SERIES CORRELATED WITH GROSS EARNINGS 22 5
and to year-to-year changes. In respect to district averages, the
ratios must of necessity be sometimes above and sometimes below
these levels; in respect to year-to-year changes, both the direc-
tions and percentage amounts vary from year to year. In spite
of the fact that the distribution of deviations of gross earnings
ratios from the country level tend to remain fixed for the differ-
ent districts, it is still significant to correlate for paired series
the types and amounts of percentage deviation. While it is ap-
parent that gross earnings in relation to earning assets vary geo-
graphically, the ratios in the newer sections of the country gener-
ally being higher than those in the older parts, it is not self-evi-
dent that, in general, the districts having high and low ratios of
gross earnings have high and low ratios, respectively, of total
expense and of loans and discounts. And yet such are the facts
for member banks, by districts, for the years 1919-1925, as shown
in Table 135. Moreover, similar relations were found when the
bases of comparison were percentage deviations from district
levels and year-to-year changes.
On the other hand, the relations between variable gross earn-
ings and net earnings ratios are not so clear. On the average,
net earnings tend to be low when gross earnings are high, and
high when they are low, but these relations do not hold for all
of the dispersion-groups. Neither are the relations unmistakable
between variable gross earnings and variable demand deposits
expressed, first, as a proportionate part of total deposits, and sec-
ond, in terms of earning assets. But let Table 1 35 speak for itself,
the reader remembering the manner in which it is constructed and
the bearings which district locations have on interest rates, and
hence on gross earnings.
It is of interest to compare the distribution of the yearly paired
ratios of gross earnings with loans and discounts and of gross
earnings with total expense—all ratios being expressed in terms
of earning assets—for (1) member banks in the Boston district
for the different years 1919 to 1923, (2) those in all districts for
1925 alone, and (3) those for all districts for the years 1919-
1925. This is done in Table 136. Gross earnings and total ex-
pense, and gross earnings and loans and discounts, are positively
correlated in all districts for the year 192 5, as well as for the
years 1919-192§.
Comparisons in this chapter are made between district ratios
