INTRODUCTION 
quite arbitrarily chosen, and is not to be looked on as a year of ‘ normal ’ 
prices. It was in fact a year in which prices were considerably inflated 
by the South African wart 
44. Whatever be the cause of changes in index-numbers, the facts 
underlying those changes modify, and sometimes to an important 
degree, the significance of the values given for exports and imports. 
For example, if we take the periods of five years for which the average 
value of imports into the United Kingdom from 187 1-75 to 1906-1910 
is given on p. 702 we find that there is only one, namely, the period 
1886-90, which shows a decline in value as compared with the previous 
quinquénnium—in round numbers £390 a gainst £400 millions. Butif we 
apply the Board of Trade index-number, base 1900, to these figures the 
values become changed to £333 millions in 1881-5 and £379 millions in 
1886-90, showing an increase in the latter period of nearly 14 per cent. 
instead of a decrease of about 2} per cent. Now with an index-number 
calculated as explained, this shows that during thelatter period consider- 
ably more supplies of food and raw materials must have been coming 
into the country than in the one before. That being so, we may be sure 
that those increased supplies would find their way into the hands of 
consumers. Stocks are not kept on indefinitely in the hope of better 
prices. Perishable goods cannot be. So far as the increased imports, 
then, were food-stufs, they must have been a direct benefit to the 
consumers ; so far as they were raw materials, increased supplies must 
have helped to maintain the demand for labour, for they were imported 
in order to be used, and manufacturers still found their advantage in 
using them in spite of the fact that they did not see their way to sell 
the products at former prices. It is not even a necessary consequence 
that the lower selling prices of the products meant lower profits to the 
manufacturers. © The exports of a manufacturing country,’ observed 
Sir Robert Giffen, ¢ may be nominally affected by a change in the value 
of the previously imported raw material, although there is no real 
change in the native produce exported, or when the real change may 
be the opposite of the nominal one. Say that one-quarter of the exports 
consists of previously imported raw material, then a decline of 50 per 
cent. in the value of the raw material would produce a decline of 12} 
per cent. in the aggregate export, which would be entirely nominal. 
If at such a time the exports were apparently stationary, the real fact 
would be that they had increased 12} per cent., or rather about 17 per 
sent., allowing that the increase really takes place on three-fourths 
only of the nominal total.” 2 
45. Further, when these tables are used for making comparisons of 
the trade of different countries one may be led into error in various 
1 The still greater inflation of the great war having brought into relief the 
defects of former index-numbers, the Board of Trade adopted in 1921 an index- 
number on a new principle, an account of which is given by Mr. Flux in the 
Journal of the Royal Statistical Society, vol. Ixxxiv., pp. 167-99 (March 1921). 
2 Journ. of the Stat. Soc., vol. x1v. (1882), pp. 197-8. 
18