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

Approved for Release: 2019/07/19 C05210421
from gross output. Milk fed by other means is includ-
ed.“ Part of the problem of understanding milk,
statistics remains, however, because a large share—
nearly 40 percent—of milk output is not marketed
and must be estimated from survey data. Moreover,
much of the data on milk marketings is a calculated
milk equivalent of marketed cream. Mistakes in the
conversion factor could result in erroneous figures for
total milk output. -
Production data used in the CIA index for most
commodities are the official Soviet gross output fig-
ures, with some surrogates. In the case of vegetables,
the USSR publishes only a single figure for total
output of vegetables. Production of individual types of
vegetables must be estimated by applying the percent-
age distribution of state vegetable procurements to the
figure for total output. For sugarbeets, tobacco, mak-
horka, and tea, official procurement statistics provide
a reasonably reliable measure of net output. We
assume home consumption of these products to be a
very small share of total output.
Although scattered statistics on the use of agricultural
commodities have become available, the CIA index
must incorporate a number of estimates of seed, feed,
and waste in order to have continuous series. Seed
estimates are based on officially “normed” seeding
rates which may well differ from the true rates.
Statistics on waste of agricultural commodities are
especially scarce.“ Ideally an index measuring agri-
cultural output should incorporate waste estimates for
grain, sunflower seed, potatoes, fruits, vegetables, and
eggs. Although the Soviet press often reports exam-
ples of high loss rates for these commodities, the data
are so incomplete that the index only deducts waste
from production of grain and sunflower seed. Output
of other commodities is therefore overstated. How-
ever, to the extent that the share of gross output
wasted remains roughly the same over time, the
failure to make waste adjustments will not affect
trends.
2’ Selkhoz I971, p. 685; A. M. Bryanskiy, Statistika zhiv0tno—
vodstva, Moscow, 1956, p. I22; V. Starovskiy, Vest stat, no. 4,
I961, p. I05.
“ Waste is broadly defined to include extraneous matter, damaged
output, and losses in handling as well as excess moisture (that is,
moisture above a standard level). Waste therefore can include
usable product. Broken kernels, for example, in grain would be
classified as waste, but would also be usable.
Estimates of agricultural commodities fed to livestock
also suffer from lack of data. The CIA index deducts
from gross output estimates of the grain, potatoes,
vegetables, and whole milk used for feed. The release
to a USDA feed-livestock delegation to the USSR in
I971 of data on quantities of concentrates by type fed
to livestock simplified the task of estimating grain
fed.” Other products fed, however, must be estimated
using commodity balances that attempt to reconcile
output statistics with estimates of utilization. For
potatoes, however, statistics on utilization are often
inconsistent so that a satisfactory independent bal-
ance cannot be constructed. For example, for some
years, the sum of estimated uses of potatoes for seed,
feed, and industrial purposes exceeds total production
as reported in statistical handbooks. (See appendix C.)
B. Accuracy
The accuracy of Soviet agricultural statistics has been
debated by Western scholars for decades.“ Inac-
curacy is believed to result in part from reporting
errors and in part from deliberate falsification. De-
spite the extensive writing on the shortcomings of
Soviet agricultural statistics, the evidence is still
ambiguous concerning the degree of accuracy of
published official series.
" The data were published in USDA, ERS Foreign 355, Livestock-
Feed Balancesfor the USSR, (undated). Grain is the major
component (roughly 85 percent) of concentrated feed fed in the
USSR. Concentrates are low in fiber and high in total digestible
nutrients. Aside from grain, concentrates include milling byprod-
ucts, oilseed meal, and alfalfa meal. Part of the grain is fed as a
component of mixed feed, while the remainder is fed “straight.”
1° A complete review of the debate is beyond the scope of this paper.
The most important sources of the last 40 years begin with Naum
Jasny‘s work in the l940s, and continue with writings by Kahan,
Volin, and Richter in the 1960s. Finally, in I972, a volume dealing
with Soviet economic statistics was published containing three
chapters on agricultural statistics. See Naum Jasny, The Socialized
Agriculture of the USSR, Stanford University Press, Stanford,
California, I949; Naum Jasny, “Interpreting Soviet Statistics,"
Soviet Survey, October-December, I958, p. 9.; Naum Jasny,
“Some Thoughts on Soviet Statistics: An Evaluation,“ Internation-
al Afiairs, no. I, I959, p. 53.; Lazar Volin, “Agricultural Statistics
in Soviet Russia,“ The American Statistician, April-May and
June-July issues, 1953.; Luba O. Richter, “Some Remarks on
Soviet Agricultural Statistics," The American Statistician, June,
I961; Arcadius Kahan, “Soviet Statistics of Agricultural Output,"
Soviet Agricultural and Peasant Affairs (Roy Laird, ed.), Universi-
ty of Kansas Press, Lawrence, Kansas, 1963.; Luba O. Richter,
“Commentary” (on Arcadius Kahan‘s article), p. l65.; V.G. Treml
and J.P. Hardt, eds., Soviet Economic Statistics, Duke University
Press, Durham, N.C., I972. See the chapters by Philip M. Raup,
Eberhard Schinke, and Karl-Eugen Wadekin, all dealing with the
validity of Soviet agricultural statistics.
261
93-892 O - 82 - 18
Approved for Release: 2019/07/19 C05210421