NET EARNINGS IN DISTRICT I k3 4 
centrate generally become smaller as the city and the volume 
groups increase in size. Similar conditions obtain in each of 
the years. It is apparent, therefore, that to present the average 
ratios by years and by groups of banks, as is done in Table 185, 
is not to disclose the variations in the net earnings of these insti- 
tutions. It is because of this fact that the details in Table 186 
are presented. 
In Table 183, it is seen that the average ratios for the banks, as 
grouped, increased between 1924 and 1925. Was this tendency 
general for the banks individually, and is it in any way related 
to the size of the ratios in 1924? The fact of variation is evident 
from the details in Table 186. Some are high, some “average,” 
and some low. Did those falling within these several categories 
increase between 1924 and 1925? To answer this question, stand- 
ards of reference from which to measure the amounts of differ- 
ence from type, and measures of change must be selected. Let 
us use those already employed in the analyses of gross earnings 
and of total expense ratios, follow through the processes of 
analysis, and observe the results. These may then be used as 
bases for certain generalizations. 
A convenient and logical standard is the average ratio for 
the 408 banks® in each of the years 1924 and 1925. With such 
amounts as points of departure for measuring dispersion, the 
process of determining the types and amounts of change between 
£924 and 1925, used above in the analysis of gross earnings ratios, 
may be employed.* The results of carrying it out are shown in 
the first section of Table 187. 
But the average ratio for all of the banks in a given year does 
not agree with the yearly averages for the banks in the respective 
city and volume groups. Indeed, the averages for the banks in 
four of the five city groups in 1924, and in three of the five in 
1925, are less than the corresponding yearly averages for all of 
the banks. Accordingly, since banks by city groups have different 
yearly average ratios, these may be taken as standards of refer- 
ence with which to compare, group by group, the ratios of the 
individual banks. If this is done, and if the yearly ratios are 
treated in a manner similar to that used immediately above, the 
8 For 1924, two of the 410 banks had total expenses exceeding gross earnings. 
t See page 273.