INTRODUCTION 
J 
proached and on the questions which the evidence collected 
is designed to answer. If merely a rough generalization or 
trend is desired, the data naturally may be much less refined 
and complete than where, for example, wages are to be 
fixed in exact accord with the changes in living costs. 
Moreover, it must be borne in mind that there is adifference, 
also, between expenditures and costs. Expenditures are 
what is paid out, and involve other considerations than the 
price level; costs are a measurement of the price level. To 
measure changes in expenditures is to measure to some extent 
changes in the standard of living; to measure changes in 
prices assumes the maintenance of the same standard of 
living.2 
The present volume has been written for the purpose of 
bringing together and interpreting existing information on 
the cost of living, and particularly on changes in the cost of 
A number of industrial establishments both in this country and abroad have 
agreements with their workmen whereby a movement of a minimum number of 
points up or down in the cost of living scale will result in a compensatory adjust- 
ment of wages. When prices began to fall, an interesting situation arose in some 
olants where, prior to the development of a cost of living index, wages were ad- 
Justed with wholesale prices. The price decline in 1920 was so obviously not reflected 
in retail prices that the agreements to reduce wages accordingly could not be carried 
out and other adjustments had to be made. The same thing would have happened 
if food prices only had been used, for, since November, 1920, the retail food price 
increase since 1914 has been well below the increase in the total cost of living. 
Estimates of “real wages” made on this basis have, of course, since that time shown 
an increase far beyond the true occurrence, just as when food costs had gone up 
higher than the cost of living, “real wages’ appeared to be greatly depressed. 
For these reasons, estimates of the economic status of wage earners based on “real 
wages” derived from price data other than a total cost of living are subject 
to considerable qualification. See, for example, Edward Taylor Bullock, “Did 
Labor Prosper During the War?” The Survey, October 15, 1921, pp. 75-76; Paul 
H. Douglas and Frances Lamberson, “The Movement of Real Wages, 1890-1918,” 
American Economic Review, September, 1921, pp. 409 ff.; Paul H. Douglas, “The 
Movement of Wages and the Future of Prices,” Academy of Political Science, 
Proceedings, January, 1925, p. 90. Douglas later completely changed his conclusions 
regarding the movement of real wages and, on the basis of an index which he com- 
uted from prices of other items in addition to food, although still not a complete 
>udget, he reversed his earlier position that wage earners were in no better position 
as regards real earnings in the 1920’s than they had been in the 1890s and decided 
that as a matter of fact the real earnings of employed factory workers in 1924 were 
28%, higher than in the 90's; of transportation workers, 22% higher and of all 
workers studied, 27%, higher. The figures themselves are not so important as 18 
the emphasis given by this complete reversal of a very widely quoted conclusion, on 
the basis of a more complete and sounder basis for computation. See paper pre- 
sented by Paul H. Douglas at the thirty-eighth annual meeting of the American 
Economic Association, New York, December 28, 1925, entitled, “The Movement 
of Real Wages and its Economic Significance.” The American Economic Review, 
March, 1926, Supplement, pp. 17-53. 
¥ See Chapter I.