In this Special Issue, we have reported the findings of a recent
large-scale survey of Russian industrial enterprises. (1) The project
looked at the factors determining and constraining enterprise
performance in the period around the 1998 financial crisis. When the
project was planned, in early 1998, there were the first signs of growth
in the Russian industrial economy since the start of transition, and it
was hoped that a survey could help to identify the characteristics of
firms, their markets and their business environment that would enable
enterprises effectively to exploit the improved conditions. In practice,
the project was blown off course by the 1998 crisis, and we decided to
wait until early 2000 to collect data. Our data therefore refer to the 3
years up to January 2000, with retrospective questions taking us back to
the point of privatisation up to 7 years earlier.
Despite the shock to trading conditions caused by the 1998 crisis,
the survey identifies encouraging as well as discouraging signals from
the Russian enterprise sector. According to Kuznetsov in Bevan et al
(2001), 75% of the firms in our sample were able to generate positive
profits in all our sample years, while some 50% managed to maintain
output, in nominal terms, over the period. However only 30% had invested
at a level in excess of the rate of depreciation. In terms of
restructuring, our measures are qualitative but still provide evidence
of considerable managerial effort, with the bulk of enterprises engaged
in 'deep' as well as 'defensive' restructuring.
However, as had been the case for most of the 1990s, on average real
output and productivity declined between 1997 and 2000.
There is a wide variation in performance between the sampled
enterprises in terms of productivity, financial performance and
restructuring; Russia clearly had successful as well as unsuccessful
firms in the industrial sector in this period. However, the empirical
results in this Special Issue do not support many of the theoretical
explanations from transition economics for this dispersion, which would
focus on ownership, competition and soft budget constraints (see eg
Roland, 2000; Djankov and Murrell, 2002). In particular, we find that
ownership and performance were not yet well correlated in Russia, and
there is no evidence of outsider-owned firms outperforming insider-owned
ones. These results are consistent with those of other studies of the
former Soviet Union (see eg Estrin and Wright, 1999), which usually
interpret such findings as being caused by capital market imperfections
and governance deficiencies (see Nellis, 2000). The evidence from our
sample is consistent with this view, in indicating only limited
correlation between ownership and perceived control over enterprise
decision-making. While insiders are perceived to have control over most
insider-owned firms, insiders are also believed by our respondents to
control nearly half of outsider-owned firms, and more than a third of
state-owned firms (see Bevan et al, 2002). This is perhaps a consequence
of the high levels of dispersion of outsider ownership in Russia (see
Earle and Estrin, 1997).
Our findings on competition are slightly more encouraging, in that
they identify a positive association between restructuring activity and
the competitiveness of the market environment. Domestic competition
spurs more deep restructuring, and to some extent more defensive
restructuring as well. Foreign competition is still an insignificant
factor in improving enterprise performance in Russia, although it plays
more of a role in stimulating investment. However, the positive effects
of competition in the regressions explaining qualitative indicators of
performance like restructuring are not significant in augmented
production or profit functions. Thus, we conclude that economic measures
of enterprise performance like profitability or productivity are not
correlated with either ownership or market structure in Russia in this
period.
On the financial side, our study indicates that company size is a
relevant factor in obtaining short but not long-term credit, while
equity financing is equally rare for all firm types in Russia. There is
little direct evidence of soft budget constraints, in the sense of
government subsidy, although financial disciplines are lax. Overall,
most Russian firms (almost 70%) face quite serious financial constraints
because of a combination of limited access to credit, poorly developed
capital markets and weak cash flow. Financial constraints appear to be
highly correlated with corporate performance and behaviour, and
investment is inversely related to the degree of financial constraint.
However, the causality is complex, and given the limited possibility of
recourse to external financing from any source, it may also run from
restructuring to financial status.
The relationship between enterprise performance and the various
exogenous variables specifying the factor and product market
environment, as well as the ownership structure, is likely to be
inter-related. For example, whether state-owned firms perform worse than
private ones will depend on both the measure of performance used
(profitability or total factor productivity, for example) and the market
structure. State-owned firms operating in highly competitive markets may
appear to perform relatively better than privately owned firms because
bankruptcy constraints bind for the latter but not the former.
Similarly, financial constraints may be more relevant for private than
state-owned firms, especially those operating in competitive markets.
The papers in this Special Issue have treated these issues in depth but
for the most part independently. In this concluding paper, we analyse
them together and explore their inter-relationships.
We therefore conclude our study by bringing together the various
determinants of enterprise performance in a single unified framework.
Our approach is to estimate an equation of the form:
(1) [P.sub.i] = [P.sub.i]([O.sub.i], [C.sub.i], [F.sub.i],
[X.sub.i])
where [P.sub.i] is an indicator of enterprise performance, for
example productivity, profitability or restructuring; [O.sub.i]
represents the form of majority ownership; [C.sub.i] the competitiveness
of markets; [F.sub.i] financial constraints and [X.sub.i] is a vector of
control variables.
The specification of equation (1) largely follows the methodology
employed in the other papers in this Special Issue. For performance, we
use a wide variety of quantitative measures--mark-up (sales over total
cost), return on equity (profit over equity capital), return on fixed
assets, sales per worker and the rate of investment (investment over
fixed assets). Accounting problems and data reliability suggest that we
should avoid concentrating on any single performance indicators. We also
use two of the qualitative measures of restructuring: the rate of
restructuring 'in any year', and 'in every year'
(see Estrin and Angelucci, 2003).
To control for alternative ownership structures, we divide the data
set into three categories of majority ownership--insider, outsider and
state-owned firms. Majority ownership is defined as the owner holding
more than a 40% stake, and holding the largest individual stake. In the
regressions, we omit the insider ownership dummy, and so we expect
[P.sub.i] to be improved by outsider ownership and probably reduced by
state ownership (see Earle and Estrin, 1997). The specification of the
competition variable parallels the methodology in Estrin and Angelucci
(2003) with three categories of domestic competition: high, medium and
low (ie monopoly), and a variable indicating the presence of significant
import competition. The low competition group is the excluded category
in the regressions, so we expect performance in terms of productivity
and restructuring to be reduced by monopoly power and import
competition. We hypothesise that profitability and measures of financial
performance will be improved, however, by monopoly power because of
higher product prices, but made worse by import competition. Our
measures of financial constraints follow the definitions in Bevan and
Fennema (2003), with the lowest level of financial constraint being the
omitted category. We would expect the performance of firms to
deteriorate as financial constraints become more binding. (2)
We use a number of control variables (X) in equation (1), most
importantly dummy variables for three-digit sector and for regions (see
the Appendix). These variables control for demand and institutional
differences by sector and region, which could affect revenue and
profitability. Since economic performance is also likely to be
influenced by the size of the firm, we include the three size
categories, omitting the smallest firms (see Bevan and Fennema, 2003).
We expect many measures of performance, including mark-up, productivity
and perhaps restructuring, to increase with firm size. The regressions
were run in cross-section form using 1999 data. (3)
We hypothesised above that different elements of the business
environment and the ownership structure may interact in their influence
on the firm. Several examples have already been noted: product market
competition may fail to elicit improved productivity in state-owned
firms, or financial constraints might only bind under private ownership.
We address these issues by estimating versions of equation (1) with
interactive effects between state ownership and competition, and between
state ownership and financial constraints. We do not hypothesise any
interactions effects between competition or financial constraints on the
one side, and insider versus outsider ownership on the other.
The results are reported in Table 1, which shows the basic
performance equations (without interactive effects) using the five
economic measures of performance. We do not report the interactive
regressions in Table 1 because these terms are never significant in any
equation, and F-tests supported their exclusions from the reported
regressions. The level of fit in these equations is, for the most part,
poor, ranging from around 7% to 11%. The only exception is the sales per
worker equation, in which the [R.sup.2] reaches about 0.46. However, all
the equations are significant at the 99% level.
Table 1 suggests that few economic variables explain the variance
in level of enterprise performance in 1999. Sales per worker are greater
in large firms, suggesting the influence of either scale economies or
monopoly power in price determination (although the effect is only
weakly significant, at the 90% level). Highly financially constrained
firms also have significantly lower sales per worker, though, as noted
above, the causality is not entirely clear. It is possible that
financial constraints are preventing firms from developing their
technology, product or sales in such a way as to increase productivity.
On the other hand, in an environment where external funding sources are
scarce and expensive, low productivity may derive from poor sales, which
may also be the cause of tight financial constraints. Either way, the
results suggest that firms can enter a downward spiral of poor
performance and financial distress, which constrains the ability of
managers to turn things around.
There are no other significant economic variables in the sales per
worker regression. The bulk of the rather good fit derives from industry
and regional dummies, which Estrin and Bhaumik (2002) argue can be
interpreted as demand and institutional factors. In particular, we find
no evidence that measures of competition or ownership structure have any
effect on sales per worker.
This pattern is repeated in most of the other regressions in Table
1, in particular for financial performance (ROE or RFA) and mark-up. As
in the regression for sales per worker, we find some weak size effects
in the mark-up equation, which supports the market power interpretation
that larger firms are charging a higher price. Financial constraints are
found to have a significant negative effect on both ROE and RFA; indeed,
both partial financial constraints and (strong) financial constraints
are separately significant, with the coefficient on the latter being
significantly greater than that on the former. This is consistent with
the idea of a recursive structure, from poor financial performance to
financial constraints to worse performance. Regional controls are
significant, as proxies for local demand and business environment
conditions (eg transport and heating costs, legal infrastructure,
enforcement, etc), and industrial dummies are relevant in explaining the
RFA, probably because of the sectoral variance in capital-labour ratios.
Once again, we fail to isolate any significant effect on financial
performance or mark-up from ownership or market structure.
Finally, we look directly at one measure of restructuring; the rate
of investment as a proportion of fixed assets (IFA). The results are
very similar to those of the previous equations. Most strikingly, we
find IFA to be significantly lower in firms with financial constraints,
especially these facing (strong) financial constraints. Given the
restricted ability to raise the finance externally, the causality in
this case seems likely to run unambiguously from the financial
constraints to investment. We also isolate a weak but significant
positive effect of import competition on investment rates; firms facing
more competition perhaps find that, whatever the financial constraints,
they do need to invest more. However, domestic market competition and
ownership remain insignificant in this equation.
We next turn to our qualitative indicators of enterprise
performance. The results on restructuring in Table 2 have slightly more
economic interpretation and parallel the findings in the Estrin and
Angelucci (2003) paper. Commencing with the basic specification, without
interactive terms, we find that only the regional and industry dummies
are significant in explaining restructuring in any year, although we
also identify size effects in the regression for restructuring in every
year. However, the columns of Table 2 which include, interactive terms
(columns 2 and 4) reveal that ownership and competition do influence
restructuring, but only when considered together. Thus, we find that
state ownership reduces the level of restructuring, when combined with
domestic monopoly power. Moreover, financial constraints actually
increase the pace of restructuring in state-owned firms, although not in
private ones. This suggests that the state as owners may want to
restructure enterprises which have serious financial constraints, in a
way that private owners are either unwilling or unable to do. Perhaps
this is with a mind to privatisation in the future. However, even for
the restructuring variables, we are unable to identify any significant
impact on performance from outsider as distinct from insider ownership,
or from import competition.
Taken together, these regressions largely confirm, in a unified
framework, the results suggested from the more detailed studies in the
other papers in the Special Issue. While there is considerable
dispersion in the performance of firms, measured by financial
indicators, productivity or restructuring (including investment), these
are not yet well explained by the factors stressed in the transition
literature--ownership structure, domestic market competition or import
competition. The underdevelopment of the capital market, however, does
mean that firms are heavily constrained by their financial situation in
their efforts to improve their performance. The major factors
determining all aspects of performance in Russian firms in this period
are found to be demand and local institutional conditions, proxied by
the regional and industry dummy variables. The strong significance of
the former, in particular, highlights that the Russian economy does not
yet function as a unified market.
Overall, our research therefore suggests three major policy
conclusions. Firstly, this study--in accordance with other
studies--demonstrates that Russian enterprises tended even in 2000 to
concentrate their activity on the local and regional markets. While our
data suggest that Russian firms do face some domestic competition, the
fact that this is not reflected in financial performance, and the
indications that monopoly power does enhance mark-ups, strongly intimate
that competitive market pressures are insufficient to motivate improved
performance. This is probably related to the regional fragmentation of
the Russia market. We do find evidence of some degree of domestic
competitive pressure, implying that the effect of barriers limiting the
ability of enterprises to enter other regional markets is less acute
than might have been expected. However, many enterprises--particularly
those in the food industry--maintain that they face difficulties in
entering regional markets. This may reflect two issues: regional
administrative barriers to entry and infrastructural deficiencies
following a long period of under-investment in transport infrastructure,
thereby impeding cross-regional competition. In the former case, the
Putin administration has made significant strides in recent years to
rebalance the federal-regional relationship, and the consequent
reduction of the regional power base may have helped reduce such
administrative barriers. In terms of infrastructure, the scale and
potential for regional fragmentation of the Russian market suggests that
the government should make efforts to alleviate transportation
bottlenecks, both through physical investment in infrastructure and
regulatory reform, in order to reduce transaction costs and encourage
domestic competitive pressure.
Secondly, our results suggest that competition from imported goods
is likely to play an important role in the future development of
competitive pressures and thereby improved financial performance in
Russian industry. Naturally, given the period under study, competitive
pressures were to some extent limited by the substantial devaluation of
the rouble following the August 1998 financial crisis. Consequently, the
subsequent real appreciation of the rouble will increase competitive
pressures from imported goods, particularly in the consumer goods
sectors. Enhancing international competitive pressures also requires the
reduction of import barriers and barriers to entry for foreign firms
particularly in view of strong tendencies of integration and
cross-ownership in Russian industry. Prospective Russian entry to the
WTO therefore has a potentially strong role to play in integrating
Russia into the world economy, and in providing a reform impetus akin to
that of the EU accession candidates. Indeed, it could prove advantageous
if the Russian authorities were to consider adopting certain elements of
the Acquis Communautaire including in areas such as licensing and
product labelling. This could serve both as an anchor for future reform
and improve the competitiveness of Russian goods in the enlarged EU
market.
Finally, our study highlights the importance of the development of
Russian financial markets and the associated importance of corporate
governance arrangements. Our results uncover strong evidence that
self-financing of investments is not yet a feasible choice for the
majority of Russian enterprises due to relatively small profit margins.
At the same time, firms are unable to obtain sufficient access to bank
finance owing to the underdevelopment of the banking sector. Hence,
while some firms, predominantly those in the natural resource sector and
hence excluded from our sample, have been able to rely on retained
earnings to finance their investment activities, insufficient financial
intermediation has precluded other firms from potentially profitable
investment. As harnessing long-term sustainable growth in Russia
increasingly requires investment both for restructuring purposes and for
capital accumulation to enhance the productive base of the economy, the
need for a robust system of financial intermediation becomes more
pronounced. Hence, the results of our analysis suggest that the
government and the Central Bank should continue with their efforts to
increase regulation and competition in the sector, enhance the capital
base and increase confidence in the financial system, which would likely
be aided by the proposed deposit insurance scheme currently under
consideration.
While equity and corporate bonds could become an alternative source
of finance in the future, the widespread development of securities
markets in Russia will require further reforms to ensure more
transparent and efficient corporate governance. The empirical evidence
from this paper shows that the average board composition of our sampled
enterprises neither reflected the ownership structure nor did it
correspond to standards from other countries. Instead employees,
especially management and sometimes regional authorities are
over-represented while other groups of stock and stake holders are
under-represented. This contradiction between the ownership and control
base of companies in our sample probably explains why we do not find
evidence of outsider owned firms performing significantly better than
insider-owned ones. While ownership arrangements have increasingly
shifted in favour of outsider ownership, improved governance structures
will be needed for this to be translated into improved company
performance. Many Russian firms are increasingly recognising the
importance of corporate governance both for securing access to
securities markets and for improving the terms of such access. Moreover,
the government's endorsement of the new Corporate Governance Code
in March 2002, which is based on OECD principles, is a significant step
forward in this regard. Nonetheless, adoption of the Corporate
Governance Code remains voluntary at present. Therefore, further
improvements and broader adherence to sound corporate governance
standards are required before Russian firms are able to harness such
additional sources of finance.
(1) The views in this paper are those of the authors and do not
necessarily reflect the official positions of their respective
institutions.
(2) We noted above that the casualty is not entirely
straightforward since in the Russian environment, where access to
capital markets is so limited, poor previous performance may be the
cause of currently binding financial constraints. Given the short time
period of the panel, we are unable to address this issue in our work.
(3) Although there are data for 3 years, the pattern of missing
values reduces the number of observations if we use the economic data
back to 1997. Moreover, some variables, for example competition, are
only measured for 1999. There are also serious problems of relative and
absolute price changes, which we are unable to address because of the
absence of appropriate disaggregated producer price indices. Hence, we
have chosen to report the cross-section results for 1999. Regressions in
rate of change form for the years 1997-1998 and 1998-1999 yield the same
pattern of results.
REFERENCES
Bevan, A, Estrin, S, Fennema, J, Kulzatsov, B, Schutler, ME,
Angelucci, M and Mangiarotti, G. 2002: The determinants of privatised
enterprise performance in Russia. CEPR discussion paper no, 3193, Centre
for Economic Policy Research: London.
Bevan, A and Fennema, J. 2003: Finance, restructuring and
performance in privatised russian companies. Comparative Economic
Studies 45: 117-147.
Djankov, S and Murrell, P. 2002: Enterprise restructuring: a
quantitative survey. Journal of Economic Literature 40: 739-792.
Earle, J and Estrin, S. 1997: After voucher privatization: The
structure of corporate ownership in russian manufacturing industry. CEPR
discussion paper no. 1736, Centre for Economic Policy Research: London.
Estrin, S and Angelucci, M. 2003: Ownership, competition and
enterprise performance, Comparative Economic Studies, this issue.
Estrin, S and Bhaumik, S. 2002: Why transition paths differ:
russian and chinese enterprise performance compared. CNEM discussion
paper no. 29, London Business School, Centre for New and Emerging
Markets: London.
Estrin, S and Wright, M. 1999: Corporate governance in the former
soviet union: an overview. Journal of Comparative Economics 27, 398-421.
Nellis, J. 2000: Privatization in transition economies: what
happened? What's next? World Bank: Mimeo, Washington, DC.
Roland, G. 2000: transition and economics: politics, markets and
firms. MIT Press: Washington, DC.
SAUL ESTRIN (1) & ALAN BEVAN (2)
(1) Centre for New and Emerging Markets, London Business School,
Regent's Park, London NW1 4SA, UK. E-mail: sestrin@london.edu;
(2) Office of the Chief Economist, European Bank for Reconstruction
and Development, One Exchange Square, London EC2A 23N, UK. E-mail:
bevana@ebrd.com
Table 1 Regressions of economic and financial performance
Mark-up Return on Return on fixed
equity (ROE) assets (RFA)
Medium firms 0.0404 * 0.0591 * 0.0710
(0.219) (0.0354) (0.0781)
Large firms 0.0526 ** 0.0365 0.0163
(0.0247) (0.0400) (0.0889)
Outsider owned 0.0074 -0.0485 0.1229
(0.0316) (0.0528) (0.1154)
State owned 0.0542 0.0437 -0.0159
(0.0357) (0.0573) (0.1267)
Medium domestic monopoly 0.0191 0.0262 0.0527
(0.0210) (0.0334) (0.0752)
Domestic monopoly -0.0146 -0.0152 -0.0661
(0.0313) (0.0502) (0.1115)
Import competition 0.0087 0.0434 0.0701
(0.0193) (00306) (0.0689)
Partial financial constraint -0.0244 -0.0955 *** -0.1882 **
(0.0220) (0.0354) (0.0784)
Financially constrained -0.0782 *** -0.1546 *** -0.3145 ***
(0.0228) (0.0367) (0.0820)
Constant 0.0870 ** 0.2469 *** 0.6165 ***
(0.0415) (0.0676) (0.1388)
Regional controls Yes *** Yes ** Yes
Industry controls Yes Yes Yes **
No. obs 292 277 286
Adj. [R.sup.2] 0.1132 0.1112 0.0667
F 2.38 *** 2.28 *** 1.75 **
Sales per Investment/fixed
worker assets (IFA)
Medium firms 2.5586 0.0109
(15.0750) (0.0188)
Large firms 33.1430 * 0.01723
(16.8841) (0.0213)
Outsider owned -11.4828 -0.0032
(21.5787) (0.0273)
State owned -10.0123 -0.0289
(24.6371) (0.0299)
Medium domestic competition -6.4635 0.0206
(14.4438) (0.0179)
Domestic monopoly -20.8590 0.0105
(21.4092) (0.0268)
Import competition 12.0472 0.0282 *
(13.0983) (0.0161)
Partial financial constraint -12.1087 -0.0328 *
(15.0558) (0.0186)
Financially constrained -52.2791 *** -0.0695 ***
(15.6447) (0.0197)
Constant 419.7697 *** 0.1492 ***
(28.6391) (0.0375)
Regional controls Yes *** Yes
Industry controls Yes *** Yes ***
No. obs 297 280
Adj. [R.sup.2] 0.4568 0.0867
F 10.22 *** 1.98 ***
*** Significant at 1% level,
** Significant at 5% level,
* Significant at 10% level.
Table 2 Regressions of restructuring activity
Restructuring (in any year)
No With
interaction interaction
Medium firms -0.2240 -0.0802
(0.6087) (0.6120)
Large firms 0.8096 0.8118
(0.6764) (0.6774)
Outsider owned -0.8256 -0.9472
(0.8557) (0.8542)
State owned -0.7144 -3.5390
(1.0013) (2.2769)
Medium domestic competition 0.3267 0.3247
(0.5792) (0.6002)
Domestic monopoly -0.2820 0.1284
(0.8401) (0.8594)
Import competition -0.2672 -0.2453
(0.5276) (0.5423)
Partial financial constraint 0.9307 0.7028
(0.6035) (0.6152)
Financially constrained 0.0292 -0.1414
(0.6334) (0.6427)
Constant 7.8061 *** 7.8368 ***
(1.1579) (1.1660)
Regional controls Yes *** Yes ***
Industry controls Yes * Yes *
Interaction of
State ownership & medium -- -1.1334
domestic competition
(2.1665)
State ownership & domestic -- -8.0677 **
monopoly
(3.5259)
State ownership & import -- -0.5973
competition
(2.0641)
State ownership & partial -- 6.1555 **
financial constraint
(2.7224)
State ownership & -- 4.8032 *
financially constrained
(2.9004)
No. obs 303 303
[R.sup.2] 0.1235 0.1349
F 2.58 *** 2.47 ***
Restructuring (in every year)
No With
interaction interaction
Medium firms -0.0209 0.0761
(0.4953) (0.4956)
Large firms 1.4738 *** 1.3827 **
(0.5504) (0.5487)
Outsider owned -1.0159 -1.0713
(0.6963) (0.6918)
State owned -0.1155 -3.2525 *
(0.8147) (1.8442)
Medium domestic competition 0.2763 0.3483
(0.4713) (0.4861)
Domestic monopoly -0.9523 -0.7099
(0.6835) (0.6960)
Import competition -0.2550 -0.3550
(0.4293) (0.4392)
Partial financial constraint 0.2805 0.1145
(0.4911) (0.4983)
Financially constrained 0.2328 -0.0767
(0.5154) (0.5206)
Constant 5.6257 *** 5.8184 ***
(0.9422) (0.9444)
Regional controls Yes *** Yes ***
Industry controls Yes Yes *
Interaction of
State ownership & medium -- -2.0778
domestic competition
(1.7548)
State ownership & domestic -- -4.4071
monopoly
(2.8558)
State ownership & import -- 0.3788
competition
(1.6718)
State ownership & partial -- 4.5852 **
financial constraint
(2.2050)
State ownership & -- 6.7568 ***
financially constrained
(2.3492)
No. obs 303 303
[R.sup.2] 0.1031 0.1227
F 2.29 *** 2.32 **
*** Significant at 1% level,
** Significant at 5% level,
* Significant at 10% level.