CIA HISTORICAL REVIEW PROGRAM RELEASE IN FULL
stment in the Sovietomparative Perspective
Scholar in Residence Office of Economic Research Central Intelligence Agency July 8, 1
TABLE OF CONTENTS
Summary 1
I. Introduction . . .
IX. Price Ratios Used in the
A. Bias in the Computer
III. Comparison of Soviet and US Investment6
IV. Comparative Trends in Soviet and US. 15
V. The Soviet Investment Effortultilateral
. . 20
VI. Investment Trendsultilateral Perspective 28
VII. Effectiveness of Investment 30
VIII. Price Relationships for Construction and Producer . . 33
APPENDIXES
of Ruble-Dollar Ratios for Construction
of Ruble-Dollar Ratios for Producer Durables
of Values for Soviet and US Investment in
of Multilateral Comparisons of Per Capita
TABLES
Ratios for Categories of Fixed Investment in
and USSR: Comparison of Fixed Investment Expenditures
in9
and USSR: Comparison of Fixed Investment Expenditures
6 11
i i
and USSR: Comparison of Construction Components of
Fixed Investment Expenditures6 12
and USSR: Comparison of Producer Durables Components
of New Fixed Investment6 14
US and USSR: Indexes of Fixed Investment Expenditures . . 16
US and USSR: Comparison of Fixed Investment5 . . 17
US and USSR: Comparison of Per Capita Fixed Investment in
19
Capita Levels of 21
Multilateral Comparison of Fixed Investment Ratios
Multilateral Comparisons of Per Capita Fixed Investment
and Its Principal Components in US Dollars6 . . 27
Comparisons of Trends in Per Capita New Fixed
Investment, 29
Comparative Return on Non-Residential Investment . 32
Composition of Non-Residential Capital. 34
Multilateral Comparisons of PPP Ratios for Construction
and Producer Durables 36
APPENDIX TABLES
A-l. Revised Ruble-Dollar Rates by Type of
A-2. Derivation of Soviet-Weighted Construction Ruble-Dollar
Ratio, 0 43
A-3. Derivation of the US-Weighted Construction Ruble-Dollar
Ratio, 0 45
A-4. Derivation of6 Construction Weights . 49
A-5. Derivation of Soviet Weighted-Construction Ruble-Dollar
Ratio,6 50
A-6. Derivation of Soviet-Weighted Construction Ruble-Dollar
Ratio, 6 51
iii
Investment the Sovietomparative Perspective
Summary
The discussion of Soviet consumption had emphasized its low priority in resource allocation. By contrast, capital investment hasigh priority since central planning was firstalf century ago. While Soviet per-capita consumption levels have barelyhird of US levels in recent years, Soviet per-capita investment levels have approached their US counterparts and have oxceeded them in the aggregate. Soviet performance has been particularly favorable in the producer durables component of fixed investment.
In dynamic terms, too, the strenuous Soviet investment priority has been conspicuous. 5 Soviet fixed investment has increased approximately three times as rapidly as the US rate, while at the same time prices of investment goods and services have risen much more gradually for the USSR.
If Soviet investment performance is evaluated in the perspectivearger grouping of the principal industrial market economies and socialist Hungary, additional interesting feutures emerge. Both the USSR and Hungaryigher proportion of their national product to investment than their relative levels of per capita CNP would predict in terms of economic theory. If residential investment if excluded, their investment proportions appear even more abnormal.
While Soviet par capita investment levels have attained parity with those of the United States, they remain considerably below those for France, Germany, and Japan. They are glaringly low for housing investment and rank high only for producer durables. As for investment dynamics, Soviet investment growth rates are higher for all types of capital Investment other than housing.
However, the more rapid Soviet investment growth rates and the high investment proportions mix have not yielded commensurately higher returns in terms of CNP growth. The Increments to CNP obtained by cumulative non-residential investment in bothor the USSR have been significantly below similar returns for France, Germany, Italy, and Japan, although higher than the yields for the United Kingdom and the' United States. Furthermore, the rate of return has deteriorated markedly The possible explanations for this unfavorable position and trend include (a) the heavy construction content of Soviet investment, (b) the relative neglect of replacement investment with its more rapid payoff, and (c) the structural composition of investment with its unusually large agricultural emphasis, ector beset by marginal weather conditions and chronic organizational problems.
The high investment priority with its connotation of relative production efficiency is also reflected in the relatively high purchasing power parity (PPP) of the ruble for this typo of
expenditure. 6 the number of rubles per dollar required to purchase investment goods and services4 for the Soviet mix4 for the US mix. By contrast, the equivalent PPPa for consumption3 Within the investment grouping the PPP of the ruble wan especially favorable forespectively; for the inefficient construction component the respective ratios0 The inefficiency of the Soviet construction sector is illustrated in the multilateral comparison by the lower PPP for construction than for producer durables. By contrast, for the five market economies and Hungary, construction has the higher PPP.
I. Introduction
The current study represents the second comprehensive estimation of comparative investment expenditures in the USSR and the US. Previous studies relied upon comprehensive comparisons5 rubles and dollars. The present comparisons are based upon updated ruble-dollar ratios. These ratios, in turn, reflect the last comprehensive Soviot price reformith modifications for producer durables The US prices are those Both Soviet and US base year prices have been updated6 by appropriate price indexes.
The5 comparison was undertaken by the Office of
1. CIA/RRorBp_.irlfton of Capital Investment in the US and the USSR, . February,
Economic Research of the Central Intelligence Agency,he producer durables portion of which was based upon research of Abraham Becker.s explained later in this study, the5 comparisons was later revised by Rush Greenslade in an unpublished manuscript. Subsequently, another setcomparisons with similar results was prepared by Professor Abram Bergsonajor input into his estimates of Soviet and US levels of gross national product.ergson based his estimates on Becker's studyIA study of construction ruble-dollar ratios. 4/
II- Price Ratios Used in the Comparisons Although the objective of this analysis is an evaluation of the Soviet investment effortomparative perspective, it is first desirable to discuss tho ruble-dollar price ratios which underlie the comparative measures of investment expenditures.
The general methodology for derivation of the ratios has been described in detail in the foregoing discussion of Soviet consumption and need not be duplicated here. Therefore, the discussion will be limited to specifics of derivation of the investment ratios.
Rand Corporation, Prices of Producer Durables in tho United States and the USSR5
Abram Bergson, "The Comparative National Incomes of the Soviet Union and the US" in D. J. Dalynternational Comparisons of Prices and Output, National Bureau of Economic Research,
CIA/RR, 5 Ruble-Dollar Ratios for Construction,
Ratios have been derived separately for the two basic components of capital investment--construction and producer durables. The detailed derivation of the construction ratios is described innd the derivation of the producer durables ratios in Appendix B. The basic construction ratio study was prepared by Ray Converae and the producer durables study by John Keilty and Earl Rubenking of the Office of Economic Research, Central Intelligence Agency, For the interested reader the details of the derivation of the respective ratios are to be found in the basic studies referenced in the appendixes. Refinements of the basic findings are also described in tho appendixes.
6 the number of rubles per dollar required to purchase investment goods and services4 for Soviet-weighted4 for US-weighted ratios, while the equivalent required for purchase of consumer goods and services3espectively. Within the investment grouping the ruble-dollar ratios were especially favorable forespectively. For the leas efficient construction sector tho respective ratios0
The ruble-dollar ratios6 for the full spectrum of investment expenditures are preaentod in table 1.
Although the same basic methodology has been used inS6 comparisons, the information base was considerably larger and more representative for the later study, as noted in
detail in thend B. Therefore, any detailed comparisons of56 findings would have to be qualified for the foregoing reason to such an extent that the conclusions would have dubious validity. Nevertheless, in subsequent discussion the earlier ratio will be used to derive implicit price deflators for major investment categories.
A. Bias in the Computed Ratios Any aggregate economic comparison between countries with considerably differing degree of economic developmentuilt-in bias. While the lesser developed economy produces goods and services utilizing technologies which have been suparncded by more advanced processes in the more developed economy, it is still possible to synthesize prices for these good3 and services in the more developed country. When the more developed nation produces products beyond tho technological capability of the leaser developed country, establishing price ratios is much more difficult. There is simply no reliable way to synthesize prices for goods in the less developed country that require technology beyond its current capability. Thus, the sample of goods and services is biased in favor of the less developed economy. The more advanced economy's production capability is understated, and thereommensurate understatement of the PPP of its currency relative to that of the less developed country.
This general qualification applies to the Soviet-US binary comparison and, by indirection, to Soviet comparisons with the other market economies in tho multilateral measures. In calculating producer durables ratios Keilty and Rubenklng estimated that lower Soviet performance standardsownward biasercent to the ratios.hey also selectively illustrate inferior durability of Soviet durables, but lack the data to suitably adjust the ratios. In the construction ratio calculations, the highest quality Soviet projects were compared against the lowest quality US examples, thus excluding higher quality US construction from the sample. hird element of downward bias is contributed by over-statement of the PPP of the ruble in capital repair expenditures (seo
II. Comparison of Soviet and US Investment6 While the United States has much higher consumption levels than the USSR, especially in per capita terms, the Soviet investment offort exceeded that of the United States in the aggregate6 and approached parity in per capita terms.esult is not unexpocted, given the high proportion of Soviet CNP that has been devoted to investment since central planning wasalf century ago
5. National Foreign Assessment Center, USSR and US Price Ratios for Rubles - 2 Dollars, Volume I, .
B
The dollar comparison gives tho Sovietecided advantage, while the lower limit ruble comparisonomewhat lesser margin in favor of the United States. The compromise geometric mean comparison shows parity for new investment,ecided Soviet advantage if capital repairs are included. The heavy reliance on capital repairs in the Soviet investment scheme contributes to the low productivity of Soviet investment, because this component of investment necessarily prolongs the lives of obsolescent assets.
If the investment comparison is recalculated in per capita terms, the Soviet margin is significantly eroded. For new investment the Soviet advantage disappears, irrespective of which country's prices are used. Even the inclusion of capital repairs yieldsoviet margin only if dollar measurements are used
Of the two principal investment components, the Soviet margin is larger for producer durables than for construction. Only in the durables category islear Soviet margin in total expenditures. Since constructionargor proportion of investment expenditure in the USSR, the ruble-dollar relationship must be less favorable for this activity. The difference reflects the relative inefficiency of the Soviet construction sector, discussed below in connection with the multilateral investment comparisons.
|
if liii.i
Although the comparisons inllustrate the high priority accorded to investment in theiner perception of the differences between Soviet and US economic priorities may be obtained if both construction and producer durables investment are decomposed into their principal components.
The contrast of construction priorities between the two economies is striking While the Soviet effort is directed toward increasing industrial and agricultural production
capacity, the US emphasis is upon consumer-oriented activities like housing, personal services, and commerce. Even the heavier US transportation investment commitment is largely explained by highway construction with its heavy consumer orientation. The low Soviet priority for housing is especially conspicuous. In these comparisons the differences between the dollar and ruble measures are minimal.
If the capital repair of buildings and structures is included among Soviet construction components, total construction would rise by aboutercent in both dollars and rubles. No conceivable distribution of this small increment among construction categories would alter the conclusion about sharply contrasting construction priorities.
A similar impression of sharply contrasting priorities is conveyed by comparison of the composition of capital Investment in producer durables in Che two economies
Soviet producer durables investment is focused on expansion of heavy industrial and agricultural production and on construction. By contrast, the US emphasis is on expansion of consumer goods production and services, and automation (purchase of instruments) of existing industrial and services technologies. The higher degree of sub-contracting in US industry and the widespread personal ownership of vehicles and electronic products probably explains the heavier emphasis on transportation and
communication equipment in the United States.
As in the comparison of construction patterns, the curroncy chosen as the unit of measurement docs not affect the impressionarked contrast in priorities, nor would tha inclusion of capital repairs of producer durables.
IV. Comparative Trends in Soviet and US Investment There are two alternative approaches to estimating relative trends in Soviet and US investment expenditures. One approach compares indexes of investment in constant priceshe other compares levels of expenditures for the two countries in different benchmark years, as calculated with the help of ruble-dollar ratios derived for those years
Ovor the, Soviet investment expenditures increased three times as rapidly as those of the United States. The Soviet margin was greater for producer durables and somewhat less for construction. For total investment the compound growth
l
H
a
i
i
hi!
rate for the Soviet Unionercent and for tho Unitedercent. The respective growth races for constructionercent and for producer durables,Sercent.
If tho aggregate geometric mean investment comparisons6 inere moved backward5 by the Indexes in tablehe ratio of Soviet to US expenditures would be reduced tohird and per capita expenditures touarter. This simplified procedure assumes, however, that relative prices have remained constant in both economies, while in actuality rapid technological progress and changes in demand have led to significant price changes.
Therefore, the second approach, which takes relative price changes into account, is more appropriate. 5 comparison based On ruble-dollar ratios computed5 the ratio of Soviet to US fixed investment is betweenndercent in the aggregate andhird and two-fifths in per capita terms depending on whether the comparison are in rubles or in dollars
If the findings in6 ruble-dollar ratio calculationsndre compared with those in5 ruble-dollar ratio calculationsver the more than two decades5 Soviet investment has risen dramatically in comparison with that of the United States. From an aggregate proportion of two-fifths of the US effort (geometric mean
6 it was higher. In par capita terras it has risen to near parityroportionit morehird as large.
V. The Soviet Investment Effortultilateral Perspective Although the bilateral comparison of Soviet and US investment patterns provides perspective on Soviet policies, an evon more informative impression may be obtained if the comparison is expanded to include the other najor market economies, as-well as Hungary,maller socialist economy. rincipal reason for introducing this broader comparison is the marked differences between the investment efforts of the United States and some other market economies.
Before comparing the6 investment structures of the eight countries, it would be worthwhile to compare the countries in terms of relative per capita GNP levels
If international prices are used as the standard,he economies fall into three groups: France and Germany have aroundercent of US per capita GNP, Japannd the United Kingdom, and Italy. Hungary, and the USSR slightly below half.
This comparison of relative Income levels should be juxtaposed against the proportions of GNP devoted to capital Investment and Its principal components to gain impressions of
6. Seeiscussion of the "International prices" methodology.
TABLE 9
PER CAPITAL LEVELS OF6 a, (International prices)
Prices
United
United
a. See contribution by Margaret Hughes for derivation of estimates.
eachelative Investment efforts.
Basic economic theory, whether neo-classical or Keynesian, states that investment and savings are rising functions of per capita income. The precept assumes that consumer sovereignty prevails and that state decision makingomparatively small share of available resources. According to this principle tho investment ratio should be lower for the less developed economies in the sample.
Both the USSR and Hungaryigher proportion of CNP than the theory would predict. Of course, the ability of the state to directly control the rate of investment and savings explains this apparent aberration. However, the theory cannot explain why the US investment ratio is below that of France, Germany, and Japan, let alone that of Italy. Whatever the shortcomings of the theory, one may conclude that Soviet and Hungarian investment efforts are large.
What characterizes the Soviet investment effort is not only its aggregate ratio, but its composition. Residential constructionuch lower share of GNP and non-residential construction an unusually high share. (The differences in the attention paid to housing in the Soviet Union and Hungary compared with other countries is striking and can only be explained by further research on Soviet and East European investment policies.) If only non-residential performance is analyzed, the "abnormal" Soviet and Hungarian efforts are even more conspicuous.
the results of the Soviet investment effort are to be
compared with those of other major industrialized countries, investment efforts of each economy must be recalculatedommon currency. Ideally, the procedure would be that followed in the bilateral comparisons presented above; investment expenditures of each country in the sample would be valued in the country's own currency and in rubles with expenditure weights similarly computed. This ideal is not practical for the multilateral comparison, so an expedient has been adopted which draws upon the research of Professor Irving Kravis and his team, prepared for the United Nations.
Kravis has computed bilateral comparisons between theand other countries, except for the USSR, in thein this study.omparisons between pairs of countries
other than the United States, in effect, are calculated through the common medium of dollar denominated international prices. Of the three possible weighting schemes used in the Kravis calculation, that of international prices is the only valid method for comparing outputs of more than two countries simultaneously.
The bias introduced by using international prices is to overstate the expenditures of less developed economies and relatively understate those of the most developed. Since the sample of countries is confined to industrialized economies, the
7. Seeiscussion of the Kravis team's methodology.
degree of bias is probably not large. It is certainly less than would result from using US prices.
For the US-USSR comparison geometric means were used to compare investment expenditures by category. This procedure closely resembles international prices in its results, as shown by such close identity in the Kravis study.
For total fixed investment the effect of using the other countries own prices rather than US prices is on the ratio of investment in the given country to investment in the United States. The difference isercent for the Soviet Union and onlyndercent for the other market economies and Hungary. For construction alone, the difference in using own country prices rather than US prices differs little among thepercent for the USSR comparedangeoercent for the other six countries. For producer durables investment, however, the spread in comparisons in different prices is much wider for theercent comparedangeoercent for the other market economies and Hungary.t must, therefore, be concluded that the bias introduced by using US prices particularly overstates the relative size of the Soviet investment effort.
ultilateral comparison Soviet per capita total fixed investment expenditures straddle those of the United States,. They are considerably below those for France,
8. Seeor the derivation of these estimates.
Germany, and Japan and far above these of Italy, the United Kingdom, and Hungary. Soviet housing investment is far below housing investment in other countries in the comparison. Including socialist Hungary, tending to confirm the general Impression of neglect of housing by Soviet planners.
for non-residential construction Soviet results are more respectable, but still far below those for the three market economies with the most concerted investment efforts. Again, Soviet inefficiency in construction is particularly marked, when one considers that the French and Germans devote half the Soviet proportion of GNP to this purpose.
In per capita investment in transport equipment tho Soviets rank far behind the United States, but at about the same lovel as the United Kingdom and Japan. Even in investment in other durables the Soviets fall below France, Germany, and Japan. Relative Soviet production efficiency in this area is evident from comparisons with France and Germany. While Soviet per-capita GNP is onlyercent of these two economies, the Soviet per capita investment ratio for other producer durables is just below parity.
Soviet investment performance is best conveyed in the comparison of non-residential investment. Here it straddles the United States, but is considerably below France, Germany, and Japan. Thus, Soviet allocation priorities not only induce an investment effort higher than that which would probably prevail
TABLE 11
MULTILATERAL COMPARISONS OF PER CAPITA FIXED INVESTMENT AND ITS PRINCIPAL COMPONENTS IN US DOLLARS6 <INTERNATIONAL PRICES) /
Franco Germany Japan
United Kingdom
ItalyI Hungary
Residential
Other
Total
Country France Germany Japan
United Kingdom ItalyI Hungary '
.
Other'oi
. a.
Total
2
Total fixed
a. See Appendix D. In the comparisons involving the USSR,xclud. capital repair and Variant II includes capital repair.
under consumer sovereignty, but further magnify this effort by restraining the investment claim of the housing sector.
VI. Inveatasnt Trend'sultilateral Perspective The more rapid growth in capital investment expenditures in tho USSR, as compared with the United States, can be extended to comparisons with the other major market economies in somewhat lesser degree. In bothndhe Soviet economy hasaggard in residential investment, but has increased non-residential construction and producer durables investment considerably more rapidly than most of the major market economies. The Soviet growth in total capital investment was conspicuously high, especially during.
Nonethelees, Soviet investment growth rates were matched and even exceeded by Hungary and other Eastern European economies.he economies with particularly high rates of growth in investment, such as Poland and Hungary, have also been faced with serious balance of payments disequilibria resulting from large purchases of investment goods from Western Europe, Japan, and the United States.
9. Unlike the Soviet investment indexes in which the construction component is independently derived from construction materials inputs, estimates for Eastern Europe countries are based on official statistics. Although there are grounds for believing these official estimates to be inflated, the degree of overstatement would not be large enough to negate the conclusions stated in the text.
VII. fectlvene3S_ cf Investment Tho high Soviet investraent/CNP ratios have meantelatively low proportion of national product has been available to the consumer, as highlighted by the consumption section of this study. The greater burden levied on the consumer to sustain an unusually high ratio of investment might be justified if thereommensurate payoff in higher growth rates. At least the consumer could be comforted by the claim that he or she were enduring austerity today so as toarger production capability tomorrow.
But during the past two decades this resource allocation strategy has failed to met its promise. The strenuous Soviet investment effort has not yielded viaably highor growth rates. Duringer capita CNF increasedlower rate in the USSR than in Japan, France, Italy, and Germany. In the first eight years of the seventies the Soviet growth rate has been exceeded by Japan and France and nearly equaled by Germany and the US.
Since economic growthunction of increases in employment, as well as capital stock (investment in its dynamicny analysis which purports to relate growth to Investment should abstract from the contribution to growth of increases in employment, especially for the Soviet Union, whose rates of increase in employment have considerably exceeded those of the Western European economies and Japan. In addition,
residential investment should be deducted from the total, as it contributes little to incroased output. Therefore, in the following analysis of investment effectiveness comparative returns on investment are measured by the ratio of growth of GNP per employed person to cumulative non-residential investment for the.
By this testhe return on investment in the USSR in terms of the per unit increase in GNP per worker obtainednit of non-residential investment was higher than for the United States and the United Kingdom, but considerably lower than for France, Germany, Japan, andurthermore, the rate of return fell significantly in. Therofore, the strenuous Soviet investment effort has not yielded comxensurately higher returns.
There may be some clues in the structure of Investment which, at least in part, explain the relatively low return on Soviet investment. As noted earlierhe Soviet construction investment ratio was particularly high This propensity was also evident in earlier years. Construction assets are longer lived than durables assets, hence the return per ruble investediven time period will be lower. Thus, the asset structure of Soviet investment has been biased toward lower rates
he low return in the United States is reflected in average GNP growth and unusually rapid growth in employment, and that in the United Kingdom in sluggish GNP growth. France and Germany enjoyod substantial GNP growth, but employment growth was slow.
of return. The same conclusion applies to the propensity of Soviet planners tomaller proportion of investment to replacement of obsolescent capital stock than the market economies.
Another explanation lies in the sectoral structure of Soviet investment. Tho outstanding feature of Soviet investment policy has beon the unusually heavy allocation to agriculture. This sector has been beset by unfavorable weather and inefficient organization. Moreover, much of investment in agriculture has been labor-substituting rather than output-increasing. Any belated attempt to rectify investment starvation in transportation, trade, and services will have unfavorable consequences for rates of return on investment in future years. Transportation has the highest capital requirements per unit of output of all the sectors in the analysis. Both the trade and services sectors tend to be construction-heavy with similar portents for return on investment.
VIII. Price Relationships for Construction and Producer Durables Revealed in Multilateral PPP Comparisons
An interesting feature of "the multilateral investment comparisons is the apparent inefficiency of the Soviet construction sector. While the Soviet share of GNP devoted to this purposes higher than for the market economies
its effective construction effort is far below those for France, Germany, and Japan.
This finding can be supportedultilateral comparison of foreign currency/dollar price ratios for construction and producer durables. What stands out so clearly inomparison is the lower purchasing power of the ruble in construction compared with producer durablos Investment, as compared with other economies.
The relatively high purchasing power (relatively low domestic currency/dollar ratios) for foreign currencies in constructione relatively high labor intensity of that expenditure, as compared with producer durables. Given the US real wage levels, it would be expected that the goods and services with, high labor intensity would be producedomparativedisadvantage in the United States. The US relative advantage, as the purchasing power ratios confirm, lies in the production of capital-intensive durables.
Since Soviet wage levels lie at or near the bottom of those for the countries in the sample, the ruble construction PPP ratio should reflect this condition. Instead, the Soviet construction PPP ratio is below that for producer durables. The most plausible explanation is the striking inefficiency of the Soviet construction sector. This hypothesis is consistent with other observations, both by Soviet and foreign analysis. Construction iu an activity which is particularly dependent upon efficient
)
Purchasing power values in durables column divided by values in construction column.
organization for its proper functioning. The Soviet system functions poorly in sectors like agriculture and construction in which organization of the work force is crucial.
By contrast the USSR appears to be relatively efficient in manufacturing producer durables. Here its effective effortomewhat exceeds its relative investment share. However, this conclusion must be qualified somewhat. The disequilibrium nature of Soviet pricing and the deficiencies inherent in Marxian price theory tend to understate the opportunity costs of manufacturing producer durables. The long omission of interest as an ingredient in cost and its continuing undercosting particularly underprices durables, given the high capital intensity characteristic of their Therefore, the atypical Soviet PPP relationship may reflect understatement of durables prices as well as inefficient construction organization.
11. In5 reform, an interest rate was introduced under the euphemism, "capital charge." However, the rate is set below equilibriumercent on the average and is differentiated perversely, with lower rates for heavy industry.
API KNDIX A
DERIVATION OF RUBLE-DOLLAR RATIOS FOR CONSTRUCTION The derivation of ruble-dollar ratios for construction has proceeded in three stages: erivation of ratios0 rubles and dollars;ovement of the foregoing derivation to one expressed in6 rubU-:; .md dollar:;, and 3) re grouping 6 estimates into classifications providing consistency between Soviet and US construction classifications and derivations of aggregate ratios.
The basic calculation of the ratios in detailonstruction sample0 rubles andhe choice0 was determined by the availability of Soviet data prepared for the capital stock revaluation23 and specifically denominated0 prices. Expenditure weights were also prepared for that year in the Soviet National Accounts study of the Central Intelligence
Derivation of Ruble-Dollar Ratios6 Prices
9 the ruble-dollar ratios were updated6 both by revisions in the expenditure weights and by appropriate ruble and dollar price indexes. Before doing that it was decided to reconsider the0 ratios to see whether they could be Improved in light of additional evidence gathered We
Central Intelligence Agency, Research Aid: Ruble-Dollar Ratios for Construction, ,
Central Intelligence Agency,
have decided that the original ratios can be Improvedo ways described below.
Revision0 Ratios
The first revision involves refining the comparison of multi-story US and Soviet buildings. Based on further investigation of Soviet buildings and designs, wo have decided that the comparisons of buildings with elevators could be improved over the original study. While elevators areelatively omallfour percent--of total construction costs in the US, Soviet elevators are expensive and frequently constitute more than seven percent of construction costs. In addition, the Soviet elevators purchased at proportionately greater expense are smaller, lees numerous, and leas reliable. Structures like those in the USSR simply would not be built in the US because they would not meet building codes. Thus, the comparisons of some US and Soviet buildings in the study result in downward biased ruble-dollar ratios.
The ideal approach to correct this deficiency is not feasible due to data limitations. It would involve costing Soviet buildings with US-style elevators and US buildings with Soviet-style elevators. Then two comparisons could bene for buildings with US-style elevators and one for buildings with Soviet-style elevators. Since the data preclude this kind of adjustment, the comparisons are improved by
the costs of Soviet buildings with more than four
stories to allow for improved elevators. The Soviet construction handbooks report the share of total construction costs incurred due to elevators. The average share of cost attributable to elevators is computed for each type of building and number of stories. Then we assume that the coat implied by this share would be double that amount to construct elevators comparable to US practices. While this assumption is highly subjective, it probably errs by being too conservative. By re-estimating the ruble costs of buildings with elevators, the ratios for office buildings, schools, and housing increase slightly. The revised ratios0 resulting from this procedure appear inwenty percent allowance for the average cost overrun of
builders in excess of the estimates.
A second refinement of the original study is the method by which the aggregate ratios are computed. There were some categories of construction lacking computed ratios that previously were omitted in the determination of the aggregate ratios for construction. This is not important if the ratios for the omitted categories approximately equal the average ratio for the types of construction that are already incorporated in the sample of ratios. In retrospect, this view seems unwarranted; rather the ratios for the omitted categories probably fall towards the high end of the spectrum of ratios observed in the construction sample. Fortunately the construction
Revised Ruble-Dollar Rates by Type of Construction,0
a/
Buildings
a/ The hospital ratio is the geometric mean of the housing, office building, and school ratios. The computed ratio for hospitals was rejected in the original study. See pagef Ruble-Dollar Ratios for Construction.
characteristics of some of these omitted categories are sufficiently close to some computed ratios to be included in the determination of the aggregate ratio for construction.
The two deleted categories In the Soviet case are construction for agriculture and the construction industry itself. 0 these combined categories received an8 percent of construction and installation work. 6 this share had grown8 percent, so their omission has gained added importance. The construction activity in the omitted categoriesubstantial amount of work that is analogous to industrial construction, roads, and airfields, especially oarthmoving and excavation. Therefore, the geometric mean of these three ratios is selected arbitrarily to represent construction for agriculture and construction. This change raises the aggregate ratio with Soviet weights because the three types of constructionigh earthwork component, work at which the Soviets operateelative disadvantage. The revised Soviet-weighted ratios are shown in.
Similar changes have been made in the US-weighted ratio. First, the basis of the weights has been changed from the series for "new construction put into place" to the "purchase of structures" in order to make the data parallel to the national income accounts. Second, the original study also ignored certain types of construction on the US aide. The coverage of the
weights could be improved significantly by the inclusion of construction for public utilities less railroads, sewer, water, and other municipal work. Based on the composition of these two categories, it is assumed that "public utilities less railroads" could be approximated by the geometric mean of the ratios for industry, streets, and airfields. Moreover, sewer, water, and other municipal work could be represented by the geometric mean of industry, street, airfields, and railroads. Both of these inclusions raise the US-weighted rublo-dollar ratio. The derivation of the0 US-weighteddisplayed in.
Derivation6 Ratios
Price indexes for both the Soviet and US sides are needed to update the ratios6 prices. While fairly decent price indexes may be derived from US data, the indexes for the Soviet side must be determined indirectly. Implicit GNP deflators for purchases of structures as computed by the Department of Commerce are used to adjust the US prices. The wealth of detail in the implicit deflators allowed applying unique price indexes to different types of construction, In some cases it is necessary to combine price deflators for private and public constructionertainospitals. For these cases, the data for the value of purchases06 in both current prices2 dollars are used to estimate an implicit price deflator
I?
=SI I
. I
si
* a
for the private and public sectors combined. The correspondence between categories is as follows:
Hospitals
Housing
Office Buildings Schools
Industry
Roadn
Airfields
Railroad
Category of GNP
Deflator
Private Hospitals and Institutions and Government Hospitals
Private and Government Residential.
Commercial
Private and
Government
Educational
Private and
Government
Industrial
Highways and Streets
Highways and Streets
Railroad
The derivation of the ruble price change06 is presented elsewhere in this volume in the article by John Pitzer. In brief heynthetic index of construction activity based on estimated material inputs in constant prices with an official Soviet measure that reports construction activity in current prices. By dividing the current price aeries by the constant price series, an implicit price deflator drops out. The
resulting deflator Implies that Soviet construction costs6 haveercent above thoseince Pitzer's methodology cannot discern between relative prices indexes for different types of construction,ercent is applied uniformly to all construction.
The ruble-dollar ratios computed6 allow for the Soviet propensity to incur cost overruns above estimate of approximatelyercent. The ratios by type of construction are:
djusted Ruble-Dollar Ratio
Mean
Buildings
his estimate may be conservative. An article in Stroitel'nava Gazetasserts the prices of construction and installationercentssuming the growth of prices was smooth throughout the period, this would imply construction costs6 are3 percent
The aggregate construction ratios6 are computed in the same way as6 Soviet weights are derived innd the Soviet-weighted construction ratio based on each category's price-adjusted ratio is given in. 6 US weights are derived as they were The US-weighted constructed ratio) uses0 ratios for the US categories, adjusted by the implicit price deflators for each country.
The ruble-dollar ratios06 in current prices
are summarized below:
Ruble-Dollar Ratios for Construction
Geometric
ew of the limitations in this update exercise should be mentioned. First, the overall increaso in Soviet construction costs is uncertain. While the results of John Pitzor's approach seem plausible, they are simply crude estimates. ore serious limitation is that the Soviet price adjustment is applied evenly across categories because the structure of the estimated cost changes is unknown. While this mightinor impact on the aggregate ratio, it could have substantial impact on the ratios for functional categories. Second, the methodology to update these ratios ignores possible
SI
iii iiii
changes in the mix of constructioniven category. For example, the distribution of housing by number of stories is held constant0 levels. While exact data are not available. It is apparent that Soviet buildings are now being built taller than previously. This means that buildings with elevators are more important, so that the ratios should be somewhat higher. Third, the update ignores changes in the Locational pattern of construction in either country. The average construction location in either country probably has changedlthough these three failings represent potential problems with tho ratios, some solace can be taken from the evidence that the ratios are not highly sensitive to these factors.
Regrouping6 Estimates into Consistent Classificatipns The derivation of ruble-dollar ratios for the principal categories of construction requires that there be consistency in definition between the two economies. In order to achieve this requirement some revisions have been introduced in the Soviet expenditure weights. These revisions and the revised category ratios are as follows:
USSR
States
a/ Ratio b/
Expenditurec/ Ratio d/
1.7
7
h/
Not separatelyJ/
APPENDIX B
DERIVATION OF RUBLE-DOLLAR RATIOS FOR PRODUCER DURABLES Ruble-dollar ratios for machinery and equipment (producer durables) are derived in three stages. Since the most recent general Soviet price revisions for durables occurred7 and the most recent information on the US investment shares of producer durables is found in the US input-output tablehe detailed ruble-dollar comparison is in terms7 rubles2 dollars. The detailed ratios are derivedublication of the National Foreign Assessment Center, Central Intelligence
The next step requires that the comparisons be moved forward In the absence of comprehensive price quotations and information on the final distribution of producer durables investment it is necessary to resort to appropriate price indexes to shift2 comparison forwardince information on US price trends is much more extensive and direct than that available from Soviet sources, different producers have been followed for each economy. The updating of dollar prices was prepared by Margaret Hughes and the updating of ruble prices by Stanley Cohn.
ational Foreign Assessment Center, USSR and US: Price Ratios forollars. Vols I and II.
Derivation of Dollar Price Indexes for Producer
Dollar price indexes for producer durables are based on line
item indexes calculated by the Bureau of Labor Statistics (BLS)
and by the Bureau of Economic Analysishe line item
indexes are aggregated into the larger groupings, necessary to
facilitate Soviet and US comparisonsy US expenditure
weights
The price indexes for component aggregated groupings, are as follows:
Transportation equipment
and agricultural machinery
and power equipment
machinery
oil field, and machinery
compressors, and materials
equipment
industrial equipment
aircraft
line items may be found in NationalCenter, op.olume I.
The BLS price indexes are published in Wholesale Prices and Price Indexes. 7 and the BEA indexes in annual issues of the Survey of Current Business.
Department of Commerce, Bureau of EconomicInput-Output Structure of the US Economy,
The aggregation of line items into component groupings is as
Transportation Equipment -
Trucks, buses, and truck trailers Autos
Ships and boats Railway equipment
and Agricultural Machinery -
Instruments -
Energy and Power Equipment -
Construction Machinery -
Mining, Oil field, and Metallurgical Machinerycomponents
Office, computing, and accounting machinery instruments
Fabricated metal products Engines and turbines Electrical transmission and distribution equipment Electrical equipment, nec.
Construction machinery
Mining machinery
Oil field machinery
Machine tools
Forges and presses
Casting machinery and equipment
Tools and dies
Compressors, and Materials Handling Equipment -
Ceneral industrial machinery Hoisting-transport equipment
Industrial Equipment and Aircraft -
Textile and apparel industry equipment Food industry equipment Printing industry equipment Construction materials industry equipment
Service industry machinery Communications equipment Aircraft
Other machinery and equipment
Derivation of RublePrice Index for Producer Durahlos,
While detailed group price indexes are available for moving2 dollar priceso reliable information is published regarding trends in Soviet producer durables prices2 If these ruble prices are not updated, the producer durables ruble-dollar ratios6 would not reflect the inflationary trends which have characterized Soviet machinery prices over the past dozen years.
The methodology selected to estimate price changes for machinery parallels that used to estimate the movement of ruble construction prices in Appendix A. Since no information is provided on the current costs of investment in producer durables or the factor inputs used in this investment process, the proxy of machinery production in current and constant values is usedeasonable approximation. The main conceptual flaw in this procedure is the assumption that price trends in total machinery production, which includes consumer durables, military durables, and net exports in addition to producer durables, do not differ significantly from such trends in producer durables output.
Machinery production in constant prices is approximated by the Soviet industrial production index constructed by the Office of Economic Research of the Central Intelligencehe current value of machinery production has been estimated for the
. Douglas Whitehouse, Ray Converse, "Soviet Industry: Recent Performance and Future Prospects," in Joint Economic Committee, Soviet Economy in Time of Change, p. .
eriod in unpublished estimotoo of Robert Abbott of the Office of Strategic Research, CIA, and foreriod by Williambbott's index has been moved forward7 by the methodology followed in the estimation of the construction price index foreriod When the constant and current price indexes are compared for the, an implicit machinery price index is derived. This index, is presumed to apply across all machinery categories equally, again paralleling the construction price estimate procedure.
The successive adjustments in ruble-dollar ratios for the producer durables component at each of the three stages are as follows:
Base
ollars
ollars
ollars
Ratios and
Lure
for
Component of
Expenditures
addition to shifting the ruble-dollar price ratios forwardt is also necessary to shift the expenditure weights forward. 6 dollar-ruble and ruble-dollar ratios and
illiam Lee, The Estimation of Soviet Defense
Expenditures, : An Unconventional Approach. p. ?5
expenditure weights for eight durables sectors are as follows:
States
Expenditure Dollar- Expenditure Ruble
Component(Percent)Ratio(Percent)Ratio
Transportation
Tractors and
Energy and
Construction
Mining, oil field, and metallurgical
Pumps, compressors and materials handling
Specializedand aircraft
Aggregate ruble-dollar
a. Reciprocal of dollar-ruble ratio, which
APPENDIX C
ESTIMATION OF VALUES FOR SOVIET AND US INVESTMENT6 Values for Soviet Investment
Soviet Investment in producer durables is largely composed of direct investment by state institutions, as reported in the annual economic handbook. This amount is supplemented by outlays by budgetary organizations (for education, health,he net change in uninstalled machinery andinally, the combined total, which is measured in9s recalculated6 price baseeflator previously calculated in the estimation of the ruble-dollar ratio for durables
Billion
Producer Durables
Budgetarychange in inventory of
uninstalled
Total9
producer durables investment
The construction component of investment Is composed of the officially reported value, supplemented by assumptions regarding construction expenditures included in the official Investment classifications "design-geological work" and "other capital work
CIA, p.
Those are defined on estimate prices adjusted for new equipment wholesale prices introduced on January 1,
and expenditures." Again the combined total, which Is measured in9s adjusted6 price baserice deflator originally constructed to calculate the construction ruble-dollar ratio
Billion
Publishedof design-geologicalof other capitalc/
Total construction investment
.
It is assumed that geological prospectingof capital investment type outlays, but that projectare an integral part of the investmentsome design expenditures relate to machineryit is assumed that the bulk are devoted toplans for buildings and structures.
to official instructions ink razrabotke qoaudarstvennykh planovkhoziaistva, the category of
other capital work and expenditures" includes outlays which would be classified as construction in United Nations and OECD accounts and other outlays which would not be considered of an investment nature. Included among the construction grouping are drilling costs, construction research and engineering costs connected with regional adaptations of techniques, drainage cost, land preparation expenditures, land protection costs, and permanent plantings. Excluded items include training costs for construction workers, administrative costs of construction control organizations, and additions of working livestock. It is assumed that two-thirds of total outlays in this miscellaneous grouping consist of construction type expenditures.
hese are defined as estimate pricesdjustedconstruction coefficients introduced on January 1,
G2
Reclassificationinstallation Expenditures
Before the official Soviet estimates on construction and producer durables investment can be used in international comparisons they must be adjustod. In the investment statistics for market economies, installation outlays are classified no equipment expenditures; Soviet statistics combine them with construction. The reasons for this difference are largely institutional. In most market economies firms that manufacture machinery and equipment include installation in their services; in the Soviet Union construction organizations are responsible for installation of machinery and equipment. Installation expenditures must therefore be shifted into the machinery and equipment category, with offsetting reductions in construction estimates.
Two' Soviet economists estimated that5 installation costsercent of total productivenclusion of installation would raise the producer durables component of investment from4 percent. Since nearly all producer durables are allocated to productive sectors, this adjustment is applicable to total investment as well. In thisid-range upward adjustment3 percent is assumed. Thare is an offsetting reduction in construction investment equal to tho absolute annual upward adjustment in the other investment category.
23. S. A. Efremov and P. D. Saeokhin, Nonnirovamo mcty. p..
6 the official value for the producer durables component of investment estimate is raised95 billion rubles and the official value for construction is reduced59 billion rubles. The same procedure is followed for other years.
Values for US Investment
Estimates of the construction and producer durables component of US investment expenditures6 are obtained from.figures published for the Bureau of Economic Analysis of the Department ofrivate sector investment consists of new private construction) and private purchases of durables Government sector investment consists of government non-defense purchase of structures and government non-defense purchases of durables
Dollars
(new) Government (new) Total
Investment
Dollars
Government Total
24. Survey of Current Business,
Soviet accountinguaai-investment concept not found in US national accounting practice, that of capital repairs. They are defined as outlays involving major replacement or renovation of parts of existing assets. Current repairs, on the other hand, are defined as essentially preventive maintenance or routine servicing expenditures. In the United States minor repairs would be charged against production cost and major repairs would be classified as new investment.
The estimated total value of capital repairs before adjustment is derived as follows:
Billion Rubles
Amortization1/
Budgetary2/
Collective farm3_/
Narodnoe khoziaistvo SSSR,.
Union-republic budget outlaysillion rubles (Gosudarstvennyi Biudzhetre increased by the average annual growth rateercent. This union-republic budget totalillion rubles is inflatedational (state) budgetary totalillion bynion-republic to state budget ratio5 percent, which pertained to the combined total for education and culture, health, science and state administration outlays (Gosudarstvennyi Biudzhet SSSR.9.
5 estimatea increased bynnual average growth rateercent (L. S. Golimon, Finansy
vo . Finansy, p.
To tho extent that capital repairsnvestment lives, they should be included as an investment component. The problem becomes one of determining what proportion cf capital repairs to include and how to distribute that share between construction and producer durables. One Soviet economistample survey in whichhird of recorded capital repairs of durables were restorative in nature, with the other two-thirds consisting of routinef this sample is assumed to be representative and if it further be assumed that most (two-thirds) of capital repairs of buildings and structures are genuine, given the nearly equal distribution of capital repairs between durables and plant andhen about half of recorded capital repairs fall within the definition of fixed investment expenditures. In this study half of capital repairs will bo so included. Total investment will be measured both inclusive and exclusive of repairs.
Ruble-Dollar Ratio of Capital Repairs
Given the assumptions concerning the composition of genuine capital repairs between construction and producer durables assets, it will also be assumed that the Soviet-weighted ruble-dollar ratio for repairs is determined by assigning
Iu. V. Kurenkov and D. M. Palterovich, Tekhnicheskiiptima 1'noe obnovler.ie proizvodstvennoqo p. .
Unpublished manuscript of Scot Butler, The Growth of Capital Ropair In the USSR, , March
construction double tho weight of producer durables.
These assumptions probably lead to overstatement of the ruble PPP for capital repairs. Repairabor-intensive activity. Total labor cost comprise aboutercent of total costs Cor repairs2 compared with labor cost share of onlyercent in the production of machinery andercent foro comparisons of repair activity in the USSR and United States are available, but qualitative impressions point to significant differences in technology. One major reason for the high labor input into repair in the USSR is that replacement spore parts often must be fabricated on the site while in tho United Stated spares are produced in adequate quantities by firms which manufacture the complete units. The high labor intensity and low productivity of Soviet repair operations, compared with US counterparts, wouldigher ruble-dollar ratio than those computed for new investment. Therefore, the assumed ruble-dollar ratio is likely overstated and becomes another source of downward bias in the ratios.
npublished collation of the2 Soviet input-output matrix, prepared by the Foreign Demographic Analysis Division, US Bureau of the Census.
APPENDIX D
DERIVATION OF MULTILATERAL COMPARISONS OF PER CAPITA INVESTMENT
Sources of Multilateral Investment Comparisons
The basic source for6 multilateral comparisons of
investment Is3 estimates calculated by Professor Irving
Kravis and his team for the United Nations International. Comparisonravis haseries of bilateral comparisons between the United States and selected
foreign economies of per capita expenditures of various components of GNP in dollars and in the other country's own currency. (His comparisons do not include the USSR; the US-USSR comparisons are based on the PPP ratios derived in Each component is further comprised of several representative products or services weighted by respective US and other country expenditure weights. What emerges from his calculations are per capita expenditure ratios comparing outlays in the United States and other economies in both dollar and the prices of these otherravis terms the ratio in dollar prices as "US weight" and that in the other country's currency as that country's weight, for example, "France weight."
II.
29. Ibid.. Summary binary tables starting on p. .
For multilateral comparisons the Kravis team uses the methodology of international prices. In this procedure dollar prices are established for each representative product or service. In order to overcome the quality comparison problem, prices are quality adjusted by linkage of countries with not widely differing levels of development rather than bilateral linkagearticular country and the United States. These separate country prices are then quantity-weighted by each country's production in order to obtain an international prices for each category of products.
In order to measure per capita investment expenditure ratios6 in the absenceetailed survey, it was necessary to adopt an expedient used by Kravis, Milton Gilbert, and other scholars engaged in international comparisons research. In its national accounts publications the OECD presents time series in both current and constanthe bilateral comparisons (foreign country vs. United Statea) may be updated36 by multiplying3 conversionsiven component of GNP by the ratio of the real growth in that componentn the foreign country to the real growth in tho component in the United States in the same time
National Accounts, .
example, the index of per) for Francond forStates if the ratio of the French to the) is multiplied by3 ratio ofUS prices), atio of2 6 is obtained.
Differences in Ratios of_Per_Capita Investment Expenditures Arising from Differing National Price Structures
The ratio of investment' expenditures of another country to those of the United States depend upon the prices selected. For reasons explained earlier, valuation in dollars almost always yield higher ratios than those in the country's currency. The greater the degree of similarity between relative prices in the various categories of Investment, the closer will be the ratios yielded by the use of alternative prices. The differences resulting from using different prices have been termed tho spreads in expenditure ratios. Other things being equal, tho nearer two economies are in their stages of economic development the smaller will be the spreads.
Empirically the spreads are calculated by comparing expenditure ratios in dollars and in other currencies for each component of investment. In terms of the present study this procedure would be equivalent to comparing component expenditure ratios in both sets of prices. 6 the implicit valuation spreads are as follows:
Spreads Between Investment Expenditure Ratios a/
Fixed
Construction Construction Durables Investment
United
8
9
9
Comparisons in dollars divided by comparisons in prices of other country.
Original document.
Comment about this article, ask questions, or add new information about this topic: