ON VAKIATIONS IN THE VALUE OF THE MONETART STANDAUD. 143 



UBfinished articles may be taken as more or less perfect representatives 

 of consumable commodities. 



The case of exports may be thus fitted to this interpretation. It is to 

 be assumed that, given the steadiness in the course of trade which we 

 have postulated, an increase in tlie volume of exports normally corresponds 

 to an increase in that of imports. Thus exports afford a measure of tho 

 advantage derived from foreign trade of the same sort as that which 

 imports afford.' 



There is a special difficulty in the case of those articles which are 

 imported like cotton in order to be re-exported at a subsequent stage of 

 manufacture. Take the extreme case, mentioned by Mr. Giffen, of tea, 

 which figures as part of the domestic produce exported from France. 

 The French of course derive some advantage from tho handling of this 

 article. But the interest which they have in the tea thus transmitted is 

 not proportioned to the value of the article in the same sense as the value 

 of a genuinely native export measures its importance to the nation. 



With regard to this special difficulty, and indeed the whole computation, 

 it is to be remarked that we are concerned — not so much with the 

 absolute volume of trade — as the relative volume in one year as compared 

 with another. The relative volume as already stated may be regarded as 

 a sort of mean of the ratios between the quantity (in tons, gallons, &c:) 

 of each commodity in one year and the coiTCSponding quantity. It is 

 a weighted mean, the weights being the respective values of the 

 commodities.^ Grant, now, that in the proposed case of tea transhipped 

 from France the weight is exaggerated. Yet, as pointed out by the 

 writer in a former Memorandum, some inaccuracy of the weights is not 

 likely to affect the result much. It is only in the case of the larger values, 

 notably cotton imported into the United Kingdom as the material for 

 future manufacture, that the difficulty is serious. Such items ought no 

 doubt to be placed in a separate category, and considered on their own 

 merits ; not merely on account of their inaccuracy, but also on account of 

 their mere magnitude. The domineering pre-eminence of one or two items 

 is fatal to the application of the Calculus of Probabilities which flourishes, 

 so to speak, only in a republic of numerous independent not very unequal 

 constituents. 



Implicated with this definition of the volume of trade there is a de- 

 finite method of measuring the variation in the value of money. This 



' It may be objected that the volume of trade is per se, and apart from hypo- 

 thesis, an interesting datum, or rather qiicesittim, as sifiording the measure of profits 

 accruing to the country, or for some such reason. This remarlc seems just if the cor- 

 rections of the Monetary Standard which are made for the purpose of estimating the 

 volume of trade are based upon some principle extraneous to the trade, or at least 

 some other principle than that of assigning to each article an importance pro- 

 portioned to the value exported or imported. All that is contended here is that the 

 received method of measuring the trade by itself, so to speak, postulates a certain 

 analogy between this species of Index-number and the more general one which is 

 based on national consumption. Indeed, it is partly on account of this analogy 

 that the subject appears to deserve such full treatment here. 



' In the symbols to be presently introduced the ratios of quantity are of the form 



£iw, ?^, &c. 

 The correspondiDg weights are ^ar i'on ffsr Pbxt &c. Thus the weighted moan is 



Qcu Pax + Iby 2'bx + &0« 



