Skip to main content
Log in

Domestic plant productivity and incremental spillovers from foreign direct investment

  • Article
  • Published:
Journal of International Business Studies Aims and scope Submit manuscript

Abstract

We develop a simple test to assess whether horizontal spillover effects from multinational to domestic firms are endogenous to the market structure generated by the incremental entry of the same multinationals. In particular, we analyze the performance of a panel of 10,650 firms operating in Romania in the period 1995–2001. Controlling for the simultaneity bias in productivity estimates through semi-parametric techniques, we find that changes in domestic firms' total factor productivity are positively related to the first foreign investment in a specific industry and region, but get significantly weaker and become negative as the number of multinationals that enter in the considered industry/region crosses a specific threshold. These changing marginal effects can explain the lack of horizontal spillovers arising in traditional model designs. We also find these effects to vary between manufacturing and services, suggesting as a possible explanation a strategic change in technology transfer decisions by multinational firms as the market structure evolves.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Figure 1

Similar content being viewed by others

Notes

  1. For example, limiting our attention to transition economies, the studies of Djankov and Hoekman (2000) on the Czech Republic, and of Konings (2001) on Bulgaria, Poland and Romania, either fail to find a significant positive effect or even detect a negative impact that multinational enterprises generate on the performance of domestic firms in the same sector. The situation is slightly different for developed countries, where some studies have found evidence of positive intra-industry spillovers (e.g., Haskel, Pereira, and Slaughter, 2007, using UK plant-level data).

  2. In the “horizontal” case, the most commonly used indicator of MNEs' presence is the MNEs' share of total employment within the industry considered. Such a practice might be itself subject to some criticism, as discussed in subsequent sections.

  3. The authors point to the fact that sectors in which there is a high level of foreign presence tend to be those in which Chinese firms are rather weak and have relatively low productivity. This causes the positive relation to weaken across sectors and eventually become negative.

  4. Smarzynska Javorcik and Spatareanu (2008) examine, for a sample of Romanian firms, whether the degree of spillover from foreign direct investment is affected by the foreign ownership modality and share in investment projects. We discuss the issue of foreign ownership and its impact on our results in our results section.

  5. Information on the FDI stock up to 1994 has been retrieved from the PECODB data set, a firm-specific collection of 4200 FDI operations undertaken in the countries of Central and Eastern Europe in the period 1990–2002, also based on the intermediate version of AMADEUS (2003) and developed by ISLA-Bocconi University. In terms of validation, the database is able to account for almost 70% of the region's total FDI inward stock in the early years of transition, as registered by official statistics.

  6. Since our sample does not include all NACE industries (and agriculture in particular), we have subtracted from official regional GVA data the output of those industries not present in our data set. The correlation between our sample and the official regional data comprising all NACE industries is instead 0.73.

  7. The classification is known as Pavitt classification, and makes it possible to divide industries into different technological patterns: economies of scale, traditional, high tech and specialized industries, plus services. The same grouping has been used by Davies and Lyons (1996) to divide industries into high, medium and low sunk costs. The classification therefore allows us to consider market structures, and hence prices, as relatively homogeneous within each industry.

  8. Imposing common input elasticities for firms belonging to different industries would in fact result in an overestimation of productivity for firms operating in sectors that have higher returns. The shortcoming of an industry-specific estimation is that, in a few cases (i.e., NACE16, NACE20), the number of firm-level observations available for each industry has not allowed a proper identification of the input coefficients. Accordingly, TFP measures from firms belonging to these industries have not been considered in the follow-up of our exercise.

  9. The LP methodology has been criticized on the grounds that the conditional demand for materials itself depends on the productivity shock, and thus materials are not a valid instrument to solve the simultaneity bias. The OP methodology does not suffer from this shortcoming, since the investment function is entirely determined before the productivity shock takes place. However, a major assumption of the OP approach is the existence of a strictly monotonous relationship between the instrument (investment) and output. This means that any observation with zero or negative investment has to be dropped from the data, thus potentially inducing a selection bias in the TFP estimation.

  10. Taking the dependent variable (TFP) in first differences also allows us to control for the unobserved firm-specific heterogeneity that may affect the correlation between firm productivity and foreign presence (e.g., Smarzynska Javorcik, 2004).

  11. Eslava, Haltiwanger, Kugler, and Kugler (2004) discuss this issue in their analysis of the productivity of Colombian firms, where they can exploit the availability of firm-specific information on prices and quantities. De Loecker (2007) provides a formal econometric discussion of the omitted price variable bias.

  12. Starting from firms' i revenues Y expressed as quantities time prices, and considering PI as the industry average price index, taking logs of the deflated revenue we have y i PI=q i +p i PI. To the extent that some domestic firms price below the industry average, we have that (p i PI)<0, and thus our observed deflated revenue y i PI is downward biased, leading to a similar bias in the TFP measure.

  13. Vertical spillovers would then be measured by weighting the horizontal penetration index with the input–output coefficients, as in Smarzynska Javorcik (2004).

  14. The average capital of foreign firms (proxied by total fixed assets) in our sample is around €2 million, but with a large standard deviation. The same is true for employment (average of 259 employees). Also note that, given our sector classification, we have excluded from our sample all foreign affiliates acting only as promotion agencies or sales representatives.

  15. Note that when assessing the overall impact of spillover as α+βCumFDI ijt −1, the coefficient α can be interpreted as the effect of the first investment on domestic firms' TFP changes.

  16. The minimum efficient scale has been calculated as the median employment of the firms in each industry.

  17. Since the restrictions to test are non-linear, the test is based on a Wald statistic (χ2-distributed) constructed through the estimated covariance matrix obtained from the unrestricted (linear) models (Greene, 2003: 176).

  18. We have also tested for backward and forward linkages, as in Smarzynska Javorcik (2004), finding weak evidence of vertical spillovers.

  19. Denoting by αS and βS the coefficients for Dzjt−1 × Serv and Dzjt−1(CumFDIzjt−1/MES z ) × Serv respectively, the χ2 and p-values of the tests H0: α+αS=0 and H0: β+β S =0 are 0.03 (0.85) and 0.02 (0.88), as retrieved from column 5 of Table 5.

  20. Albeit not significant, the combined effects of our estimates for α and β in services would point in the direction of positive spillovers, consistent with the findings of Vahter and Masso (2007), who, using a similar semi-parametric measure of TFP, find some evidence of higher spillovers in the services industry with respect to manufacturing in Estonia.

  21. The industry average proxy for absorptive capacity ranges across domestic firms in our sample from 0.1% (car production) to 6.9% (computer industry).

  22. Consistent with our previous results, the threshold for services as retrieved from column 5 of Table 5 turns out to be negative and not significantly different from zero.

References

  • Aitken, B. J., & Harrison, A. E. 1999. Do domestic firms benefit from direct foreign investment? Evidence from Venezuela. American Economic Review, 89 (3): 605–618.

    Article  Google Scholar 

  • Aitken, B. J., Harrison, A. E., & Lipsey, R. 1996. Wages and foreign ownership: A comparative study of Mexico, Venezuela and the United States. Journal of International Economics, 40 (3–4): 345–371.

    Article  Google Scholar 

  • Alcacer, J., & Chung, W. 2007. Location strategies and knowledge spillovers. Management Science, 53 (5): 760–776.

    Article  Google Scholar 

  • Arnold, J., & Smarzynska Javorcik, B. 2005. Gifted kids or pushy parents? Foreign acquisitions and plant performance in Indonesia. CEPR Discussion Paper No. 5065, London.

  • Baltagi, B. H. 2001. Econometric analysis of panel data. New York: Wiley.

    Google Scholar 

  • Belderbos, R., Lykogianni, E., & Veugelers, R. 2005. Strategic R&D location by multinational firms: Spillovers, technology sourcing and competition. CEPR Discussion Paper No. 5060, London.

  • Bhargava, A., Franzini, L., & Narendranathan, W. 1982. Serial correlation and the fixed effects model. The Review of Economic Studies, 49 (4): 533–549.

    Article  Google Scholar 

  • Blalock, G., & Gertler, P. 2008. Welfare gains from foreign direct investment through technology transfer to local suppliers. Journal of International Economics, 74 (2): 402–421.

    Article  Google Scholar 

  • Buckley, J., Clegg, J., & Wang, C. 2002. The impact of inward FDI on the performance of Chinese manufacturing firms. Journal of International Business Studies, 33 (4): 637–655.

    Article  Google Scholar 

  • Buckley, J., Clegg, J., & Wang, C. 2007. Is the relationship between inward FDI and spillover effects linear? An empirical examination of the case of China. Journal of International Business Studies, 38 (3): 447–459.

    Article  Google Scholar 

  • Cantwell, J. A., & Santangelo, G. D. 2002. The new geography of corporate research in information and communications technology (ICT). Journal of Evolutionary Economics, 12 (1–2): 163–197.

    Article  Google Scholar 

  • Capar, N., & Kotabe, M. 2003. The relationship between international diversification and performance in service firms. Journal of International Business Studies, 34 (4): 345–355.

    Article  Google Scholar 

  • Castellani, D., & Zanfei, A. 2006. Multinational firms, innovation and productivity. Cheltenham: Edward Elgar.

    Book  Google Scholar 

  • Caves, R. E. 1971. International corporations: The industrial economics of foreign investment. Economica, 38 (1): 1–27.

    Article  Google Scholar 

  • Damijan, J., Majcen, B., Knell, M., & Rojec, M. 2003. The role of FDI, R&D accumulation and trade in transferring technology to transition countries: Evidence from firm panel data for eight transition countries. Economic Systems, 27 (2): 189–204.

    Article  Google Scholar 

  • Davies, S., & Lyons, B. 1996. Industrial organization in the European Union. London: Clarendon Press.

    Google Scholar 

  • De Loecker, J. 2007. Product differentiation, multi-product firms and estimating the impact of trade liberalization on productivity. NBER Working Paper No. 13155, Washington, DC.

  • Djankov, S., & Hoekman, B. 2000. Foreign investment and productivity growth in Czech enterprises. World Bank Economic Review, 14 (1): 49–64.

    Article  Google Scholar 

  • Driffield, N., & Love, J. H. 2007. Linking FDI motivation and host economy productivity effects: Conceptual and empirical analysis. Journal of International Business Studies, 38 (3): 460–473.

    Article  Google Scholar 

  • Eslava, M., Haltiwanger, J., Kugler, A., & Kugler, M. 2004. The effects of structural reforms on productivity and profitability enhancing reallocation: Evidence from Colombia. Journal of Development Economics, 75 (2): 333–371.

    Article  Google Scholar 

  • Görg, H., & Greenaway, D. 2004. Much ado about nothing? Do domestic firms really benefit from foreign investment? World Bank Research Observer, 19 (2): 171–197.

    Article  Google Scholar 

  • Greene, W. 2003. Econometric analysis. Englewood Cliffs, NJ: Prentice-Hall.

    Google Scholar 

  • Griffith, R., Redding, S., & Simpson, H. 2002. Productivity convergence and foreign ownership at the establishment level. CEPR Discussion Paper No. 3765, London.

  • Griliches, Z., & Mairesse, J. 1995. Production functions: The search for identification. NBER Working Paper No. 5067, Washington, DC.

  • Haskel, J., Pereira, S., & Slaughter, M. J. 2007. Does inward foreign direct investment boost the productivity of domestic firms? Review of Economics and Statistics, 89 (2): 482–496.

    Article  Google Scholar 

  • Katayama, H., Lu, S., & Tybout, J. R. 2003. Why plant-level productivity studies are often misleading, and an alternative approach to inference. NBER Working Paper No. 9617, Washington DC.

  • Klette, T. J., & Griliches, Z. 1996. The inconsistency of common scale estimators when output prices are unobserved and endogenous. Journal of Applied Econometrics, 11 (4): 343–361.

    Article  Google Scholar 

  • Konings, J. 2001. The effects of foreign direct investment on domestic firms: Evidence from firm level panel data in emerging economies. Economics of Transition, 9 (3): 619–633.

    Article  Google Scholar 

  • Levinsohn, J., & Petrin, A. 2003. Estimating production functions using inputs to control for unobservables. Review of Economic Studies, 70 (2): 317–342.

    Article  Google Scholar 

  • Liu, X., Siler, P., Wang, C., & Wei, Y. 2000. Productivity spillovers from foreign direct investment: Evidence from UK industry level panel data. Journal of International Business Studies, 31 (3): 407–425.

    Article  Google Scholar 

  • Liu, Z. 2008. Foreign direct investment and technology spillovers: Theory and evidence. Journal of Development Economics, 85 (1–2): 176–193.

    Article  Google Scholar 

  • Markusen, J., & Venables, A. 1999. Foreign direct investment as a catalyst for industrial development. European Economic Review, 43 (2): 335–356.

    Article  Google Scholar 

  • Olley, S., & Pakes, A. 1996. The dynamics of productivity in the telecommunications equipment industry. Econometrica, 64 (6): 1263–1298.

    Article  Google Scholar 

  • Sinani, E., & Meyer, K. E. 2004. Spillovers of technology transfer from FDI: The case of Estonia. Journal of Comparative Economics, 32 (3): 445–466.

    Article  Google Scholar 

  • Smarzynska Javorcik, B. 2004. Does foreign direct investment increase the productivity of domestic firms? In search of spillovers through backward linkages. American Economic Review, 94 (3): 605–627.

    Article  Google Scholar 

  • Smarzynska Javorcik, B., & Spatareanu, M. 2008. To share or not to share: Does local participation matter for spillovers from foreign direct investment? Journal of Development Economics, 85 (1–2): 194–217.

    Article  Google Scholar 

  • Syverson, C. 2004. Market structure and productivity: A concrete example. Journal of Political Economy, 112 (6): 1181–1222.

    Article  Google Scholar 

  • Vahter, P., & Masso, J. 2007. Home versus host country effects of FDI: Searching for new evidence of productivity spillovers. Applied Economics Quarterly, 53 (2): 165–196.

    Google Scholar 

  • Van Biesebroeck, J. 2005. Exporting raises productivity in sub-Saharan African manufacturing firms. Journal of International Economics, 67 (2): 373–391.

    Article  Google Scholar 

Download references

Acknowledgements

We wish to thank Rene Belderbos, Davide Castellani, Chiara Criscuolo, Johannes Moenius, Alberto Pozzolo, two anonymous referees, and participants at the 2005 CNR Workshop in International Economics (Urbino, Italy), the 32nd EARIE Conference (Porto), the 7th ETSG meeting (Dublin), the 3rd EIIE Conference (Koper, Slovenia), WIIW (Vienna) and CEPII (Paris) for helpful comments and suggestions. The usual disclaimer applies.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Enrico Pennings.

Additional information

Accepted by Arjen van Witteloostuijn, Area Editor, 8 March 2008. This paper has been with the authors for three revisions.

Appendices

APPENDIX A: DATA SOURCES AND ISSUES

Balance sheet data for domestic firms and information on the presence of MNEs have been retrieved by the AMADEUS data set, commercially available from the Bureau Van Dijk (www.bvdep.com). The latter is a comprehensive, pan-European database developed by a consulting firm, Bureau van Dijk. It contains balance sheet data in time series on 7 million public and private companies in 38 European countries (2004 edition). The data set comes as a modular product: a version including the top 250,000 companies, the top 1.5 million (employed in this paper), or all 7 million companies in the considered countries. When using these data, three issues are worth pointing out:

  1. 1)

    The available data sets tend to exclude small firms (mainly domestic) from the records, and thus yield a lower proportion of domestic firms vs multinationals with respect to the Romanian population of firms. The latter issue does not necessarily distort our spillover measure: although it is true that smaller firms could be characterized, in principle, by a lower absorptive capacity of technological spillovers, they also tend to grow faster in terms of productivity. Since we use TFP changes as our control variable, the latter entails a conservative measure of productivity.

  2. 2)

    In AMADEUS the information on ownership is recorded only for the last available year (2000 or 2001), thus implying that some of the firms that we consider as foreign in 2001 might have been domestic in the years before. In order to gauge the magnitude of this issue, we have compared different yearly releases of AMADEUS, finding that, given an MNE in year 2000 or 2001, there is a 15% chance that the same firm is a domestic one before that year, whereas the probability of the opposite event (a firm switching from MNE to domestic) is negligible. However, the issue is not critical for our exercise, since the aim is to test the impact of the entry of MNEs on the average productivity of a sample of domestic firms. If we incorrectly attribute the multinational status to that 15% of firms that some time before 2001 were still domestic, we de facto exclude them from our dependent variable (domestic firms' TFP). The latter exclusion leads to a more conservative TFP measure, if we assume that MNEs acquire the most productive domestic firms (Arnold and Smarzynska Javorcik, 2005). Moreover, considering as MNEs some firms that for a certain number of years have remained domestic would lead to a more modest spillover effect, as we expect domestic entry to have a lower impact on domestic productivity than foreign entry. Thus, if anything, these potential measurement errors would lead to a more conservative assessment of the spillover effect.

  3. 3)

    In terms of the entry and exit dynamics of both domestic and foreign firms, the entry rate retrieved from our sample (see Table 1) matches very closely the official entry rate recorded by the Romanian Chamber of Commerce in the period considered. The lower exit rate reported in our sample is probably due to the large-firm bias of the data set, since in transition economies larger firms tend on average to benefit from softer budget constraints and display higher survival rates than small firms. Again, the latter issue does not affect our exercise: if soft budget constraints play a role, then our TFP is measured conservatively, because the selection effect driving out inefficient domestic firms works less intensively.

APPENDIX B: CLASSIFICATION OF INDUSTRIES

The model includes a total of 48 NACE two- and three-digit industries, grouped as follows:

  • Economies of scale industries: 10, 11, 12, 13 and 14 (mining of coal, metals and stone; extraction of petroleum and natural gas); 21 (paper and pulp); 22 (publishing and press); 241 and 242 (basic chemicals and agrochemicals); 246 and 247 (other chemical products and synthetic fibers); 251 (rubber products); 26 (other non-metallic products); 27 (metallurgy); 297 (domestic appliances); 31 (electrical appliances, excluding domestic); 321 (electronics); 322 and 323 (communication equipment); 341 (car production); 343 (car components); 351 (ship building); 352 and 354 (railways; motorcycles); 40 (energy).

  • Traditional industries: 151 and 152 (production and transformation of meat and fish); 153 and 155 (vegetables, milk and dairy products); 156 (grains); 157 (pet food); 158 (fabrication of bread, tea, coffee); 159 (drink and beverages); 16 (tobacco); 17 (textiles); 18 (clothing); 19 (leather); 20 (wood); 28 (metals); 361 and 362 (furniture); 363 and 365 (musical instruments and toys); 366 (other general manufacturing).

  • Specialized industries: 252 and 262 (plastic products); 291 (mechanical machinery); 292 (general machinery); 293 (agricultural machines); 294 and 295 (machine tools); 334 and 335 (optics, photography, clocks); 45 (construction).

  • High-tech industries: 243 and 245 (paintings and pharmaceuticals); 244 (pharmaceuticals); 30 (office machines and computers); 331 and 332 (medical and precision instruments); 642 (telecommunication).

  • Services: 55 (hotels and restaurants); 65 and 66 (financial intermediation and insurance); 72 (computer and related activities); 73 (research and development); 92 (cultural and sporting activities).

Rights and permissions

Reprints and permissions

About this article

Cite this article

Altomonte, C., Pennings, E. Domestic plant productivity and incremental spillovers from foreign direct investment. J Int Bus Stud 40, 1131–1148 (2009). https://doi.org/10.1057/jibs.2008.99

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1057/jibs.2008.99

Keywords

Navigation