SERIES CORRELATED WITH EARNING ASSETS 169

That is, in so far as positions relative to the district averages
and directions of change from year to year are concerned, perfect
 positive correlation occurs between loans and discounts and
six other series; for these measures and for deviations from the
country’s averages, it obtains between loans and discounts and
gross earnings and total expense.
Graphic illustrations of the relations, district by district and
for the country as a whole, between ratios of loans and discounts
to earning assets, and (1) ratios of demand deposits to total
deposits, (2) gross earnings to earning assets, (3) total expense
to earning assets, and (4) net earnings to earning assets, are
shown by Charts 38, 39, 40, and 41, respectively. These charts
are drawn on a ratio basis and are to be interpreted in keeping
with the details relating to such charts given in Chapter VIII.
They are based upon the data used in constructing Tables 97, 98,
and roo. Not only is there marked correlation between loans
and discounts and the series illustrated, as summarized in tabular
 form for all districts and years, but such correlation obtains
for each of the districts.

2. SERIES CORRELATED WITH RATIOS OF INVESTMENTS TO
EARNING ASSETS

The two parts of earning assets, as used in this study, are
loans and discounts, and investments. Accordingly, when the
loans and discounts are expressed as percentages of the earning
assets, the complements of the amounts are the ratios of investments
 to earning assets. Moreover, when the ratio of loans and
discounts to earning assets for a given year for a given district deviates
 plus (4) or minus (=) from the district or country level, or
increases or decreases from year to year, the corresponding ratio
of investments to earning assets deviates minus (=) or plus (4),
or decreases or increases. Accordingly, it is unnecessary to discuss
 the relations of the different statistical series to the ratios
of investments to earning assets, the signs of deviations and
of directions of change in the two cases invariably being opposite
to those for loans and discounts. The percentage amounts of
change in the two sets of ratios are, however, different, but, inasmuch
 as, in general, the relations of each of the series to investments
 are opposite to those to loans and discounts—the paired