What happens to R&D in domestic multinationals after foreign acquisition?

Roger Bandick, Holger Görg y Patrik Karpaty, VOXEU, 14 January 2011

The recent waves of cross-border mergers and acquisitions have raised the question as to what will happen to the domestic firms, and in particular the headquarter activities of domestic multinationals, once they are acquired by a foreign owner.

In Sweden – a nation with particularly good data on its multinationals – this is an important issue. Former flagship Swedish multinationals such as Volvo, Saab, Asea, and Astra have all been acquired by foreign owners and, therefore, are no longer Swedish. And of course similar events are observed or mooted in many other countries.

Does this change in ownership imply that high skill intensive headquarter activities will now be no longer carried out in Sweden? This is an important question that is not only of academic interest, but also has strong policy implications. Recent theoretical models by Nocke and Yeaple (2008) and Bertrand et al. (2009) provide a framework for thinking systematically about the impact of such changes.

New evidence from Sweden

In a recent CEPR research paper (Bandick et al. 2010), we use Swedish firm-level data from 1993 to 2002 to examine what happens to headquarter activity, specifically R&D activity, in a firm once it gets taken over by a foreign multinational.

Figure 1 shows the importance of foreign and domestic multinationals in the manufacturing sector. Both types of multinationals account together for about 80% of total employment in the sector. However, the relative contribution of the two groups has changed markedly over the last decade. While Swedish multinationals accounted for the majority of employment in the early 1990s, the early 2000s see foreign multinationals as the largest group. This change is mainly due to acquisitions of Swedish multinationals by foreign owners.

Figure 1. Employment shares in Swedish manufacturing

Bandick and Görg (2010) and Bandick and Karpaty (2010) find that such takeovers do not result in any reductions in employment or higher probability of plant shut down. However, an important question is what happens to headquarter activities, or more specifically, domestic R&D activity, of previously Swedish owned multinationals. After all, the acquirer has, by definition, its headquarter activities in its own home country. Will the R&D of the target be relocated to the home country of the new owner, depleting Sweden of its high skill activity? Or will the R&D location be maintained and perhaps even extended as a result of the foreign takeover? And, how do these effects compare to the R&D effects of acquisitions of domestic non-multinationals – is there a special “multinational effect”?

In a snapshot comparison, foreign and Swedish multinationals have significantly higher R&D intensities (measured as the ratio of R&D expenditure over total sales in a firm) than non-multinationals. There is no statistically significant difference in the average R&D intensity between foreign and Swedish multinationals, however. This simple comparison does, of course, not allow any conclusions about a potential causal effect of a foreign takeover on R&D activity in the takeover target.

Figure 2. R&D intensity in 2002

In order to address this issue, Bandick et al. (2010) employ a propensity score matching approach combined with difference-in-differences. In a nutshell, the empirical strategy involves the careful construction of a valid counterfactual control group, i.e. firms that have similar characteristics to the acquisition targets, but that were themselves not taken over. For those two groups of acquired and control group firms, the difference in the growth of R&D intensity before and after the incidence of acquisition are then compared.

The estimation results suggest that R&D activity in Swedish firms does not decline after being acquired by a foreign owner. To the contrary, point estimates from the preferred econometric specifications suggest that R&D intensity increases by between 3% to 10% after foreign acquisition. While these effects are stronger for the acquisition of domestic non-multinationals than for Swedish multinationals, it is important to stress that even for the acquisition of Swedish multinationals the effect on R&D is generally positive, never negative. Hence, the results suggest that fears that the acquisition of large Swedish multinationals by foreign owners may lead to a relocation of headquarter and in particular R&D activities abroad appear unfounded.

Policy implications

The key policy implication of the analysis is that foreign acquisitions can have beneficial effects for domestic R&D activity. Hence, there is no need for fears and therefore no need for policymakers to start thinking about limiting international merger and acquisition activity. Quite the contrary, foreign acquisitions may be an important way to generate new knowledge and contribute to boosting the level of technology in the domestic economy.

References

- Bandick, R and H Görg (2010), “Foreign acquisition, plant survival and employment growth”, Canadian Journal of Economics, 43:547-573.

- Bandick, R, H Görg and P Karpaty (2010), “Foreign acquisitions, domestic multinationals, and R&D”, CEPR Discussion Paper No. 8081.

- Bandick, R and P Karpaty (2010), “Employment effects of foreign acquisitions”, International Review of Economics and Finance, forthcoming.

- Bertrand, O, K-N Hakkala, P-J Norbäck, and L Persson (2008), “Should countries block foreign takeovers of R&D champions and promote greenfield entry?”, IFN Stockholm WP 772.

- Nocke, V and SR Yeaple (2008), “An assignment theory of foreign direct investment”, Review of Economic Studies, 75(2):529-557.