USSR: Measures of Economic Growth and Development, 1950-80

Approved for Release: 2019/07/19 C05210421
fluctuations in Soviet GNP are captured fairly accu-
rately in our data. The continued close relationship of
both measures after 1978 suggests that our estimates
of the growth rate of total GNP are accurate, and
that the apparent underestimate of outlays n.e.c. after
1978 noted above results from an overestimate of
other end-use components.
Comparisons With Other Western Estimates
A number of other estimates of Soviet GNP have
been compiled by Western scholars, although none is
as comprehensive or recent as those presented here.
Most of these estimates concern the 1950s or early
1960s. The tabulation below compares the average
annual rate of growth of total GNP derived in these
other Western studies with the corresponding value
derived in this study: 32
Period Other Western Studies
of Soviet GNP
Author Average Annual
Rate of Growth
(Percent)
1951-55 Bergson 7.6 5.5
1951-61 Moorsteen and 7.4 5.7
_ Powell
1951-65 Kaplan 6.3 5.5
1951-69 Cohn 5.5 5.3
1956-75 Lee 7.7 5.0
1959-64 Becker 5.8 4.8
Average Annual
Rate of Growth
Derived in This
Study (Percent)
It is readily apparent that in each case the growth rate
derived in this study is lower, and in some cases the
difference is considerable. The studies by Bergson,
Becker, and Lee estimate the growth of GNP as the
sum of end-use expenditures. About 60 percent of the
“ Bergson, 1961, p. 149; Moorsteen and Powell, Soviet Capital
Stock, pp. 623-624; Kaplan, Soviet Economic Growth, p. 14; Cohn,
“Growth of the Soviet Economy," p. 17; W. T. Lee, “USSR Gross
National Product," p. 413; and Becker, 1969, p. 128.
difference between the growth rates computed by
Becker and in this study arises from the lower growth
rate of consumption derived here (4.7 versus 3.5
percent per year). This difference in turn results
largely from Becker’s use of Soviet price indexes to
deflate the retail sales data as opposed to the use of
physical quantity consumption and production data in
this study. Similarly, most of the difference between
Bergson’s estimate and the one obtained here is the
much lower growth rate of consumption derived in
this study. Again, Bergson uses Soviet price indexes to
deflate the retail sales component of consumption.
Lee’s data are essentially in current prices and this
undoubtedly accounts for most of the difference be-
tween his results and those obtained here.
The studies by Moorsteen and Powell, Kaplan, and
Cohn all use a set of base-year, sector-of-origin
weights and a set of corresponding production indexes
in order to estimate the growth rate of GNP. The
growth rate obtained here is lower than that of
Moorsteen and Powell primarily because of different
weights. The 1950 implicit weights for industry (a
high-growth sector) in this study are lower and the
weights for agriculture and housing (low-growth sec-
tors) are higher. The Kaplan growth rate is higher
mainly because industry is assigned a higher weight
and the average growth rate of industry is a full
percentage point higher. The Cohn growth rate is
nearly the same as the one obtained here.
26
Approved for Release: 2019/07/19 C05210421