COST OF LIVING INDEXES COMPARED 107 
dard of living a larger proportion of the budget would neces- 
sarily be spent for those items the prices of which had gone 
up the most, and a relatively smaller proportion for those of 
which prices had advanced more conservatively. 
It 1s quite likely, however, that the pre-war standard was 
not retained, and that in 1918 and 1919, in the great bulk of 
communities where the survey of the cost of living was made 
by the Bureau of Labor Statistics, more luxury goods, of a 
more expensive type, were purchased than was the case 
under conditions prevailing in preceding years; in other 
words, that American workmen whose incomes were greatly 
increased during these years spent some of their earnings 
for better clothing, more expensive furnishings and other 
sundries.) These items of expenditure are most elastic of 
all in their adjustability to income. It is only natural to 
suppose, therefore, especially as very little housing was built 
to meet demands for better living quarters,® that American 
workmen expressed their improved economic status in better 
clothing, household furnishings and equipment. An investi- 
gation of incomes and expenditures, most of which covered 
the year 1918, would inevitably have reflected this in both 
the quality and the quantity of goods consumed. 
From the point of view of total expenditure, this meant 
that rents were relatively a less important item and that 
clothing and sundries were relatively more important than 
1This does not agree with a conclusion reached on the basis of union rates of 
wages. (Carr, Journal of the American Statistical Association, December, 1024, 
op. cit., p. 506.) The union scale, however, applies only to a limited proportion of all 
workers and rates leave out of account losses from unemployment and payment for 
overtime. In December, 1918, the Bureau of Labor Statistics’ cost of living index 
number for the country as a whole was 171.5 based on July, 1914 as 100, and at the 
same time New York State factory earnings were 183, based on June, 1914 as 100. 
On this basis, instead of being 25% worse off, as contended, in 1918, workers were 
8% better off; in some of the exclusively war industry centers, their position was 
probably even more favorable. Observations of conditions at that time indicated 
that such was the case. A new computation by Paul H. Douglas of real earnings of 
employed American workers shows a 3%, improvement of conditions in 1918 over 
1914; "among employees in manufacturing, 8%; in transportation, 12%; among 
certain groups, as iron and steel or textiles, 19% and 13%, respectively. Douglas. 
The American Economic Review, March, 1926, Supplement, op. cit., p. 53. 
? It is a well-demonstrated fact that as total expenditures increase, the proportions 
spent for clothing and for sundries increase. 
® This was one of the complaints during the war period, where workingmen’s 
families with greatly increased incomes wanted better living quarters and could 
not find them.