NORMS AND TRENDS IN EXPENSES 05
with the seven-year district levels, (2) with each other from year
to year and over the period 1919 to 1925, and (3) with the aver-
age yearly amounts for the country. Moreover, having expressed
the expenses as ratios of both earning assets and of gross earnings,
and having found norms and tendencies peculiar to them as thus
measured, we may summarize the similarities and differences. If
a summary is to be brief, it must be drawn in general terms and
in broad outline. Exceptions to the rules suggested are accounted
for in what has already been said.
Norms
1. Total expense measured in terms of earning assets and of
gross earnings, relative to the district levels for the period 1919
to 1925, was prevailingly low in the early and generally high in
the later years in the various districts. Both measures agree in
showing the ratios to be low in 1919 and 1920, and to be high in
1924. In the other years, agreement is less complete. In 1921,
expenses were high in terms of earning assets and generally low
in terms of gross earnings; in 1923, the ratios according to both
measurements were low in the East and hich in the other parts of
the country.
2. Total expense, when expressed as percentages of earning
assets and of gross earnings and when compared with the coun-
try’s yearly average level, was low in the eastern, high in the
western, and about average in the central districts. The ratios in
New York and Philadelphia were below and those in Chicago,
St. Louis, Minneapolis, and Kansas City above the country level
during each year from 1919 to 1925.
Trends
1. The trend of total expense, as percentages of earning assets
and of gross earnings for the country as a whole, was irregularly
upward between the beginning and the close of the period. To
this general rule there are no exceptions for the different dis-
tricts for the amounts based on gross earnings, and only one for
those based on earning assets. Both measures indicate increases
to be general among the districts between 1920 and 1921, and be-
tween 1923 and 1924; in the other pairs of years they show con-
flicting trends.
