THE ECONOMIC IMPACT OF SOVIET MILITARY SPENDING

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP08S01350R000602110001-7
• Release Decision: RIPPUB
• Original Classification: U
• Document Page Count: 37
• Document Creation Date: December 22, 2016
• Document Release Date: September 27, 2012
• Sequence Number: 1
• Publication Date: April 1, 1975
• Content Type: REPORT

Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
FOR OFFICIAL USE ONLY
Intelligence Report
The Economic Impact of Soviet Military Spending
ER IR 75-3
April 1975
Copy No. 332
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
STAT
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Q
FOR OFFICIAL USE ONLY
, The Economic Impact
of Soviet Military Spending
KEY JUDGMENTS
The most important conclusions concerning the influence of economic
considerations on Soviet military spending are:
? The Soviet leaders have not acted ? as though costs have been a major
factor in their military decisions. Defense programs have been well
funded, even during periods of lagging economic growth, and the
followthrough on new programs has been strong.
? Defense costs are unlikely to constrain the Soviets unduly in the future.
Political priorities favor continued emphasis on defense, and expected
economic growth should permit continued increases in defense spending.
Defense spending as a share of gross national product is often used as a measure
of the burden of a nation's defense effort. In fact, however, no single measure
adequately describes the economic impact of the Soviet defense effort.
Furthermore, the notion of burden implies a perception of the political-economic
benefits to be derived from the use of economic resources for defense, compared
with those to be derived from other uses. The weight of the burden therefore
varies according to the outlook of the observer.
? In terms of shares of national economic aggregates, the Soviet defense
effort currently takes less than 8% of GNP, less than 10% of the labor
force, 10%15% of industrial production, and about 20%-30% of
machinery output. All of these proportions have declined over the past
decade even though defense expenditures have grown by about 3% per
year.
? Looked at from a different perspective, defense impacts heavily in the
high technology areas where, it exerts a priority claim on manpower and
output. For example, defense still takes the lion's share of high-grade
scientific and technical talent, and in electronics, defense requirements
account for almost the total output of integrated circuits.
Note: Comments and queries regarding this report are welcomed. They may be
directed to the Office of Economic Research,
STAT
STAT
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
? Finally, the peculiarities of the Soviet economic system introduce
additional uncertainty to the evaluation of the economic impact of
defense programs. In the USSR the character of planning, management,
and incentives poses substantial and persistent obstacles to the efficient
use of resources released from. the military sector.
Soviet perceptions of the burden of defense spending could be shaped by
considerations quite different from those that influence Western perceptions.
? Unfortunately, we have little direct evidence on how Soviet leaders view
the burden of defense.
? On the rare occasions that Politburo members have addressed this
question, they have simply noted that, although defense spending is a
burden and the resources could be better used elsewhere, defense
requirements will be met as long as necessary.
Although the Soviet perceptions of their defense burden are elusive, the
magnitude of Soviet military programs can be assessed against the yardstick of
US programs and US costs:
? The dollar costs of Soviet defense programs measured in 1973 prices
have grown steadily since the mid-1960s.
? They have exceeded US expenditures each year since 1971 and amounted
to about $93 billion in 1974 ? 20% higher than US outlays.
0
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
DISCUSSION
Introduction
I. The nature and dimensions of the burden of Soviet defense programs
on the USSR have been the focus of intense interest in the intelligence community
for almost 30 years. Conceptually, there are two kinds of factors that bear on
the assessment of the Soviet defense burden. The first are "objective" measures
of the economic impact of defense programs. The second are the "subjective"
valuations that Soviet leaders attach to defense programs of various sizes, compared
with their valuations of benefits to be derived from the application of economic
resources to nonmilitary programs. Although this political-economic calculus
probably is not performed explicitly by Soviet decisionmakers, these factors must
play some part in the Soviet decisionmaking process, however imperfectly and
implicitly.
2. This report begins by setting out a number of "objective" measures of
the economic impact of Soviet defense programs. Even these measures are not
easily understood. (There are, for example, frequent allusions to the so-called gross
national product (GNP) paradox: with Soviet GNP about one-half that of the United
States and with the scale of Soviet military programs equal to or greater than
those of the United States, how can the share of GNP devoted to defense be
roughly the same in the two countries?) Then, the report considers the quite
separate question of how Soviet leaders perceive the cost of defense expenditures
in terms of lost alternatives. In part, this is an exercise in politics rather than
economics, with the aim of discovering whether the political impact of defense
spending on Soviet policymaking might be of different weight and direction than
the economic effects.
The Nature of the Burden
3. The commonsense notion of burden is an obligation or expense that
constrains freedom of action. The burden of defense programs on a country has
been looked at in the same way. Military programs use manpower, raw materials,
and production capacity that could otherwise be employed to increase economic
growth and improve living conditions.
4. Typically, the cost of military programs is accepted as the measure of
the value of nonmilitary goods and services that a country forgoes because of these
programs. The resources needed to produce, say, one billion rubles of military
goods and services are assumed to provide one billion rubles of consumer goods
or new investment if transferred to civilian uses.
1
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
5. This assumption will not hold unless
? prices of goods and services in all sectors of the economy are proportional
to the costs of the resources (labor, materials, etc.) used in producing
them,
? the amount of resources transferred to a given sector is not so large
as to reduce sharply the marginal productivity of additional labor and
capital, and
? political or institutional factors do not foster systematic differences in
the productivity of resources in different economic sectors.1
6. As we argue in a following section, the Soviet pricing system blurs the
accounting of gains and losses, while sharp differences in the management of
military and civilian production in the USSR tend to make the transfer relatively
less effective than in Western economies. Even in the United States, the transfer
would cause dislocations and take time, although the economy is far more flexible
than the Soviet economy and US prices are a better approximation of resource
cost than Soviet prices.
7. In considering the economic impact of military programs, some important
spillover effects must also be kept in mind. In both the United States and the
USSR, military outlays have resulted in substantial gains in civilian technology.
The development of computers, microelectronics, and successive models of jet
passenger aircraft has depended heavily on military R&D and production experience.
In addition, the Soviet and US armed forces have trained successive generations
of servicemen in a wide range of occupational skills. This training has been
particularly useful in the USSR, where a large share of the draftees possess only
the most elementary technical skills before callup. The benefits to the civilian
economy of these kinds of technology transfer and manpower training are important
and must be taken into account in assessing the net cost of military programs.
8. Perhaps the most serious obstacle to arriving at an agreed-upon burden
measure is that assessments differ because perceptions vary. The point of view
of the economist is not that of the political leader. More important, perceptions
of burden inevitably depend on national institutions. Certainly the US leaders
responsible for shaping policies must deal with a wider range of political constraints
than Soviet leaders. Even so, Soviet leaders too must weigh many conflicting
interests in assessing the burden imposed by their military programs.
I. For a description of the institutional setting, see Appendix A.
2
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
"Objective" Measures of the Burden
9. Despite the qualifications noted above, there is little choice but to use
military program costs as a measure of the civilian goods and services forgone.
To place these costs in perspective, they are almost always presented in relative
terms ? the share of defense expenditures in GNP, the share of defense procurement
in total industrial production, and the like.
10. The indicators of defense burden in the United States and the USSR
may be compared in this manner. Table 1 shows the shares of GNP, industrial
production, machinery output, the working-age population, and employment that
were required to support the military establishments of the USSR and the United
States in 1973.2 The defense share of Soviet GNP is stated at factor cost because of
problems in pricing that are discussed below. Moreover, the shares are presented
as a range because of the possible downward bias in our estimates due to problems
of coverage and pricing (see paragraphs 22-33).
Table 1
USSR and the United States:
The Defense Share of Selected National Aggregates
1973
Percent
USSR'
United
States
Gross national product
6-82
6
Industrial production
10
6
Production of machinery
/2
10
Population of working age
6
3
Employment in:
Industry
11
6
Machinery industries
20
13
I. CIA estimates. The GNP share is at factor cost; the other shares are based on 1-0
calculations and therefore must use established prices.
2. The lower end of the range represents current estimates of Soviet defense spending
as a share of GNP; the higher end takes into account a possible 20% underestimate of
defense outlays due to the factors discussed in paragraph 32.
2. For a description of some of the alternative measures of burden, see Appendix B. The year 1973 is
chosen for comparison because it is the most recent year for which reasonably firm data are available. Even
in 1973, however, the US figure includes $6 billion of Vietnam-related expenditures, which have no
counterpart in the Soviet figure. Defense expenditures include the pay and allowances of uniformed and civilian
personnel, the operation and maintenance of forces, the procurement of weapons systems, and military research
and development. The spending definition is that of the Department of Commerce, which excludes, for example,
transfer payments such as pensions.
3
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
11. Excluding manpower costs, the defense shares of GNP would be 4% in
the USSR and 3% in the United States, reflecting the fact that Soviet military
pay is low relative to US military pay. The share of GNP devoted to defense in
both countries has declined since the mid-1950s, when it was more than 12% in
the USSR and 9% in the United States.
12. The defense share of industrial output and particularly the output of
the machinery or engineering industries is a key aspect of the economic impact
of military programs because it highlights the potential tradeoff between military
goods and investment goods. Soviet industry in 1973 supplied a larger percentage
of output to defense than did US industry. In the middle and late 1960s, however,
the US share was closer to the Soviet share of 10%-12% because of the large
Vietnam-related US purchases of military hardware.
13. A comparison of Soviet and US uses of machinery output for defense
purposes provides a sharper contrast. In the USSR, the defense share of machine
building and metalworking output has declined from about 30% in 1960 to 20%
today, whereas the US defense establishment normally takes only 10%-12% of the
output of the comparable US engineering industries. The much higher Soviet share
results from the fact that consumer durables are relatively much less important
in Soviet machinery output than in US machinery output. During the Vietnam
involvement, US purchases peaked at around 15% of machinery production.
Although reliable data are hard to come by, we have estimated that in the USSR
about 90% of the output of integrated circuits went to defense in 1970, 75%
of the production of the aircraft and shipbuilding industries, 60% of electronics,
50% of communications equipment, and 35% of numerically controlled machine
tools.
14. Because of the well-founded suspicion of Soviet prices, comparisons of
the manpower preempted by military programs have been frequently used in
comparisons of the burden of military programs in the United States and the USSR.
Uniformed and civilian employees of the Ministry of Defense in the USSR and
the Department of Defense in the United States, employees in the defense industries
of the two nations, and employees in nondefense industries that support defense
production accounted for more than 8% of all employment in the USSR in 1973
and 6% of all US employment. All defense needs preempted 6% of the population
of working age (more than 14 years) in the USSR and 3% in the United States.
15. But manpower comparisons have their own defects. The shares of
defense-related manpower in total manpower will be a measure of relative burden
only if labor productivity is the same in military and civilian production in both
4
- Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
the USSR and the United States. Indeed, the estimates of manpower engaged in
military-related production in the USSR depend on the assumption that labor
productivity in military production equals that in civilian production.
The GNP "Paradox"
16. Before considering the reliability and meaning of the measures of burden
presented in Table 1, an alternative measure must be considered ? and rejected.
Over the years, it has been estimated that
? the dollar value of Soviet GNP has increased from slightly less than one-
half to slightly more than one-half of US GNP, and
? Soviet military expenditures valued in US prices have climbed from
somewhat less than US defense outlays to somewhat more than US
outlays.
Some observers concluded from these estimates that the economic impact of defense
programs must be roughly twice as large in the USSR as in the United States.
17. The fact is, however, that the Politburo spends rubles, not dollars, on
defense. Defense goods and services are relatively cheap in the USSR, compared
with the United States. As a result, the share of defense in GNP in dollar prices
is much higher than in ruble prices3 (see Table 2).
18. The outcome is not surprising. First, defense manpower is paid at a low
rate in the USSR compared with average military manpower costs in the United
States. Outlays for military manpower are 25% of total military expenditures in
rubles; the dollar share would be 45%.
19. Military production is also relatively less expensive than much of civilian
production in the USSR. Throughout its history the USSR has concentrated on
the development of war-supporting industries to the neglect of consumer-related
3. Most of the dollar comparisons of the Soviet defense burden have been wrong on two counts. They
have used dollar prices instead of ruble prices, and they have compared a dollar value of defense with a
different kind of dollar GNP. Most often, defense is valued in dollar prices. The dollar GNP usually employed
is the result of multiplying US GNP in dollars by the geometric average of comparisons of US and Soviet
GNP calculated alternatively in dollar and ruble prices. In 1973, for example, the dollar value of defense
spending ($90 billion) would be compared not with the dollar value of GM' ( $914 billion) but with a lower
value ($697 billion) that uses the geometric mean of size comparisons estimated first in ruble prices
(USSR/US = 0.40) and, second, in dollar prices (USSR/US = 0.70)
5
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Table 2
USSR: Gross National Product
1973
Billion 1955 Rubles
Billion 1973 Dollars
Value
Percent
Value
Percent
Total
353
100
914
100
Consumption
212
60
475
52
Investment
96
27
269
29
Defense
23
7
90
10
Other
22
6
80
9
production. This has resulted in what has often been called a dual economy ?
that is, a relatively efficient heavy industrial sector standing alongside a relatively
inefficient consumer sector. In addition, the most important element of
consumption in the Soviet Union ? food and other farm products ? has been
dogged by the inefficiencies of collectivization, the draining away of the most
productive manpower, and the high costs associated with a poor natural resource
endowment. These factors all lead to differences in relative prices that make Soviet
defense a larger proportion of GNP in dollar prices than in ruble prices.
20. Although ruble costs should be used in dealing with burden, dollar costs
of defense spending are extremely useful in their own right. To compare the scale
of effort represented by military programs of different countries, the programs
must be priced alike. A comparison of the dollar cost of US and Soviet defense
spending in a given year conveys a sense of the size of defense programs in the
two countries in terms of the relative prices prevailing in the United States.4 In
1973 dollar prices, Soviet defense programs have grown steadily since the
mid-1960s, from an estimated $72 billion in 1964 to about $93 billion in 1974.5
Cumulatively, Soviet defense programs measured in dollar cost terms amounted
4. Scale of effort, however, is several steps away from military capability. For example, the number of
weapons available at a given point in time is the result of procurement programs carried out over a period
of years as well as related hardware retirement policies. Capability also depends on the training and morale
of the forces, geography, and a great many other factors.
5. In contrast, the trend in US defense outlays reflects involvement in Southeast Asia. From a level of
about $87 billion in 1964 (1973 dollars), US defense spending dropped in 1965, rose to a high of almost
$109 billion in 1968, and then declined, first gradually and then sharply, to about $84 billion in 1971.
Expenditures continued to decline, although more gradually, to about $79 billion in 1973 and 1974.
6
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08S01350R000602110001-7
to about $900 billion in 1964-74 ? about 90% of the US total. Since 1971 the?
dollar costs of Soviet defense programs have exceeded US defense spending ? by
about 20% in 1974. Logically, the comparison of scale of effort should be carried
out in ruble prices as well. To construct a ruble comparison, US defense
expenditures must be converted to rubles by appropriate average ratios of US and
Soviet prices for the various categories of US defense spending. According to a
very rough calculation, the ruble cost of US defense programs in 1974 was about
23.4 billion rubles.6 Soviet outlays in 1974 are estimated at about 25.5 billion
rubles ? about 10% higher than the level estimated for the United States.
21. The quantitative results ? that the comparison in dollars is more favorable
to the USSR and the comparison in rubles is more favorable to the United States ?
reflect the fact that each country is better equipped to produce its own pattern
of output than that of the other country. Practically all international GNP
comparisons have found that the ratio of prices of goods between two countries
is inversely related to the ratio of quantities produced. In other words, goods that
are produced in large quantities in either country tend to sell at low prices in
that country, and vice versa. Compared with Soviet prices, ? for example,
consumption goods in the United States are relatively cheap in comparison with
goods produced for investment and defense. Within the defense accounts, manpower
is relatively cheap in the USSR compared with its cost in the United States;
sophisticated equipment tends to be relatively more expensive for the Soviet
military establishment. Thus dollar prices place a greater weight on those items
in which the USSR has an advantage while ruble prices give greater weight to
activities in which the United States has an advantage. Since the relative prices
for military goods and services differ in the two countries, no unambiguous
comparison of their defense efforts is possible, except in the unlikely case that
both are buying an exactly proportional mix of goods and services.
Reliability of the "Objective" Measures of Burden
22. The measures of total GNP, civilian industrial production, civilian
employment, and manpower used in the estimates of defense impact in Table 1
6. The ruble comparison has to be rough and dirty mainly because (1) the structure of the US defense
spending accounts does not lend itself to this lcind of calculation (it is almost impossible to isolate individual
hardware items, for example) and (2) the available ruble-dollar conversion ratios have been developed to fit
the Soviet procurement mix (we do not have ratios for categories of equipment that are bought by the
US Department of Defense but not by the USSR Ministry of Defense).
7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
are reasonably firm, probably subject to errors of no more than plus or minus
5%-10%, and such errors would not change the indicators much. But the accuracy
of the measures does depend on whether all of the rubles spent on defense in
the USSR have been counted and whether the ruble prices reflect the full cost
of the resources employed in defense programs.
Coverage Problem
23. Since the USSR reveals nothing about its defense spending aside from
a single "defense budget" total of uncertain coverage and doubtful reliability, CIA
uses a direct costing approach to estimate Soviet military expenditures. Judgments
with respect to the number of weapons and forces are based chiefly on what is
counted or estimated from observations of Soviet military activities. These numbers
are then multiplied by estimates of their unit cost in rubles.
24. The estimates of Soviet defense spending are generally good on the large
visible items such as missiles, aircraft, and ships and on the pay-and-allowance bill
for uniformed personnel. These items account for about 50% of total spending,
excluding R&D. The estimates are weaker for activities that are not readily
observable ? such as R&D, operating, and general support costs.
Pricing of Military Products
25. The evidence on Soviet pricing practices in military production is scarce
and often contradictory. The goal of the 1967 price reform was to set prices so
that they would cover direct costs and a reasonable profit based on the value
of fixed and working capital used in the manufacture of a given item. In the past,
military production ? like many sectors of civilian production ? probably has been
subsidized. There is no evidence that this is the case today. The armed forces
seem to pay the same price that industry does for common-use products such
as electric power, coal, and petroleum products. Sometimes the military reportedly
pays a higher price for a product than a buyer in the civilian sector. Upon further
examination, the higher price usually turns out to be justified by the higher quality
of the goods sent to the military. Military representatives at a defense plant closely
inspect finished products to ensure their quality before accepting delivery, a practice
without parallel in civilian industry.
26. While defense production is not subsidized, profit margins in defense
production may be lower than in civilian production. Several apparently
knowledgeable sources assert that the strong hand of the military consumer forces
8
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
a periodic repricing of newly introduced military hardware. As costs fall, prices
are renegotiated. Prices for civilian machinery, on the other hand, frequently are
kept at the levels established when production first starts ? despite regulations
calling for the setting of permanent prices after an initial break-in period. To the
extent that tighter military control squeezes the profit margins of military industry
to a degree not typical of civilian industry, the comparison of military and civilian
-production is distorted. The bounds of the potential bias perhaps can be estimated
by examining profit rates in the machine building and metalworking sector, where
most of the new product pricing problems arise. In 1972 the average profit rate
in all machinery production was 14%. If profits on military machinery had been
held to 4%, the raising of profit margins on military products to the level prevailing
in civilian machinery manufacture would raise the value of military purchases from
the machinery sector by 15%.
27. An additional source of error in the estimate of Soviet defense spending
results from the sampling procedures used to get around the scarcity of information
on actual ruble prices for military hardware. For a substantial part of military
hardware procurement ? about 40% of total defense spending ? Soviet weapons
are priced in dollars (an estimate of what they would cost the US Department
of Defense) and then converted into rubles by using estimated ruble-dollar ratios.
These ratios are obtained from samples of prices of a few military goods and of
civilian goods with similar characteristics. Errors from this procedure can arise as
a result of
? the estimation of the dollar prices,
? the unrepresentativeness of the sample from which the ruble-dollar ratios
are calculated, and
? differences between the ruble prices used in the sample and the actual
prices charged in the year for which Soviet defense expenditures are being
estimated (ruble-dollar ratio calculations have been based on data from
different years and updated by price indexes for fairly broad categories
of goods).
28. The net effect of these possible errors is difficult to evaluate, but we
believe they probably result in an understatement of Soviet defense spending. This
judgment is based on partial completion of a recalculation of the ruble-dollar data
base. The work completed to date has raised the estimate of purchases of new
equipment and spare parts by about 10%.
9
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Evaluation
29. Reasonable adjustments for the vagaries of Soviet pricing or for possible
errors in estimating Moscow's defense bill do not greatly change most of the
objective measures of defense burden. Our estimates of Soviet defense spending
are more likely to be too low than too high, but errors probably have not caused
an underestimation of total defense spending of more than 20%. Increasing the
current estimate by 20% would raise the share of Soviet defense spending in GNP
by slightly more than 1 percentage point ? or from. less than 6% to about 7%
in established prices.
30. To eliminate some of the distortions in Soviet pricing, CIA has carried
out an adjustment of the GNP to put it on a factor cost basis. We have imputed
the average blue collar wage to conscripts, subtracted indirect taxes and added
subsidies, and imputed an average return of 12% to fixed and working capital in
place of the varying profit rates that characterize the Soviet economy. These rough
adjustments increase the cost of defense somewhat, but not greatly. The defense
share of GNP is raised by less than one-half of 1 percentage point.
31. Independent., estimates of Soviet GNP have resulted in like differences
between the share of defense in GNP in established prices and the share at factor
cost. Studies undertaken for the RAND Corporation by Abram Bergson found that
in 1950 and 1955 the defense share of GNP was 11% at factor cost and 9% in
established prices.7 Abraham Becker, also working for RAND, calculated the
defense share of GNP as 6% in 1960 in both established prices and at factor cost;
whereas, in 1964, defense was 6% of GNP in established prices and 7% at factor
cost.8 (All of these estimates of the defense shares were based on the explicit
defense budget and are lower than the CIA calculations for these years.)
32. Although most of the estimates presented in this report are single valued,
the various "objective' measures of the impact of defense spending clearly should
be thought of as a range rather than a single number. To review some of the
underlying uncertainty:
? Total GNP, industrial production, employment, and the like probably
have a range of error of plus or minus 5% (although the estimates are
more likely to be too low rather than too high because of the difficulty
of measuring private activity completely).
7. Abram Bergson, The Real Income of the Soviet Union Since 1928, Harvard University Press, 1961,
p. 150 (for 1950); "The Comparative National Income of the USSR and the United States," International
Comparisons of Prices and Output, D.J. Daly, edr., Columbia University Press, 1972, p. 148 (for 1955).
& Abraham Becker, "Soviet National Income 1958-64," National Accounts of the USSR in the Seven Year
Plan Period, University of California Press, p. 96 (for 1960 and 1965).
10
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
? Estimates of defense spending in Soviet established prices conceivably
could be understated by as much as 20%.
? If the profit rate on defense hardware production is less than that on
civilian production, a defensible adjustment might raise the factor cost
of Soviet military hardware by up to 15% or so, and total defense at
factor cost by about 9%.
Because of these uncertainties, the ratios in Table 1, which represent some of the
facets of defense burden, are shown as a range reflecting the cumulative effect
of such errors in measurement.
33. Although other factor cost adjustments might be entertained, no set of
defensible changes would greatly alter the picture. _Factor cost prices, however,
are not a wholly adequate guide to the effects of resource transfer in the USSR.
As we argue below, the persistent disequilibrium in the Soviet economy means
that there are as many opportunity costs as there are alternative uses of resources
employed in defense. But, as Becker notes, factor cost prices require only that
prices of factors "be proportional to their relative productivities on the average
for the economy as a whole, not in each and every use."
Assessment of Static Measures
34. The traditional static measures of the economic impact of defense
programs ? even when adjusted for anomalies in the relative prices of military
and civilian goods and services ? do not provide a reliable guide to the effect
of a change in defense spending. First, static measures do not account for differences
among countries (or over time in the same country) in the ease with which resources
can be transferred or assimilated. More particularly, they do not convey the fact
that a transfer of one billion rubles can be accommodated more readily than a
transfer of 10 billion rubles. Second, static measures have nothing to say about
the influence of shifting economic policies and priorities or changes in the degree
of tautness in the economy, although such factors surely are important in
determining the effect of a transfer of resources out of or into the defense sector.
35. Aside from frictional difficulties and the sensitivity to policy changes,
two basic problems interfere with the interpretation of the static measures. in the
USSR:
9. Abraham Becker, Some Methodological Issues in US-USSR Defense Outlay Comparisons, RAND
Corporation, December 1973, p. 24.
11
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
? disequilibrium in resource allocation is the rule in the economy, and
? military production tends to be more efficient than civilian production.
36. The disequilibrium results from the lack of an effective mechanism for
distributing factors of production so as to reduce differences in the value of their
marginal products. Because the marginal products of factors of production (valued
at prevailing ruble prices) vary widely among alternative uses, the value of the
civilian goods and services that could be produced by transferring resources from
military to civilian employment depends on where the resources go. This general
feature of the economic landscape prevents a single-valued measure of the
opportunity cost of defense programs.
37. Static measures of defense impact must also contend with a particular
facet of the disequilibrium in the Soviet economy ? the institutional differences
between military and civilian production. Soviet military-related production and
R&D are carried on under special rules.
? They have priority in obtaining materials and labor.
? The buyer has a powerful voice in determining what will be produced,
in sharp contrast to the buyer-be-damned attitude prevailing in civilian
industry.
? Military R&D has closer ties with production than does civilian R&D.
The tighter project management stems in part from the fact that military
producers face keen competition from product development in the West.
? Party interferences in military production and R&D seems to be minimal
compared with the petty tutelage exercised over civilian ministries and
enterprises.
38. Because of these institutional disadvantages, resources transferred to
civilian pursuits are unlikely to be as productive as they were in military production.
There is no reason why military plants producing, say, tanks cannot be efficiently
converted to producing tractors, if they are given a like priority. Conceivably,
moreover, the preferential treatment granted to some part of military production
or military R&D might be withdrawn and granted to a civilian activity. The transfer
of such advantages would, however, have to be concentrated along a narrow front;
otherwise, the priorities would be ineffective. Large-scale transfers of resources from
12
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
defense, which would involve production of a wide variety of civilian goods, would
be subject to the usual inefficiencies in civilian economic management. 10
Impact of Defense Spending on Economic Growth
39. In assessing the burden of Soviet defense outlays since 1950, the basic
historical trends should be kept in mind. The central fact is that while Soviet
GNP has more than tripled over the past two decades, total defense expenditures
have increased by only 40%. As a result, the share of GNP devoted to defense
fell from 15% in 1951 to 6% in 1974. Defense spending patterns since 1950 reflect
four major trends.
? During 1951-55, outlays increased markedly.
? The demobilization in the late 1950s led to a sharp decline in spending
in 1956-60.
? During the 1960s, Soviet defense spending increased again, although
irregularly. The path of spending was influenced heavily by two waves
of strategic weapons procurement as the USSR sought to close the
strategic gap with the United States.
? After leveling off between 1970 and 1972, Soviet defense outlays rose
rapidly in 1973, as the USSR entered a new cycle of strategic
procurements. Total Soviet defense spending is expected to grow faster
in 1974-76 than at any time since the early 1950s.
40. The ups and downs of defense outlays offer a distinct contrast with the
trends in the rates of growth of GNP, GNP excluding agricultural output and
services, and industrial production (see Table 3 and the chart). In the late 1950s
the decline in defense spending was associated with a slump in rates of growth
of major economic aggregates. On the other hand, the upward trend in defense
spending in 1961-65 occurred against a backdrop of decline in the rates of growth
of GNP, GNP excluding agriculture and services, and industrial production. Since
1965, however, the growth of defense spending has generally moved in concert
with the growth of the economy.
10. According to general belief, defense industry and military R&D have access to high-quality resources
denied to the civilian sector. With the large growth in graduate education and the expanded assortment of
industrial production, this advantage is less than it was a decade ago. In any case, the efficiency of military
production is related to the ability to command resources when they are needed, not to the command over
better resources.
13
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Table 3
USSR: Average Annual Growth in National Output and Defense
(Factor Cost)
Percent
1951-55
1956-60
1961-65
1966-70
1971-72
Preliminary
1973-74
GNP
6.0
6.1
5.1
5.5
3.3
5.4
NANS GNP1
10.9
9.2
6.5
6.7
6.0
6.2
Industrial
production
10.9
8.2
6.5
6.5
5.3
6.4
Defense
4.8
-2.7
2.0
3.9
0.4
4.8
I. Nonagriculture, nonservice GNP.
41. Clearly, economic trends in the USSR do not provide convincing evidence
of the tradeoff between defense expenditures and economic growth. In this
connection, regression analysis or other statistical techniques have not established
strong connections or tradeoffs between the growth of GNP and the growth of
defense spending, between the growth of defense and the growth of investment
or consumption, or even between the growth of production of military hardware
and the growth of output of producer durables. To some extent, the lack of
correlation could be expected:
? year-to-year fluctuations in overall Soviet military spending have been
relatively small;
? whereas we believe that our estimates of Soviet military spending reflect
basic trends quite well, they are less successful in identifying annual
variations;
? the movement of other economic aggregates in the USSR has been
relatively smooth;
? the impact of a change in military spending on the civilian economy
involves lags of varying length; and
? the Soviet practice of maintaining the military production base ? the
factories and equipment, managers, and workers ? during periods of slack
demand blurs the measurement of tradeoff over time.
14
- Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Index 1970=100
120
100
USSR: Growth in National Output
and Defense Spending
o
o
o
0?0o Defense o
00000 000 o ,o0n
80
o
0
o o0
crj00000??o oo(i- e' uu o?
o
oo
o
o o 0
o e
0 oe
o 0,
000?000 eo9 NANS GNP*
60
GNP 00?
oo
oo
40
000?0?
000?
Industry oe
\o"
o
200.
o?
* Non-agricultural, non-service GNP
0L1111111111111111111111__Lj
1950 55 60 65 70 74
569503 3-75
42. In addition, important complementary ties between defense spending and
civilian production must be taken into account. Military-space demands were mainly
responsible for the development of many important sectors of Soviet industry ?
especially in electronics, nuclear power, and the production of higher quality
materials. The unprecedented program of training scientists and engineers might
not have been sustained at such a vigorous pace were it not for the impetus supplied
15
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
by the military competition with the United States.11 Thus, given the emphasis
on quantity rather than quality in civilian production and the tremendous needs
of postwar reconstruction, the average level of Soviet technology might well be
lower than it is today except for the indirect effects of military-space programs
of the past 20 years.
43. Nonetheless, the main reason for the lack of correlation between defense
spending and other economic aggregates is the intervention of other factors tending
to curb Soviet economic growth. For example, the USSR carried out a sharp
reduction in the length of the workweek in 1956-61, the extension of the cultivated
area came to an end with the settling of the New Lands, and the postwar recovery
characteristic of most European countries ran its course. Later on, the predominant
influence on growth was a marked slowdown in the rate of increase of the
productivity of labor and capital.
44. Despite the difficulty of isolating the impact of military expenditures
on the economy, program changes must have been important in certain periods.
For example,
? The military buildup in the early 1950s occurred when returns on new
investment in civilian programs were potentially large. As a result of war
and incomplete recovery, investment could have exploited a backlog of
unexploited technological gains and eased structural bottlenecks.
? The military demobilization of the middle and late 1950s undoubtedly
helped to ease critical shortages of grain and housing. More than
one-fourth of the 4 million men demobilized went to the New Lands
area while others went into construction and into the remote regions
of the USSR. The demobilization also eased the pressure generated by
a low (1% per year) rate of growth of the labor force in 1956-65 ?
the result of depressed wartime birthrates ? and by the 15% cut in the
length of the workweek.
? In the early 1960s, Khrushchev's drive to modernize the armed forces
may have preempted an unusually large share of specialized manpower
and electronics production capacity.
11. Between 1950 and 1972 the number of Soviet scientists and engineers with graduate degrees increased
from 54,000 to 298,000.
16
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Simulation of the Burden
45. Because direct measurement of the opportunity cost of military
expenditures proved to be unreliable, we tried to simulate the tradeoff. The
simulation is based on an aggregate production function for the Soviet economy
and an assumed pattern of civilian uses for resources hypothetically released from
the military sector in 1951-71.12
46. In the simulation, defense spending was held to 6% of GNP in all years ?
about its present level ? and the resources represented by the difference between ?
estimated and hypothetical defense spending were put in economic sectors other
than agriculture and services, mainly into industry and construction.13
47. This redistribution of resources increases the average annual rate of
growth of GNP, excluding agriculture and services, in 1952-71 from 7.9% to 8.4%;
the corresponding growth of GNP increased from 6.1% per year to 6.4% per year.
Although the hypothetical reduction in defense production does not raise the rate
of growth of output much, the accumulated changes in GNP and in major 'end
uses of GNP are substantial:
Billion Rubles
GNP
477
Consumption
418
Investment
139
Defense
-80
48. Because of the assumptions of the simulation (the share of defense in
GNP is held to 6%), most of the cuts in military outlays are concentrated in the
early part of 1951-71. The gap between actual and hypothetical GNP peaks at
about 30 billion rubles in the late 1960s and stays close to that level through
1971.1 4 The hypothetical GNP in 1971 is 419 billion rubles, 5-1/2% larger than the
12. The production function is a constant elasticity of substitution (CES) function, estimated from data
on output, manhours worked, and fixed capital in the non-agricultural non-service economy in 1951-71.
Compared with other functions like a Cobb-Douglas or the frequently used incremental output-capital ratio,
the CES function explains more of the variation in output.
13. For example, if the savings equaled 20% of defense outlays in a given year, 20% of military manpower
would be added to employment in the non-agricultural non-service sectors, and investment in these same
sectors would be increased by the value of 20% of procurement of military machinery and 20% of military
construction.
14. An alternative calculation using a Cobb-Douglas function (with weights based on factor shares) changed
the average annual rate of growth of GNP in 1952-71 from 6.1% to 6.2% The aggregate absolute change
in GNP was 232 billion rubles.
17
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
estimated actual GNP of 397 billion rubles (1970 established prices). If the
simulation results are even approximately correct, the Soviet population has been
forced to shoulder a sizable cumulative defense burden over the past 25 years.
49. In the simulation, the estimated elasticity of substitution of capital for
labor was quite low (0.37). Since the capital-labor ratio has been rising steeply,
returns to new investment have been declining through time ? an observation that
fits well with what Soviet leaders have been saying about their economic problems.
Table 4 shows the increase in GNP (in sectors other than agriculture and services)
that can be attributed to a unit increase in fixed capital, at different points in
time in (a) the simulation model and (b) as reconstructed from Soviet investment
data and CIA estimates of GNP.
Table 4
USSR: Average Output-Capital Ratios'
1951-55
1956-60
1961-65
1966-70
(a) Simulation model
0.66
0.43
0.26
0.28
(b) Incremental
output-capital model
0.70
0,53
0.35
0.35
1. Nonagriculture, nonservice GNP.
50. Thus the nub of the Soviet growth problem is faltering growth in
productivity, due perhaps to a lack of effective persOnal and managerial incentives
and a cumbersome bureaucratic apparatus. To date, however, Brezhnev and Kosygin
have done little more than tinker with the administrative structure while stepping
up efforts to acquire Western technology.
51. Even though the objective measures of burden are in the main imprecise,
certain propositions are clear: defense outlays have been declining in relative
importance in the economy. More important, the USSR ? or at least its leaders ?
has not considered the burden heavy enough to deflect it from supporting a large
and growing military budget and, in the process, accumulating huge inventories
of missiles, ships, aircraft, and land armaments. One possible exception that might
be noted ? the demobilization of the late 1950s ? suited both Khrushchev's
strategic predilections and his desire to do more for the consumer.
52. In sum, the central conclusion to be drawn from economic measures
of burden of military expenditures on the Soviet economy is (1) whatever the
18
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
strain is, it has probably eased over the years and (2) historical and analytical
evidence suggests that the present rate of growth of military outlays can be sustained
or accelerated without taking a larger share of national product. These conclusions,
however, are drawn from Western analyses conducted within the framework of
? Western economics. Soviet perceptions of the impact of defense spending ? the
"subjective" burden ? could be shaped by quite different considerations. And, for
the Western policymaker interested in the pressures on his counterparts in the
USSR, these Soviet perceptions are all important.
Soviet Views of the Defense Burden
53. Although the most important perceptions of the Soviet defense burden
are those of the Soviet government, these perceptions are but dimly seen from
the West. Moreover, the range of perceptions of the defense burden in the USSR
is surely very wide. The population, party and government officialdom, and the
leadership have special interests that affect their outlook on military programs.
Popular Views
54. The Soviet leadership, while increasingly sensitive to living standards, tells
its population almost nothing about defense outlays and consistently employs all
media to justify the need for an ever stronger defense. The Soviet people, as a
consequence, perceive the burden of defense as extremely large.15 For years they
have been told that the need to strengthen the nation's defenses required sacrifices,
and the veil of secrecy cast over defense matters has given the population an
exaggerated idea of the scale of defense activities.
55. Actually, the defense burden in the USSR is larger from the consumer's
point of view than from the planner's. For economic planning purposes, the
government uses prices based on costs of production while the goods consumers
buy are priced to clear the market and often include a large tax. This means that
a million rubles of resources, if shifted from defense uses to consumption, is likely
to bring a good deal more than this on the consumer market.
Perceptions of Midlevel Officials
56. The party and government officials below the very top level apparently
do not know much more about the dimensions of Soviet military programs than
15. According to numerous reports, the average citizen even feels that the relatively small foreign aid programs
to countries like Cuba impact seriously on his consumption.
19
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
the general population does. Although we have little direct evidence of their .views,
they probably believe that the defense burden is fairly large. Indeed, the officials
in responsible positions in enterprises and in government are likely to have
day-to-day knowledge of how defense programs impinge on their interests ? a plant
manager cannot get a special variety of steel because he has been bumped from
the delivery schedule by a defense plant, or a high republic official is directed
to ensure that the military facilities under. his jurisdiction get the fuel and power
they need when shortages occur. Nevertheless, we do not believe that a mechanism
exists by which midlevel officials can affect decisions on military policy and
programs.
Attitudes of the Top Leadership
57. Thus, the perceptions that count in the USSR are those of the top
leadership ? in particular the 23 men in the Politburo. They do not say a great
deal about defense burden, and what they say mostly follows the party line ("We
do not want to spend those large sums on defense, but we will do what is
necessary" ).16 Nikita Khrushchev was a notable exception to this rule. In some
of his public statements while he was in power and later in his memoirs, he described
concretely how the drive to build up Soviet strategic forces in the early 1960s
curtailed consumer-oriented programs in housing and agriculture. Of course, the
appeal to the demands of national s security helped stave off criticism of
Khrushchev's failure to make good on a number of promises ? failures that can
be traced mainly to factors other than military spending.
58. In the absence of convincing testimony on leadership perceptions of the
defense burden, the views of the men who decide defense allocations must be
inferred from their actions. And in general, the leadership has not acted as though
the defense burden had a major influence on their decisions. Defense programs
have been well-funded, and the followthrough on new military programs has been
strong even in the face of economic setbacks like the agricultural and chemical
industry crises of the early 1960s and the poor crop of 1972. To the extent that
the Politburo is worried about a defense burden, the concern probably centers
on the unknown costs of competition in future strategic programs beyond those
registered in today's rough US-Soviet parity.
16. For a summary of the available evidence on leadership perceptions, see Appendix C.
20
- Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
APPENDIX A
SOVIET DEFENSE BURDEN: THE INSTITUTIONAL SETTING
Management of the Economy
Under the Soviet "command economy," the Politburo of the Central
Committee of the CPSU establishes priorities and guidelines for the annual and
five-year plans. These policies are fleshed out in the form of detailed plan documents
by the Council of Ministers through a major staff arm, the State Planning Committee
(Gosplan). Early drafts of a plan are normally discussed with subordinate economic
units and modified according to bureaucratic pressures and later developments. The
Council of Ministers implements the final plan decisions and checks on results.
It has the responsibility for determining the output of all major commodities and
for setting all important prices and wages.
Subordinate to the Council of Ministers are 42 industrial ministries,
19 nonindustrial ministries, and 28 state committees and other agencies with
ministerial status. The ministries at the all-union level, such as the Ministry of
the Defense Industry, are located in Moscow and directly supervise production
facilities throughout the country. Union-republic ministries such as the Ministry
of Agriculture have central headquarters in Moscow and subordinate ministries in
the republics, with the central ministry directing major enterprises and subordinate
ministries administering the remainder. Subordinate to the ministries are the
thousands of individual economic units ? enterprises, farms, and research
organizations ? that are responsible for production, supply, sales, and
transportation in accordance with plan targets.
The Defense Economy
Organization and Management
Of the 42 industrial ministries, 8 have primary roles in the development and
production of weapons and military equipment. Four of these eight ministries
appear to be prime developers of weapons systems; the remaining four are for
the most part subsystem and component producers. Substantial support in the form
of materials, equipment, components, and technology is furnished by other
ministries and the USSR Academy of Sciences.
21
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
The defense-industrial ministries engage in a wide spectrum of activity, from
basic and applied research to series production and to ongoing monitoring and
overhaul of existing weapon systems. In addition to military output, they produce
a substantial quantity and assortment of civilian goods. Over a long period of time
they have integrated vertically in order to control many of the resources ? raw
materials, machinery, and R&D activities ? needed for the development and
production of their military hardware lines. In many cases, the defense-industrial
ministries themselves are the only suppliers of the capital equipment needed in
their own production processes. Military R&D is greatly aided by special prototype
and other facilities not present in the civilian sector.
The Soviet defense establishment, including the defense production sector,
is strongly represented at the highest political levels. The Politburo, assisted by
the Defense Council ? an advisory body on military affairs ? is the locus of
national defense policymaking. It determines the priority to be placed on defense
as a general economic activity and on specific defense programs. The views of
the defense establishment find their most unequivocal expression in the Politburo
in the person of Defense Minister Grechko, a full member since 1973; at the same
time, the other members of the Politburo put defense requirements at the forefront
of priorities, though not with equal vigor.
The de facto supermanager of all defense industrial matters is candidate
member of the Politburo D.F. Ustinov, also believed to be a member of the Defense
Council. He is assisted by the Defense Industry Department of the Central
Committee, which reports directly to him. His effectiveness as a supervisor is
enhanced by his longstanding familiarity with many of the top defense-industrial
managers, who began their careers under him and share his perspectives.
The formal administrative agency of the highest level for military R&D and
production programs is the Military Industrial Commission. A daily working
committee of the Council of Ministers, it contains representation from the Ministry
of Defense and each of the defense- industrial ministries. The Commission combines
into one unique organization wide supervisory, planning, and coordinating powers.
Presumably, it is responsible for directing the defense-industrial ministries,
coordinating defense-related industrial projects within other ministries, and
providing interface between the Ministry of Defense and the, industrial sector of
the economy. In addition, the Commission probably plays a key role in the technical
and economic evaluation of proposed new weapons programs.
22
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
The service branches of the Ministry of Defense are the customers for weapons
developed and produced within the industrial sector. The technical directorates
of each service are charged with the responsibility for originating weapons
requirements and specifications, monitoring research and development, performing
acceptance testing, and enforcing quality control during series production. Military
representative teams, the indispensible assets of the technical directorates, are
present at every facility having substantial military R&D or production contracts.
The chief military representative is a high-ranking and specially trained technical
officer whose responsibilities include the annual development of independent cost
estimates of weapons and equipment produced by that facility.
The weapons development and acquisition process is initiated by the technical
directorates of the individual branches of service. Requirements generated here are
assembled by the General Staff of the Ministry of Defense and then forwarded
to the Military Industrial Commission for review. The Commission coordinates the
requirements with the industrial sector and arrives at some judgment regarding
program feasibility. It is the likely forum for most technical discussions and debates
and probably solicits expertise and advice from a variety of sources.
Operating Characteristics
Defense-related ministries function according to many of the same rules as
do civilian industrial ministries. The principle of financial autonomy applies to all
enterprises, and we believe that the Soviets intend that the costs of weapons in
series production be fully defrayed by the procurement prices paid by the Ministry
of Defense. Nevertheless, military production is different from civilian production.
Military goods and components are manufactured to higher specifications and are
subject to stringent quality control standards. Production schedules are also much
more rigid. Higher prices are charged the Ministry of Defense in order to
accommodate these factors, although cost accounting procedures may not be
adequate, particularly in the case of joint military-civilian production, to fully
reflect quality differences. Another salient characteristic of the defense sector is
the existence of excess capacity, which has often been noted at major defense
facilities. The longstanding campaign to have defense enterprises produce a specified
quota of consumer goods in part is an attempt to employ production and design
teams temporarily idled by defense program phasing.
Soviet defense industries are more technologically advanced than their
counterparts in the civilian sector. This situation is the natural outcome of a history
of preferential investment for heavy industry ? within which the defense sector
23
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
comprises a privileged enclave. In addition, premium wages, salaries, and incentives
have attracted a highly qualified body of scientists, technicians, and production
workers. Since high-level interest assures priority in obtaining scarce equipment
and materials, some of the bureaucratic obstacles that plague the remainder of
the economy pose a less serious problem to the defense industries.
Another important source of high-level performance is that Soviet defense
industry ? unlike civilian industry ? engages regularly in product competition with
the West. The confrontation between the United States and the Soviet Union has
long stimulated the defense industries to produce and apply the advanced
technology necessary for the development of strategic weaponry. This pressure is
greatly enhanced by the need to respond to the views and demands of the Soviet
defense establishment ? a nearly sovereign consumer whose influence extends from
the Politburo to the production line.
As a result of these competitive pressures, the defense industries have played
a key role in the introduction into the Soviet economy of advanced capital
equipment as well as an array of products whose assimulation would otherwise
have been delayed. Specific examples include computers, automation equipment,
and advanced metal-forming machine tools. Many of these goods were developed
in, and continue to be produced only in, facilities subordinate to the
defense-industrial ministries. The effectiveness of the defense-industrial sector in
innovation must be understood as relative to the remainder of the Soviet economy;
compared with the United States, Western Europe, and Japan, the USSR has great
difficulty in (1) translating products from the development stage into successful
series production and (2) motivating industrial managers to improve the quality
and variety of their output and to modernize their production processes.
24
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08S01350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
APPENDIX B
MEASURES OF THE DEFENSE BURDEN
This Appendix briefly treats the methods used to estimate (1) Soviet GNP,
(2) the defense shares of total industrial production and of the output of individual
industries, and (3) the demands of the Soviet Ministry of Defense and the defense
industries for uniformed and civilian personnel.
Estimates of Soviet GNP in Established Prices
The CIA estimates of Soviet gross national product are constructed according
to Western defmitions of GNP using Western accounting procedures. Official
national income and "gross social product" statistics in the USSR are couched
in terms of the Marxian concept of "material product" ? i.e., they exclude the
value of services except for their "material" content ? and gross social product
includes much double-counting of production.
Total Soviet GNP in ruble terms is estimated as the sum of values added
by each producing sector of the economy ? i.e., the wages, profits, depreciation
allowances, and taxes less subsidies. Published Soviet data on retail trade, new fixed
investment, and government outlays and specialized monographs on agricultural
topics and household expenditures also provide the basis for reasonably reliable
estimates of consumption and investment.
Information is inadequate, however, on expenditures for research and
development and inventory changes. Estimates of total GNP arrived at by summing
end use items are consequently less reliable than those detailed by summing the
sectors of origin. The estimates of economic aggregates in this report agree fairly
well with independent calculations made by several Western students of the Soviet
economy.
Estimates of Soviet GNP at Factor Cost
The pattern of resource use in the Soviet economy is not accurately reflected
when GNP is valued in Soviet "established" prices. Commodity prices do not fully
reflect charges for primary factors of production ? land, labor, and capital ?
because of the liberal Soviet use of indirect taxes and subsidies and because of
the exclusion of important production costs in arriving at prices (the state budget,
for example, finances much investment, and Soviet capital charges are quite
arbitrary).
25
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
CIA, therefore, adjusts Soviet GNP to a factor cost basis by (1) substituting
a 12% charge on capital for official profit figures, (2) excluding indirect taxes
(chiefly the turnover tax), (3) including subsidies along with money income and
income-in-kind estimates, and (4) substituting the average blue-collar wage for the
pay of conscripts in military services. Unlike the situation in the United States,
where indirect taxes and subsidies are relatively minor, Soviet GNP at factor cost
is about 10% less than GNP at established prices. The distributions of GNP in
established prices and at factor cost in 1970 are compared in Table B-1.
Table B-1
USSR: Gross National Product
1970
Billion 1970 Rubles
Established Prices
Factor Cost
GNP
381
340
Consumption
220
195
New fixed investment,
capital repair, and
civilian R&D2
110
98
Defense and military
space
23
21
Other end uses and
statistical dis-
? crepancy
28
26
1. These figures are for GNP, by sector of origin, which is used as a control total.
2. Including expenditures on civilian space projects.
Defense Demands on Industrial Output
The defense expenditures were used in conjunction with the Soviet
input-output tables to estimate defense demands on specific industrial sectors
(Table B-2). The use of input-output analysis permits the measurement of the
indirect as well as the direct costs of military programs. For example, in the last
two columns of Table B-2 the share of oil and gas production used to satisfy
military requirements reflects both (1) direct purchases of petroleum and (2) the
oil and gas used by industry and other branches of the economy in producing
goods for the Ministry of Defense.
The Soviets published a truncated version of their 1966 input-output table
in purchasers' prices. The Research Analysis Corporation (RAC) reconstructed the
table, and the US Department of Commerce converted the table to producers'
26
- Declassified in Part - Sanitized Copy Approved for Release 2012/09/27 : CIA-RDPO8S01350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Table B-2
USSR: Military Shares of Sector Output in 1970 Prices1
Percent
Share
of Deliveries
to Final Demand2
Percentage
Share of
Gross Output3
Sector
1966
1970
1966
1970
Total material production
8
7
8
7
Ferrous metallurgy
7
6
15
Nonferrous metallurgymetallurgy
5
8
'213.-
19
Coal
3
2
8
10
Oil and gas
23
19
12
11
Electric power
13
17
10
12
Machine building and metalworking
32
CP
26
22
Chemicals
24
22
13
11
Woodworking
3
4
5
5
Paper and pulp
12
19
7
8
Construction materials
17
16
4
3
Soft goods
3
3
4
4
Processed foods
2
2
2
2
Other industry
2
2
3
3
Construction
1
1
1
1
Agriculture
2
2
2
2
Transportation and communications
2
2
7
6
Trade and distribution
2
2
Other branches
2
2
5
4
I. Total military expenditures, including the estimated material expenditures for military and military-space
R&D but excluding outlays for services (mainly the pay of military forces and civilian employees of the Ministry
of Defense). Material production under the Soviet definition excludes the value of services except for the
material components of such services.
2. Direct deliveries to private and public consumption, investment, defense, and other end uses of national
income (Soviet definition).
3. Share of total sector output that directly or indirectly supports deliveries by the listed sectors to the
military-space establishment.
prices. CIA developed input-output tables for 1966 and 1970 in 1970 prices based
on the RAC and Department of Commerce tables and used its own defense and
GNP estimates to derive the shares shown in Table B-2.
According to estimates in Table B-2, the share of gross output going to defense
uses (7%-7-1/2%) is about the same as the share of GNP going to defense. Considering
both direct and indirect military demands, the impact of defense is greatest on
27
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
the machinery, metals, and chemical industries. The fuels and power, transportation
and communications, and paper and pulp industries also contribute fairly heavily
(5%-10% of total output) to defense programs.
Labor Force (see Table B-3)
Table B-3
USSR: Defense-Related Employment
Sector
1966
1970
Million
Persons
Percent
of Total
Million
Persons
Percent
of Total
Total
10.4
8.7
11.2
9.0
Industry
3.8
13.5
3.9
12.0
Other branches of
material production
1.4
2.5
1.3
2.1
Services (including
science)
1.7
7.6
2.1
7.8
Military forces
3.5
100.0
3.9
100.0
Uniformed
3.3
100.0
3.7
100.0
Civilian
0.2
100.0
0.2
100.0
The intelligence community estimates the number of uniformed and civilian
personnel directly employed by the Soviet Ministry of Defense from order-of-battle
and table-of-organization information. Estimates of the number of civilians
employed by the Ministry of Defense are based on US analog information.
Using the Soviet input-output tables for 1966 and 1970, the share of
defense-related employment in total employment has been higher than the defense
share of GNP. Aside from the military forces themselves, the defense bite is
particularly large in industry.
28
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
APPENDIX C
SOVIET STATEMENTS ON THE DEFENSE BURDEN
The brief and guarded statements by Soviet officials on the burden of defense
indicate that differences exist on this sensitive question among individual leaders
and organizations within the USSR.
The Politburo
Politburo members addressing the question admit routinely that spending for
defense is a burden that requires material sacrifices by the Soviet people. At the
same time, they clearly state that the USSR will continue to spend all that is
necessary for an adequate defense ? a formulation that notably fails to provide
us with any useful information about their force goals. Among the leadership,
Brezhnev, Kosygin, and Gromyko have most often alluded to the desirability of
diverting military expenditures into other areas in order to make improvements
in the economic life of the Soviet citizenry.
At a dinner held at the Soviet Embassy in Washington during the June 1973
US-USSR summit meetings, Brezhnev emphasized that further progress in strategic
arms limitations will allow "our countries to rechannel greater assets for creative
purposes, for improving the life of peoples." Shortly thereafter, Kosygin assessed
the results of these talks in a speech presented to the Austrian-Soviet Society in
Vienna. Further progress in arms control, he claimed, will serve to "alleviate the
burden of military expenses" and will "liberate vast material resources for useful
and creative aims."2
In contrast, Minister of Defense Grechlco has avoided any mention of
eventually redirecting funds from the military to the civilian sector of the economy.
In a 1974 article, Grechko suggested by use of a historical example that "the
sole correct Leninist road" is to continue to strengthen military defenses despite
the burden this places on the national economy.3 He emphasized the need for
economy and efficiency in the military sector and stated that the maintenance
I. Tau, 22 June 1973. ?
2. Toss, 4 July 1973.
3. "Leading Role of the CPSU in Structuring the Army of a Developed Socialist Society," Voprosy Istorll
KPSS, No. 5, May 1974.
29
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
of the "high technical state" of the military requires "great efforts, huge capital
investment" and "considerable labor by the Soviet people." He cautioned that
"miscalculations in this sphere could lead to unjustified expenditures of funds and
resources."
The members of the top Soviet leadership have not clearly articulated a specific
set of conditions that, if attained, would commit them to reductions in defense
expenditures. In a 1968 interview for Life magazine, Kosygin made the point that
exemplary behavior by the United States and a reduction in international tension
would be only one factor that might allow the Soviets to cut defense spending
in favor of consumerism.4 In a letter to the UN General Assembly in 1972, however,
Gromyko listed the renunciation of the use of force and the "banning" of nuclear
weapons "for all time" as two of the prerequisites for a possible reduction in
military expenditures.5
The Military
In line with decisions apparently made at the October 1964 and September
1965 Plenums of the CPSU Central Committee and at the 23d Party Congress
in April 1966, military authors began writing about the application of operations
research methods to the weapons procurement process. An important consideration
according to these writers was the need to apply economic criteria to military
decisionmaking. They observed that the resources available to any state for the
fulfillment of its military policy were limited. Achievement of the maximum
possible results within these limits was the goal.
In the August 1968 edition of the semimonthly journal of the military's Main
Political Administration, Major General M. Cherednichenko, a proponent of the
application of comprehensive planning and systems analysis in defense procurement,
warned that the sharp acceleration in weapons obsolescence and the reduced "life
cycle" of modern weapons systems increase the potential cost of defense.6
Nonetheless, he insisted that the qualitative aspects of a weapon system, rather
than cost, must always remain the determining criteria.
In a September 1971 article in the same journal, Cherednichenko explained
that the defense industry is in the forefront in developing advanced technology
and thus provides an important contribution to general Soviet economic growth.7
4. Life, 2 February 1968.
5. Tars, 16 September 1972.
6. "The Economy and Military-Technical Policy," Kommunist vooryzhennykh sil, No. 15, August 1968.
7. "Contemporary War and the Economy,"Kommunist vooryzhennykh sil, No. 18, September 1971.
30
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
He also declared that "an improved international situation will permit a portion
of the capability of the defense industry to be diverted to nonmilitary production."
The use of this formulation by a military writer is highly unusual. Of equal interest
was Cherednichenko's statement that "our state had to economize in the allocation
of resources for satisfying defense requirements" because of the decision reached
at the 24th Party Congress to improve the people's living standards.
The issue of preferential treatment for the civilian sector of the economy,
to which Cherednickenko alluded, was explicitly dealt with in a number of articles
published in the military press during the same period. A November 1971 article
in Red Star by Gosplan official F. Kotov claimed that the new investment priorities
of the Ninth Five-Year Plan had not altered the traditional preference given
producer's goods in the economy as a whole.8 Through his abstruse arguments,
Kotov seemed to be telling his military readers that the "profound shift" in the
direction of economic planning was little more than rhetoric. Kotov claimed that
"preferential development is being accorded those sectors of heavy industry" which
determine "technical progress."
The May and June 1972 editions of Communist of the Armed Forces ,carried
articles dealing with the question of defense spending. The treatment accorded
this subject, however, differed radically, thus suggesting that the theme of defense
spending is a contentious one. The first article, coauthored by a Colonel M. Gladkov
and candidate of economic sciences B. Ivanov, exhibited concern over the impact
of defense spending and the efficiency with which resources are used.9 To buttress
their arguments the authors cited Lenin.
Lenin stated that under conditions of a threat of imperialist aggression
we must, for the sake of the army, be willing to make "certain severe
sacrifices, of course, strictly defining the extent of these sacrifices"
(Complete Works, Vol. 45 p.112), and that in peacetime the military
organization must not be too burdensome to the national economy.
Economic work in the armed forces also assumes observance of the party's
general requirement concerning economic activities, or, precisely, concern
for the needs of the consumer and for economy in social labor. The
proper handling of this problem means the subordination of departmental
interests to general state interests. [emphasis added.]
8. "Structural Changes in the Country's Economy," Krasnaya zvezda, 30 November 1971.
9. "The Economy and Military-Technical Policy," Kontnnudst vooryzhennykh ill, No. 9, May 1972.
31
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
The second article, by a Lieutenant Colonel D. Volkogonov, which was signed
to press on the eve of President Nixon's visit to Moscow, emphasized the existence
of "favorable opportunities for improving our country's defensive might."1 0
Volkogonov's brief treatment of the relationship of military development to the
national economy was designed to support his call for further strengthening of
the armed forces.
The Economists
In 1964 a highly critical assessment of Soviet economic performance was
reportedly presented to the Central Committee by Abel Aganbegyan, a young Soviet
economist.11 He cited a two-thirds decrease in the rate of growth (presumably
in 1959-64), which he blamed mainly on internal mismanagement and wastage of
resources and, to a lesser extent, on the heavy commitment of resources to defense.
In this connection, Aganbegyan stated, "It is very difficult to compete with the
United States since American expenditure on defence and ours are approximately
equal, while our economic potential is only half as big as that of the Americans."
Aganbegyan's comments come to us secondhand in the form of incomplete
notes purportedly haken at the lecture in which he reviewed his report to the
Central Committee. Aganbegyan presumably did not expect his statements to be
made public. Understandably, few statements by Soviet economists are available
on the ticklish burden issue.
In a 1969 article appearing in the journal of the Ministry of Finance, V.N.
Semenov stated that "exacerbation of the international situation has prevented our
making the full quota of appropriations intended for agricultural investment." 2
He pointed out that appropriations for agriculture "by no means approximate the
amounts stipulated for agricultural development." The tone of these statements
suggests some dissatisfaction with the resource allocation policy at that time.
Semenov is the head of the Agricultural Finance sector of the Ministry's scientific
research institute.
Gosplan
In a 1970 lecture, Nikolai Baryshnikov, Chief of an unidentified Gosplan
office, was particularly gloomy about the social problems in the Soviet Union and
10. "Urgent Questions Concerning Soviet Military Organizational Development in Light of the Decisions of
the 24th CPSU Congress," Kommunist pooryzhennykh ail, No. 11, June 1972.
11. Socialist Commentary, October 1965.
12. "Agriculture - A Stable Economic Base," Finansy SSSR, No. 3, March 1969.
32
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
blamed them in part on the defense burden. Baryshnikov indicated that Soviet
defense expenditures were "not less" than US defense expenditures and that,
because of the lower Soviet national income, Soviet defense outlays created a bigger
burden for the USSR than American defense spending did for the United States.
In September 1971, Gosplan official A. Dorovskilch referred to the existence
of basic disagreements among economists over the new investment priorities of
the Ninth Five-Year Plan.13 The author alluded to the existence of a group of
economists who stressed that preferential treatment accorded light industry, at the
expense of heavy industry, and implicitly to the detriment of defense-related
industries, was dangerous and could create a lag in the production potential of
the country.
Also in 1971, a number of Gosplan officials complained that the escalating
cost of new weapons systems and their tremendous demands for trained manpower
and high-quality equipment presented major problems for Gosplan. The officials
made it clear, however, that the needs of the defense sector were given first priority
and that the defense establishment had no reason to complain about its share of
the national pie during the coming five-year period.
The Think Tanks
Members of the USA Institute in Moscow have alluded on several occasions
to the advantage that will accrue to the domestic economy from cutbacks in the
military expenditures. Noting that the cost of armaments is rising "in truly
geometric progression," Director G.A. Arbatov has pointedly warned that US
military outlays have turned into "an enormous unproductive part of the
economy."14 Following Arbatov's lead, G.A. Trofimenko, a civilian military policy
analyst with the Institute, noted that limitation and reduction of strategic
armaments would enable both the USSR and the United States "to dispose of
bigger resources more freely in solving their own domestic problems." 5
Recent Election Speeches
In the most recent (May-June 1974) round of Supreme Soviet' election
speeches, Kosygin made the only reference to the sensitive burden issue, stating
13. "On the Relationship of the 1st and 2d Subdivisions of Social Production,"Planovoye khozyastvo, No.
9, September 1971.
14. "US Foreign Policy and the Scientific and Technical Revolution," SShA ekonondka, politika I kkologlya,
No. 10, October 1973.
15. "The USSR and the United States: Peaceful Co-existence as the Norm of Mutual Relations," SShA
ekononttica, politika i kkologiya, No. 2, February 1974.
33
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7
that those in the West who suggested increased defense expenditures in a time
of detente condemned mankind to an unending wastage of resources.16 Other
Politburo members, while noting the desirability of detente and the need for
continued progress in arms control, avoided any mention of the defense burden.
Kosygin's statement possibly was an indirect challenge to some members of the
Soviet leadership.
16. Moscow Domestic Service, 1355 GMT, 12 June 1974.
34
- Declassified in Part - Sanitized Copy Approved for Release 2012/09/27: CIA-RDP08501350R000602110001-7