THE PROBLEM—ITS METHODS OF STUDY

7

sive evidences of order in our banking system, its several parts,
geographical and group, responding in much the same manner to
change, and its several processes being interrelated in a definite
manner. Moreover, the norms, trends, and correlations seem to
be broadly predictable. In general outline, they repeat themselves
 year after year in all parts of the Federal Reserve system.
They are quantitative evidences of the processes of competitive
effort in a competitive system.
Are these facts significant? They appear to be. Banks individually
 and collectively need standards or norms of operation
which will serve as guides. They need also knowledge of trends
and of the relative levels upon which banking in its various phases
is carried on in different parts of the country. They need to
know that forces are constantly in operation tending to bring rates
of earnings, for instance, to conditions of equilibrium, and that
these forces, interrelated with the fortunes of business, are operating
 in the same direction at a given time in all parts of the country.
They need measures of the nature and of the degree of the interrelation
 of banking processes, representative of groups broad
enough to give them statistical significance. They should welcome
facts which indicate for groups of banks the combinations of conditions
 concerning the composition of their earning assets and of
their deposits, and so forth, tending to be associated with relatively
 high or low gross earnings, expense, and net earnings. It
is facts of this and of other types which the study discloses, partly
as a direct and partly as a by-product of its major purpose—the
search for order in our economic system.
The results of the study may have general economic significance,
 inasmuch as they seem to demonstrate that, for the years
covered and for the institutions involved, the results of widely
dispersed competitive effort can be charted and measured. Moreover,
 they disclose patterns, repeated again and again, associated
with, if not causally related to, such effort. Then, too, the pattern
 of regression to type, in so far as expenses of operation are
concerned, is essentially the same as that found in the field of
retail distribution,? the details concerning which, with respect to
2 See the writer's Competition in the Retail Distribution of Clothing—a Study of
Expense or “Supply” Curves, The Bureau of Business Research, Northwestern
University, Chicago, 1023; Expense Levels in Retailing—A Study of the “Representative
 Firm” and of “Bulk-Line” Costs in the Distribution of Clothing, The