GROSS EARNINGS IN DISTRICT I 283 
upon differences in interest rates, and these in turn are functions 
of the demand for and the supply of capital funds, of risk, and 
so on. Similarly, district differences in ratios of total expense 
are traceable largely to interest rates—the rates paid on deposits 
and on borrowed money—and to wage and salary scales. As 
between districts, this common element, interest, serves only in 
part to explain the direct relation between the two series. But 
does it explain the positive correlation between gross earnings 
and total expense ratios within a given district, when the entire 
member banking structure or any significant part, as banks in 
tity or volume groups, is considered, and when presumably rates 
of interest, as expense and as revenue, vary less widely from 
bank to bank over the entire district or within any city or volume 
(earning assets) group than they do between districts? Addi- 
tional reasons, it is believed, must be sought to explain the 
phenomenon. These are found in the varying proportions of 
time deposits to total deposits and of investments to earning assets 
—topics discussed at length in Chapter X.1° 
Positive correlation between gross earnings and net earnings 
cannot be explained even in part on the basis of a common ele- 
ment such as that suggested above for that of gross earnings 
and total expense. Other things being equal, if gross earnings 
ratios are high or low, those of net earnings are high or low. 
But other things are not equal, inasmuch as total expense and 
net earnings ratios are negatively correlated, and net earnings 
are the difference between gross earnings and total expense. 
Accordingly, net earnings ratios are dependent upon the rela- 
lions between variable gross and variable total expense ratios, 
each of which series itself is dependent upon variable operating 
conditions. 
Data are not available with which to trace out for individual 
sanks the relations between these antecedent conditions and 
ratios of net earnings. We shall content ourselves here with the 
statement that gross earnings ratios are positively correlated 
and total expense ratios negatively correlated with net earnings 
ratios, and leave to a later chapter the measurement of the 
relative effect on net earnings of these divergent factors. 
10 See pages 190-204, particularly.