The effectiveness of investment incentives: the Slovenian FDI Co-financing Grant Scheme

Abstract

This article analyses the effectiveness of the main instrument of Slovenian FDI policy, the 'FDI Co-financing Grant Scheme'. We look at the post-grant performance of foreign subsidiaries that received grants in 2000–09 using a double-track approach: calculation of performance premia of subsidised foreign subsidiaries, based on financial statements data, as suggested by Bernard and Jensen, and a questionnaire survey to tackle those qualitative aspects of subsidiaries' operations which are used in the official evaluation of grant applications. Subsidised foreign subsidiaries on average show better performance than comparable local companies and better qualitative characteristics than non-subsidised foreign subsidiaries. The main objectives of the scheme, creation of new capacity and jobs in export-oriented activities, have been achieved. The quality of this quantitative increase is more of a question; the data do not indicate any real breakthroughs in technological intensity, human resource development or productivity. In terms of technology and skills subsidised FDI projects remain more or less on the level of average Slovenian firms.

Acknowledgements

This analysis was been funded by contract no. PP 76 / 2011 (JN-S-0061/2011-POG-4300) between the Public Agency of the Republic of Slovenia for Entrepreneurship and Foreign Investment and the Faculty of Social Sciences, University of Ljubljana.

Notes

1. More detail on the contents and characteristics of the scheme may be found in JAPTI (2010).

2. This drop is due to the fact that in t ₀ a number of small firms, newly established for the purpose of subsidised projects, enter the population of subsidiaries analysed.

3. We also tested the impact of grant allocation on recipents' perfomance indicators by coarsened exact matching analysis. We do not report the results because the number of observations is too low for coarsened exact matching analysis to allow really reliable conclusions. Still, the results point to the same conclusions as the Bernard and Jensen method. The results of coarsened exact matching analysis are available on request.