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

Approved for Release: 2019/07/19 C05210421
are based on deflated retail sales or value of produc-
tion in “constant” prices may impart an upward bias
to the index, because of the dubious reliability of the
official price indexes used as deflators. These two
sources of bias tend to offset each other. Judging from
a mass of anecdotal evidence from the Soviet press
concerning the poor quality of consumer goods and
hidden price inflation, however, one might conclude
that the upward biases are predominant. There is no
good way to assess the matter.
Category Weights
The expenditure weights used to aggregate the 1 l
major categories of goods and services are those for
1970 estimated in the calculations of GNP by end use
in established prices. A full description of the sources
and methods of their derivation is given in CIA, GNP
I970. The weights shown there have been revised
considerably to take into account additional or more
recent information and an improved understanding of
Soviet data; the revisions are described in Appendix D
of JEC, GNP, 1950-80. The revisions reduced total
consumption by 4 percent and raised the share of
goods as opposed to services. A summary description
of the revised weights follows:
\
Goods. The weights for food and soft goods are the
sum of (1) retail sales reduced to reflect only house-
hold purchases for consumption, (2) consumption in
kind estimated in physical units valued at average
prices realized by producers (“farm gate” prices) and
(3) military subsistence estimated by valuing rations
in 1970 prices. The weight for durables and miscella-
neous goods represents retail sales reduced to reflect
only household purchases for consumption.
Household services. The weights represent estimated
household expenditures on these services based on a
wide variety of Soviet sources, mainly studies by
individual researchers published in books and journal
articles.
Communal services. The weights are the sums of
estimated expenditures for education and health (1) by
households based on a variety of Soviet and other
sources and (2) by the government, derived from
official data on wages and employment and on budget
expenditures for other items, net of investment outlays
and stipends. -
On the whole, the expenditure weights are considered
to be quite reliable. Although nominal consumption
may be overstated because of inability to remove some
non-consumption expenditures from retail sales of
goods and services, the total can be reconciled fairly
well with official Soviet values for consumption. The
total, of course, does not take into account black
market prices and other activities carried on in the
illegal “second” economy. Such activities are not
included in GNP in other countries either. In the
Soviet case, moreover, they probably generate mainly
price increases and income transfers rather than
additions to the supply of real goods and services." As
for individual category weights, which are based on
specific Soviet sources and assumptions, their magni-
tudes generally are supported by other information
that is available. Also, the revealed expenditure pat-
tern agrees well with that given by official statistics
based on family budget surveys (Narkhoz I977, pp.
409-410), and, in general, with what one would expect
on the basis of international comparisons for a coun-
try at the USSR’s level of development as measured
by per capita GNP and consumption.“ Data from
surveys of the family budgets of urban Jewish emi-
grees also support the relationshipsfwhen allowance is
made for the unrepresentative composition of the
sample."
The 1970 weights used to aggregate the component
indexes, both for major categories and for their sub-
categories, are probably about as accurate as can be
obtained with the data base and understanding of its
meaning that are available at present. Although a
number of estimates and assumptions had to be made
in the process of constructing these weights, the
impact of any errors cannot be large. If anything, the
values for consumption as a whole and for some
categories" are probably overstated, a tentative conclu-
sion stemming from the fact that successive revisions
in the weights, based on new information invariably
produced lower values. The data underlying the
weights for individual sub-categories are of uneven
quality, a matter that will be considered in the
following section.
" Many personal services are provided privately, both legally and
illegally. The expenditure weights for repair and personal care and
health and education services include a substantial allowance for
such services, based on a variety of information.
“ Data on comparative expenditure patterns are given in Schroeder
and Edwards, 0p. cit.; International Labour Office, Household
Income and Expenditure Statistics, N0. 3, I968-I976, Geneva,
1979; V. Cao-Pinna and S. S. Shatalin, Comparison Patterns in
Eastern and Western Europe. New York, Pergamon Press, 1979.
" Gur Ofer and Aaron Vinokur, Private Sources of Income ofthe
Soviet Urban Household, RAND, R-2359-NA, August I980, pp.
53-69. -
330
Approved for Release: 2019/07/19 C05210421