Gregory: Russian Structural Change 5/28/2004
Structural Change in Russian Transition
Paul R. Gregory
Professor of Economics
University of Houston
May 25, 2004 Final Version Gregory: Russian Structural Change 5/28/2004
INTRODUCTION
Socialist economies such as the Soviet Union practiced centralized distribution of resources according to what became known as “planner’s preferences”.1 Planner’s preferences were essentially the preferences of the dictatorial government rather than of consumers. The distribution of resources was also affected by special interests such as the military-industrial complex or the heavy industry lobby. In addition, rigidity of material balance planning or “planning from the achieved level,” produced additional deviations from market-like resource allocations. Finally, relative autarky of the socialist economies (primarily limited to intra- COMECON trade) prevented them from realizing their comparative advantages and created additional biases. Consequently, the patterns of resource allocation and structural proportions (as observed in the structure of GDP, consumer budgets, foreign trade, and so on) that emerged in the Soviet period differed significantly from those of market economies at similar levels of development. These structural distortions produced inefficiencies that were in part responsible for inadequate levels and distorted structures of consumption in Eastern Europe and former Soviet Union and for the stagnation of these economies in the 1970s and 1980s.2 Transitional economies, including Russia, started their transformation with initial conditions inherited from the pre-existing socialist economies. It is therefore natural to expect that these economies, after the removal of the constraints of planning, should move in the direction of more “normal” proportions of market economies in the process. International openness should also lead to greater role of foreign trade and to deeper international specialization. Indeed studies show that transition “success” varied inversely with the proximity to and duration of the Soviet core model.3 The goal of this report is twofold. First, we study structural change in the Russian economy to compare its starting points with its current makeup. Second, we seek to elicit the structural differences between Russia and various country groups at the start of transition (around 1990), trace the pattern of structural change, and to capture the extent to which Russian structural
1 The term was introduced by Abram Bergson, See, for example: Abram Bergson, The Economics of Soviet Planning, Yale University Press, 1964. 2 Padma Desai, The Soviet Economy: Problems and Prospects, Blackwell, 1987; Paul Gregory and Robert Stuart, Russian and Soviet Economic Performance and Structure, 7th ed., Addison Wesley, 2001; Steven Rosefielde (ed.), Efficiency and Russia’s Economic Recovery Potential, Ashgate, 1998. 3 Stuart, Robert and Christina Panayotopouolos (1999). “Decline and Recovery in Transition Economies: The Impact of Initial Conditions,” Post-Soviet Geography and Economics, vol. 40, no. 4.
Gregory: Russian Structural Change 5/28/2004 distortions have been reduced since 1991. This exercise accomplishes a number of goals: First, it provides a measure of transition “success” which is grounded in quantitative rather than subjective indicators, such as those of the EBRD. Second, it points the way to future sectoral growth patterns under the assumption that remaining structural distortions will continue to be removed. Third, it points to the type of market economy towards which Russia is moving, given the great uncertainty as to the Russian economy’s “final” shape and form. We know from previous studies that Russia’s (the USSR’s) deviations from “normal” structures of market economies were substantial, such as the greater shares of heavy industry, the low shares of services, the high shares of food, consumption, and the underutilization of foreign trade.4 We also know from previous studies that the pace of change, both structural and institutional, has been rapid since 1990 in Russia.5 What is lacking is a major study of Russian structural change in the comparative context of other countries.6 Our general conclusions are: 1) The changes observed in Russia’s value added and labor force structures appear to be market driven in accordance with productivity differentials. 2) Russia is becoming increasingly similar in terms of sector-of-origin structure to upper-middle income countries and to the lower tier of high-income countries such as Greece, Portugal, and South Korea. This conclusion is remarkable since per capita GDP measures place Russia well below these groups.
4 Ofer, Gur (1973). The Service Sector in Soviet Economic Growth (Cambridge, MA: Harvard University Press); Gregory, Paul (1970), Socialist and NonSocialist Industrialization Patterns (New York: Praeger); Kuznets, Simon (1963). “A Comparative Appraisal,” in Abram Bergson and Simon Kuznets (eds.), Economic Trends in the Soviet Union, (Cambridge, MA: Harvard University Press); Schroeder, Gertrude and Imogene Edwards (1981), Consumption in the USSR: An International Comparison, Joint Economic Committee, U.S. Government Printing Office, Washington, D.C. 5 Schroeder, Gertrude (1998), “Dimensions of Russia’s Industrial Transformation, 1992-1998: An Overview,” Post- Soviet Geography and Economics (39, no. 5 (May), 243-271; Shinichiro Tabata (1996), “Changes in the Structure and Distribution of Russian GDP in the 1990s,” Post-Soviet Geography and Economics, 37, no. 3, 129-144; Paul Gregory and Robert Stuart (1998), Russian and Soviet Economic Structure and Performance, 6th ed. Reading, Mass.: Addison Wesley Longman, chaps. 16-18. See also, World Bank and State Statistics Committee of the Government of the Russian Federation (October 1995), Russian Federation: Report on the National Accounts.
6 The World Bank, From Transition to Development, April 2004, has conducted a “benchmarking” study of Russia’s changing structure of value added, broken down into four major sectors. Gregory: Russian Structural Change 5/28/2004
3) The Russian labor force has been much slower to adjust than output, especially in agriculture. The main labor force adjustments have taken place in terms of movements from manufacturing to market services. The movement out of agriculture was limited to the first few years of transition. There are preliminary signs of renewed movement out of agriculture. 4) The structure of manufacturing still feels the imprint of initial Soviet conditions. Unlike other parts of the economy, Russian manufacturing resembles the structure of high income countries, a fact that works against successful restructuring. The sector of manufacturing that suffered an absolute collapse when confronted with an open economy and market prices was light industry. 5) Russian GDP by end use has adjusted to begin to remove the major distortions of the Soviet period – too high investment shares and too low consumption shares. However, there is a remarkable lack of convergence in the major end-use shares, which we attribute to over adjustments, such as investment shares going from too high to too low. We interpret the extraordinary current account surpluses as a partial attempt by Russian citizens to cushion the decline in real incomes and to guard against uncertainty. 6) International comparison data show that the structure of Russian consumption remains highly distorted due to the lack of price reform in housing, energy, and in health and education. 7) Russia’s participation in trade has increased dramatically, but it remains similar to a low-middle income in its participation in trade. It will retain its “backward” structure of exports (reliance on raw materials) because of its comparative advantages. Russia has also not emulated lower income countries by producing “low tech” manufacturing, which explains its relative lack of manufacturing exports. 8) Long term data show that Russia began to undergo rapid structural change in the late perestroika period after a lengthy period of planning from the achieved level. The mini- restructuring of late perestroika, however, appeared to move the economy away from a market economy. 9) We conclude that Russia’s own estimates of the structure of GDP are more reliable than alternate estimates, such as the substantial adjustments proposed by the World Bank. This report is organized as follows: Part I examines changes in the structure of the Russian economy between 1990 and the present. It examines the changing structure of GDP and labor force by sector of origin and by end use as well as changes in the structure of trade. Part II examines these structural changes in international perspective by comparing Russia with international data of countries grouped by level of economic development. Gregory: Russian Structural Change 5/28/2004
PART I: RUSSIAN DATA AND TRENDS Overview We draw data from the period 1989/1990 to 2001. It should be remembered that the Soviet Union was disbanded in December of 1991 and Gaidar’s liberalization of prices began in January of 1992. Therefore, the years 1989-1991 are used to establish a base point from which transition began. These pre-transition benchmarks are not ideal because of the immense disruptions and upheavals of the Soviet economy in the period 1988 to 1991. The second period that the data covers is the period of transition recession (or depression) from 1992 to 1997, which was also accompanied by near hyperinflation between 1992 and 1994. The third period is the period of growth resumption which followed the ruble crisis of August 1998. Thus, our data set captures at best four years of economic growth, which means that we lack a sufficiently long record of growth to judge its long-term pattern. Data The Russian data are from Gosskomstat (GKS) publications. Since the early 1990s, Russia’s GKS has been replacing the Soviet Material Product System with the international System of National Accounts (SNA). In making this conversion, GKS has had to change over from an accounting system that focused on material production and gross outputs to one that counted both material products and services not directly associated with production and value added rather than gross output. The transition to a market economy created problems with respect to measurement of depreciation (how rapidly to write off obsolete Soviet-era equipment), profits (which now vary often dramatically by industry rather than being percentage add-ons to prices), and activities (such as business services, small business, and real estate) that did not even exist in Soviet times. Complicating matters was the hyperinflation of the early 1990s, which made value calculations tricky with prices that were changing significantly by month and even by week. The switchover from a system of automatic subsidies for unprofitable businesses, profits transfers among enterprises, and from turnover taxes to income, value added, excise, and export/import taxes also created significant differences between value added in producers’ prices (now called “basic prices” by GKS) and prices paid by buyers that include taxes of various sorts, minus subsidies, and including trade and transport margins.7
7 For methodological explanations, see Goskomstat Rossii (1995), Natsional’nye scheta Rossii v 1989-1994 gg. Moscow: Gosskmonstat, 5-15; Goskomstat Rossii, Natsional/nye scheta Rossii v 1995-2002 godakh (2003). Moscow: Goskomstat, 11-18. Gregory: Russian Structural Change 5/28/2004
In the turbulent early years of transformation, GKS produced some contradictory figures, added significant imputations for uncounted wages and for small and underground businesses.8 It is difficult to find consistent figures in current prices for the period 1990 to 1995, but the data for the period 1995 to present appears to be calculated on a consistent basis. Those relying exclusively on official GKS publications are hampered by the paucity of GKS data on the breakdown of industry, broadly defined to include mining and electricity. Typically, the official national accounts report only one figure for industry, although they separately report breakdowns of manufacturing. Thus researchers must compile their own breakdowns of industry, usually using the more detailed annual input-output tables of GKS.9 The Russian data used in this study are described in Appendix A and Appendix B, which shows data sources, adjustments, and reclassifications. Value shares can change because of changes in relative prices and because real outputs are changing due to the relative movement of labor and capital resources. We have chosen to use primarily current price data in this study, which combines price and real resource movements, because economic decisions are based upon current prices, taxes, and profits, as well as those expected in the future.
Value Added and Labor Force Shares We record in Table I-1 (end of paper) the value added and labor force data organized into 23 sectors compiled from GKS sources. The value added data are in current (“basic”) prices (which adjust for indirect taxes, subsidies, and transportation and trade margins). These data are combined into eight sectors for value added, labor force, and relative productivity in Table I-2 (end of paper). We summarize in Figure I-1 results for the eight sectors, which break industry into “fuels” and the rest of industry (primarily manufacturing). The six other sectors include agriculture and forestry, trade, transportation and communication, construction, transportation, trade, business services and “other” which includes public administration, health, science,
8 GKS began including “unreported wages and salaries” starting in 1993 and GKS raised the value of the capital stock much slower than inflation after 1991. Both practices affect the changes in VA shares in the early 1990s, but they seem like reasonable adjustments. Otherwise, we did not find significant methodological changes.
9 Natsional’nye scheta (National Accounts), and Sistema Tablits (System of Input-output Tables available in electronic form from www. Eastview.com, except for Promyshlennost. 1996 and Natsionalye scheta Rossii v 1989- 1994 gg (GKS 1995 – Chapter on input-output tables) which were used in hard copy.
Gregory: Russian Structural Change 5/28/2004 education, and welfare, which we denote as “PA-HES.” PA-HEAS captures those services, including activities of the bureaucracy/administrative apparatus, traditionally offered by the state. Hence it is a rough measure of the size and scope of the state sector. It should be noted that it is extremely important that the sector definitions be identical for labor force and for output; otherwise, there can be large errors in the estimation of relative productivity (the value added share divided by the labor force share). Presumably errors would be greatest for narrowly defined sectors and least for broadly defined sectors.
Figure I-1A: Shares of Value Added, Russia
40.00%
35.00%
30.00% industry excluding electricity and fuels
electricity and fuels (includiong geology, metereology) 25.00% agriculture
construction 20.00% transport and communication
trade 15.00% business, financial, information services
PA-HEAS 10.00%
5.00%
0.00% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Figures I-1A and B show that the changes in value added shares exceed the relative changes in labor force shares. This is an expected result in that labor force mobility, both geographically and sectorally, is subject to rigidities, which would be prominent in the Russian transition case. Both manufacturing and agriculture experienced significant declines in value added shares (manufacturing’s was reversed briefly as an apparent consequence of the ruble devaluation in 1998), but agriculture’s share has remained relatively stable after an initial plunge since the mid 1990s. The generally rising (albeit not monotonically) sectors are trade, business services, and fuels. PA-HEAS experienced a substantial drop in the first two years (perhaps a consequence of the collapsing state budget), then increased thereafter, suggesting that the share of the state in economic activities has actually been rising. Construction and transportation maintained relatively stable shares. Gregory: Russian Structural Change 5/28/2004
Figure I-2 shows that the value added share of fuels has been generally linked to world energy prices (as proxied by the world price of crude oil), after an initial period where most fuels were sold domestically at artificially-low domestic prices.
Figure I-1B: Shares of Labor Force, Russia
35.00%
30.00%
industry excluding electricity and fuels 25.00% electricity and fuels (incl geology, meterology) agriculture 20.00% construction
transport and communication 15.00% trade
business, financial, information services 10.00% PA-HEAS
5.00%
0.00% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Compar ed to value added shares, there were lesser changes in labor force shares. The employment shares of manufacturing and construction generally fell but less than their output shares, while agriculture’s and transportation’s labor shares remained remarkably stable. PA-HEAS’s employment shares also rose, but the most dramatic increase in employment shares was in trade, suggesting a “real” relative movement of resources into trading activities. Gregory: Russian Structural Change 5/28/2004
Figure I-1C: Relative Labor Productivity, Russia
600.0%
500.0%
industry excluding electricity and fuels
electricity and fuels (including geology, 400.0% meterology) agriculture
construction 300.0% transport and communication
trade
200.0% business, financial, information services
PA-HEAS
100.0%
0.0% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 The relative labor productivity figures (Figure 1c) divide sectors into those with higher or lower than average labor productivity (where the economy average equals 100). Those sectors experiencing upsurges in relative productivity – business (commercial, information, financial) services, trade, and fuels made their major gains between 1990 and 1992, after which they either maintained their position (fuels), converged towards the average (trade), or declined to experience another surge after devaluation in 1998 (business services). Business services, which had been absent from the planned economy, proved extremely productive when the transition began, but then declined with the level of real economic activity, only to surge again with the pickup in real economic activity. Notably manufacturing remained near the average throughout the period as did construction; while agriculture and PA-HEAS remained stuck below the average. Gregory: Russian Structural Change 5/28/2004
Figure I-2: Scatter Diagram: Fuel Shares vs. World Crude Oil Price
35
30
25 ) $
( 20 ice r p y g r
e 15 en
10
5
0 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% value added share
The labor force share of public administration, health, education, and welfare rose from 28 to 32 percent between 1990 and 2001, contrary to the popular image of collapsing state revenues. This rising share was in economic activities whose relative productivity fell from 62 percent to 44 percent of the economy average. Figure II-3 confirms that the share of government administration employees – the bureaucracy – in fact was the growth employment industry of the second half of the 1990s. The number of bureaucrats, according to some estimates, in Russia today is greater than in the Soviet period and their productivity is lower than in the Soviet period.10
Figure II-3: Government Employment as Share of Total
10 E. Larina, The Russian Journal, Dec, 7-13, 2001. Gregory: Russian Structural Change 5/28/2004
5%
4%
3%
2%
1%
0% 1970 1975 1980 1985 1990 1995 2000 2005
The Structure of Manufacturing Table I-3 (end of paper) summarizes value added, labor force, and relative productivity for manufacturing sub sectors. Figures I-4A, B, and C capture the dramatic change in value added shares as the Russian economy began the liberalization of prices. The shares of machinery and light industry plunged in the first two to three years, after which they remained relatively stable. Conversely, the value added shares of fuels and electricity soared in the first few years of transition, after which the energy share depended on the world price of oil (as noted above), and the electricity share dropped after 1998. The shares of metals remained relatively stable until the 1998 devaluation. As in the case of the major economic sectors, the manufacturing labor force shares moved less than the value added shares. The declines in machinery’s and light industry’s labor force shares were most prominent among declining employment shares. In the Soviet period, light industry and food products were neglected as low-priority branches. Presumably, their shares of manufacturing would be expected to increase in the course of transition. The experiences of light industry and food were notably different. Between 1991 and 1992, light industry’s share of value added collapsed from 19 percent to 2 percent, a low level from which its share failed to recover. Its loss of employment share was also substantial but less dramatic resulting in a collapse of relative labor productivity in light industry. Light industry dropped from being 77 percent above the economy-wide productivity average to 20 percent. On the other hand, food manufacturing more-or less held its own with a rising share of value added and of relative productivity starting in the mid 1990s as Russian food manufacturers began to compete more effectively with foreign imports. The collapse of light industry’s value added and labor force shares suggests an industry coddled by relatively high prices and shielded from foreign competition that collapsed when price were freed and the economy was opened.
Gregory: Russian Structural Change 5/28/2004
Figure I-4A: Manufacturing Value Added Shares, Russia
35.0%
30.0%
25.0% electricity fuels ferrous 20.0% nonferrous chemicals machinery 15.0% forestry&wood processing construction mats light food 10.0%
5.0%
0.0% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Figure I-4B: Manufacturing Labor Force Shares, Russia
50.00%
45.00%
40.00%
35.00% electricity fuels 30.00% ferrous nonferrous chemicals 25.00% machinery forestry&wood processing 20.00% construction mats light food 15.00%
10.00%
5.00%
0.00% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
Trends in relative productivity within manufacturing mirror those of output share changes given the relatively slow movement of relative labor force changes. Those sectors above the 100% line represent manufacturing sectors with higher than average productivity. Of those that Gregory: Russian Structural Change 5/28/2004 were significantly above average in 1990, fuels and electricity’s relative productivities soared during the early years of transition, with electricity eventually returning approximately to its initial position. Non-ferrous metals basically maintained its relative position until 1998 after which it rose. Light industry fell from well-above-average productivity to the lowest productivity manufacturing sector. Other sectors generally maintained their initial positions throughout the transition near or below the economy average.
Figure I-4C: Manufacturing Relative Productivity, Russia
600.0%
500.0%
electricity 400.0% fuels ferrous nonferrous chemicals 300.0% machinery forestry&wood processing construction mats light 200.0% food
100.0%
0.0% 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
It appears that the Soviet legacy of extreme priority for metallurgy, machinery and investment goods was not fully misplaced for an open, market economy. The emphasis on fuels has generally served the Russian transition economy well, although the antiquated infrastructure and lack of technology make it relatively backward as compared to its industrialized country counterparts. Russia’s international competitiveness in metallurgy products appears to depend upon the exchange rate (given its recovery after the 1998 devaluation). With a low ruble exchange rate, Russian metals appear to be competitive and even raise pressure in the United States for protection. The main burden of the Soviet past has been felt in machinery, whose relative productivity fell from 70 percent to a low of 50 percent of the economy average before staging something of a recovery to 55 percent. Indeed the major adjustments of both value added and labor force have been in the machinery sector. It could be that these resource re-allocations have weeded out the lowest productivity machinery producers so that the increase in relative productivity may have fundamental causes, rather than simply a result of ruble devaluation. From a policy standpoint, it is clear that labor did not move in adequate numbers. One reason is the extreme geographical concentration of industry in Russia. For example, the largest share of textile enterprises is located in the Central Industrial Region, where textile employment Gregory: Russian Structural Change 5/28/2004 accounts for the absolute majority of labor force (as in Ivanovo oblast). Resource-extracting industries are located primarily in remote Eastern areas of the country. The costs of moving from the former to the latter may be prohibitive for many workers, especially in the conditions of transitional recession. The Russian government, instead of stimulating retraining and relocation of labor from labor-excessive sectors/areas, has been protecting existing jobs by subsidizing failing enterprises. As Figure I.1B shows, “light industry” (textiles, apparel, etc.), despite the total collapse of output, lost about one half of its workers. Similar trends of a lesser magnitude are characteristic of other consumer goods sectors except food processing. At the same time, tripling of the share of “fuels” sector was not nearly matched by modest increase in the share of employment.11 The differences in the rate of reallocation of output and employment in industry can be summarized using an index of structural change, defined as:
2 D = Σi (S90i – SYi) ,
th where S90i is the share of i sector in the Russian economy in 1990 and SYi is the share of this sector in the current year, Y. The formula is applied to value added and employment data separately to compare their respective rates of adjustment. Increasing values of the index of structural change mean that the current structure is farther away from the initial conditions, all sectors considered (See figure below). The much weaker climb of the employment index show the much more limited restructuring of the labor force than of output.
11 Practically all the increases in value added and employment are the contribution of oil and gas sectors, while larger coal mining sector remains stagnant. Gregory: Russian Structural Change 5/28/2004
0.14
0.12
0.1
0.08 Value Added 0.06 Labor
0.04
0.02
0 1990 1992 1994 1996 1998 2000 2002
Figure II.1.14. Indices of structural change in Russian industry. Source: Goskomstat. Calculations by the authors.
An Evaluation of Sector-Of-Origin Structural Changes Figures I-1 through 4 paint a picture of an economy that has been subjected to three shocks: price deregulation, the opening of the economy to trade and hence to world prices, and the ruble devaluation of 1998. The first two shocks meant that branches, whose outputs had been highly valued under planning, would now receive relatively low valuations in a market setting and vice versa. The most notable case of an industry that virtually collapsed when confronted with the first two shocks was light industry manufacturing. Moreover, the transition to a market economy created a demand for activities not present in the planned economy, the supply of which could not be created over night -- examples being trade and business, finance, and information services. Those factors with “new” skills would be among the most valued and productive resources in the transition economy. As might be expected, the “rents” from scarcity appeared to decline over the years, such as in trade whose relative productivity declined from almost four times the economy’s average in 1992 to only 75 percent above the average in 2001. Whereas the shifts in output shares took place quickly (within two to three years), the shifts in labor force shares were slow to materialize given the initial lack of organized labor markets, barriers to movement, and the continuation of patronistic behavior by large enterprises. Nevertheless, the labor force did generally redistribute itself, albeit slowly, according to economic principles, namely according to relative productivity. The charts also show the large, albeit transitory, Gregory: Russian Structural Change 5/28/2004 impact of the exchange rate on economic activity. The ruble devaluation of 1998 caused a reversal of the decline in manufacturing (particularly of metallurgy) as Russia’s industrial products became more competitive in world markets. Sectors of industry that were more productive on average (in terms of value added per worker) in 1990 remained so a decade later. The exception is “light industry” which fell from a higher than average position to the lowest one. It can be concluded therefore that, despite the significant restructuring of industry, efficiency losses resulting from misallocation of resources in the previous-period Soviet economy were not eliminated but probably were made even higher due to governmental policies that failed to make labor markets more flexible. In looking to the future, the charts suggest that labor will continue to move into electricity and fuels, to business services, trade, and transport and communications. The relative productivity of fuels will remain high and will fluctuate according to world energy prices. The attractiveness of business, finance, and information services will depend on real economic growth; while that of metals will hinge on exchange rates unless there are real technological improvements which lower the domestic costs of doing business. The World Bank Critique The tables and figures cited above are based upon official GKS statistics, whose validity has been disputed by World Bank studies.12 Two related World Bank studies argue that GKS’s distribution of value added between the oil and gas sector (which accounts for the bulk of the electricity and fuel sector above) grossly overstates the trade share and grossly understates the oil and gas value added share. The World Bank studies do not alter the distribution of labor force shares; hence, they suggest very substantial upward revisions of the relative productivity of oil and gas and substantial downward revisions of trade’s relative productivity. The crux of the World Bank argument is that Russian oil and gas firms use low transfer prices (either to conceal their true income or, more likely, to reduce their tax burdens) and thereby shift value added to trade and to transportation. By selling energy at transfer prices that are “too low” value added that truly belongs to oil and gas is shifted to trade in the form of inflated profits, which are the primary component of trade’s value added. Empirical support for this conclusion is found in (what the World Bank considers) to be inflated trade and transport margins relative to other energy producers. The World Bank compares transport and trade margins in Russia with those of the UK and the Netherlands (where transport and trade margins
12 World Bank, From Transition to Development: A Country Economic Memorandum for the Russian Federation, April 2004 (draft), 60-73; World Bank, Russian Economic Report. February 2004, www.worldbank.org.ru Gregory: Russian Structural Change 5/28/2004 are calculated together) and with trade margins in Canada (where trade margins are given separately). The key trade margins (for Canada and Russia) for the year 2000 are recorded in Table I- 4:
Table I-4: Trade Margins, Oil and Gas Sectors, Russian and Canada, 2000 Russia Canada Oil extraction 30.7% 0.0% Oil refining 36.6% 17.2% Gas 63.1% 0.0% Source: World Bank, From Transition to Development, p. 62.
The discrepancy between Russian oil and gas trade and transport margins are even larger as compared to those of the UK and Netherlands, but the World Bank recognizes that their trade and transport conditions are quite different from Russia and that Canada serves as a better benchmark. Another notable proof of Russia’s exaggerated oil and gas margins is said to be that margins for other branches of the economy are similar to those of Canada, UK, or the Netherlands. Table I-4 does raise the interesting puzzle of zero margins for oil extraction and natural gas, products that are traded across borders and transported long distances. It would be advised to examine definitions to determine how the zero rate is arrived at. When World Bank specialists shifted the “excess” of the trade (or trade and transport margins) from profits in trade to profits in oil and gas, the result is a substantial change in the distribution of value added in favor of oil and gas and away from trade, as is shown by Table I-5. This adjustment results in changes in other sector shares, but such changes are modest and are ignored in the following discussion.
Table I-5: Goskomstat and World Bank Estimates of Value Added Shares, 2000 Goskomstat Canadian UK margins Netherlands margins margins Oil and gas 7.8 19.2 25.2 24.9 Trade 27.3 14.6 11.0 9.4 Source: World Bank, From Transition to Development, 63, Russian Economic report, February 2004, p. 15.
Figure I-5 shows the effects of the World Bank value added adjustments on the official estimates of relative labor productivity by sector. Insofar as the World Bank adjustments affect Gregory: Russian Structural Change 5/28/2004 primarily trade and fuels, the relative productivities of the other branches remain relatively unchanged. Manufacturing, transport and construction remain near the economy average, agriculture and PA-HES remain well below the economy average.
Figure I-5: Sector Productivity Comparisons: World Bank vs. Goskomstat Estimates
9
8
7
6
5 Productivity-WB Productivity-GKS 4
3
2
1
0 mfg fuels agriculture construction transport trade busines, PA-HES Services finance, Combined information
World Bank specialists have used these figures to argue that Russia’s industrial sector is still more productive than its service sector and implicitly that resources should flow to “industry” rather than to “services.” The benchmark year 2000 may not be representative given that the positive effects of devaluation on manufacturing and metals were still being felt. However, the validity of the argument does not really rest on the choice of benchmark year. Figure I-5 shows that the relatively low productivity of services (under the World Bank variant) is the consequence of exceptionally low productivity in public administration, health, education and science, which is compensated in the combined service figure by the very high productivity of market service sectors. As a guide to resources allocation, Figure I-5 suggests that resources should flow into mining and market services and perhaps transportation infrastructure and not into manufacturing. Moreover, in what is largely a market economy, resources will likely flow spontaneously to those sectors with higher returns anyway. Moreover, remaining Russian subsidies are largely for manufacturing anyway; so public policy already favors manufacturing over services. The gradual decline of manufacturing employment provides a rather clear signal that its productivity is low relative to other pursuits. The basic assumption of World Bank analysts is that the high oil and gas transport and trade margins are pure tax avoidance (or worse) schemes without economic foundations. In the Gregory: Russian Structural Change 5/28/2004
Russian context, however, one must ask, to the contrary, whether high trade and transport margins are the consequence of natural economic processes. A number of sources suggest that it is difficult to distinguish between value added attributed to trade and to transport. Those marketing and transporting the product can attribute the revenue either to trading activities or to transport activities. Therefore it makes sense to combine trade and transport margins in the discussion that follows.
Figure I-6: Payments for Trade and transport for oil and gas exports and for total product, 2000 (bil rubles)
3000
2500
2000
margins for exports 1500 total margins
1000
500
0 oil production oil refined products gas oil and gas total economy total
Figure I-6 shows the combined Russian trade and transport payments for oil and gas products and for the entire economy in 2000 broken down into payments for exported products and for the product as a whole (domestic uses plus exports). It shows that, except for refined oil products (gasoline, jet fuel, heating oil), which are used primarily domestically, more than half of the trade and transport fees are paid on oil and gas products that were exported. Does this make any economic sense, or does it suggest simply a device to avoid taxes? The answer depends, to a degree, on where scarcity lies in an economy. Is the scarce resource oil and gas as a natural resources or is it marketing knowledge and/or the transportation infrastructure? In Russia, the natural gas monopoly (Gazprom) both produces gas and owns the transportation system, but its export pipelines go through two sovereign nations before reaching their ultimate export markets. In the oil sector, most export pipeline capacity is controlled by a state-controlled entity (Transneft), which is in a position to extract rents in the form of marketing or transportation fees from oil producers. In Russia, where domestic prices of energy are well below world prices, those who are in a position to extract rents between the well head and the Gregory: Russian Structural Change 5/28/2004 final distribution point are those who control access to export markets, through their control of export licenses, access to pipelines, ability to gain official permissions and so on. Assuming that domestic energy prices roughly equal the cost of production plus a normal profit, the economic rents in the energy market (per physical unit of product) roughly equal the difference between the world and domestic prices, which, in the case of natural gas, is a ratio of four to one. The margins in Figure I-6 may simply reflect the relative scarcities of modern Russia in which wellhead oil or natural gas is not a particularly valuable resource compared to other resources required to get it to market. One way to judge whether the pattern of trade and transport margins makes economic sense is to examine trends over time. Figure I-7 compares the 1991 trade and transport margin (combined) 13 with that of 2000 and reveals that the 2000 margin for oil and gas combined was only slightly above that of 1991. Given that the year 1991 counts as a pre-transition year, we would expect tax-avoidance and other value-added-redistributing activities to differ from those of 2000; therefore, it is remarkable that margins were about the same in the two years. Figure I-7 also shows that margins remained relatively stable although payments for export margins increased dramatically as a percent of total margin payments, which itself is explained by the fact that exports as a percent of total energy expanded substantially as well. 14
13 Goskomstat, Natsional’nye scheta Rossii v 1989-1994 gg (Moscow, 1995), 14 Note that the above figures measure export and production in value terms. We do not know how these calculations are made in light of the large differences in domestic and world prices. Gregory: Russian Structural Change 5/28/2004
Figure I-7: Export margin payments, total margin payments and reported trade and transport margins, 1991 and 2000
0.7
0.6
0.5
0.4 1991 2000 0.3
0.2
0.1
0 export margins as percent of total exports as a percent of total GKS reported margins
It would require a detailed study to determine how trade and transport margins are calculated by GKS given that domestic energy prices differ from export prices. According to GKS, the margins are calculated as a percent of their “full cost” in the prices paid by the user.15 This definition raises conceptual issues: For example, in 1991, the trade and transport margin paid by the electricity sector in acquiring oil and gas products was 49%, the economy average was 36%, and the export margin was 33 percent.16 The higher margin paid by electricity for oil and gas may therefore be the result of a higher margin payment (in rubles) or it may be the result of a lower price. In fact, given that the export price of natural gas is currently some four times that of the domestic price, a 48 percent margin on the low domestic price may represent a relatively small payment. This technical issue requires considerable study beyond the scope of this particular study. A final skeptical word on the World Bank criticism: The industrial structure of any country reflects its institutions, including how its enterprises organize themselves (into conglomerates or by product divisions), their tax systems, and the degree of price controls. U.S. economic statistics, for example, will likely show a relatively low percentage of corporate dividends (due to their double taxation) and a relatively large home building sector (due to mortgage interest deduction). If the Russian tax system encourages energy concerns to separate
15 Goskmostat, Natsional’nye scheta Rossii v 1989-1994 gg, p. 127. 16 Ibid., p. 126-129. Gregory: Russian Structural Change 5/28/2004 trading and transport functions from the functions of production and refining, this fact should be reflected in Russian production statistics. If the Russian state dictates that domestic prices be well below world prices, this creates opportunities to earn economic rents and makes those with the skills and knowledge to capture such rents into scarce resources. Buyers of refined products or of natural gas would be willing to share rents with intermediaries in order to buy at the low domestic price. Sellers would be willing to share rents with intermediaries in order to sell at the world price. Given that such pricing policies are part of Russian institutions, they should be reflected in the structure of Russian production. Economic theory also underscores the ambiguity of this issue. In an economy dominated by monopolistic structures, the distribution of rents/profits is the outcome of a bargaining or gaming process, the results of which depend upon time and circumstances. Thus, in the Russian economic environment of the period 1990 to present, virtually any distributional result is possible, based purely on economic considerations. For these reasons, and until further detailed studies of the above issues are completed, we should continue to use the official Goskomstat statistics to represent the “true” structure of the Russian economy. The Structure of End Use We can measure the structure of an economy either by its production structure (as done above) or by how it uses that production in the end uses of private consumption, public consumption, investment, and net exports. The comparative literature on the Soviet system showed (as noted above) that the end-use structure was “distorted” relative to market structures by its relatively low personal consumption shares, its high investment rates and, within personal consumption, by high food expenditures and low services expenditures. Table I-5 (end of chapter) shows GKS’s official estimates of the end-use structure of GDP. Gregory: Russian Structural Change 5/28/2004
Figure I-8: Distribution of GDP by End USe, Russia (current prices)
100%
80%
60%
net exports inventories 40% fixed capital Government Consumption Total, Personal consumption
20%
0% 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
-20%
Figure I-8 shows the distribution of GDP by end use by private consumption, government consumption, net exports, and dividing investment into fixed capital formation and inventory accumulation.17 Given that the Soviet investment rate was higher than “normal” and that the consumption rate was below “normal,” we would expect to see the private consumption share rising and the investment rate falling in the course of transition. Indeed, the gross investment rate fell from 33 percent in 1989 to a low of 15 percent in 1999 before starting to experience a rising share. Personal consumption expenditures rose from 46% in 1989 to 57 percent in 1998 before dropping as a consequence of the ruble devaluation. The investment rate (the sum of fixed capital and inventory accumulation) rose in the final few years of the Soviet period (1989 to 1992) while the consumption share fell. For some reason, the chaos of the final years of perestroika reinforced the long-standing “Soviet” pattern of resource allocation. The rise of the consumption share and the fall of the investment share are evident immediately after the start of transition. The share of crucial fixed capital investment fell from almost 30 percent in 1990 to slightly less than 15 percent in 1999, while inventory investments were generally rising and were sporadically large until 1993. This would suggest a form of involuntary investment associated with the inability to sell goods in the early years of transition.
17 Goskomstat provides a category of private consumption called “non-commercial establishments serving personal consumption”, the meaning of which is not clear. We include this small category in private consumption. Gregory: Russian Structural Change 5/28/2004
The investment figures suggest an extreme lack of fixed capital investment in the first seven years of transition, which can be attributed both to a lack of supply of investment finance and to a lack of demand for domestic investment due to the huge risks associated with poorly defined property rights and other forms of uncertainty. The relative decline of fixed investment is an entirely expected result of extreme uncertainty, and the rise in the share of fixed investment after 1998 suggests both a growing expectation of future profits and of declining uncertainty.
Figure I-9: Distribution of GDP by End Use, Personal Consumption and Net Exports Combined
100%
80%
60%
inventories fixed capital 40% Government Consumption personal consumption plus net exports
20%
0% 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
-20% Figure I-9 provides another view of Russian household behavior: The most striking feature of the end- use data is the enormous rise in the share of net exports (the current account surplus), which was almost 15 percent of GDP in 1992 and 20 percent of GDP in 2000. To have a relatively low income country earning such enormous current account surpluses is highly unusual given that such countries are typically capital importers. In effect, a current account surplus of 20% of GDP means that residents are acquiring net assets abroad equal to 20 percent of total income. These results are confusing until it is recognized that the acquisition of foreign currencies is a form of accumulating foreign assets. These figures suggest that Russian citizens chose to use their falling real incomes both to consume and to acquire foreign currencies as a shield against uncertainty. Figure I-9 shows that when personal consumption and net exports are combined, they constitute a generally rising share until the currency crisis of 1998, after which the combined share fell. Russian End Use Data from the ICP Gregory: Russian Structural Change 5/28/2004
Russia and the USSR (1990) were participants in the International Comparison Project (ICP) which compares the end-use structures of a large number of countries using a common methodology. In the ICP, GDP by end use is valued both in local (domestic) prices and in “international” prices, stated usually in dollars. Hence, the ICP data allows researchers to view the structure of the use of GDP for participating countries both in their own relative prices and in relative “international” prices. To illustrate: If country X’s private consumption share is 50 percent in domestic prices and 70 percent in international prices, this tells us that private consumption goods are “cheap” in country X relative to other countries. If country X had the price structure of other countries, its residents would pay higher relative prices for consumer goods and private consumption would constitute a larger share of GDP. It is important to realize that the following comparisons deal with relative prices, not absolute prices. Thus to say that consumption is expensive means to say that consumption is expensive relative to other types of goods, such as investment goods, government consumption, or net exports. The use of ICP data for Russia is complicated by the fact that the 1990 data are for the USSR not Russia; however, we can use a study from the early 1990s that compares the GDP structures of Russia and the USSR to adjust the 1990 ICP data to reflect Russia, not the USSR.18 Table I-6 (end of chapter) provides the Russian ICP data.
18 A. Tretyakov and Barry Kostinsky, Grosss National Product Accounts of the Newly Independent States of the Former Soviet Union, 1987-1990, Appendic c, Center for International Research Bureau of Census, December 1992 Gregory: Russian Structural Change 5/28/2004
Figure I-10 Components of GDP, 1990
70% 60% 50% 40% 30% 20% 10% 0% -10% international domestic
CONSUMPTION GOVERNME NT CONS UMPTION GROSS FIXED CAPITAL IMPORTS/EXPORTS
Components of GDP, 1993
70% 60% 50% 40% 30% 20% 10% 0% internationa domestic
CONSUMPTION GOVERNME NT CONS UMPTION GROSS FIXED CAPI TAL IMPORTS/EXPORTS Gregory: Russian Structural Change 5/28/2004
Components of GDP, 1996
70% 60% 50% 40% 30% 20% 10% 0% international domestic
CONSUMPTION GOVERNMENT CONSUMPTI ON GROSS FIXED CAPITAL IMPORTS/EXPORTS
Components of GDP, 1999
70% 60% 50% 40% 30% 20% 10% 0% international domestic
CONSUMPTION GOVERNMENT CONSUMPTI ON GROSS FIXED CAPITAL IMPORTS/EXPORTS
Figure I-10 gives ICP data for Russia for the benchmark years 1990, 1993, 1996, and 1999. It is hoped that data for a later year (2003) will be made available by ICP shortly. Figure I- 10 shows the shares of major end use categories – private consumption (consumption of the population), collective consumption of the government, investment in construction and investment in equipment, and net exports (“balance of exports and imports”) in both domestic and international prices. Figure I-10 compares shares in domestic and international prices. If a share in domestic prices is smaller than in international prices, this means that the relative price of that sector is “cheap”, namely, below the economy-wide average and vice versa. For example, the 1999 government consumption share is smaller in domestic than in international prices, suggesting that the Russian government provides government consumption services more cheaply (relative to Gregory: Russian Structural Change 5/28/2004
other products) than other types of end use. On the other hand, gross fixed capital’s share is larger in international prices, suggesting capital goods are “expensive” in domestic prices. Table I-7 provides the numbers behind the above figures.
Table I-7 Russia: Components of GDP, % International domestic prices
1990 1993 1996 1999 1990 1993 1996 1999 CONSUMPTION 55 65 63 60 58 52 61 60 GOVERNMENT 15 17 25 27 10 9 11 9 CONSUMPTION GROSS FIXED CAPITAL 32 11 11 9 39 23 20 14 CONSTRUCTION 30 9 9 9 MACHINERY & 8 3 3 2 EQUIPMENT IMPORTS/EXPORTS 0 2 2 4 -1 8 4 17
The international price and domestic price figures show different trends: In international prices consumption rose (from 55% to 65% and then fell back to 60% in 1999), while in domestic prices it fell from 58% in 1990 to 52% in 1993 and then rose to 61 percent in 1996. In international prices, the share of government consumption almost doubled, while its share in domestic prices remained bogged in the neighborhood of 10 percent. In domestic prices, gross investment fell from 39% in 1990 to 14 percent in 1999, while in international prices it fell from 32% to 9 percent. Net export shares were higher in domestic prices than in international prices. As Table I-7 and Figure I-10 show, there are remarkable differences in relative prices between Russia and the “international” price system. The higher shares of government consumption in international prices suggests that government services are “very cheap” in Russia; the lower shares of gross fixed capital in international prices suggests that capital goods are “relatively expensive” in Russia, The latter result is perhaps the consequence of the fact that capital goods are increasingly imported. The higher share of net exports in domestic prices suggests that Russians are paying relatively more for foreign imports than residents of other countries. Table I-8: Break down of personal consumption, per cent (Revised Numbers)
Year 1990 1990 1993 1993 1996 1996 1999 1999 Gregory: Russian Structural Change 5/28/2004
Prices Domestic International Domestic Inter Domestic Interna Dome Internati natio tional stic onal nal 15 8 13 6 11 7 11 4 CLOTHING & FOOTWEAR 37 26 35 21 34 18 39 18 FOOD,BEVERAGES, TOBACCO 6 14 10 22 9 24 6 28 GROSS RENTS, FUEL & POWER 7 6 4 3 5 2 1 0 HOUSEHOLD EQUIPMENT & OPERATION MEDICAL CARE 6 13 6 23 8 23 7 21 11 12 11 6 9 5 7 5 MISCELLANEOUS GOODS & SERVICES TRANSPORT & 6 5 6 5 11 5 11 7 COMMUNICATION RECREATION, 12 18 12 19 12 23 10 18 EDUCATION
Table I-8 and Figure I-11 show the “distortions” in the Russian pricing system, which are the result of remaining price controls, the failure to liberalize prices, and implicit policies of not collecting revenues from households for necessities for political reasons. The general pattern is that goods freely bought and sold (such as food, beverages, and tobacco and clothing and footwear) are expensive in Russia by international standards (the domestic-price share exceeds the international price share). On the other hands, goods provided by government, by utilities, or necessities (such as medical care) are very cheap by international standards. The remarkable feature is that discrepancies between the domestic price system and international prices have been growing over time. For example, the discrepancy between gross rents, fuel and power was 7 percentage points in 1990 (6% vs. 13%), while it was 24 percentage points (6% vs. 30%) in 1999. These figure underscore the urgency of further deregulation of the power sector, education, and medical care. In effect, the low (sometimes virtually zero) prices of some products placed little or no downward pressure on consumption of those items; so that Russian households consumed virtually equal shares of food, beverages and tobacco, rent and power, medical care and education measured in international prices, but relatively small shares of the last three measured in domestic prices. The large discrepancies between domestic and international prices with respect to housing, power, rents, medical care and education impede the development of Russian consumption to a more “normal” pattern as will be discussed in Part II. The distribution of expenditures looks somewhat different if we look at private consumption of households as provided by the Russian Longitudinal Monitoring Survey Gregory: Russian Structural Change 5/28/2004
(RLMS), the annual household survey conducted by the University of North Carolina in a number of Russian territories.19 Unfortunately, the first round of this survey study was conducted only in Fall 1992, when structure of consumption had been already modified significantly by the hyperinflation. It does, however, show the trends through the rest of the 1990s. RLMS data (Figure I-12) show that food, beverage, and tobacco remain the largest expenditure category of Russian households, although the downward trend is apparent here, too. The major difference between the patterns of total and private final consumption of households lies with the rent and utilities. Continuing subsidization of utilities and residential maintenance coupled with mass housing privatization produced the result, whereby this category still plays marginal, albeit growing, role in private expenditure. Thus RLMS, which of course is denominated in Russian domestic prices, paints the same picture of price-induced distortions in the pattern of Russian household consumption derived from a source that is independent of Goskomstat.
19 Mroz, T., L. Henderson, M. Bontch-Osmolovsii, and B.M. Popkin. “Monitoring Economic Conditions in the Russian Federation: The Russia Longitudinal Monitoring Survey 1992-2002.” Report submitted to the U.S.A.I.D. Carolina Population Center, University of North Carolina at Chapel Hill. March 2003. Gregory: Russian Structural Change 5/28/2004
Figure I-11: Personal consumption, 1990
40% 35% 30% 25% 20% 15% 10% 5% 0%
domestic prices international prices
Personal consumption, 1993
40% 35% 30% 25% 20% 15% 10% 5% 0%
domestic prices international prices Gregory: Russian Structural Change 5/28/2004
Pe rsonal consumption, 1996
40% 35% 30% 25% 20% 15% 10% 5% 0%
domestic prices international prices
Pe rsonal consumption, 1999
45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
domestic prices international prices Gregory: Russian Structural Change 5/28/2004
Figure I-12: Structure of Russian Household Consumption (RLMS)
100%
80%
60%
40%
20%
0% Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02
rent & utilities durables services, incl. dining, payments for tution etc. nondurables food, alcohol, tobacco
International Trade A distinctive feature of the Soviet pattern of industrialization in the USSR and Eastern Europe was the underutilization of trade potential. USSR and Eastern European economies had smaller shares of exports plus imports divided by GDP than did market economies at similar levels of development. It was extremely difficult to measure trade ratios (exports plus imports as a percent of GDP) for the Soviet period because of the problem of combining trade flows to socialist bloc countries which were conducted in artificial prices with hard currency trade. Data for the early 1960s estimated Soviet trade ratios at a very low 26 percent.20 It is also difficult to calculate trade ratios for the early years of transition because of exchange rate problems and hyperinflation. Russia’s market exchange rate was grossly undervalued relative to purchasing power parity, while most exports (and probably most imports) were traded in hard currencies. Thus trade would have to be converted back into rubles at artificial market exchange rates, yielding exaggerated high trade ratios. Figure I-13 captures the distorted ruble exchange rate in 1992, as the trade ratio rose from slightly over 20 percent in 1991 to over 100 percent in 1992. 21 Thereafter, the trade ratio
20 Paul Gregory, Socialist and Nonsocialist Industrialization Patterns, p.112. 21 In the early 1990s, Russia was exporting to hard currency countries earning dollars, say, for its energy exports. While at the same time, the ruble exchange rate was grossly undervalued. Nevertheless, if we take the total Gregory: Russian Structural Change 5/28/2004 settled at from slightly below 60 percent in the mid 1990s, rising to a high of 71 percent in the aftermath of the 1998 ruble devaluation. Note that Russia has consistently run a positive balance on current account since 1992, a point discussed above. Figure I-14 provides another view of the importance of trade to the Russian economy – per capita imports and exports in constant 1995 dollars. This data (along with the rest of the data in this section) was gathered from the World Bank data base, and all figures are in U.S$. GKS itself reports its own trade figures in dollars; therefore, the use of dollars is standard practice both for international organizations and for Russia. It shows that exports have ranged between the low $500s and the mid $700s between 1992 and the present. Exports per capita exceed imports per capita, but, as is usual in international trade, if export earnings are high, there is an opportunity for import earnings also to be high. At first glance, for an economy listed as having a GDP per capita of slightly over $4,000 in the mid 1990s, exports would equal some 20 percent or more of GDP per capita. With the Russian exchange rate at 69 percent of the PPP rate (1997), the ratio as measured in domestic prices would be in the neighborhood of 30 percent. In Part II, we consider whether such rates are high or low by international standards. The many panels of Figure I-15 tell the story of the composition of Russian trade. Panel A shows that Russia is primarily an exporter/importer of merchandise (rather than services), The share of merchandise rose from 57 percent in the mid 1990s to 90 percent in the late 1990s. The share of merchandise imports similarly rose (from 15% to 72%). Of these merchandise items, Russia spends about 14 percent of its imports on food (in a country that is rich in agricultural resource) and the 1998 crises resulted only in a temporary dip in the share of food imports. Panel C shows the dominance of Russian exports by fuels, which rose from 37 percent of exports in 1996 to 49 percent in 2001. Panel D shows that Russia is both a manufacturing exporter and importer, although its imports increasingly outweigh its exports. Manufacturers rose from 36% to 45% of exports, while manufacturing imports held roughly steady at around 20 percent. Panel E shows Russia as a major exporter of ores and metal products, with shares fluctuating between 7 and 14 percent of exports. The fluctuating share shows the instability of this export market for Russia.
value of exports in 1991 (Natsional’nye scheta, p. 112), assume a balance of exports and imports and divide by the nominal value of GDP (p. 33), we get a trade proportion for 1991 of 28%.. The same exercise yields a trade proportion of only 15 percent for 1992 (natsional’nye. Scheta, p. 33, 155). We do not know how Goskomstat made these calculation, and the figure for 1992 drawn from the World Bank data base gives an entirely different result. Hence we would argue that we really do not know the trade ratios for the early 1990s. Gregory: Russian Structural Change 5/28/2004
Panel F shows Russia as a substantial importer of services, which account for 20 to 27 percent of the total, but it is also a not-insignificant exporters of services (at between 9 and 14 percent of the total). Panel H shows Russia as an arms exporter, whose market share (as a percent of exports) fluctuates wildly between 2.5 and 8 percent) probably as a consequence of erratic orders and sales. Panel I shows one of the fruits of the new market economy – the ability of Russian nationals to travel abroad. What used to be a one-way street in the Soviet period is now one dominated by Russian tourists abroad. In the second half of the 1990s, Russian international tourism accounted for between 10 and 14 percent of Russian imports, while international visitors to Russia accounted for between 3 and 9 percent of Russian exports. In Part II, we examine whether Russian trade ratios, per capita trade, and trade patterns of exports and imports fit into established trade patterns of other countries. For now, we can say that these figures point towards a marked move forward in the integration of the Russian economy into the world market. Russian fuel exports are making a significant dent in world energy markets (as they did in the Soviet period), Russians are eating imported foods, and are buying foreign-produced merchandise.
Gregory: Russian Structural Change 5/28/2004
FIGURE I-13: Russian Trade Ratios Exports/imports of goods & services (% of GDP)
Year Exports Imports 1990 18.1617 17.9444 1991 13.2712 12.9852 1992 55.5673 50.4439 1993 35.5011 31.6193 1994 27.5249 23.7318 1995 27.681 24.5126 1996 24.8072 20.7182 1997 24.0597 21.4674 1998 30.8944 26.3571 1999 43.7648 27.3158 2000 44.5075 24.0627 2001 36.81 24.1454
Gregory: Russian Structural Change 5/28/2004
FIGURE I-14: EXPORTS AND IMPORTS OF GOODS AND SERVICES PER CAPITA (constant 1995 US$) Year Exports Imports
1990 1096.27 1684.68 1991 765.936 900.68 1992 546.054 602.076 1993 558.224 541.529 1994 567.832 472.228 1995 631.027 558.799 1996 696.663 611.307 1997 729.168 701.174 1998 714.719 593.944 1999 705.337 426.679 2000 730.513 515.516 2001 753.887 603.907
Gregory: Russian Structural Change 5/28/2004
Panel A
Figure I-15
Merchandise Exports and Imports, % of Total Exports (imports) (based on current US$) Year Exports Imports 1990 1991 1992 14.8456 1993 26.4175 1994 57.2075 46.376 1995 67.7357 52.0107 1996 85.3178 79.3612 1997 85.6849 80.0337 1998 85.8374 79.5239 1999 89.3363 75.0076 2000 91.3104 71.9431 2001 90.3649 71.9678
Gregory: Russian Structural Change 5/28/2004
Panel B Food Exports and Imports as % of Total Exp/Imp (curr US$) Year Exports Imports 1990 1991 1992 1993 1994 1995 1996 1.51 14.15 1997 1.31 14.88 1998 1.37 13.67 1999 0.90 14.04 2000 1.14 10.83 2001 1.31 14.74
Gregory: Russian Structural Change 5/28/2004
Panel C Fuel Exports and Imports as % of Total Exp/Imp (curr US$) Year Exp Russia Imp Russia 1990 1991 1992 1993 1994 1995 1996 36.7784 2.18608 1997 39.2091 2.6014 1998 32.8288 2.01081 1999 37.3069 1.47248 2000 46.8219 2.38009 2001 48.6996 1.75001
Gregory: Russian Structural Change 5/28/2004
Panel D Manufactures Exports and Imports as % of Total Exp/Imp (curr US$) Year Exports Imports 1990 1991 1992 1993 1994 1995 1996 22.1814 35.9364 1997 19.8436 36.752 1998 24.02 34.6263 1999 22.0879 31.3548 2000 20.1945 30.3385 2001 19.8048 45.2577
Gregory: Russian Structural Change 5/28/2004
Panel E Ores & metals Exports and Imports as % of Total Exp/Imp (curr US$) Year Exports Imports 1990 1991 1992 1993 1994 1995 1996 8.57143 1.33327 1997 9.67611 1.38728 1998 13.7181 1.25249 1999 10.4593 1.43444 2000 8.49066 1.86379 2001 7.20863 1.84222
Gregory: Russian Structural Change 5/28/2004
Panel F Service Exports and Imports as % of Total Exp/Imp (based on curr US$) Year Exports Imports 1990 1991 1992 1993 1994 9.39019 19.9555 1995 11.3044 24.4084 1996 12.7897 21.5204 1997 13.6581 21.7709 1998 14.1792 22.1052 1999 10.7003 25.2441 2000 8.63329 26.7629 2001 9.55677 25.9287
Gregory: Russian Structural Change 5/28/2004
Panel G Communications,computer,etc.,% of Total Exports / Imports(curr US$) Year Exports Imports 1990 1991 1992 1993 1994 2.29398 6.70752 1995 2.57464 6.31425 1996 2.37388 6.84711 1997 3.0322 7.10013 1998 2.95973 6.19924 1999 2.61424 6.9789 2000 2.1171 7.36961 2001 2.02554 7.30996
Gregory: Russian Structural Change 5/28/2004
Panel H Arms Exports / Imports (% of total Exports / Imports) Year Exp Russia Imp Russia 1990 1991 1992 5.8 1993 7.9 1994 2.5 1995 4.3 1996 3.4 0.1 1997 2.9 1998 2.9 0.1 1999 4.2 1.1 2000 2001
Gregory: Russian Structural Change 5/28/2004
Panel I International tourism, receipts (% of total Exports / Imports) Year Exp Russia Imp Russia 1990 1991 1992 1993 1994 3.16328 10.7639 1995 4.61271 14.0069 1996 6.61373 11.8377 1997 6.94944 10.1763 1998 7.45845 11.1171 1999 8.86318 14.0561 2000 2001
Gregory: Russian Structural Change 5/28/2004
Part II: RUSSIAN TRANSITION IN INTERNATIONAL CONTEXT The first part of this paper discussed the changes in the structure of the Russian economy as measured by the production structure, the end use structure, patterns of consumption, and patterns of foreign trade that took place in the course of transition. This second part examines the changing structure of the Russian economy in international perspective; namely, in terms of comparisons with other countries at different levels of economic development. What we wish to learn is the extent to which the Russian economy in the course of transition has become more like a market economy, and, if so, more like what type of market economy – a low middle-income country (which is where international organizations peg the current Russian economy), or a higher income country? Such comparisons shed light on the eventual outcome of Russian transition. When it is all over, will Russia be an affluent economy like industrialized Europe, will it be a troubled low to middle income country, such as Turkey, or will it be at the fringes of affluence, such as a Greece or a Portugal? To make such comparison, we require panel data on value added, labor force, and foreign trade distributions for a large number of countries that we can then combine into country groups for comparison with Russia. We also require both Russian and international data coded into a common set of branches and categories, which means that we must restrict ourselves to the lowest common denominator of sectoral classifications used in large samples of countries. We must also insure that the Russian data is coded according to the international standards, such as that of the International Labor Organization as prescribed by the generally-accepted International System of Industrial Classification (ISIC). We have indeed compiled a number of international panels of countries at different levels of development as measured by real per capita income in purchasing power parities. The sources of data and classification issues are discussed in Appendix B. Prior to the start of transition, the Russian economy, like other socialist economies, was characterized by significant structural deviations from market economies at similar levels of development. These distortions resulted from the decisions made by central planning authorities with little respect to consumer preferences. This research aims at determining the extent to which transition has removed such distortions. Aggregate level (Industry, Agriculture and Services: We begin at the highest level of aggregation: the trichotomy of agriculture-industry- service sectors. It has been long established that most important proportions of national economies correlate with real per income – the most common measure of the level of economic development: In particular, the share of the service sector (both in terms of value added and employment) grows with per capita GDP in economies past the initial phase of industrialization. Gregory: Russian Structural Change 5/28/2004
22 Although this relationship holds for any group of countries (for the postwar period) including former Soviet Union and Eastern Europe, the trend for the socialist economies was significantly displaced with respect to the dominant pattern exhibited by the rest of the world. Moreover, the shares of agriculture and industry, especially heavy industry, in centrally planned economies were higher than elsewhere. In other words, these economies were lagging behind market economies in the rate of structural modernization.
% 90
, 80 70 60 l employment a t 50 o 40 30
services in t 20 10
Share of 0 1000 10000 100000 Per capita GDP, 1990 intl. $
market economies transitional economies Trend (market economies) Trend (transitional economies)
% 80
, 70
60
l employment 50 a t o 40
30
20 services in t 10
Share of 0 1000 10000 100000 Per capita GDP, 2000 intl. $
market economies transitional economies Trend (marketconomies)
Figure II.2. Level of development and service sector employment, 1990 and 2000. Source: WDI.
22 See, for example: Hollis Chenery, Sherman Robinson, and Moshe Syrquin,. Industrialization and Growth: A Comparative Study, Oxford University Press, 1986. Gregory: Russian Structural Change 5/28/2004
Figure II.1 shows the relationship between the level of development and the service sector as a share of the total employment for both transition and market economies at the start of transition (1990). Trend lines are imposed for market and transition economies separately, which show that transitional economies as a group had smaller service sectors. Although the slopes of the trend lines are essentially the same, the transitional-economy trend line is characterized by a downward shift.23 Figure I.2 revisits this relationship after a decade of transformation. For all practical purposes, the difference in the relationship has been removed on average, even though East and Central Asian transitional economies still fall below the general trend. The Russian economy, denoted by a larger gray square in Figures II.1 and Figure II. 2, followed the general restructuring pattern. A typical centrally-planned economy with a deficient service sector in 1990, Russia becomes “normal” ten years later (the corresponding point lies close to the trend line for the market economies in 2000). Similar patterns of change characterize the contribution of the service sector to value added. The relationship between per capita GDP and the service sector share of value added is significantly different for market and transitional economies in 1990, with the latter again being shifted downward relative to market economies. By 2000, the differences are practically wiped out (Figure II.3). Russian transition is again a typical case (again represented by the larger gray square). It lies on the transition trend line in 1990 and is close to the common trend line in 2000.
23 Regression equation for transitional economies, Ss = 12.3 ln (Y) - 73.5, differs significantly from the regression for market economies, Ss = 12.6 ln (Y) - 58.9, only in the intercept. The slopes do not differ significantly. In the equations above, Ss denotes the share of service sector in the total employment and Y denotes GDP per capita in 1990 USD measured at purchasing power parity. Gregory: Russian Structural Change 5/28/2004
80 70 % P, 60 50 40 30 services in GD 20
Share of 10 0 1000 10000 100000 Per capita GDP, 1990 intl. $
market economies transitional economies Trend (market economies) Trend (transitional economies)
80
70 %
P, 60
50
40
services in GD 30
20
Share of 10
0 1000 10000 100000 Per capita GDP, 2000 intl. $
market economies transitional economies Trend (marketconomies)
Figure II.3 Scatter Diagram of Service Sector Shares, 1990 and 2000 Gregory: Russian Structural Change 5/28/2004
Table II.1: The Structure of Value Added, Labor Force, and Relative Productivity, Russia and Country Groups (by income level)
Value added, % of total Employment, % of total Productivity index (total economy =1) Country Year group Agricultur Industry Services Agricultur Industry Services Agricultur Industry Services e e e
1990 H 4.6 33.0 62.4 8.3 29.6 61.8 0.55 1.11 1.01 2001 2.4 30.3 67.3 5.1 26.9 67.9 0.48 1.13 0.99 1990 LH 8.4 34.6 57.0 12.5 29.8 57.6 0.67 1.16 0.99 2001 3.9 33.9 62.2 8.8 27.4 63.6 0.45 1.24 0.98 1990 UM 8.6 39.4 52.0 18.2 26.4 54.4 0.47 1.50 0.96 2001 6.0 36.3 57.7 15.6 23.3 60.6 0.38 1.56 0.95 1990 LM 17.3 32.6 50.1 31.1 22.9 44.6 0.56 1.43 1.12 1999 13.9 31.3 54.8 30.4 20.8 48.8 0.46 1.50 1.12 1990 CEE 14.0 45.1 40.8 20.2 40.2 39.3 0.69 1.12 1.04 2000 7.3 33.1 59.7 12.6 31.0 56.5 0.58 1.07 1.06 1990 Russia 16.5 48.6 34.9 13.9 40.1 41.0 1.19 1.21 0.85 2001 7.0 37.8 55.2 13.0 31.6 55.4 0.54 1.20 1.00 Notes: 1) H, UM, LM stand for high, upper-middle, and lower-middle income groups respectively as defined by World Bank. LH is a lower subdivision of high-income group that includes countries such as Spain, Greece, and South Korea. CEE includes for transitional economies of Central and Eastern Europe (CIS is not included). 2) The numbers in the table are imprecise due to changing availability of data for separate countries and intended only to show principal trends. 3) Employment data for high-income countries are for the year 2000. Sources: WDI, Goskomstat. Table X provides a listing of the countries that fall in each group, with Russia falling into the lower-middle income group.
Table II-1 covers an eleven year period for five groups of countries and for Russia. In all countries, agriculture’s and industry’s shares of value added and labor force declined while the shares of services rose. In all countries, agriculture’s relative productivity fell and stood at around half the economy-wide productivity level. In all countries, the service sector had a relative productivity that did not deviate significantly from the economy average. The more developed the country, however, the closer the relative productivities of industry and services, suggesting that affluent countries produce “high end” services. Gregory: Russian Structural Change 5/28/2004
The structural changes in the transition countries of central Europe and Russia, were more pronounced than in other countries and followed the general trend of rising services and falling industry and agricultural shares. Russia’s transformational changes were percentage-wise about the same as in CEE, but Russia started from more “backward” initial conditions (relatively high shares of agriculture and industry). The message of Table II-1 is that the transition economies underwent much more significant change over the studied 11 year period, while moving in the expected direction. The extent of reallocation of labor force across the three aggregate sectors is lower in Russia than in Central Europe, particularly in agriculture, reflecting lower labor mobility due to much larger size of the country, higher concentration of labor, and less flexible labor marker institutions. It should be added that employment shares in Russia remained virtually unchanged until about 1994 (See Part I above). Moreover, the share of agricultural employment even increased (to about 15%) in mid-1990s due to low mobility of agricultural labor against the background of surging unemployment in industry. The resistence of the agricultural sector to change appears to contradict the general modernization trends in the Russian transitional economy. However, it serves as another piece of evidence that the structure of the Soviet economy was driven by central planners rather than by individuals. The removal of artificial labor incentives combined with economic disorganization could explain the slow change. It is reasonable to expect, given the comparative patterns of change presented in Table II.1 and the relative decline of agricultural productivity in Russia, that the outflow of labor from agriculture is about to resume.24 Index of Structural Deviation We summarize the dynamic of structural change in the Russian economy with respect to other economies using an “index of structural deviation,” which we define as follows:
2 Dk = Σi (SRi – Ski) ,
th th where SRi is the share of i sector in the Russian economy and Ski is the average share of i sector in the economies of country group k. This index measures the “distance” between the Russian economy and other economies taking into account all the components of value added (or labor force) simultaneously. Proximity of D to zero would mean that the Russian economy is insignificantly different from the economies of country group k. We again classify countries into high, lower- high, upper-middle, and lower-middle income countries (see above and Table X).
24 New data (Russian Statistical Yearbook 2003) suggest that this process is already under way: the share of agricultural employment proper (without forestry and fishing) declined from 13% to 11.8% in 2000-2. Gregory: Russian Structural Change 5/28/2004
Aggregate value-added and employment indices are presented in Figure I1.4 and Figure I1.5 respectively.25
1500
1000 High "Low er High" Upper Middle Low er Middle 500
0 1990 1992 1994 1996 1998 2000 2002 Figure II.4. Indices of structural deviation. Value-added shares of Agriculture, Industry and Services Source: WDI, Goskomstat. Calculations by the authors.
1000
High "Low er High" 500 Upper Middle ` Low er Middle
0 1990 1992 1994 1996 1998 2000
Figure II.5. Indices of structural deviation. Employment. Shares of Agriculture, Industry, and Services Source: WDI, Goskomstat. Calculations by the authors.
25 Index of structural deviation, Dk, measures only the relative distance between the Russian economy and that of a corresponding country group. Absolute values of the index, as well as position of various country group curves in the graph bear no meaning. In particular, the nearly equal values of the indices for the lower-high and lower-middle countries do not mean that these are countries similar, only that the Russian economy is equidistant in structural terms from both groups. Gregory: Russian Structural Change 5/28/2004
Figure II.4 shows that the composition of Russian value added (by industry, services, and agriculture) differed greatly from that of any other group of market economies at the start of transition. The structural distortions (deviations) were eliminated rapidly during the early phase of transition, 1992-95, and stabilized thereafter. The average deviation of the Russian economy from upper-middle income countries has been insignificant since 1995. Until the ruble crisis of 1998, the lower and lower-high income countries were closest to the Russian pattern, but after the currency crisis, the lower-high income countries are slightly closer to Russia. Figure II.5 shows that the employment shares follow a different path, as might be expected from the slower movement of employment. On the eve of transition in 1990, the Russian labor force distribution closely resembled that of lower-middle income countries despite a significantly higher level of income, while deviations from closer income groups were relatively large. The first few years of transition had a small effect on the labor force distribution; changes became notable only since 1994. The general tendency in the subsequent years is toward convergence with upper-middle and lower-high income country groups, with corresponding separation from the lower-middle income group. Employment adjustment slowed down after 1998.26 In Part I, we examined the factors behind the relative growth of Russia’s service sector, such as the explosive growth of business services which were virtually non-existent in Soviet times, in contrast to “old” service sectors such as health and education, which did not experience relative increases. New market-oriented service sectors clearly led the expansion with banking and insurance increasing almost four-fold in ten years and the small real estate sector doubling in size every two years. This employment growth in services was fed largely by movement from industry and from traditional services. Part I also described the surprising resilience of employment in government and social services, which proved to be one of the fastest growing employment areas in the second half of the decade..27 The share of government employment increased from the low of 2.1% in 1992 to 5.0% in 2000.28 Absolute employment also increased, most rapidly in the mid-1990s. This development is especially remarkable since this indicator in the Soviet Union was stable for decades, 1.7-2.1% from the late 1950s till the late 1980s. Although the increase in government employment compared to the Soviet period may be a
26 Due to the lack of consistent data Figure 1.6 is insufficient to make such a conclusion. However, detailed labor distribution data presented later in this project also exhibit this tendency. 27 “State administration” (upravlenie) in terms used by Goskomstat. 28 CIS Statistical Committee. According to the most recent Goskomstat publication, Russian Statistical Yearbook 2003, this share reaches 4.5% in 1990 and remained at that level thereafter. Gregory: Russian Structural Change 5/28/2004 statistical artifact, its rapid expansion in the mid-1990s is significant and can be viewed as signifying the retrenchment of the new Russian bureaucracy during Yeltsin’s “glory years”.
International Comparisons Nine Sectors : The “aggregate” (agriculture, industry, services) patterns of Figures II.4 and II.5 – rapid adjustment of value-added shares in the early 1990s, lagged and less pronounced reallocation of labor shares, with subsequent stabilization of both processes – can be further studied using more detailed data on the distribution of value added and labor in the Russian economy by more finely disaggregated sectors. Although Goskomstat produces detailed data of this kind (see for example the 23 sector classification in Table 1, Part I), we must work with more modest sectoral classifications (as discussed in Appendix B) that make direct comparisons with international data possible. As Appendix B discusses, we use a nine sector classification in this subsection, which allows us to compare Russia’s value added and labor force share distributions with those of a large number of countries again aggregated into income groups. Basically we use the ILO classification system (described in Appendix B) to make international comparisons for the eight sectors used in the figures in this section.
Table II.2 Employment Shares, Eight Economic Branches (ILO categories)
Country Agricult Commun Construc Electricit Financin Manufac Mining Transpor Wholesale Year group ure, ity, tion y, Gas g, turing and t, and Retail Forestry Social and Insuranc Quarryin Storage Trade and and and Water e, Real g and Restaurant Fishing Personal Estate Commun s and Services and ication Hotels Business Services 1990 H 5.7 28.0 7.3 1.0 9.2 20.8 0.5 6.9 20.0 2001 3.8 31.1 7.2 0.7 12.7 16.8 0.4 7.2 20.0 1990 LH 15.1 23.4 7.4 0.8 6.0 22.9 0.5 5.7 17.9 2001 8.9 24.6 8.8 0.7 9.9 17.5 0.2 6.0 22.4 1990 UM 15.6 31.5 6.6 1.0 4.4 18.0 1.8 5.6 18.9 2001 9.3 29.5 8.0 0.7 6.1 17.0 0.5 6.8 22.1 1990 LM 28.1 24.9 5.3 0.6 2.7 15.3 1.0 4.9 16.9 2001 22.1 22.4 6.5 0.6 4.5 13.9 0.8 6.0 23.0 1990 Russia 13.2 21.9 13.4 0.8 0.8 29.2 1.5 7.7 7.8 Gregory: Russian Structural Change 5/28/2004
2001 13.4 28.1 8.8 1.5 2.4 20.2 1.3 7.8 14.6 Source: ILO
Trends in the nine sector (ILO) classification are less pronounced than those in the three sector classification. In all country groups, the employment shares of agriculture (including fishing and forestry) fell, the share of utilities (electricity, gas, and water) was roughly stable, the share of financing, insurance, and real estate rose, the share of manufacturing fell, the share of transportation and communication rose modestly, and the shares of trade rose in all groups except the high income countries. Russian employment shares moved in the same direction for those sectors where trends were evident, although Russia’s trends were more pronounced given its more “backward” initial employment shares. Even after eleven years of adjustment, Russia’s 13 percent employment share in agriculture (broadly defined) places it above all country groups except lower-middle income countries. Even after a tripling of employment shares in financing, insurance and real estate and a doubling of employment shares in trade, Russia still remains far behind other country groups including even lower-middle income countries.
0.070
0.060
0.050 High 0.040 "Low er high"
0.030 Low er Middle Upper middle 0.020
0.010
0.000 1990 1992 1994 1996 1998 2000 2002 Figure II.6. Indices of structural deviation. 8-sector distribution of labor. Source: ILO, Goskomstat. Calculations by the authors.
Figure II.6 quantifies these employment share changes in terms of the index of structural deviation. The four curves show a clear pattern of convergence of the Russian 9-sector labor force distribution toward that of market economies. Market adjustment process have made the Russian economy more similar to all other groups, although the rate and degree of convergence is the highest for the upper-middle income country group (remember, the closer to the horizontal axis, the lesser the “distance”). The lower-highincome group is next closest to that of Russia. As in the case of the aggregate distribution of labor, discussed above, the upper-middle income and Gregory: Russian Structural Change 5/28/2004 the lower-high income countries appear to constitute the reference group for the contemporary Russian economy.
Much of the convergence tendency, it appears, can be explained by the contraction of the oversized Russian manufacturing sector. However, the largest contribution to the reduction of the deviation from Russia’s nearest counterparts, lower-high and upper-middle income countries, is made by the service sector category denoted in ILO classification as “Wholesale and Retail Trade and Restaurants and Hotels”. In particular, this category is responsible for 50% of the reduction in the “distance” between the Russian economy and the lower-high income group between 1990 and 2000. Despite the strong growth in this sector, the Russian economy still lags substantially behind typical market economies with respect to “trade” (broadly defined), as the data presented in Figure II.7 show.
25%
20%
"Lower high" 15% Lower Middle Upper middle 10% Russia
5%
0% 1990 1992 1994 1996 1998 2000 2002 Figure II.7. Employment Shares in “Trade, Restaurants and Hotels” sector. Russia vs. three country groups. Source: ILO, Goskomstat.
The behavior of the tails of the index curves in Figure II.6 suggest that Russian economy might have started moving away from all groups but upper-middle after 1998. However, the robustness of this trend cannot be established given the relatively few observations.. It is apparent, however, that the rate of structural change in the Russian economy slowed down in the late 1997, although it had not yet reached adequate proportions. Future development will show whether this was a transitory slowdown, possibly related to the recession of 1998-9, or a Gregory: Russian Structural Change 5/28/2004 permanent stabilization at a state removed from the Soviet point of origin but not fully consistent with the parameters of a normal market economy.
Trends in industry: International Comparisons As noted in Appendix A, Goskomstat does not provide much official data on the breakdowns of Russian industry. We are further limited by the availability of data on the structure of industry in international panel data sets, such as that of the World Bank. International data bases are particularly weak in terms of manufacturing labor force shares. Hence, unlike earlier comparisons that included both labor force and value added shares, we can present only value added analysis. We cannot offer international comparisons of developments of labor force shares. Table II.3 supplies internationally-comparable data on value added in manufacturing split into four large sectors (plus a fifth “other” category).
Table II.3 Manufacturing value added by sector, percent total. Year Country Chemicals Food, beverages Machinery and Textiles and Other group and tobacco transport clothing manufacturing equipment
1990 H 9.9 15.6 27.8 7.2 40.8 2000 9.0 14.9 29.9 7.6 38.6 1990 LH 7.7 16.0 18.5 11.5 46.4 2000 8.2 18.4 22.1 10.8 40.5 1990 UM 11.7 28.5 11.0 7.9 41.0 2000 7.9 30.8 14.1 5.8 41.4 1990 LM 9.2 26.1 13.9 16.0 39.4 2000 7.4 30.1 7.2 17.5 38.3 1990 Russia 3.7 7.5 34.6 20.8 33.4 2001 6.6 13.9 28.3 1.6 49.7 Source: WDI. There are few distinctive trends within manufacturing. Food, Beverages and tobacco increased its share in all but the highest income group; machinery and transport equipment increased its value added shares except in the lower-middle income group. What stands out is Russia’s peculiar initial conditions – an extremely low share of food, beverages and tobacco, an extraordinarily high share of machinery, a very high initial share of textiles, and a low initial share of trade. Gregory: Russian Structural Change 5/28/2004
1500
1000 High "Lower High" Upper Middle Lower Middle 500
0 1990 1992 1994 1996 1998 2000 2002
Figure II.8. Indices of structural deviation. Manufacturing value added. Source: WDI. Calculations by the authors.
Figure II.8 again applies our structural convergence index measure to the structure of manufacturing value added. It shows a quite unusual pattern, in which the Russian manufacturing structure on the eve of industrialization was most like that of high income countries and was quite similar to lower-high income countries. In the first few years of transition, Russia became even closer to these two income groups. The Russian distance from upper-middle and lower- middle income countries was much greater. In the course of transition, however, Russian manufacturing did not exhibit strong convergence trends. It remained practically equidistant from high- income economies,29 and even tended to diverge in the late 1990s. The convergence trend with respect to upper-middle income countries was more pronounced, but the structure of Russian manufacturing remained far removed its usual reference group (as noted above) -- the upper-middle income countries. In sum, Russian industry has moved slower toward “normal” proportions than the economy as a whole. This may be a result of the higher degree of regulation of industry than service sectors. Whereas price liberalization and demonopolization of international trade led to significant changes in the composition of the value of industrial output, labor market rigidity produced less adequate labor force reallocation. While assessing trends in Russian industry, we should also take into account that implicit subsidization of inefficient enterprises may further distort data: the effect of so called “virtual economy” concept introduced by Ickes and Gaddy,30
29 The only large drop in the indices between 1990 and 1991 obviously cannot be attributed to the effect of transformational reforms and therefore bears no significance. This may have been the consequence of a “pre- transformation” transformation as a consequence of the collapse of the planning and ministerial system after 1989. . 30 Clifford Gaddy and Barry Ickes. Russia’s Virtual Economy, Foreign Affairs, vol. 77, no. 5, 1998. Gregory: Russian Structural Change 5/28/2004 which implies that only a few resource-extracting sectors generate value added, while the rest of the industry effectively destroys the value. The most recent research, however, suggests that one of the supposed “value creating” sectors – electricity – itself may be operating at a loss..31 Given the publication of a complete list of annual input-output tables, the effect of subsidies can be removed from the picture, insofar as the input-output tables provide rather detailed sectoral data on subsidies,32 but such research remains to be done. Patterns by End Use In Part I, we described changes in the pattern of end use of Russian GDP. We used official Russian data on the distribution of GDP among personal consumption, government consumption, investment and net exports, and we examined trends in personal consumption using International Comparison Project (ICP) data in both domestic and international prices using four rounds that included Russia (1990, 1993, 1996, 1999). The Russian ICP data were presented in Part I in both domestic prices, which capture Russian relative scarcities but also distortions in Russian pricing, and international (PPP) prices, which imposes international relative prices on Russian real products. In Part I, we showed that some sectors of the Russian economy, particularly health care, housing and power, remain subject to an extremely distorted pricing system, which yields structural distortions in Russia’s use of GDP.
Composition of GDP (Consumption, Investment, Net Exports) in International Comparisons: The Russian data, presented in Table 1-6 (end of paper) , are analyzed here in conjunction with the trends in market economies divided into three income groups. The lower- middle income group is dropped due to reasons of data availability in the ICP (see definitions in the note under Table 1I.1). 33 In Figure II.9 we calculate the index of structural deviations for Russia vis-à-vis high, lower high, and upper middle income countries using international prices. This figure shows how the structure of end use of GDP in Russia differs from that of other groups of countries assuming that all countries use “international” prices rather than their own domestic
31 Vlad Ivanenko. “Searching for the Value-subtraction in the Russian Economy”, Journal of Comparative Economics, Vol. 32, 2004, pp. 88-104. 32 In fact, the whole subsidy issue may not affect these tables and figures insofar as they are supposed to be in “basic” prices which eliminate subsidies as well as indirect taxes. 33 This analysis does not involve lower-middle income group for the following reasons. First, between 1990 and 1993 the list of high-income and upper-middle countries covered by the ICP broadened significantly. For example, the 1990 list included only European high-income countries. Second, the lower-middle income countries represented in ICP are mostly transitional economies that cannot be included with the reference market economies. Gregory: Russian Structural Change 5/28/2004 prices. Thus the relative overpricing of Russian consumption goods and the relative underpricing of government consumption are removed from this comparison.
0.070
0.060
0.050
0.040 high lower high 0.030 upper middle
0.020
0.010
0.000 199 199 199 199 199 199 199 199 199 199 200 0 1 2 3 4 5 6 7 8 9 0
Figure II.9 Indices of structural deviation. End use of GDP, International Prices Source: ICP.
The index of structural deviations shows that, although the Russian structure by end use categories converged during the early or mid 1990s towards that of the three country groups, Russian GDP (by end use) was about as far from any of the three country group averages in 1999 as it was before the start of transition. We hope that a new wave of ICP results will be soon published for a later year (2003), which will allow us to determine whether Russia continues its lack of convergence towards other groups of countries. 1999 is far removed from an ideal year for study in that it immediately follows the currency crisis of August 1998. The lack of convergence in the composition of Russian GDP by end use in international prices does not necessarily imply a reversal of structural adjustment. In fact, the structure of Russian GDP has moved from one extreme to another. The major contribution to this pattern is made by the rapidly declining share of investment. Figure II.10 shows that investment declined in Russia by 1993 temporarily bringing its share in GDP down to more “normal” rates, but the continuing decline thereafter led to increasing deviations. The investment collapse may be explained by overlapping transformational and recessionary effects. The lack of commitment of the Russian government of the 1990s to the creation of a secure business environment meant that the collapse of domestic investment could not be partially offset by increasing foreign direct investments as in Central and Eastern Europe. From the available data, we cannot tell which of the two factors dominated and whether Russian investment will settle at a “normal” level after recovers to the pre-transitional level. Gregory: Russian Structural Change 5/28/2004
35%
30%
25% High 20% Lower-high 15% Upper middle Russia 10%
5%
0% 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Figure II. 10. Shares of gross capital formation in GDP, PPP-based. Source: WDI.
Another factor responsible for the lack of convergence between Russian and market economies is Russia’s large and growing government. Figure II.11 shows that the share of government consumption in Russian GDP exceeded average for all reference groups. As in the case of investment, there is no reason to believe that government consumption is going to be reduced soon leading to normalization of GDP use proportions.
30%
25%
20% High Low er-high 15% Upper middle 10% Russia
5%
0% 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Figure II.11. Shares of government consumption in GDP, International Prices Source: ICP.
Figure II-12 computes the structural deviation index for the end-sue categories in terms of nominal domestic prices to capture both domestic relative scarcities and distortions introduced by subsidies or other distorting practices. These data are drawn from World Bank databases and should be conceptually the same as the ICP data in domestic prices for the benchmark years (1990, 1993, 1996, 1999); the advantage of the World Bank data is that they are annual. It should Gregory: Russian Structural Change 5/28/2004 be noted that economic decisions in countries are not based on international prices but on domestic prices. Hence, ICP domestic price data may be a better test of structural convergence. Figure II-12, unlike the previous chart, shows a substantial convergence between 1991 and 1993 towards all three country groups. Thereafter followed periods of slight divergence and convergence. From 1993 to 2000, Russian end-use patterns were most like those of the lower- high countries. After 1998, there was a reversal of convergence as a consequence of the currency crisis of 1998.
2000
1500 High "Low er High" 1000 Upper Middle Low er Middle
500
0 1990 1992 1994 1996 1998 2000 2002
Figure II.12. Indices of structural deviation. End use of GDP, nominal values based. Source: WDI.
Household consumption The ICP also provides also a window into the restructuring of final household consumption. Soviet consumption patterns were characterized by abnormally high proportion of goods (more than two thirds) and in particular by a high share of food expenditure, 43 to 47 per cent depending on the method of calculation.34 General tendencies in Russian final household consumption in comparison with market economies are summarized in Table II.4 for the year 1999. Figure II.13 provides the index of structural deviation with 1999 as the end point. Both the table and the figure show that Russia is close to an average upper-middle income country in its structure of final household consumption. At the same time, the gap between Russia and the more developed countries seems to be increasing. When assessing the vector of change, we should take into account that the latest ICP
34 Gertrude Schroeder. Soviet Consumption in the 1980s: A Tale of Woe. In: Michael Ellman and Vladimir Kontorovich (eds.), The Disintegration of the Soviet Economic System, Routlegde: London and New York, 1992. Gregory: Russian Structural Change 5/28/2004 data come from 1999, the aftermath of the financial crisis. Some of the features of Russian consumption pattern may reflect transitory effects of the 1998 crisis rather than a turn of the trend. The new round of ICP, for the year 2003, should be expected to provide a better picture of the direction of change in Russian GDP use structure. Table II.4. Components of final household consumption, percent. 1999, International Prices (Revised Numbers)
High Lower Upper Lower Russia high middle middle Food, beverages, tobacco 12 16 18 18 18 Clothing and footwear 4 4 2 4 4 Gross rents, fuel and power 19 17 28 25 28 Household equipment and operation 3 3 2 3 0 Medical care 17 15 22 15 21 Transport and communication 11 10 6 9 7 Recreation, education 20 24 32 24 18 Miscellaneous goods & services 16 15 9 9 5 Net purchases abroad -1 -3 -1 0 0
. . Figure II.13 Indices of Structural Deviation, Structure of Household Consumption, International Prices
Gregory: Russian Structural Change 5/28/2004
In sum, unlike patterns of value added and labor force distributions, which have been changing in the direction of reference countries (upper middle and lower-high income countries), the Russian pattern of consumption remains difficult to characterize. It has moved from being an over-investor and under-consumer to an over-consumer and under-investor. Patterns of household consumption behave even more erratically with domestic prices deviating significantly from international prices. If anything, Russian household consumption resembles the pattern of lower-middle and upper-middle income countries, but changes have been erratic. The apparent culprit is the lack of price reform in housing, education, medicine, and utilities, which means that Russian households are operating under a system of non-scarcity prices.
Foreign trade patterns The removal of the governmental monopoly over foreign trade with the transition to market economy ended the near-autarkic position of Russian economy. The increasing volume of foreign trade in conjunction with domestic price adjustment and changes in domestic demand has led substantial changes in the volume and structure of Russian exports and imports, as discussed in Part I. As shown in Part I, we lack foreign trade data for the entire transition period; the data we have covers sub periods and unsually starts in the mid 1990s. 35 Part I showed that Russian trade ratios (the ratio of exports plus imports to GDP) increased, that Russia became a major food importer and a raw material exporter, that Russia had a very large positive balance on current account, and that Russian had become major consumers of international travel. In this section, we try to place these developments in international context.
Table II.5. Foreign trade indicators: Russia vs. Country Groups Year Country Exports of goods and Imports of goods Exports of goods Imports of goods and group services per worker, and services per and services (per services (per cent of constant 1995 USD worker, constant cent of GDP) GDP) 1995 USD
1990 H 13406 13185 37.2 37.9
35 Data used here originate in the WDI database and come in their from Goskomstat publications. The latter started to collect foreign trade statistics using customs records only in 1994. Gregory: Russian Structural Change 5/28/2004
2001 28179 25392 51.6 47.0 1990 UM 3166 3238 37.1 38.9 2001 4535 4441 42.0 37.9 1990 LM 1231 1638 27.5 33.0 2001 1546 1625 34.2 36.7 1990 Russia 1037* 1143* 18.2 17.9 2001 1406 1126 36.8 24.1 Source: WDI. * We use 1992 figures because the earlier figures may be grossly distorted.
It should be recalled that the period 1990 to 2001 was one of great turbulence for Russia, with a major recession/depression characterizing the period 1993-1996. If the real GDP figures truly capture real income trends, the would not expect to see a large rise in imports. Indeed, imports per worker remained relatively static, while exports per worker (which depend not on domestic income but on foreign demand) increased. The volume of exports or imports per workers appear to be very close to that of a low middle income country – the income level at which Russia is calculated. The ratio of exports to GDP shows a real opening of the Russian export market, with the ratio of exports to GDP nearly doubling. The ratio of imports to GDP also rose by six percentage points, roughly rivaling the percentage point increase of the high income countries that were experiencing the gains of rapid globalization during this period. In international perspective, Russian exports ratios appear “normal” relative to other countries, but we require an econometric investigation to establish the parameters of “normality,”36
36 For example, large countries such as Russia tend to have lower trade ratios. We must therefore use econometric methods to adjust for size as well as for per capita income. Gregory: Russian Structural Change 5/28/2004
Table II.6 Structure of exports, percent of total exports. Goods Services
Year Country Food Fuel Manufact Ores and Communi Insurance Transport Travel Other group ures metals cations, & services services commerci computer financial al services , etc. services
1994 H 8.8 7.9 55.5 2.4 8.1 1.3 5.9 7.8 8.6 1996 8.4 8.4 56.0 2.4 8.0 1.2 5.3 7.3 8.6 2001 5.6 3.6 66.8 1.8 8.5 1.9 5.5 5.8 9.8 1994 UM 17.0 22.7 25.0 4.3 5.9 0.5 19.2 6.9 8.4 1996 18.1 23.8 26.2 4.4 4.6 0.8 17.3 6.6 7.7 2001 15.4 16.1 34.7 4.0 4.4 1.3 21.5 7.8 10.8 1994 LM 14.7 9.7 38.3 2.5 9.3 0.6 26.8 5.9 12.6 1996 13.9 15.2 33.0 2.2 9.5 0.7 26.4 5.1 12.7 2001 12.4 9.7 44.3 2.9 7.4 2.0 26.1 5.1 12.8 1994 Russia 2.3 0.1 4.3 2.7 2.4 1996 1.5 36.8 22.2 8.6 2.4 0.1 3.5 6.8 2.5 2001 1.3 48.7 19.8 7.2 2.0 0.2 4.1 3.3 2.0 Source: WDI
Table II.6 shows the structure of Russian exports in 1996 and 2001 and compares these structures with other income groups by assorted years. From the limited coverage of the data, we cannot obtain a good picture of change, but it is evident that Russia stands out by its failure to export food products, by its substantial fuel, ore and metals exports, and by its relatively low share of manufacturing exports. Table II-7 on the comparative structure of Russian imports shows that Russia stands out with its high proportion of food imports, and its low proportions of manufacturing imports, and its extremely low proportions of imports of financial services and finance. The weight of fuels (primarily oil) has been steadily growing since 1996. This development can be seen in favorable light as Russia exploiting its natural comparative advantage. This also reflects of low competitiveness of Russian manufacturing. This applies even to the traditionally more developed arms industry, whose weight in the total exports declined from about 6 per cent in the early 1990s to 3 per cent in the end of the decade. Gregory: Russian Structural Change 5/28/2004
Table II.7 Structure of imports, percent of total imports.
Goods Services
Year Country Food Fuel Manufact Ores and Communi Insurance Transport Travel Other group ures metals cations, & services services commerci computer financial al services , etc. services
1994 H 8.2 5.7 61.4 2.7 7.8 1.1 8.0 6.7 7.1 1996 7.7 5.9 62.4 2.5 7.8 0.9 8.0 6.2 7.0 2001 6.5 7.3 61.9 2.7 10.2 1.3 10.8 6.1 6.9 1994 UM 7.2 7.2 55.8 2.0 7.3 1.2 7.4 8.2 5.0 1996 8.6 9.3 62.4 2.1 6.8 1.5 7.6 6.9 5.3 2001 7.4 10.5 66.0 1.7 6.2 1.4 7.1 6.0 5.2 1994 LM 10.8 7.1 57.6 2.2 6.0 1.2 6.4 7.7 4.1 1996 12.0 8.4 57.4 2.4 5.4 1.3 5.9 7.8 4.2 2001 9.8 9.8 57.1 2.1 5.7 1.6 6.6 7.2 5.2 1994 Russia 6.7 0.2 6.9 3.9 9.2 1996 14.2 2.2 35.9 1.3 6.8 0.1 7.0 3.0 11.5 2001 14.7 1.8 45.3 1.8 7.3 0.8 7.1 4.0 13.8 Source: WDI
5000
4000
3000 High Upper Middle 2000 Low er Middle
1000
0 1995 1996 1997 1998 1999 2000 2001 2002
Figure II.14. Indices of structural deviation. Exports, based on 1995 USD. Gregory: Russian Structural Change 5/28/2004
Source: WDI.
2000
High
1000 Upper Middle Low er Middle
0 1995 1996 1997 1998 1999 2000 2001 2002
Figure II.15. Indices of structural deviation. Imports, based on 1995 USD. Source: WDI. Structure of imports and exports also makes Russian noticeably distinct from the reference groups. Moreover, index of structural deviation for exports (Figures II.14- II-15) shows that Russian economy actually diverges from all other groups, retaining the closest position to the upper-middle group. In terms of imports structure (Figure 3.4), Russian economy is closest to lower-middle income countries, while no strong convergence trend can be identified for the 1990s. Structural proximity of Russian imports to lower-middle income group is determined largely by relatively high shares of food imports – 13-14 per cent throughout the period of 1996-2001. Peculiarity of the composition of Russian exports is largely explained by heavy weight of raw materials: fuels 49% and ores 7% of total value of imports in 2001. Summing up, Russian foreign trade looks relatively backward in both scope and structure, although there has been a considerable opening of the Russian market to world trade. . Foreign trade indicators tie Russia to lower-middle income countries and probably low income. One possible explanation for this situation is export tariffs and protectionist policies of Russian governments. Moreover, the reliability of data also may be deficient, in particular because of underreporting of the volume of imports by tax-evading traders. We must be cautious in characterizing the structure of Russian trade because the trade of countries reflects their relative resource endowments, and it is difficult to find two sets of countries with exactly the same endowments. Clearly, Russian exports are dictated by its comparative advantage in raw materials, including oil, natural gas, and metals. Russia’s dependence on these types of exports is the result of natural market processes not of a specified state policy. The United States, for example, has a comparative advantage in agriculture, which is often characterized as a “backward” export; yet Gregory: Russian Structural Change 5/28/2004 no one claims that the U.S, trade pattern is “backward” because of it reliance on agriculture. It seems as if we must approach Russian exports in the same fashion.
Russian Structural Change: A Long-Run Perspective We have a relatively long time horizon for describing structural change in the area of the world that Russia now occupies. Given that Russia accounted for more than sixty percent of Soviet GDP, we can combine the Russian and Soviet experiences from the end of World War II to the present to determine if Russian structural changes can be divided into distinctive periods. Figure II.15 shows the annual sectoral growth rates of the Soviet/Russian economy for five major sectors (industry, construction, transport and communications, trade, and services) from 1950 to the present.37 Figure II.15 shows that the Soviet economy experienced a postwar restructuring as the new post-Stalin leadership took command. Trade and construction grew rapidly and the supply of neglected services improved somewhat. Starting in the early 1970s, the Soviet economy was frozen into a set distribution of resources in an extended period of “planning from the achieved level” during which sectors tended to grow at uniform rates, only with industry exhibiting variation in growth rates. . Unlike market economies, negative sectoral growth rates were rare. The only issue was how positive the growth rate would be, not whether it would be positive or negative. Figure II-15 shows that planning from the achieved level collapsed in the late 1980s as a number of sectoral growth rates became negative. The deep transition recession/depression, which began in the early 1990s, was headed by the collapse of construction, industry, and transportation and communication. It was this deep downturn that generated the major restructurings that are the subject matter of this study. The restructuring has continued into the early 2000s, and it remains to be seen what the new equilibrium will look like. The more market- like the future Russian economy, the more variation in growth rates as the economy responds to changes in tastes and technology and to external shocks, such as oil prices or changes in international financial markets and foreign direct investment flows.
37 The growth rates are are drawn from Joint Economic Committee, USSR: Measures of Economic Growth and Development, 1950-1980 (Washington, D.C., December 1982), Table A-1; World Bank and Goskmostat, Russian Federation: Report on National Accounts, October 1995, Table 2-2, Goskmostat, Natsional’nye scheta Rossi v 1995-2002 godakh, Moscow, 2003, Table 1.5. Gregory: Russian Structural Change 5/28/2004
Figure II-15: Annual Growth Rates of Sectors, USSR and Russia
0.3
0.2
0.1
0 industry construction transport and communication trade -0.1 services
-0.2
-0.3
-0.4
3 5 7 1 3 5 7 9 1 7 1 5 7 1 5 7 9 51 5 5 5 59 6 6 6 7 73 75 7 79 83 8 8 89 9 93 9 9 9 01 9 9 96 9 96 9 9 9 98 9 9 0 1 19 19 19 1 1 19 19 1 1 1 1 19 19 1 1 19 19 19 1 19 1 19 19 19 2 Gregory: Russian Structural Change 5/28/2004
Appendix A: Sources of Russian Data
All publications are updated annually. The latest available editions were of 2002. Table numbers and title are given as they appear in 2002 editions. A summary version of the Russian data (classified into 23 sectors) is found in the file AB-RussiantablesMay6.
Indicator Publication Table number (if applicable) and title Value Official Statistics of the Macroeconomic indicators/GDP by added; Countries industries of economy/ Gross value added by sector of the Commonwealth of
See also: Sodruzhestvo Russia/MAIN SOCIO-ECONOMIC Nezavisimykh INDICATORS/GROSS DOMESTIC Gosudarstv. PRODUCT BY Statisticheskii' BRANCHES OF CBNE ezhegodnik. (paper publication) Labor; Official Statistics of the Labour market/ by sector Countries Average number of employed persons on of of the Commonwealth of payroll of enterprises by industries of economy Independent States (CD- economy ROM); CIStat and STATPRO
See also: Russia/MAIN SOCIO-ECONOMIC Sodruzhestvo INDICATORS/DISTRIBUTION OF Nezavisimykh EMPLOYED Gosudarstv. POPULATION BY BRANCHES OF Statisticheskii' ECONOMY ezhegodnik. (paper publication) Gregory: Russian Structural Change 5/28/2004
Labor in Promyshlennost Rossii 1.6. ×ÈÑËÅÍÍÎÑÒÜ ÏÐÎÌÛØËÅÍÍÎ- industry; [Russian industry]; ÏÐÎÈÇÂÎÄÑÒÂÅÍÍÎÃÎ ÏÅÐÑÎÍÀËÀ ÏÎ by sector Goskomstat ÎÒÐÀÑËßÌ ÏÐÎÌÛØËÅÍÍÎÑÒÈ (òûñÿ÷ ÷åëîâåê) [Labor employed by industrial sectors; (thousands)]
Value Promyshlennost Rossii 1.13. ÂÀËÎÂÀß ÄÎÁÀÂËÅÍÍÀß ÑÒÎÈÌÎÑÒÜ added; [Russian industry]; ÏÎ ÎÑÍÎÂÍÛÌ ÎÒÐÀÑËßÌ ÏÐÎÌÛØËÅÍÍÎÑÒÈ (â shares of Goskomstat ôàêòè÷åñêè äåéñòâîâàâøèõ öåíàõ; â ïðîöåíòàõ industrial ê èòîãó) sectors, [Gross value added by major industrial sectors % (current prices; percent of total)]
Gregory: Russian Structural Change 5/28/2004
Appendix B. Russian sectoral data and international classifications.
Distribution of value added and labor across sectors of economy is available for all years from 1990 onward from the State Statistical Committee (Goskomstat) publications. Industry in the economy-wide tables has one aggregate entry. In addition, value added and labor data are available for the sectors of industry. Distribution of labor across industrial sectors is available in annual issues of Promyshlennost for all years starting with 1990. Distribution of value added is available from 1995 onward. Data on value added for 1990-4 were calculated from the data on the costs and accounting profits (by sector of industry) collected in Promyshlennost 1996. Classification used by Goskomstat in the economy-wide tables is given in Table A1. In addition, industrial statistics uses the breakdown of industry into 11 major sectors as shown in Table A2.
Table A1. Classification in Goskomstat economy-wide tables. Agriculture (includes fishery) Business services Construction (construction services only; building materials which is included in “industry”) Credit and insurance Education, culture and arts Forestry General government Geology and meteorology Health services, physical culture, and social security Housing, communal, and personal services Industry Information and computer services Real estate Science and scientific services Trade, catering, material and technical supply, and procurement Transport and communication "Other spheres of material production " (mostly small-scale consumer goods production not covered by “normal” industrial statistics)
Sums of labor by sectors reported in industrial statistics falls short of total industrial labor reported in economy-wide tables. For the years starting 1994, the discrepancy does not exceed 10% of total industrial labor. The source of this discrepancy is unclear. We include the residual labor in our tables as Industrial labor not included in the 11 sectors of industry (no match in value added data). Since value added distribution across industrial sectors is given with respect to the 11 sectors listed above (in percentages only, not in rubles), these distributions presumably do not take into account value added produced by the residual labor.
Gregory: Russian Structural Change 5/28/2004
Table A2. Sectoral breakdown in industrial statistics. Electricity Fuels (coal mining, extraction of oil, gas; primary oil processing) Ferrous metallurgy (iron ore mining and steel industry) Nonferrous metallurgy (other metals: mining and processing) Chemicals (includes petrochemical industry; excludes pharmaceuticals) Machinery Wood processing (includes furniture, mobile wooden homes) Construction materials (includes some quarrying) Light (textiles, leather, apparel) Food Other (includes pharmaceuticals, medical equipment, printing, glass and porcelain, and some other minor branches)
Composition of “Other” differs somewhat from table to table in Promyshlennost. There fore, it is not particularly reliable. However, it should not affect relative positions of the remaining industrial sectors, since its weight is relatively small: around five per cent of labor covered by industrial statistics. Note that “Other” in industrial statistics is different from “Other spheres of material production” reported in the economy-wide tables. For the latter, unlike the former, absolute data on both labor and value added are available. As far as international comparison is concerned, the major problem with sectoral breakdowns in Goskomstat industrial statistics is the inseparability of mining. The largest share of it is, however, located in the “fuels” category. Data on labor in iron ore mining and in quarrying for building materials production can be found in detailed tables in Promyshlennost (e.g. Table 1.4. in Promyshlennost 2000). However, no data on value added for this sub sectors are available. Consequently, mining sector has to be limited to “fuels” although the coverage of “mining” thus constructed is incomplete. For the purposes of comparison with international datasets using ILO classification, we use the following recoding scheme (Table A3). Note that due to the lack of separate data in Russian sources on public utilities ILO category “Electricity, Gas and Water” is filled with the data on gas and electricity industries only. Water services are included in Goskomstat data in “Housing and communal services…” Separate data on gas industry are, however, available from Goskomstat input-output tables. Goskomstat category “Science and scientific services” is placed in ILO residual category “Activities not Adequately Defined” since output of this quasi-sector belongs in parts to practically all other categories.
Table A3. Conversion of Goskomstat to ILO Gregory: Russian Structural Change 5/28/2004
ILO Goskomstat
Agriculture, Hunting, Forestry and Fishing agriculture + forestry
Mining and Quarrying fuels - gas + geology and meteorology
Manufacturing industry - electricity - fuels - construction mat + other kinds of activities of material production sphere Electricity, Gas and Water electricity + gas
Construction construction + construction materials
Wholesale and Retail Trade and Restaurants and trade, catering, material and technical supply and Hotels procurement Transport, Storage and Communication transport and communication
Financing, Insurance, Real Estate and Business information and computer services + Services credit and insurance + real estate + commercial market activities Community, Social and Personal Services general administration + education, culture and art + health services, physical culture and social security + housing and communal services and public utilities Activities not Adequately Defined Science and scientific services
For the purposes of comparison with international datasets using World Bank Development Indicators database (WDI), we aggregate ILO categories as Table A4 shows. Correspondence between sub sectors of manufacturing used in WDI – Chemicals, Machinery, Food, and Textiles – and Goskomstat industrial sectors is transparent. “Other Manufacturing” used in WDI equals “Manufacturing” introduced in Table A3 minus the four specific industrial sectors listed above.
Table A4. Conversion of ILO/ Goskomstat to WDI.
Gregory: Russian Structural Change 5/28/2004
WDI ILO Goskomstat
Agriculture Agriculture, Hunting, Forestry and agriculture + forestry Fishing Industry Mining and Quarrying fuels - gas + geology and meteorology Manufacturing industry - electricity - fuels - construction mat + other kinds of activities of material production sphere Electricity, Gas and Water electricity + gas
Construction construction + construction materials Services Wholesale and Retail Trade and trade, catering, material and Restaurants and Hotels technical supply and procurement Transport, Storage and Communication transport and communication
Financing, Insurance, Real Estate and information and computer services Business Services + credit and insurance + real estate + commercial market activities Community, Social and Personal general administration + education, Services culture and art + health services, physical culture and social security + housing and communal services and public utilities Activities not Adequately Defined Science and scientific services
The major difficulty in establishing the consistency of WDI data is that WDI contains no data on manufacturing in general in Russia, while the data on sub sectors of manufacturing are given only in percentages. However, data on the highest level of aggregation – Agriculture, etc.; Industry; Services - derived from Goskomstat data using recoding schemes in Table 3 and Table 4 are close to WDI data. This is no surprise given that WDI inputs come normally from national statistical agencies. For consistency, we replace WDI structural indicators with the numbers we obtain directly from Goskomstat data. Gregory: Russian Structural Change 5/28/2004
Appendix C
Guide to Russian International Data Base, 1990 to 20001 OVERVIEW
The data are divided into two general groups: 1) Recalculated (regrouped) data for the Russian economy for the period 1990 to 2001. 2) International data bases that include this Russian data as observations in international panel data sets. The Russian data are used both for international comparisons and for study of Russian trends for the period 1990 to 2001. It is expected that this data base will be periodically updated. Data sources are given and methodologies are described.
PART I: RUSSIAN DATA
All the Russian data are taken from Gosskomstat (GKS) publications, rather than from international publications that cite Russian data (which we found to be unreliable). The GKS data is useful in its own right for evaluating structural change, but it must be recoded to make it useful for international comparisons. Some weaknesses of the Russian data are described in Appendix A.
File #1: AS-GKS(Economy and Industry)
This file contains GKS data on value added and employment by 25 economic branches. GKS’s major weakness is that is does not provide a breakdown of industry into manufacturing, mining, and electricity, gas, and water. It does give breakdowns for manufacturing branches. This file reports the original GKS data broken down into 25 branches (See Table 1 below) and then regrouped into 28 branches, including 11 manufacturing branches) as in Table 2 below.
This file is of value in its own right because it provides data for calculating value added shares (in current prices) and labor force shares to show the extent of structural change using GKS’s own sector classifications. VA and LF are classified in the same way so that we can make relative productivity calculations by dividing the VA share by the LF share. Relative productivity values above unity suggest that the branch has a higher productivity than the economy-wide average. We supply a number of charts which show trends in structural change.
Gregory: Russian Structural Change 5/28/2004
GKS now uses standard value added concepts, and one can trace the derivation of VA through rather remarkable annual input-output tables. 38 GKS began including “unreported wages and salaries” starting in 1993 and GKS raised the value of the capital stock much slower than inflation after 1991. Both practices affect the changes in VA shares in the early 1990s, but they seem like reasonable adjustments. Otherwise, we did not find significant methodological changes.
The GKS sectoral codes are given in Table 1a for economy branches and for industry branches in Table 1b.
Table 1a: The GKS Value Added Categories (note only one category for “industry”) industry other services social organizations general government finance/credit/insurance/pension funds science and scientific services art and culture education health/sports/social security communal and personal services housing other industries of material production sphere geological research and meteo services general commercial activity serving functioning of market real estate information and computer center services
38 Natsional’nye scheta (National Accounts), and Sistema Tablits (System of Input- output Tables available in electronic form from www. Eastview.com, except for Promyshlennost. 1996 and Natsionalye scheta Rossii v 1989-1994 gg (GKS 1995 – Chapter on input-output tables)which were used in hard copy.
Gregory: Russian Structural Change 5/28/2004
procurement material and technical supply trade and catering communication transport construction agriculture fishing forestry Table 1b. Sectoral breakdown in industrial statistics. Electricity Fuels (coal mining, extraction of oil, gas; primary oil processing) Ferrous metallurgy (iron ore mining and steel industry) Nonferrous metallurgy (other metals: mining and processing) Chemicals (includes petrochemical industry; excludes pharmaceuticals) Machinery Wood processing (includes furniture, mobile wooden homes) Construction materials (includes some quarrying) Light (textiles, leather, apparel) Food Other (includes pharmaceuticals, medical equipment, printing, glass and porcelain, and some other minor branches)
File #2: AS-GKSVA-LF 28categories:
In this file, we have recoded the GKS VA and LF data into 28-Sectors by including 11 industrial branches and combining non-industrial branches according to the categories in Table 2. We report VA and LF and calculate shares and relative productivity.
Table 2: Recoding GKS into 28 sectors Economy total electricity fuels ferrous Gregory: Russian Structural Change 5/28/2004
nonferrous chemicals machinery forestry&wood processing construction mats light food other (pharmaceuticals, medical equip., printing, glassware, etc.) agriculture business services construction credit and state insurance education, culture and art forestry general government geology and meteorology health services, physical culture and social security housing, communal, and personal services information and computer services real estate science and scientific services trade, catering, material and technical supply and procurement transport and communication "other industries of material production sphere" industrial labor not included in the 11 sectors of industry (no match in VA data – See Appendix A for explanation
Data Sources for Russian Data
The GKS data are drawn from the following publications and web sites. All publications are updated annually. The latest available editions were 2002. Table numbers and title are given Gregory: Russian Structural Change 5/28/2004 as they appear in 2002 editions. Those updating the data should use these same sources. Unusually, we found the best organized LF data under the Russian data of GKS’s CIS statistics (cdRom). The standard GKS publications on labor (such as Trud) did not provide useful sectoral breakdowns. For more detail, see the table of sources in Appendix A. Much of the above data are available in the subscription statistical data base at www.eastview.com. We used as well the GKS January 2002 Web Database “Statistics of the CIS”, which we used in cdRom form. It is released annually. We also used Vienna Institute of Economic Research (WIIW), cdrom. “Countries in Transition, 2002 to update the Russian labor force statistics to 2001. (Table III2.8).
File #3: AS-RussiaILO
This file organizes the GKS data into the 11 International Labor Organization (ILO) sectors shown in Table 3 for use in international comparisons and as a convenient sectoral breakdown for assessing Russian structural change.
Table 3: ILO 11-sector Breakdown
Activities not Adequately Defined Agriculture, Hunting, Forestry and Fishing Mining and Quarrying Manufacturing Electricity, Gas and Water Construction Wholesale and Retail Trade and Restaurants and Hotels Transport, Storage and Communication Financing, Insurance, Real Estate and Business Services Community, Social and Personal Services
For those updating the data, Table 4 provides conversion tables to convert the GKS data into IOL categories.
Gregory: Russian Structural Change 5/28/2004
Table 4. Conversion of Goskomstat to ILO
ILO Goskomstat
Agriculture, Hunting, Forestry and agriculture + forestry Fishing Mining and Quarrying fuels - gas + geology and meteorology Manufacturing industry - electricity - fuels - construction mat + other kinds of activities of material production sphere Electricity, Gas and Water electricity + gas Construction construction + construction materials Wholesale and Retail Trade and trade, catering, material and Restaurants and Hotels technical supply and procurement Transport, Storage and transport and communication Communication Financing, Insurance, Real Estate information and computer services + and Business Services credit and insurance + real estate + commercial market activities Community, Social and Personal general administration + education, Services culture and art + health services, physical culture and social security + housing and communal services and public utilities Activities not Adequately Defined Science and scientific services
File #4: AS-Russia-USSR Comparisons
This file reports 1990 data for the Russian Republic when it was still a part of the USSR. The sources of this data are given in the file. The main reason for this calculation is to provide a Russian 1990 benchmark for the 1990 International Comparison Project (ICP), the 1990 version Gregory: Russian Structural Change 5/28/2004 of which included the USSR. This data will allow us to substitute Russian figures for the 1990 USSR figures in the 1990 ICP.
File#5: AS-Russianworksheets
This worksheet explains our calculations of industrial value added and labor force to obtain fuels, to divide up the industry sector, and to obtain VA data in current prices. (For the early period, GKS only reports VA in constant prices. Therefore we had to reconstruct VA data in current prices from the components of VA. We inserted these calculations into the WDI data base. We had to calculate the manufacturing share of GDP (which is not reported by GKS) and we must code the Russian manufacturing sectors in accordance with WDI ISIC codes. 39 We also checked the WDI data for the CEE countries with the WIIW Handbook of Statistics, Countries in Transition 2001 handbook and CDRom, and substituted the WIIW values for the Czech Republic, Bulgaria, Slovakia, Slovenia, Hungary and Poland.
PART II: ILO INTERNATIONAL DATA
File: AS-ILO Methodological description see Appendix A. Source of data: http://laborsta.ilo.org. For selected transition countries, we use CDrom 2002, WIIW: Handbook of Statistics (Vienna Institute of International Economic Studies), available for purchase with annual updates. Tables III-2 and III-3.
The ILO provides labor force distributions for 84 countries of all income categories for the 1990 to present period. It publishes its coding procedures (See table immediately above for ILO 10-sector categories). 40 Although the Russian labor force data reported by the ILO is supposed to be properly coded, we use our own coding as described in Appendix A. The ILO LF
.39 The ISIC codes are provided, for example, in Tables 4.2 and 4.3 in their World Development Indicators annual publication. 40 Countries are coded either as ISIC rev.2 or rev.3) and some countries are coded as both ISIC rev.2 and ISIC rev.3. We ran econometric models to uniformly code all countries according to rev. 2. For users who wish to update the series, we conclude that such econometric adjustments result in only minor changes. Hence, future updaters can feel free to use either rev.2. or rev.3 data. Appendix B explains the econometric adjustments. Gregory: Russian Structural Change 5/28/2004 data base is divided into 21 HH (high high income), 6 LH (low high income),41 19 Upper middle (UM), 30 lower middle (LM), and 8 low (L) countries. We report some data prior to 1990, but our main focus is 1990 to present. We interpolated and extrapolated missing values (reported in blue) when we felt there was sufficient data. Users can either use or ignore these extrapolated values. At the bottom of the file, we compare our own Russian estimates (described above) with data supplied by GKS to ILO.42 The comparisons (lines 927 to 958) show substantial differences between our GKS calculations and those of the ILO. We believe that our figures make more sense.43
The ILO data is arranged in a number of tabs: The annotations tab explains the methodology of creating a unified coding system. The ISIC rev2&rev3 tab provides the ILO data in its original format coded according to ISIC rev2, ISIC rev 3 or both. The simulated ISIC rev2 tab provides the ILO data coded according to the ISIC rev2 norm. Appendix B explains how these simulations were done. We use these data for our calculations. We provide two additional data tabs and five chart tabs to compare the Russian values with LH and UM averages. The most comprehensive chart compares the sum of squared deviations of the LH and UM averages from the Russian averages. These charts show a clear convergence of Russia towards to LH model. The other four charts compare the nine sector deviations of Russia from the LH and UM averages.
41 The LH category does not exist in World Bank country classifications. We have constructed it because of the possible parallels with Russia. There are six LH countries – Israel, Korea, Portugal, Greece, Spain, and Cyprus). We use the World Development Indicators grouping of countries in their “World By Income” tables. 42 In our correspondence with ILO about the Russian data they report, they claim that the Russian data has been recoded in a consistent fashion to conform to ILO standards. 43 The ILO version shows Russia starting out with an improbable 9 percent of LF in financial services, real estate and business services in 1990 declining to 4.5 percent in 2000. We show, on the other hand, an increase from less than one percent to almost 2.5 percent. These figures appear more plausible than the ILO figures. Unfortunately, there is no way to check the ILO figures that have been supplied directly by GKS. ILO will provide no information beyond the fact that the Russian data were provided by GKS and were coded according to ILO standards.
Gregory: Russian Structural Change 5/28/2004
An important note for those updating the ILO file: Although we provide an extensive statistical analysis to insure the use of uniform industrial classifications, our analysis shows that our adjustments are minor. Hence, those updating should feel free to use whatever classifications that is reported,
PART III: WORLD BANK WORLD DEVELOPMENT INDICATORS (WDI)
World Development Indicators: file:AS-WBdata Sources: The WDI data base was accessed through university data bases and is also available in CDroms. Information can be found at http://www.worldbank.org/data/wdi2003/
The WDI data cover 83 countries, which we break down into 20HH, 6LH, 14UM, 29LM, and 14L countries. According to WDI, Russia is a LM country. We use the Russian trade data reported by WDI without adjustment. We recode the other Russian GKS data as explained in Appendix A.
The GDPetc tab provides the VA distributions of GDP by major sectors (industry, manufacturing, services, agriculture and forestry). It breaks down manufacturing VA into 5 sectors (as a percent of manufacturing). Its weakness is its lack of detail on services, which are combined into one sector. WDI provides GDP by end use (Consumption, investment, exports and imports) with a breakdown by household and government consumption. It also provides certain expenditure items as a percent of GDP such as military and health spending. It also divides capital formation into gross fixed capital formation. It also provides other information on electricity production and use, inflation, employment and population, including scientific and engineering personnel. These latter variables could come in handy as explanatory variables (along with PPP GDP per capita and population) in subsequent analysis. The WDI data base provides relatively little useful information on labor force distributions.
. The ExIM tab gives exports and imports (valued in 1995$) on a per capita basis, a per workers basis, and as a percent of GDP, the structure of merchandise exports and imports (as a percent of total expenditures), 4-category breakdowns of merchandise exports/imports, and six- category breakdowns of service exports/imports. It also provides data on miscellaneous items such as arms exports/imports, high tech exports/imports, and tourism. As an exception, we have not calculated independently Russian trade data. Our research suggested that we could not improve over the data provided by the World Bank without a monumental data gathering effort. Gregory: Russian Structural Change 5/28/2004
Although the Russian trade data are relatively complete for the first bloc of variables (Exports/imports in dollars per capita, per worker and as a percent of GDP), the Russian data are largely missing for other variables for the period 1990 to 1993/5.
The file A-RusStru&TradeIndex(date) compares the Russian data with the HH, LH, UM, and LM countries with respect to the structure of GDP by major sectors (agriculture, industry, manufacturing, services), with respect to the structure of manufacturing, with respect to the volume of foreign trade (as measured by exports per capita in $ and exports as a percent of GDP). The trade tabs show that Russia’s trade volume as measured by exports per capita is like that of a low middle income country (as classified by GDP per capita), and that its exports as a percent of GDP has been subject to erratic shifts (as trade expands and GDP contracts and vice versa), such that at times, Russia’s trade ratios were above even the HH countries. The structure of trade comparisons (sums of squared residuals) cover a more limited period and do not yield conclusive result at this point.
A note for those updating the WB data file: The major problem is that Russia does not break down industry into manufacturing, mining, and electricity, gas, and water (See Appendix A). We provide our worksheet for making such adjustments in the fifth file under the Russian data section. These adjustments are no easy and may require some training.
PART IV: INTERNATIONAL COMPARISON PROJECT
FILE#1:AS-ICPdata90,93,96,99
Sources: The ICP data were provided in excel files directly by contacts at the World Bank and at the UNECE, and OECD. We can give no web sites or other official sources of information. The 1990 USSR data were supplied in hard copy form by your office. ICPs are conducted several years apart. The next ICP (for 2001?) will have to be added at that time point when it is available.
The ICP projects provide useful international comparison data for Russia. ICP data are available in current domestic prices and in dollars of PPP purchasing power. Thus they can be compared both in terms of percentage shares and in terms of absolute values (in dollars per capita). We received ICP data from World Bank sources (1996) and from UNECE and OECD for 1990, 1993, and 1999. Apparently a new round of ICP data should be available soon for 2000 or Gregory: Russian Structural Change 5/28/2004
2001. We have ICP data that directly includes Russia for 1993, 1996, and 1999. We have the 1990 USSR data from United Nations Statistical Commission, International Comparison of Gross Domestic Product in Europe 1990: Results of the European Comparison Program. The 1990 data are listed as a separate tab. The ICP file contains a large number of tabs for the raw data in current domestic prices and in PPP dollar prices for all the ICP categories. The aggregated tabs reduce the ICP categories to 53 and are more manageable for analysis. The tabs also contain a number of charts that compare Russia with the averages of various income groups. We provide only a few charts and comparisons in view of the complexity and depth of ICP. At this point, we examine only the structure of consumption in domestic prices and the structure of GDP (personal consumption, public consumption, construction, plant and equipment investment, and net exports).
File #2: AS-ICPanalysis
This file provides statistical comparison of Russia (USSST in 1990) with groupings of other countries with respect to the major GDP categories and with respect to consumption categories.
PART V: OECD AAND RUSSIAN DATA ON THE BREAKDOWNS OF GDP BY RESOURCE COSTS
File:AS-OecdGDPcostcom
Sources: We obtained this data directly from the OECD website (http://cs4- hq.oecd.org/oecd/selected_view.asp?tableId=567&viewname=ANAPart1Table31970). The Russian data are from Natsionaly’nye scheta and from Sistem Tablitz (www.eastview.com). They are also drawn from the hard copy of Natsionalynye scheta Rossii v 1995-2002 godakh (GKS 2003, Tables 1.3).
The OECD provides breakdowns of GDP by compensation of employees and gross operating surplus. We compiled such data because they may shed light on Russia’s capital formation. The gross operating surplus equals depreciation plus profits plus proprietor’s income. This is a “mixed bag” category in that it includes proprietor’s income, but this category provides a starting point for international comparisons of Russia’s internal sources of investment finance. We also present some limited comparative data on Russia’s depreciation as a percent of GDP to obtain “net operating surpluses.” GKS now provides incredibly detailed data on the breakdown Gregory: Russian Structural Change 5/28/2004 of sectoral value added by compensation of employees, wages and salaries, taxes, and subsidies, and gross operating surpluses. GKS is beginning to publish data on depreciation. We can find data on depreciation for earlier years from the input-output tables in Sistem tablitz.
PART VI: RELATIVE WAGES AND PRODUCTIVITY: RUSSIA AND TRANSITION ECONOMIES
File: AS-relative_wages_jan04 Sources: CDrom 2001, WIIW: Handbook of Statistics (Vienna Institute of International Economic Studies), available for purchase with annual updates. Tables III-1, III-2, and VI.1. The Russian data are taken from Part I. The Russian data on relative wages are from www.eastview.com.
The client requested data on productivity and wages for transition countries including Russia to determine whether Russia was following the Polish-Hungarian-Czech model or the Romanian-Bulgarian model. We have assembled data on relative wages and relative productivity by ten sectors for those transition economies that have been coded according to ISIC classifications. The relative wage is the sector average wage (including social benefits) divided by the economy-wide average. Relative productivity is the sector share of VA divided by the sector share of LF. Two charts are supplied for Russia showing the sharp rise in relative wages and relative productivity in mining and the sharp fall in agriculture. The correlation tab correlates relative productivity to relative wages for those economies for which a sufficient number of observations exist. At this juncture, no serious analysis of these data has been undertaken. Gregory: Russian Structural Change 5/28/2004
Technical Report A: Russian Transition Comparative Dataset
Valery Lazarev
Note: There is considerable duplication between this report and Appendix B.
1. Russia. Value added (VA) and labor.
Source: Goskomstat.
Distribution of VA and labor across sectors of economy is available for all years from 1990 onward. Industry in the economy-wide tables has one aggregate entry. In addition, VA and labor data are available for the sectors of industry. Distribution of labor across industrial sectors is available in annual issues of Promyshlennost for all years starting with 1990. Distribution of VA is available from 1995 onward. Data on VA for 1990-4 were calculated from the data on the costs and accounting profits (by sector of industry) collected in Promyshlennost 1996. Classification used by Goskomstat in the economy-wide tables is given in Table 1. In addition, industrial statistics uses the breakdown of industry into 11 major sectors as shown in Table 2.
Table 1a. Classification in Goskomstat economy-wide tables. Agriculture (includes fishery) Business services Construction (construction services only; building materials which is included in “industry”) Credit and insurance Education, culture and arts Forestry General government Geology and meteorology Health services, physical culture, and social security Housing, communal, and personal services Industry Information and computer services Real estate Science and scientific services Trade, catering, material and technical supply, and procurement Gregory: Russian Structural Change 5/28/2004
Transport and communication "Other spheres of material production " (mostly small-scale consumer goods production not covered by “normal” industrial statistics)
Table 1b. Sectoral breakdown in industrial statistics. Electricity Fuels (coal mining, extraction of oil, gas; primary oil processing) Ferrous metallurgy (iron ore mining and steel industry) Nonferrous metallurgy (other metals: mining and processing) Chemicals (includes petrochemical industry; excludes pharmaceuticals) Machinery Wood processing (includes furniture, mobile wooden homes) Construction materials (includes some quarrying) Light (textiles, leather, apparel) Food Other (includes pharmaceuticals, medical equipment, printing, glass and porcelain, and some other minor branches) Sums of labor by sectors reported in industrial statistics fall short of total industrial labor reported in economy-wide tables. For the years starting 1994, the discrepancy does not exceed 10% of total industrial labor. The source of this discrepancy is unclear. We include the residual labor in our tables as Industrial labor not included in the 11 sectors of industry (no match in VA data). Since VA distribution across industrial sectors is given with respect to the 11 sectors listed above (in percentages only, not in rubles), these distributions presumably do not take into account VA produced by the residual labor. Composition of “Other” differs somewhat from table to table in Promyshlennost. There fore, it is not particularly reliable. However, it should not affect relative positions of the remaining industrial sectors, since its weight is relatively small: around five per cent of labor covered by industrial statistics. Caveat: “Other” in industrial statistics is different from "Other spheres of material production" reported in the economy-wide tables. For the latter, unlike the former, absolute data on both labor and VA are available. As far as international comparison is concerned, the major problem with sectoral breakdowns in Goskomstat industrial statistics is the inseparability of mining. The largest share of it is, however, located in the “fuels” category. Data on labor in iron ore mining and in Gregory: Russian Structural Change 5/28/2004 quarrying for building materials production can be found in detailed tables in Promyshlennost (e.g. Table 1.4. in Promyshlennost 2000). However, no data on VA for this subsectors are available. Consequently, we have to limit mining to “fuels” noting that the coverage of “mining” thus constructed is incomplete. For the purposes of comparison with international datasets using ILO classification, we use the following recoding scheme (Table 3). Note that due to the lack of separate data in Russian sources on public utilities ILO category “Electricity, Gas and Water” is filled with the data on gas and electricity industries only. Water services are included in Goskomstat data in “Housing and communal services…” Separate data on gas industry are, however, available from Goskomstat input-output tables. Goskomstat category “Science and scientific services” is placed in ILO residual category “Activities not Adequately Defined” since output of this quasi-sector belongs in parts to practically all other categories. 2. Russia. Other structural data.
Source: WDI (from Goskomstat).
Other Russian data included in the dataset is taken from WDI where it comes from Goskomstat. In particular, we compared the data on the uses of GDP in WDI with those in National Accounts published annually by Goskomstat and found no discrepancies.
3. Russia. Consumption data. From RLMS
Source: Russia Longitudinal Monitoring Survey (RLMS).
The separate file with the data on the structure of household consumption in Russia uses the data generated by subsequent rounds of Russia Longitudinal Monitoring Survey (1992-2002). Specifically, summary data were drawn from: Mroz, T., L. Henderson, M. Bontch-Osmolovsii, and B.M. Popkin. "Monitoring Economic Conditions in the Russian Federation: The Russia Longitudinal Monitoring Survey 1992-2002." Report submitted to the U.S. Agency for International Development. Carolina Population Center, University of North Carolina at Chapel Hill, North Carolina. March 2003. The same research team will probably continue to review the results of future rounds of RLMS.
4. Central and Eastern Europe. Gregory: Russian Structural Change 5/28/2004
Sources: WDI, WIIW.
Data on transitional economies of Central and Eastern Europe were mostly taken from WDI database. In a number of cases, when structural data (shares of VA) were missing or incomplete in WDI, additional data from WIIW database were used. Cells that are filled with WIIW data are grayed in our data file.
Gregory: Russian Structural Change 5/28/2004
Technical Report B:
ILO Labour Force Distribution Database - Generation Report
(Onno Hoffmeister, October 17, 2003)
The database is generated from the ILO Laborsta Database. It shows the distribution of the labour force on 10 sectors of the economy in 92 countries from 1969 to 2001.
A major difficulty consists in the fact that in the Laborsta Database the sectors are not coded uniformly. Some are coded according to ISIC Rev. 2, others according to the ISIC Rev. 3 classification. Both keys are available only at the one-digit level. ISIC Rev. 3 has 18 categories, ISIC Rev. 2 has 10. I used the correspondence list available at the ILO website, in order to summarise Rev. 2 and Rev. 3 codes into comparable groups (see table 1).
Table 1: Classification system of the generated activity groups using ISIC Rev. 2 and ISIC Rev. 3 codes ISIC ISIC Rev. Rev. 2 3 Agriculture, Hunting, Forestry and Fishing 1 A, B Mining and Quarrying 2 C Manufacturing 3 D Electricity, Gas and Water 4 E Construction 5 F Wholesale and Retail Trade, Restaurants, 6 G, H Hotels Transport, Storage and Communication 7 I Financing, Insurance, Real Estate, Business 8 J, K Services Community, Social and Personal Services 9 L, M, N, O, P, Q Activities not Adequately Defined 0 X
Gregory: Russian Structural Change 5/28/2004
The Excel-sheet called "original ISIC Rev. 2 & Rev. 3" shows both ISIC Rev.2 and ISIC Rev.3 numbers for each year.
A 100% equivocal linkage between ISIC Rev. 3 and ISIC Rev. 2 codes is not possible. For example some subcategories of the ISIC Rev. 3 category C should be converted into category 3, but the identification of cases belonging to this subcategory is not possible at the given aggregation level. Thus, generating the activity codes from ISIC Rev. 3 leads to a slightly biased employment distribution. Table 2 shows, into which possible ISIC Rev. 2 categories a worker, classified according to ISIC Rev. 3, may be converted. "X" shows the Rev. 2 category which applies for the overwhelming majority of ISIC Rev. 3 subgroups. These where used for transforming the ISIC Rev. 3 codes into the generated categories. "(X)" shows the remaining ISIC Rev. 2 groups which apply for some ISIC Rev. 3 categories.
Gregory: Russian Structural Change 5/28/2004
Table 2: Possible migration from ISIC Rev. 3 to ISIC Rev. 2 categories ISIC Rev. 2 ISIC Rev. 3 A, B X) X) C X) X) D X) E F X) G, H X) I J, K X) X) X) X) X) L, M, N, O, P, Q X)
Although Rev. 2 and Rev. 3 are mostly overlapping, there are some subcategories which according to Rev. 2 would need to be coded in a different way compared to Rev. 3. Thus the Rev. 2 are slightly biased from Rev. 3.
In the Excel-sheet called "simulated ISIC Rev. 2" all ISIC Rev. 2 all ISIC Rev. 3 codes have been converted in ISIC Rev. 2 codes. For this purpose the bias between both keys was estimated. The simulated data were produced in three different ways, depending on the data source available for particular countries:
1.) Original ISIC Rev. 2 codes (Simulation Status: 3) Whenever the Laborsta Database provided ISIC Rev. 2 data these data were copied into the simulated database.
2.) Simulated ISIC Rev. 2 using the common Rev.2- Rev.3 bias (Simulation Status: 1) Gregory: Russian Structural Change 5/28/2004
For countries, which show no year with both ISIC Rev. 2 and ISIC Rev. 3 numbers, the bias between both keys has been estimated using overlapping years in the whole dataset. All cases in which ISIC Rev. 2 and ISIC Rev. 3 data are available for the same country and the same year were pooled, and the relationship between both was estimated. From table 2 it is clear that each number ISIC Rev. 2 number may be influenced by more than one Rev. 3 number. For example the number of Rev. 2 code 5 is deemed to be influenced not only by the Rev. 3 number F, but also by the sum of A/B and by C. Thus we can write a system of 9 seemingly unrelated equations:
ISIC2(1)=c(1,1)*ISIC3(1) ISIC2(2)=c(2,2)* ISIC3(2) ISIC2(3)=c(3,2)*ISIC3(2) + c(3,3)*ISIC3(3) + c(3,8)*ISIC3(8) ISIC2(4)=c(4,1)*ISIC3(1) + c(4,4)*ISIC3(4) … ISIC2(9)=c(9,8)*ISIC2(8) + c(9,9)*ISIC2(9)
where ISIC2(i) is the size of the labour force in sector i according to ISIC Rev. 2, ISIC3(j) is the size of the labour force in sector i according to ISIC Rev. 3, c(i,j) measures the influence of ISIC3(j) on ISIC2(i), i={1;…;9}, j={1;…;9}.
The coefficients c(i,j) can be estimated using Zellner's seemingly unrelated regression system. c(4,1), c(6,3) and c(7,6) showed no significant influence at the 10%-level and therefore dropped out of the model. The seemingly unrelated regressions were run with the remaining coefficients. The results are summarized in appendix 1.
The simulated ISIC Rev. 2 codes (ISIC2s(i)) were obtained by multiplying the vector of corresponding ISIC Rev. 3 codes (ISIC3) with the ith row of the matrix (c(1,1); …; c(i,j)).
3.) Simulated ISIC Rev. 2 using the Rev.2- Rev.3 bias of the individual country((Simulation Status: 2) For countries which show ISIC Rev. 2 and Rev. 3 codes in same years the bias between both keys was estimated using only the numbers of the overlapping years in the individual country. However, the number of cases was to low to obtain reliable results by a multivariate regression system as in the case 2.) above. Instead only the one ISIC Rev. 3 category, which Gregory: Russian Structural Change 5/28/2004 covers the overwhelming majority of subcategories associated with the respective the ISIC Rev. 2 category ("X", not "(X)" in table 2), were used as the independent variables:
ISIC2(1)=c(1)*ISIC3(1) ISIC2(2)=c(2)* ISIC3(2) ISIC2(3)=c(3)*ISIC3(3) ISIC2(4)=c(4)*ISIC3(4) … ISIC2(9)=c(9)* *ISIC2(9)
where c(i) measures the influence of ISIC3(i) on ISIC2(i), i={1;…;9}.
When there occurred at least two overlaps the coefficients c(i) were estimated by univariate OLS regressions. The results for the first three countries in the alphabet, for which the number of overlaps is more than one, are summarized in appendix 2. The results of the remaining ones are similar (t-values > 20; R2>0,99). When there occurred only one overlap, c(i) was calculated by division of ISIC2(i) through ISIC3(i).
Gregory: Russian Structural Change 5/28/2004
Appendix 1
Seemingly unrelated regression ------Equation Obs Parms RMSE "R-sq" chi2 P ------activa1 79 1 8.062373 1.0000 3.55e+06 0.0000 activa2 79 1 8.693406 0.9895 8557.96 0.0000 activa3 79 3 82.74974 0.9978 59908.19 0.0000 activa4 79 1 14.04304 0.9686 3388.39 0.0000 activa5 79 4 34.71097 0.9970 38162.85 0.0000 activa6 79 2 42.71638 0.9996 225594.66 0.0000 activa7 79 2 30.22396 0.9967 29029.87 0.0000 activa8 79 3 53.16444 0.9962 26194.34 0.0000 activa9 79 2 189.2483 0.9942 15855.59 0.0000 ------
------| Coef. Std. Err. z P>|z| [95% Conf. Interval] ------+------activa1 | activb1 | .9989007 .0005303 1883.72 0.000 .9978614 .9999401 ------+------activa2 | activb2 | .9571929 .010347 92.51 0.000 .9369131 .9774726 ------+------activa3 | activb2 | .4030307 .1371082 2.94 0.003 .1343035 .6717578 activb3 | .9905895 .0062495 158.51 0.000 .9783408 1.002838 activb8 | -.0649915 .0190449 -3.41 0.001 -.1023188 -.0276641 ------+------activa4 | activb4 | .9712553 .0166854 58.21 0.000 .9385525 1.003958 ------+------activa5 | activb1 | .0218591 .0040338 5.42 0.000 .013953 .0297652 Gregory: Russian Structural Change 5/28/2004
activb2 | -.2586852 .1021159 -2.53 0.011 -.4588286 -.0585418 activb5 | .9101379 .0162808 55.90 0.000 .878228 .9420478 activb8 | .0307132 .0153303 2.00 0.045 .0006664 .06076 ------+------activa6 | activb6 | 1.004325 .0041846 240.01 0.000 .9961233 1.012527 activb8 | -.0456364 .0099224 -4.60 0.000 -.065084 -.0261887 ------+------activa7 | activb7 | 1.082798 .0163353 66.29 0.000 1.050781 1.114814 activb8 | -.108343 .0100451 -10.79 0.000 -.128031 -.088655 ------+------activa8 | activb5 | .1510003 .0116011 13.02 0.000 .1282626 .1737379 activb8 | .8663478 .010226 84.72 0.000 .8463051 .8863905 activb9 | -.0224451 .0036879 -6.09 0.000 -.0296732 -.015217 ------+------activa9 | activb8 | .2279574 .0454302 5.02 0.000 .1389159 .3169989 activb9 | .9646412 .0177327 54.40 0.000 .9298858 .9993966 ------
Gregory: Russian Structural Change 5/28/2004
Appendix 2
Australia i c(i) t- R2 observation statistics s 1 1.005195 251.47 0.9999 7 2 1.086444 101.49 0.9994 7 3 0.9939649 283.07 0.9999 7 4 0.9943264 211.89 0.9999 7 5 1.006059 257.59 0.9999 7 6 1.00749 151.90 0.9997 7 7 0.9994727 2822.88 1.0000 7 8 1.013118 182.81 0.9998 7 9 0.9311359 14.23 0.9712 7
Bolivia i c(i) t- R2 observation statistics s 1 0.9447056 22.04 0.9939 4 2 1.007228 35.36 0.9976 4 3 0.9443444 27.16 0.9959 4 4 0.9302157 6.66 0.9366 4 5 0.9318611 19.55 0.9922 4 6 0.9838816 22.42 0.9941 4 7 0.9595554 29.75 0.9966 4 8 0.9673495 18.52 0.9913 4 9 0.9642001 34.25 0.9974 4
Canada i c(i) t- R2 observation statistics s 1 1.010007 202.33 0.9998 11 2 0.9227804 145.66 0.9995 11 3 1.013555 119.96 0.9993 11 4 1.123327 62.59 0.9975 11 Gregory: Russian Structural Change 5/28/2004
5 0.966713 76.77 0.9983 11 6 0.9641897 118.82 0.9993 11 7 0.8380179 186.07 0.9997 11 8 0.8351807 148.01 0.9995 11 9 1.050927 128.59 0.9994 11
Dataset Source va-ggdc- Groningen Growth and Development Centre http://www.ggdc.net/i wiiw.xls (GGDC): 60 industry database dseries.htm Vienna Institute for International Economic Studies (CD-Rom submitted by Paul Gre (WIIW): Handbook of Statistics, Countries in Transition, 2002 expendit ICP (International Comparison Programme) (Excel sheets submitted by V ure.xls Laz lf-ggdc- Groningen Growth and Development Centre http://www.ggdc.net/i wiiw.xls (GGDC): 60 industry database dseries.htm Vienna Institute for International Economic Studies (CD-Rom submitted by Paul Gre (WIIW): Handbook of Statistics, Countries in Transition, 2002 lf-ilo- International Labour Organization (ILO): Laborsta- http://laborsta.ilo wiiw.xls database Vienna Institute for International Economic Studies (CD-Rom submitted by Paul Gre (WIIW): Handbook of Statistics, Countries in Transition, 2002
Gregory: Russian Structural Change 5/28/2004
Table 1-1(A) Percentages of Gross Value Added 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 electricity 1.56% 1.63% 3.34% 3.86% 4.15% 3.46% 4.72% 5.17% 5.19% 3.11% 2.74% 2.82% fuels 2.60% 4.30% 7.70% 6.54% 5.21% 6.00% 6.12% 6.32% 5.70% 7.02% 9.10% 7.80% ferrous metals 1.37% 1.78% 2.19% 2.19% 2.04% 2.21% 1.72% 1.46% 1.75% 2.28% 2.33% 1.73% nonferrous metals 1.32% 2.12% 1.87% 2.02% 1.74% 2.39% 1.75% 1.72% 2.78% 3.88% 3.91% 3.08% chemicals 1.30% 2.50% 1.77% 1.65% 1.66% 1.83% 1.42% 1.31% 1.48% 1.72% 1.61% 1.38% machinery 12.18 11.08 7.40% 8.16% 7.73% 5.50% 6.56% 5.83% 5.85% 5.85% 5.58% 5.93% % % wood processing 2.85% 2.80% 1.90% 1.81% 1.79% 1.51% 0.98% 0.91% 1.09% 1.48% 1.39% 1.15% construction materials 1.73% 1.85% 1.16% 1.38% 1.50% 1.40% 1.42% 1.17% 1.03% 0.77% 0.76% 0.81% light 7.32% 4.88% 0.74% 0.70% 0.48% 0.73% 0.53% 0.43% 0.42% 0.40% 0.38% 0.35% food 2.65% 3.86% 2.36% 3.10% 3.13% 2.82% 3.12% 2.97% 3.59% 3.23% 2.71% 2.91% other 2.95% 1.37% 3.38% 3.02% 3.47% 1.25% 1.34% 1.29% 1.30% 1.05% 1.01% 0.83% agriculture 16.44 13.87 7.14% 8.02% 6.32% 7.04% 7.14% 6.48% 5.68% 7.56% 6.60% 6.87% % % business services 0.00% 1.36% 0.13% 1.29% 0.69% 1.27% 1.19% 1.56% 0.95% 3.07% 3.16% 2.91% construction 9.51% 9.43% 6.30% 7.94% 9.13% 8.53% 8.45% 7.99% 7.13% 6.15% 7.24% 8.24% education, culture and art 5.36% 3.82% 2.88% 3.78% 3.95% 3.80% 4.29% 4.76% 4.42% 2.94% 2.75% 2.86% finance/credit/insurance 0.85% 2.21% 4.61% 5.16% 4.44% 1.57% 0.57% 0.73% 0.52% 0.90% 1.25% 1.92% forestry 0.08% 0.11% 0.12% 0.15% 0.18% 0.15% 0.18% 0.14% 0.13% 0.11% 0.11% 0.11% general government 2.82% 2.45% 2.14% 3.14% 4.74% 5.26% 5.22% 6.25% 6.83% 4.93% 4.72% 4.83% geology and meteorology 0.00% 0.00% 0.31% 0.19% 0.24% 0.24% 0.28% 0.35% 0.32% 0.31% 0.35% 0.36% health, physical culture social security 2.75% 2.75% 1.68% 2.70% 3.22% 2.93% 3.20% 3.77% 3.10% 2.42% 2.15% 2.21% housing, communal, and personal services 3.77% 2.50% 1.94% 3.27% 3.59% 5.41% 6.08% 5.95% 5.51% 3.07% 2.73% 2.83% information and computer services 0.22% 0.09% 0.07% 0.07% 0.10% 0.07% 0.08% 0.07% 0.09% 0.08% 0.08% 0.10% real estate 0.00% 0.07% 0.23% 0.39% 0.56% 1.38% 1.25% 1.78% 2.75% 3.46% 3.18% 3.53% science and scientific services 2.72% 2.07% 0.94% 1.11% 0.98% 0.81% 1.04% 0.92% 0.89% 0.82% 0.90% 0.81% trade, catering, supply and procurement 6.02% 12.27 29.01 19.06% 18.46% 19.78% 18.40% 17.70% 19.76% 23.19% 23.64% 23.02% % % transport and communication 9.99% 7.47% 7.42% 8.66% 9.92% 11.92% 12.38% 12.32% 11.01% 9.71% 9.12% 10.12% "other industries of material production sphere" 1.26% 1.18% 1.26% 0.63% 0.56% 0.72% 0.56% 0.63% 0.74% 0.50% 0.53% 0.53% social organizations (no match in labor data) 0.37% 0.19% 0.18% 0.21% 0.41% 0.37% 0.39% 0.78% 0.99% 0.18% 0.15% 0.19% Sum of sectors (own calculations) 100% 100% 100% 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00% 100.00 % % % % % % % %
Gregory: Russian Structural Change 5/28/2004
Table 1-1B Percentage of Labor Force Shares Labor Force shares 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 electricity 0.72% 0.76% 0.87% 0.94% 1.04% 1.13% 1.20% 1.25% 1.32% 1.38% 1.42% 1.46% fuels 1.06% 1.10% 1.21% 1.25% 1.26% 1.27% 1.30% 1.27% 1.24% 1.15% 1.13% 1.25% ferrous 1.04% 1.05% 1.10% 1.11% 1.08% 1.09% 1.10% 1.06% 1.05% 1.06% 1.11% 1.13% nonferrous 0.65% 0.68% 0.74% 0.76% 0.75% 0.83% 0.81% 0.79% 0.75% 0.79% 0.87% 0.90% chemicals 1.39% 1.51% 1.59% 1.57% 1.48% 1.35% 1.29% 1.27% 1.21% 1.19% 1.22% 1.23% machinery 12.80 12.31 12.16 11.20 10.26% 9.25% 8.48% 8.09% 7.57% 7.33% 7.32% 7.21% % % % % forestry&wood processing 2.38% 2.34% 2.52% 2.32% 2.24% 2.08% 1.91% 1.76% 1.62% 1.65% 1.71% 1.64% construction mats 1.46% 1.44% 1.58% 1.55% 1.52% 1.46% 1.32% 1.21% 1.12% 1.12% 1.06% 1.05% light 3.04% 2.90% 2.56% 2.40% 2.34% 2.00% 1.72% 1.56% 1.39% 1.35% 1.32% 1.26% food 2.05% 2.08% 2.16% 2.20% 2.27% 2.27% 2.25% 2.25% 2.19% 2.25% 2.31% 2.32% other 1.29% 0.93% 1.30% 1.34% 1.24% 1.35% 1.26% 1.16% 1.17% 1.18% 1.19% 1.17% agriculture 12.91 13.18 14.03 14.26 15.01% 14.67% 14.04% 13.28 13.67% 13.28% 13.01% 11.90% % % % % % business services 0.00% 0.00% 0.19% 0.61% 0.51% 0.57% 0.57% 0.75% 0.78% 0.79% 0.78% 0.80% construction 11.98 11.49 10.94 10.08 9.91% 9.34% 8.91% 8.76% 7.98% 7.95% 7.78% 8.79% % % % % credit and state insurance 0.53% 0.60% 0.68% 0.82% 1.09% 1.23% 1.21% 1.20% 1.15% 1.16% 1.15% 1.27% education, culture and art 9.60% 9.85% 10.44 10.22 10.78% 11.01% 11.09% 11.04 11.02% 11.04% 10.91% 11.44% % % % forestry 0.32% 0.32% 0.33% 0.34% 0.36% 0.39% 0.37% 0.37% 0.37% 0.38% 0.37% 0.38% general government 2.40% 2.33% 2.11% 2.33% 2.42% 3.03% 4.24% 4.26% 4.68% 4.89% 5.00% 4.87% geology and meteorology 0.43% 0.41% 0.40% 0.36% 0.31% 0.29% 0.29% 0.27% 0.28% 0.27% 0.27% 0.27% health services, culture, social security 5.63% 5.83% 5.87% 5.99% 6.42% 6.69% 6.87% 6.83% 6.99% 7.03% 7.00% 7.05% housing, communal, and personal services 4.27% 4.28% 4.15% 4.21% 4.41% 4.48% 4.86% 5.19% 5.34% 5.26% 5.16% 5.28% information and computer services 0.24% 0.18% 0.16% 0.13% 0.11% 0.12% 0.10% 0.18% 0.16% 0.17% 0.19% 0.19% real estate 0.00% 0.00% 0.00% 0.01% 0.02% 0.03% 0.05% 0.09% 0.12% 0.18% 0.25% 0.29% science and scientific services 3.72% 3.75% 3.20% 3.16% 2.68% 2.54% 2.30% 2.21% 2.04% 1.89% 1.87% 1.91% trade, catering, supply and procurement 7.79% 7.62% 7.88% 9.00% 9.47% 10.05% 10.30% 13.49 14.59% 14.57% 14.65% 14.64% % transport and communication 7.72% 7.79% 7.81% 7.63% 7.82% 7.91% 7.92% 7.92% 7.60% 7.69% 7.79% 7.98% "other industries of material production sphere" 2.17% 2.03% 2.13% 1.50% 1.11% 1.23% 1.36% 1.12% 1.03% 1.10% 1.23% 0.81% industrial labor not included in the 11 sectors 2.40% 3.24% 1.81% 2.74% 1.66% 1.77% 2.17% 1.39% 1.55% 1.91% 1.94% 1.51%
Gregory: Russian Structural Change 5/28/2004
Table 1-1C Relative Productivity Relative Productivity 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 electricity 216% 214% 385% 410% 400% 307% 394% 413% 393% 226% 193% 193% fuels 245% 390% 638% 523% 415% 471% 471% 498% 458% 608% 802% 623% ferrous 131% 170% 198% 197% 190% 202% 156% 138% 166% 216% 211% 153% nonferrous 205% 311% 253% 264% 231% 289% 215% 218% 369% 493% 449% 341% chemicals 93% 166% 111% 106% 113% 136% 110% 103% 122% 145% 132% 112% machinery 95% 90% 61% 73% 75% 59% 77% 72% 77% 80% 76% 82% forestry&wood processing 120% 120% 75% 78% 80% 73% 51% 52% 67% 89% 81% 70% construction mats 119% 128% 74% 89% 99% 95% 108% 97% 92% 69% 71% 77% light 241% 168% 29% 29% 21% 36% 31% 28% 30% 30% 29% 27% food 129% 186% 110% 141% 138% 125% 138% 132% 164% 144% 117% 125% other B216 229% 147% 260% 225% 281% 93% 106% 111% 111% 89% 85% 72% agriculture 127% 105% 51% 56% 42% 48% 51% 49% 42% 57% 51% 58% business services 70% 213% 135% 222% 208% 208% 122% 390% 407% 365% construction 79% 82% 58% 79% 92% 91% 95% 91% 89% 77% 93% 94% credit and state insurance 1005% 642% 420% 461% 363% 307% 354% 396% 383% 252% 239% 226% education, culture and art 9% 22% 44% 50% 41% 14% 5% 7% 5% 8% 11% 17% forestry 26% 34% 37% 45% 50% 39% 48% 38% 35% 30% 29% 30% general government 118% 105% 101% 135% 196% 174% 123% 147% 146% 101% 94% 99% geology and meteorology 0% 0% 79% 53% 77% 83% 99% 128% 116% 114% 130% 130% health services,culture,social security 49% 47% 29% 45% 50% 44% 47% 55% 44% 34% 31% 31% housing, communal, and personal services 88% 59% 47% 78% 82% 121% 125% 115% 103% 59% 53% 53% information and computer services 89% 52% 42% 56% 88% 59% 78% 40% 57% 45% 45% 53% real estate 3762% 3019% 4919% 2610% 1974% 2342% 1896% 1265% 1208% science and scientific services 73% 55% 29% 35% 37% 32% 45% 42% 43% 44% 48% 42% trade, catering, supply and procurement 77% 161% 368% 212% 195% 197% 179% 131% 135% 159% 161% 157% transport and communication 129% 96% 95% 113% 127% 151% 156% 156% 145% 126% 117% 127% "other industries of material production sphere" 58% 58% 59% 42% 51% 59% 41% 56% 72% 46% 43% 65% industrial labor not included in the 11 sectors 15% 6% 10% 8% 25% 21% 18% 56% 64% 9% 8% 12%
Gregory: Russian Structural Change 5/28/2004
Table 1-2 Value Added, Labor Force, and Relative Productivity, 8 sectors Major sectors shares of value added 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 industry excluding electricity and fuels 33.68% 32.23% 22.77% 24.04% 23.56% 19.65% 18.85% 17.09% 19.28% 20.66% 19.66% 18.15% electricity and fuels (incl.geology, meteorology) 4.16% 5.93% 11.36% 10.59% 9.60% 9.71% 11.12% 11.84% 11.21% 10.43% 12.19% 10.97% agriculture 16.53% 13.98% 7.27% 8.17% 6.50% 7.19% 7.32% 6.62% 5.81% 7.67% 6.70% 6.98% construction 9.51% 9.43% 6.30% 7.94% 9.13% 8.53% 8.45% 7.99% 7.13% 6.15% 7.24% 8.24% transport and communication 9.99% 7.47% 7.42% 8.66% 9.92% 11.92% 12.38% 12.32% 11.01% 9.71% 9.12% 10.12% trade 6.02% 12.27% 29.01% 19.06% 18.46% 19.78% 18.40% 17.70% 19.76% 23.19% 23.64% 23.02% business, financial, information services 1.06% 3.74% 5.04% 6.92% 5.79% 4.29% 3.10% 4.15% 4.31% 7.50% 7.67% 8.46% PA-HEAS 17.42% 13.59% 9.89% 14.18% 16.72% 18.45% 20.12% 22.01% 21.07% 14.49% 13.60% 13.89% shares of labor force 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 industry excluding electricity and fuels 26.09% 25.24% 25.70% 24.43% 23.17% 21.69% 20.15% 19.13% 18.08% 17.91% 18.11% 17.91% electricity and fuels (incl geology, meteorology) 2.22% 2.28% 2.47% 2.55% 2.60% 2.70% 2.78% 2.79% 2.84% 2.80% 2.82% 2.99% agriculture 13.23% 13.50% 14.36% 14.60% 15.37% 15.05% 14.42% 13.65% 14.05% 13.66% 13.38% 12.28% construction 11.98% 11.49% 10.94% 10.08% 9.91% 9.34% 8.91% 8.76% 7.98% 7.95% 7.78% 8.79% transport and communication 7.72% 7.79% 7.81% 7.63% 7.82% 7.91% 7.92% 7.92% 7.60% 7.69% 7.79% 7.98% trade 7.79% 7.62% 7.88% 9.00% 9.47% 10.05% 10.30% 13.49% 14.59% 14.57% 14.65% 14.64% business, financial, information services 0.78% 0.78% 1.04% 1.57% 1.73% 1.95% 1.94% 2.22% 2.21% 2.30% 2.37% 2.55% PA-HEAS 28.03% 28.25% 28.05% 27.53% 27.93% 29.10% 30.81% 30.83% 31.25% 31.38% 31.35% 31.56% relative productivity 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 industry excluding electricity and fuels 129.1% 127.7% 88.6% 98.4% 101.7% 90.6% 93.6% 89.3% 106.6% 115.3% 108.5% 101.3% electricity and fuels (incl geology, meteorology) 187.5% 260.3% 459.2% 415.2% 368.5% 360.1% 399.5% 424.0% 394.6% 373.0% 431.7% 367.0% agriculture 124.9% 103.5% 50.6% 56.0% 42.3% 47.8% 50.8% 48.5% 41.4% 56.2% 50.1% 56.9% construction 79.4% 82.0% 57.6% 78.8% 92.1% 91.2% 94.8% 91.3% 89.3% 77.3% 93.2% 93.8% transport and communication 129.3% 95.9% 94.9% 113.5% 126.9% 150.8% 156.4% 155.6% 144.8% 126.3% 117.0% 126.8% trade 77.3% 161.1% 368.2% 211.9% 195.0% 196.8% 178.6% 131.2% 135.4% 159.1% 161.4% 157.2% business, financial, information services 136.8% 481.5% 484.5% 440.8% 334.4% 220.1% 160.1% 186.7% 195.1% 326.4% 323.9% 332.3% PA-HEAS 62.1% 48.1% 35.3% 51.5% 59.9% 63.4% 65.3% 71.4% 67.4% 46.2% 43.4% 44.0%
Gregory: Russian Structural Change 5/28/2004
Table 1-3, Value Added, Labor Force, and Relative Productivity, Industry Industrial sectors (my estimates from promyshlennost) VA, % industry total 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 electricity 4.1% 4.3% 9.9% 11.2% 12.6% 11.9% 15.9% 18.1% 17.2% 10.1% 8.7% 9.8% fuels 6.9% 11.3% 22.8% 19.0% 15.8% 20.6% 20.6% 22.1% 18.9% 22.8% 28.9% 27.1% ferrous 3.6% 4.7% 6.5% 6.4% 6.2% 7.6% 5.8% 5.1% 5.8% 7.4% 7.4% 6.0% nonferrous 3.5% 5.5% 5.5% 5.9% 5.3% 8.2% 5.9% 6.0% 9.2% 12.6% 12.4% 10.7% chemicals 3.4% 6.6% 5.2% 4.8% 5.1% 6.3% 4.8% 4.6% 4.9% 5.6% 5.1% 4.8% machinery 32.2% 29.0% 21.9% 23.7% 23.5% 18.9% 22.1% 20.4% 19.4% 19.0% 17.7% 20.6% forestry&wood processing 7.5% 7.3% 5.6% 5.3% 5.4% 5.2% 3.3% 3.2% 3.6% 4.8% 4.4% 4.0% construction mats 4.6% 4.8% 3.4% 4.0% 4.6% 4.8% 4.8% 4.1% 3.4% 2.5% 2.4% 2.8% light 19.3% 12.8% 2.2% 2.0% 1.5% 2.5% 1.8% 1.5% 1.4% 1.3% 1.2% 1.2% food 7.0% 10.1% 7.0% 9.0% 9.5% 9.7% 10.5% 10.4% 11.9% 10.5% 8.6% 10.1% other 7.8% 3.6% 10.0% 8.8% 10.5% 4.3% 4.5% 4.5% 4.3% 3.4% 3.2% 2.9% industry total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
LABOR, th. Employed (promyshlennost) 1990 1991 1992 1993 1994 1995 1996 1997 1 998 1999 2000 2001 electricity 545 563 626 666 710 750 790 810 842 880 913 942 fuels 801 815 870 886 860 846 856 821 794 738 730 806 ferrous 785 772 795 788 738 727 727 683 673 676 711 727 nonferrous 487 502 532 542 517 549 537 508 480 503 560 582 chemicals 1049 1115 1143 1109 1011 895 852 822 775 761 785 795 machinery 9639 9093 8767 7933 7029 6149 5590 5231 4833 4688 4709 4642 forestry&wood processing 1792 1725 1813 1641 1535 1383 1261 1138 1034 1057 1102 1054 construction mats 1097 1067 1136 1095 1040 973 868 783 713 718 684 677 light 2288 2145 1845 1699 1600 1332 1133 1006 888 863 849 814 food 1545 1533 1554 1556 1554 1506 1487 1454 1396 1439 1484 1492 other 970 687 939 949 846 896 833 753 745 754 767 751 industry total 20998 20017 20020 18864 17440 16006 14934 14009 13173 13077 13294 13282
Labor Force Shares 1990 1991 1992 1993 1994 1995 1996 1 997 1 998 1999 2000 2001 electricity 2.60% 2.81% 3.13% 3.53% 4.07% 4.69% 5.29% 5.78% 6.39% 6.73% 6.87% 7.09% fuels 3.81% 4.07% 4.35% 4.70% 4.93% 5.29% 5.73% 5.86% 6.03% 5.64% 5.49% 6.07% ferrous 3.74% 3.86% 3.97% 4.18% 4.23% 4.54% 4.87% 4.88% 5.11% 5.17% 5.35% 5.47% nonferrous 2.32% 2.51% 2.66% 2.87% 2.96% 3.43% 3.60% 3.63% 3.64% 3.85% 4.21% 4.38% chemicals 5.00% 5.57% 5.71% 5.88% 5.80% 5.59% 5.71% 5.87% 5.88% 5.82% 5.90% 5.99% Gregory: Russian Structural Change 5/28/2004
machinery 45.90% 45.43% 43.79% 42.05% 40.30% 38.42% 37.43% 37.34% 36.69% 35.85% 35.42% 34.95% forestry&wood processing 8.53% 8.62% 9.06% 8.70% 8.80% 8.64% 8.44% 8.12% 7.85% 8.08% 8.29% 7.94% construction mats 5.22% 5.33% 5.67% 5.80% 5.96% 6.08% 5.81% 5.59% 5.41% 5.49% 5.15% 5.10% light 10.90% 10.72% 9.22% 9.01% 9.17% 8.32% 7.59% 7.18% 6.74% 6.60% 6.39% 6.13% food 7.36% 7.66% 7.76% 8.25% 8.91% 9.41% 9.96% 10.38% 10.60% 11.00% 11.16% 11.23% other 4.62% 3.43% 4.69% 5.03% 4.85% 5.60% 5.58% 5.38% 5.66% 5.77% 5.77% 5.65% industry total 100.00% 100% 100% 100.00% 100.00% 100.00% 100.00% 100.00% 100.00% 100.00% 100.00% 100.00%
Relative Productivity 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 electricity 158.9% 152.3% 316.1% 317.3% 309.6% 254.0% 300.6% 313.0 % 269.1% 150.1% 126.7% 138.2% fuels 180.2% 276.7% 524.1% 404.5% 320.8% 389.7% 359.4% 377.1% 313.6% 404.0% 526.3% 446.6% ferrous 96.5% 120.7% 163.1% 152.3% 146.6% 167.3% 119.1% 104.6% 113.5% 143.2% 138.4% 109.6% nonferrous 150.8% 221.2% 207.8% 203.9% 178.6% 239.1% 164.1% 165.5% 252.5% 327.6% 294.4% 244.2% chemicals 68.7% 117.8% 91.5% 81.6% 87.2% 112.7% 84.1% 78.4% 83.3% 96.2% 86.4% 80.2% machinery 70.1% 63.9% 49.9% 56.4% 58.3% 49.2% 59.0% 54.6% 52.9% 53.0% 50.0% 58.9% forestry&wood processing 88.3% 85.0% 62.0% 60.4% 61.9% 60.2% 39.1% 39.4% 45.9% 59.4% 53.1% 50.4% construction mats 87.7% 90.8% 60.7% 69.1% 76.4% 79.0% 82.6% 73.4% 62.8% 45.5% 46.6% 54.9% light 177.5% 119.4% 23.9% 22.6% 16.1% 30.0% 23.7% 20.9% 20.8% 19.7% 18.8% 19.6% food 95.3% 132.0% 90.0% 109.2% 106.9% 103.1% 105.5% 100.2% 112.3% 95.4% 77.0% 89.9% other 168.9% 104.3% 213.3% 174.3% 217.2% 76.8% 80.7% 83.7% 76.0% 59.0% 55.5% 51.3% industry total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
Gregory: Russian Structural Change 5/28/2004
Table I-5 The structure of Russian GDP by End Use, current prices (percent) 1989 1990 1991 1992 1993 1994 1995 1996 1 997 1998 1999 2000 2001 2002 Personal consumption 45.6 47.7 41.4 33.7 38.4 39.6 50.3 50.4 52.6 56.8 52.2 45.2 47.7 49.8 Services by non-commercial 0.5 0.9 4.3 1.9 5.2 5.9 1.8 1.8 2 1.9 1.2 1.2 1.2 1.4 organizations Total, Personal consumption 46.1 48.6 45.7 35.6 43.6 45.5 52.1 52.2 54.6 58.7 53.4 46.4 48.9 51.2 personal consumption plus net exports 47 48.8 46 50 51.4 49.2 55.5 56.4 56.8 65.5 70.4 66.5 61.7 62 Government Consumption 19.8 20.9 16.9 14.3 17.2 21.7 19.1 19.7 21.1 19.1 14.6 14.9 16.2 16.9 Gross Investment 33.2 30.3 37.1 35.7 31.4 26.3 25.4 23.9 22.1 15.4 15 18.6 22.1 21.1 fixed capital 31.2 28.9 23.8 24.7 22.8 24.6 21 20.2 18.4 16.5 14.5 16.9 18.7 17.9 inventories 2 1.4 13.3 11 8.6 3.7 4.4 3.7 3.7 -1.1 0.5 1.7 3.4 3.2 net exports 0.9 0.2 0.3 14.4 7.8 3.7 3.4 4.2 2.2 6.8 17 20.1 12.8 10.8 imports 29.2 26.2 24.8 31.9 43.2 44.1 36.3 34.8 exports 25.9 22 22.6 25.1 26.2 24 23.5 24
Gregory: Russian Structural Change 5/28/2004
table I-6 Expenditure category Nominal Purchasing Nominal Purchasing Purchasing Nominal final 1990 1993 1996 1999 values of Power values of Power power expenditures, shares shares shares shares in final Parities final Parities parities for 1999, expenditures, (Austrian expenditures, (US final National in in in current mln. National shilling=1), mln. National Dollar=1), expenditure currency current current current rubles currency 1990 currency 1993 on GDP, rubles rubles rubles Units, 1990 Units, 1993 1996; US (USSR) dollar = 1
FINAL CONSUMPTION of POPULATION 602,504.0 0.03988 87,812,201.0 184.6 1,339,861,599.5 2,145.4 2,870,266.2 58.3% 51.7% 60.9% 59.7% (national) FOOD,BEVERAGES,TOBACCO 217,005.0 0.05573 30,527,972.0 298.8 453,357,730.1 4,117.7 1,110,738.4 21.0% 18.0% 20.6% 23.1% CLOTHING and FOOTWEAR 98,799.0 0.07858 11,808,951.0 399.9 151,282,146.0 3,717.9 327,761.2 9.6% 7.0% 6.9% 6.8% GROSS RENTS, FUEL and POWER 35,232.0 0.01767 8,767,321.0 84.6 117,076,770.8 776.8 165,819.7 3.4% 5.2% 5.3% 3.5% HOUSEHOLD EQUIPMENT and OPERATION 39,890.0 0.04478 3,787,651.0 310.6 66,420,643.2 4,682.9 40,986.2 3.9% 2.2% 3.0% 0.9% MEDICAL CARE 33,241.0 0.01726 5,671,612.0 51.5 112,341,416.3 788.8 210,925.9 3.2% 3.3% 5.1% 4.4% TRANSPORT and COMMUNICATION 36,596.0 0.04571 5,697,359.0 230.4 141,940,326.7 4,283.5 324,135.4 3.5% 3.4% 6.5% 6.7% RECREATION, EDUCATION 73,001.0 0.02638 10,544,030.0 116.9 155,532,958.2 1,082.5 275,407.6 7.1% 6.2% 7.1% 5.7% MISCELLANEOUS GOODS & SERVICES 68,741.0 0.03799 9,686,661.0 342.8 122,039,608.1 3,555.1 194,598.3 6.7% 5.7% 5.5% 4.0% NET PURCHASES ABROAD 0.0 0.06900 1,320,644.0 889.1 19,870,000.0 5,124.8 66,617.9 0.0% 0.8% 0.9% 1.4% COLLECTIVE CONSUMPTION of 107,296.0 0.02660 15,507,600.0 124.8 237,596,063.1 943.0 410,082.3 9.1% 10.8% 8.5% GOVERNMENT GROSS FIXED CAPITAL FORMATION 309,990.0 0.03510 38,656,757.0 458.0 446,507,009.0 4,169.8 690,141.8 30.0% 22.8% 20.3% 14.4% CONSTRUCTION 192,324.0 0.02334 25,180,150.0 383.5 296,387,941.7 3,223.4 347,060.6 18.6% 14.8% 13.5% 7.2% MACHINERY and EQUIPMENT 117,666.0 0.05482 13,476,607.0 543.9 155,936,324.8 5,712.8 431,199.7 11.4% 7.9% 7.1% 9.0% BALANCE of IMPORTS and EXPORTS -5,700.0 0.06900 13,200,000.0 889.1 86,852,000.0 5,124.8 824,388.1 -0.6% 7.8% 3.9% 17.2% GROSS DOMESTIC PRODUCT 1,033,222.0 0.03762 169,764,503.0 230.9 2,208.9 4,805,417.1 100.0% 100.0% 0.0% 100.0%
Gregory: Russian Structural Change 5/28/2004
Table I-6 (Revised) 1990 1990 1993 1993 1996 1996 1999 1999 domestic international domestic international domestic international domestic international prices prices prices prices prices prices prices prices
FOOD,BEVERAGES,TOBACCO 37% 26% 35% 21% 34% 18% 39% 18%
CLOTHING and FOOTWEAR 15% 8% 13% 6% 11% 7% 11% 4%
GROSS RENTS, FUEL and POWER 6% 14% 10% 22% 9% 24% 6% 28%
HOUSEHOLD EQUIPMENT and OPERATION 7% 6% 4% 3% 5% 2% 1% 3%
MEDICAL CARE 6% 13% 6% 23% 8% 23% 7% 21%
TRANSPORT and COMMUNICATION 6% 5% 6% 5% 11% 5% 11% 7%
RECREATION, EDUCATION 12% 18% 12% 19% 12% 23% 10% 18%
MISCELLANEOUS GOODS & SERVICES 11% 12% 11% 6% 9% 5% 7% 5%
NET PURCHASES ABROAD 0% 0% 2% 0% 1% 1% 2% 0%
had to adjust 1999 to equal 100 adjustment factor = .92
Gregory: Russian Structural Change 5/28/2004
Table I-6 Continued 1999ppp DATA Actual individual consumption 5.396271 Food and non-alcoholic beverages 11.1621 Alcoholic beverages, tobacco, 9.254816 narcotics Clothing and footwear 15.6847 Gross rents, fuel and power 1.03077 Health 1.763052 Transport 9.501623 Communication 10.23273 Recreation and culture 9.122166 Education 0.976667 Restaurants and hotels 12.86514 Miscellaneous goods and services 7.232799 Net purchases abroad 24.62 Actual collective consumption 1.735831 Gross fixed capital formation 8.300115 Construction 4.477723 Balance of exports and imports 24.62 Gross Domestic Product 5.413536