سرمایه گذاری مستقیم خارجی و عامل کشش تقاضا

This paper argues that the liberalisation of foreign direct investment (FDI) has made labour costs more important to domestic investment and long-run labour demand. It provides evidence from British and German data that is consistent with this view. First, high unit labour costs increase FDI outflows and lower FDI inflows. Second, the effect of unit labour costs on domestic manufacturing investment was more negative in the high-FDI 1980s than in the low-FDI 1970s, and this change was concentrated in high-FDI industries. The estimates suggest that the long-run labour demand elasticity may have risen substantially.

If higher labour costs induce firms to relocate production abroad, domestic employment will fall. In recent years, this has led some observers to argue that falling barriers to foreign direct investment (FDI) have made wage moderation more important for preserving employment. Unless wages are kept under control, the argument goes, capital will migrate to countries with lower labour costs, and unemployment will rise. If falling FDI barriers increase the elasticity of the capital stock with respect to costs, the underlying argument is perfectly consistent with standard labour demand theory. Hicks notes in The Theory of Wages that `[t]he demand for anything is likely to be more elastic, the more elastic is the supply of co-operant agents of production' ( Hicks, 1932, p. 242). This paper examines the implications of falling FDI barriers using data from Germany and Britain. The public perception of FDI liberalisation differs profoundly in these two countries. In Germany, it is seen as a threat that will either drive companies out of the country, or reduce Germany's high wages and generous social welfare benefits. By contrast, many British commentators see lower FDI barriers as an opportunity to attract companies that seek access to European markets but want to avoid continental `inflexibility'. Most commentators in both countries take it for granted that labour costs are a major determinant of FDI, and that falling FDI barriers make wage moderation more important for keeping investment and jobs at home. Britain's FDI inflows of around 2% of GDP, as well as its share of over 40% of all EU inward FDI (Eurostat, 1995), are among the British government's favourite statistics and have encouraged it to portray Britain as the `enterprise centre of Europe'. FDI inflows to Germany, by contrast, have been below 0.5% of GDP for many years, causing great concern among policymakers and pundits. However, it is less often realised that outflows are also much higher in Britain (about 3% of GDP) than in Germany (about 1.5%) so that net FDI outflows are rather similar in the two countries (see OECD, 1995). Hence, Britain's higher inflows may simply reflect a different industrial structure, with a greater role for multinationals, rather than more attractive locational conditions. Most policy debates take it for granted that FDI translates straight into physical investment.1 It is therefore interesting to see what has happened to capital formation in Britain and Germany. While the aggregate investment rate is substantially lower in Britain than in Germany, manufacturing investment, which may be more relevant to the relocation debate, is quite similar (Bond and Jenkinson, 1996). Both countries have seen their investment rates fall considerably since the early 1970s, but their relative positions have remained quite stable. These facts do not suggest that high costs have hurt capital formation in Germany as compared to Britain. Instead, it seems that if high costs have contributed to falling investment, both countries were affected in similar measure. Globally, FDI has grown dramatically since the early 1980s. The combined annual outflows from OECD countries (including flows within the OECD) have increased from less than $30 billion before 1983 to over $160 billion in every year since 1988. It seems clear that deregulation has played its part in this process. The United Nations Transnational Corporations Division (UNTCD, 1993) finds that with the abolition of exchange controls in Europe during the 1980s, outward FDI is essentially only subject to market forces. Some controls on inward FDI remain in various countries, but the liberalisation trend that `began in the mid-1970s has continued through the 1980s and early 1990s' (UNTCD, 1993, p. 17). In addition, trade restrictions have fallen, through both GATT and regional institutions such as the European Community. The impact on FDI is theoretically ambiguous. Falling trade costs increase `vertical' FDI, which is driven by production cost considerations, but reduce `horizontal' FDI, which is motivated by market access (Markusen et al., 1996). How do the elasticities of investment and labour demand with respect to labour costs relate to the level of FDI barriers? Throughout the paper, I assume that factor prices are not fully equalised by international trade, so that factor demand curves slope down. Possible reasons include trade barriers, factor endowments that lie outside the factor price equalisation cone, and imperfect product market competition. In the context of a very simplified model, Section 2shows that a partial dismantling of FDI barriers should not only increase foreign production and employment on impact, but also make both more sensitive to labour costs. Section 3investigates the effect of unit labour cost differentials on the bilateral FDI flows of Britain and Germany, and finds a significant and sizeable effect. This implies that, through the FDI channel, higher costs reduce both the long-term capital stock and, if capital and labour are quantity complements, also long-run labour demand. Section 4confirms that the effect of unit labour costs on manufacturing industry investment was substantially more negative in the 1980s than in the 1970s in both countries, and this change was concentrated in high-FDI industries. Section 5concludes by discussing the implications of a flatter labour demand curve.

This paper analyses the hypothesis that the liberalisation of foreign direct investment has made labour costs more important to domestic investment. Using a very simplified model of the multinational corporation, the paper demonstrates that falling FDI barriers will tend to make the effect of unit costs on domestic production more negative. The paper makes two empirical contributions. First, it shows that unit labour costs have a positive effect on bilateral FDI flows in both Britain and Germany. This effect is both quantitatively important and robust to the inclusion of control variables that may be correlated with the rate of return on investment. Then, the paper tests directly whether the effect of unit labour costs on domestic investment has changed between the 1970s and 1980s, a time when FDI grew substantially. And indeed, the long-run elasticity of manufacturing investment with respect to unit labour costs has risen significantly. This change is especially clear in those industries where FDI plays an important role. The estimated changes in the investment elasticity are quantitatively quite large. Both sets of results are consistent with the idea that greater openness to FDI has flattened the long-run labour demand curve by making investment location more responsive to costs. What are the implications of a flatter labour demand curve? Clearly, workers' bargaining power will be reduced, an effect which may be partly responsible for at least some of the problems experienced by the British – and to a lesser extent German – trade union movement over the last 20 years. In a general equilibrium setting, Wes (1996) shows that a flatter labour demand curve leads to lower aggregate unemployment as the markup of bargained wages over prices falls. However, it is important to note the underlying assumption that wage-setters have adjusted fully to the new labour demand environment. If wage-setting behaviour takes time to adjust, a transition phase of higher unemployment may result. Quite possibly, we are currently witnessing such a high-unemployment transition phase in both countries. It is probably too early to say whether the recent fall in British unemployment indicates that the transition phase is coming to an end.