84
RICE UNIVERSITY STUDIES
past for post-1930 data. The six points corresponding to the popu-
lation figures from 1910 through 1960 now lie on a line with very
good precision. It is possible to interpret this Type-III fluctuation
as implying that the effective duration of the Great Depression
was about 5.89 years. In other words, the поп-recoverable popula-
tion loss due to the depression was equivalent to 5.89 years of
exponential population increase at the previously and subsequently
prevailing rate. It should be observed that World War II had as
little effect on the U. S. population growth rate as it had on that of
the Japanese population (cf. Figure 10).
As a check on the notion that there was really a non-recovered
loss of population during the Great Depression, we can investi-
gate the behavior of the Gross National Product (GNP) on the
assumption that there is a close relation between the two time
series. This can be done either by assuming the 5.89-year gap and
testing the departure from linearity using this gap, or by deriving
the gap (if any) from the GNP data directly.
Figure 17 shows the GNP (in current dollars) for the period
1894 to 1965. The values up to 1921 are 5-year averages; the
subsequent values are by individual years. The data prior to
World War I and the data subsequent to World War II are quite
consistent with the general hypothesis of exponential growth. The
intervening period (1914-1945) exhibits rather wild fluctuations—
as might be expected. From this superficial examination alone it
is quite clear that GNP does not possess the stable growth pattern
shown by population growth.
Figure 18 shows the same data with the 5.89-year interval in
the 1930’s removed. Even with the larger variation of the GNP
data, it is clear that this data is not inconsistent with the “depres-
sion-gap” hypothesis. In fact, a good portion of the data for the
1920’s is included within the rough limits of variation sketched
in Figure 18. One could conjecture that the real roots of the Great
Depression are to be found in the “excessive” growth of the GNP
in World War I, and in the difficulty in guiding the GNP back
to its basic growth rate without overshooting the goal.
Derivation of the depression-gap value from GNP alone leads
to some problems. If one uses the 5-year averages for the two
periods of evident linearity before World War I and after World
War II, one obtains a depression gap of 2.05 years. This can be
explained in part by the variation introduced by the Korean War;
if the 5-year period including the Korean War is eliminated, the
estimate of the gap becomes 3.29 years. Better agreement might be