POLICY RESEARCH WORKING PAPER 1284
The Soviet Economic 'WVhat led to the relatKa Soviet
decline was reliance on
Decline capital accumulation and a
low elasticity of substitution
between capital and labor.
Historical and Republican Data Planned economies are
apparently less successful at
replacing labor effort with
capital. Tentativ" evidence
Stanley Fischer indicates that the burden of
defense spending also
contributed to the Soviet
debacle.
The World Bank
Policy Research Departnent
Macroeconomics and Growh Division
AprilI1994
POLICY RESEARCH WORKING PAPER 1284
Summary findings
Soviet growth for 196089 was the worst in the world, market economies, and whether this difficulty was
after controlling for investment and human capital. And related to the Soviets' planned economic system.)
relative performance worsens over time. Tentative evidence indicates that the burdeni ut defelns
Easterly and Fischer explain the declining Soviet spending also contributed to rhe Soviet debacle.
growth rate from 1950 to 1987 by the declining Differences in growth performance betweern tile SuvOe
marginal product of capital. The rate of total factor republics are explained by the same factors that tigure in
productivity growth is roughly constant over that period. the empirical crosssection growth iiterature: i, .al
Although the Soviet slowdown has conventionally been income, human capital population growth, anO the
attributed to extensive growth (rising capitaltooutput degree of sectoral distortions. The results Easteily and
ratios), extensive growth is also a feature of market Fischer got with the Soviet Union in the internaiurial
oriented economies like Japan and Korea. One message crosssection growth regression indicate that the planneui
from Easterly's and Fischer's results could be that Soviet economic system itself was disastrous for longi un
style stagnation awaits other countries that have relied on economic growth in the Soviet Union.
extensive growth. The Soviet experience can be read as a This point may now seem obvious but was not so
particularly extreme dramatization of the longrun apparent in the halcyon days of the 1950s, wliei, the
consequences of extensive growth. Soviet case was often cited as support for the neuLiassical
What led to the relative Soviet decline was a low model's prediction that distortions do not havL 5Leady
elasticity of substitution between capital and iabor, which state growth effects. Since a heavy degree of piloining
caused diminrishing returns to capital to be especially and government intervention exists in many countries,
acute. (The natural question to ask is why Soviet capital e, eciallv developing countries, the ill fated Soviet
labor substitution was more difficult than in Western experience continues to be of interest.
This paper  a product of the Macroeconomics and Growth Division, Policy Research Department  is part of a larger
effort in the department to study the determinants of longrun growth. Copies of the paper are available free from the Worid
Bank, 1818 H Street NW, Washington, DC 20433. Please contact Rebecca Martin, room NI 1043, extension 3 1320 (56
pages). April 1994.
The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about
development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The
papers carry the names of the authors and should be used and cited accordingly. The findings, interpretations, and conclusions are thc
authors' own and should not be attributed to the World Bank, its Executive Board of Directors, or any of its member countries.
Produced by the Policy Research Dissemination Center
THE SOVIET ECONOMIC DECLINE: HISTORICAL AND REPUBLICAN DATA'
William Easterly
World Bank, Room Ni 1043
1818 H Street NW
Washington DC 20433
and
Stanley Fischer
Department of Economics
MIT
Cambridge MA 02139
Easterly is in the Macroeconomics and Growth Division at the World Bank, and Fischer in the Department of
Economics at MIT, and a Research Associate of the NBER. Views expressed here are not to be attributed to the
World Bank.We are grateful to Karen Brooks, Alan Gelb, Eugene Gravilenkov, Barry Ickes, Barry Kostinsky,
Martha de Melo, Gur Ofer, Sergio Rebelo, Bryan Roberts, Marc Rubin, Randi Ryternan, Martin Schrenk, Marcelo
Selowsky, Michael Walton, and Alwyn Young for comments and useful discussions while writing this draft, to
seminar participants at MIT, the University of Michigan, the Federal Reserve Board, and the World Bank, to
Professor Mark Schaffer of the London School of Economics for kindly making his historical Western, official, and
Khanin data available in machine readable forn, and to Elana Gold and Mary Hallward for diligent research
assistance.
2
While the final collapse of the Soviet Union and Soviet communism now appean to have been
inevitable, It is esendal to try to pinpoint the causes of the economic decline, without which the Soviet
Union would still exist. The different accounts of the causes of declining Soviet economic growth
developed by Bergson (1387b), Desai (1987), Ofer (1987), Weitzman (1970) and others emphasize: the
Soviet reliance on extensive growth which, given the slow growth of the labor force and the failing
marginal productivity of capital, eventually ran out of payoff; the declining rate of productivity growth
or technical progress associated with the difTiculties of adopting and adapting to the sophisticated
technologics being introduced in the West (including East Asia as part of the West); the defense
burden; and a variety of special factors relating to the absence of appropriate incentives in the Soviet
system, including corruption and demoralization.
In this paper, we first place the Soviet growth performance in an international context using the
empirical crosssection growth literature. In section 1, we start with an overview of the data and of the
Soviet growth record, comparing it to other countries using a standarcd growth regression. We compare
the Soviet pattern of extensive growth (rising capital to output ratiom) to other countries. We reexamine
and update Sovietlevel production function estimates based on the official, unofficial (by the Russian
economist Khanin)l and western data, as well as examining other historical indicators of the Soviet
growth pattern. In section 11, we turn to a new data set: official data on republican output, capital
stocks (both in constant, or as the Soviets called it, comparable prices), and employment by sector.
These data have not previously figured in the Western literature on Soviet growth2. The fact that the
republics will now have to operate as independent economic units adds interest to our republicanlevel
1Ericson (1990) argues that the Khanirn data are preferable to the western data created by Bergson and others;
Bergson (199la, 1987a) criticizes the Khanin data for a poorly documented methodology and the apparent use of
unweighted averages of physical indicators. Harrison (1993) provides a more sympathetic analysis of Khanin's data,
emphasizing his attempts to adjust for the reporting biases inherent in the Soviet statistical system. We are grateful to
Professor Mark Schaffer for making the Khanin data available.
2The republican data were provided by Goskomstat of the Commonwealth of Independent States and by the Center
for Economic Analysis and Forecasting at the Ministry of Economics, and are available from Easterly at the World
Bank, 1818 H Street NW, Washington, DC 20433, We rescaled the data to reconcile different base years for the data
in compareble prices.
3
reso. We disus the patterns of growth by sector and republic, exploring crousection correlations
beow growth of the Soviet republics or sectors and conventional righthand side variables used in
growth regressions. In the conclusion, we offer some thoughts on interpretation of our results.
1. Ihe Soviet Growth Record
lhe ftndamental problem in evaluating Soviet growth is data qualir .3 Ofcial Soviet output
data overstate growth, as a result of both methodological problns  particularly in deflating nominl
d"a, and inentives to mis eport output within the Soviet system. Western analysis of Soviet growth
relies on the elsic studies of Bergson (1961), and other., as well as the CIA, which makes the
working assumption that physical quantities as presented in the otlicial data were not systematically
misreported. Thus the difference between the western estimate that per capita Soviet GNP increased
between 1928 and 1987 by 3.0 percent per annum (4.3 percent for aggregate GNP) and the official
estimate for NMP per capita of 6 percent per snnum results mostly from pricing corrections, and also
from differences m the coverage of NMP and GNP.4 The classic western estimates generally assume
that Soviet invetment nd capital data are more accurate than output data (Bergson (1987a), a view
disputed by Wiles (1982)). The western data through 1985 are conveniently summarized in Ofer
(1987).
We use four different data sets in the empirical work in this paper:
(1) The official Soviet Unionwide data on real output, industrial production, employment, and the
capital stock in the material sector in 1973 rubles, taken from official sources;
(2) Western data on output, industrial production, employment, and the capital stock, for the
Soviet Union as a whole: including
(a) the Powell(1963)/CIA(1982)/CIA (various years) series on value added and capital stocks
in indunry, and
3T1his discussion draws on material in Fischer (1992).
4The Soviet concept of Net Material Product omitted from GNP services not directly related to production, such as
pasenger transportation, housing, and the output of government employees not producing material output.
4
(b) the MoorsteenPowell(1966)/Powell(1968)/CIA(1982)/CIA (various years)/Kellogg(1989;
series on GNP, labor input, and capital stock for the entire economy. These series are chain linked,
using 1937 rubles for 1928960, in 1970 rubles for 196080, and 1982 rubles for the 198ts.
(3) Khanin's data, from Khanin (1988), also at Soviet Unionwide level, for output, employment
and the capital stock in the material sector; and
(4) Republiclevel data on aggregate and sectoral output and inputs in the material sectors for
197090 in constant rubles, which were made available by (ioskomstat.
The direct source of our datasets (1) through (3) is Gomulka and Schaffer (1991), who spliced
together series from the sources described. Note that the Khanin data are presented for the material
sectors (i.e. not including consumer services), as are the official data. Our preferred dataset for the
aggregate data will be (2); the others are presented to test the robustness of the conclusions to
alternative estimates of outputs and inputs.
A) SOURT GROWTH IN INTERMATIONAL COMPARSON
Growth rates of series in the first three data sets for different periods are presented in Table 1.
The Western output per worker growth rates are well below the official rates, with the Khanin data in
turn below the Westem data. All series show growth declining sharply since the 1950s.
How does the Soviet growth record compare to the rest of the world? We use the Western GDP
series to compare Soviet per capita growth over 196089 with World Bank per capita growth rates for
102 countries (we look here at per capita ratiner than per worker growth to enlarge the sample of
comparators and make it consistent with the crosssection growth literature). The first column of Table
2 shows that Soviet per capita growth has been slightly above the global average over both 196089 and
197489.
However, Soviet growth no longer looks respectable once we control for the standard growth
detemiinants from the empirical literature. The last column of Table 2 shows the residual from
inserting the Soviet Union into the core regression of Levine and Renelt (1992), which relates growth
to initial income, population growth, secondary enrollment, and the investment ratio to GDP. The
LevineRenelt regression including the Soviet Union is as follows:
S
Per cap ha growth 6089 0.83 + 17.49 Inwstment 6089 .35 GDP per cavita 1960 +
(85) (2.68) (14)
3.16 * Secondary enrollment 1960 .38 Population growth 6089  2.34 Dummyfor USSR
(1.29) (22) (1.43)
103 observatioas, R2 .46. (standard errors in parentheses)
Except for population growth, Levine and Renelt showed these variables to be robust to alternative
specifications in growth regressions (although concerns about endogeneity remain). The regression
results are identical to the LevineRenelt original since we are dummying out the Soviet observation.
Excepting initial income. the values of the Soviet righthand side variables should have implied
very rapid growthpopulation growth was low, and secondary education and the investment ratio were
near the top of the distribution. Growth was only average, hence the large negative residual of 2.3
percentage points in 196089. It is notable that the only countries with worse residuals are generally
both small and poor: Surinaine, Jamaica, GuineaBissau, Liberia, Zambia, and Peru. Soviet per capita
income in 1989 was only half of what it would have been if the average relationship between growth
and the righthand side variables had held over 196089.
The Soviet residual in this OLS regression is not actually significant in a twosided test at the 5
percent level. However, the presence of so many small and poor countries among the large outliers
makes us suspect heteroskedasticity. The suspicion is justified. We split the 196089 sample into thirds
on the basis of total real GDP (i.e. population times PPP per capita income) and rerun the above
regression for the top and bottom thirds ranked by total GDP. (The USSR is included in the top third
ranked by total GDP and we continue to dummny it out.) The GoldfeldQuandt test statistic for
heteroskedasticity  which is equal to the ratio of the sum of squared residuals in these two subsample
regressions and is distributed as an F statistic with the number of degrees of freedom of the numerator
and denominator corresponding to the degrees of freedom in the subsample regressions  indicates that
we can reject homoscedasticity. The test results are as follows:
Sum of squared residuals in third of sample with lowest real GDP: 88.3
Sum of squared residuals in third of sample with highest real GDP: 33.9
F (29, 28) = 2.61 (significant at 1 percent level)
6
3aod on the test results, we now perform weighted least square using the lo of total real GDP u the
weighting series. The results are now as follows:
Per capita growth 6089  0.43 + 15.93 Investment 6089 .28 GDP per capita 1960 +
( 73) (2.19) (.08)
2.56 * Secondary enrollment 1960  .24 Population g&owth 6089  2.28 Dummyfor USSR
(0.73) (.16) ;3. 48)
102 observations, R2 (weighted) .84. R2 (unweighted) =41.
The Soviet dummny becomes highly significant with weighted least squares, with a tstatistic of
4.8. Taking into account that only countries doing worst than the USSR were small economies makes
the Soviet performance look even worse. After correcting for heteroskedasticity, the Soviet economnic
performance conditional on investment and human capital accumulation was the worst in the world
over 196089.
How does the comparative Soviet performance evolve over time? Since the World Bank data
used by Levine and Renelt begins only in 1960, we compare the Soviet performance also with the
crosscountry SummersHeston (1991) dataset that extends back to ' )50. We perform a pooled time
series, crosssection regression using decade averages for the same specification as before (except that
we have to unfortunately omit the secondary enro!lment variable for lack of reliable Soviet data for the
1950s). We use the same Soviet data as in the previous regression, but now broken down by decade.
We put intercept dummies for each decade, as well as a separate Soviet dummy for each decade. We
continue to use weighted least squares with the weighting series being the log of total GDP, as the
GoldfeldQuandt statistic still indicates a significantly larger variance for small economies.5 The results
are:
5The Fstatistic for the ratio of the sum of squared residuals in the bottom third to that in the top third of the sample
ranked by total GDP (in PPP prices from SummersHeston (1991)) is F( 124,121)=2.03, which is significant at the I
percent level.
7
Pr. cpia roih by dwca  0 022 +.120 InvestmenGDP by dck  I 5E 06 GDPp capiet, initialye.y
(005) (016) (3.6E07)
.626 Popidauongrowh bydwde + .005 60sdummy .005 70sdummy.015 80 dummy
(143) (.004) (.O33) (003)
+ .024 Dummyfor USSR 50s  008 Dummyfor USSR 60s  017 Dummyfor USSR 70s
(01)) (010) (009)
. 023 DumyYfor USSR 80J
(009)
391 obueruao., R2 (wighte) .54, R2 (unweighied) .26
As is well known, world economic growth decelerated in the 70s and even more in the 80s. However
the Soviet growth deceleration is notable even by comparison with the world pattemn: Scviet economic
mrowth was significantly above the world average in the 1950s, and significantly below even the poor
world growth of the 1980s. Note especially the good performance of the Soviet Union in the 1950s,
even controlling for high investment: it suggests that whatever the weaknesses of Soviet central
planning in hindsight, these weaknesses were unlikely to have been apparent prior to 1960.
5) POSSIBLE EXPLANATIONS FOR POOR AND DECNING SOVIET GROWTH
We now consider other possible factors in the relative Soviet decline, including the defense
burden, demoralization, and Soviet disincentives for innovation. Could the poor and declining growth
performance be explained by the burden of defense on the Soviet economy? Although measurement is
problematic, the burden seems to have been high and rising. In Table 3, we show some estimates of the
Soviet defense burden as a share of GDP. Over the entire period since 1928, Soviet defense spending
has risen from 2 percent of GDP to the much higher levels of the midand late1980s, of around 1516
percent of GDP. Over the period 196089 in which the Soviet growth decline occurred, the rise in the
defense burden is more modest  from 1013% in 1960 to 12169% in the 1980s.
The international evidence for adverse effects of defense spending on growth is ambiguous 
see Landau (1993) for a recent survey. Landau (1993) himself finds an inverted U relationship: military
spending below 9 percent of GDP has a positive effect on growth, but defense above 9 percent of GDP
has a negative effect on growth. To see whether this affects the Soviet dummy in the growth
regressions, we insert defense spending into the decadeaverage growth regressions performed earlier.
We also include a variable meaurin war casualties per capita on national torritory to inswu that the
military spending variable is not simply proxying for wars. Because the military spending data is only
available for recent periods, we use data from the 1980. only.6 The regression including a quadratic
function of military spending is as follows:
Per capita growth 198C88  0.003 + i 27 InvwtmeP/GDP 198088 2. 7E064 GDP per capita 1980 +
(017) (038) (. IE06)
1. 34 Population growth, 198088 + .007 Secondary enrollment 1970 + .0081 MilItory spending/GDP
(38) (017) (.0024)
.00041 (Miltary sp.ndlng/GDP)2 0.746 War caswalti per capita .0155 Dummyffor USSR
(0001) (0.343) (.0268)
IS wigshted by log of total GDP. 77 observation,, R2 (weighted) .59, R2 (unrnwighted) .30 Standard ecors in
penthes.
We confirm Landau's result of an inv'ted Ushaped relationship between growth and defense
spending. Military spenduig reduces the magnitude and significance of the Sovie; dummy. However, as
Landau also noted 'is result is not very robust  omitting Syria and Israel from our sample eliminates
the significance of military spending. The defense explanation for the Sov.et decline is plausible but
not firmly established with crosssection data. We will test the defense hypothesis further with the
Soviet time series in the production function estimates below.
Another hypothesis about the Soviet growth decline is that it was related to the increasing
demoralizatioi. of the population, or alternatively to the increasing breakdown of worker discipline.
This breakdown of discipline could have resulted from the gradual opening up of the Soviet systern,
and the declining reliance on state terror.
Demoralization is obviously hard to measure, but we present some fragments of evidence. One
statistic relevant to demoralization is shown in Figure 1, which represents the results of a survey of
emigres which asked how satisfied they had been with the standard of living in the USSR. The young
had been less suisfied than the old. Among the many possible explanations for these results is that
6Landau only coven developing countries, so we use instead dat from Hewitt (1993) dut covers all countries
(including the USSR itself). The data for both Landau and Hewitt is mainly from SIPRI (the Stockholm Intemational
Peace Resarch Institute). The data on war casualties is from Eaterly, Kremer, Pritchett, and Summers (1993).
9
declining growth and disappointed expectations among the young were mutually reirfbrcinr.7
Other indicators of life in the Soviet Union also support the idea of a system breaking down.
Westem specialists were amazed to leam that Soviet male life expectancy actually declined in the 1970s
while other countries' (male) life expectancy rates were rising(Fiiure 2). Soviet life expectancy wu
declining even though per capita income growth was slightly above the world average, as we have
seen. There was a recovery in Soviet life expectancy in the 80s, but the USSR was stUi supassed
during the decade by developing countries like Mexico.'
Another possible explanation for poor and declining Soviet growth could be adverse incendves
under central planning for technological innovation (Berliner (1976)). A recent theoretical and
empirical literature argues that endogenous technological innovation, as measured by resources devoted
to research and development (R&D), significantly explains reladve growth performance across
countries (Coe and Helpman (1993), Lichtenburg (1992), Romer (1989); see Birdsall and Rhee (1993)
for a dissenting view).
Westem estimates of the Soviet research effort, presented in Figure 3, show R&D spending
rising as a share of GNP. The R&D share is above the 23 percent in the leading industrialized
countries. In 1967, about 1.5 percentage points of this was estimated to be for defense and space
(Bergson, 1983). The share of defense and space R&D in total R&D is believed to have fallen over
195984 (AclandHood (1987)), implying an even steeper rise in civilian R&D. The data on Soviet
R&D thus go in the wrong direction to explain either poor Soviet growth on average or the fall of
Soviet growth over timne.
7Among the other explanations: the young are chronic complainers; the old remember the period of much lower
consumption before the rapid Soviet growth of the 1950s; the authorities resisted emigration by the young, so that
any young emigre had to be more determined and disgruntled than the average emigre. Al>, since the original
source did not report standard deviations within the sample groups, we are unsure whether the differences are
statist,cally significant.
'One factor could have been the sharply rising consumpdon of alcohol in the 60s, 70s, and early 1 980s, which itself
may be an independent indicator of demoralization (TremI 1991). However, we are reluctant to make too much of
this since some countries with rapid income increases  like Korea  ilso had shaply rising alcohol consumption.
10
C) THE EA7ENSIVE GROWrH HYPOTHESIS
As noted, in the introduction, the conventional hypothesis for the Soviet growth decline is the
pattem of extensive growth, defined by Ofer (1987) as a rising ca; italoutput ratio, Figure 4 shows the
evolution of the capitaloutput ratios implied by the alternative data series for 195087.' The wstern
series shows the capitaloutput ratio increasing two and & half times between 1950 and 1987. The
official series also rises steadily beginning at the end of the S0s, more thar. doubling tetween 1958 ad
1987. The Khanin data, by contrast with the other two series, show only a small increase in the
capitaloutput ratio between the early 1950s and 1987. The capitaloutput ratios in industry first
decline in the S0s and then rise sharply after 1960, according to both Western and official estimates.
The behavior of the capitaloutput ratio is central to the debate about whether reliance on
extensive growth was the Achilles' Heel of Soviet industrialization, as the conventional wisdom has it.
In the neoclassical model, a rising capitaloutput ratio implies capital deepening during the transition to
a higher steady state, but this capital deepening will sooner or later run into dininishing returns that
will cause growth to slow or stop. The Soviet reliance on capital deepening is implicitly contrasted with
market economies, where according to the famous Kaldor stylized fact, capitaloutput ratios remain
relatively stable (recently reaffirmed by Rorrer (1990)). A constant capitaloutput ratio is consistent
with neoclassical steady state growth with laboraugmenting technical change. King and Rebelo (1993)
argue that capital deepening cannot account for much of sustained economic growth in the neoclassical
model without implying imnplausibly high rates of return to capital early in the transition process.
Nevertheless, recent research on capital accumulation in market economies casts doubt on
whether the Soviet extensive growth experience was unique. Appendix 1 lists the per annum growth
rates of the capitaloutput ratios in a selection of recent growth accounting studies and a few older ones.
All studies agree that the capitaloutput ratio in the U.S. has remained remarkably constant, which
9We begin the graphs in 1950 because we wre puzzled by the extreme volatility of all of the capitaloutput series
before 1950. We conclude that more even than the usual caution should be attached to results that rely on pre1950
data.
11
perhaps accounts for the conventional wisdom that Kaldor's stylized fact holds. However, a number of
recent studies point to capitaloutput ratios rising at Sovietstyle rates in Jpn and in some of the E'st
Asian NICs such as Korea (Young (1993b), Kim and Lau (1993), King and Levine (1994), Benhabib
and Spiegel (1992), Nehru and Dhareshwar k,.993)).
Moreover, the latter three, crosscountry, studies show that rising capitaloutput ratios are a
feature of growth for many countries. 10 The three studies compute capital stocks for a large sample of
countries, using a variety of data sources (SummersHeston versus World Bank) and a variety of
assumptions about initial capital stocks and depreciation rates. The three concur that rising capital
output ratios are by no means rare: the median capitaloutput ratio growth of their respective samples is
around 1 percent per annum, and fully a quarter of the samples' capitaloutput ratio growth rates are
over 1.7 percent per annum. 11 Nor is it only developing countries that are shown to have rapid
capitaldeepening. For example, the studies concur that capitaloutput ratios in Austria and France
increased at over 1.5 percent per annum. The literature on extensive growth as the bane of Soviet
development did not recognize that extensive growth also occurred in market economies, and
sometimes with striking results as in Japan and Korea. What is notable about the Soviet experience was
not the extensive growth, but the low payoff to the extensive growth.
As either a cause or a consequence of the low payoff, the level of the Soviet capitaloutput (K
Y) ratio had become extreme by the 1980s. The KY ratio as measured by the Westem GDP and total
capital stock series was 4.9 in 1985, which is higher than any of the 1985 KY ratios in the Benhabib
Spiegel and KingLevine exercises. In the NehruDhareshwar sample, there are only four countries
with a KY ratio above the Soviets in 1985, none of which seem especially relevant as comparators 
Guyana, Zambia, Jamaica, and Mozambique.
One other implication of the extensive growth model is that investment ratios have to rise over
I See also Judson (1994), who shows the capitaloutput ratio rising systematically with income.
I IFor the two studies that use SummersHeston data (Benhabib and Spiegel 1993 and King and Levine 1994), we
omit Africa from the sample because investment to GDP ratios are implausibly extreme (both high and low) in the
1950s.
12
time if growth is to be maintained while the capit%l output ratio rises. As has previously been
highlighted in the literature (see Ofer (1987)) the Soviet investment share doubled between 1950 and
1975. as can be seen in the CIA estimates presented in Figure 5. After 1975, the investment share
continued to increase, but more slowly.
How unusual is the doubling of the investment rate over a 25 year period? In the Summers
and Heston (1991) international database for 195075, 8 out of 52 countries  most notably Japan and
Taiwan  had a doubling or more of investment rates.12 Shifting the sample period forward by 10
years to expand the sample, 6 out of 72 countries had a more than doubling of investment over 1960
85, among which Korea and Singapore are of particular interest. Soviet investment mobilizadon wa at
a level that was above average, but not unknown, among market economies.
The standby of Soviet industrialization, machinery and equipment investment, also increased
sharply as growth declined. The importance of this sector to growth has been emphasized by de Long
and Sununers (1991, 1992, 1993); the Soviet data suggest a high ratio of machinery investment to GNP
is not sufficient to generate growth.
D) PRODUCTION FUNCTIONS AND EXTENSIVE GROWTH
Another way to evaluate the extensive growth hypothesis is to do the traditional total factor
productivity calculation. For the TFP calculation, there is little difference between the official and
westem data on factor input growth while Khanin shows substantially lower rates of growth of capital
(Table 1). This is a consequence in part of Khanin's view that hidden inflation is as serious in capital
goods industries as in consumer goods, a view shared by the "British school" of Hanson (19U), Nove
(1981), and Wiles (1982).
In Table 4 we show summary statisics for productivity growth for the USSR, ;lculated
assuming a CobbDouglas production function with labor's share equal to .6 and the share of capital
equal to .4 (slightly above that used by Bergson and the CIA, but within the conventional range for
12We contiue to exclude Afiica. countries from this and the following sample.
13
developing countries) for alternative data series. 13 With the assumption of CobbDouglas production
(unit elasticity of substitution between capital and labor) we see a strongly declining trend in TFP
growth after the 1950's.
The most interesting aspect of Table 4 is that the 1950s once more stand out as an exceptional
period in Soviet growth. It is especially striking that even the western data for the industrial sector
Lnply productivity growth in that decade of more than 6 percent per annum. Note the remarkable
divergences of views about perfornance in the 1930s that emerge from Table 4, with official Soviet
data showing extraordinary rates of productivity growth and Khanin and western GNP data imnplying
negative rates.
Westem GNP data present the most pessimistic assessment of Soviet productivity performance,
implying that productivity growth in the Soviet Union was positive only in the 1950s. The Khanin data,
which uniformily exhibit lower overall growth than western GNP data, by contrast imply positive
post1950 productivity growth, a result of the lower rates of growth of capital in the Khanin series.
The data in Table 4 point to one extremely important feature of the Soviet growth slowdown:
estimated productivity growth for the industrial sector was positive until the 1980s. This locates the
major slowdown of productivity in the nonindustrial sector. Looking ahead to Table 7, using the
aggregate of the republican data for 197090, the biggest problem was in agriculture, where
productivity appears to have declined by 4 percent per annum, with construction and trade and
procurement showing small positive rates of productivity growth.14
Following the pioneering work of Weitzman (1970, 1983) and later zontributions (Desai (1976,
13t has long been a stylized fact in the development literature that capital shares are higher in developing than
developed countries (see for example De Gregorio's (1992) estimate that the capital share is between .4 and .55 for
Latin America). Westem estimates of Soviet per capita income suggest it was a developing rather than a developed
country.
14Previous estimates of productivity growth in agriculture were less drastic but still showed poor performance.
Diamond, Bettis, and Ramson (1983) show productivity growth in agriculture of 0.2 percent over 197179. Brooks
(1983) show d zero agricultural productivity growth over 196080. We are not sure how our calculation of negative
agricultural productivity growth relates to Desai's (1992) evidence that weatheradjusted grain yields were rising
rapidly in the 1980s, unless the increase in yields was obtained through massive increases in inputs.
14
1987), Bergson (1979), and others), we also investigate whether CES functions provide a boeer
representation of the data than the CobbDouglas production function inposed in calculating the TFP
estimates in Table 4.
Weitzman's basic finding was that a CES producdon function with a low eluticity of
substitution of 0.4 fit the data better than the CobbDougla, and that the hypotheis that the ebsticity
of subsdtution was one could be rejected. Bergson (1983) cridcized this result, on the grounds that it
implied implausibly high estimates of the marginal product of capital in earlier yea. Desai (1987)
concurred with Weitzman's finding for aggregate industry, but claimed that CobbDouglas was an
adequate representadon for some branches of industry.
Estimation of production functions in industrial countries is the subject of a large literaure.
The usual method is to estimate parameters of factor demands derived from the cost funcdon, the dual
of the production function (see Jorgenson (1983) for a survey). This is obviously inappropriate for a
nonmarket economy like the Soviet Union. Direct estimadon of production funtions is usually
thought to be tainted by endogeneity of the factor supplies, particularly capital; we believe this would
be much less of a problem in the nonmarket system of the USSR. Table 5 shows elasticities of
substitution estirnated by nonlinear least squares (see Appendix 2 for the regressions), and recalcubted
TFP growth rates for 195087 (assuming Hicksneutral technical progress) for subperiods with the CBS
form:
ln(Y/L) = cl *Time*D5059 + c2*Time*D6069 + c3*Time*D7079 + c4*Time*D8087
+ c5* ln[c6*(K/L)lI/c5 + (1c6)] + c7
We find indeed that, with the exception of the estimatea based on the Khanin data, all of the altermative
estinates of Soviet output and capital growth per worker lend themselves to the CES form with low
elasticities of substitution between capital and labor (signiflcantly below one).l5 The results are less
ISA clasic uticle by Diamnond, McFadden, and Rodriguez (1967) shows that it is in general impossible to identify
sepurely a timevarying elasticity of substitudon pamneter and the bia of technical change (neutral verus labor
augmenting etc.) We identify the substitution paraneter by presuming that it is constant over time and that technical
change is neutral.
1s
sharp when we use the entire sample 192887, where as indicated earlier the data before 1950 are
volatile. Serial correlation is a problem for most of the estimates, with the significant exception of the
results using our preferred Western GDP series for 195087.
The results with the Khanin data are intriguing because they support a story of unit elaticity of
substitution and poor (though not necessarily declining) productivity growth. According to the Khanin
data, growth declines mainly because capital growth slowed (see Table 1 again). Given Bergson's
criticisms and the limited information about the methodology behind the Khanin data, these differing
results can only point to the need for further research into Khanin's approach to see wheher his work
represents a valid criticism of the Western estimates. For the moment, we are forced to regard the
Khlanin story as unproven. 16
The low elasticity of substitution from the other data series gives us an important insight into
the lack of success of the Soviet extensive growth strategy compared to the high payoffs from capital
deepening in Japan, Korea, and other market economies. The literature does not find the elastdcity of
substitution between capital and labor in market economies to be greatly below one (see for example,
Berndt and Wood (1975) and Prywes (1983) for discussion of theirs and other results for U.S.
manufacturing). A recent study estimating the elasticity parameter from the convergence behavior of
the crosssectional national per capita income data even argues that the elasticity of substitution is
slightly ABOVE one (Chua (1993)). Diminishing returns to extensive growth were much sharper in the
USSR than in market economies because the substitutability of capital for labor was abnormally low.
In the concluding section, we will speculate why substitutability may have been low in a planned
economy.
Another striking feature of Table S is that the implied rates of TFP growth show no significant
decline between the 50s and 80s. Thus, freeing up the functional form of the production function rules
16We would have liked to examine the implications of the "British school" of Hanson (1984), Nove (1981), and
Wiles (1982), who somewhat similar claims to Khanin's. However, we cannot do so since those researchers do not
provide alternative time series for output and capital. Note that a lower estimae for the growth rae of capital over
the entire period, as implied by the "British" arguments, would imply higher TFP growth but does not imply anything
about the estimated elasticity of substitution that would resuk from using such data.
16
out the collapsing productivity growth explanation for declining growth: in Table 4, both extensive
growth and declini productivity growth account for the overall fall in growth; in Table 5 the fault lies
endrely with dlminishing retuan to capital.
Table S also shows that the level of TFP growth is more plausible after we control for the
shrply falling marginal product of capital with a low elasticity of substitution. Our preferred estimates
are toe yielded by the Western GDP estimates in the last column in Table S for the 195087 sanple.
Those estimates yield a constant TFP growth rate of I percent per annum, in contrast to the negative
TFP growth implied by the CobbDouglas estimates in Table 4 for the 60s through the 80s. We find a
positive rate of TFP growth with falling returns to capital more plausible than a negative rate of TFP
growth.
In Figure 6 we examine a second implication of the estimates in Table 5: these are estimates of
the *share of capital" implied by the alternative columns in Table 5 for 195087, assuming marginal
productivity pricing. In the graph, we present only the more reliable western data. For total GNP, the
share of capital falls steadily throughout. For industrial output, the implied share of capital would have
been close to one until the mid50s, and it then would have begun a sharp decline to close to zero by
1980.
In Figure 7 we present closely related data, on the marginal product of capital implied by the
CES estimates. The western GNP data imply high rates of return to capital in the early 50s, declining
to about 3 percent in 1987. The 1950 marginal product of capital in industry is lower than that implied
by the GDP estimates. It stays constant throughout the 50s, then declines sharply to zero by the late
70s.
The data presented in Figures 6 and 7 suggest that a market economy could not have gone
through the growth process of the Soviet economy between 1928 and 1987. The very low wage shares
in the early period would probably have prevented any but a subsistence wage equilibrium in those
periods. The essentially zero marginal product of capital in industry (estimated using western data) by
the mid80s would have been inconsistent with equilibrium, and would have meant that investment in
industry and the capitallabor ratio would have been lower.
17
What would have happened in the early years if there had been a market economy? One
posibility is that some methodsuch as trade unionswould have been found to divorce factor
payments from marginal productivities. Another possibility is that different technologies would have
been adopted. Similarly, in the later period, there may well have been other technologies available that
yielded a positive return to capital. It is also possible that if the extensive growth route had been closed
off in a market economy, there would have been more incendve for Soviet entrepreneurs to attempt to
imnprove technology.
The high capital share in the CES production functions before 1960 has one other implicadon
we find interesting. A CES function with a high capital share acts much like a linear function of
apital, so that the marginal product of capital can stay flat for as long as the capital share is high (see
the line for indutrial capital's marginal product in the SOs in figure 7). With a very capitalintensive
producdon of goods, including capital goods, the Soviets were close for a while to the model of
growth through rapid reproduction of capital  described by Feldman in the 1920s as using "machines
to make more machines" (see Dornar (1957)).17
However, as the capital share begins to fall, the marginal product will begin to decline. The
decline can be precipitous when the elasticity of substitution is particularly low (see the industrial
marginal product in the late 50s and early 60s). While we find the extreme values of the marginal
product of capital and capital's share in Figures 6 and 7 surprising, they do not logically rule out the
CES formthe capitallabor ratio in a nonmarket economy could be driven to levels that would not be
observed in a market economy.
E) COMBINING REGRESSION EVIDENCE WITH PRODUCTION FUNCTION ESTIMATES
As a final exercise, we insert the other apparent correlate of declining growth  defense
spending _ into our production function estimates (we take the midpoint of the Brada and Graves
estimates in Table 3 spliced together with the Steinberg estimate for 198587). Specifically, we allow
17Rebelo (1991) shows formally that constant retums to reproducible factors in the capital goods sector is sufficient
to geneate a constant, sustained rate of growth even without TFP growth.
18
the Hicksneutral rate of technological progress to depend linearly on the share of defense spending in
GDP in the production hnction estimated with Western GDP and capital stock data:
ln(Y/L)  cl*nme + c2*nme*(Defenre Spending) + c3* ln[c4*(KfL)l/C3 + (1c4)] + C5
The results are shown in Appendix 2. We find defense spending does indeed have a significant and
negative effect on the rate of incr.ase in the total productivity term in the production function.
However, the effect is not very quantitatively important: every additional 1 percent of GDP spent on
defense lowered productivity growth by .07 percent. The increase over 196087 of 2.2 percentage
points in the defense share thus would have lowered growth by .15 percentage points. Moreover, the
parameters of the CES function are virtually unchanged from our earlier regression so the low
substitutability, diminishing returns story still holds We also tried equipment investment and R&D
spending as independent influences on the technical progress term, but both gave insignificant results.
How do we reconcile our production function estimates with our earlier crosssection growth
regression evidence using the LevineRenelt specification? The Soviets' high capitaloutput ratio and
low substitutability of capital for labor implies a lower derivative of growth with respect to the
investment rate than in other countries with lower KY ratios and more substitutable capital for labor.
To see this, assume zero depreciation and labor growth for simplicity and set labor = 1 by choice of
units. Assume a CES function Y=A(yKP+ Iy)(l/P). Growth will be given as a function of the
investment ratio (I/Y = AK/Y) as follows:
AY/Y = y (I/Y) [ (K/Y) (y+(ly)KP) ]I
As is well known, a higher K/Y implies a lower marginal effect of the investment ratio on growth
simply because a given investment rate translates into lower capital growth. With a unit elasticity of
substitution (p=O), this is the only way that the level of capital influences the marginal effect of
investment. With a less than unit elasticity of substitution (p < 0), higher capital has an even stronger
negative effect on the coefficient on investment in a growth equation. Although obviously not the only
explanation, this is consistent with the large negative residual for the USSR  and increasingly negative
residuals over time  in the crosssection regressions. (With only one observation, we cannot
distinguish between a Soviet slope dununy on the investmnent coefficient and a Soviet intercept dummy.)
19
We conclude from our reexamination of the aggregate data that the original Weitzman story
holds up. Soviet growth declined because of diminihig returns to capital accumulation, and not
because of a slowdown in TFP growth. The average growth performance was poor when we take into
account the rapid capital growth and high education levels. The general extensive growth hypothesis of
the literature on Soviet growth is not sufficient explanadon by itself, because in a comparative context
we find that Soviet extensive growth was not that unusual. It was the low substitutability of capital for
labor, rather than extensive growth per se, that was the fatal weakness of the Soviet development
strategy.
11. R _Ub_ican RMlts  Capitl Growth and _ow la Growth
The republican time series cover the period 197090, and provide detailed data to describe the
economic decline in the final years of the Soviet system, by republic and by sector. The data are for
NMP. Table 6 shows leastsquares estimates of real NMP growth per worker in the USSR and the
republics, overall, and by branch of industry, for the years 197090. Growth rates in the Central Asian
republics were well below those in the rest of the Soviet Union, with Belarus and Georgia having the
highest rates of output growth. Among the Central Asian republics, Turkmenistan experienced
negative growth of per worker output over the 20year period; the growth rate of per worker output of
Kyrgyzstan, the most rapidly growing of the Central Asian republics was nonetheless a full percentage
point below that of the slowestgrowing of the other republics, Azerbaijan. Growth performance in the
Baltics does not stand out relative to the Soviet average.
Across branches, output per worker grew most rapidly in industry, and at a negative rate in
agriculture; output growth in the service sectors, transport and communications, and trade and
procurement, was positive. Belarus shows the highest rate of growth of output per worker in industry;
Lithuania, where aggregate growth was relatively low, also shows rapid output growth in industry.
The slow growth of output in the Central Asian republics clearly owes much to the poor performance
of agriculture in these relatively agricultural republics.
Table 7 shows estimates of TFP growth (computed assuming a 0.4 capital share for all sectors
20
and a CobbDouglas form) by sector and by republic. 18 Judging by TFP growth, industry did
relatively better than other sectors in the European USSR Oust as industrial productivity growth is
usually higher than other sectors in the West), while agriculture was a disaster everywhere. Transport
and communications did well in the border regions of Belams and the Baltics. Productivity growth in
construction and in trade is uneven and generally close to zero. Central Asia is an almost unrelieved
tale of woe for all sectors, with Kyrgyzstan standing out again as having the best performance in that
region.
For the entire material sector's TFP growth, Georgia and Belarus did the best over 197090,
Armenia, Azerbaijan, Latvia, and Estonia, the next best, and Central Asia the worst. The relative
success of Belarus and Georgia was due entirely to industry, with productivity performance in
agriculture still disastrous, and performance in the transport/communications/construction sectors
generally poor.
The relative performance of the republics shown here is broadly consistent with previous
studies focusing on earlier periods. The ranking of TFP growth by republic for the period 196075 in
Koropeckyj (1981) is similar to that in Table 7: Belarus is at the top, and Central Asia at the bottom.
Whitehouse (1984) presents similar findings for 196170: Belarus and Georgia are third and fourth in
productivity growth (Latvia and Estonia are at the top), with Central Asia again firmly ensconced at the
bottom. The bad Central Asian outcome is well known in the literature (for example, Rumer(1989)).
A) EXPLWNING RELATIVE PERFORMANCE WITHIN THE USSR
Growth by republic is correlated with some of the same factors  human :apital, initial income,
and population growth  that have been singled out in recent crosssectional growth regressions (Barro
(1991), Barro and Lee (1993), Levine and Renelt (1992)). We first examine simp'e correlations
between estimated productivity growth over the period and these factors, and then present the results of
a multiple regression.
I tWe are rather embarrassed to resort to the CobbDouglas form for the republican sectors after rejecting it for the
Soviet Union as a whole in section 1. The republican data series are too short to lend themselves to CES estimation
of individual production functions. The CobbDouglas TFP growth rates are still useful descriptive statistics.
21
Pigue 8 shows the associadon between one measure of human capial  the percentage of
specblists with higher education per capita  and productivity growth. The negative productivity
growth of the Central Asian republics is associated with a low level of higher education, while the
relatively high growth of Georgia, Latvia, Estonia, and Amenia is strikingly correlated with a high
proportion of highlytrained specialists. Belarus is well above the implied regression line, reflecting its
relatively low dwhre of higher education specialists. Sinilarly, the Central Asian republics' poor record
is associated with rapid population growth (Figure 9).
We have also run a standard crosssectional growth regression for the fifteen republics. While
we have not found republican data to match the international data in the LevineRenelt regression we
presnted In Section 1, we can do a similar crosssection across republics. The results, presented in
Table 8, are similar to those obtained in the standard crosscountry regressions. The educational
variable has a posidve coefficient, while those on initial incomne (relative to the Soviet average) and
population growth are negative. The coefficient on population growth is much larger than is normal in
crosscountry regressions.
The coefficient on initial relative income implies that the rate of convergence between the
Soviet republics is over 4 percent per year (taking the derivatives at the Soviet average). This implies a
rate of intermal convergence for the Soviet republics considerable faster than the convergence
coefficients found by Barro and Sala i Martin (1992), which are around 2 percent per annum for both
US states an* a sample of 98 countries. Chua (1993) shows that convergence is more ranid the lower
is the elasticity of substitution between capital and labor. The rapid convergence of Soviet republics to
each other (though admittedly b. ed on the tenuous evidence of 15 observations) is yet another
confumation of the stronger force of dirninishing retums (and possibly the lower elasticity of
substitution) in the Soviet Union compared to market economies. 19
19An altemnative explanation for faster convergence among Soviet republics is that Soviet policymakers placed more
emphasis on regional redistribution than did Westem policymakers.
22
B) SECTOPiAL PATTRNS AND PRODUCV/7 GROWTH IN THE REPEUCS
Table 7 shows enormous variations in productivity growth between sctors and republics; in
this section, we examine whether these variations are related to degrees of sectoral distortion. It is well
known that the Soviet Union (and socialist economies in general) had distorted sectoral structures of
production (Ofer (1987)). Ofer compared the sector employment and output shares in the Soviet Union
to those that would have been expected for a country of its pce capita income level, using the patterns
of sectoral shares and income established by Chenery, Robinson, and Syrquin (1986). He showed that
the Soviet services sector was smaller than normal, Soviet agriculture was larger than normal, and
Soviet industry roughly normal for a country of its per eapita income level. The atrophied service
sector has been documented also with recent data (Easterly, de Melo, and Ofer (1994)).
Table 9 shows the difference between the sectoral shares of employment that would have been
predicted by the republics' respective per capita incomes and their actual sectoral shares. The per capita
incomes are derived from the estimates of relative incomes by the World Bank, and then applied to the
per capita income for the Soviet Union as a whole in Bergson (1991b). The employment shares of the
comparators are taken from International Labor Organization (various years) for a sample of about 70
developing and developed countries. We see the basic pattern confirmed: agriculture is larger than
average in all of the republics for their respective income levels, and trade is smaller. Transport is
larger than expected in the republics. Industry and construction do not diverge as sharply from the
expected patterns.20
Are the sector imbalances related to relative productivity growth performance? The answer is
yes  sectors that were "too large" had poor productivity growth. Figure 10 shows a scatter of
productivity growth 197090 for the 5 sectors and 15 republics against the sectoral employment
20These measures are extremely crude and obviously reflect other factors besides "distortions". For example, the
fertile soils of Ukraine and Moldova might imply a larger agriculture share than per capita income alone would
predict. The variation in employment shares in the intemational data set is enormous and the residuais shown in
Table 10 are generally not statistically significant in OLS regressions (with the exception of many of the deviations
in republican trade shares). The sectoral employment deviations nevertheless remain useful descriptive statistics for
the nature of the republican economies. Note that employment statistics cover the entire economy, and so are
preferable to NW shares that only cover the material sector (not to mention the pricing problems).
23
deviations in 1970. Republican agricultural sectors that were above the predicted employment shares
also had poor (negative) TFP growth, while industrial and trade sectors below predicted shares of
employment had positive TFP growth. The slope coefficient in figure 10 is .11 (ten percentage points
excess employment share lowers TPP growth in that sector by 1.1 percentage points). The coefficient is
strongly significant. This result is reminiscent of the finding in market economies that distorted
sectoral price incentives or other measures of departure from comparative advantage are negatively
related to growth (Barro (1991), Easterly (1993), Edwards (1989), Fischer (1993)).
III. Conclusion: internretin the results on Soviet Growth
Our results confirm and update the results of Weitzman (1970) on the low Soviet elasticity of
substitution between capital ard labor, which combined with the Soviet attempt at extensive growth, is
sufficient to explain the decline of Soviet growth. The natural question to ask is why Soviet capital
labor substitution was more difficult than in Westem market economies, and whether this difficulty was
related to the Soviets' p!anned economic system.
Recent work on models of endogenous economic growth stresses the notion of a broad concept
of capital, including human capital, organizational capitol, and the stock of knowledge, which can
substitute easily for raw labor and perhaps replace it altogether (Rebelo 1991, Jones and Manuelli
1992, Parente and Prescott 1991). Conversely, one possible explanation for the Soviets' substitution
problems would be that, under an autocratically directed economic system, they accumulated a narrow
rather than a broad range of capital goods. Some forms of physical or human capital that were missing
would have been marketoriented entrepreneurial skilis, marketing and distributional skills, and
informationintensive physical and human capital (because of the restrictions on information flows). It
is more difficult to substitute more and more drill presses for a laborer than it is to substitute a drill
press plus a computerized inventory and distribution system for a laborer. There is nothing that
explicitly supports this conjecture in our results, but it is an interesting direction for further research.
The other message from our results could be that Sovietstyle stagnation awaits other countries
that have relied on extensive growth, a point that has been made forcefully for those extensive growers,
the East Asian Tigers, in several articles by Young (1992, 1993a, 1993b). After all, the USSR had its
24
period of rapid growth in the 30s through S0s when it appeared to be following a linear outputcpital
production function, as we have shown. If East Asian capitaloutput ratios keep rising until they reach
the extreme Soviet levels, they too could experience a drastic slowdown. Even if diminiing returns
are weaker in the East Asian economies (if, following our conjecture, they have been accumulatizn a
broader range of capital goods and experiencing higher substitutability between capital and labor),
diminishing returns would still eventually cause a growth slowdown. The Soviet experience can be
read as a particularly cxtreme dramatization of the longrun consequences of extensive growth.21
The crosssection results on republics, although based on a small number of datapoints, support
the idea that some cf the same factors that are argued to determine growth in the recent empirical
crosssaction literature  human capital, population growth, initial income, sectoral distorticns  also
mattered under Soviet central planning. Our results with the Soviet Union in the international cross
section growth regression indicate that the planned economic system itself was disastrous for longrun
economic growth in the USSR. While this point may now seem obvious, it was not so apparent in the
halcyon days of the 1950s, when the Soviet case was often cited as support for the neoclassical model's
prediction that distortions do not have steadystate growth effects. Since a heavy degree of planning
and govermnent intervention exists in many countries, especially in developing ones, the illfated Soviet
experience continues to be of interest.
21 Weitzman (1990) describes Soviet growth (as analyzed by Ofer (1990)) as the best application of the Solow
neoclassical model ever seen.
25
BIBLIOGRAPHY
AcuridHood, Mary. 1987. 'Estimating Soviet Military R&D Spending' in Jacovsen, ed., Tle Snviet Def1EePM page
136147.
Anderson, Barbara and Brian Silver (1990). 'Trends in Mortality of the Soviet Population", Saxig.Ecin=. Vol. 6, No. 3,
pp. 191251.
Barro, Robert J. (1991). 'Economic Growth in a Cross Section of Countries', ru Journualnf , 105, 2 (May),
407444.
 and Jongwha Lee (1993), "International Comparisons of Educational Attainment", of _Mn et ami
3?(3', 363393.
Barro, RJ. and X. SalaiMartin. 1992. "Convergence', Journal of Poliical Economy 100, 223251.
Benhabib, Jess and Mark Spiegel. 1993. "The Role of Human Capital in Economic Development: Evidence from Agregte
Crosscountry and Regional U.S. Data", forthcoming, InumalofMgna,
Berndt, Ernst and Dwid 0. Wood. 1975. 'Technology, Prices, and the Derived Demand for Energy". Ile Review
EconoMiCs and Statistics, Vol. LVII. No. 3, 259268.
Bergson, Abram (1979). 'Notes on the~ Production Function in Soviet Postwar Industrial Growth", '. aLLa C
'.connnmics, 3, 2 (June), 116126.
(1983). 'Technological Progress" in A. Bergson and H. Levine (eds), 'Tbe Soviet Economy: Toward the Year 'Q.
London: Allen & Unwin.
(1987a). "On Soviet Real Investment Growth', SgyieLStudine, 39, 3, (July), 406424.
(1987b). 'Comparative Productivity: The USSR, Eastem Europe. and the West. Americmn.Economic Review, vol. 77,
no. 3, pp. 342357.
 (1991a). "Real National Income Measurement In Soviet Perspective", paper prepared for the Bureau of Labor Staistia
(BLS) and Eurostat conference on economic statistics for Economies in Transition. Washingtcn DC, February.
(1991b). "The USSR Before the Fall: How Poor and Why", Iournal of Economic Persectives, 5, 4 (Fall), 2944.
Berliner, Joseph S. (1976). The _nnovatior Decision in Soviet Tndus . Cambridge, MA: MIT Press.
Birdsall, Nancy and Changyong Rhee, 1993, "Does Research awd Development Contribute to Economic Growth in Developing
Countries?', Policy Research Working Paper 1221 (World Bank: Washington D.C.)
Brada, Josef C. and Ronald L. Graves. 'Slowdown in Soviet Defense Expenditures." Southeconmic aurN. vol. 54 pp.
96984, 1988
Brooks, Karen McConnell. 1983. "Productivity in Soviet Agriculture". in D. Gale Johnson and Karen Brooks, eds.,
Prospects for Soviet Agriculture in the 1980s.
Chenery, Hollis, Sherman Robinson and Moshe Syrquin (1986). IndusMiaL4&aisnlidGrowh. New York: Oxford University
Press.
Chua, Hak. 1993. "Factor Substitution and Slow Convergence'. Economic Growth Center Discussion Paper, Yale University.
CIA (1982), USSR Measures of Economic Growth and Developmnen (Washington DC: US Government Printing Office)
CIA (various years), Hsndbook of Economic Stktistics (Washington DC: US Goverrnnent Printing Office)
Coe, David T. and Elhanan Helpman (1993), "International R&D Spillovers". NBER Working Paper No. 4444, Augm.
De Gregorio, Jose. 1992. "Economic Growth in Latin America" urnsl o_lm, Vol. 39 No.1, pp. 5985.
26
De Long, J. Bradford and Lawrence Summers. 1991. 'Equipment, Investment, Relative Prices, and Economic Growth.
Quarteriy lounrl of Ecorgomic, May 1991, Vol CVI, Issue 2, pp. 445502.
and (1992). 'Equipment Investment and Economic Growth: How Robust is the Nexus?' Ri r
on rcnmic Activity, 2, 157211.
and _ (1993). 'How Strongly Do Developing Countries Benefit from Equipment Investment?7, InurnLnf
Mnncsarj,Ecorimiza 32, 395415.
Desai, Padma (1976). 'The Production Function and Technical Change in Postwar Soviet Industry: A Reexawnatioa',
American Economic Review, 66. 3 (June). 372381.
 (198T). The Soviet Economy. ProbILm a nd PrQtpets. Oxford: Basil Blackwell.
 (1992). 'Reforming the Soviet Grain Economy: Performance, Problems, and Solutions', ricn EonomiC Reviaw,
vol. 82, no. 2, pp. 4954
Diamond, Douglas, Lee Bettis, and Robert Ramsson. 'Agricultural Production' in A. Bergson and H.S. Levine, ods.. Mm
,Soviet Fronomy: Toward the Year 2000, Allen & Unwin, Boston.
Diamond, Peter, Daniel McFadden. and Miguel Rodriguez. 1967, 'Measurement of the Elasticity of Factor Substitution and
Bias of Technical Change", in...
Domar, Evsey (1957). 'A Soviet Model of Growth', in Essayc in the Theory of Economic rrowth. New York: Oxford
University Press.
Easerly, William, Martha de Melo, and Gur Ofer. 1994. 'Services as a GrowthPromoting Sector in the Former Soviet
Union', in Mario Blejer, ed. _Macroeconomic ad Strumral Anpets of Ru_aiq c miC Reform, Intenutjoal
Center for Economic Growth.
EAsterly, William. 1993. 'How Much Do Distortions Affect Growth?' InalofMD i 32: 187212.
Michael Kremer, Lant Pritchea, and Lawrence H. Summers, 1993, 'Good policy or good luck? Country growth
performance and temporary shocks', Journal of Monetary Econnmics 32, 459483.
Edwards, Sebastian. 1992. 'Openness, Outward Orientation, Trade Liberalization, and Economic Performance', NBER
working paper.
Elias, Victor J. Sources of Growth! a Studv of Seven latin American Economies, International Center for Economic Growth,
1992.
Ericson, Richard E. (1990a). 'The Soviet Statistical Debate: Khanin vs TsSU', in H.S. Rowen and C. Wolf, Jr. (eds) Mm
Impoverished Superpower. San Francisco, CA: ICS Press.
Feschbach, Murray. 1983. 'Population and Labor Force' in A. Bergson and H.S. Levine, eds., The So Toward
the Year 2000, Allen & Unwin, Boston.
Fischer, Stanley (1992). 'Russia and The Soviet Union: Then and Now', NBER Conference on Trasition in Eastern Europe.
_____. (1993) 'The Role of Macroeconomic Factors in Growth', Joura of Mn om 32 (3), 485511.
Gomulka, S. and M. Schaffer 'A New Method of LongRun Growth Accouuing, with Applicaios to the Soviet Economy
192887 and the US economy 194978. for Eonomic Perforinne Discuscion Paper 14. Jazuay 1991.
Haon., Phillip (1984). 'New Evidence about Soviet Investment', Ri h; 13 September.
Harrison, Mark (1993). *Soviet Economic Growth Since 1928: The Alternative Statistics of G.1. Khanin', EwzmAsAa
Sidi, 45,1, pp. 141167.
Hewitt, Daniel (1993). 'Miliary Expenditures 19721990, The Reasons Behind the Post 1985 Fall in World Militay
27
Spending, IMF WP/93/18
International Labor Organization (various years), Handbwk of LAhor Statics
IMF, World Bank, OECD, and EBRD. 1991. A Study nf the Sokiet Pconomy (3 volumes).
Jones, L. and R. Manuelli. 'A Convex Model of Equilibrium Growth", 1 _urnal of PolitinAL cion 98, 10081038.
Joint Economic Committee (1990), Measures of Soviet GNP (U.S. Government Printing Office: Washington D.C.)
Jorgensen, Dale. 1986 "Econometric Methods for Modeling Producer Behavior", in Z. Griliches nd M. Intriliptor,
Hand,book of Roonomerrica, Vol. 3. Amsterdam: NorthHolland.
Judson, Ruth. 1994. "Do Low Human Capital Coefficients Make Sense? A Puzzle and Some Answers", mimeo, Departnent of
Economics, MIT.
Kellogg, Robert L. 'Inflation in Soviet Investment and Capital Stock data and its impact on meaurement of total factor
productivity", CIA Office of Soviet Analysis, May 1989.
Khanin, G. 1988. *Ekonomicheskii Rost: Al'ternativnaia Otsenka", Kammnais, (No. 17)
Kim, Jongl1 and Lawrence J. Lau. 1993. "The Sources of Economic Growth of the East Asian Newly Industrialized
Countries', Stanford University Mimeo.
King, Roben and Ro3s Levine. (1994). 'Capital Fundamentalism, Economic Development, and Economic Growth',
forthcoming. CarnegieRochester Series on Public Policy, forthcoming.
King, Robert and Sergio Rebelo (1993). "Transitional Dynamics and Economic Growth in the Neoclassical Model', &Mariin
Fconomic Review, September. Vol. 83, No. 4. pp. 908931.
Kingkade. W.W. (1987) 'Demographic Trends in the Soviet Union", in Joint Economic Committee, Gorbachev's Fcnomie
liAn. Volume 1, U.S. Government Printing Office, Washington D.C.
Koropeckyj, I.S. (1981). "Growth and Productivity' in I.S. Koropeckyj and Gertrude Schroeder (eds), ERornmic.q of Snviet
Regions. New York: Praeger.
Landau, Daniel. 1993. "The Economic Impact of Military Expenditures'. Policy Research Working Paper 1138, World Bank.
Levine, Ross and David Renelt (1992). "A Sensitivity Analysis of CrossCountry Growth Regressions", An
Review, 82, 4 (Sept), 942963.
Lichtenberg, Frank R., 'R&D Investment and International Productivity Differences," NBER Working Paper #4161,
September 1992.
A. Maddison, The World Economy in the 20th Century. Development Center of the Organization for Economic Cooperation
and Development, Paris, 1989
Marer, Paul, Jaos Avray, John O'Connor, Martin Schrenk, and Daniel Swanson. 1992, n_ie A
adg theDam, World Bank: Washington D.C.
Millar, J.R. and E. Clayton. 1987.Quality of life: subjective measures of relative satisfaction, in J.R. Millar, ad. Politics,
work, daily life in the USSR. Cambridge University Press.
Moorsteen, Richard H. and Raynond P. Powell (1966). rne Soviet Caitl Stock 19231962. Honewood. IL: Irwin.
Nehru, Vikram and Ashok Dhareshwar (1993), "A New Databse on Physical Capital Stock: Sources, Methodology, and
Results." Revista de AnaiV Eo l Vol. 8, No. 1, 3759.
Nove, Alec (1981). "A Note on Growth, Investment, and Price Indices", Soviet uSlim, 23, 1 (Jan), 143145.
Ofer, Gur (1987). "Soviet Economic Growth: 192885, 1a u_ L _, 25, 4 (Dec), 17671833.
28
 (1990). Macroeconomic Issues of Soviet Reforns', _ A , 297339.
Parente, Stephen L., and Edward C. Prescott, 1991. "Technology Adoption and Gromwth, N13ER Work Pper No. 3733.
Powell, Raymond P. (1968) *The Soviet Capital Stock and Related Stuaical Series for the War Yean', Me Econom
Growth Center, Yale University.
(1963). 'Industrial Production' in Abram Bergson and Simon Kuzneu, eds., T in the Soviet huani,
Canbridge MA: Harvard University Press.
Prywes, Menahem. 1983. 'A Nested CES Approach to CapitalEnergy Substitution'. Federal Reserve Dank of New York.
Rebelo, Sergio (1991). 'LongRun Policy Analysis and LongRun Growth', u f a 99, 3 (Jun),
500.521.
Romer, Paul (1990). 'Capital Accumulation in the Tneory of LongRun Growth', in R. BDra (ed), Mmax
M=. Cambridge, MA: Harvard University Press.
______ (1989). 'What Determines theRate of Growth and Technical Change?', World Bank PRE Working Paper #279,
September 1989.
Rumer, Boris Z. (1989). Soviet Central Asia A Tragic FaeiMe. Boston, MA: Unwin Hyman.
Solow, Robert M. (1956). 'A Contribution to the Theory of Economic Growth". f rterlX oumnal Of9knomia. 70, 1
(Feb). 6594.
Summers, Roben and Alan Heston (1991). The Penn World Table (Mark 5): An Expanded Set of Internaional Comparisons
195088'. Quarterly Journal of Economics CVI, 369406.
Steinberg, Dmitri. 1987. 'Estimating Total Soviet Military Expenditures: an Alterntive Approach Based on Reconstructed
Soviet National Accounts.' in Jacobsen, ed., *Te Snviet Defe
Treml, Vladimir G. 1991.'Alcohol Abuse'.in Jones, Connor nd Powell, eds. Soviet Social Prhlems
UNESCO (1975), UNESCO Staigicil Yearbook (United Nations: New York)
Weitznan, Martin L. (1970). 'Soviet Postwar Economic Growth and CapitalLabor Substitution', AzjiaEc4
Review, 60, 5 (Dec), 676692.
(1983) "Industrial Production' in A. Bergson and H.S. Levine, eds., The Soviet Rconomyc Toward the Year I=, Allen
A Unwin. Boston.
(1990). 'Comment', NR,R MacroecgRaMiGLAnna, 339342
Whitehouse, F. Douglas (1973). 'Demographic Aspects of Regional Economic Development in the USSR' in V.N. Bandera
and Z.L. Melnyk (eds). The Soviet Rconomy in ReganaI Persecixe. New York: raeger.
Wils, Peter (1982). 'Soviet Consumption and Investment Prices nd the Meaningfiuness of Real Investment', Snai%t.Si%d,
24, 2 (April), 289295.
Young, Alwyn. 1993b,'The Tyranny of Numbers: Confronting the Statistical Realities of the EAt Asian Growth Experinc'e',
mimeo MIT
' 1993s, 'Lessons from the East Asian NlCs: A Contrrian View', fortcoming,
M 1992, 'A Tale of Two Cities: Factor Accumulatiob nd Tecbial Chaunge in Hong Kong and Singapore', In
Olivier J. Blinchard and Stanley Ficher, eds., _ M A, Cambridge MA: MIT Preus.
29
Figure 1: Index of satisfaction with standard of living in USSR
reported in survey of emigres, 1983
2.8
2.7 Index ranges from 4 (very satisfied) to 1 I
(very unsatisfied)
2.6
5^2.5**_
2.4
2.3
Over 54 4154 3140 Under 31
Age group in survey
Source: Millar and Clayton (1987)
Figure 2: Male life expectancy at birth: USSR and world median
68
64
62
USSR (Source: Anderson and Silver (1990))
60
58
56
Median for 150 countries (Source: World Bank Economic andi
Social Database)
52
50
%n %O %O O % O % @ %  r ~00ac 0 0 0000 00
… /
!,       El !L O7 _ .1 12 t t toF0  !S ^ et !e C P40
1950 
1952
1954
c
19560
1958
1960
1962 1
0
1964 0r
1966 C.
0 a~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~C.
o1968
() 1970
0 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1
1972
1974
1976 a
II 1978
1980
1982
19840
1986
32
Figure 4: Capitaloutput ratios, Alternative Estimates
5,2
4.5
I CIA (GDP)
4 Khanin (NMP)
Ofricial (NMP)
3.5 __ _ _ _ _ _ _
3.
2.5
1.5
ob N °h °@ 0 N q ' @0 N w '
~I, ~ri i ',~ ~ C 'C "0 C "0 I~ 1 ~ @00 000
3.2 
3  Industry
(official)
2.8 Industry (West)
2.6
2.4
2.2
2
1.8 *. . _ _
1.6
o N t 'C eo o N § C@0 o N q WI 0 N q
0 0 un 0 b O V
1950
1952
1954
1956 \
1958 CA
1960 n
1962 \
1964
1966U
1968 S
o 1970
1972 0
B~~
U~~
1974 nr
1976 2
1978 .
1980
1982
1984
1986
Share of capital in output
0 0 0 0 0 0 0 00 0
c  i Lo . v Ub 0t J @z0 vo _
1950 _ _ _
1952
1956
1956 4
0 r~~~~~~~~~~A
1958 3
1960 c
1962
1964
1966 e0.
1968 C,
1970 u
1972 0
1974
1976 jCJ
1978
1980
1982
1984 ~ ~ ~ ~ .~
1986__ J
Marginal products of capital (including rising TFP level)
o o o o o
0 , _~ ,o "j o w> C. .
0 k A  U 4i U h w t I A
1950 t ,
1952
1954 ~ 1
1956 CL
1958 I
1960 U
1962 .
1966 1
196_ 0
00
1970
1972 0
1974
1976
1973 .
1980 O
1982
1984 0
1986 / _ Ci
Un
36
Figure 8: Growth and Human Capital, Soviet Republics
3%_
Belarus Anena GF
g2% AmiaG zl
; ~~~~~~~~Azerbaijan Latvia 8
Moldova a U
1% Lithuania Esoi
Kyrzstan * Uknaine * Russia Estonia
E0% a
1 0% fajikstan Uzbekistan
*° 1%1
Kazakhstan
2% a
Turkmenistan
3%
120 140 160 180 200 220 240 260 280 300 320
Specialists with higher education per 10,000 inhabitants, 1965
37
Figure 9: Growth and Population Growth, Soviet Republics
3%_
Belarus Georgia
_2t% n Aglenia
% ,Latva Estonia Azerbaijan
a Ukraine Moldova U
Russia LMIuania KyrgyZstan
0% iikt t
° I* 1%  ° Uzbekistan
h ! Kazakhstan
.2% I
Turkmenistan
0% 1% 2% 3%
rate of natural incrase (percent), 1970
Figure 10: Sectoral employment sbares and TFP growtb, 5 sectors and 15 republics
3% a a
a ~~~~S.ckn we apcu**e. cinshucfKit uak'y.
 .% Gwga a~ab ~wxjb rI and Sckr W. dmvegs
Gowmn Wed 1970 per cape*swm (Summs4iosn) gm fw4o~
U 1%  'mu *t e
%~~~~~~~~~~
i1% I* w* T '
5%
14%
15% 10% 5% 0% 5% 10% 15% 20% 25% 30%
Deviation from sector shae predicted by per capita incomw
co
39
Table 1: Soviet Growth Data, 192847
Period Industry, official  Industy,  Material sectors,  Material sctors, 1 Total economy,
Westerm hsnin official Western
Growth rates of output per worker, alternative estimates
192887 6.3%s 3.4% 2.1% 6.0% 3.0%
192839 12.S% 5.0% 0.9% 11.4% 2.9%
194049 0.1% 1.S% 1.0% 2.1% 1.9%
1950459 8.9% 6.2% 5.3% 8.3% 5.8%
196069 5.7% 2.8% 2.7% 5.4% 3.0%
197079 5.2% 3.4% 1.2% 4.1% 2.1%
198087 3.4% 1.S5% 0.2% 3.0% 1.4%
Growth rtes of capital per worker, altermative estimates
192887 6.2% 3.2% 2.3% 6.1% 4.9%
192839 11.9% 6.5% S.9% 8.7% S.7%
194049 1.5% 0.1I % 1.3% 2.7% 1.5%
195059 8.0% 3.9% 3.5% 7.7% 7.4%
196069 6.1% 3.4% 3.8% 7.1% 5.4%
197079 6.3% 4.1% 1.9% 6.8% 5.0%
198087 S.6% 4.0% 0.1 % 5.3% 4.0%
Note: growth rates ar logntbmc wrtsquares estimates.
Table 2: The Soviet Union in the LevineRenelt (1992) Growth Regression, 196089
Per capita
income, 1960 Population Investment
Per capita (Summers  growth,1960 Secondary ratio to GDP, Growth
growth, 196089 Heston PPP) 89 enrollment, 1960 196089 residual
Average for sample excluding
Soviet Union 2.00 1792 2.07 21% 21%
Soviet Union 2.36 2796 1.05 58% 29%/e 2.34
Rank of Soviet Union in
sample (out of 103
observations) 45 24 81 10 7 97
Sources: Datafor all countries except Soviet Unionfrom Levine and Renelt (1992).
Soviet data: Per capita growthWestern GDP described in text, updated to 198889 with Marer et al. (1992)
Per capita incomeBergson(1 991b) for 1985 PPP, bockcast to 1960 with per capita growth infirst column
Population growth: Feschbach (1983), Kingkade (1987), IMF et al. (1991). Marer et al. (1992)
Secondary education: UNESCO (1975 Statistical Yearbook), 1970from Marer et al. (1992)
Investment rate: Joint Economic Committee (1990), updated for 198889 with Marer et al. (1992) (JEC series available at 5 yr
intervalsfrom 196075, interpolated in between)
Growth residual: Residualfrom LevineRenelt regression offirst column on other columns
0
41
Table 3: Soviet defeme burden a share of GDP
Jwa6 and Gmws Brda and Gmvs
Oir (1987) (1988) High (1988)  Low Steinberg (1990)
(cW'rent rubles) (constant rubles) (constant rubles) (constant rubles)
1928 2%
1950 9%
1960 12% 13.34% 9.90%/
1961 13.86% 10.60%
1962 14.93% 11.39%
1963 15.49% 12.32%
1964 15.03% 12.17%
1965 14.49% 11.79%
1966 14.11% 11.54%
1967 14.40% 11.95%
1968 14.45% 12.14%
1969 14.61% 12.08%
1970 13% 13.83% 11.48% 13.28%
1971 13.56% 11.30% 13.76%
1972 13.80% 11.34% 13.61%
1973 13.33% 11.03% 13.14%
1974 13.71% 11.28% 13.15%
1975 14.14% 11.53% 13.57%
1976 14.32% 11.62% 13.30%/e
1977 14.07% 11.26% 12.98%
1978 14.00% 11.09% 13.08%
1979 14.53% 11.43% 13.05%
1980 16% 15.06% 11.82% 13.91%
1981 15.48% 11.75% 14.03%
1982 15.36% 11.70% 14.58%
1983 15.51% 11.63% 14.36%
1984 15.55% 11.57% 14.37%
1985 14.79%
1986 14.49%
1987 14.63%
42
Table 4: Toa factor productivity srowth rmtu, alteative smin, USSR
period Khanin Official Westem ot Official Weder at.
material acton material industrial indutrial GNP
sectors sctor s_ctor
192840 1.7 7.2 1.7 7.2 1.2
194050 0.2 2.5 1.1 1.7 0.2
19S040 3.8 6.0 6.1 4.1 1.3
196070 I.S 2.9 1.9 3.q 0.1
197080 0.4 1.4 2.4 1.7 0.8
198087 0.4 0.7 0.1 1.1 1.2
ounces: seW earlier description in text
43
Table 5: Elasitkis of substtuton and TFP growth with utmated CES hmcdows
Watern
Xanln Official estfmata Ocfidal
material material Industrial Indiustrial Western
sectors sectors sector sector GNP
For 1950L87 sampk:
Eaticity of subsitudon 1.11 0.37 * 0.13 * 0.40 * 0.37 *
TFP grwth In:
195059 0.11% 2.93% * 2.40% * 3.72% * 1.09% *
196069 0.07% 2.88% * 2.36% * 3.60% * 1.10% *
197079 0.30% 2.98% * 2.51% * 3.74% * 1.16% *
198087 0.35% 2.92% * 2.43% * 3.62% * 1.09% *
For entir sample period, 192887
Easticity of substitution 1.11 0.38 * 0.22 0.45 * 0.81
TFP growth In:
192839 2.03% * 3.34% * 1.38% 0.72% 0.52%
194049 1.17% * 2.18% * 0.72% 0.51% 1.32% *
195059 0.18% 2.96 % * 0.36% 1.48% * 0.21%
196069 0.33% 2.97% * 0.40% 1.27% 0.15%
197079 0.30% 3.05% * 0.43% 1.38% 0.18%
198087 0.22% 2.97% * 0.37% 1.28% 0.33%
Notes: * indicates elasticity of substitution significantly different than one or TFP growth
rates significantly different than zero. Full regression results given in appendix.
44
Table 6: Growtb rats of NM? pew worker 197090 cmoutt prkc ________
_ Total Industry Agriculture Trasort n& Construcimon Trade &
_ Commurncation Procurelment
USSR 2.8% 3.4% 1.3% 3.1% 2.7% _ 2.1%
Slavic: .
Russia 3.0% 3.5% 2.1% 3.2% 3.1% 2.4%
Ukraine 2.9% 3.2% 0.3% 3.1% 2.2% 2.2%
Belarus 4.5% 5.4% 0.3% 3.5% 2.9% 2.4%
Baltic/Moldavian:
Estonia 3.1% 3.8% 1.7% 3.6% J 2.6% 2.5%
Latvia 3.3% 4.3% 0.8% 5.6% 1.0% 2.0%
Lithuania 2.8% 4.9% 0.6 % 3.9% 1.4% 1.0%
Moldova 3.3% 3.3% 0.5% 4.1% 2.2% 2.4%
Transcaucasian:
Georgia 3.9% 4.5% 2.5% 3.1% 2.0% 2.5%
Armenia 3.4% 3.4% 0.8% 5.2% 2.7% 2.5%
Azerbaijan 2.7% 3.9% 1.9% 0.7% 2.7% 1.6%
Central Asian:
Kazakhstan 0.7% 0.6% 4.4% 1.9% 1.9% 0.4%
Turkmenistan 0.3% 0.6% 2.8% 0.9% 1.6% 1.4%
Uzbekistan 1.2% 2.2% 1.8% 2.7% 1.0% 1.8%
Tajikstan 1.0% 1.7% 1.8% 3.1% 0.6% 1.8%
Kyrgyzstan 1.7% 3.1% 2.4% 3.8% 1.4% 1 .1 %
45
Tablh 7: Total factor productivtty growth by sector and republik, 197O.90
Total lndusuy Apictlnws Transport & Conm ucdton Trade &
___________ ______Cormnmunication Procurement
USSR 0.8% 1.1% 4.1% 0.8% 0.2% 0.3%
SOWic:
Runsia  0.8% 0.9% 5.3% 0.7% 0.5% 0.4%
Ukraine 1.0% 1.3% .2.7% 1.0% 0.4% 0.6%
Belau 2.1% 3.0; 3.3% 1.3% 1.8% 0.3%
Blaic/MoldaWan:
Estonia 1.3% 1.8% 4.4% 1.6_% 0.2_% O.S_ %
LAuvia 1.4% 2.1% 3.3% 2.9% 0.9 % 0.2 %
Lithuania 0.6% 2.6% 4.0% 2.0% 0.9% 0.6%
Moldova 1.0% 1.2% 2.7% 2.0% 0.4% 0.5%
Tscaucasian:
Georgia 2.3% 2.6%  0.1% 1.3% 0.3% 1.0%
Armenia 1.8% j 1.8% 3.1% J 2.8% 1.1% 0.4%
Azerbaijan 1.4% 2.5% X 0.1% 1.2% 0.3% 0.2%
Central Asian:
Kazakhstan 1.1% 1.5% 6.4% 0.2% 0.3% 1.2%
Turkmenistan 2.0% 3.0% I 4.0% 2.1% 0.3% 0.3%
Uzbekistan 0.4% 0.5% 3.7% 0.6% 0.6% 0.1 %
Tajikstan 0.4% 0.3% T 2.9% 1.8% 1.1% 0.9%
Kyrgyzstan 0.2% 1.1% 3 .9% 1.7% 0.2% 0.7%

46
Table 8: Crosssecdonal growth regresion, 15 Soviet repubUcs
m.uu..uu.m.u.u. u u..mumnm.. m m.u..munu..u
LS // Dependent Variable is Total Factor Productivity Growth,197090
Number of observations: 15
VARIABLE COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
C 0.0303710 0.0244123 1.2440874 0.2393
SPECHI65 0.0001098 5.011E05 2.1910647 0.0509
INCOM60 0.0002895 0.0001477 1.9598548 0.0758
NATINC70 1.4371064 0.5552033 2.5884325 0.0252
Rsquared 0.686424 Mean of dependent var 0.006709
Adjusted Rsquared 0.600904 S.D. of dependent var 0.012018
S.E. of regression 0.007592 Sum of squared resid 0.000634
Log likelihood 54.25'78 Fstatistic 8.026420
Prob(Fstatistic) 0.004105
= r.rmnn.i.i.i..r.ir..rnrr..~ur...inrn.ini.imu.nmrm.uumiUU
Variables:
SPECHI65  number of specialists with higher education per 10,000 inhabitants, 1965
INCOM60  income per capita relative to USSR, 1960
NATINC70 rate of natural increase of population, 1970
Source: official data.
47
Table 9: Sectoral employment Imbalances by republic
DeviationJfom employment shares predicted by per capita income, 1970
Transport
Industry Construction Agriculture and comm Trade
USSR 3% 1% 6% 2% 7%
Slavic:
Russia 6% 1% 1% 2% 7%
Ukraine 2% 0% 11% 1% .8%
Belarus .1% 0% 17% 0% 8%
Baltic/Moldavian:
Estonia 7% 2% 3% 3% 7%
Latvia 7% 0% 4% 2% 7%
Lithuania 1% 2% 14% 1% 9%
Moldova .9% 1% 29% 1% 9%
Transcaucasian:
Georgia 6% 0% 16% 1% 8%
Armenia 3% 3% 4% 0% 8%
Azerbaijan 4% 1% 10% 2% 7%
Central Asian:
Kazakhstan 4% 3% 6% 4% 7%
Turkmenistan 11% 3% 16% 2% 7%
Uzbekistan 9% 1% 19°% 0% 7%
Tajikstan 9% 0% 20% 0% 8%
Kyrgyzstan 3% 0% 11% 1% 7%
Sources: International Labor Organization (various years)for international
employment data; see textfor sources on Soviet republics
Note: Share deviations are calculated by regressing employment shares in 1970 for
nonsocialist countries on 1970 per capita income (SummersHeston), dummying out
the Soviet republics.
48
Appendix 1: Trends in capitaloutput ratios in growth accounting atdiu
Per annwn percent change in
County Period capital output ratio
This paper
USSR (Western GDP and Capital Stock Estimates) 195087 2.53%
Maddison (1989)
France 195084 .0.45%
Germany 195084 0.16%
Japan 195084 0.91%
United Kingdom 195084 0.62%
United States 195084 .0.07%
China 195084 2.48%
India 195084 1.54%
Korea 195084 0.03%
Taiwan 195084 0.34%
Argentina 195084 0.61Y
Brazil 195084 0.86%
Chile 195084 0.22%
Mexico 195084 0.50%
USSR 195084 3.75%
Yowig (1993)
Hong Kong 196691 0.84%
Singapore 197090 2.79s
South Korea (excluding agriculture) 196690 3.62%
Taiwan (excluding agriculture) 196690 2.55%
Kim and Lau (1993)
Hong Kong 6690 1.11%
Singapore 6490 1.38%
South Korea 6090 3.50Y
Taiwan 5390 3.13%
France 5790 0.68%
W. Germany 6090 1.16%
Japan 5790 3.19%
U.K. 5790 0.68%
U.S. 4890 0.19¶/'
Ela (1992)
Argentina 195080 0.39f
Brazil 195080 0.54%
Chile 195080 0.39%/0
Colombia 195080 0.79%
Mexico 195080 0.44%
Per 195080 1.22%
Venezuela 195080 0.75%
49
Trads in apitaloutput ratdos in growth aceoutldg studis (Appeudix I conL)
Per annum percent change in
Counny Period capital output ratio
Chenery, Robinson,_and Syrquin (1986)
Canada 4773 0.66%
France 5073 0.13%
Genmany 5073 0.33%
Italy 5273 0.63%
Netherlands 5173 0.17%
United Kingdom 4973 0.69%
United States 4973 0.00%
Benhabib and Spiegel (1992) (using Summers Heston data)
United States 196585 0.63%
Japan 196585 2.56%
Hong Kong 196585 0.20OYo
Korea 196585 2.78%
Singapore 196585 2.41%
Taiwan 196585 2.97%
75th percentile of sample (77 countries in sample,
excluding Africa) 196585 1.72%
50th percentile 196585 0.80%
25th percentile 196585 0.21%
Nehru and Dhareshwar (1993) (World Bank data)
United States 195090 0.20%
Japan 195090 2.70%
Korea 195090 3.70%
Singapore 196090 1.09%
Taiwan 195090 2.08%
75th percentile of sample (72 countries in sample) 195090 1.64%
50th percentile 195090 1.06%
25th percentile 195090 0.38%
KI.ng and Levine (1994) (SummersHeston data)
United States 195088 0.40%
Japan 195088 2.33%
Hong Kong 195088 0.80/o
Korea 195088 3.05%
Singapore 195088 2.94%
Taiwan 195088 2.63%
75th percentile of sample of 74 countries excluding
Africa 195088 1.69%
50thpercentile 195088 0.95%
25th percentile 195088 0.23%
Appendix 2: Nonlinear 1WM squars esImaon of CU h _om
Variable name for nHonllnr regressionl
dummy for 1928.39 D2839
dummy for 194049 D4049
dummy for 195059 D5059
dummy for 196069 D6069
dummy for 197079 D7079
dummy for 198087 D8087
Capitallabor ratio, industry, official KLINO
Capitallabor ratio, industry, KLINW
Western est
Capitallabor ratio, Khanin KLKHAN
Capitallabor ratio, material sectors, KLOFF
oMcal
Capitallabor ratio, whole economy, KLWES
Wtstern
Log of output per worker, industrial LYLINO
sector, official numnbers
Log of output per worker, industrial LYLINW
sector, Westeu estimates
Log of output per worker, material LYLKHAN
sectors, Khanin
Log of output per worker, material LYLOFF
sectors, official
Log of output per worker, whole LYLWES
economy, Western estimates
TIME*D2839 T2839
TIME*D4049 T4049
TIME*DSOS9 TS059
TIME*D6069 T6069
TIME*D7079 T7079
TIME*D8087 T8087
Ratio of defense spending to GDP DEFGDP
TIME (1,2,3,4,ETC.) TIME
(1) Results for 1928V7 ample
51
NLS // Dependent Variable is LYLKHMAN
SMPL range: 1928  1987
Number of observations: 60
LYLKHANC(1)*T2839+C(2)*T4049+C(3)*T5059+C(4)*T6069+C(5)*T7079+C(6)*T808
7+C(7)*LOG(C(S)*KLKHANA(1/C(7))+1C(8))+C(9)
.........m.. mamma am.... ma.  mmmaumumamma. a m a amam..  mm.umm
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
C(1) 0.0202626 0.0073460 2.7583265 0.0080
C(2) 0.0116552 0.0034600 3.3685319 0.0014
C(3) 0.0017588 0.0025248 0.6966280 0.4892
C(4) 0.0033136 0.0029598 1.1195481 0.2682
C(5) 0.0030160 0.0031696 0.9515282 0.3458
C(6) 0.0022460 0.0029139 0.7707873 0.4444
C(7) 9.8304594 113.76212 0.0864124 0.9315
C(S) 0.6102850 0.1464251 4.1678994 0.0001
C(9) 0.0916414 0.0363921 2.5181703 0.0150
Rsquared 0.983100 Mean of dependent var 0.471133
Adjusted Rsquared 0.980450 S.D. of dependent var 0.397586
S.E. of regression 0.055592 Sum of squared resid 0.157612
Log likelihood 93.12259 Fstatistic 370.8535
DurbinWatson stat 1.024675 Prob(Fstatistic) 0.000000
NLS // Dependent Variable is LYLOFF
SMPL range: 1928  1987
Number of observations: 60
LYLOFF C(1)*T2839+C(2)*T4049+C(3)*T5059+C(4)*T6069+C(5)*T7079+C(6)*T808
7+C(7)*LOG(C(8)*KiOFFA (1/C(7))+1C(S))+C(9)
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
ama a a a a a.... a a a a mamma a ama mamma mum mmma a mama a a aa a a mamma a a mammaS a..m..mamma.a a a
C(1) 0.0333843 0.0080809 4.1312477 0.0001
C(2) . 0.0218378 0.0044657 4.8901520 0.0000
C(3) 0.0296170 0.0047890 6.1844430 0.0000
C(4) 0.0296679 0.0059473 4.9884344 0.0000
C(S) 0.0304758 0.0058937 5.1708799 0.0000
C(6) 0.0297401 0.0054773 5.4296908 0.0000
C(7) 0.6260722 0.1706363 3.6690441 0.0006
C(S) 0.5403918 0.1322528 4.0860527 0.0002
C(9) 1.3706613 0.1645207 8.3312377 0.0000
Ruquared 0.996612 Mean of dependent var 2.117937
Adjusted Rsquared 0.996080 S.D. of dependent var 1.061674
S.E. of regression 0.066469 Sum of squared resid 0.225327
Log likelihood 82.40014 Fstatistic 1875.115
DurbinWatson stat 1.101054 Prob(Fscatistic) 0.000000
a mmmaa aa amama. mama am a a a amm, a a amm mmmaa mmmaamaa a ma a aassssssst=s
NLS // Dependent Variable is LYLWES 52
Date: 7201993 / Time: 23:35
SMPL range: 1928  1987
Number of observations: 60
LYLWES C(1)*T2839+C(2)*T4049+C(3)*T5059+C(4)*T6069+C(5)*T7079+C(6)*T806
7+C(7)*LOG(C(8)*KLWES^(1/C(7))+1C(S))+C(9)
Convergence achieved after 2 iterations
mummm u. u.mu m.mu......mu. m.... ua U.UUU r.. mm...................... mum ur......
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
m...u..................m...... m..........m...m.m......................n.m m....
C(1) 0.0051988 0.0086239 0.6028340 0.5493
C(2) 0.0131666 0.0045147 2.9163642 0.0033
C(3) 0.0020808 0.0048916 0.4253758 0.6724
C(4) 0.0014649 0.0058862 0.2488673 0.8045
C(S) 0.0018068 0.0059853 0.3018751 0.7640
C(6) 0.0033362 0.0060313 0.5531465 0.5826
C(7) 4.1327746 5.8211962 0.7099528 0.4810
C(S) 0.7331452 0.1523563 4.8120442 0.0000
C(9) 0.2293690 0.0861945 2.6610638 0.0104
mu.... mm...un... mmm...m... ma mm m.. m u...... mm
Rsquared 0.989484 Mban of dependent var 0.934737
Adjusted Rsquared 0.987835 S.D. of dependent var 0.549799
S.E. of regression 0.060641 Sum of squared resid 0.187543
Log likelihood 87.90639 Fstatistic 599.8574
DurbinWatson stat 1.118814 Prob(Fstatistic) 0.000000
NLS // Dependent Variable is LYLINW
SMPL range: 1928  1987
Number of observations: 60
LYLINW C(1)*T2839+C(2)*T4049+C(3)*T5059+C(4)*T6069+C(5)*T7079+C(6)*T803
7+C(7)*LOG(C(8)*KLINwA (1/C(7))+1C(S))+C(9)
mm.. m mm.m.... m. mm....... rn.....mm..m.m.m... mm....mm.....m..... mm...... m. m
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
C(1) 0.0137604 0.0079630 1.7280338 0.0900
C(2) 0.0071615 0.0038608 1.8549179 0.0694
C(3) 0.0035802 0.0026946 1.3286169 0.1899
C(4) 0.0040201 0.0038213 1.0520246 0.2977
C(S) 0.0043616 0.0051544 0.8461905 0.4014
C(6) 0.0036720 0.0054903 0.6688182 0.5066
C(7) 0.2743993 0.3211317 0.8544760 0.3968
C(8) 0.3542810 0.3325082 1.0654807 0.2917
C(9) 1.2839549 0.3090338 4.1547401 0.0001
,,,,,m.u m..... mmm.mum..... mm..mmmmm....m...mum.
Rsquared 0.979397 Mean of dependent var 1.911255
Adjusted Rsquared 0.976165 S.D. of dependent var 0.610544
S.E. of regression 0.094259 Sum of squared resid 0.453126
Log likelihood 61.44156 Fstatistic 303.0437
DurbinWatson stat 1.034563 Prob(Fstatistic) 0.000000
u mmm.... mm.m.. m.m...mm...m... mm...m..mu..
NLS // Dependent Variable is LYLINO 53
Date: 7201993 / Time: 23:38
SMPL range: 1928  1987
Number of observations: 60
LYLINO C(1)*T2839+C(2)*T4049+C(3)*TS059+C(4)*T6069+C(S)*T7079+C(6)*T808
7*C(7)*LOG(C(8) *KLINOA (1/C(7) )+1C(8) )+C(9)
Convergence achieved after 2 iterations
uummmmmmmmmmmmmmmmmmmmmmmminmummmummuummmmmmmminu=uuinmmuuminumumummuummmm
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
C(1) 0.0072341 0.0086047 0.8407205 0.4044
C(2) 0.0051121 0.0053042 0 9637879 0.3397
C(3) 0.0148348 0.0050212 2.9544124 0.0047
C(4) 0.0126798 0.0064987 1.9511147 0.0565
C(5) 0.0137831 0.0069631 1.9794368 0.0532
C(6) 0.0128241 0.0069723 1.8392831 0.0717
C(7) 0.8243762 0.3706599 2.2240773 0.0306
C(8) 0.4472637 0.1562502 2.8624840 0.0061
C(9) 1.4963393 0.37q1015 3.9470674 0.0002
s mwmmmum m mmmm=mmmmuummmmmmmmmmmmmmmuuw.Sumi==nmmu==
Rsquared 0.996675 Mean of dependent var 2.191933
Adjusted Rsquared 0.996154 S.D. of dependent ver 1.111757
S.E. of regression 0.068948 Sum of squared resid 0.242448
Log likelihood 80.20302 Fstatistic 1911.111
DurbinWatson stat 1.317886 Prob(Fstatistic) 0.000000
(2) Results for 195087 4mmpIe
NLS // Dependent Variable is LYLKHAN
SMPL range: 1950  1987
Number of observations: 38
LYLKHANmC(1)*T5059+C(2)*T6069+C(3)*T7079+C(4)*T8087+C(5)*LOG(C(6)*KLKHAN
A(1/C(S))+1C(6))+C(7)
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
C(1) 0.0011270 0.0048291 0.2333864 0.8170
C(2) 0.0006563 0.0051628 0.1271136 0.8997
C(3) 0.0030198 0.0050557 0.5972997 0.5546
C(4) 0.0034716 0.0045055 0.7705352 0.4468
C(S) 9.7629283 220.16876 0.0443429 0.9649
C(6) 1.1237026 0.2228861 5.0415996 0.0000
C(7) 0.1735591 0.1109200 1.5647225 0.1278
mmmm................... u.............mum..............mum........... m mu mu u u =. mum... mu m m = mum..... mu m.. mum,.
Rsquared 0.987796 Mean of dependent var 0.707439
Adjusted Rsquared 0.985434 S.D. of dependent var 0.305213
S.E. of regression 0.036836 Sum of squared resid 0.042064
Log likelihood 75.39730 Fstatistic 418.1935
DurhinWatson stat 0.505965 Prob(Fstatistic) 0.000000
mum........... mum mum mmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm...........m...mm.m, mmm m....mmm....mm.m
54
NLS // Depenoent Variable is LYLOFF
SMPL range: 1950  1987
Number of observations: 38
LYLOFF C(1)*T5059+C(2)*T6069+C(3)*T7079+C(4)*T8087+C(5)*LOQ(C(6)*KLOFF*
(1/C(5))+1C(6))+C(7)
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
C(1) 0.0292335 0.0027887 10.482690 0.0000
C(2) 0.0288169 0.0029999 9.6060223 0.0000
C(3) 0.0297372 0.0030072 9.8888315 0.0000
C(4) 0.0291584 0.0028214 10.334843 0.0000
C(S) 0.5812015 0.0565917 10.270078 0.0000
C(6) 0.5715239 0.0456052 12.531981 0.0000
C(7) 1.3912334 0.0869462 16.001090 0.0000
Rsquared 0.999488 Mean of dependent var 2.812825
Adjusted Rsquared 0.999389 S.D. of dependent var 0.594913
S.E. of regression 0.014707 Sum of squared resid 0.006705
Log likelihood 110.2862 Fstatistic 10084.79
DurbinWatson stat 1.390053 Prob(Fstatistic) 0.000000
na..... nm..... m. ann mmnam maanmn aama,amm.... an ..
NLS // Dependent Variable is LYLWES
SMPL range: 1950  1987
Number of observations: 38
LYLWES =C(1)*T5059+C(2)*T6069+C(3)*T7079+C(4)*T8087+C(5)*LOG(C(6)*KLWESA
(1/C(5) )+1C(6) ) +C(7)
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
C(1) 0.0109709 0.0032629 3.3623384 0.0021
C(2) 0.0109651 0.0035797 3.0631462 0.0045
C(3) 0.0116165 0.0035835 3.2417041 0.0028
C(4) 0.0109505 0.0034259 3.1963828 0.0032
C(S) 0.5958245 0.0966218 6.1665625 0.0000
C(6) 0.9598616 0.0134483 71.373976 0.0000
C(7) 0.8162872 0.0832581 9.8042949 0.0000
m..a mmm .mmnan m.mnu.a m. mm.. m...... aaa...., m...... mm.... ns
Rsquared 0.998747 Mean of dependent var 1.285964
Adjusted Rsquared 0.998505 S.D. of dependent var 0.356312
S.E. of regression 0.013777 Sum of squared resid 0.005884
Log likelihood 112.7701 Fstatistic 4119.769
DurbinWatson stat 1.922959 Prob(Fstatistic) 0.000000
a..anm.... mna... m...am.... m n... a..aa... m,. .. m...., naa wm
NLS // Dependent Variable is LYLINW 55
SMPL range: 1950  1987
Number of observations: 38
LYLINW .C(1)*T5059+C(2)*T6069+C(3)*T7079+C(4)*T8087+C(S)*LOG(C(6)*KLINWA
(1/C(S))+1C(6))+C(7)
 n....mm. m. .m.... m...fu.... m..........
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
.m... .......................... ..............m......mm...m......... m.........
C(1) 0.0240169 0.0021467 11.187968 0.0000
C(2) 0.0236218 0.0023585 10.015738 0.0000
C(3) 0.0251244 0.0021509 11.681057 0.0000
C(4) 0.0243103 0.0018522 13.124830 0.0000
C(S) 0.1441544 0.0330823 4.3574480 0.0001
C(6) 0.0020469 0.0031944 0.6407644 0.5264
C(7) 2.4449147 0.1045291 23.389801 0.0000
Rsquared 0.997683 Mean of dependent var 1.535708
Adjusted Rsquared 0.997234 S.D. of dependent var 0.417767
S.E. of regression 0.021971 Sum of squared resid 0.014965
Log likelihood 95.03323 Fstatistic 2224.312
DurbinWatson stat 0.506999 Prob(Fstatistic) 0.000000
NLS // Dependent Variable is LYLINO
SMPL range: 1950  1987
Number of observations: 38
LYLINO C(l)*T5059+C(2)*T6069+C(3)*T7079+C(4)*T8087+C(5)*LOG(C(6)*KLINO^
(1/C(s) )+1C(6) )+C(7)
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
m..u....m m...mmunau.... m_... m...am...... mm.,.,mm.
C(1) 0.0371878 0.0049959 7.4436178 0.0000
C(2) 0.0359650 0.0053985 6.6619911 0.0000
C(3) 0.0373821 0.0054797 6.8219713 0.0000
C(4) 0.0361438 0.0052913 6.8308024 0.0000
C(5) 0.6618038 0.1296726 5.1036531 0.0000
C(6) 0.1022757 0.0669595 1.5274265 0.1368
C(7) 2.7672364 0.2907247 9.5184097 0.0000
................ m... mm..um ..u....m.. .m.m.mu.mUu m.m.m....................mm................... mum
Rsquared 0.999228 Mean of dependent var 1.481138
Adjusted Rsquared 0.999079 S.D. of dependent var 0.657494
S.E. of regression 0.019953 Sum of squared resid 0.012342
Log likelihood 98.69529 Fstatistic 6690.964
DurbinWatson stat 1.008779 Prob(Fstatistic) 0.000000
56
(a) Regrsian wth Defense qpmndfng/rnP
NLS // Dependent Variable is LYLWES
SMPL range: 1960  1987
Number of observations: 28
LYLWESC(1)*TIME+C(2)*TIME*DEFGDP+C(3)*LOG(C(4)*KLWESA(1/C(3))+1C(4))+C
(5)
mmm u u mNm.. m.....m.. mm m mm mmmm am a m....... mm, mmmm.m.m., mm
COEFFICIENT STD. ERROR TSTAT. 2TAIL SIG.
mm..... m... mmmmmmmmmam...ammmmmmmiam.mmmmmmmammmmmmm
C(l) 0.0207134 0.0059611 3.4747771 0.0021
C(2) 0.0727455 0.0108381 6.7120112 0.0000
C(3) 0.5785414 0.2470464 2.3418330 0.0282
C(4) 0.9690962 0.0396033 24.470094 0.0000
C(5) 0.8963781 0.2645463 3.3883604 0.0025
Rsquared 0.996629 Mean of dependent var 1.468442
Adjusted Rsquared 0.996042 S.D. of dep)endent var 0.179790
S.E. of regression 0.011310 Sum of squaied resid 0.002942
Log likelihood 88.52034 Fstatistic 1699.821
DurbinWatson stat 2.232942 Prob(Fstatistic) 0.000000
mm... am....... a mam mmm ms.m.m mm. ... m. amimammmmma..ammmmmmm
Pollcy Research Working Paper Series
Contact
Title Author Date for paper
WPS1264 A Rock and a Hard Place: The Two J. Michael Finger March 1994 M. Patenia
Faces of U.S. Trade Policy Toward Korea 37947
WPS1265 Parallel Exchange Rates in Miguel A. Kiguel March 1994 R. Luz
Developing Countries: Lessons from Stephen A. O'Connell 34303
Eight Case Studies
WPS1266 An Efficient Frontier for International Sudhakar Satyanarayan March 1994 D. Gustafson
Portfolios with Commodity Assets Panos Varangis 33732
WPS1267 The Tax Base in Transition: The Case Zeljko Bogelic March 1994 F. Smith
of Bulgaria Arye L. Hillman 36072
WPS1268 The Reform of Mechanisms for Eliana La Ferrara March 1994 N. Artis
Foreign Exchange Allocation: Theory 3abriel Castillo 38010
and Lescons from SubSaharan John Nash
Africa
WPS1269 UnionNonunion Wage Differentials Alexis Panagides March 1994 I Conachy
in the Deveioping World: A Case Harry Anthony Patrinos 33669
Study of Mexico
WPS1270 How LandBased Targeting Affects Martin Ravallion March 1994 P. Cook
Rural Poverty Binayak Sen 33902
WPS1271 Measuring the Effect of External F. Desmond McCarthy March 1994 M. Divino
Shocks and the Policy Response to J. Peter Neary 33739
Them: Empirical Methodology Applied Giovanni Zanalda
to the Philippines
WPS1272 The Value of Superlund Cleanups: Shreekant Gupta March 1994 A. Maranon
Evidence from U.S. Environmental George Van Houtven 39074
Protection Agency Decisions Maureen L. Cropper
WPS1273 Desired Fertility and the Impact of Lant H. Pritchett March 1994 P. Cook
Population Policies Lawrence H. Summers 33902
WPS1274 The New Trade Theory and Its Asad Alam March 1994 A. Alam
Relevance for Developing Countries 87380
WPS1275 FemaleHeaded Households, Ricardo Barros March 1994 K. Binkley
Poverty, and the Welfare of Children Louise Fox 81143
in Urban Brazil
WPS1276 Is There Persistence in the Growth Ashoka MY' y March 1994 M. Patefia
of Manufactured Exports? Evidence Kamil Yilmaz 37947
from Newly Industrializing Countries
WPS1277 Private Trader Response to Market Steven Jaffee March 1994 C. Spooner
Liberalization in Tanzania's Cashew 32116
Nut Industry
Policy Research Working Paper Series
Contact
TitIS Author Date for psper
WPS1278 Regulation and Commitment In the Ahmed Galal March 1994 B. lMoore
Development of Telecommunications 38526
in Chile
WPS1279 Optimal Hedging Strategy Ravisited: Ying Oian March 1994 S. Lipscomb
Acknowledging the Existence of Ronald Duncan 33718
Nonstationary Economic Time Series
WPS1280 The Economic Impact of Export Wendy E. Takacs March 1994 M. Pateha
Controls: An Applbcatlon to Mongolian 37947
Cashmere and Romanian Wood Products
WPS1281 Human and Physical Infrastructure: Emmanuel Jimenez Aprl 1994 L. Longo
Public Investment and Pricing Policies 37786
in Developing Countries
WPS1282 Copper and the Negative Price of Donald Frederick Larson April 1994 A. Kim
Storage 33715
WPS1283 Interest Rates in Open Economies: Dipak Das Gupta April 1994 B. Kim
Real Interest Rate Parity, Exchange Betoy Das Gupta 82467
Rates, and Country Risk in Industrial
and Developing Countries
WPS1284 The Soviet Economic Decline: William Easterly April 1994 R. Martin
Historical and Republican Data Stanley Fischer 31320