ECONOMIC GROWTH DUE TO EXPORT EXTERNALITIES: A SPATIAL ECONOMETRIC ANALYSIS FOR RUSSIAN REGIONS, Sergey Kadochnikov Dr. Sc., Professor Director of Graduate School of Economics and Management UrFU Anna Fedyunina M. Sc., Research fellow Graduate School of Economics and Management UrFU

Motivation What does foster economic growth in a region? Does export diversification matter? 3 groups of explanations 1.Traditional Explanations Avoidance of dutch disease (Corden, Neary, 1982) Avoidance of the voracity effect (Lane, Tornell, 1996; Tornell, Lane, 1999) Learning-by-doing and learning-by-exporting spillovers ( Greenaway, Kneller, 2004) 2.New Explanations Countries that export expensive goods associated with higher productivity levels (according to the world export basket) grow more rapidly (Hausmann, Hwang, Rodrik, 2006) 3. New New Explanations Rate of economic growth depends on the density of the product space near the area where each country has developed its comparative advantage (Hausmann, Klinger, 2006; 2010), therefore there are two types of externalities affecting economic growth: Externalities from related industries Spatial externalities from related industries

RUSSIAN FEDERATION Need to be considered ITALY (Boschma, Iammarino, 2009) Value added + Employment : + Labor productivity: + THE NETHERLANDS (Frenken, Oort, Verburg, 2007) Value added : ? Employment : + Labor productivity: 0 THE UNITED KINGDOM (Bishop, Gripaios, 2010) Value added: ? Employment: 0 Labor productivity: ? Note: «+» - positive significant impact; «-» - negative impact; «0» - insignificant impact; «?» – the impact hasnt been considered SPAIN (Boschma, Minondo, Navarro, 2010) Value added : + Employment : 0 Labor productivity: 0 GERMANY (Brachart, Kubis, Titze, 2011) Value added : + Employment : + Labor productivity: + Impact of related variety on economic growth: recent empirical results

REGIONAL GROWTH BASED ON RELATED VARIETY Value added growth (Boschma, Iammarino, 2009; Boschma et. al., 2010) Employment growth (Frenken, Oort, Verburg, 2007; Boschma, Iammarino, 2009) Labor productivity growth (Boschma, Iammarino, 2009) ECONOMIC GROWTH Entry of new industries (Hidalgo et. Al., 2007; Hausmann, Klinger, 2007) EXPORT DIVERSIFICATION RELATED INDUSTRIES Technological convergence Absorption of spillovers Knowledge sharing Resource reallocation

Methodology PRODY: implied productivity level of each industry weighted average of exporting regions GRP per capita, where the weights are each exporting regions revealed comparative advantage EXPY: degree of export structural sophistication implied productivity level PRODY for industry i multiplied by its share in export structure of region c Ω : potential export capability average weighted PRODY of all industries to be potentially produced and exported in a region for any region с holds distance between industries

Database Electronic copies of cargo customs declarations of the Federal Customs Service of Russia Export value: f.o.b. export prices Customs treatment: export (exclusive of re-export) Industry classification: 4-digit codes of Customs Nomenclature of the Russian Federation foreign economic activity Period: Territories: 77 Russian regions Russian Federal State Statistics Service, the Database on Socio-economic figures in Russian regions Data: Socio-economic indicators for Russian regions Period: Territories: 77 Russian regions

Export of goods and services (% of GRP) in Russian regions, % of national export belongs to regions with quota >15%

Export productivity EXPY in Russian regions, 2003 Potential export capability Ω in Russian regions, 2003

Export diversification and economic growth in Russian regions: Does hold Basic result of Hausmann, Hwang, Rodrik (2007) Adjusted export productivity results in strong dependence Weak or zero dependence Strong positive dependence Doesnt hold GRP per capita Ω export potential GRP per capita EXPY Ω export potential EXPY Average annual growth rate of GRP per capita

Empirical estimation where: Variable (in logs)Proxy - Average annual growth rate of the dependent variable between year t and t+5 - GRP per capita - GRP per person employed - Employment - Dependent variable reflecting economic growth in year t - Export productivity measure in year t Potential export capability - Urbanization economies in year t Population density - Human capital in year tNumber of students enrolled in higher educational institutions per 10,000 population

GRP per capita growth GRP per person employed growth Employment growth Value added, initial Labor productivity, initial Employment, initial ** Export potential Ω *** *** *** Urbanization economies Human capital *** Constant N 78 Log likelihood Akaike info criterion Schwarz criterion Morans I (error) *** Lagrange Multiplier lag * Robust LM lag Lagrange Multiplier error *** Robust LM error *** Lagrange Multiplier SARMA ** Note: Significance at the 1, 5 and 10% level is signaled by ***, ** and *, respectively. Diagnostics of spatial dependence in economic growth models

MiningManufacturingConstruction Wage in region-industry 0.24* -0.15*** 0.03 Wage in region-industry, spatial lag 0.23* 0.15*** 0.05 Potential Export productivity, -0.04* 0.03*** 0.02* Potential Export productivity, spatial lag -0.08* -0.04*** -0.02* Human development Human development, spatial lag 0.12* Initial Employment -0.05* -0.01* Average employment growth outside -0.26* 0.10* -0.22** Cons N 77 R R2 adj AIC BIC Note: Significance at the 1, 5 and 10% level is signaled by ***, ** and *, respectively. Regression results for industrial sectors

Results Verification of new explanations Potential export productivity results in economic growth in contrast to actual export productivity EXPY that doesnt. Regions with high potential export productivity are locomotives of economic growth. It is necessary to distinguish regions with high potential export productivity from regions with high actual export productivity level. Verification of New new explanations In case of economic growth in adjacent regions negative labor mobility effects on export diversification are higher than positive inter-regional technological spillovers.