In the early 1990s Russia embarked on the transition to a market economy as an independent country, but with institutions and other economic structures inherited from the Soviet Union. The combination of a regional differentiated industrial structure and lack of incentives for profitability has influenced Russia’s regional development in productivity. How the sectors and firms have adjusted to the market system and their ability to restructure varies, which could have influenced the regional distribution of industrial productivity levels.
In my thesis I have analyzed how relative productivity levels across the regions of Russia have developed during the period 1996–2004. As a measure on productivity level I have used labour productivity, defined as value added per worker employed in industry. I have focused on what has been traditionally regarded as the main Russian industries: oil and gas extraction, electricity production, mining and quarrying, together with manufacturing.
I have studied whether the Russian regions have converged or diverged in productivity level and how the observed pattern resonates with economic theory. Have investments and technology flowed into the least capital-intensive and less technologically advanced regions, such that the initially less-productive regions have caught up with the relatively more productive ones, – as predicted by the neoclassical Solow-model and the technology-gap model by Barro and Sala-i-Martin? Or will we find a tendency for agglomeration, as predicted by the Krugman (1991) model? In presence of economies of scale or local spillover effects the result could be divergence rather than convergence and give rise to development of highly productive economic clusters. I have also tested the importance of regional differences in industrial structure and resource endowments on the regional growth performance.
To answer my questions I have applied an empirical method using regional data from Goskomstat, which is the issuing body of official statistics in Russia. Since the analysis could be sensitive to regional and time specific effects I have applied panel data techniques, in addition to the cross-sectional analysis widely used in the literature, to control for these effects. I identify and control for some of regional specific factors in the cross-sectional analysis. Such factors could be domestic saving rate, the rate of foreign direct investments, population growth and migration, openness for trade and investment in human capital and R&D-activities. Especially, regarding Russia’s huge geographical area and the economy’s dependence on resource intensive industry, investigating spatial economic links and the role of the natural resource endowment on regional economic performance are interesting.
In my analysis I have found support for conditional convergence predicted by the Solow- model and the technology-gap model by Barro and Sala-i-Martin (1997), related to endogenous growth theory. Regional openness for trade and investment prove to be most important explaining the observed regional differences in growth rates. Industrial structure prove to be insignificant for regional growth, only the variable indicating regional resource endowments have some explanatory power. The effect is positive, but not robust and is sensitive for the other variables included in the analysis. There is no clear geographical pattern in which regions, which is showing best growth performance, they are rather scattered around the country.
The analysis supports a pattern of conditional convergence in productivity among Russian regions, but since the analysis only cover nine years it is difficult to draw any clear conclusion about the long term trend. When time goes it would be interesting to analyze the regional development of the Russian industry under market conditions over a longer time horizon.