demand, the wage elasticity in MNEs’ subsidiaries is expected to be higher than that in purely national
firms since MNEs have the option of relocating activities or shifting employment across countries in
response to wage changes. This asymmetric position of MNEs as employers has a number of
consequences for the countries hosting FDIs. First, the degree of labour turnover in the host country
may increase as inward FDIs become more abundant. Second, there may be effects related with
changes in the demand elasticity for particular skill categories. In particular, if demand for unskilled
labour is more elastic in MNEs’ subsidiaries, union mark-ups may fall and a greater share of the wedge
between labour costs and wages (payroll taxes,...) will be effectively paid by workers (see, e.g., Rodrik
(1997) on this points).
While the relation between foreign ownership, productivity and wages has been extensively
investigated empirically, there is little work analysing the impact of inward FDI on labour demand
adjustment and on the elasticity of labour demand.4 The aim of this paper is that of contributing to fill
this gap by providing a cross-country perspective to the firm-level analysis of the relation between
foreign ownership and labour demand. The paper focuses on the relation between foreign ownership
and the characteristics of labour demand in given firms, while the effects on the pattern of firms’
shutdowns/openings are not investigated. In particular, we address the following questions: Is the
adjustment of employment systematically faster in MNEs’ subsidiaries? Is there any significant
difference in labour demand elasticities between national firms and MNEs’ affiliates? How does the
different employment behaviour of MNEs relate with country-level labour market regulations?
Our data set is an unbalanced panel covering more than ten thousand companies located in
eleven European countries observed in the period 1993-2000. The source of our data is the “Amadeus”
database collected by Bureau Van Dijk. We estimate firm-level, constant output labour demand
equations separately in each country by means of Arellano-Bond (1991) GMM estimators for dynamic
panel data. The ownership status of firms (foreign or national) is assumed to potentially affect the
speed of technical progress, wage and output labour demand elasticities, and employment adjustment
costs.
Our estimates are largely supportive of the view that employment adjustment in foreign owned
firms is faster than in purely domestic companies, after controlling for sector and size effects.
Employment changes are in fact significantly less persistent for MNEs’ subsidiaries, and this is a feature
that characterises all countries analysed. As for the wage elasticity of labour demand, there is clear
4 One of the few exceptions is Fabbri, Haskel and Slaughter (2002), who compare the evolution of labour demand
elasticities for MNEs and national firms in the UK.