346 
BANKING STANDARDS 
rather than random behavior, a consistency extending alike to 
all of the twelve districts. 
3. The tendency for ratios in the various districts and for the 
different groups of banks in given districts to regress to type. 
This phenomenon shows itself in two ways. First, ratios which 
are high or low in a given year tend, respectively, to decrease or 
to increase in the following year. As respects the type of change, 
increase or decrease, it is the ratios which are high—that is, above 
their average levels—which tend to decrease, and it is those 
which are low—that is, below their average levels—which tend 
to increase. Second, the net percentage changes in the ratios from 
year to year vary directly with the percentage amounts by which 
they deviate from their own average levels. The rule is for 
direction of change from year to year to vary inversely and for 
percentage amounts of change to vary directly with the positions 
which district and group ratios occupy, in the first of each pair 
of years, relative to their own average levels. 
4. The tendency for ratios in the various districts and for 
the different groups of banks in given districts to be consistently 
above or below the country and the district averages, respectively. 
That is, in the case of the different districts, geographical differ- 
ences in the ratios in the various series of data tend to obtain, 
while for groups of banks within a given district, differences asso- 
ciated with bank size and location persist from year to year. 
In general, therefore, and in summary of the tendencies for 
the individual series analyzed, it is concluded as follows: Dis- 
trict ratios relative to their own levels tend to be high or low at 
the same time; relative to the levels for the country as a whole, 
they are consistently high or low during the entire period studied; 
and they change in the same direction from year to year, those 
above their levels in a given year tending to decrease, and those 
below these levels tending to increase in the following year, the 
percentage amounts of decrease or of increase varying directly 
with the percentage amounts of displacement from their own 
average levels. 
The foregoing summary has to do with series of banking ratios 
treated individually. But such series are not independent of each 
other in respect to bank operation, nor are they unrelated in so 
far as deviation from type—district or country—, change from