Global activities and plant survival: the case of Sweden
Helpman et al. (2004). There is consensus in the recent theoretical and empirical literature about the clear ordering of firm types; the best being outward investors, the next-best being exporters and the least equipped firms remaining in the domestic market.7 This may have implications for the survival prospects of the plants as well. However, as described above there is no theoretical evidence on the impact of firm ownership structure on plant survival. On the one hand, we may expect plants of MNEs to have lower survival rates than other plants, since MNEs can easily shift production from one country to another whenever the conditions in the home or host country change to their disadvantage. We may also expect the plant survival rate to be lower for foreign MNEs than for domestic MNEs, since the former are less rooted in the local economy and may respond more quickly to adverse shocks. On the other hand, foreign MNEs may face substantially higher sunk costs of setting up new plants abroad, as do comparable indigenous firms, which should lead to lower exit rates for their plants. Finally, exporting firms are less dependent on the domestic product market. Therefore, if the domestic market is hit by a negative shock, plants of exporting firms are better equipped to survive, given their ability to cushion the adverse effects through the export market. The results estimating equation (2) on a sample of Swedish manufacturing plants in firms with 50 or more employees in the period 1993-2002 are reported in Table 13.3. First, the result in column 1 reveals that plants of exporting
non-MNEs have almost 10 per cent higher chances of surviving as compared to plants belonging to non-exporting non-MNEs. This result is in line with the theoretical prediction developed by Bernard et al. (2003) and Melitz (2003), arguing that exporting firms are less likely to fail than non-exporters due to the higher productivity of the former. However, controlling for productivity and other factors on plant, firm and industry level that may affect the survival ratio, the result in column 2 still suggests that plants of exporting non-MNEs are less likely to exit the market than plants belonging to domestic-oriented firms. One explanation for this could be that plants of exporting non-MNEs face higher sunk costs of exiting the market, since, as shown in Table 13.2, they are significantly larger in terms of number of employees, are more skill-intensive and have a higher productivity relative to their non-exporting counterparts. Moreover, a negative shock in the domestic market has a stronger effect on plants of purely domestic firms, in that they are more dependent on their home market. The result in Table 13.3, column 1 also reveals that the survival prospects not only differ between MNE and non-MNE plants, but also between the MNE plants depending on whether they are foreign-or domestic-owned. Consistent with the existing empirical findings when conditioning on plant-specific differences (e.g. in Bernard and Sjöholm, 2003; Görg and Strobl, 2003a; Van Beveren, 2007), the results support the perception of ‘footloose’ behaviour of MNEs, but only for Swedish MNE plants, which have the highest exit rates among all plants in the Swedish manufacturing sector. Controlling for other factors affecting the exit rate in column 2, the results suggest that Swedish MNE plants have a 4 and 24 per cent higher probability of exiting as compared to non-exporting non-MNE plants and exporting non-MNE plants, respectively. Moreover, it seems that foreign MNEs are more committed to being located in Sweden than Swedish MNEs; the hazard ratio is larger for Swedish MNEs than for foreign MNEs. The result in column 2 shows that the former have an approximately 24 per cent higher exit probability than their foreign counterparts. Plants of foreign MNEs also seem to have lower exit rates than non-exporting non-MNEs, while the exit rate of foreign MNE plants and exporting non-MNE plants seems not to differ.8 The results in Table 13.3 may indicate that since an investment abroad entails such large sunk costs, even if the conditions deteriorate in a country, foreign MNE plants (and plants of exporting non-MNEs) tend to stay longer on the market than domestic MNE plants. Taking into account that the sample in Table 13.3 includes plants in larger firms (i.e. firms with 50 or more employees)9 and that foreign MNEs are more concentrated in sectors where product differentiation and R&D intensity are very high (see Table 3 in Bandick, 2010), it appears that foreign MNEs may face substantially higher sunk costs than do comparable domestic MNEs. In addition, given that domestic MNEs have been established in the domestic market for a long time, they would face lower sunk costs in shutting down some of their plants as compared to foreign MNEs. This may explain the result that foreign MNE plants have lower exit rates than plants of domestic MNEs.