Report No. 45486-RU Russian Federation Regional Development and GrowthAgglomerations The LongerTerm Challenges of EconomicTransition in the Russian Federation A Country Economic Memorandum for Russia January 16, 2009 Poverty Reduction and Economic Management Unit Europe and Central Asia Region Document of the World Bank CONTENTS EXECUTIVESUMMARY ............................................................................................................ i 1. REGIONALECONOMICGROWTHINRUSSIA: AN ASSESSMENT OF RECENT TRENDS ................................................................................................................................. 1 B EvolvingPatterns of Regional Growth............................................................................... A. Introduction........................................................................................................................ . 1 D. A StatisticalTest for Convergence..................................................................................... C. Do Growth Patterns inRussian Regions Favor Convergence or Divergence?...................4 3 . 8 F. Conclusion ........................................................................................................................ E The Geographic Distributionof Growth............................................................................. 9 14 2. THE SOURCES OFREGIONALGROWTHAND AGGLOMERATIONEFFECTS15 A. RelatedLiterature onRegional GrowthinRussia andAgglomeration Effects................15 B. UnitofAnalysis andPPP Correction............................................................................... 16 D. Spatial InteractionsandMarket Potential......................................................................... C. Trends inthe Regional Variation of GRP......................................................................... 16 E. StatisticalAnalysis: A Summary ofthe Results............................................................... 20 F. Conclusion........................................................................................................................ 23 26 3. RAPIDREGIONALDEVELOPMENTINRUSSIA:THE CASES OFROSTOVAND TVER .................................................................................................................................... 29 A. Economic Background..................................................................................................... B. The Business Survey ........................................................................................................ 29 C. Conclusion ........................................................................................................................ 32 44 4. CONFRONTINGTHE CHALLENGESOFREGIONALDEVELOPMENTINTHE RUSSIANFEDERATION .................................................................................................. 47 A. Recent Initiatives inRussiaRelatedto Regional Development....................................... 47 B. A Typology ofPolicies for Affecting RegionalDevelopment......................................... C. A National Regional Strategy for Russia?........................................................................ 49 58 LIST OF TABLES Table 1.1: Dispersion of Industrial GrowthAcross Russian Federation(RF) Regions..................3 Table 1.2: Convergence: Industrial Production and InitialIncomes ............................................. 8 Table 1.3: The Dynamics of Regional GrowthinRussia............................................................. 11 Table 1.4: The Distribution o f Growthby Okrug ......................................................................... 13 LIST OFFIGURES Figure 1.1: Industrial Growth. 1999-2003 ...................................................................................... 4 Figure 1.2: Industrial Growth, 2004-2006 ...................................................................................... 4 Figure 1.3: Industrial Growthvs InitialGRPPer Capita, 1999-2003 ............................................. 5 Figure 1.4: Industrial Growthvs InitialGRP Per Capital, 2004-2006 ........................................... 6 Figure 1.5: Industrial Growthvs Initial GRP Per Capita, 1999-2006............................................. 6 Figure 1.6: Retail Growthvs InitialGRP Per Capita, 1999-2003 and 2004-2006 ........................ 7 Figure 1.7: Growthof Manufacturing, 2005-2006 ........................................................................ 7 Figure 1.8: Growthof Construction, 2005-2006 and Growthof Other Paid Services, 2005-20067 Figure 1.9: Partial Correlation o f Industrial Growth, 1999-2003 and 2004-2006 ......................... 9 Figure 1-10: Industrial Growth Rates at the Okrug Level ............................................................ 10 Figure 1.11: Weighted Average of GrowthRates of Regions Within EachFederal Okmg.... ..-10 .. Figure 2.1: Average Annual Regional Growth in GRP Per Capita in the Russian Federation: 1999-2004 ................................................................................................................. 17 Figure2.2: NominalandReal Coefficients ofVariation ofRussianRegional GRP ... ...............18 Figure 2.3: 1999-2004 Per Capita GRP GrowthinRussianRegions: Leaders and Lagers .......19 Figure2.4: The "Market Potential" ofRussianRegions: 1999-2004......................................... 21 Figure 2.5: Spatially Determined Growthinthe Russian Federation: 1999-2004 ..................... 26 Figure3.1: Net Percentage of Tver Entrepreneurs Claiming that the Business Environment in their Region is Better (Positive) or Worse (Negative) than ina Given Other Region .................................................................................................................................. 33 Figure3.2: Net Percentageof Rostov Entrepreneurs Claiming that the Business Environment in their Regions is Better (Positive) or Worse (Negative) than ina GivenOther Region .................................................................................................................................. 33 Figure 3.3: The Share of EntrepreneursIndicating that a GivenProblemis a Serious Impediment to Business (Tver (2007), Rostov (2007), Russia (2005)) ........................................ 34 Figure 3.4: The Share of Medium and Large Enterprises in Tver and Rostov that Indicated a GivenConstraint is a Serious Impediment to their Business.................................... 35 Figure3.5: Perceptions of Rostov Entrepreneurs o f Changes in the Business Climate in the Regions During 2005-2006 (Net Percentage Indicating Worsening (Negative) or Improvement (Positive) ............................................................................................ 36 Figure 3.6: Perceptions of Tver Entrepreneurs of Changes in the Business Climate in the Regions During 2005-2006 (Net Percentage Indicating Worsening (Negative) or Improvement (Positive) .........., ..............,..........,......................................................37 . Figure 3.7: The Share of Respondents in Rostov and Tver that Indicate a Given Administrative Barrier i s a Significant Impediment to Business (% ofResponses).......................... 38 Figure 3.8: The Share (%) of Entrepreneurs in Rostov, Tver, and Russia (2005) that Believe it Possible to Overturn a Decision in Violation of Norms or Laws made by a State Bribery...................................................................................................................... Official Through an Appeal to Another Body or Higher Authority Without Resort to 39 Figure3.9: The Net Assessments of Rostov and Tver Entrepreneurs on the Influence of Different LevelsofGovernment on Their Business ................................................. 39 Figure 3.10: Perceived Competitive Pressures in Rostov and Tver: The Share (%) of Respondents Indicating that Competitive Pressures from a Given Source are Significant ..,. ...,.....................,,...........,.................., ...,.,,..,, ...,,...,..,..,..,..,..,, ..,.,,.,.. 1 ,, ,, , 4 Figure3.11: The Share (%) of Respondents in Rostov Engaging in a Given Activity Differentiatedby Responsesto Questions on Competitive Pressures . ..................,..42 Figure 3.12: The Share (%) of Respondents in Tver Engaging in a Given Activity Differentiated by Responsesto Questions on Competitive Pressures.............................................. 42 Figure3.13: Competitionand Growth inRostov and Tver Oblasts: The Share (%) o f Firms that Recently Expanded Output Differentiated by Responses to Questions on CompetitivePressures ............................................................................................... 43 Figure 3.14: Competitionand Investment in Rostov and Tver Oblasts: The Share (%) of Firms that Recently Expanded their Capital Differentiatedby Responses to Questions on CompetitivePressures............................................................................................... 44 Figure4.1: ProjectedPercentage Change inWorking-Age Populationby RegioninRussia: 2006- 2026 .......................................................................................................................... 52 LIST OF BOXES Box 1.1: A Note on the Data........................................................................................................... 3 Box 3.1: RostovOblast ................................................................................................................. 30 Box 3.2: Tver Oblast ..................................................................................................................... 31 Box 4.1: World Development Report2009: Territorial Development........................................ 49 APPENDIX ................................................................................................... 60 REFERENCES ............................................................................................... 61 ACKNOWLEDGEMENTS This Country Economic Memorandum (CEM) was drafted by Task Team Leader, John Litwack on the basis o f the following contributions: Chapter 1 was written by Faruk Khan and John Litwack, with assistance from Sergei Ulatov and Olga Emelyanova. Chapter 2 i s based on work performed by Evgeny Polyakov, with the assistance of Oleg Lugovoi, and Denis Fomchenko under cooperation between the World Bank and the Moscow Institute for the Economy in Transition, and was co-financed by the Canadian InternationalDevelopment Agency (CIDA) 0 The surveys in Tver and Rostov for Chapter 3, together with the basic analysis of the results, were led by Victoria Golikova and Boris Kumetsov. The survey work in the respective regions was conducted by Maksimum-Inform (Tver) and the Southern Russian Research Center FACTOR (Rostov). Other contributions to this chapter came from Sergei Ulatov and Olga Emelyanova. 0 Chapter 4 was written by John Litwack. Background information for migration was providedby Timothy Heleniak. Technical support for this CEM was provided by Olga Emelyanova, Elena Lukyanchikova, Judy Wiltshire, Vladislava Nemova, and HelenaMakarenko. Peer reviewers for this report were Sergei Guriev, Mikhail Dmitriev, Chorching Goh, Deepak Bhattasali, and William Tompsen. Valuable feedback on earlier drafts o f this report also came from Asad Alam, Zeljko Bogetic, Marsha Olive, and the participants of a seminar devoted to the draft CEM held at the Moscow World Bank Office on July 17, 2008 (including Natalia Zubarevich, EvgenyiYasin, Leonid Grigoriev, Sergei Artobolevsky, Vladimir Nazarov, and Igor Pilinenko. Rosstat providedimportant support inthe provision o fregional data. EXECUTIVESUMMARY Introduction 1. Vast geographic and cultural diversity gives a unique dimension o f complexity to economic and political transformationinthe Russian Federation. The richnatural endowments o f the country are distributedhighly unevenly among Russian Subjects o f the Federation (regions). Seventy years of Soviet planning left a geographic distribution of settlements and capital that i s often poorly related to economic comparative advantage. This includes significant settlements and capital inextremely cold areas, where economic efficiency i s inherentlylow. The large size and ethnic diversity of Russia has always posed strong challenges for central governance, and policies of Moscow toward the regions can be associated with important political, as well as economic, complexities. 2. As Russia continues to make rapid general progress inbuildingmodemmarket institutions, the complex spatial dimension o f transformation in the country will continue to present strong policy challenges for many years to come. In fact, these challenges are arguably becoming progressively stronger. As confirmed inthis Country Economic Memorandum, economic growth has become muchmore unevenly distributed across the Russian Federation inrecent years, with a sharp contrast opening up between growth agglomerations in much o f the Western part o f the country and much more modest development in the East and North. A significantly decreasing and aging working population also imposes severe constraints to the sustainability o f growth in regions that are unable to attract a large number of migrants. Given limitations in the speed of migration, and significant uncertainty about the economic potential of many regions, the regional dimension of economic growth has major implications for sustaining the current strong momentum inpoverty reduction. 3. The Russian government is aware that it faces considerable future challenges inthe spatial dimensions of economic development. This was the motivation for creating a Ministry of Regional Development in 2004, and bestowing on this Ministry augmented authority and influence in 2007. Much time and effort in recent years has been devoted to the drafting o f a National Regional Strategy that would facilitate the adoption and coordination of policies in the area of regional and spatial development. However, considerable disagreements within the government and policy circles over the nature of this strategy have delayed its adoption. Primary sources of controversy include the degree to which the Strategy should focus on supporting growth points, as opposed to lagging regions, and on the particular policy roles o f the federal, regional, and local levels o f government. 4. This Country Economic Memorandum (CEM) focuses on regional development in the Russian Federation. It first conducts an examination of recent trends inthe regional dimension of growth based on available economic data for the country as a whole. Particular attention is given to the topics of convergence, agglomeration, and the sources of regional growth. In this endeavor, the analysis distinguishes between two quite different recent periods o f economic growth: 1999-2003 and 2004-2006. A regional-level survey allows for the examination of 1 perceptions of entrepreneurs working in two poorer-than-average regions that have been exhibiting impressive rates o f growth inrecent years: Rostov and Tver oblasts. The conclusions obtained inthese investigations are thenrelatedto the question of what type o f a national regional strategy, ifany, mightbe useful for Russia. 5. An examination of recent growth trends reveals important shifts in the pattern of regional growth inRussia. The regionally broad-based growth inthe years following the financial crisis of 1998-1999has beenreplaced by muchmore geographically uneven growth that i s concentrated in agglomerations inthe Western part of the country. As a number o f Western regions experiencing relatively rapid growth have below-median income levels, and growth has slowed in some resource-rich regions, regional growth patterns inRussia still have a strong pro-poor component. Yet, the emergence o f Western growth agglomerations and the growing East-West divide suggest strong challenges in managing the future regional dimension of economic development in the country. The problem of lagging regions inthe East should attract increasingpolitical attention. In the emerging growth agglomerations in the West, the sustainability of rapid growth is threatened by labor shortages, infrastructure constraints, and a protectionist lobby against growing competitive pressures from all-Russian and world markets. 6. Inthe light ofthese challenges, the CEMrecommends that Russia continueto contemplate a coordinated national regional strategy to meet these considerable challenges. This strategy should encompass four main tasks (a) the facilitation of movements of labor and capital to areas of greater economic promise (in particular, the current pace of migration appears to be inadequate), (b) the provision of opportunities for regional (local) administrations and entrepreneurs in lagging regions to realize their own initiatives for improvingthe local economic situation, (c) the maintenance o f minimal social and living standards across Russian regions, and (d) the provision ofkeyinfrastructureto better linkand integrate the national market. 7. Russia i s already pursuing a number of policies that are consistent with these goals. This includes some changes in migration policies, the defense o f a unified economic space in the country, the implementation of more effective formulae for intergovernmental transfers, the drafting of development strategies at the regional level consistent with federal priorities, and the general prioritization of infrastructure, human capital, and other productivity-enhancing investments over subsidies to low productivity firms. 8. Nevertheless, an effective national regional strategy for Russia will also require a shift of emphasis incertain areas. Migration and other relatedpolicies at the national and regional levels will most likely need to be adjusted to accommodate Russia's growing needs inlabor supply. An enormous degree of uncertainty exists about the economic potential o f many regions, and competition for labor and capital coming from emerging growth agglomerations in other regions can seriously complicate the realization of this potential. In this context, opportunities for genuine decentralized initiatives at regional and local levels can be critical for the successful economic development o f regions with genuine potential, as well as for economizing on the high social and economic costs o f mass internal migration. Given the strong continuing trends in Russia toward political centralization, supporting appropriate incentives for decentralized initiatives at the subnational level promises to be a challenge. On the one hand, a stronger orientation o f the regions toward Moscow creates enhanced opportunities for coordination and cooperation inthe realization o f regional initiatives. On the other hand, too strong a hand o f the center can also undermine incentives for local and regional officials to pushaheadwith their own initiatives or elicit the initiatives o f local entrepreneurs. A fundamental principle for the success of a regional development strategy in Russia for lagging regions i s that the system rewards innovation and genuine decentralized initiatives as opposed to passive obedience. World .. 11 experience provides a clear warning that highlycentralized attempts by the central government to choose future points o f growth or design its own detailed programs for implementation at the regional level are likely doomed to failure. RecentTrends inRegionalDevelopmentinRussia 9. Chapter One investigates trends in regional development based on output data, primarily for industrial production, which were available through 2006, with a goal o f distinguishing between the periods of 1999-2003 and 2004-2006 (The analysis in chapter two employs data on gross regional product (GRP), which were available only through 2004.). The motivation for distinguishing between 1999-2003 and 2004-2006 i s the perception that factors behind economic growth in Russia were quite different during these two subperiods. The former period witnessed a rapid economic recovery and import-substitution-based industrial expansion in the context o f a weak ruble, cheap labor, and substantial excess capacity. A slowdown in growth in 2001-2002 can be related to a weakening o f all of these factors along with economic recovery. In more recent years, economic growth received a strong boost from increases in export prices, most particularly oil and gas, and from booming domestic demand. The benefits o f these latter factors are distributed far more unequally across Russian regions than the factors that supported the earlier broad-based growth. This i s visible in industrial production data by region for these two sub-periods. The standard deviation in regional growth in Russia almost doubled in the latter time period (Figures A 1 and A2). The examination o f output data on retail trade volume yield very similar conclusions. FigureA1 FigureA2 I industrialGrowth 1999-2003: Industrial Growth 2004-2006: Disperslonacross FtF Regions Dispersion across RF Regions I 20 I I 20, R I I - 5 J 10. The greater dispersion in regional ind-stria1 growth in Russia in recent years can a h be associatedwith a growing gap between most of Western Russia (The Central, North-Western, and Southern Federal Okrugs) and the more Eastern part of the country (Urals, Siberia, Far East). While industrial growth slowed for Russia as a whole during2004-2006, it actually accelerated in muchofthe Western part ofthe country. iii Figure A3 Figure A4 IndustrialGrowth:2000-2003 by FederalOkrug Industrial Growth:2004-2006 by Federal Okrug (Annualized,%) (Annualized,"/p) a 6 4 2 0 Central West South Volga Vrak Siberia FarEast Central West South Volga Urals Siberia FarEast 11. The greater dispersion in Russian regional industrial growth favoring Western regions echoes concerns often expressed by specialists about growing regional inequality. However, a formal analysis of convergence gives no evidence that inequality among regions has increased, on average, during either the 2000-2003 or 2004-2006 periods. Despite a higher variance in regional growth, and a greater differentiation between the Eastern and Western parts of the country, it does not appear to be the case, in general, that poorer regions got poorer and richer regions got richer. Infact, many of the Westernregions that achieved higher industrial growth in the 2004-2006 period have lower-than-median average income (GRP per capita), while many o f the Eastern regions that achieved lower-than-average growth in the 2004-2006 have significant resource wealth and higher-than-average GRP per capita. A significant pro-poor component o f Russian regional growth patterns i s suggestedby Figures A5 and A6, which plot Russian regions according to initial GRP per capita (in log) and industrial growth for 1999-2003 and 2004-2006. Graphs that use retail trade or other service growth inplace o f industrialproduction look similar. Figure A5 Figure A6 industrial Growth vs initialGPP per capita1999-03 IndustrialGrowth VI initial GRPper capita2004-06 20 *1 f 15 8 f 10 E . ' 1 5 g -I YI O - * -5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 Q 10 11 12 13 Ln ( GRP per capita 1998) Ln ( GRPper capita2003 ) 12. An examination of the recent data also suggests growing agglomeration effects in the Western part of the country in the form o f spillover effects from proximity to other rapidly growing regions. The geographical correlation of growth i s most likely increasing. This i s particularly visible in the Central Federal O h g , where the number o f relatively poor regions iv achieving above-averagegrowth increased inthe 2004-2006 period relative to 1999-2003 jumped from 6 to 9. The question o fregional spillover effects i s examined indetailedinChapter 2. 13. To summarize the results of Chapter 1: While the pace of average industrial growth has fallen, the dispersion of growth among Russianregions has increased. Growth has become increasingly concentrated in the Western part o f Russia, particularly in the Central, North Western, and Southern Okrugs. In fact, industrial growth has actually accelerated in this part of Russia, in contrast to the trend for the country as a whole. Despite the greater dispersion in growth duringthe more recent period of 2004-2006, output data do not suggest any shift toward a trendof divergence among regions. This was due mostly to rapid growth in a number o f regions in the Central Okrug with lower than average per capita GRP, as well as slower growth ina number o f resource- richregions inthe Eastwithrelatively highper capita GRP. 0 The number of regions experiencing higher-than-average growth inthe Central Okrug i s increasing, consistent with the possible strengthening o f agglomeration (spillover) effects. Industrial and service output growth appears more regionally concentrated in 2004-2006 than in 1999-2003. 14. Chapter 2 explores the questions of sources o f regional growth and agglomeration effects in more detail on the basis of data on gross regional product (GRP) per capita corrected for regional price differentials. It employs a spatial econometrics approach that has been used in recent research for the analysis o f regional spillover effects on economic growth inthe European Union. This variable picks up geographically concentrated effects on economic growth that are not captured by other specified exogenous variables. The focus of Chapter 2 is on the period of economic growth from 1999-2004, as this is the period for which data on GRP were available. 15. The analysis reveals a strong significance of regional spillover effects in explaining regional GRP growth. Regions in close proximity to rapidly growing regions receive a significant boost intheir growth. This can be due to boththe advantages of proximity to booming markets as well as direct spillovers from the expansion of growth agglomerations. The estimated regional spillover effects are not quite as large for Russia as comparable measuresrecently made for the European Union. However, the analysis o f Chapter 1suggests that these effects may have become stronger in recent years. Figure A3 below illustrates the size of the estimated part of annual regional growth that i s induced from growth in regions in geographic proximity, i.e. it i s the component of regional growth that can be explained by growth inother regions. As indicated on the map, these effects are generally much stronger in the European part o f the country, where one percent of economic growth in other regions weighted for geographical proximity produces an estimated 0.3 to 0.5 percent growth ina givenregion. V Figure A3: Spatially Determined Growth in the Russian Federation Spatially growth rate: 1 ~ 16. The model in Chapter 3 also investigates the contributions o f a number o f factors to regional growth inthe Russian Federation. The results are summarized below: Initial Income and Convergence: Both with and without controlling for other variables, the level o f initial income in a region (GRP per capita) appears to have a negative and statistically significant effect on growth. This provides evidence for both absolute convergence (poorer regions are catching up on average) and conditional convergence (the relationshipremains significant, and becomes even stronger, when controlling for the investment rate and other variables). This i s consistent with the conclusions reached in Chapter 1. Investment Rate: As might be predicted, the investment rate (investment as a share of GRP) has a positive and statistically significant effect on growth, suggesting that regions which invest a higher share o f their income experience faster growth. Urban Aadomeration: The size (population) o f the main city in a region i s found to be positively related to growth and significant in a wide variety o f specifications. This i s robust to the removal o f Moscow and Saint Petersburg from the sample. Shares o f the regional population living in larger cities also appear to be positively related to growth, although their significance i s more variable over different specifications. However, the share o f the population living in the single largest city o f the region did not appear significantly related to growth statistics. Human Capital: The share o f populationwith higher education appears to have a positive and significant effect on growth, although this still statistically significant relationship weakens somewhat when Moscow and Saint Petersburg are dropped from the sample. vi Infrastructure: Measures related to infrastructure include stock variables in absolute and per capita terms (rail density, road density, length o f roads and rail, andphone lines) and quality variables (the share of hard roads and share o f improved hard roads). These measures of infrastructure intensity did not generally correlate well with regional growth. However, this does not imply that infrastructure i s not important for regional growth, as infrastructure intensity i s often not a very good proxy for infrastructure constraints, i.e. a lot of infrastructure does not mean that infrastructure constraints are not binding, and visa-versa. Investment Climate: Indicators of the investmentclimate based on an OPORA Survey of small businesses that i s carried out in the majority o f Russian regions did not emerge as significant in the conducted statistical estimations as an explanatory factor for either growth or investment. Part of the problem may concern the fact that these data were only available for 2006, and may be a very inaccurate proxy for business climate constraints at the regional level during 1999-2004. However, a regional investment risk index compiled by Expert did show a positive relationship to regional growth in some specifications. Ethnic fragmentation: The index of ethnic fragmentation (number and share in the population of ethnic minorities) can be related to regional growth in a negative and statistically significant manner. This effect i s robust to the inclusion of human capital in the specification and to the dropping of federal cities and southern republics from the sample. However, ethnic diversity tends to be correlated with other variables, such as resource endowments, which makes the interpretationof this result somewhat difficult. Fuel Endowment: A dummy variable was used for those regions with a fuel share of GDP exceeding 14 percent. For Russian regions, this cut-off point provides a very convenient bifurcation of hydrocarbon-oriented and other territories. Only a few regions are close to this 14 percent border line. This fuel endowment dummy variable has a positive and statistically significant effect on growth inthe 1999-2004 period, particularly in an equation that corrects for urban agglomeration and human capital. On the other hand, Chapter 1suggests that this dynamic may have weakened inmore recent years. Year-Round Port: The presence of a year-round port has a strong positive and statistically significant effect on regional growth. This effect is robust to various specifications that include or exclude investment, human capital, and other determinants of growth. Climate (Average Temperature, Permafrost): Of a number o f climate variables explored, a dummy variable used for regions with no permafrost (warmer regions) appears to have a positive and statistically significant relationshipwith growth. Rapid Regional Development inRussia: the Cases of Rostov and Tver 17. Chapter 3 turns to questions pertaining to the business climate and potential constraints to sustaining growth in two relatively poor Russian regions that have been experiencing rapid growth in recent years. Rostov and Tver are part of the story o f the strong pro-poor element in Russian regional growth examined in the previous two chapters. Rostov has been a center o f growth in the South of Russia since 1999, whereas growth inTver has accelerated only recently with the spread of the Moscow and Saint Petersburg agglomerations. A regional level enterprise survey conducted in both regions offers information on the perceptions and activities of vii entrepreneurs. For comparative purposes, the sample o f 535 firms in Rostov and 421 firms in Tver was conducted according to the same methodology as the national Business Enterprise Environment Performance Survey (BEEPS) that i s conducted in Eastern Europe and the CIS every three years since 1999. This survey gives proportional representation to different sectors and geographical areas, although small businesses o f less than 50 employees comprise a large share o fthe sample. 18. The surveys reveal some very interesting similarities, as well as differences, between the Tver and Rostov economies. Rostov has a particularly strong environment for small business and entrepreneurship, reflecting in part cultural traditions that were not entirely suppressed in the Soviet period. Larger businesses in the region often work in close partnership with authorities. InTver, on average, larger businesses function more independently. For somewhat different reasons, both regions are currently facing tensions related to the opening o f their regional economies to outside investment, both from abroad and from the growing Moscow-Saint Petersburg agglomerations. InRostov, perceptions o f close relations between incumbent firms and regionalflocal authorities may have limited the size o f outside investment relative to neighboring Krasnodar and other regions. In Tver, the booming economy o f Moscow and Moscow oblast, together with a new leadership in the region that has placed a priority on attracting outside capital, have increased outside investor interest in the region considerably. Some o f the effects o f this rapid process have been very disruptive for established local business, in particular the very rapid increase in property values and rentals. Inboth cases, sustaining rapid growth will depend on successfully navigating a difficult transition to a more open regional economy focused increasingly on all-Russian and export markets: The survey i s consistent with the picture o f Rostov and Tver as growing regions largely hospitable to expanding business activities. Perceptions o f barriers to business in these regions are consistent with, and somewhat more favorable, than those documented for Russia as a whole in 2005. Perceptions o f changes in the business climate inrecent years were not very favorable. Part o f the reason for this may be the increased competitive pressures that both regions have faced in recent years. The pessimism in both regions about progress in the development o f formal market institutions (the court system, competition policy, corruption) i s somewhat disturbing, although it follows the general pattern inrecent business surveys for Russia. 0 Inthe Tver region, perceptionsof thebusiness climate are fairly uniformacross firms o f different size. In Rostov, by contrast, smaller firms give significantly more favorable assessments o f the business climate than do larger firms. Medium and larger-sized firms in Rostov are much more concerned about business climate constraints than are their counterparts in Tver. Interestingly enough, both Tver and Rostov entrepreneurs complain about excessive paper work more than any other administrative barrier to business, including various inspections. Rostov entrepreneurs generally rank the business climate in their region favorably in comparison to surroundingregions, whereas Tver firms do not. 0 Although Rostov and Tver share the skepticism and mistrust o f their counterparts in other regions about the court system, an impressive number o f entrepreneurs in both regions do appeal to the courts for the settlement o f many disputes. The skepticism becomes greater for overturning violations o f laws or norms by government officials, particularly inRostov. Tver entrepreneurs are notably more optimistic in this regard. In general, Rostov and Tver entrepreneurs have very different perceptions of the influence o f various levels o f government on economic activity, with the balance o f influencedperceived as positive inRostov and negative inTver. V...l l l The results o f the Rostov and Tver surveys give strong additional evidence o f a positive relationship between competition (competitive pressures), on the one hand, and modernization (technical progress) and business expansion, on the other. Those f i r m s that claim to face significant competitive pressures from firms outside o f the region engage in significantly more activities associated with modernization, restructuring, and investment. 19. How should these results be interpreted for assessing the present and future prospects for the development of the Rostov and Tver regions? Both o fthese regions face social and economic tensions associatedwith rising outside competition. Despite the short-run costs of policies aimed at facilitating the integration of regional with larger markets, the survey provides direct evidence of potential benefits. It appears that many firms in both regions that offers the most hope for sustaining rapid growth (those actively engaged in modernization, restructuring, and output expansion) are already competing on all Russian or international markets. Rostov and Tver will need to manage the short-run transition costs, which are particularly notable in Tver, but heavy- handed regional-level protectionist policies would most likely end up rewarding less promising firms and hinderingthe activities of the most promising. 20. The survey is also consistent with the general picture ofRostov as aregionwhere informal relations can often be o f more importance than explicit institutions. At the same time that many entrepreneurs note improvements inrelations with authorities, there i s a widespread perception o f deterioration during 2005-2006 o f precisely those variables that relate to the functioning o f explicit institutions: unfair competitive practices, violations o f contractual agreements, criminality, corruption, and the court system. Attitudes are particularly pessimistic in Rostov about possibility of overturning unfair decisions or violations made by state officials. The sustainability o f Rostov's rapid growth into the medium term will most likely depend on the strengthening of explicit (impersonal) institutions relative to informal relations, which would make the region more attractive to outside investment and fair competition. Rostov's recent record in economic growth has been so far been much more impressive than its record in investment growth. This cannot be sustainedindefinitely, and outside capital i s needed. 21. Although the situation in Tver oblast looks to be somewhat different to that described above for Rostov, the future challenges are somewhat similar. An even larger share of firms in Tver look to be oriented only to local markets, and many operate under very limited competitive pressures. Furthermore, in sharp contrast to Rostov, Tver entrepreneurs consider the business climate in their region to be generally worse than in surrounding regions. This difference could be due as much to being surrounded by regions with better business climates as opposed to a worse business climate inTver than inRostov. Yet this i s part of Tver's dilemma. They are in a very advantageous geographical position for participating in the growing growth agglomeration incentralRussia,but so are anumber ofother regionsthat are indirect competition withTver for attracting capital and labor. Furthermore, many of these other regions got a head start on Tver by attracting substantial outside capital to their regions before 2005. This has made Tver's recent transition more radical and more disruptive to local interests than ina number o f other regions in central Russia. The political divisions within Tver oblast, and the lack o f a single omnipotent source o f state power like in Rostov, appears to have made even larger businesses in the region more independent, and more willing to rely on the courts and other explicit institutions than in Rostov. Tver firms were much less favorable intheir assessments of the influence o f government on their business, although these answers are difficult to interpret in the current period of transition when many firms believethat role o f regional and local authorities shouldbe protecting them from risingrentals or outside investors. ix 22. The rapidly-growing regions o f Rostov and Tver unquestionably have strong future potential. So far, both regions have grown in a largely self-contained manner in response to rising demand in the region and neighboring regions. The formation of agglomerations in the southern and central parts o f Russia has reinforced this growth. Yet both regions have yet to make the transition to developing highly competitive niches on all-Russian or world markets. The survey provides evidence suggesting that unleashing the potential of Rostov and Tver inthat direction will depend on promoting competition and providing a business climate that i s hospitable to market entry and investment. The cases o f Rostov and Tver also provide information on potential obstacles to sustaining rapid development in the growth agglomeration areas of the Russian Federation. In particular, they confirm labor supply and competition, includingthe openness ofregionalmarkets, as critical issues. Confrontingthe Challengesof RegionalDevelopmentinthe RussianFederation 23. Chapter 4 examines the question of a national regional strategy in the light of the conclusions of the previous chapters and international experience. Such a strategy should generally balance a number o f important economic and political priorities, including (a) the facilitation of growth points and agglomerations for increasing national economic growth, competitiveness, and living standards, (b) supporting minimal social standards inpoorer areas of the country, (c) providing opportunitiesfor laggingregionsto realizetheir economic potential, (d) increasing opportunities for migration from regions with poor economic prospects, and (e) promoting economic development in regions of particular political priority. National policies aimed at improving the overall business climate, supplying key public goods, and improving education have important general implications for realizing all o f these goals. In addition, the policy mix should generally include: Spatial policies. Policies to defend the unified market and facilitate the mobility o f labor and capital to areas o f agglomeration and greater economic promise. Interconnecting policies. Infi-astructure policies in areas such as transportation and telecommunications can link regional markets to the benefit o f economic development. (Social) equalization policies. Policies to ensure the provision o f minimal social services and support inpoorer regions. Regional (development) policies. Policies aimed at facilitating growth (productivity) inspecific regions. Education policies. Beyond the guarantee of minimal standards through equalization policies, additional investment (and reform) in education can increase overall economic opportunities for the population. Spatial Policies 24. Both the inherited Soviet legacy and the spatial nature o f recent growth in Russia imply that the attraction of people and capital to areas of agglomeration and greater economic promise i s perhaps the single most critical regional issue for the successful rapid development o f the country. Despite substantial internal and external migration to the Western regions o f Russia, and important progress in the developing o f markets to facilitate the movement o f capital, Russia nevertheless faces some significant challenges in sustaining and fostering growth agglomerations. Outside of Moscow and Saint Petersburg, Russia still lacks large cities by international standards, X and migration flows may not be adequate to compensate for an aging and declining population. Expert opinions suggest that the resettlement process from East to West inRussia has been rather slow, given the high differentials in living standards between different areas in the Russian Federation. Despite the relaxation of urban registration requirements and the presence of much greater regional inequality, the pace o f internal migration has not increased significantly in the post-Soviet period. Evidence exists that this may be partly due to poverty traps; many poor citizens inremote regions lack the resources to move to more promising areas. The current pace o f foreign migration, even accounting for the large number o f informal migrants, would not be sufficient to offset substantial population declines inmost o f the Western part of Russia over the medium term. The surveys in Tver and Rostov suggest growing constraints on growth from tightness inthe labor market. 25. Inevery country, large (voluntary) movements of people take time, and can be associated with substantial social and psychological stress. Nevertheless, the pace o f internal migration in Russia appears to be constrained by problems inhousing and labor markets. The housing market i s a particular constraint. The privatization process transferred the majority of housing to occupants, which became a windfall for those living in areas of economic promise, but has left many other citizens inremote regions without the means to afford increasingly expensive housing in areas of economic booms. The booming cities of Moscow and Saint Petersburg have faced serious challenges in increasing housing and other infrastructure in pace with economic growth. While the Russian labor market exhibits a significant amount o f (largely informal) flexibility, some experts suggest that important rigidities nevertheless remain that require attention. In particular, even after the adoption of new and more liberal Labor Code, there appear to remain excessive costs and hassles for employers associated with releasing workers, which find reflection in a common cautious attitude toward hiring. Important economic policies for the resolution o f the above issues include priority infrastructure projects for growth points to accommodate inflows of migrants (including affordable housing), labor market policies, and direct programs to assist (potential) migrants. 26. Another important policy direction concerns the enforcement of the unified economic space of the Russian Federation (competition policy) to combat regional protectionism that can limit competition and restrict the free movement of capital and labor to the detriment of sustaining rapid growth. The cases of Rostov and Tver, as examined inChapter 3, are instructive of the political and social pressures surrounding the opening up o f Russian regions, many of which have been very internally oriented in the past. Those regions also provide evidence that competitive pressures and competition on all-Russian or foreign markets can be associated with more technical progress and investment at the firm level. 27. Like a number o f other countries that attract a substantial number of foreign migrants, Russia must weight difficult social, political, and economic issues while adopting policies inthis area. ConsideringRussia's demographic situation, the country is quite fortunate that its relatively highstandard of living inthe greater region makes it a magnet for migrants within the CIS. An aging and shrinkingpopulation makes Russia comparable to the countries o f Western Europe in this regard: there is a strong need for migrants to augment the labor force, but social tensions place limits on policies to liberalize the (formal) migration regime. In this regard, it should be noted that the balance of net advantages to Russia from external migrants may be greater than those in Western European countries for several reasons: (a) the demographic crisis (aging and population decline) i s greater in Russia than in Western Europe, and the need to sustain rapid growth i s also more pressing, (b) Russia has a large pool of potential migrants with a fluent command o f Russian and other similar (post-Soviet) social attributes that allow for a more harmonious assimilation to work and life in Russia, and (c) Russia does not have an expensive xi and generous social safety net for unemployedmigrants (a primary source of tension surrounding migration inthe West). 28. A World Bank Policy Note of 2005 outlines important directions for improving migration policy, in particular the liberalization of the official regime along with a crackdown on illegal migration. This policy mix would have the potential o f attracting a migrant pool with better skill profiles for Russia, and o f alleviating social tensions that currently exist around illegal migrants. Russia i s indeed following a policy consistent with these two directions, although there i s a riski s that the official regime could remain sufficiently restrictive to negate the net benefits from a crackdown on illegal migrants, i.e. it i s not in Russia's economic interest to adopt a regime that would cause a decline in net inflows of migrants, especially over the medium term when the demographic crisis inthe country will reachits apex. Interconnecting Policies 29. Infrastructure has long been recognized as a key factor in regional and economic development, and represents a primary channel through which the government can promote competitiveness and productivity growth. Improvements in infrastructure may not only improve the development prospects o f a given regions, but may have benefits for multiple regions, including critical linkages inmarkets and the reduction o f transportation costs. 30. On the other hand, the results of Chapter 2 suggest that infrastructure density, in and of itself, may not contribute much to growth at the regional level. The 2009 World Development Report also emphasizes that, in world experience, a number of expensive and ambitious infrastructureprojects have had little impact inpractice. It i s therefore critical that decisions on infrastructural investments are made on the basis o f a careful assessment of the overall development impact of each specific project. This assessment should focus on the degree to which the absence of key infrastructure is currently a genuine constraint on growth or the integration of markets. (Social) Equalization Policies 3 1. Measures to ensure minimal social (and living) standards inpoorer regions properly belong to the category of social policy, as opposed policies specifically targeting to regional development. Yet the coordination o f these policies with a national strategy for regional development would be important. Equalization policies aimed at supportingthe social sphere are often confused in Russian debates with policies aimed at equalizing development or growth in different regions. Such latter measures belong to category of regional (development) policies discussed below. Confusion between these two types o f equalization policies may have been a primary source o f contention in the discussion o f the various drafts of the National Regional Strategy in Russia during 2005-2007. While a regional strategy focused on promoting growth points and agglomerations i s indeed inconsistent with alternative policies aimed at equalizing regional development, the need for social equalization policies could become even greater under the former strategy. Given limitations inthe speed of migration, promoting unbalanced growth could actually complicate the challenge o f addressing poverty in lagging regions. Given the fact that basic social expenditures are largely the responsibility of regional and local governments in Russia, equalization transfers have major significance for maintaining minimal standards in laggingregions. 32. The actual levels and stability o f equalization transfers do have important direct implications for regional development, however. If the goal i s to promote migration to growth xii points, benefits should not be set at levels highenough to discourage out migration. Ifthe goal i s to encourage initiatives aimed at igniting regional growth and development, then benefits should not be "soft" in the sense o f adjusting quickly to reallocate funds from more successful to less successful regions. The equalization formula was developed inRussia with the aim o f satisfying both of these criteria, although the additional allocation of transfers and federal loans outside of the formula-based equalization fund shouldalso be scrutinized from this point of view. Regional Development Policies 33. State policies aimed at igniting or promoting economic development in specific, usually poorer, regions represent one o f the most controversial areas of economic policy. Any country considering embarking on this path would do well to study the lessons of a very mixed, and predominantly disappointing, history. The reasons for past failures are numerous, including limitedinformation inthe government about true regional economic potential, distortions of the allocation of factors away from true comparative advantage, the adverse effect of subsidies on economic incentives, and the inevitable politicization of decisions about which regions to support. A number o f experts argue that investments in people (education, health, migration) can be a more effective tool for addressing social distress inlaggingregions. 34. Nevertheless, many larger countries with regional development problems have pressed on with regional policy, althoughthe lessons of the past have supported a consensus around a set of new principles, summarized below, that are strikingly different than in the past. Arguments in favor o f a regional policy for Russia could cite the following considerations: Some areas of the country (the Far East, the South, border areas) are prioritized for economic development for purely political as well as economic reasons. Given the constraints of whatever political decisions are made in this regard, Russia should pursue the most economically efficient means for realizing the implied development priorities. e Russia i s in need of strong urban agglomerations outside o f Moscow and Saint Petersburg, and alternative cities face demographic challenges for attracting and accommodating sufficient migrants. Under certain circumstances, state interventions (social infrastructure, for example) mighthelp support such agglomerations. 0 At this point in transition, a significant amount of uncertainty exists about the economic potential of many Russian regions. This i s arguably greater than the uncertainty that exists in countries where the initial allocation of capital and settlements followed market forces. Increasingreturns associated with agglomeration economies can prevent regions that currently lack such agglomerations to compete effectively with other regions and realize their potential. Thus, it may be expedient to seek and promote mechanisms for revealing and realizing this hidden potential where it exists. 0 The substantial social and economic costs of massive internal migration increase the opportunity costs in Russia of not revealing hidden potential in lagging regions. Although many settlements in Russia may have not been established under market forces alone, that in itself i s not proof that they should be dismantled. Given that people and capital already exist inthese places, finding successful development paths inat least some ofthemcouldbeofconsiderable value to Russia. 35. The extent to which Russia adopts regional development policies to address these issues will reflect a difficult political choice with significant pros and cons. To the degree that Russia xiii does decide to pursue policies of this nature, it would be well advised to base its approach on the principles of the so-called new regional policy, which incorporate important lessons from past experience. The example of the European Union (EU) may provide the outline of a good basic model for policy design inRussia, althoughthe focus inthe EUi s exclusivelyon laggingregions. Perhaps the most important distinctions of the new approach to regional policy followed in the EUand elsewhere are: The overall goal o f new regional policy is to increase productivity, not provide subsidies to compensate for lower productivity, as was common inthe past. Under the new regional policy, government provides local public goods (infrastructure, regulation, coordination) as opposedto financial subsidies to economic organizations. The new regional policy supports initiatives coming from regional or local levels, consistent with national priorities, through a competitive process. The goal i s reveal information about promising projects and to coordinate decision-making around their realization. The new regional policy monitors results, cuts off unsuccessful projects before social losses accumulate, andre-allocates resourcesinthe direction of successes. Education Policies 36. While social (equalization) policies can have a direct impact on the quality o f education, a number of countries have adopted additional programs aimed at boosting educational opportunities in lagging regions. Although these polices can be regionally oriented, this represents investment in people rather than places. A better educated and qualified population will generally have better opportunities not only for employment within a given region, but for migration to other more promising regions. Ireland and India are among countries that many observersbelieve hadparticularly successful national education programs. 37. Most important for the success of such education policies i s that the training provided i s linkeddirectly to the labor market and the needsofthe economy. The common absence of such a strong link has been a particular problem inRussia during the transition period, and i s the major target of education reforminthe country. A National Regional Strategy for Russia 38. The above considerations suggest some central elements o fa constructive nationalregional strategy for Russia. O f course, the particular chosen mix of these elements will require the consideration of complicatedtradeoffs. Sustaining rapid growth and welfare improvements inRussia depends critically on the continued flow o f capital and migrants to areas o f high economic promise and agglomeration. These flows appear to be insufficient and imperfectly regulatedat the present time. Thus, additional government policies to facilitate these flows could be important in areas such as migration policy, housing and social infrastructure, labor market policies, and competitionpolicy. Maintaining strong and stable social equalization transfers will be a means o f sustaining minimal living and service standards in lagging regions. These transfers xiv should not be so high or unstable as to discourage either out migration from these regions, or the active pursuit by local authorities and entrepreneurs o f a successful development strategy. This support may decline over time in harmony with other policies to realize out migrationor regional development initiatives. Given the extent of informational and social problems associated with settlements in lagging regions, it i s critical that local authorities and entrepreneurs are given incentives and opportunities to realize their own initiatives aimed at improving their situation. Furthermore, Russia could make a political choice to support these initiatives consistent with national objectives through federal programs based on the principles o f new regional policy. Given the current nature of agglomeration effects (increasing returns) in Russia today, regions may not succeed in realizing their potential through market forces alone. Key infrastructural investments to improve market linkages, facilitate factor mobility, and improve the prospects for regions with genuine economic promise. 39. These considerations are broadly consistent with many o f the directions that Russia has been moving in recent years. This includes the drafting of a comprehensive National Regional Strategy, the solicitation o f regional strategies in harmony with national priorities, and the allocation o f a number o f national instruments to regons on a competitive basis, including the location of special economic zones and IT parks and infrastructural investments from the InvestmentFund. New developments interritorial planning place a strong emphasis on fostering a broad dialog between different levels of government and local business. Special national projects have been introduced for the priority regions o f the South and the Far East. Other measures have targeted migration, including the provision o f amnesty to a large number of illegal migrants. Equalization transfers are concentrated in poorer regions by formula that limits the adjustment o f transfers to current economic conditions. Education reform has also been prioritized. Recent policy declarations on regional policy prioritize infrastructure, human capital, and other productivity-enhancing investments over subsidies to low productivity firms. However, Russia has not yet adopted a strategy for regional development that would allow it to employ these and other policy instruments ina coordinated and consistent way. 40. The important goal of giving incentives and opportunities to lagging regions for realizing their potential implies a particular challenge for Russia. As stressed above, the inherently decentralized nature o f information on regional potential implies that success in this area will depend greatly on genuine decentralized initiatives for increasing productivity. While agglomeration points are key to regional growth, an attempt by the central government itself to choose future points o f growth and design its own detailed regional programs would likely be highly unsuccessful. Indeed, experts attribute some o f the failures of regional policy inthe past preciselyto top-down technocratic planning and a lack o f local responsibility. 41. Since 2000, the Moscow has assimilated considerable power relative to the regions. Perhaps the most important decision in this regard was the elimination of direct gubernatorial elections in2004 in favor of the presidential appointment of regional governors. Inthis context, the balance of political priorities of the regional leadership has shiftedinthe direction of pleasing superiors in Moscow relative to constituents within the region. On the one hand, the new willingness of the regions to cooperate strongly with Moscow offers better opportunities for coordination and cooperation in the realization o f regional initiatives. On the other hand, too strong a hand of the center can also breed risk aversion among local and regional officials for pushing ahead with their own initiatives, or eliciting the initiatives o f local entrepreneurs. The exercise in drafting regional growth strategies inRussia was informative from this point of view. xv Some regions took this process very seriously and were deeply engaged. Other regions did little more than go through the motions as a means o f complying with the demands o f Moscow. A fundamental principle for the success o f a regional development strategy in Russia for lagging regions i s that the system rewards innovation and genuine decentralized initiatives as opposed to passive obedience. 42. The particular problems o f regional development and inherited spatial imbalances make economic transition in the Russian Federation an inevitably long and complicated process. The speed with which Russia has been developing modem economic institutions inrecent years offers hope that the government will also succeed inmanaging the difficult spatial transformation o f the country inthe decades ahead. xvi 1. REGIONAL ECONOMIC GROWTHINRUSSIA:AN ASSESSMENT OF RECENT TRENDS' A. INTRODUCTION 1.1 Vast geographic and cultural diversity gives a unique dimension o f complexity to economic and political transformationinthe Russian Federation. The richnaturalendowments of the country are distributedhighly unevenly among Russian Subjects o f the Federation (regions). Seventy years o f Soviet planning left a geographic distribution of settlements and capital that i s often poorly related to natural economic comparative advantage. This includes significant settlements and capital inextremely cold areas, where economic efficiency i s inherently low.' 1.2 The size and diversity of the Russian Federation also presents major challenges in governance and central control. The weakening o f central power inthe Soviet Union during the Gorbachev era of the late 1980scan be associatedwith a corresponding rise inregional economic power, which increased even further to inthe early 1990s. Recent years have witnessed a major consolidation o f power at the federal level. Nevertheless, the size and diversity o f Russia imply that a large number of important decisions will continue to be made (formally or informally) at lower levels o f government. 1.3 Regional development during the transition period in Russia can be broken up roughly into four distinctperiods: 1991-1994: Economic decline and a major increase in regional inequality. The rapid introduction of markets and opening of the economy to foreign competition represented a huge shock to manufacturing and other tradable sectors of the economy. The general sharp decline ineconomic activity was distributed highlyunevenly across the Russian Federation. The weakness of the federal government also hindered the adoption of measures to counteract widening inequalities. Regions endowed with oil, gas, and other naturalresources fared relatively well, whereas regions concentratingin industrial manufactured goods generally experienced sharp declines and social distress. 1995-1998:Demonetization and crisis. Relations between the federal government and the regions became codified and somewhat more uniform in the latter 1990s. State subsidies in the form of directed credits were largely discontinued which fed into payment problems for a large share o f Russian enterprises, and the rise o f the so- called "virtual economy." An economic boom in Moscow in 1996-1997, fueled by huge short-term capital inflows to the government bond market, stood in sharp ' Some preliminaryresultsinthe chapter were publishedas "Russian RegionalGrowthand Agglomeration Effects" as Part2 of World BankRussian Economic Report #14,2007. 2 On this topic, see GaddyandHill(2003). I contrast to the depressed economic situation in most regions. The reversal o f short- term capital flows in 1998 generated a general financial crisis. e 1998-2003: Broad-based growth and trend toward regional equality. The sharp depreciation o f the Russian ruble, combined with tight financial constraints imposed by the 1998 crisis, provided both opportunities and incentives for economic growth and restructuring. The growth in this period was industry-led, and largely through import substitution. Many o f the most depressed industrial regions with the largest excess capacity exhibited exceptionally high growth rates. The consolidation o f power at the federal level also helped to counteract the strong former tendencies toward regional inequality. Changes inresource taxes and the federal transfer formula had a progressive effect inthis regard. 2004-2007: Accelerated and more regionally-dzflerentiated growth. By 2002, economic growth had slowed considerably in the face o f the strong real appreciation o f the ruble, growing capacity constraints, and years o f low fixed capital investment. But the sharp increases inoil, gas, and other export prices provided a major stimulus to domestic demand, which has been the primary catalyst for growth in recent years. Changes in sharing rules ensured that the majority o f revenues from exceptionally highoilprices accruedto the federal budget (and StabilizationFund). 1.4 This chapter examines changes inthe patterns o fgrowth across the regions o fthe Russian Federation in recent years by contrasting the latter two subperiods listed above, primarily on the basis o f data on industrial production and other output (See Box l.l).3 This analysis yields a number o f interesting conclusions. Data suggest increasing differentiation in growth between Russian regions in the more recent period, in contrast to 1998-2003. Nevertheless, it does not appear that a trend toward divergence (i.e. greater inequality) in per capita regional output has accompanied this greater differentiation. This i s due to the fact that a number o f poorer regions have exhibited relatively rapid growth and a number o f richer regions have grown more sluggishly. Thus, the profile o f regional growth inRussia would appear to still have a strong pro- poor component. This stands in contrast to some suggestions in the literature that regional inequality has increased. 1.5 On the other hand, the changing pattern regional growth in Russia does suggest some increasing challenges. A widening disparity has emerged, between most o f Western Russia (Central, North Western, and Southern Okrugs) and the rest o f the country. Many o f the poorer regions that exhibited rapid growth in recent years are located in the Central Okrug. While aggregate industrial growth slowed inRussia during2002-2007, it actually accelerated inmuch o f the Western part o f the country. The relatively poorer regions that have achieved higher-than- average growth inrecent years are concentrated in the European part o f Russia. For this reason, the sustainability o f the current trend toward convergence is questionable. As the West becomes richer relative to the East, regional disparities inRussia could become increasingly important. Chapter3 providesamore detailedlook at the factors of regionalgrowthfor the entireperiodo f 1998-2004. 2 Box 1.1: A Noteonthe Data Data on GRP were available for this CEM only through 2004. Thus, in order to explore the more recent period, this chapter makes use o f industrial production data, as well as data on retail trade, construction, and other market services. I s industrial growth a good proxy o f regional growth? The following figure shows that industrial growth and GRP growth are highly correlated for the 1999-2004 period. Furthermore, industry accounts for a significant share of GRP in most regions o f Russia. Thus, looking at industrial growth should give us a sense o f what i s happening to growth at the regional level during the more recent period. Industrial Growth and GRPGrowth 1999-2004 . -5 0 5 10 15 GW Growth 1999-2004 B. EVOLVING PATTERNS OF REGIONAL GROWTH 1.6 What has happened to industrial growth across the regions o f Russia inrecent years? The table below presents some basic statistics. The primary message i s that while overall growth has fallen, the dispersion o f growth across the regions o f Russia has increased considerably. Table 1.1: Dispersionof IndustrialGrowthAcross RussianFederation(RF') Regions Industrial Growth (%) 1999-2003 2004-2006 1999-2006 Russian Federation 6.64 5.24 6.12 Mean across RF Regions 7.22 6.58 7.00 Median across RF Regions 7.65 5.85 6.93 Standard Deviation 3.76 6.13 3.80 Coefficient o f Variation 0.52 0.93 0.54 1.7 Russian industrial growth on aggregate fell from 6.6 % during 1999-03 to 5.2 % during 2004-06. The mean rate of industrial growth across the regions of Russia fell by 0.6% points from 7.22% to 6.58%, while the median fell by much more from 7.6 % to 5.8 %. Furthermore, the standard deviation almost doubled from 3.8 % to 6.1 %. As a result, the coefficient of variation of industrial growth increased dramatically after 2003, suggesting that while growth has fallen, the distribution o f growth has become more uneven in the more recent period. This i s confirmed visually in the figures below, which depict the dispersion of industrial growth across the regions of Russia for the 1999-2003 and 2004-06 periods. Inthe more recent period, there are 3 fewer regions with growth above 10%. But average growth among the remaining fast-growing regions i s higher. Average growth among slower-growing regions i s also lower. Thus, the distribution of industrial growth rates across the regions o f the RF has become more skewed duringthe 2004-06 period, as the pattern of growth has evolved from largely broad-based growth to the concentration of growth ina smaller number of regions. Figure 1.1: Industrial Growth, 1999-2003 Industrial Growth 1999-2003: Dispersion across RF Regions I -5J Figure 1.2: Industrial Growth, 2004-2006 Industrial Growth 2004-2006: Dispersion across RFRegions 1.8 An important question concerns the impact of changes in the overall pattern of regional growth on the convergence or divergence o f income levels across regions. This question is examined below for the 1999-03 and 2004-06 periods. c. DO GROWTH PATTERNS INRUSSIAN REGIONS FAVOR CONVERGENCE OR DIVERGENCE? 1.9 While growth has become much more uneven across the regions o f the Russian Federation, this does not imply, in and of itself, that a process o f divergence has set in, i.e. that the gap betweenricher andpoorer regions is increasing. Thiswould betrue ifricher regions were growing faster than poorer regions. Growing divergence among Russian regions may appear intuitively compelling when comparing poorer regions in Southern or Northern Russia with the some booming and relatively wealthy areas like Moscow and Saint Petersburg. For this reason, it 4 i s a common conjecture in some articles and the popular press. But this intuition can be deceiving for Russia as a whole. The evidence presented below suggests that no overall trend o f divergence has emerged among Russian regions, even during more recent years o f much more differentiation inregional growth. 1.10 An interesting recent study by the Moscow Independent Institute for Social Policy presents evidence that, on the contrary, regional disparities in GDP have actually increased in recent years: This study takes a different analytical approach than this Country Economic Memorandum. It does not estimate convergence according to differential regional growth, but employs Gini coefficients for the examination o f relative GRP per capita over time (with a price correction for differences inprices across regions). The increase inregional inequality according to this measure i s most likely due to strong relative price movements that increased the relative current value o f GRP in regions with a concentration in resource extraction. Most o f these regions already had relatively high GRP per capita. While this i s indeed consistent with one interpretation of growing regional inequality, measures of convergence or divergence based on differential regional growth arguably give a clearer picture of the actual development trend, particularly as the vast majority o f additional resource income (GRP) due to oil price increases in Russia i s taxed away from regions. Measures of regional household income per capita also do not show a notable trend toward divergence inrecent years. A forthcoming poverty analysis by the World Bank also shows no increase the coefficient of variation o f average regional consumption per capita between 2002-2006, although weighing regions by population would show a slight increase, as the difference in average consumption between the highly populated richer regions of Moscow and Saint Petersburg, on the one hand, and the highly populatedpoorer regions of the South, on the other, has indeedincrea~ed.~ 1-11 Output data suggest that many regions with lower initial GRP per capita experienced faster industrial growth for both the 1999-03 and 2004-06 periods. The figures below plot Russian regions on the basis o f industrial growth and initial GRP per capita (in 1og)for the 1999- 03 and 2004-06 periods, as well as for the entire 1999-06 period. Figure 1.3: Industrial Growthvs Initial GRP Per Capita, 1999-2003 Industrial Growth vs initial GRP per capita 1999-03 20 I 9.0 9.5 10.0 10.5 11,O 11.5 12.0 12.5 Ln ( GRP per capita 1998) 'ARussian Social Atlas of Russian regions: Thematic Overviews (2008). Federation: Promoting Equitable Growth - A Living Standards Assessment (2008), World Bank, forthcoming. 5 Figure 1.4: Industrial Growth vs Initial GRP Per Capital, 2004-2006 industrial Growth vs initial GPPper capita 2004-06 ** * * 9 10 11 12 13 Ln ( GRP per capita 2003 ) Figure 1.5: Industrial Growth vs Initial GRP Per Capita, 1999-2006 industrial Growth vs Initial GRfJper caplta 1999-06 ' 20 e 15 c, 10 ic 5 - $c 0 + -5 9.0 10.0 11.0 12.0 Ln (GRP per caplta 1998 ) 1.12 Once we ignore a few outliers,6 it i s possible to decipher a negative relationship between initial GRP per capita and subsequent industrial growth inthe figures above for boththe 1999-03 and 2004-06 periods. In other words, on average, poorer regions appear to have experienced faster industrial growth. Retail Trade,Manufacturing, Construction, Other Paid Services 1.13 Does this negative relationship between initial income and subsequent growth apply to other sectors of the economy? Figures on retail trade also suggest evidence o f convergence, particularly during2004-2006, despite some few notable outlyingregions. These outliers are Moscow city and Tyumenobl., which have GRP per capita levels far in excess of any other region, and some southern republics which have low GRP per capita and highly volatile growth: Ingush Rep., Kalmyk Rep.DagestanRep. Finally, Kalingradhadunusuallyhigh industrialgrowthof 32% during 2004-06. 6 Figure 1.6: Retail Growthvs Initial GRP Per Capita, 1999-2003 and 2004-2006 RelallTrade Growth vs lnltlalGRP per caplla 1999.03 RelallTrade Growth vs lnltlalGRP per uplta 2004-06 1 5 1 I * * 1 2 0 - 1 15- (3 ge 10- * * I I 8.5 9.5 10.5 11.5 12.5 9.0 10.0 11.0 12.0 13.0 Ln ( GRP per uplta 1998) Ln ( G W per capita2003 ) 1.14 This relationship also looks to be visible indata on manufacturing and services, although construction growth, as an exception, would appear not to follow this pattern. When GRP data become available for these years, it will be possible to give a more full treatment to the question of convergence. Yet, the preliminary evidence would suggest the absence o f a growth trend toward divergence inboththe 1999-2003 and 2004-2006 periods. Figure 1.7: Growthof Manufacturing, 2005-2006 Growthof Manufacturlng2005-06 30 _I 6 -10- Figure1.8: Growthof Construction,2005-2006 and Growthof Other PaidServices, 2005-2006 Oruuth of Conrtructlon2005.06 e w t h &Other PaidServlces 200546 30 251. 8 20- t ,I, , - 8.5 100 10.5 11.0 11.5 12.0 12.5 9.5 10.0 10.5 11.0 11.5 12.0 12.5 In (OW per capita 2004) In( G P per caplta2004) 7 D. A STATISTICALTESTFORCONVERGENCE 1.15 This section employs a formal statistical test for convergence that formalizes the relationships visible in the above figures. The basic equation relates the industrial growth of a given region to initial GDP per capita and the investment rate over the same given period. As the number ofyears under consideration for thesetwo subperiods are very few, and output rather than GRP data are being used, the significance of this test is limited, and the preliminary results obtained here will needto be confirmed infuture work. Industrial Growth, =a, a,ln(GRPper capitai) a, ln(InvestmentRatei)+ ci + + 1.16 This equation is estimated for the regions of the Russian Federation, both with and without the investment rate. The estimation i s carried out separately for the 1999-2003 and 2004-2006 periods using OLS'. For both periods, a few outliers are removed from the regressions due to their potential excessive effects on the results. For 1999-2003, Moscow, Saint Petersburg, the Southern Republics, and Tyumen were removed from the sample. For 2004- 2006, Chukotka, Kaliningrad, and Arkhangelsk were also removed for the same reason. DependentVariable: DependentVariable: IndustrialGrowth1999-2003 IndustrialGrowth2004-2006 Constant 26.191** 23.461** 32.043** 29.338** (9.812) (9.381) (14.097) (13.852) ln(Initia1 GRPper capita) -1.869* -2.692** -2.481* -3.149** (0.964) (0.962) (1.343) (1.356) ln(1nvestment Rate) 3.890** 3.279* (1.377) (1.654) R2 0.054 0.157 0.051 0.106 N 68 68 66 66 1.17 The above results verify that regions with lower levels o finitial GRPper capita generally experienced faster industrial growth, and that this effect i s statistically significant for both the 1999-2003 and 2004-2006 periods (Columns 1 and 3). Furthermore, this relationship becomes stronger when controlling for the effect of the investment rate on industrial growth (Columns 2 and 4). This i s depicted graphically inthe following figures, which present the partial correlation between industrial growth and initial income from the estimations that included the investment rate. 7 The next chapter employs a maximum likelihood technique that is consistent in the presence of (apparently significant) spillovereffectsthat link growthbetweendifferentregions. 8 Figure 1.9: Partial Correlation of Industrial Growth, 1999-2003 and 2004-2006 '$i Partial Correlationof Industrial8owth 1999.2003 PartialCorrelationof IndustrialGrowth2004-06 and and initialGRP per capita InitialGRP per capita ... 8 1 EE -l ' . ' e I,*:* ' 5 - 2 5 .e 'r. ..,... 0 - .* I I - P J 9 0 9 5 100 105 110 115 9.5 10.0 10.5 11.0 11.5 12.0 h(GWper capitalB98) In(GFIPper capita2003) E. THEGEOGRAPHICDISTRIBUTION OF GROWTH 1.18 Although high growth has become increasingly concentrated in a smaller number o f regions of the Russian Federation, output data do not suggest that an overall process o f divergence has set in. This indicates that a number o f poorer regions must be experiencing high output growth andor that a number of richer regions must be experiencing lower growth. This section examines how growth has evolved within individual regions o f the Russian Federation and how the distribution o f growth has shifted inrecent years. Data confirm a very strong trend toward the concentration o f growth inthe Western (European) part of Russia, most particularly in the Central, Northwestern, and Southern Federal Okrugs. The continued trend toward convergence i s explained primarily by the facts that a number o f highgrowth regions inWestern Russia were relatively poor, while a number of resource-rich regions in Eastern Russia have experienced slower growth. 1.19 The figures below depict how the pattern o f industrial growth across the seven federal okrugs has evolved since 1999. Giventhat federal okrug level data are only available from 2000, the periods of 2000-03 and 2004-06 are compared below on the basis of this data (Option 1). In addition, estimates of the 1999-03 and 2004-06 periods are presented using estimates based on a weighted average of the growth rates of the individual regions within each federal okrug (Option 2)* 1.20 Both options confirm that growth in the most recent period has become more concentrated inthe Central, North Western, and Southern Federal Olaugs. It is interestingto note that, although rates of industrial growth slowed in the Russian Federation as a whole during the 2004-06 period, the pace of growth actually accelerated in these okrugs. At the same time, growth rates have fallen notably inthe Urals, Siberia, and Far East. Inthe Volga Okrug, growth increases slightly under option 1 and falls slightly under option 2. The increasing concentration o f growth inthe Western part o f the country is consistent with evidence presented by Brown et al, (2008) that Russian enterprises activity increasingly westward, where productivity i s generally higher. 9 IndustrialGrowth:2000-2003 by FederalOkrug IndustrialGrowth:2004-2006 by FederalOkrug (Annualized, 9'4 (Annualized, 9'4 10 a - Central West South Volga hals Sikria FarEast Central West South Volga Uals Siberia FarEast IndustrialGrowth: 1999-2003 by FederalOkrug IndustrialGrowth:2004-2006 by FederalOkrug (Annualiied,Y4 (Annualized, 9'4 Central West South Volga Urals Siberia FarEast Central W e s t South Volga Urals Siberia FarEast 1.21 How i s the concentration of growth toward the Central, North Western, and Southern Okrugs related to the absence o f a tendency toward divergence across the regions o f Russia? Table 1.3 below divides Russian regions into four groups according to initial per capita GRP and the pace of industrial growth during 2004-2006. The majority of Subjects of the Federation fall into the categories of either poorer (below median GRP) regions with faster growth or richer regions with slower growth. In addition, the poorer regions with higher growth are concentrated in the Central and Southern Okrugs: Tver, Bryansk, Tula, Kostroma, Ryazan, Voronezh, Ivanonvo, Saratov, Vladimir, Kaluga, and several regions in the South. The North Western Okrug has only one such region, Kaliningrad, but has a number o f the richer regions that have continued to experience higher growth. Conversely, a high number o f richer regions that grew slower than average during 2004-06 belong to the Siberian and the Far East O h g s : Irkutsk, Kemerovo, Krasnoyarsk, Tomsk, Khabarovsk, Sakha (Yakutia), Chukotka, Kamchatka, Magadan, Amur. 10 Table 1.3: The Dynamics of Regional Growth in Russia Industrial Growth GRP per capita 199943 2004-06 1998 2003 Growth above median, GRPper capita above median-2004-06 A indicatesindustrial growth rose above median in 2004-06 from below median in 7999-03 Nwest Arkhangelsk oblast 8.99 22.79 34,121 55,260 Nwest Leningradoblast 10.09 17.49 27,096 51,252 Fareast Primorskiikrai 0.74 15.26 A 28,206 34,492 Fareast Sakhalinoblast 9.07 15.20 60,349 94,955 Central Moscowcity 14.69 15.19 105,263 155,712 Central Moskowoblast 14.20 14.80 26,945 38,415 Volga Mordovia republic 13.90 10.13 22,999 35,693 Nwest Karelia republic 6.04 10.03 A 31,654 44,793 Central Belgorodoblast 11.43 9.11 23,594 33,628 Siberia Novosibirskoblast 7.92 8.85 23,508 36,866 Central Oryol oblast 8.46 8.08 23,960 34,849 Volga Orenburgoblast 8.22 7.73 30,485 44,714 Urals Sverdlovskoblast 9.74 6.75 31,534 46,058 Nwest Novgorod oblast 7.75 6.66 26,993 35,857 Urals Chelyabinskoblast 8.10 6.49 30,695 42,710 Nwest Vologda oblast 4.47 5.92 A 43,748 61,140 Growth above median, GRPper capita below median-2004-06 Nwest Kaliningradoblast 7.57 32.17 A 20,786 32,044 South Dagestanrepublic 1.35 18.62 A 7,549 12,233 South Adygeya republic 11-72 16.68 11,877 14,194 South Rostovoblast 12.00 14.41 17,302 27,776 Volga Mari-El republic 5.20 13.61 A 16,570 18,791 Siberia Omsk oblast 9.06 11.51 20,692 31,666 South Kabardino-Balkarrepublic 12.95 10.58 13,981 22,106 Central Bryansk oblast 9.96 10.11 16,600 23,030 South Krasnodarkrai 8.93 9.67 22,224 31,299 Volga Chuvash republic 5.22 9.33 A 17,508 22,357 Central lvanovo oblast 6.22 9.24 A 12,631 17,530 Volga Saratov oblast 11.33 8.97 20,910 30,772 Central Tver oblast 4.51 7.97 A 21,773 30,249 Central Tula oblast 7.72 7.97 22,563 29,440 Central Kostromaoblast 6.06 7.73 A 20,513 26,523 Central Ryazan oblast 6.84 7.40 A 21,346 29,468 Central Voronezh oblast 5.79 7.36 A 18,629 25,492 South Stavropol krai 7.83 6.69 18,955 26,057 Volga Penza oblast 8.96 6.26 15,204 21,398 Siberia Tuva republic 4.86 6.19 A 10,820 14,859 Siberia Altai republic 12.13 6.12 16,165 24,120 Central Kaluga oblast 13.50 6.11 21,730 29,697 Central Vladimir oblast 11.07 5.96 18,438 26,275 Siberia Buryat republic 11.46 5.85 18,900 26,901 Mean 7.22 6.58 25,646 35,967 Median 7.65 5.85 22,999 33,314 Russian Federation 6.64 5.24 36,060 51,958 Growth below median, GRPper capita above median-2004-06 V indicatesindustrialgrowth fell below median in 2004-06 from abovemedian in 7999- 03 Volga Tatarstan republic 5.20 5.79 46,661 66,445 11 Sentral Lipetsk oblast 10.35 5.56 V 34,673 49,241 Sentral Yaroslavl oblast 8.87 5.23 V 27,198 40,806 Volga Bashkortostanrepublic 5.04 5.13 35,141 47,124 Siberia lrkutsk oblast 6.46 5.07 37,451 45,501 Urals Tyumen oblast 5.86 4.83 171,087 234,989 Siberia Kemerovooblast 6.38 4.59 27,135 37,896 Nwest Komi republic 3.07 4.40 56,774 69,863 Volga Nizhny Novgorodoblast 2.44 4.33 27,015 38,906 Chukotka autonomous Fareast okrug 11.67 4.23 V 61,546 177,478 Nwest St. Petersburgcity 10.32 3.71 V 36,631 59,136 Volga Samaraoblast 7.35 3.56 43,129 56,089 Fareast Kamchatkaoblast -0.63 2.66 47,192 48,845 Volga Perm krai 2.32 36,789 50,234 Nwest Murmansk oblast 2.81 2.20 53,897 65,707 Fareast Sakha (Yakutia) republic 3.58 1.61 72,312 91,741 Volga Udmurtia Republic 2.35 1.55 30,310 40,116 Siberia Krasnoyarskkrai 4.45 1.48 63,959 85,478 South Astrakhan oblast 5.75 1.06 24,682 37,092 Siberia Tomsk oblast 10.95 -1.79 V 35,647 56,431 Fareast Khabarovskkrai 11.65 -2.01 V 37,258 56,970 Fareast Amur oblast 2.45 -2.64 25,329 35,155 Fareast Magadanoblast 2.46 -5.48 60,034 72,098 Growth below median, GRPper capita below median-2004-06 Urals Kurgan oblast 3.91 5.76 16,616 21,591 Karachaevo-Cherkess South republic 12.80 5.67 V 11,886 16,220 Central Kursk oblast 4.25 5.44 22,295 31,240 Central Tambov oblast 7.80 5.14 V 16,404 26,341 Siberia Khakasia republic 4.15 5.07 29,051 33,314 Volga Ulyanovskoblast 2.74 4.26 20,655 26,645 South Volgograd oblast 3.80 3.94 22,298 31,713 Volga Kirov oblast 2.09 3.84 21,642 25,992 Central Smolenskoblast 8.05 3.30 V 20,584 32,594 Nwest Pskov oblast 8.04 3.03 V 17,135 24,910 Siberia Altai krai 7.94 1.52 V 15,914 22,600 Fareast Evrei autonomousoblast 11.43 1.35 V 17,128 24,710 South North Osetiya republic 6.45 0.05 11,602 18,076 Siberia Chita oblast 6.64 -0.82 20,830 30,939 South lngush republic -2.92 -6.07 16,249 12,267 South Kalmyk republic 0.98 -6.37 16,363 20,671 1.22 As indicated in Table 1.3, a number of regions where industrial growth increased from below-median growth in 1999-2003 to above-median growth in 2004-2006 were also regions with incomes below the median in 2003: Kaliningrad, Dagestan, Mari-El Republic, Chuvash republic, Ivanovo, Tver, Kostroma, Ryazan oblast, Voronezh, and Tuva republic. O f course, this i s not universally true. Several regions that fell from above-median growth in 1999-03 to below- median growth in 2004-06 had incomes below the median in 2003: Karachaevo-Cherkess republic,Tambov, Smolensk, Pskov, Altai krai, and the Jewish Autonomous Oblast. 1.23 Another interesting question i s the distribution of industrial growth within the various groups of regions presented above. The higher growth rates experienced in the Central, North Western, and Southern Okrugs could be the result of an increasing number of regions in those areas achieving higher growth, or because of an acceleration of growth in the same particular subset of regions. Table 1.4 shows that the number of poorer regions in the Central O h g 12 experiencing above-median growth increased from 6 during 1999-2003 to 9 during 2004-06. Thus, the continued evidence for convergence in the more recent period is related to a larger number of poorer regions in the Central Okrug growing at above-median rates. By contrast, the same number of poorer regions in the Southern Okrug (6) continue to experience above-median growth in the more recent period. Three of those six regions remain the relatively developed areas o f Rostov, Krasnodar, and Stavpropol. Table 1.4 also shows that the number o f poor and lagging (low-growth) regions decreased (from 6 to 3) in the Central Okrug, remained the same number (5) in the Southern Okrug, and increased (from 2 to 4) in the Siberian and Far East Okrugs combined. Table 1.4: The Distributionof Growthby Okrug Number of NumberOf Total Number Regions Regions of Regions in 1999-03 2004-06 Federal Okrug Growth above median, GRPper capita above median Central 6 4 18 NWest 4 5 10 South 0 0 12 Volga 1 2 14 Urals 2 2 4 Siberia 2 1 12 FarEast 3 2 9 Growth above median, GRPper capita below median Central 6 9 18 NWest 1 1 10 South 6 6 12 Volga 3 4 14 Urals 0 0 4 Siberia 4 4 12 FarEast 1 0 9 Growth below median, GRP per capita above median Central 0 2 18 NWest 4 3 10 South 1 1 12 Volga 6 6 14 Urals 1 1 4 Siberia 4 4 12 FarEast 5 6 9 Growth below median, GRP per capita below median Central 6 3 18 NWest 1 1 10 South 5 5 12 Volga 4 2 14 Urals 1 1 4 Siberia 2 3 12 FarEast 0 1 9 13 1.24 A greater numberofregions incentralRussiaexperiencing higher-than-average growthis consistent with the emergence and strengthening o f agglomeration effects for regions in proximity to Moscow and Moscow Oblast, which have been growing at particularly impressive rates. These questions are taken up inChapters 2 and 3. A preliminary statistical examination of the data suggests that industrial growth has become more regionally correlated in 2004-2006 relative to 1999-2003.' F. CONCLUSION 1.25 The analysis in this chapter explored recent shifts in the patterns of growth across the regions of the Russian Federation primarily on the basis of data on industrialproduction, which i s available for more recent years than GRP data. The evidence suggests that 0 While the pace o f average industrial growth has fallen, the dispersion of growth among Russianregionshas increased. 0 Industrial growth has become increasingly concentrated inthe Western part of Russia, particularly in the Central, North Western, and Southern Okrugs. In fact, industrial growth has actually accelerated in this part of Russia, in contrast to the trend for the country as a whole. 0 Despite the greater dispersion in growth during the more recent period of 2004-2006, output data do not suggest any shift to a trend o f divergence among regions. This was due to rapid growth in a number of regions in the Central Okrug with lower than average per capita GRP, as well as slower growth inanumber o f resource-richregions inthe Eastwithrelatively highper capita GRP. 0 The number of regions experiencing higher-than-average growth inthe Central Okrug i s increasing, supporting the possible presence of agglomeration (spillover) effects. Industrial growth appears statistically more related to growth in regions of close geographic proximity in2004-2006 than in 1999-2003. 1.26 The policy implications of these results, as well as the results of the Chapter 3, will be examined in detail in Chapter 4. On the one hand, the absence of a notable trend toward regional divergence inthe Russian Federation i s encouraging, and runs counter to the suggestions in a number of articles that regional inequalities are increasing sharply in the country. On the other hand, as the emerging divide between the Western and Eastemorthem part of the country increases, and the corresponding inequalities become more serious, the overall trend could very well revert to divergence. Some poorer regions in the East and North are already facing increasing challenges. 8 Although the subperiods inquestionare too short for a compellingstatisticalexaminationofthis question, ifregional industrial growth in 1999-2003 and 2004-2006 is regressed on initial GRP per capita and "market potential," or spillover effects from the growth rates in other regions weighted by distance from capital city to,capital city (see Chapter 2 for a detailedexplanation), the estimatedcoefficient on spillover effects is highly significant for 2004-2006 andless significantfor 1999-2003. 14 2. THE SOURCES OFREGIONAL GROWTHAND AGGLOMERATION EFFECTS' 2.1 This chapter continues the investigation of regional growth of Chapter 1. Instead of industrial production data, this chapter utilizes GW data (close to a regional equivalent of GDP) which, at the time of writing, were available only up to 2004. For this reason, this chapter only examines the period o f economic growth (1999-2004) with available data as one unit, as opposed to the division into growth subintervals performed in Chapter 1. On this basis, a more detailed investigation i s conducted on the subjects of regional convergence, the factors o f regional growth, and the presenceof spillover (agglomeration) effects between regions. 2.2 The results are largely consistent with, and complementary to, those o f Chapter 1. Additional evidence i s presented for the hypothesis that regional differentials have not increased, on average, duringthe period o f economic growth inRussia. On the contrary, initial income level enters consistently as a significant negative coefficient in the growth estimations. Explanatory factors of growth that appear significant include agglomeration effects (urban agglomeration and spillover effects from proximity to rapidly growing regions), human capital, climatic conditions, initial income, fuel endowments, and a year-round port. By contrast, data under consideration on infrastructure intensity or the business climate did not appear significantly related to the regional growth data. However, this could be due to problems in the measurement of genuine infrastructure or business climate constraints to growth over the period inquestion. 2.3 This chapter only presents the basic results ofthe econometric work. Those interested in the details of the investigations are referred to a background paper for this Country Economic Memorandum, (Polyakov (2007)). A. RELATED LITERATUREREGIONAL GROWTH RUSSIAAND AGGLOMERATION ON IN EFFECTS 2.4 Before the work conducting for this CEM, not much statistical research has been conducted on the sources of regional growth in Russia. Two studies ((Ahrend (2002) and Yudaeva et a1 (2004)) examined the 199Os, a period of sharp economic decline in Russia. The period o f economic growth since 1999 examined inthis chapter features a very different dynamic. Drobyshevsky et a1 (2005) examined regional growth during 1994-2002, which does include the initial period of growth. They investigate primarily the questions o f convergence and the link between regional growth to labor and capital constraints. They find evidence neither for strong regional convergence nor divergence duringthis period. 'This chapter was prepared under the joint support of the World Bank and the Canadian International Development Agency (CIDA). Work supported by CIDA was conducted in the Moscow Institute for the Economy in Transition. Some preliminary results from this chapter were published as "Russian RegionalGrowth and Agglomeration Effects" as Part 2 of World BankRussian EconomicReport #14,2007. 15 2.5 None o f the previous studies cited above attempt to account for agglomeration or spillover effects between regions. Research in the so-called new economic geography has spumedincreasinginterest inthe investigation of agglomerationeffects inexplaining regional or local growth, although the empirical literature has emerged and developed very slowly. One such literature examines urban agglomerations on the basis o f local (municipal level) data.' Suchdata was not available for this CEM, and hopefully this topic will be examined in future research through the generation o f appropriate Russian city-level data sets. Another literature examines agglomeration effects at the larger regional level in the form o f growth spillovers from surrounding regions. Getis et al. (2004) and Fingleton (2002, 2003) applied a concept of spatial econometrics to the analysis of such regional growth spillovers in the EU. This chapter largely follows their approach for the case of Russia. B. UNIT OFANALYSIS AND PPPCORRECTION 2.6 Russian statistics provide data for territorial unitsunder different definitions: the subjects of the Federation (republics, hays, oblasts, and autonomous okrugs); larger subjects of the Federation (which incorporate autonomous okrugs); and the units at the municipal level (rayons and citiedtowns). This report limits analysis to growth at the level of larger subjects o f the Federation, i.e. republics, hays, and oblasts, which incorporate respective autonomous okmgs. The choice o f territorial unit reflects the availability of data. Due to the absence of data, the Chechen Republichas been excludedfrom analysis. 2.7 The GRP per capita indicators usedinthis study were adjusted to correct for a substantial variance in regional price levels and the purchasing power o f the ruble across the Russian Federation. Russian official statistics do not offer PPP estimates for regions. Nevertheless, Granberg and Zaitseva (2003) calculated composite regional PPP indices based on data for three main GRP components (household consumption, social consumption and savings). This study employs this index as a base, and then adjusts GRP for other years through an implicit GRP deflator that i s calculated on the basis o f official data for nominal and real rates o f GRP growth for each region. The same analysis was also performed without a PPP correction with very similar results. Population and other demographic indicators correspond to recently revised official estimates that make use of the 2002 Census. c. TRENDS INTHE REGIONALVARIATION OF GRP 2.8 The analysis o f the previous chapter, based on primarily on industrial production data, suggested that growth in Russia was becoming increasingly concentrated in the Westem part o f the country, particularly in the Central, North Western, and Southern Okmgs. A preliminary glance at the distribution of regional per capita GRP growth from 1999-2004, presented inFigure 2.1, looks to be consistent with this view. Regions o f higher GRP growth duringthis period are represented with a darker green color. In addition to faster growth in the European part of Russia, regions with coastal locations also appear to enjoy an advantage, as i s consistent with the findings o f Sachs et al. (1998). For a recent survey of the empirical literature, see Roberts and Setterfield (2007) 16 Figure 2.1: Average Annual Regional Growth in GRP Per Capita in the Russian Federation: 1999- growth rate: 2.9 Some recent literature on Russian regions highlights what are perceived as growing regional disparities in income levels in the country. For this question, Figure 2.2 shows the coefficient of variation of regional GRPper capita inbothnominal andreal terms (PPP corrected) from 1995-2004. This coefficient of variation, which equals to the ratio o f the standard deviation to the mean, i s a commonly usedmeasure o f di~persion.~ 2.10 As indicated inFigure 2.2, while the coefficient of variation o f nominal GRP rose by 59 percent from 1995-2002, the coefficient of variation o f real GRP was essentially flat, rising by only 9 percent due to an increase during 2000-2003.4 The statistical significance o f variation in the coefficient of real GRP was tested through a likelihood ratio test (Verrill and Johnson (2005)).' This test suggests that, although there was a statistically significant increase in Alternative measures o f dispersion o f regional per capita GRP (Gini coefficients, inter-quartile range (the difference between the upper and lower quartiles o f the Log GRP per capita)), or range (the difference between maximum and minimumo fthe LogGRP per capita) givethe same consistent picture as the coefficient o fvariation. 4 These results are slightly different to those reported in Klotsvog and Chernova (2004) for the period o f 1996-2001, where it was argued that the coefficient o f variation inreal terms was increasing inthe mid- 1990s. This may be due to slight differences inthe data or methodology used. Yet Ktotsvog and Chernova also reach the general qualitative result that, after a price correction, no major trend toward divergence is detectable inthe coefficient o f variation. See ht@://wwwl.fpl.fs.fed.us/covtestk.html. This i s a test o f the equality of the coefficients o f variation o fk normally distributed populations for the 79 Russian regions. The text was also performed with the exclusion of two atypical regions: Chukotka (the highest growth rates) and Ingushetia (the lowest). This test is based on a Likelihood Ratio, and i s robust to log-normal distributions if the number o f such distributions tested i s less or equal 50 (which i s true for this 17 variation of nominal GRP from 1995 to 2004, all annual changes in the coefficients of variation were statistically insignificant. Therefore, it appears that the regional variation inper capita GRP didnot change significantly from 1996to 2004. Growing regional disparities innominal GRP per capita can be explained almost entirely by differential regional inflation rates, with richer regions experiencinghigher inflation on average. As concluded in Chapter 1, production data on growth in industry,trade, and other services actually indicates evidence for regional convergence over the period in question. As indicated below, initial GRP per capita also appears significantly and inverselyrelatedto regional growth, particularly after correction for a number of other factors. Figure 2.2: Nominal and Real Coefficients of Variation of Russian Regional GRP GRP per capita: Coeff. of Variation 0.850 - 0.800 - 0.750 - .z -----I- ~ r-* 0.700 - 0.650 - 0.600 - 1995 1996 1997 1998 1999 2000 2002 2003 2004 -Ncininal GRP -2001-Real + GRP Note:Sample size-79regions 2.11 Some groups o f regions did show significant convergence toward, or divergence from, the regional mean in GRP per capita during 1999-2004, as indicated in the map below (Figure 2.3). In this map, each region i s represented according to changes in its GRP per capita level relative to the mean over the period under consideration. IfY; (Y?)i s real GRP per capita at the end (beginning) of the period under examination for region i,and Y' (YO) i s the average o f per capitaregional GRP at the same time, then: (1) Aui ==-=. y.t q.O Y' YO 2.12 The map identifies regions that contributed the most to either regional convergence or divergence. The regions that contributed in the direction of an increase in overall variation (regional diversion) were o f two types: (i)leading regions further expanding their lead case, as 10 years are under consideration). This test was conducted for GRP per capita and its logarithms, with similar results. 18 (= >1,Ayi > 0), and(ii) regionsfurther falling behind(A<1,Ayi <0). The regions Ti' lagging Y.' Y' Y' that contributed the most to decrease increase in overall variation (regional conversion) were also o f two types: (iii) lagging regions catching up (= < l,Ayi > 0), and (iv) leading regions yi' Y' digressing to the mean (I The map below presents regional changes inrelative Y.t >1,Ayi <0). Y' GRPper capita and the relative level o f GRPper capita inthe beginning o fthe period (in 1998). Figure 2.3: 1999-2004 Per Capita GRP Growth in Russian Regions: Leaders and Lagers -60 01 -26 35 - ia ia+aei I&de?aa, -20 49 -8 03 - e Adaaias, 1135-4904 1r - C I +I-0 00 +/-I+I-2 84 42 2.13 Leading regions further expanding their lead with the largest change inrelative GRP per capita include St. Petersburg, Arkhangelsk oblast, Khabarovsk krai, Tyuminoblast, Perm oblast, and Tomsk oblast. These are competitive industrial regions with significant mineral extraction sectors (except for St. Petersburg and Yaroslavl Oblast). As emphasized in Chapter 1, however, this picture appears to have changed fundamentally for the period o f 2004-2006, with many central regions close to Moscow experiencing highgrowth and a number o f resource-rich regions experiencing lower growth. 19 2.14 Lagging regions that fell further behind include the Buryatiia republic, Altai krai, and Primorskii krai. These are remotely locatedregions with limited resources, although the latter i s a major transportation hub for the Far East. 2.15 Lagging regions that were catching up in 1999-2004 include Smolensk oblast, Rostov oblast, Saratov oblast, Astrakhan oblast, and the Mordovo republic. These are relatively developed regions with a favorable geographic position. A number of additional regions in the central part of Russia with lower-than-average per capita GRP have experience accelerated growth since 2004 (See Chapter 1). Higher per capita GRP regions that experienced slower growth include Irkutskoblast, Kamchatka oblast, Magadan oblast, Komi republic, and the Sakha republic. These are resource-rich regions with difficult cold and/or peripheral geographic positions. D. SPATIALINTERACTIONSANDMARKET POTENTIAL 2.16 The analysis in Chapter 1 suggested the presence of spatial dependence in regional economic growth, namely spillover effects from neighboring regions on the growth o f a given region. This chapter does a detailed investigation of this question of the degree to which fast- growing regions induce growth in neighboring areas. The economic advantages for such areas include proximity to markets of high demand and the expansion o f activities from centers of growth to peripheral areas due to congestion or cheaper available factors of production. It i s interestingto note that proximity to wealthy or growing areas can be a disadvantage as well as an advantage for growth. For example, it was perceived inRussia that the initial economic boom in Moscow had the effect o f attracting the best human and physical capital from surrounding regions. Only more recently has the greater Moscow agglomeration expanded well beyond its borders to surrounding regions. 2.17 Spatial spillover effects are measured by a variable that weighs surrounding regions by the size oftheir markets (GRPper capita) and distance. Distance is measured alternatively by the geographical distance o f regional capital cities from each other, or of estimated transportation time between these cities. This variable is commonly referred to in spatial analysis as "market potential (MP)," and i s based on the gravity concept that the strength (or probability) o f economic interactions between regions i s positively related to the size of their respective GRPs and inverselyrelatedto distance betweenthem (Isard, 1960: 424-457). where GRPj is the GRP o fregionj and dVisthe distance ofregionj from region i. 2.18 As might be expected, market potential, definedinthis way, is particularly concentrated inCentral Russia and the Northwest, particularly around Moscow and St. Petersburg. It should be noted that spillover effects could also be present in regions close to foreign borders from markets inother countries. Sucheffects are beyondthe scope of this study, however. 20 ure 2.4: The "Market Potential" of RussianRegions: 1999-2004 Map 3: Market Potentialof RussianRegions(basedon the 1998 GRP) The spatialmodelfor regionalgrowth estimations 2.19 Growth equations for 1999-2004 were estimated for Russian regions using methods o f spatial econometrics that account for correlation inregional growth (spillover effects) that are not captured inthe specified exogenous variables. The basic spatial growth regression model can be specified as follows:6 (3) Y=a+XP+p W Y + E "N(0, 821) where Y is the vector of regional growth rates, X i s the vector of exogenous growth factors, W i s the geographic weight matrix that corresponds to "market potential" as defined above, and a: 6, and p are estimated parameters.' In this framework, the parameter p measures the impact on growth from relative market potential (proximity to rapidly growing regional markets). 2.20 The presence of nontrivial spatial effects implies stochastic dependence in the regional growth equations, whichmake ordinary OLS estimates biased and inconsistent. Indeed, standard tests indicate that spatial spillover effects for the case of Russian regions are significant.8 For this reason, a maximum likelihood estimation procedure fiom spatial econometrics i s applied that is bothunbiasedand consistent. This i s referred to as the "spatial lag model" in the literature. An alternative model, the "spatial error model," assumes a first order autoregressive process for the spatial errors. Both o f these models were considered for the estimations that are reported inthis chapter. 'For normalization purposes, inthe matrix, W, Cj wij = 1. These tests are the Moran's 1, Geary statistics, and Moran's scatter plot (See Polyakov (2007)) for details. 21 2.21 The primary exogenous variables related to regional growth in the estimations were initial GRP per capita, the investment rate, urban agglomeration, human capital, infrastructure, the investment climate, ethnic fragmentation, resource endowments, the presence or absence o f a port, climatic conditions, andthe size of federal transfers. The measurements and results from the estimations for each o f these variables and regional spillovers are summarized briefly below. For details, the reader i s referred to Polykov (2007). DefinitionandMeasurement ofthe ExogenousVariables 0 Initial income is measured as GRP per capita at the beginning of the period in question. It i s usedprimarily to assessthe convergence hypothesis as examined above and inChapter 1. 0 Urban agglomeration. While one type of agglomeration effect (spillovers from growth in other regions) i s to be captured in the coefficient on "market potential," as defined above, growth clusters can take the more direct form o f urban agglomeration within a given region.' Evidence inthe literature suggests that productivity within an urban area i s an increasing function of population within a significant range. Some studies argue that urban agglomeration has indeedbecome o f increasingimportance in Russia inrecent years (Nefedova, 2004). The statistical analysis considered a number of variables related to urban agglomeration: size (population) of the largest city, share of the regional population living in the largest city, the share of regional population living incities withpopulationof 100,000 or more, or, alternatively, 50,000 or more. 0 Human capital. As in Chapter 1, this is measured by the share of the regional population with higher education. 0 Infrastructure. Measuresrelatedto infrastructure include stock variables inabsolute and per capita terms (rail density, road density, length o f roads and rail, and phone lines) and quality variables (the share of hard roads and share of improved hard roads). Investment climate variables are represented by the OPORA survey of small businesses for 2005, which measures perceptions of bribery, criminal activity, corruption, unfair competitive practices, and other variables related to the investment climate in the majority of Russian regions. A regional index o f investment risk distributedregularlybyExpert was also considered. 0 The degree of ethnic fragmentation i s approximated by a index that accounts for the share and number of ethnic minorities in the population. Sachs and Warner (1995) suggest that high ethnic fragmentation can affect growth through economic policies that must seek compromises between various ethnic interest groups. In the Russian Federation, ethnic fragmentation is particularly common inthe poorer republics of the Southern Federal Okrug. Fuel endowments are measured through a dummy variable for those regions with a fuel share o f GDP exceeding 14 percent. For Russian regions, this cut-off point provides a very convenient bifurcation of hydrocarbon-oriented and other territories. Only a few regions are close to this 14percent border line. Year-roundport. A dummy variable is usedfor regions that have suchaport. For arecentdetaileddiscussion ofurbangrowth andagglomeration, see OECD (2006). 22 Climate. One of the most important variables that differentiates Russian regions is climate. In particular, regions with severely cold climates have economic disadvantages inhigher production, transportation, and investment costs. The climate variables explored include average January temperature, the average o f January and July temperatures, and an indicator of whether or not a region is subjected to any permafrost. 0 Federal financial aid. During the period in question, most federal transfers to Russian regions did not have the explicit aim o f increasing growth, but of supporting the social sector in poorer areas. Nevertheless, per capita federal transfers to the regions were examined inthe growth analysis. Infact, some recent studies o f the EU finda significant positive impact of central transfers onregional growth. E. STATISTICALANALYSIS:A SUMMARYOFTHE RESULTS 2.22 The statistical analysis examined the significance of the above variables in a variety of approaches and specifications. The details o f this work can be found in Polyakov (2007). The general findings are summarized below and ina summary table inthe Appendix: 2.23 Initial Income and Convergence: Both with and without controlling for other variables, the level of initialincome inaregion appearsto have anegative and statistically significant effect on growth. This provides evidence for bothabsolute convergence (poorer regions are catching up on average) and conditional convergence (the relationshipremains significant, and becomes even stronger, when controlling for the investment rate and other variables). This i s consistent with the conclusions reached inChapter 1. 2.24 Investment Rate: As might be predicted, the investment rate (investment as a share o f GRP) has a positive and statistically significant effect on growth, suggesting that regions which invest a higher share of their income experience faster growth. The relationship between the investment rate and growth strengthens substantially if Moscow, Saint Petersburg, and the southern republicsare excluded from the sample, suggesting that these outliers were adding noise to the relationship. 2.25 Urban Agglomeration: The size (population) of the main city in a region i s found to be positively related to growth and significant in a wide variety o f specifications. This i s robust to the removal of Moscow and Saint Petersburg from the sample. Shares of the regional population living inlarger cities also appear to be positively relatedto growth, although their significance is more variable over different specifications. However, the share o f the population living in the single largest city o fthe region didnot appear significantlyrelatedto regional growth statistics. 2.26 Human Capital: The share o fpopulation with higher education appears to have a positive and significant effect on growth, although this still statistically significant relationship weakens somewhat when Moscow and Saint Petersburg are dropped from the sample. Human capital, thus defined, i s also very highly correlated with measures of urban agglomeration used here, and therefore these two variables were not examined inthe context of a single equation. 2.27 Infrastructure: The measures of infrastructure intensity did not correlate well with regional growth. O f these indicators, only the total length o f roads and the total length o f rail come out positive and significant in some statistical estimations. Yet the interpretation here i s somewhat problematic due to the strong relationship of these variables to regional size. Infrastructure intensity does not appear closely related to regional growth. But this does not 23 imply that infrastructure is not important for regional growth. Infrastructure intensity i s most likely a poor proxy for genuine infrastructure constraints on growth, i.e. a rapidly growing region might have relatively more infrastructure, but still not enough, while a stagnating region may have yet to experience sufficient growth to feel a strong bite from infrastructureconstraints. The impact o f infrastructure on growth can also be highly non-linear inthe form o f a "big push." For example, if a new road or a rail network i s essential to ignite regional growth, then a marginal expansion o f infrastructure that falls short o f this goal mighthave little impact on growth. 2.28 Investment Climate: Indicators o f the investment climate based on the OPORA Survey did not emerge as significant in the conducted statistical estimations. This was the case for specifications that related them directly to regional growth, or as part o f a two-stage estimation that made the investmentrate a function o f these variables. Part o f the problem may concern the fact that these data were only available for 2006, and may be a very inaccurate proxy for business climate constraints at the regional level during 1999-2004. On the other hand, the regional investment risk index o f RA Expert did appear significant and positive in a number o f specifications. 2.29 Ethnic fragmentation: The index o f ethnic fragmentation can be related to regional growth in a negative and statistically significant manner. . This effect is robust to the inclusion o f human capital in the specification, and to the dropping o f federal cities and southern republics from the sample. However,, ethnic diversity tends to be correlated with other variables, such as resource endowments, which makes the interpretation o f this result somewhat difficult. 2.30 Fuel Endowment: This fuel endowment dummy indicator has a positive and statistically significant effect on growth in the 1999-2004 period. This relationship weakens when the investment rate i s also included in the specification due to multicollinearity between the investment rate and fuel abundance. However, the relationshipwith the fuel indicator strengthens again when agglomeration or human capital i s also included in the specification . Overall, the results suggest that an abundant endowment o f petroleum had a significant positive effect on regional growth -this i s particularly true o f the 1999 2004 period that has witnessed significant increases in fuel production and prices. On the other hand, Chapter 1 suggests that this dynamic may have weakened inmore recent years. 2.31 Year-Round Port: The port indicator has a strong positive and statistically significant effect on regional growth. This effect is robust to various specifications that include or exclude investment, human capital, and other determinants o f growth. Introducing the port indicator in the specification results in a weakening of the relationship between growth and the investment rate, which again suggests that the investment rate i s likely highly correlated with the port indicator. 2.32 Climate (Average Temperature, Permafrost): Of the climate variables explored, the "no permafrost" indicator appears to have a positive and statistically significant relationship with growth. This relationship weakens somewhat when urban agglomeration or human capital is included in the specification, due to the fact that these phenomena are common to regions with more favorable climates. The inclusion o f a permafrost indicator also weakens somewhat the negative relationship between growth and initial GRP per capita in some specifications. Interestingly enough, the average January and July temperatures did not show significance as exogenous variables in the growth equations. However, the average January temperature does have a significant relationship with the investment rate, and therefore becomes significant in a sequential 2-stage least squares estimation o f the investment rate and growth rate. 24 2.33 Federal Financial Aid: Inall specifications considered, the federal financial aid variable was statistically insignificant or negative. This result i s robust to the exclusion o f the national republics of the Southern Okrug and the capital cities. Hence, it does not appear that federal budgetary transfers to the regions had a positive effect on growth. 2.34 Regional spillover effects: Our results show a significant and robust dependence of regional growth on growth in other regions o f geographical proximity. Estimates o f spatial spillover coefficients (the coefficient p on market potential) suggest that one extra percentage point of weighted average growth of all regions adds 0.35-0.45 percentage points to growth of an average individual region. This magnitude of spatial spillover effect i s in line with estimates made for regions inthe EU. For example, the spatial spillover coefficient was estimated at .5-.7 for the EU-15 regions (inthe NUTS classification) for 1980-95 period (Fingleton, 2003, L e Gallo and Dall'erba, 2003). The estimates here for Russian regions are somewhat lower, although two facts shouldbe kept inmind(a) EUregions are on average smaller (by area) than Russian regions and probably more integrated, (b) the analysis of Chapter 1suggests that spatial spillover effects in Russia may have become stronger in the 2004-2006 period. This fact might be confirmed further when later GRP data become available. It i s not unreasonable to conjecture that regional spillover effects in Russia are not too different now from those in the EU. A recent study (Brown et al, 2008) that makes use of enterprise level data offers important complementary evidence for the significance of regional spillover effects as measured in this CEM. In their study, access to markets, as measured through market potential, emerges as a major explanatory factor for the productivity of Russian enterprises. 2.35 The regression model (3) divides growth into three components: a spatially-determined component, a component explained by other exogenous variables, and a random component. Figure 2.5 below presents estimates of so-called spatially-determined growth rates for Russian regions during 1999-2004. This corresponds to the first component described above, and estimates the amount of growth ina givenregion that can be explainedby growth inother regions through the "market potential" term. As seen in the figure, spatial spillovers are generally more significant in the Western part of the country, where one percent of economic growth in other region. . regions weightedby market potential generates an estimated 0.3 to 0.5 percent growth in a given 25 Figu e 2.5: Spatially DeterminedGrowthinthe RussianFederation: 1999-2004 Spatially determined growth rate - 1024 - 1 026 1021 1 024 1026 - 1 027 1 027 1 028 - 1028 - 1 030 1030 1 032 - 1032 - 1 033 1 033 - 1 036 1 036 1 051 1 051 1051 -- No Data F. CONCLUSION 2.36 This chapter conducted growth equation estimates for regional growth inRussiabased on GRP data for the period o f 1999-2004. The estimations including a number of exogenous variables related to growth as well as a component that accounts for spatial spillover effects from the growth inregions o f geographic proximity. The primary conclusions are: 0 Consistent with the results of Chapter 1, no evidence was found for general divergence inGRP per capita over the period inquestion. Onthe contrary, growth has been generally pro poor in its regional dimension. Initial GRP per capita appears negatively relatedto regional growth over the period inquestion. 0 Regional spillover effects are estimated as being quite significant for Russia over the period o f 1999-2004, although not quite as strong as those estimated for the EU-15 regions from 1980-1995. However, the analysis in Chapter 1suggests that these that these spillover effects may have become stronger inRussia inmore recent years. The investmentrate, humancapital, urban agglomeration (size ofthe largest city), fuel endowments, a warm climate (no permafrost), and a year-round port appear to have a significant and positive impact on regional growth. Ethnic fragmentation and initial GRP per capita, by contrast, look to be negatively related to regional growth. Variables considered for infrastructure intensity, the investment climate (OPORA), and federal financial aid did not appear linkedto regional growth statistics, although, for the case of the former two variables, this i s likely due to an insufficient link 26 between the measures considered and infrastructure or business climate constraints at the regional level. The policy implications of the results of this chapter are considered later in Chapter 4. 27 3. RAPID REGIONAL DEVELOPMENT INRUSSIA: THE CASES OFROSTOVAND TVER A. ECONOMIC BACKGROUND 3.1 Rostov in the Southern Federal Okrug and Tver in the Central Okrug represent two Russia regions that have experienced quite rapid economic growth in recent years.' Both o f these regions still have average income levels below the median inthe Russian Federation. Thus, their rapid growth is part of the continued story of a strong pro-poor element inRussianregional growth examinedinthe previous two chapters; the relatively rapidgrowth inRostov and Tver has enabled these regions to make progress incatching up with average living standards elsewhere in the country. The Rostov economy has been expanding rapidly throughout the 1999-2007 period o f growth inRussia, while Tver has shiftedfrom sluggishto rapidgrowth only relatively recently. Both of these regions are both part of what appear to be emerging larger growth agglomerations inthe South andCentralparts ofRussia(Boxes 3.1 & 3.2). 3.2 There are a number of important differences betweenthe economies of Rostov and Tver. Perhaps most importantly, Rostov i s a central point of growth inthe South of Russia, whereas the economic picture in Tver has been profoundly affected by an agglomeration that i s centered in Moscow. In fact, a good number of people in Tver living close to the Moscow Oblast border actually work in Moscow. Relative to Tver, Rostov boasts a strong entrepreneurial tradition that was never completely subdued inthe Soviet period, as well as a stronger comparative advantage in agriculture. Rostov i s also close to the potentially unstable areas o f the Southern ethnic republics. Substantial migration into Rostov from those areas may help to alleviate the labor market constraints that are becoming increasingly serious in Russia, but can also be focus of social tensions. On the other hand, Tver faces the same demographic problems o f stagnant or declining population growth as many other Russianregions. 3.3 As Moscow, Saint Petersburg, and international capital become increasingly interested in Rostov and Tver, certain associated social and political tensions have emerged in these regions. This is particularly pronounced inTver (See Box B.2). Rostov may not face the same degree of social tensions in this regard, but the attraction and penetration o f outside capital i s also a key issue for this region. The Rostov economy has functioned largely on the basis of close ties between larger local businesses and the regional administration. Such ties helped to facilitate trust during the time of exceptionally weak institutions for contract and law enforcement. Yet a perception of favoritism toward local or politically connected businesses can also adversely affect the expectations of outside investors, and weaken competitive pressures. Like Tver and other growing Russian regions, Rostov faces strong competition for the attraction of physical and human capital, and its future success should depend on how well they can manage a transition to a more hospitable economic environment for outside capital and migrants. ' The terms "Rostov" and "Tver" are commonly used to refer either to the entire region (oblast) or the capital city o f the region. Inthis chapter, we usethe convention o f "Rostov" or "Tver" to refer to the entire region, and"Rostov city" or "Tver city" to refer to the municipality. 29 Box 3.1: Rostov Oblast Rostov oblast has a temtory o f 101 thousand square kilometers and a cument population o f 4404 thousand people. It is situated in the Southern Federal Okrug of Russia, bordering the regions o f Voronezh, Volgograd, Krasnodar, Stavropol, Kalmykia, and the country o fUkraine. Rostov has beenone o f the more rapidly growing Subjects o f the Federation since the beginning o f economic growth inRussia inthe late 1990s. Cumulative Rostov GRP growth from 1996 to 2006 amounted to an estimated 108 percent, as compared to an average o f 67 percent for Russia and 73 percent for SFO. InKrasnodar Krai, a comparable more developed southern region, GRP growth (60 percent) was only roughly halfo f that inRostov Oblast. Figure B.1: Cumulative GRPgrowth (YO),1996=1 . 2 0 6 This rapid growth reflects a number o f economic advantages relative to many other regions in Russia. Rostov has a favorable geographic location, with a relatively warm climate, fertile agriculture, an abundance o f coal and some other natural resources, and a foreign border. Rostov also has a strong SME sector, and i s a significant center o f education. Rostov has experienced an uncommon degree o f political stability under a strong regional leadership for the last 16 years. Conflicts between regional and municipal bodies, which have been problematic in a number o f Russian regions, appear much less pronounced in Rostov. In contrast to a number o f large Russian cities that have been experiencing marked population decline, the population o f the city o f Rostov has been relatively stable, which reflects its position as a center o f economic opportunity inthe densely-populated Russian South. As the capital o f the strategically important Southern Federal Okrug, the city o f Rostov also receives exceptional political attention from the federal government, which hopes that a thriving Rostov, as well as Krasnodar, may help to pull up neighboring ethnic republics that continue to suffer from poverty, internal conflicts, and political instability. Interviews with entrepreneurs, business, and civil society confirm perceptions o f many o f these advantages inRostov, particularly the relative stability of "rules of the game" for business. Under these rules of the game, a number o f smaller businesses are able to compete relatively free o f interference, which has contributed to a relatively well developed service sector in the region. On the other hand, larger firms typically operate inclose cooperation with the regional administration. While corruption is perceived as significant, larger businesses nevertheless praise the fact that informal relations between business and state organs have become more stable andpredictable. Rostov still faces some serious development challenges. Although a rapidly growing region, Rostov i s relatively poor, with average incomes 25 percent below the Russian average. Some o f this discrepancy is likely due to a relatively large informal economy and unreported incomes, which i s a challenge to the region, in and o f itself. Investment rates at 20 percent o f GRP are rather low in Rostov, considering the rapid growth that the region has been experiencing. The neighboring Krasnodar region, for example, invests 28 percent o f GRP, and had attracted cumulative foreign investment o f US$ 3.2 billion from 1996- 2006, as opposed to only US$ 344 million in Rostov. A large share o f economic activity in Rostov i s oriented only to the regional market. Like a numbero f other regions inRussia, Rostov has yet to cross the threshold from a growth agglomeration feeding largely off the internal regional market to a region with strong competitive niches on the greater Russianand world markets. 30 Box 3.2: Tver Oblast Tver oblast has a territory o f 84.2 thousand squarekilometers and a populationo f 1390.4 thousand people. I t s large geographic size can be associated with a lot o f diversity within the region. Some parts o f Tver are virtually part o f the greater Moscow region, while other parts are quite remote and provincial. Tver is a regionrichly endowed with water and forests, i s the main natural source o f fresh water for Central Russia, and possessesa substantial stock o f mineral resources. Tver industry i s characterized by a number o f very large plants inherited from the Soviet period, including electricity and the production o f train cars. Although the Tver region, like Rostov, has average household income that is about 25 percent below the Russian average, it also has a relatively low poverty rate (According to estimates using NOBUS data, the poverty headcount inTver stood at 20 percent in2003, as opposed to 23 percent for Russia as a whole). Incontrast to Rostov, Tver's takeoff into rapid growth has been a very recent phenomenon. Tver was a relative underperformer in the growth that began from 1999-2002. Cumulative Tver GRP growth from 1996-2006 amounted to an estimated 46 percent, which can be compared to 67 percent for Russia as a whole. Growth has since accelerated, however, particularly since 2006, when GRP growth apparently exceeded 10percent. The recent growth is no doubt related to increasing demand and economic interest in Tver coming from the booming greater Moscow region and, to some degree, the agglomeration around Saint Petersburg. An important change in political leadership in Tver oblast in 2004 supported a shift toward a more outward looking economic strategy aimed at attracting investment to the region. FigureB.2 CumulativeGFU GrowthinTver Oblast, the CentralFederalOkrug, andtheRussian Federation , 1 0 I / I 1.7- L.5- 1.3- 1.1- 0.9 _I Cumulative per capita foreign investment in Tver oblast for 1996-2006amounted to only US$ 35.5, as compared to an average inthe Central Federal Okrug (excluding Moscow) o fUS$250. FDIhas picked up notably in recent years, however, along with the economic expansion. In 2007 alone, Tver received an estimated US$ 77.9 million inforeign direct investment. The rapid recent penetration o f Tver by Moscow and other outside capital can be associated with significant social and economic tensions. Rentals and property values have soared in the Tver region in recent years, particularly in areas closer to Moscow. This has placed a number o f local entrepreneurs who rent their premises ina very difficult position. There have been open protests in the region with demands for more protection o f local businesses. Although the Tver region leadership appears intent on realizing an economic transition that would bring more outside capital to the region and fully integrate it with the surrounding booming economy, it faces a difficult task in managing the social tensions and short-run costs o fthis strategy. While Tver has a very advantageous geographical position on the road between Moscow and Saint Petersburg, it i s still in competition with other regions inthe Central Federal Okrug that possess the same kindo f advantages. A primary question for Tver's future is the degree to which it succeeds inbecoming a center o f economic growth itself, as opposed to a peripheral area that services Moscow and Saint Petersburg. 31 3.4 This CEM conducted business surveys in both the Tver and Rostov regions with the primary goal of identifying the sources of, and constraints to, growth at the regional level. Surveys o f 421 firms in Tver and 535 firms in Rostov were conducted in 2007 according to the same methodology as BEEPS that has been conducted by the World Bank and the European Bank for Reconstructionand Development (EBRD) inEast Europe and the former Soviet Union every three years since 1999. The survey also included other questions that are more specific to issues of regional development and the activities of firms. More complete write-ups of these surveys, including details on the samples, can be found in World Bank (2008a, 2008b). This section summarizes a few o f the results of this survey that are particularly relevant for understanding the challenges facing emerging fast-growing Russianregions such as Rostov andTver. B. THEBUSINESSSURVEY 3.5 The sample for the survey was chosen according to the same methodology as BEEPS, and corresponds closely to the structure o f the regional economies. The breakdown o f the samples by sector was similar for Rostov and Tver, although Rostov has a somewhat greater representation intrade relative to goods production incomparisonwith Tver, corresponding to the structures o f the regional economies. BEEPS givesparticular attentionto entrepreneurs and small businesses. Seventy and fifty two percent of the firms in the Rostov and Tver surveys, respectively, had fewer than 50 employees. This i s important for the interpretation of the results, as economic developments often affect larger and smaller businesses differently. When necessary, an attempt is made to distinguishthese cases. The BusinessClimateinTver andRostov GeneralPerceptions - 3.6 Entrepreneurs were askedtheir general assessments of the business climate of the region, how it has changed in recent years, and its comparison with Moscow, Saint Petersburg, and surrounding regions. This question was meant to capture subjective perceptions, as opposed to actual reliable information, as many businesses may have only very limited information about business conditions in other regions. In fact, close to 20 percent of entrepreneurs inboth Rostov and Tver refused to answer this question for that reason. Nevertheless, these sorts of perceptions themselves can be key elements of the business climate. 3.7 On this note, it i s interesting that entrepreneurs in Rostov evaluate their local business climate relative to surrounding regions significantly more favorably than those in Tver (Figures 3.1 and 3.2). Two important reasons for this may be (a) regions surrounding Tver are generally in a more advantageous economic situation than those surrounding Rostov and (b) the particularly strong tensions in Tver associated with the recent spread of the Moscow agglomeration. Entrepreneurs in both regions generally perceive that the business environment in Moscow or Saint Petersburg i s better than in their region. Yet, Rostov entrepreneurs felt that their own business environment i s preferable to that in the geographically close regions o f Voronezh, Volgograd, Kalmykia, Stavropol, and comparable to that in Krasnodar. On balance, Tver entrepreneurs rated their local environment worse than in neighboring Vologda, Novgorod, Pskov, Yaroslavl, and Smolensk. Furthermore, pessimistic attitudes toward the business climate of the region relative to its neighbors were common among firms o f all size inTver, incontrast to Rostov, where larger firms were notably more optimistic inthis regard. Ineach case, the number on the chart measures the net percentage of answers that rate the climate of the region better (plus) or worse (negative) than ina givenneighboringregion. 32 Figure 3.1: Net Percentage of Tver Entrepreneurs Claiming that the BusinessEnvironment in their Region is Better (Positive) or Worse (Negative) than in a Given Other Region r I , I I , , , I I I -45 -40 -35 -30 -25 -20 -15 -10 -5 0 Figure 3.2: Net Percentage of Rostov Entrepreneurs Claiming that the BusinessEnvironment in their Regions is Better (Positive) or Worse (Negative) than in a Given Other Region -25 -20 -15 -10 -5 0 5 10 15 20 25 3.8 Followingthe BEEPSsurvey, entrepreneurs inTver and Rostov answered questions about barriers to business in a number of different categories. The Figurebelow indicates the shares of firms in Rostov, Tver, and Russia (2005) as a whole that indicated a particular problem as being a significant barrier to business. 33 Figure3.3: The Shareof EntrepreneursIndicatingthat a GivenProblemis a Serious Impedimentto Business(Tver (2007), Rostov (2007), Russia(2005)) 60% 50% 40% 30% 20% 10% 0% 14 9 15 8 2 1 20 13 17 11 21 16 18 10 7 6 19 12 5 4 3 Russia2005 Tver 0Rostov 1-Access to finance; 2-Cost of finance; 3-Telecommunications; 4-Electricity; 5-Transport; 6-Access to land; 7-Land right or lease; 8-Tax rates; 9-Tax policy; 10-Rules of customs and trade; 11-Licenses and approvals; 12-Labor legislation; 13-Work habits and education level inthe labor market; 14-Instability of macroeconomic policy; 15-Instability of macroeconomic conditions (inflation, exchange rate); 16- Functioning of the judicial system; 17-Corruption; 18-Street crime/robbery/disorder; 19-Organized crimeimafia; 20-Monopolistic practices; 21-Breaches o f contractual agreements. 3.9 As can be seen inFigure 3.3, the evaluation o f the business climate by entrepreneurs in Tver and Rostov i s largely consistent with that for Russia as a whole in 2005, although more favorable in a number of areas. Some o f these more favorable responses, including the macroeconomic stability, macroeconomic policy stability, and taxation, might be explained by time as opposed to geography. These areas have improved for the country as a whole since the all-Russia survey o f 2005. Similarly, the Russian labor market has tightened considerably since 2005, which i s no doubt reflected in a larger share o f complaints from both Tver and Rostov about the labor market and shortages of slulled labor. The forthcoming national BEEPS survey o f 2008 will most likely confirmthese trends for the entire country. 3.10 A most notable difference between Rostov and Tver concerns the perceptions of larger and smaller businesses. In Tver, this difference was not large. In Rostov, by contrast, small entrepreneurs gave considerably more optimistic assessments o f the business climate than did their larger counterparts. This i s visible inFigure 3.4, which graphs only the assessments only of medium and large-sized enterprises o f the barriers listed above. Medium and large-sized enterprises in Tver appear considerably less concerned about barriers to business than their counterparts inRostov. As argued by Shaffer et al, this does not, inand o f itself, imply that these barriers are more serious for larger businesses inRostova2But it does imply that alleviating these Shaffer et a1 (2007) demonstrate a very important regularity that more productive and competitive firms tend to complain more about business climate constraints (other than access to finance) than do other firms. Thus, more 34 constraints would probably have a particularly strong positive impact on larger businesses in Rostov. Figure3.4: The Shareof Mediumand LargeEnterprisesinTver and Rostov that IndicatedaGiven Constraintis aSerious Impedimentto their Business "Is" 1-Access to finance; 2-Cost o f finance; 3-Telecommunications; 4-Electricity; 5-Transport; 6-Access to land; 7-Land right or lease; 8-Tax rates; 9-Tax policy; 10-Rules o f customs and trade; 11-Licenses and approvals; 12-Labor legislation; 13-Work habits and education level inthe labor market; 14-Instability o f macroeconomic policy; 15-Instability of macroeconomic conditions (inflation, exchange rate); 16- Functioning of the judicial system; 17-Corruption; 18-Street crimehobberyldisorder; 19-Organized crimeimafia; 20-Monopolistic practices; 21-Breaches of contractual agreements. 3.11 Interestingly enough, entrepreneurs inboth Tver and Rostov did not give very favorable responses about changes in the regional business climate since 2005-2006. Many entrepreneurs in both regions perceived little change. Among those that did, the balance o f answers was negative in all areas except access to outside finance and utilities (Figure 3.5). Responses in Rostov were somewhat more pessimistic than those in Tver. Part o f the explanation for these negative perceptions may be the increased competitive pressures that many firms in Tver and Rostov have faced inrecent years associated with the appreciation o f the ruble, the penetration o f the local market by outside capital, and the tightening o f the labor market. It is somewhat disturbing, however, that some o f the areas of greatest perceived deterioration correspond to variables related to market institutions and law and order: violations o f contractual agreements, unfair competitive practices, corruption, criminality, and the functioning o f the court system. While greater competitive pressures, in and o f themselves, might explain much o f the perception o f an increase in "unfair competition," most of these other variables should have a positive dynamic in a region that i s making steady progress to an open market economy. One possible reason for this regularity i s that, as the Russian economy experiences rapid growth and complaints in one region relative to another could be due to worse constraints (negative) or to the presence of more highly productive firms (positive). 35 development, the demand for developed market institutions i s increasing, and thus institutional weaknesses are becoming more notable and problematic. Figure 3.5: Perceptions of Rostov Entrepreneurs of Changes inthe BusinessClimate in the Regions During 2005-2006 (Net Percentage Indicating Worsening (Negative) or Improvement (Positive) 16.0 11.0 6.0 1.o -4.0 -9.0 -14.0 -19.0 -24.0 1.. ^___._^I__ ._,".__^"",..I"^" ~ I_ ^, QRostov ~~ ~~ ~ Electricity, 6 - Unfair competition o f other market players, 7 - Macroeconomcsvolatility ( Inflation, Exchange 1 - Cost of Funds ( interest rate, fees), 2 - Telecommunication, 3 - Financing, 4 - Labor legislation, 5 - rate), 8 and leasing, 13 - Street crime/theft/disorder, 14 - Land access, 15 - Skills and education level on labor market, - Corruption, 9 - Organized crime/ Mafia, 10 - Tax rates, 11 - Functioning ofjudiciary, 12 - Land title 20 - Contractual arrangements violations by customers and suppliers, 21 - Tax policy 16 - Customs and trade regulations, 17 - Economic policy volatility, 18 - Licensing & pemts, 19 - Transport, 36 Figure 3.6: Perceptions of Tver Entrepreneurs of Changes in the Business Climate in the Regions During 2005-2006 (Net Percentage Indicating Worsening (Negative) or Improvement (Positive) 6 1 -4 -9 -14 -19 0Tver - Exchange rate), 8 - Corruption, 9 - Organized crime/ Mafia, 10 Tax rates, 11 Functioning ofjudiciary, 12 - Electricity, 6 - Unfair competition of other market players, 7 - Macroeconomics volatility 1 - Cost of Funds ( interest rate, fees), 2 - Telecommunication, 3 - Financing, 4 Labor legislation, 5 - (Inflation, Land title and leasing, 13 - - -- Skillsand education level on labor market, 16 - Customs and trade regulations, 17 - Economic policy volatility, 18 - Licensing & permits, 19 Street crime/theft/disorder, 14 - Land access, 15 - Transport, 20 - Contractual arrangementsviolations by customers and suppliers, 21 Tax policy - I Administrative Barriers to Business 3.12 Many business surveys give particular attention to so-called "administrative barriers to business," as these refer to processes or institutions that are largely under control o f government: red tape and administrative procedures for registration, licensing, certification, land access or ownership, price regulations, consumer or ecological standards, and various state inspections. A "debureacratization" campaign o f the federal government since 2001 has sought to reduce these barriers. This campaign appears to have met with some success, although progress since 2003 has been much more m ~ d e s t . ~Such administrative barriers continue to be cited by businesses as important impediments to their activities throughout Russia. Perceptions in Rostov on the relative severity o f such barriers mirror closely those in Tver (Figure 3.7). Interestingly enough, in both regions, the largest share of entrepreneurs complained about a major burden of "paper work" inpreparing various required statements and accounts. This problem also received a larger number o f complaints in Rostov than in Tver, as did the general problem o f various inspections and procedures for gaining access to land. A relatively high number o f complaints in Rostov about various inspections i s consistent with evidence, also revealed inthe survey, that larger firms inRostovexperience more inspections, on average, than inTver or Russia as awhole. SEFIR(2007). 37 Figure3.7: The Share of RespondentsinRostov and Tver that Indicatea GivenAdministrative Barrier is a Significant Impedimentto Business(% of Responses) -CRostov Oblast Registration Mandatory reporting requirements "excessive paperwork" Certification Price regulation regulation Sanitary-epidemiological semi Purchase of premises The CourtSystem and Relations with State Authorities 3.13 The development o f an effective and impartial judiciary system has proved to be one o f the most difficult tasks o f economic transition. Progress i s certainly visible. Russian firms have been using the courts much more often for the resolution o f disputes. This i s very much true for Rostov and Tver. In fact, a considerably larger share o f Tver (45%) and Rostov (40%) firms reported to have been either defendants or initiated court cases in the last two years than was the case for Russia as a whole in 2005 (27%). It i s an interesting question as to the degree that this mightbe the general case in rapidly growing regions. At the same time, like their counterparts in the rest o f Russia, entrepreneurs in Rostov and Tver are rather skeptical about the impartiality o f the courts. Rostov entrepreneurs appear a bit more optimistic: Forty eight percent o f Rostov entrepreneurs and 40 percent o f Tver firms agreed partially or fully with the statement "I feel confident that the judicial system will defend my property rights and business agreements." Only minorities o f firms inRostov and Tver believe that the court system can be characterized as fair, uncorrupted, or easy to access. Despite this pessimism, these numbers were slightly better than those obtained for Russia as a whole in2005, particularly those for Rostov. 3.14 The relative optimism o f Rostov entrepreneurs vanishes, however, when asked about their vulnerability to the whims o f state officials. When asked about the possibility o f overturning a decision inviolation o f norms or laws made by a state official through an appeal to another body or higher authority without resort to bribery, Rostov firms showed even more pessimism than their counterparts in Tver or Russia. The situation in Tver from this point o f view i s relatively encouraging, with 33 percent confident o f overturning such a decision made by a local government and 25 percent for the regional or federal government. Perhaps this i s related to stronger divisions in the power structure within the Tver region that could enhance opportunities for appeals to alternative bodies. 38 Figure3.8: The Share(%) ofEntrepreneursinRostov,Tver, and Russia(2005) that Believeit Possibleto Overturna DecisioninViolation of Normsor Laws madeby a StateOfficialThrough an Appeal to Another Bodyor Higher Authority Without Resortto Bribery 40.0 - 35.0 - --.- 9, F 30.0 - 25.1 25.0 126.3 25.0 - 20.0 -- 15.7 15.0 -- 10.0 -- 5.0 -- 0.0 -I Federal I Oblast Local O Tver Rostov I3 Russia (2005d 3.15 Despite the results reported in Figure 3.9, Rostov firms appear to have a much more favorable assessment of the influence o f different levels o f government on their business than do firms in Tver. The assessments o f larger firms in Rostov were also substantially more positive than those o f smaller firms. This i s quite consistent with anecdotal evidence that the business model in Rostov i s based on close relations between larger firms and government. In Tver, by contrast, the balance o f perceptions o f state influence on business i s negative. When asked about perceptions o f changes in business-government relations over the past two years, there was a strong difference inresponses inboth Tver and Rostov between small and larger businesses, with larger businesses perceiving net improvements, particularly in relations with local and regional authorities, and smaller businesses perceivinglittle change or net deterioration. Figure3.9: The Net Assessmentsof Rostov and Tver Entrepreneurson the Influence of Different Levels of Governmenton Their Business 10 5 0 -5 -10 -7.2 -15 -20 16.3 I 1 Tver a Rostov 1 39 Competition and Growth 3.16 The questionnaire included a number o f questions on competition, competitive pressures, and their relationship to firm behavior. A primary result of the recent Investment Climate Assessment (ICA) of the Russian Federation is that competitive pressures can be positively associated with technical progress, i.e. those firms facing greater competitive pressures engage more frequently in innovative activities or m~dernization.~A related study that made use of the ICA dataset suggested a certain bifurcation of Russian enterprises into very different groups, including highly competitive firms, usually active on all-Russian or international markets, and firms of muchlower productivity orientedonly toward local market^.^ 3.17 Rostov and Tver oblasts appear consistent with this picture. Inboth Rostov and Tver, one third of all firms responded that they do not experience significant competitive pressures at all. A slightly higher percentage o f firms in Tver (43 percent) than in Rostov (38 percent) indicated that they experience significant outside competitive pressures from Moscow, Saint Petersburg, or foreign firms. The remaining group o f firms (between % and 1/3) responded that they experience competitive pressures only from local firms operating inthe region.6 3.18 Ingeneral, entrepreneurs inRostov are a bit more concerned with competitive pressures than their counterparts in Tver. This i s illustrated in Figure 3.10 below. Among types o f competitive pressures, only those from Moscow and Saint Petersburg are perceived as roughly comparable inTver as inRostov. Given Tver's geographic position, it i s rather surprisingthat the pressures from Moscow and Saint Petersburg are not perceived as greater than in Rostov. Pressures coming from import competition, other firms in the region, and firms in other regions are all perceivedas significant by a larger share of firms inRostov than inTver. 3.19 As mightbe expected, industrial firms and larger firms report somewhat more significant outside competition. Medium and larger enterprises also report more competition with imports and outside firms on average than do smaller enterprises inboth regions. But even less than half of larger firms inRostov and Tver report significant competitive pressures from imports. Over 50 percent o f medium-large Rostov firms, and close to 70 percent o f medium-large Tver entrepreneurs, do report significant competition with outside firms, however. Desai and Golberg (2007). Russian Firms at the Crossroads (2007). 6 The surprising low share of Russian firms experiencing competitive pressures is consistent with at least two other recent surveys carried out inthe country, the Investment Climate Assessment survey (the results written up inRussian Firms at the Crossroads...(2007)) and Fry and Yakovlev (2007), who conducted a survey of firms in Moscow, Bashkortostan, Voronezh, Nizhny Novgorod, Sverdlovsk, Smolensk, and Tula. 40 Figure 3.10: Perceived Competitive Pressures inRostov and Tver: The Share (YO)of Respondents Indicating that Competitive Pressures from a Given Source are Significant Sverd 60.0 7 I 52 3 50.0 40.0 30.0 20.0 10.0 0.0 from import from Moscow or Saint Petersburg from other firms in the region from firms of other regions 3.20 The ICA presented evidence that the presence of competitive pressures inRussia can be a stimulus to restructuringand technical progress at the enterprise level. The data from Rostov and Tver are quite consistent with this notion, and give even stronger such evidence. Figure 16 illustrates the relationship between the status of "no competitive pressures," "only local competitive pressures," and "outside competitive pressures" with the percentage of enterprises that undertook activities associated with restructuring, technical progress, or innovation. More competitive pressures can be associated with a greater share of such activities. The l o w share of restructuring or modernization activities taken by the "no competition" group in Rostov i s particularly notable. In Tver, no real distinction emerged between the "no competitive pressures" and "only local competitive pressures," but the distinction between these categories and "outside competitive pressures" i s even more pronounced (Figures 3.10 & 3.11). 41 Figure 3.11: The Share (YO)of RespondentsinRostovEngagingin a GivenActivity Differentiated by Responsesto Questionson CompetitivePressures The inkduclbnofa wwpro&cl -0 competitive pressures +criblccdcmpetnivepressures -0ttsLJecompeflve pesstres Figure3.12: The Share (YO)of Respondentsin Tver Engagingin a GivenActivity Differentiatedby Responsesto Questionson Competitive Pressures The introduction of a new product Quality certification obtained License obtained +tdo compel Ive press-res +On y oca1compel w e pressms OLIslde compelat ve pressuresI 42 3.21 To what degree can greater competitive pressures be associated with growth in output or investment in specific firms? The shares of Rostov and Tver firms in the sample reporting growth in output in 2005-2006 were 58 and 69 percent, respectively. As could be expected, reported output growth i s strongly correlated with reported investment, although roughly half of all firms in both regions reporting significant output expansion did so without increasing their capital. The relationship between competitive pressures, output expansion, and capital expansion i s given in Figures 3.13 and 3.14. This correlation i s notably positive, particular for the case of outside competition and the expansion o f capital. The same difference between Rostov and Tver noted for Figure 3.12 above i s present in figures 3.13 and 3.14: the difference between "no competitive pressures" and "only local competitive pressures" is significant for Rostovbutnot for Tver. On the other hand, the group of firms subject to outside competition is quite distinct inthe degree that they have expanded output and (especially) capital, particularly inTver. Figure3.13: CompetitionandGrowth inRostov andTver Oblasts: The Share (%) of Firms that RecentlyExpandedOutput DifferentiatedbyResponsesto Questions on CompetitivePressures 50 i 45 Rostov 40 Twr 35 I I 30 25 20 15 10 5 0 No Competitive Pressures Only Local Competive Outside Competitive Pressures Pressures 43 Figure3.14: CompetitionandInvestmentinRostov andTver Oblasts: The Share (YO)of Firms that RecentlyExpandedtheir CapitalDifferentiatedbyResponsesto Questions on Competitive Pressures 5 0 0 Rostov Tver .It 5 I I 'O 5 0 , No Competitive Pressures Only Local Competive Outside Competitive Pressures Pressures ~ ~~ ~ C. CONCLUSION 3.22 This chapter examined questions pertaining to the business climate o f two relatively poor Russian regions that have been experiencing rapid growth in recent years. Rostov has been a center o f growth in the South o f Russia since 1999, whereas growth in Tver has accelerated only recently with the spread o f the Moscow and Saint Petersburg agglomerations. A regional level enterprise survey conducted inbothregions offers information on the perceptions and activities of entrepreneurs inbothregions. 0 The survey i s consistent with the picture o f Rostov and Tver as growing regions largely hospitable to expanding business activities. Perceptions of barriers to business in these regions are consistent with, and somewhat more favorable, than those documented for Russia as a whole in 2005. Perceptions o f changes in the business climate inrecent years were not very favorable. Part o f the reason for this may be the increased competitive pressures that both regions have faced in recent years. Pessimism in both regions about progress in the development o f formal market institutions (the court system, competition policy, corruption) i s somewhat disturbing. However, it follows the general patterninrecent business surveys for Russia. Inthe Tver region, perceptions ofthe business climate are fairly uniformacross firms o f different size. In Rostov, by contrast, smaller firms give significantly more favorable assessments o f the business climate than do larger firms. Medium and larger-sized firms are much more concerned about business climate constraints than 44 are there counterparts in Tver. Interestingly enough, both Tver and Rostov entrepreneurs complain about excessive paper work more than any other administrativebarrier to business, including various inspections. Larger Rostov firms would also appear to face a higher-than-average burden for Russia of various inspections. Rostov entrepreneurs generally rank the business climate in their region favorably incomparison to surrounding regions, whereas Tver firms do not. Although Rostov and Tver share the skepticism and mistrust of their counterparts in other regions about the court system, an impressive number o f entrepreneurs inboth regions do appeal to the courts for the settlement o f many disputes. The skepticism becomes greater for overturning violations of laws or norms by government officials, particularly inRostov. Tver entrepreneurs are notably more optimistic inthis regard. In general, Rostov and Tver entrepreneurs have very different perceptions of the influence of various levels of government on economic activity, with the balance of influencedperceived as positive inRostov and negative inTver. 0 The results o f the Rostov and Tver surveys give strong additional evidence o f a positive relationship between competition (competitive pressures), on the one hand, and modernization (technical progress) and business expansion, on the other. Those firms that claim to face significant competitive pressures from firms outside of the region engage in significantly more activities associated with modernization, restructuring, and investment. In Rostov oblast, the difference between firms facing no significant competitive pressures and local (within the region) competitive pressures i s also notable. 3-23 How should these results be interpretedfor assessingthe present and future prospects for the development o f the Rostov and Tver regions? Both of these regions face social and economic tensions associatedwith rising outside competition. Despite the short-run costs of policies aimed at facilitating the integration of regional with larger markets, the survey provides direct evidence of potential benefits. It appears that the many of the firms inboth regions that offers the most hope for sustaining rapid growth (those actively engaged in modernization, restructuring, and output expansion) are already competing on all Russian or international markets. Rostov and Tver will need to manage the short-run transition costs, which are particularly notable in Tver, but heavy-handed regional-level protectionist policies would most likely end up rewarding less promising firms and hindering the activities of the most promising. 3.24 The survey i s also consistent with the general picture o f Rostov as a region where informal relations can often be o f more importance than explicit institutions. At the same time that many entrepreneurs note improvements in relations with authorities, there i s a widespread perception of deterioration during 2005-2006 of precisely those variables that relate to the functioning of explicit institutions: unfair competitive practices, violations o f contractual agreements, criminality, corruption, and the court system. Attitudes are particularly pessimistic in Rostov about possibility of overturning unfair decisions or violations made by state officials. The sustainability of Rostov's rapid growth into the mediumterm will most likely dependon the strengthening o f explicit (impersonal) institutions relative to informal relations, which would make the region more attractive to outside investmentand fair competition. As noted inBox 3.1, Rostov's recent record in economic growth has been far more impressive than in investment growth. This cannot be sustainedindefinitely. Outside capital i s needed. 3.25 Although the situation in Tver oblast looks to be somewhat different to that described above for Rostov, the future challenges are somewhat similar. An even larger share of firms in Tver look to be oriented only to local markets, and many operate under very limited competitive pressures. Furthermore, in sharp contrast to Rostov, Tver entrepreneurs consider the business climate in their region to be generally worse than in surrounding regions. This difference could be due as much to being surrounded by regions with better business climates as opposed to a worse business climate inTver than inRostov. Yet this i s part o f Tver's dilemma. They are ina very advantageous geographical position for participating in the growing growth agglomeration incentral Russia,butso are anumber ofother regions that are indirectcompetitionwith Tver for attracting capital and labor. Furthermore, many o f these other regions got a head start on Tver by attracting substantial outside capital to their regions before 2005. This has made Tver's recent transition more radical and more disruptive to local interests than in a number of other regions in central Russia. The political divisions within Tver oblast, and the lack of a single omnipotent source o f state power like in Rostov, appear to have made even larger businesses in the region more independent, and more willing to rely on the courts and other explicit institutions than in Rostov. Tver firms were much less favorable intheir assessments of the influence o f government on their business, although these answers are difficult to interpret in the current period of transition when many firms believe that the role of regional and local authorities should be protectingthem from risingrentals or outside investors. 3.26 The rapidly-growing regions of Rostov and Tver unquestionably have strong future potential. So far, both regions have grown in a largely self-contained manner in response to rising demand inthe region and neighboring regions. The creation of growth agglomerations in the southern and central parts of Russia have reinforced this growth, as described in Chapter 3. Yet bothregions have yet to make the transition to developinghighly competitive niches on all- Russian or world markets. The survey provides evidence suggesting that unleashing the potential o f Rostov and Tver in that direction will depend on promoting competition and providing a business climate that i s hospitable to market entry and investment 46 4. CONFRONTING THE CHALLENGES OFREGIONAL DEVELOPMENT INTHE RUSSIANFEDERATION 4.1 The exceptionally large size and diversity o f the Russian Federation, together with the inherited legacy of Soviet planning, presents a somewhat unique set of challenges in regional development. While regional development has long been a major topic of debate inthe country, the Russian government has yet to articulate a comprehensive strategy for meeting these challenges. This reflects the economic and political complexity o f the issues involved, as well as some significant disagreements inRussianpolicy circles about the type o fregional strategy that i s needed. This chapter outlines some of the central economic issues involved, in light of the conclusions of the preceding chapters, and makes suggestions concerning some o f the key building blocks for constructive strategy. A typology of economic policies is proposed, and particular attention i s given to the critical and controversial issues of migration and regional development policy. 4.2 Infact, Russiahasbeen already adoptinga number ofeconomic policies that are roughly consistent with common or best practices in promoting regional development, although the absence of a unifyingnational strategy may compromise coordination and consistency. Giventhe exceptional degree of uncertainty inRussia on the economic potential o f many regions, the large number of settlements in lagging regions, and current increasing returns in areas of agglomeration, it i s important that depressed regions are given the opportunity to realize initiatives themselves aimed at boosting productivity and growth. But an over-centralized approach that stifles regional and local initiatives will likely generate disappointing results. A. RECENTINITIATIVES INRUSSIA RELATEDTO REGIONALDEVELOPMENT 4.3 The federal government had quite limited means in the 1990s to pursue any sort of regional development agenda. Major regional inequalities that emerged during this time, given very unequal inherited endowments for market competitiveness, were reinforced by the delegation of most social expenditures to the regional level. In this context, wealthier regions could maintain relatively generous per capita social expenditures, as opposed to many o f the poorest regions that may have needed them the most.' 4.4 As indicated inChapters 1and 2, the very broad-based economic recovery of 1999-2003 can be associated with quite favorable conditions for rapid industrial growth in poorer regions. Thus, even without government intervention, poorer regions, on average, achieved relatively rapid growth and poverty reduction. Government policy initiatives concentrated on national issues of importance to most or all Russian regions, including macroeconomic stabilization, tax reform, and the business climate. Significant progress was also made infiscal federalist relations, including a clarification inthe division of budgetary autonomy, revenues, and expenditures. The formula-based transfer system was also modified to concentrate more federal assistance in the poorest regions. 'On thispoint, see OECDEconomicSurveys...(2000), Chapter 3. 47 4.5 Inmore recent years, concernshave increased inRussia over perceived growing regional inequality and disparities. The analysis of Chapters 1 & 2 suggests that, contrary to popular opinion inmany circles, no major general trendtoward regional inequality emerged during2004- 2006. Duringthis time, quite a number of poorer Russian regions continued to exhibit relatively rapid growth. On the other hand, the analysis also confirmed that regional growth has become much more unevenly distributedamong Russian regions, and that a considerable gap appears to be opening up between growth agglomerations in the European part of the country (Central, North Western, and Southern Okrugs) and most o f the Eastern part of Russia. Furthermore, the evidence presented for the emergence o f growth agglomerations inRussia suggests that there may be no automatic market mechanism for correcting inequalitiesbetween regions over the medium term in productivity, growth, and welfare.2 Thus, future trends may very well revert to divergence. Some poorer regions in the Eastern and Northern part of the country are already sufferinginthisregard. 4.6 The Russian government has taken a number of initiatives in recent years to boost the policy agenda for regional development. In2004, it created a Ministry of Regional Development. This Ministry began work almost immediately on a National Strategy for the Socio-Economic Development of Russian Regions. The first draft Strategy presented an ambitious agenda under the banner o f rejecting equalization policies in favor of supporting points of growth and promoting agglomerations. It also called on individual regions to create their own development strategies in a manner consistent with the national ~trategy.~ As o f early 2007, Russian regions had submitted 54 such strategies, 13 of which were approved by Moscow. 4.7 The draft National Strategy of the Ministry o f Regional Development generated a lot of controversy inRussianpolicy circles, particularly among those citingboth economic and political reasons against the neglect of poorer or less fortunate regions. On the contrary, a number o f specialists argue, growing contrasts among more and less fortunate regions present a threat to national cohesion and the welfare of a significant number of settlements! In2006, an alternative draft strategy was prepared by the Working Group for Complex Social-Economic Planning o f the Development o f Regions of the Government Council of the Russian Federation. This draft placed its emphasis on the "joint coordinated interaction o f federal and regional organs for developing programs for the comprehensive social-economic development o f the region^."^ The concept placed a lot of emphasis on the restoration of Soviet-type territorial planning at the federal and regional levels, and also generatedsignificant controversy. 4.8 Inthiscontext, theMinistryofRegionalPolicycontinuedtopushfor ageneralconsensus around a National Strategy, organizing a number o f seminars in late 2006, encouraging debates within regions, and a convening a national conference in early 2007. A finalized National Strategy was to emerge following this conference. Butremaining controversies proved too strong to achieve this objective. Neoclassicaleconomic theory presumes decreasingreturns to scale to labor and capital within a given regional area. Under these conditions, the reallocationof capital form less productive to more productive regions will decrease the relative marginal returns in the leading regions, thus pushing the economy in the direction of an equilibrium where marginalreturns are equalized. The presence of strong agglomerationeffects, as suggestedinthe analysis in Chapters 1 and2 for the case of Russia, implies increasingreturns to scale over a wide range, and thus the gap between leading andlaggingregions cannaturally be expectedto grow over time. "Concept for a Strategy..".(2006). See, for example, Zubarevsky(2007a). "On aMechanism..."(2006). 48 4.9 The Ministryof RegionalPolicy receivedanimportant boost inits power andauthority in 2007 for the (infrastructure) Investment Fund (formally under the sole management of the Ministry of Economic Development and Trade) and federal transfers to the regions (formally under the sole management of the Ministry of Finance). The new leadership of the Ministry of Regional Development would appear to have somewhat different views about the desired nature of a national strategy for regional development. In a recent interview, the new Minister of Regional Development indicatedthat achieving regionally balanced development i s one of the top priorities of the Ministry.6 Thus, a consensus strategy for regional development has still yet to emerge within the Russian government. Box 4.1: WorldDevelopmentReport 2009: TerritorialDevelopment The World Bank World Development Report for 2009 is devoted to the subject o f territorial (regional) development, and will provide a large amount o f valuable information on recent world experience and policy lessons. A few o fthe general emerging conclusions from this report are: 0 Agglomeration effects are important. The spatial concentration o f growth is common in most countries in the world, and is even stronger in developing countries. The concentration o f economic mass is inevitable and generally desirable. With economic development, spatial concentrationusually first rises, and then levels o f fas economies reachhigher levels o f income. 0 Spatial disparities inliving standards are neither desirable nor inevitable. They can be addressed through economic integration. Labor mobility i s the strongest natural mechanism to enhance agglomeration economies and facilitate convergence. Balanced development may very well require unbalanced growth. 0 Measures to facilitate integration include spatially-blind policies (not targeted to specific regions and facilitating the movement o f people and capital toward opportunities), spatially connecting policies (infrastructure to better link regions and markets), and spatially-targeted policies (to support growth inspecific regions). 0 Policies for lagging regions should prioritize investments in people over measures to promote growth in the local economy. This includes social support (basic education, health, water and sanitation) for poverty reduction and the facilitationo f out migration. Policies for leading regions should prioritize investments in places: transportation, telecommunications and other infrastructural investments that can increase the productivity o f firms. B. A TYPOLOGY POLICIESFORAFFECTINGREGIONAL DEVELOPMENT OF 4.10 A regional development strategy for Russia cannot avoid the goal of balancing various important economic and political priorities. This includes (a) the facilitation of growth points and agglomerations for increasing national economic growth, competitiveness, and living standards, (b) supporting minimal social standards in poorer areas of the country, (c) providing opportunities for lagging regions to realize their economic potential, (d) increasing opportunities for migration from regions with poor economic prospects, and (e) promoting economic development in regions of particular political priority. To balance these objectives, the policy mix should include: "Three Projects are Already Approved, Six More are inLine to be (2007). 49 Spatial policies. Policies to defend the unified market and facilitate the mobility of labor and capital to areas of agglomeration and greater economic promise. Interconnecting policies. Infrastructure policies in areas such as transportation and telecommunications can linkregional markets to the benefit of economic development. (Social) equalization policies. Policies to ensure the provision of minimal social services and support inpoorerregions. Regional (development) policies. Policies aimed at facilitating growth (productivity) in specific regions. Education policies. Beyond the guarantee of minimal standards through equalization policies, additional investment (and reform) ineducation can increase overall economic opportunities for the population. 4.11 In addition to these areas, national policies aimed at economic stabilization, improving the overall business climate in the country, supplying public goods, and boosting the education and health levels o f the population have obvious important general consequences for the development o f Russianregions. SpatialPolicies 4.12 Many spatial policies to facilitate the mobility of labor and capital to areas of greater economic promise correspond to the category of "spatially-blind policies" in the 2009 World Development Report that do not target or favor any particular region. For the Russian case, some important spatial policies could also beregionally targeted. For example, the desire to economize on infrastructure costs could motivate special programs for assisting the out migration of entire populations from specific cold and remote areas. 4.13 Both the inherited Soviet legacy and the spatial nature of recent growth inRussia imply that the attraction of people and capital to areas of agglomeration and greater economic promise i s perhaps the single most critical regional issue for the successful rapid development o f the country. The Soviet legacy concentrated settlements and capital in a number o f remote or cold areas with very high costs o f productiodtransportation and the maintenance o f infrastructure. A number ofthese regions have substantial resource wealth, althoughthe exploitationo fthis wealth does not require the type of extensive settlements present inRussia.' Inthis context, growth and welfare improvements inthe transition periodhave beenclosely related to a natural and important process of migration from East to West, as well as the concentration of non-resource related private investment inthe Western part o f the country. Nevertheless, important barriers remain to the speed and efficiency of this process. The efficient territorial allocation and re-allocation of capital in Russia will depend on the continued rapid development o f capital markets in the country. As the prospects for this development are good, this section will concentrate on the other particularly critical spatial issue for Russia: migration. 4.14 The most recent census of 2002 indicates a strong flow o f migrants from Northern and Eastern regions to Western Russia. Despite a negative natural rate o f population growth, the Central Federal District registered a net population gain of an estimated 0.2 percent between 1989 Onthispoint, see Gaddy and Hill(2003). 50 and 2002 due to migration. On the other hand, the Eastern regions o f Magadan and Chukotka experienced population declines of over 50 percent during the same period.' Thus, important progress in the spatial allocation o f labor has already occurred. Nevertheless, many expert opinions suggest that this resettlement process has been too slow, given the high differentials in living standards between different areas in the Russian Federation. Despite the relaxation of urban registration requirements and the presence o f much greater regional inequality, the pace of internal migration has not increased significantly inthe post-Soviet peri~d.~ At the current pace, a substantial share o f the populationwill remain incold and remote areas for some time to come. 4.15 Along with internal migration, foreign migration into Russia represents another critical factor for future growthinthe country. Russia i s currently only second to the United States inthe attraction o f foreign migrants. The census of 2002 shows 11 million new migrants to Russia since the previous census o f 1989 (5.6 million net migration). Actual migration has been significantly larger than this, as a good share o f migration to Russia i s informal. Experts believe the number of informal migrants during this period to have been in the range of 4-5 million individuals, or almost half the legal migrant number." A large number o f legal migrants have been ethnic Russians movingto the RussianFederation after the breakup o f the Soviet Union. In fact, legal migration to Russia peaked in 1994 at 978,000, and declined to an average of only 87,000 from 2003-2005. While foreign migration to Russia has been quite substantial, and has been a key factor in maintaining a sufficient active labor population to support growth in the country, many argue that it i s nevertheless inadequate for Russia's current needs. A number of experts expect Russia's population to shrinkby close to 20 percent by 2050, which would require a compensating inflow of one million migrants a year, or three times the average official inflow over the last 15 years and five times the average official inflow inrecent years." 4.16 Eveninthe more promising Europeanpart o fthe country, most Russian cities continue to experience population decline, which i s inconsistent with maintaining rapideconomic growth and agglomerations. Outside of Moscow and Saint Petersburg, Russia i s lacking larger cities relative to other countries o f comparable population size." The small net increase in population in the Central District betweenthe 1989 and 2002 censuses i s due almost entirely to the rapid expansion o f Moscow and Moscow oblast. Virtually every other central Russian region experienced population decline during this period, and the loss in population reached over 10 percent in Pskov, Ivanovo, Tula, and Tver.13 The evidence fi-om Rostov and Tver, as presented in Chapter 3, suggests that labor market constraints are becoming an increasing constraint to sustaining growth in the agglomeration areas o f Russia. Given the complex demographic situation in Russia, regions unable to attract significant numbers o f internal and external migrants will most likely be unable to sustain rapid growth. Thus, the facilitation of migration, both internal and foreign, i s crucial for sustaining rapid (regional) growth inRussia.14 8 Heleniak(2003). Gerber (2005) examines 7167 residentialhistories, and comes to the conclusionthat migration flows in Russiahave not accelerated in response to greater post-Soviet freedom o f mobility, greater wage differentials, or unemployment rates. lo Andrienko andGuriev (2005). Ibid. l2 Russia fails to follow Zipfs law that is observed inmost countriesof the world. This law shows a common almost linear relationshipbetweenthe rank order of cities by populationsize andthe log of the population. Russialacks large cities other than Moscowand Saint Petersburg, andtherefore lies below Zipfs curve. See Gaddyand Hill(2003). l3 Heleniak(2003). 14 Perov (2008) also argues that the 2003 Law on Local Self-Governmentmay also be restricting the growth of larger citiesby makingmunicipal boundariesmorerigid. 51 4.17 This conclusion is confirmed by projections for regional growth of the working age population through the year 2026 that were performed for this CEM under a rather optimistic set o f assumptions relative to expert opinions: an overall decline in population by 8.4 million (5.9 percent), a steady rise inthe fertility rate to 1.65, a gradual increase in average life expectancy to 68.2 years, and an a slight acceleration o f net migration from the Eastern to the Western part o f the country consistent with past observed patterns. Even under these assumptions, the workmg age population would decline by over 15 percent inmost o f central Russia (Figure 4.1). Regions that sustain rapid growth over this period will most likely need to compensate for this decline through the attraction o f substantially more migrants than inthe past.15 Figure 4.1: Projected Percentage Change inWorking-Age Population by Region inRussia: 2006- 2026 4 Percentchange inworking-age 0 1 - 2 4 8 -29 9 - -15 0 Note: Working agas ate males 16to 64 population,2006-2026 -14 8 - 0 0 -55 5 - -30 0 and females 16 to 59. 4.18 What are the barriers to more rapid internal migration flows? In every country, large (voluntary) movements o f people take time, and can be associated with substantial social and psychological stress. Thus, even under an optimal policy mix to facilitate migration, the problem o f numerous poor and backward settlements in Russia will persist as a key issue for many years to come. Andrienko and Guriev (2004) present evidence to the effect that migration out o f poorer areas in Russia may actually be an increasing function o f local income levels, suggesting that a number o f Russian citizens are stuck in poverty traps. They lack the minimal income necessary to migrate to a more promising area. This would also be consistent with the evidence discussed above to the effect that the pace o f migration has not picked up in the post-Soviet period, despite a widening o f income and employment differentials. Taken from Heleniak (2006) (Background paper for the CEM). 52 4.19 The housing market is certainly a key constraint on migration in Russia. The privatization process transferred the majority of housing to occupants, which became a windfall for those living in areas of economic promise, but has left many other citizens inremote regions without the means to afford increasingly expensive housing in areas o f economic booms. More generally, booming cities such as Moscow and Saint Petersburg have faced serious challenges in increasing housing and other infrastructure inpace with economic growth. Other constraints on migration can be associated with the labor market. While the Russian labor market exhibits a significant amount of (largely informal) flexibility, some authors suggest that important rigidities nevertheless remain that require attention. In particular, even after the adoption o f a new and more liberal Labor Code, there appear to remain excessive costs and hassles for employers associated with releasing workers, which find reflection in a common cautious attitude toward hiring.16 Finally, there are the issues of regional protectionism that can limit competition and the free movement of capital to the detriment of sustaining rapid growth inthe most promising areas. The cases of Rostov and Tver, as examined inChapter 3, are instructive of the politicaland social pressures surrounding the opening up o f Russian regions that had been formerly very internally oriented. 4.20 Important economic policies for the resolution of the above issues include priority infrastructure projects for growth points to accommodate inflows of migrants (including affordable housing)," labor market policies, and effective implementation o f the Competition Law to defend and preserve the unified space of the Russian Federation. Special government programs for potential migrants may also play a supportingrole. 4.21 Like a number o f other countries that attract a substantial number o f foreign migrants, Russia must weight difficult social, political, and economic issues while adapting policies inthis area. ConsideringRussia's demographic situation, the country is quite fortunate that its relatively highstandard of living inthe greater regionmakes it a magnet for migrants within the CIS. An aging and shrinkingpopulation makes Russia comparable to the countries o f Western Europe in this regard: there is a strong need for migrants to augment the labor force, but social tensions place limits on policies to liberalize the (formal) migration regime. In this regard, it should be noted that the balance of net advantages to Russia from external migrants may be greater than those in Western European countries for several reasons: (a) the demographic crisis (aging and population decline) i s greater in Russia than in Western Europe, and the need to sustain rapid growth i s also more pressing, (b) Russia has a large pool of potential migrants with a fluent command of Russian and other similar (post-Soviet) social attributes that allow for a more harmonious assimilation to work and life in Russia, and (c) Russia does not have an expensive and generous social safety net for unemployed migrants (a primary source of tension surrounding migration inthe West). 4.22 A World Bank Policy Note o f 2005 outlines important directions for improvingmigration policy, in particular the liberalization of the official regime along with a crackdown on illegal migration." This policy mix would have the potential of attracting a migrant pool with a better skill profiles for Russia and alleviating social tensions that currently exist around illegalmigrants. Russia i s indeedfollowing a policy consistent with these two directions, although there i s a risk i s that the official regime could remain sufficiently restrictive to negate the net benefits from a l6This isdescribedindetailinGimpelsonandKapeliushnikov ed. (2005), Chapter 1. Russiahas indeedmade affordable housing a priority policy area, althoughthe focus so far has been on facilitating the purchaseof housing, rather than the provision of affordablerentals. For potentialmigrantsfrom poorer regions,the latter is also quite important(See RentalChoiceandHousingPolicy.. ,2005). 18Andrienko andGuriev (2005). 53 crackdown on illegal migrants, i.e. it i s not in Russia's economic interest to adopt a regime that would cause a decline in net inflows of migrants, especially over the medium term when the demographic crisis inthe country will reach its apex. 4.23 Attitudes toward external migrants in regional and local administrations are typically quite cautious or negative. While these attitudes are understandable, there nevertheless appearsto be a common insufficient understanding at the regional level of the importance of attracting strong inflows of labor, as well as capital, to points of potential or emerging growth. Many regions have come to understandthat their economic future depends very much on the attraction of outside capital. Yet this future dependsjust as much on attractinglabor.'' Interconnecting Policies 4.24 Infrastructure has long been recognized as a key factor in regional and economic development, and represents a primary channel through which the government can promote competitiveness and productivity growth. Improvements in infrastructure can not only improve the development prospects of a given regions, but may have benefits for multiple regions, including critical linkages inmarkets and the reduction o f transportation costs. 4.25 On the other hand, the results of Chapter 2 suggest that infrastructure density, in and of itself, may not contribute much to growth at the regional level. The 2009 World Development Report also emphasizes that a number of expensive and ambitious infrastructure projects have had little impact inpractice. It i s therefore critical that decisions on infrastructural investments are made on the basis o f a careful assessment of the overall development impact of each specific project. This assessment should focus on the degree to which the absence of key infrastructure i s currently a genuine constraint on growth or the integration o f markets. The benefits of infrastructure for connecting markets should also be weighed against alternative directions of infrastructure investments. Often, infrastructure investments in areas of growth agglomeration can offer a higherreturnthan those inmore remote regions.20 4.26 Better transportation linkages betweenregions should generally have a positive impact on economic growth and development. The specific impact at the level of a single region can sometimes be ambiguous, however. There exist examples from the EU where better transportation linkages of a lagging region with a leading region actually had a negative impact on economic development in the former. Given the better transportation linkages, many businesses actually moved their operations out of lagging region in favor of using the improved transportation network for accessing the market from the leading region, where advantages of agglomeration can be exploited. (Social) Equalization Policies 4.27 Measures to ensure minimal social (and living) standards in poorer regions properly belong to the category o f social policy, as opposed to regional development policies. Yet the coordination o f these policies with a national strategy for regional development would be important. Equalization policies aimed at supporting the social sphere are often confused in Russian debates with policies aimed at equalizing development or growth in different regions. Such latter measures belong to category of regional (development) policies discussed below. Confusion between these two types of equalization policies may have been a primary source of Some of the prevailingattitudes towardmigration are summarized and criticized inKarachurina (2006) 20 See Brown et a1(2008). 54 contention in the discussion of the various drafts o f the National Regional Strategy in Russia during 2005-2007. While a regional strategy focused on promoting growth points and agglomerations does indeed entail the rejection o f alternative policies aimed at equalizing regional development, the need for social equalization policies could become even greater under the former strategy. Given limitations inthe speed of migration, promoting unbalanced growth can actually increase the challenge o f addressing poverty in lagging regions. Given the fact that basic social expenditures are largely the responsibility of regional and local governments in Russia, equalization transfers have major significance for maintaining minimal standards in laggingregions. 4.28 The actual levels and stability o f equalization transfers do have important direct implications for regional development, however. If the goal i s to promote migration to growth points, benefits should not be set at levels highenough to discourage out migration. Ifthe goal i s to encourage initiatives aimed at igniting regional growth and development, then benefits should not be "soft" in the sense of adjusting quickly to reallocate finds from more successful to less successful regions. The equalization formula was developed inRussia with the aim o f satisfying both of these criteria, although the additional allocation o f transfers and federal loans outside of the formula-based equalization find shouldalso be scrutinizedfrom this point ofview. Regional Development Policies 4.29 State policies aimed at ignitingor promoting economic development in specific, usually poorer, regions represent one of the most controversial areas of economic policy. Any country considering embarking on this path would do well to study the lessons o f a very mixed, and predominantly disappointing, history. The reasons for past failures are numerous, including limited information inthe government about true regional economic potential, distortions of the allocation o f factors away from true comparative advantage, the adverse effect o f subsidies on economic incentives, and the inevitable politicization o f decisions about whichregions to support. Long histories of disappointment include decades o f efforts inItaly and Brazil to ignite growth in the respective Southern and Northern regions of the countries, which did not succeed inreducing significantly the North-South gap in either case.21 The greatest success story in regional policy may be China. However, the circumstances inthis case were unusual. China introduced market reforms and lowered taxes only in the Southern regions of the country, which produced an economic boom in those regions at a time when Northern areas remained largely under the planned economy. Market reforminChina began inthe South, and gradually movedNorth. 4.30 The World Bank produced a report in 2007 on the experience of regional development policies and their potential role for poverty reduction in Eastern Europe and the former Soviet Union.** This report concluded that the efficacy of regional development policy is rather questionable, and that other policies may prove more effective in addressing the problems of regional pockets of poverty (facilitating migration, social spending, and investinginpeople rather than places (education policies)). This conclusion i s also echoed inthe preliminary results o f the On the experience of Italy, see Barca (2001). For the case of Brazil, see Ferreira (2005). Ferreira concludes that even some o f the isolated so-called successes in Brazil can be called into question because of their economic costs. For, example a free economic zone succeeded in increasing investment and growth in Manaus, but at the expensive of economic development in surrounding regions. For an interesting and balanced comparison of the experiences in S ain Italy, and Brazil, see Perez and Rowland(2005). 2PDil;inger (2007). 55 forthcoming 2009 World Development Report: "countries should prioritize investments in people."23 4.3 1 Nevertheless, many larger countries with regional development problems have pressed on with regional policy, although the lessons of the past have supported a consensus around a set of new principles, summarized below, that are strikingly different than in the past. Given the complexities, risks, and past disappointments, Russia might first ask itself the question as to whether or not it really needs regional development policies. Some would argue that measuresto assist the working of market forces inthe efficient allocation and reallocation of factors, together with the provision ofpublic goods, social equalizationand education policies, would be sufficient to ensure a healthy process of regional development inthe country. On the other hand, arguments infavor of aregionalpolicyfor Russiacouldcitethe following considerations: 0 Some areas of the country (the Far East, the South, border areas) are prioritized for economic development for political as well as economic reasons. Given the constraints o f whatever political decisions are made in this regard, Russia should pursue the most economically efficient means for realizing the implieddevelopment priorities. 0 Russia i s in need of strong urban agglomerations outside of Moscow and Saint Petersburg, while alternative cities face demographic challenges for attracting and accommodating sufficient migrants. Under certain circumstances, state interventions (social infrastructure, for example) could help to support such agglomerations. This concerns the type o f regional policy outlined in the former draft National Strategy that i s focused on growth points rather than 1agging.regions. 0 At this point intransition, a significant amount of uncertainty exists about the economic potential of many Russian regions. This i s arguably greater than the uncertainty that exists in countries where the initial allocation of capital and settlements followed market forces. Increasing returns associatedwith agglomeration economies can prevent regions that currently lack such agglomerations to compete effectively with other regions and realize their potential. Thus, it may be expedient to seek and promote mechanisms for revealing and realizing this hiddenpotential where it exists. 0 The substantial social and economic costs o f massive internal migration increase the opportunity costs in Russia of not revealing hidden potential in lagging regions. Although many settlements in Russia may have not been established under market forces alone, that in itself i s not proof that they should be dismantled. Given that people and capital already exist inthese places, finding successful development paths in at least some o f them couldbe o f considerable value to Russia. 0 Some recent studies emphasize the possible expedience o f cooperation between the federal government, subnational administrations, and the local business community for the revelation of local information and coordination o f decisions, particularly those pertaining to the particular mix of localpublic goods neededfor productivity growth.24 4.32 The extent to which Russia adopts regional development policies to address these issues presents a difficult political choice with significant pros and cons. To the degree that Russia does 23 World Development Report2009 (2008). 24 A strong argument for this approach is made in Barca (2007b). It also reflects parallel thinking about the "new industrialpolicy" (Rodrik, 2004). 56 decide to pursue policies o f this nature, it would be well advised to base its approach on the principles of the so-called new regional policy, which incorporate important lessons from past experience. For the case of Russia, the recent experience o f the European Union inthe allocation o f Structure Funds i s particularly instructive, and consistent with trends in regional policy elsewhere inthe Although the explicit focus o fregional policy inthe European Union i s to promote development in lagging and poorer regions, the overall approach developed could be potentially generalized to the case of regional policies targeted to leading regions and growth agglomerations. It should be emphasized that the results o f the new regional policy inpractice in the EU and elsewhere are controversial and, for many observers, unconvincing,26 although it couldbe argued that the time horizon i s still too short for proper empirical testing. 4.33 The new regionalpolicy begins from the premise that past traditional regionalpolicy has been largely ineffective. Inthe words of Fabrizio Barca (2007a), the traditional policy could be characterized as compensating for regional gaps in productivity by providing sectoral public goods and/or subsidies to f m s or labor." By contrast, the new regional policy aims at "increasing productivity by providing integrated baskets of local public goods through (mostly) local and regional projects devised and selected according to standards agreed at the national and/or supernational level." This complex definition has the following key component parts: 0 The overall goal of new regional policy i s to increase productivity, not compensate for lower productivity. 0 The new regional policy generally provides local public goods (infrastructure, regulation, coordination) as opposed to financial subsidies to economic organizations. 0 The new regional policy supports initiatives coming from regional or local levels, consistent with national priorities, through a competitive process. 4.34 These principlesare similar to those associated with the "new industrial policy.'' Policy in this context can be conceived as ajoint attempt by federal, regional, and local authorities to work together with investors and the local business community to reveal information about projects of potential promise for regional development, and then coordinate decisions around their realization. As the federal government has only very limited information about regional potential, the actual initiatives shouldbe quite decentralized, coming from regionalflocal officials, investors, and entrepreneurs. The federal government selects proposed projects according to a competitive process inline with nationalpriorities. ., 4.35 The primary role of government here is incoordinatingthis overall process andproviding key local public goods for the success of the project. "Local public goods" can be understood very broadly as ranging from key physical infrastructure, to the implementation and enforcement of key regulations, and to helping attract key investors to the region. A critical element to the success o f this type of policy i s the realization that quite a number (probably the majority) o f such projects will not succeed, and a strong mechanism i s needed to ensure unsuccessful projects will not be funded for too long a time. The general goal i s the continual reallocation of support from less successful to more successful projects. As an example, this i s what the EU has done in the case of Southern Italy (the Mezzogiorno) through a particular system o f sanctions and 25 For a general critical survey o f the EU approach, see Funk and Pizzati ed. (2003). See also Barca (2001, 2007). Granberg (2000) offers a general examination o f the relevance o f the experience o f the EUinregional development for Russia. 26See, for example, see the discussion inthe special issue of the Journal of Regional Studies no. 2, April, 2006. 57 premiums.*' By implication, the new industry policy requires an effective mechanism for monitoring results. EducationPolicies 4.36 While social (equalization) policies can have a direct impact on the quality of education, a number of countries have adopted additional programs aimed at boosting educational opportunities in lagging regions. Although these polices can be regionally oriented, this represents investment in people rather than places, per se. A better educated and qualified population will generally have better opportunities not only for employment within a given region, but for migration to other more promising regions. Ireland and India are among countries that many observers believe hadparticularly successful national education programs. 4.37 Most important for the success of such education policies i s that the training provided i s linkeddirectly to the labor market and the needs ofthe economy. The common absence of such a strong link has been a particular problem in Russia during the transition period, and i s the major target o f education reform inthe country. C. A NATIONAL REGIONAL STRATEGY FOR RUSSIA? 4.38 The above considerations suggest some central elements of a constructive national regional strategy for Russia. Of course, the particular chosen mix o f these elements will require the consideration of complicatedand important tradeoffs. 0 Sustaining rapid growth and welfare improvements in Russia depends critically on the continued flow o f capital and migrants to areas o f high economic promise and agglomeration. These flows appear to be insufficient and imperfectly regulated at the present time. Thus, additional government policies to facilitate these flows could be important in areas such as migration policy, housing and social infrastructure, labor market policies, and competition policy. 0 Maintaining strong and stable social equalizationtransfers will be a means of sustaining minimal living and service standards in lagging regions. These transfers should not be so high or unstable as to discourage either out migration from these regions, or the active pursuit by local authorities and entrepreneurs of a successful development strategy. This support may decline over time in harmony with other policies to realize out migration or regional development initiatives. Given the extent of informational and social problems associated with settlements in lagging regions, it i s critical that local authorities and entrepreneurs are givenincentives and opportunities to realize their own initiatives aimed at improving their situation. Furthermore, Russia could make a political choice to support these initiatives consistent with national objectives through federal programs based on the principles of new regional policy. Given the current nature of agglomeration effects in Russia today, regions may not succeedinrealizing their potential through market forces alone. Key infiastructural investments to improve market linkages, facilitate factor mobility, and improve the prospects for regions with genuine economic promise. 27 Barca (2007a). 58 4.39 These considerations are broadly consistent with many of the directions that Russia has been moving in recent years. This includes the contemplation and attempted drafting o f a comprehensive National Regional Strategy, the solicitation of regional strategies inharmony with national priorities, and the allocation o f a number of national instruments to regions on a competitive basis, including the location of special economic zones and IT parks and infrastructural investments from the Investment Fund. New developments in territorial planning place a strong emphasis on developing a broad dialog between different levels o f government and local business. Special national projects have been introduced for the priority regions of the South and the Far East. Other measures have targeted migration, including a simplification of registration procedures for some cases and the provision o f amnesty to a number o f illegal migrants. Equalization transfers are concentrated inpoorer regions by a formula that allows only limitedadjustments in transfers to current economic conditions. Education reformhas also been prioritized. Recent policy declarations on regional policy prioritize infrastructure, human capital, and other productivity-enhancing investments over subsidies to low productivity firms. However, Russia has not yet adopted a strategy for regional development that would allow it to employ these and other policy instrumentsina coordinated and consistent way. 4.40 For the important goal of giving incentives and opportunities to lagging regions to realize their potential, Russia currently faces a particular challenge. As stressed above, the inherently decentralized nature of information on regional potential implies that success in this area will depend greatly on genuine decentralized initiatives for increasing productivity. While agglomeration points are key to regional growth, an attempt by the central government itself to choose future points of growth and design its own detailed regional programs would likely be highly unsuccessful. Inthis regard, Gainfranco Viesti, an acute observer of regional policy in Italy and the EU, lists three o f the primary reasons for the failure of past regional policy to be "top-down technocratic planning, no local responsibility, and political intermediation of resources.''2* 4.41 Since 2000, Moscow has assimilated increasingly greater power relative to the regions. Perhaps the most important decision in this regard was the elimination o f direct gubernatorial elections in2004 in favor of the presidential appointment of regional governors. Inthis context, the balance of political priorities of the regional leadership has shiftedinthe direction o fpleasing superiors in Moscow relative to constituents within the region. On the one hand, the new willingness of the regions to cooperate strongly with Moscow offers better opportunities for coordination and cooperation in the realization of regional initiatives. On the other hand, too strong a hand of the center can also breed strong risk aversion among local and regional officials for pushing ahead with their own initiatives. The exercise in drafting regional growth strategies inRussiawas informative from this pointofview. Someregionstook thisprocess very seriously and were deeply engaged. Other regions did little more than go through the motions as a means o f complying with the demands of Moscow. A hndamental principle for the success of a regional development strategy in Russia i s that the system rewards innovation and genuine decentralized initiatives as opposed to passive obedience. 28 GianfrancoViesti lists these as three of six primary reasons that regionalpolicy was unsuccessful from 1951-1992. The other three are "the independence of regional policy from overall sectoral and growth policy, the centrality of financial grants to firms, and increasingemphasis on transfers to households.Viesti (2006). See also Haussmannand Rodrik (2003) on the importanceof decentralizedinitiatives (self-discovery) for the rationaleof developmentpolicies. An interestingdiscussionofthe problemofpotential overcentralizationin aRussianapproachto regionalpolicy canbe found inZubarevich(2005). 59 Appendix Summary table for the statistical regression, See Polyakov (2007) for more details The left column indicates the exogenous variables considered in the regression, the middle column indicates the share o f variation in regional growth explained by the regression, and the right column shows the coefficient value for the estimated spatial spillover, i.e. if this number i s .35, then one additional percentage o f growth in other regions weighted by the market potential matrix will generate -35percent growth on average ina given region. Coefficient Significance Spat Squared Corr. Spatial Spillover Coefficient (its p-value if over .lo) 60 References Ahrend, Rudiger. 2002. "Speed of Reform, Initial Conditions, Political Orientation, or What? Explaining Russian Regions' Economic Performance," DELTS Working Papers2002-10 DELTA (Ecole Normale SupCrieure). Andrienko, Yuri, and Guriev, Sergei (2004), "Determinants o f Interregional Mobility in Russia: Evidence from Panel Data," Economics of Transition 12:1, pp. 1-27 Andrienko, Yuri and Guriev, Sergei (2006), "Understanding Migration in Russia: A Policy Note," hm://www.cefir.ru/index.php?l=eng&id=130 World Bank Policy Note, CEFIR Policy Note Series Barca, Fabrizio (2007a), "The Rationale and Complex Governance o f Place-Based Policies," mimeo Barca, Fabrizio (2007b), "Regional Development Policy, Rationale, Multi-Level Governance and Evaluation," Powerpoint presentation for Moscow seminar, January 26 Barca, Fabrizio (200l), "Rethinking a Partnership in Development Policies: Lessons from a European Policy Experiment," http://www.sociologiadip.unimib.it/mastersqs/dida1/testicinclueharca develop.pdf Brown, David, Fay, Marrianne, Felkner, John, Lall, Somik, and Wang, HyoungGun (2008), "The Death of Distance? Economic Implications o f InfrastructureImprovementinRussia," mimeo "Concept for a Strategy for the Social-Economic Development of the Regions of the Russian Federation" (2006), Draft document o f the Ministry of Regional Development (In Russian) Desai, Raj. and Goldberg, Itshak, ed (2007), Enhancing Russia 's Competitiveness and Innovative Capacity (Investment Climate Assessment of the Russian Federation) World Bank Dillinger, William (2007), "Poverty and Regional Development in Eastern Europe and Central Asia," World Bank Working Paper 118 Drobyshevsky, S., Lugovoy, Oleg, Astafyeva, E., Polevoy, D, Kozlovskaya, A,, Trunin, R., and Lederman, L. (2005) Factors of Economic Growth in Russian Regions (in Russian), Consortium for Applied Economic Analysis, Moscow Ferreira, Pedro Cavalcanti (2004), "Regional Policy in Brazil: a Review" http://www.fgv.br/urofessor/ferreira/ReaionalPolicyFerreira.pdf Fingleton Bernard ed. (2003), European Regional Growth, Springer, Berlin. Fingleton, Bernard (2002). "Theoretical Economic Geography and Spatial Econometrics: Bridgingthe Gap between Theory and Evidence" mimeo Frye, Timothy and Yakovlev, Andrei (2007), "Reforms inRussia through the Eyes o f Business: What has Changed in Seven Years," Higher School of Economics (in Russian) http://d1.hse.ru/data/587/45611234/Yakovlev-Frye16aug2007.pdf 61 Fujita, Masahisa and Krugman, Paul (2004), "The New Economic Geography: Past, Present, andFuture," PapersinRegional Science, 83: 139-164 Funk, Bernard and Pizzatii, Lodovico (2003), European Integration, Regional Policy, and Growth, World Bank Gaddy, Clifford and Hill,Fionna (2003), The Siberian Curse: How Communist Planners Left Russia in Out in the Cold, BrookingInstitution Press Gerber, Theodore (2005), "Individual and Contextual Determinants of Internal Migration in Russia, 1985-2001," Global Development Network http://www.eldis.ore;/e;o/topics/resource- guides/mi~ration/internal-migration&id=21123&~e=Document Getis Arthur, Mur Jesus, and Zoller, Henry (eds.). 2004. Spatial Econometrics and Spatial Statistics, Palgrave: London. Gimpelson, Madimir, and Kapeliushnikov, Rodislav (2005), Nonstandard Employment, Higher School of Economics Press(InRussian) Granberg, Aleksandr, ed., (2000) Regional Development: The Experience of Russian and the European Union, (inRussian) Tacis Project, Ekonomika. Granberg, Aleksandr. andZaitseva, Yulia. (2003), "Production andthe Use of Gross Regional Product: Interregional Comparisons," Russian EconomicsJournal No. 1,pp. 39-54 (inRussian) Haussmann, Ricardo and Rodrik, Dani (2003), "Economic Development as Self-Discovery," Journal of Economic Development, December Heleniak, Timothy (2003), "The 2002 Census in Russia: Preliminary Results," Eurasian Geography and Economics, 4416 pp.430-442 Heleniak, Timothy (2006), "Russia's Population in the Future: National and Regional Scenarios," Backgroundpaper for the CEM Independent Institute for Social Policy (2008), A Social Atlas of Russian Regions: Thematic Overviews (2008) (InRussian), http://atlas.socpol.ru/overviews/econ_condition/index.shtml Isard, Walter (1960), Methods of Regional Analysis: An Introduction to Regional Science. MIT Press. Karachurina, Kilia (2006), "The Economics of Migration in Russian Cities: Myths and Reality," (inRussian) inChanging Cities,Rossiisckotekspertnoeobozrenie, No. 2(16) Center for Strategic Researchof the NorthWest. Kltosvog, F.N. and Chernova, L.S (2004). "An Assessment o f the Level o f Development o f Russian Regions Relative to a Group o f Western Countries (in Russian), Russian Academy of Sciences, Institute of Economic Forecasting, Laboratory for Forecasting Regional Economic Proportions. http://www.ecfor.dpdf.php?id=pub/klot02 62 L e Gallo, Julie and Dall'erba, Sandy (2003), "Spatial Econometric Analysis o f the Evolution of the European Regional Convergence Process: 1980-1999," UrbadRegional Papers Series, http://129.3.20.4 l/eps/urb/~a~ers/03 11/0311001.pdf Nefedova, Tatyana (2004), "To See Russia (Uvidet Rossiu)" http://www.strana- oz.ru/authors/?author=3 82 OECD(2006), Competitive Cities in the GlobalEconomy, OECD OECDEconomic Surveys:Russian Federation (2000) "On a Mechanism for the Interaction of Federal and Regional Organs of State Power For the Formulation of Programs for the Comprehensive Social-Economic Development of Regions," (2006) Document of the Working Group on Questions of Comprehensive Social- Economic Planning o fthe Development o f Regions, July, Moscow (inRussian) Perez, Gerson Xavier and Rowland, Peter (2005), "Regional Economic Policies: Four Country Cases," Polyakov, Evgenny, "Structural and Spatial Aspects o f Russian Economic Growth," Background Paper for CEM Perov, Maksim (2008), "Central Problems in the Development o f Russian Cities, and Tasks for http://www.csr-nw.ru/content/news/default.asp?shmode=2&ids=28zida=20 16 City Planning." (inRussian), Gradoustroistvo, No. 0 Rental Choice and Housing Policy Realignment in Transition: Post-Privatization Challenges in theEurope and CentralAsia Region (2005), World Bank Roberts, M a r k and Setterfield, Mark (2007), "Endogenous Regional Growth: A Critical Survey," Cambridge Centre for Economic and Public Policy Working Paper Rodrik, Dani (2004), "Industrial Policy for the Twenty First Century," http://ksnhome.harvard.edu/-drodnkWNIDOSep.pdf Russian Firms at the Crossroads: What is Preventing Our Firms From Becoming Competitive (2007), Higher School of Economics, (InRussian) Russian Federation: Promoting Equitable Growth - A Living Standards Assessment (2008), World Bank, forthcoming "RussianRegional Growth and Agglomeration Effects (2007)," part 2 of Russian Economic Report #14, World Bank Sachs, Jeffery, Gallup, John Luke, and Mellinger, Andrew (1998), "Geography and Economic Growth," h~://www.~dsnet,or~/classes/SachsGallupGeonaphy~owth.udf Sachs, Jeffrey and Warner, Andrew (1995), "Natural Resource Abundance and Economic Growth," NBERWorking Paper 53981 63 "Three Projects are Already Approved, Six More are in Line to Be (2007)," Interview of Regional Minister Dmitrii Kozak by Vesti, November 16 (in Russian) htt~://www.minregion.ru/WorkItems/DocItem.aspx?DocID=521&PageID=376 Verrill, Steve, and Richard A. Johnson. (2005). "Confidence Bounds andHypothesisTests for Normal Distribution Coefficients o f Variation," Research Paper, USDA Forest Products Laboratory, Madison, Wisconsin http://www 1.fpl.fs.fed.us/cov.asym.pdf Unleashing the Potential: Growth and Investment in Russian Regions (2006) Stockholm Institute for Transition Economics (SITE) and Centre for Economic and Financial Research (CEFW, htt~://www.hhs.se/NR/rdonlyes/ADE66EOE-D6AD-42C8-BOBA- 79C0564B63CE/O/Unleashing the uotential.pdf Viesti, Gianfranco, "Place-Based Policies for Regional Development (2006)," Powerpoint presentation for World Bank Regional Roundtable, Dec. 13 "World DevelopmentReport: 2009" Powerpoint presentation of preliminary results, presented inMoscow, 2008 Yudaeva, Ksenia, Maria Gorban, Natalia Volchkova, and Vladimir Popov. (2004). "Up and Down the stairs: Paradoxes of Russian Economic Growth", In The Economic Prospects o f the CIS: Sources of LongTermGrowth, Gur Ofer andRichard Pomfied(eds.), Oxford Press. Zubarevich, Natalia (2007a), "The Index of Human Potential o f a Region Depends Greatly on its Economy," Demoskop Weekly, no. 273 Jan. 22-Feb. 4 (inRussian) Zubarevich, Natalia (2005), "Regional Policy in the Context o f a Strengthening Government: How can the Reincamation o f Gosplan be Prevented?," (in Russian), Rossiisckoe Ekspertnoe Obozrenie, No. 1:13 http://www.rusrev.org/content/review/default.asp?s~ode=2&ids=32&ida=806 Zubarevich, Natalia (2007b), "The Social-Economic Development of the Regions," Powerpoint presentation made at the Carnegie Foundation, Moscow, (inRussian) 64