It may be noted that when prices change by reason of differences in the demand for and supply of com modities, some commodities may rise in price, while others fall. Such readjustments in prices of com modities are constantly taking place. On the other hand, should price changes be wrought by changes in the value of money, we should expect uniform action upon prices, a general rise of prices or a general fall.
12. Measuring price changes.—Simple as these propositions appear to be, no little difficulty is ex perienced in demonstrating their truth by an appeal to the facts. In actual experience the two sets of forces fixing prices, those affecting commodities and those affecting money, are never separated. Partic ular price movements run counter to general price movements and obscure them. Like those who are unable to see the forest for the trees, there have been some who were unable in the shifting of many prices to discern any general movement whatsoever. They have therefore been led to question whether changes in the monetary situation have had any effect upon prices and have been ready to ascribe such changes to almost anY cause but the true one.
13. Index numbers.—The study of prices has long been a subject of particular solicitude, not only to the men actually engaged in business but to the students of business affairs. To separate the general from the particular, economists have invented what is known as the index number. The purpose of this device is to ascertain the average change in prices of a group of commodities. The prices of a given period or a given time are selected as a base and the price of each com modity is deterinined.
The advantage of taking as a base the average price for a series of years is that this course makes it possible to eliminate as far as possible the influence which, in a shorter period, unusual conditions might have had on the price of a particular commodity. When the base price is ascertained other prices during, before and after the period of the base are expressed as percentages of this base. Commodities A, B, c and D, will each have a different base, but when the base is turned into a percentage it becomes uniform for all, namely 100. We may suppose that at a later period A is represented by B by 106, c by 102, and D by 98. Now if these figures be averaged the result is 104, and this represents for the group an average rise in price of 4 per cent over the base. One of the
commodities indeed declined in price, but this does not alter the fact that the general tendency is to ad vance the price.
In following the method, briefly indicated,. of con structing an index number, each article has been given an equal importance in determining the change. If instead of abstract designations B, C and D, con crete articles had been named, such as wheat, iron, wool and indigo there might be some doubt as to the propriety of giving each an equal importance in fix ing the result. To avoid such questionings various methods have been devised to give an appropriate weight to the different articles; such weights have been arranged in accordance with the foreign trade, or with the estimated national consumption or the expenditure of workingmen's families.
None of these highly complicated methods have, however, led to substantially different results from the simple arithmetical average. It is clear that if each article moved upward in price exactly 5 per cent it would not matter how many articles were added to the list nor how they were combined, and one could not get any other general result than a five per cent advance. If some articles increased four per cent and others six they would, if equally divided, show an average of five per cent, but no method of combination could make the advance less than four, or more than six per cent. In other words, in a group of prices the general tendencies outweigh the particular ones, and no combination of different re sults according to any plausible system of weights will emphasize the particular tendencies at the cost of those which are general.
14. Course of prices.—Various index numbers have been compiled in many countries. Differences due to a different choice of articles, different initial bases for the calculations, and occasionally different methods of grouping or averaging the price indexes of the different commodities produce different price levels for the same date in the different indexes; but the movements of prices show a remarkable similarity. The oldest of these calculations are those of the Lon don journal, The Economist. _Another English com putation of wide repute, that of Mr-Augustus Sauer beck, has been continued by Sir George Paish in The Statist.