S. Kadochnikov, P. Vorobev, Y. Legkaya, N. Davidson

S. Kadochnikov, P. Vorobev, Y. Legkaya, N. Davidson

UralsStateUniversity, Ekaterinburg, Russia

FDI Concentration in Russian Regions: the Impact on Enterprise Productivity

Introduction. We examine the impact of regional FDI concentration onproductivity of Russian enterprises. Motivation is search for sources of economic development in Russian regions,whichdiffer significantlyin initial conditions and development; findingan optimum combination of industries in a region; finding business concentration effect on productivity; finding overall productivity gain from FDI concentration at an industrial and regional level; separating FDI effects from concentration effects.Our objective is to reveal determinants and effects of regional FDI concentration. We investigate dynamic pattern of FDI inflow in sectors of Russian economy and estimate regionalFDI concentration effect. Hypotheses are: enterprise activities are affected by industrial and regional factors; agglomeration economies exist in Russian cities; FDI concentration may have negative localization effects, but positive urbanization ones.

FDI effects for Russian economy are an important research area.FDI inflow into Russiabecame$27.797 bln in 2007compared to $4.429 bln in 2000. Production share for FDI sector was31.84% in 2007. In 2000-2004 labor productivity in FDI sector was 4 times higher than in Russian economy as a whole; in 2005-2006 productivity gap reached 6.5 times (Rosstat).

S. Ledyaeva (2007) points out ten Russian regions - leaders in attracting FDI
in 1995-2006.FDI determinants are oil and gas, agglomeration advantages, industrial development, market size and large cities advantages, transit region and protection of investors by the local legislation. Herfindal-Hirschman index for 2000-2006 revealed growing regional FDI concentration: 2000 – 19%, 2006 – 27% (Rosstat data).

Previous research. Agglomeration is defined as concentration of economic activityand implies an industrial, geographic and temporal scope(Rosenthal and Strange, 2004). Quigley (1998), Ottaviano and Thisse (2004), Bekele and Jackson (2006) provide an overview of research. Analysis of enterprise location dates back to Weber (1909).New economic geography explains agglomeration combining trade costs with scale economies (Krugman, 1991). Geographic agglomeration implies eitherlocalization,measured by industry scale (Marshall)orurbanizationmeasured by population (Marshall,1890; Jacobs,1969; Glaeser et al, 1992; Henderson et al, 1995).Doubling the city brings 3-27% advantage in labor productivity (Shefer, 1973; Sveikauskas, 1975; Rosenthal and Strange, 2004).

Duranton, Puga (2004) and Ottaviano and Thisse (2004)provide an overview of agglomeration microfoundations.There is evidence of agglomeration economiesintroduced by A. Marshall(1890): input sharing, labour market pooling(matching mechanisms, lower risks) and knowledge spillovers. Other agglomeration economiesare: home market effects, economies in consumption, competition.Agglomeration in Russia is investigated byK. Gonchar (2008),N. Zubarevich (2008), Lappo and Polyan (2007). Research reveals agglomeration economies.

Concerning FDI effects, direct effectsimply that FDI are a source of financing, training of labor force, technological transfer within the firm. Indirect FDI effects are localization effects(Blomström and Kokko, 2003; Markusen, 1995;Markusen and Venables, 1999; Markusen and Trofimenko, 2007), competition effects (Markusen and Venables, 1999) and urbanization effects.PositiveFDI spilloversfor Russia are found by Yudaeva et al. (2000);Kadochnikov/ Kulakova (2002);Bessonova et al. (2003);negativeFDI spillovers are revealed by Yudaeva et al.(2000); Kadochnikov/Kulakova(2002);negative FDI linkages: Yudaeva et al.(2003),Kadochnikov/Drapkin (2007).

Data. We study firm-level data on medium-sized enterprises for the period 2003-2005 for 10530 firms with revenue$10-80 mln. Indicators are revenue, profit, capital, structure of ownership. There are 28 industries, 68 sub-industries in the database; regional structure is representative. Our source is database made by “Expert-Data” based on Rosstat data. Regional and city-level datasource is Rosstat.

Empirical model. We consider external factors affecting productivity on industrial and regional levels, andcompetition. Industrial level factors are activities of national and foreign firms (firms with FDI) in an industry in a region. Presence of FDI is associated withinfrastructure development, large number of workforce in an industry in a region and large demand for final goods. FDI enhance productivityvia stimulating development within an industry.Factors considered on the regional level: activity of national and foreign firms in various industries, presence of regional infrastructure, presence of large number of workforce, presence of large demand from the final goods consumers. We construct the empirical model:

Yi/Ki = αiXi + βjZj + γkVk + δjkWjk + εi

Yi – revenue of the company; Ki – fixed assets of the company; Xi – resources of the company(fixed assets, old fixed assets, age of an enterprise); Zj – resources in an industry (revenue of national enterprises with higher productivity, revenue of foreign enterprises, revenue of foreign enterprises with higher productivityin an industry in a region); Vk – resources in a region(revenue of national enterprises with higher productivity, revenue of foreign enterprises, the volume of deposits by physical persons, transport development indicators, average job search period, the number of employees relative to the number of individuals in a sample, human capital quality); Wjk – competition in an industry in a region(competitiveness index, small business support index and price regulation index for a region). Below are the results of estimation.

Regression results (Random-effects GLS)

Dependent variable / Logarithm of revenue divided into fixed assets
Independent variables / Coefficient / z - statistics
Fixed assets, mln. Rubles / -0.00174 / -47.18
Year of firm’s establishing / 0.052799 / 7.75
HHI for an industry / -9.09411 / -3.57
Logarithm of revenue share of firms with FDI in a city / 0.116231 / 2.21
Logarithm of revenue share for firms with FDI in a city in an industry / -0.08936 / -5.18
Logarithm of population in a region / 0.03544 / 2.58
Logarithm of bank deposits in a region / 0.127612 / 4.12
Logarithm of GRP per capita / 0.086279 / 7.44
Logarithm of railway density in a region / 0.099837 / 2.1
Logarithm of employees involved in research in a region / -0.05378 / -1.63
Logarithm of time period for finding a job (months) in a region / 0.578454 / 4.07
Observations / 18527
R2 / 0.277

Conclusion. We conclude that agglomeration economies do exist in Russia. There are positive urbanization effects due to input sharing (infrastructure), home market effects and economies in consumption. Effects from labor market pooling are not revealed. FDI concentration leads to positive urbanization effects and to negative localization and competition effects. Thus region-wide economic policy aimed at attracting FDI is relevant; infrastructureis one of the key factorsfor attracting FDI and for promoting positive FDI effects.

Key words: FDI, agglomeration, regional development, enterprise productivity