by some additional imperfections of measurement. 



The character and worth of such a mean variation of price as we here desider- 

 ate might be illustrated by an imaginary example of another sort of mean one 

 obtained by taking the average temperature for the same day over a period of 

 years. We have known old ladies who each year discontinued and resumed fires 

 on the same days of the year. Suppose that they had affected even o-reater 

 precision, and had burned each day a quantity of fuel based upon the mean 

 temperature for that day averaged over a period of years. It is clear that in a 

 climate like oms those who adopted this arrangement would some days suffer 

 from too great heat and other days from too great cold. The arrangement would 

 be so very defective that it would not be sensibly deteriorated by some imperfec- 

 tions in the method of averaging the temperatures. Suppose, for instance, that in 

 the different years the thermometrical measurements had been effected with diflerent 

 degrees of completeness. For the earlier years there might be (for a given day) 

 only sample readings of the thermometer, made two or three times a day. For the 

 later period there might be a more continuous record of temperature. Theoretically, 

 in combining such data more weight should be given to the more complete measure- 

 ments. But practically for the purpose in view such elaboration would be nuo-atorv. 



To look at the matter more closely, let us suppose with sufficient accuracy that 

 the income of a particular class of producers depends mainly on the prices of a 

 certain group of articles, so that it woidd be convenient for that particular class 

 that the standard for deferred payment should be regulated by the movement of 

 those particular prices. Roughly speaking, the desideratum for that class is 



that the unit should be proportioned to some mean of those prices ; say ?^F ~M"" ' 



mp n 

 where p and n are prices of products and agents of production respectivelv- But 

 in fact the unit must be based on the prices (and quantities) of all kinds of 

 1887. T 



