How big are productivity gains from FDI?

Sebnem Kalemli-Ozcan, Christian Fons-Rosen, Bent Sørensen, Carolina Villegas-Sanchez, Vadym Volosovych 04 June 2013

a

A

During the decades of globalisation, flows of foreign direct investment (FDI) to both developed and emerging markets have surged in parallel with extensive policy momentum to increase FDI at the expense of debt as the major source of external financing.

This pro-FDI policy urge is justified by the perceived benefits of FDI on the host country. For example:

  • Foreign firms are expected to transfer new technologies and management practices to local incumbent firms – in particular in the case of joint ventures.
  • Local firms can be created by workers previously employed by foreign companies, leading to a more dynamic labour market in which locals learn from foreign firms and subsequently implement this knowledge in their newly opened firms.
  • Foreign corporations require a number of inputs to produce their products and they will source some of these inputs locally giving domestic input suppliers opportunities to engage in conversations and contracts with foreign firms who will be willing to transfer knowledge to their suppliers. 

Nonetheless, the entry of foreign firms can also have detrimental effects on the home economy.

  • Research has found a 'business stealing' effect where foreign-owned firms gain markets shares potentially leading to exit of domestic firms due to the more intense competition (see Harrison et al. 2011 and Bloom et al. 2013).

This is particularly likely to happen if local and foreign firms compete in the same narrowly defined product category.

  • Foreign firms may compete for resources, in particular labour.

One can imagine a situation in which a number of highly skilled employees relocate to foreign firms attracted by higher wages or better conditions, or demand higher wages to stay, so that local firms can end up being worse off after opening up to FDI investment.

The evidence

The net effect of foreign entry is an empirical question and so far the literature has found conflicting results. In the case of developed countries, there is evidence of positive spillovers from foreign firms to local companies operating in the same sector of activity, so-called horizontal spillovers (see Haskel, Pereira and Slaughter (2008) for the UK or Keller and Yeaple (2006) for the US). In the case of developing countries, however, the results are less encouraging. For example, Aitken and Harrison (1998) find negative spillovers for Venezuela.

These divergent results for developed and emerging countries have led researchers to suggest explanations which can encompass both cases. 

  • First, some developing countries may lack the necessary economic preconditions to benefit from FDI, for example, countries may only benefit from FDI if they reach a certain level of human capital and local financial development.

For example, a local blacksmith may not learn how to operate computer controlled welding machines or a software company may not get off the ground without start-up funds (Alfaro et al. 2004).

  • Second, foreign-owned firms may have an incentive to help firms which buy their products and, in particular, firms which supplies inputs and, therefore, positive effects of foreign investment may be found for such firms.

Using input-output tables and firm level data from Lithuania, Javorcik (2004) indeed finds that local firms which supply multinationals benefit from their presence.

New research

We try to address the question of whether the net aggregate gain from FDI is positive using a very large panel of firms from 30 European countries. Most of the literature focuses on just one country.

There are two main advantages of the multi-country setting. First, results from one country may be due to special conditions in that country and cannot be generalised to others (external validity compromised). In particular, differences in results between developed and emerging countries calls for a study with harmonised data for both types of countries such that it can be verified if the differences found between these groups of countries are due to differences between the particular countries and not an artefact of subtle differences in the implementation of the econometric studies.

Second, multi-country studies can hedge against various potential biases that single-country studies cannot. For example, certain sectors may have high productivity growth (e.g. telecommunications due to recent technological advances) and such sectors may at the same time attract foreign investment. Such a pattern would result in foreign investment and productivity growth being positively correlated, but this correlation would not reflect a causal impact of FDI on productivity. Using data from various countries, we can control for such sector-level patterns. Similarly, high growth and investment phases of particular countries, which might results in potentially spurious correlations, can be accounted for including country-year fixed effects.

One of the main advantages of our dataset is the highly detailed information on foreign ownership at the firm level. Firms provide information on the percentage of equity owned by foreign investors. Moreover, additional information is provided on the type of each foreign investor. In particular, it is possible to identify which investors are 'industrial' (foreign-owned companies operating in the industrial sector) and 'financial' (foreign-owned companies operating in the financial sector, such as banks, mutual funds, etc … ).

We exploit this ownership information to control for potential reverse-causality effects (i.e. it is possible that increases in foreign ownership are correlated with firm productivity even if foreign ownership does not cause productivity increases because foreign investors tend to  'cherry pick' better performing firms). While both industrial and financial investors seek high-productivity firms to invest in, we conjecture that only industrial foreign investment brings along the necessary production changes that can lead to productivity increases.

Our findings

We verify that:

  • Foreign-owned firms increase their ownership stakes in more productive domestic firms and that such investments are accompanied by higher market shares.

In terms of spillovers, we can sort out competition versus knowledge spillovers better than previous work because our dataset provides a very detailed sector classification of firms (the four-digit level, according to the NACE classification, where more previous research was limited to the two-digit level).

We identify competition spillovers as externalities resulting from the presence of multinational companies in the same four-digit sector of activity while we identify knowledge spillovers as externalities resulting from foreign-owned firms operating in the same two-digit sector but outside the four-digit sector.

Similarly to the early literature, we find different results for developed and developing countries.

  • In developed countries, local firms suffer from increased competition, while benefiting from knowledge spillovers from non-direct competitors.
  • In developing countries, local firms also suffer from increased competition, but there is no benefits found for non-direct competitors; i.e., there appears to be negative knowledge spillovers.

We believe that competition for scarce resources in developing countries may explain this discrepancy. If only a limited number of, say, engineers are available and some of these get hired by the foreign firms, local firms may be left with less expertise.

Finally, we explore heterogeneity; i.e., it the effects of foreign investment are different for different types of firms. We explore whether spillovers to local firms vary with:

  • The involvement of foreign owned companies in the local target measured by the percentage of foreign-owned equity; and
  • The initial level of productivity of local firms.

We do not find interesting results in the case of developing countries where competition effects seem to dominate. However, there are interesting results in the case of developed countries where all local firms suffer from the increase in competition; however, highly productive local firms further increase productivity due to knowledge spillovers from foreign companies. The greater the foreign ownership stake in these companies, the more they contribute to this beneficial impact for such local firms.

Aggregate results

Having obtained firm-level estimates, we can move to our main goal of evaluating the aggregate (country-level) net effects of FDI.

  • Our estimates imply that even very large increases in FDI are not important for country-level productivity growth.

For example, doubling FDI from its current levels results in a productivity increase of about 0.01% in developed countries, and a drop of 0.01% in emerging countries. These numbers incorporate both direct and spillover effects.

It is worth pointing out that even if we do not find significant aggregate total productivity effects from FDI, our study is silent about other potential effects of FDI. For example, FDI may generate employment, provide capital, and improve risk sharing and consumption smoothing. FDI may generate healthy competition in the labour market and FDI might even have growth-enhancing indirect benefits by impacting on structural policies.

References

Aitken B and A Harrison(1999) Do Domestic Firms Benefit from Direct Foreign Investment?," The American Economic Review, 89(3), 605-618.

Alfaro L, A Chandra, S Kalemli-Ozcan, and S.Sayek (2004) FDI and Economic Growth: the Role of Local Financial Markets," Journal of International Economics, 64, 89-112.

Bloom N, M Schankerman and J Van Reenen (2013) Identifying Technology Spillovers and Product Market Rivalry," Econometrica, forthcoming.

Harrison A, L Martin and S Nataraj (2011) Learning Versus Stealing: How Important are Market-Share Reallocations to India's Productivity Growth?," NBER Working Papers 16733.

Haskel J, S Pereira, and M Slaughter (2007) Does Inward Foreign Direct Investment Boost the Productivity of Domestic Firms?," The Review of Economics and Statistics, 89(3), 482-496.

Javorcik, B (2004) Does Foreign Direct Investment Increase the Productivity of Domestic Firms? In Search of Spillovers through Backward Linkages," The American Economic Review, 94(3), 605-627.

Keller W and S Yeaple (2009) Multinational Enterprises, International Trade, and Productivity Growth: Firm-Level Evidence from the US," The Review of Economics and Statistics, 91(4), 821-831.

a

A

Topics:  Development International trade

Tags:  productivity, FDI

Neil Moskowitz Endowed Professor of Economics, University of Maryland; Research Fellow, CEPR

Assistant Professor of Economics at Universitat Pompeu Fabra

Lay Professor of International Economics, University of Houston and CEPR Research Fellow

Assistant Professor at the Economics Department , ESADE Business School

Associate Professor of International Finance, Erasmus University Rotterdam

Events