EXPENSES IN DISTRICTS I AND II 201 
must be given to these questions. The details in Table 165 show 
that if ratios for banks in one group increase from year to year, so 
likewise do those in all groups. To this general rule there are some 
exceptions, but these are insufficient to invalidate the rule. 
If, for the various groups of banks, account is taken both of 
the position of the expense ratios in the first of each pair of 
years, relative to the group level, and of the direction of change 
from year to year, it is found, as shown in Table 166, that group 
ratios which are high in a given year decrease, and those which 
are low in a given year increase in the following year—*“high” 
and “low,” respectively, meaning positions above or below the 
four-year average group ratio for the period 1922-1925. This 
condition, while generally true, most clearly obtains for the 
grouping by size of city of location, despite the fact that, in gen- 
eral, as shown by Table 165, the trend in the group ratios from 
vear to year 1s upward. 
TABLE 166 
CoMPARATIVE POSITIONS AND YEAR-TO-YEAR DIRECTIONS OF 
CHANGE IN GrouP RATIOS OF ToTAL EXPENSE TO 
EARNING Assets, MEMBER BANKS, BosToN 
FEDERAL RESERVE DISTRICT, 1922-1025 
Positions Relative 
to City-Group 
Averages 
1022-102°%¢ 
~~ -*isp of Change [rc:* 
par t » yaar 
Positions Relative 
‘0 Volume-Group 
Averages, 
1022-102¢ 
Direction of Change from 
year to vear 
Te. 
'One group = in position. 
{One group = in direction. 
One group == in hoth direction and position. 
This tendency for ratios of groups of banks in the Boston 
district to regress to type is analogous to that already found for 
member banks by districts for the country as a whole, as shown 
in Table 59, (page 85). But the similarity of tendency, as be- 
tween the two bodies of data, is not restricted to directions of 
change with respect to positions held. Not only do the group 
ratios by directions of change regress to type, but the net per- 
centage amounts of change vary directly with the percentage 
amounts of deviation from type. This fact, most pronounced