Firm competitiveness and regional disparities in Georgia

Abstract

There are many challenges to building firm competitiveness in posttransition economies, particularly with the intensification of as global trade integration. Intranation variations in firm competitiveness are also stark, highlighting the need for policies to overcome the legacy of pretransition economic structures. Utilizing data from Georgia's annual firm census and household surveys, this paper analyzes the nature of the country's competitive landscape—measured as labor productivity—over the period 2006–2012. The results of our empirical estimations reveal that although a large proportion of a firm's competitiveness is associated with its own characteristics (sorting and compositional effects), location‐specific factors are also highly relevant. In particular, the extent of agglomeration, human capital endowments, and local expenditures—such as transport infrastructure investments—play a significant role in conditioning firm‐level competitiveness. Given current regional endowments, these findings highlight the significant attention that needs to be paid to building capacities in less‐favored areas, not only to ensure that trade integration does not harm Georgia's less‐favored regions, but also to make further progress in developing the country's private sector and fully maximize the export potential across its full stock of enterprises.

Key words:

• competitiveness
• productivity
• firms
• Georgia transition economies

Support from the European Research Council under the European Union's Seventh Framework Program (FP7/2007‐2013)/ERC grant agreement n° 269868 is acknowledged. The authors are also grateful to the editor and the anonymous reviewers for comments and suggestions to earlier versions of the manuscript. Rashmi Shankar, of the World Bank, and Nino Mosiashvili, of the International School of Economics (ISET) at Tbilisi State University, encouraged this research and facilitated access to data.

Support from the European Research Council under the European Union's Seventh Framework Program (FP7/2007‐2013)/ERC grant agreement n° 269868 is acknowledged. The authors are also grateful to the editor and the anonymous reviewers for comments and suggestions to earlier versions of the manuscript. Rashmi Shankar, of the World Bank, and Nino Mosiashvili, of the International School of Economics (ISET) at Tbilisi State University, encouraged this research and facilitated access to data.

Notes

Support from the European Research Council under the European Union's Seventh Framework Program (FP7/2007‐2013)/ERC grant agreement n° 269868 is acknowledged. The authors are also grateful to the editor and the anonymous reviewers for comments and suggestions to earlier versions of the manuscript. Rashmi Shankar, of the World Bank, and Nino Mosiashvili, of the International School of Economics (ISET) at Tbilisi State University, encouraged this research and facilitated access to data.

1. This is not unambiguously accepted, however. Timm (2013) suggests that these accolades hide a multitude of potential negatives, including increasing interference in economic activities, infringement upon property rights, and the creation of uncompetitive business contexts, all unlikely to aid in economic and social progression within Georgia.

2. The nine regions include Tbilisi (1), Adjara (2), Guria (3), Imereti, Racha‐Lechkhumi and Kvemo Svaneti (4), Kakheti (5), Shida Kartli and Mtskheta‐Mtianeti (6), Samagrelo‐Zemo Svaneti (7), Samtskhe‐Javakheti (8) and Kvemo Kartli (9). Data limitations force the noted aggregation of regions 4 and 6, consistent with the methodology employed by the Georgian Statistics Office. The Autonomous Republic of Abkhazia is excluded from the analysis due to a lack of data.

3. i.e. to overcome the potential bias evident in OLS productivity estimates, we instrument using intermediate inputs ‐ materials and energy expenditures. However, due to insufficient data on various firm inputs – including firm investments which precludes computing TFP via the Olley‐Pakes method – this imposes strong limits on our sample size, and are likely to introduce systematic bias. We therefore prefer to present the results of the analysis using the simpler measure of labor productivity, and seek to control as best as possible, given existing data limitations, for cross‐sectoral variation and differences in capital intensity.

4. We break each region into 1m square tiles, take an average elevation for each and construct an indicator that measures the coefficient of variation in these elevation data across the region's area.

5. KIS and HTM sectors include: KIS (NACE codes 64, 72, 73) and HTM (NACE codes: 24.4, 30, 32, 33)

6. For example, heads of households engaged in (often subsistence) agriculture are recorded as ‘individual entrepreneurs’, and family members as ‘unpaid family business workers’.

Additional information

Funding

European Research Council

European Union's Seventh Framework Program

Notes on contributors

Andrés Rodríguez‐pose

Dr. Andrés rodrÍguezpose is a professor at the Department of Geography and Environment, London School of Economics, London WC2A 2AE; [[email protected]].

Daniel Hardy

Mr. daniel hardy is a researcher at the Department of Geography and Environment, London School of Economics, London WC2A 2AE; [[email protected]].