FDI and the Multinational Corporation:
New Theories and
Evidence
Training and Mobility of
Researchers Network
Workshops and
Conferences
The Network's third Workshop...
A CEPR/LdA Workshop, entitled ‘Foreign Direct Investment and the Multinational Corporation’, was held in Turin on 18/19 May 2000. The Workshop was organized by Giorgio Barba Navaretti (Università di Ancona and Centro Studi Luca d'Agliano), Riccardo Faini (Università degli Studi di Brescia, IMF and CEPR), Anna Falzoni (Università di Bergamo and Centro Studi Luca d'Agliano) and Anthony Venables (London School of Economics and CEPR). It formed part of the activities of the TMR Network of the same name.
The first paper presented was ‘Optimal Location of Capital’ by Anthony Venables (London School of Economics and CEPR). In his paper ‘The Optimum Town’, Mirrlees (1972) posed the question, what is the optimal spatial organization of individuals around a town centre in which they work? Venables’ paper also addresses the spatial organization of activity, but whereas Mirrlees’ context was urban, Venables’ is international, and he develops a simple geographical structure in which the location issue can be addressed. In the model there are many countries at varying distances from an economic centre. The more remote countries have lower real incomes. Factor prices are determined by the interactions between factor endowments and the factor intensity of products, and between the distance from the centre and the products transport intensity. Does capital mobility benefit or hurt these low-income countries? A precise condition is found to determine the result: it is likely that capital is drawn to countries that are close to the centre, so are already relatively rich. This may take the form of capital moving from remote countries to more central ones, causing wages in remote countries to fall. Alternatively if there is an increase in the overall supply of capital, then this additional capital will locate in countries close to the centre, benefiting them and leaving remote countries unaffected – a finding consistent with observed flows of foreign direct investment. This equilibrium outcome contrasts with the socially optimum outcome in which the poorest economies are allocated greater capital stocks.
‘The District Goes Global: Export vs. Delocation’ was presented by Gianmarco Ottaviano (Università Bocconi, Milan, and CEPR) and co-authored Giorgio Basevi. The paper examines when a Marshallian Industrial District (i.e. a local agglomeration of small firms producing similar products with local learning spillover effects) begins to relocate production activity outside the district. Ottaviano depicts an industrial district as a centre for innovation in which local technological externalities sustain the endogenous invention of new goods by profit-seeking firms. After invention, firms face a crucial choice between reaching distant markets by export or by plant delocation. The paper shows how firms, in their attempt to circumvent the obstacles of goods and plant mobility, overlook the impact of their decisions on innovation activities inside the district, thus generating a suboptimal mix of export and delocation. Due to local technological spillovers from plant to R&D labs, firms’ choices to delocate their production abroad slows down the pace of innovation – i.e. the growth effect. Conversely, for the same reason, it increases local wealth – i.e. the wealth effect. And finally, it relaxes the competition among local producers to the detriment of the consumer surplus – i.e. the competition effect. The first and the third effect pull towards too much delocation, while the second effect causes too little delocation from a world welfare perspective. Ottaviano argued that, for high trade and low delocation barriers, the miscalculation by firms of the growth and competition effects dominates the wealth effect so that, from the perspective of the district, the market outcome overprovides foreign plants and underprovides exports. The reverse is true for when obstacles to trade are low and barriers to delocation are high. Nevertheless, decentralized decision making by firms results in suboptimal choices for the district as a whole. Frank Barry (University College Dublin) asked for the externalities – i.e. intersectoral rather than intra sectoral spillovers. Reinhilde Veugelers (Katholieke Universiteit Leuven and CEPR) asked what the impact on the results were if multi-plant operations were allowed for.
Kristof Dascher (University College Dublin) presented the paper ‘Trade, FDI, and Congestion – The Small and Very Open Economy’. A number of stylised facts of the boom of the Irish economy (used as an example of a very open economy with both free trade and free factor mobility) are explained by a model that merges features of neo-classical trade theory and regional and urban economics. Typically, a small and open economy trades goods at given world prices. Dascher presents a model of a small and very open economy where capital and labour are also internationally mobile. When targeting mobile capital (i.e. FDI) the economy’s government attracts not only mobile capital but also mobile labour. These inflows reinforce each other and contribute to rising welfare for the economy ’s indigenous population. But these benefits must be traded off against rising land rents, rising inequality and increasing exposure to adverse shocks.
Jean-François Ruhashyankiko (London School of Economics and CEPR) presented the paper ‘Ownership, Information Technology, and Multinational Activities’. There are two different types of FDI-theories: those that build on ownership, location, and internalization considerations; and those that build on the new trade theory. The first type of theory fails to explain the growth of FDI. The second fails to account for the bulk of FDI being among similar countries rather than different countries, relies on transport cost reductions as the sole case of FDI growth, and ignores two modes of production (licensing and contract manufacturing). Ruhashyankiko’s model contributes to the existing literature on multinationals and FDI by explaining how the level of FDI is affected by mergers and acquisitions; by considering improvements in information technology as a reason for the rapid growth in FDI; by analysing the efficiency of four alternative modes of production (i.e. technology licensing, multinational corporations, national corporations, and contract manufacturing); and by providing predictions consistent with North-North directions of FDI. Alessandro Turrini (Università Bocconi, Milano, and CEPR) questioned whether the variance of the productivity shock was the right way to capture the information technology revolution. Lucia Tajoli (Politecnico di Milano) remarked that the North-South nature of sub-contracting was not captured in the model.
Enrico Pennings (Universitat Pompeu Fabra, Barcelona) and Leo Sleuwaegen (Erasmus Universiteit, Rotterdam) presented the paper ‘Exit, Downscaling or International Relocation of Production’. Poorly performing firms need to improve their profitability through restructuring their operations. In many cases this results in downsizing by means of a collective layoff of employees. Based on a unique sample of firms reporting a collective layoff in Belgium, the paper analyses whether a firm dismisses all employees (exit), a significant proportion of its employees (downscaling), or closes down part of its activities and moves production abroad (international reallocation). The choice of downsizing approach differs following the strategic options and characteristics of the firm. A multinomial logit analysis is undertaken to relate the three different types of restructuring to a set of explanatory variables. It is found that those firms that relocate are typically more profitable firms, have invested more in the recent past and belong to a multinational group. Downscaling and relocation are more likely in the manufacturing than in the service industry. Downscaling firms are more capital intensive than relocating firms. Exiting firms are less profitable, smaller, younger, financed more by debt and more labour intensive than downscaling or relocating firms. Thus relocation may be driven by international production cost comparisons, while downscaling and exiting are motivated by loss of profitability.
Beata K. Smarzynska (The World Bank) presented the paper ‘Technological Leadership and the Choice of Entry Mode by Foreign Investors’. Developing country governments tend to favour joint ventures (JVs) over other forms of FDI, since they believe that local participation facilitates transfers of technology and marketing skills. Smarzynska’s paper assesses the potential of JVs for such transfers by comparing the characteristics of foreign investors engaged in JVs and wholly owned projects in eastern Europe and the former Soviet Union in the early 1990s. In contrast to the previous literature, the paper focuses on intra- rather than inter-industry differences in R&D and advertising intensities. The empirical analysis shows that foreign investors who are technological or marketing leaders in their industries are more likely to engage in wholly owned projects than to share ownership. These effects are present in high and medium technology sectors but not in low R&D industries. Smarzynska concluded that it is inappropriate to treat industries as homogeneous in studies of investment techniques. He suggested that JVs in high R&D sectors may present a lower potential for the transfer of technology and marketing techniques than wholly owned subsidiaries. Giorgio Barba Navaretti suggested that there may have been a shortage of local partners, since their technology capability may be restricted with respect to Western technology. Helen Louri (Athens University of Economics and Business) suggested using firm profitability as a control variable.
Helen Louri (Athens University of Economics and Business) and Marina Papanastassiou (Athens University of Economics and Business) presented ‘Inward Direct Investment in Greece: Home Country Determinants’, which was co-authored by Raymond Loufir. The paper tests neo-classical determinants (i.e. comparative advantage) of FDI flows and new economic geography determinants (i.e. home-market) of FDI flows against each other. The dataset consists of a 1997 Bank of Greece survey on the FDI-stocks of 343 foreign affiliates of nine OECD home countries located in Greece. Among the neo-classical determinants are the bilateral real exchange rate, wages in manufacturing, and bilateral exports and imports. Among the new economic geography determinants are GDP, GDP per capita, and R&D expenditure. First, a cross-section model is estimated using only neo-classical determinants. Second, a model is estimated using only new economic geography determinants. In both cases, the determinants are significant. Third, the neo-classical and new economic geography determinants are nested in a single estimation model, where the geography determinants become insignificant. However, the complementary relationship between FDI and trade indicates some new economic geography influence. Finally, industry specific estimations are run for the sub-sample of food and chemical industries to avoid mixing up constant returns to scale and increasing returns to scale industries, and geography determinants again become significant. Overall, inward FDI into Greece appears to be driven by the attraction of labour intensive industries that exploit cost advantages and by its use as an export platform to some Eastern European markets.
Stephen Pavelin (University College Dublin) presented ‘Firm Interdependence in Foreign Production: Leading UK Firms in 1986 and 1993’. The paper estimates econometric models explaining the foreign production of leading UK firms for the period 1986–93. Essentially the paper addresses two questions: what effect does one UK firm’s operations have on the foreign operations of another firm? And what effect does the UK operations of a foreign firm have on the foreign operations of UK firms. Pavelin tests three main hypotheses. First, UK firms that have the largest market share in the UK also invest more abroad – i.e. firms with sufficiently large market shares pass the scale that is sufficient to cover the fixed costs of operating an additional affiliate. Second, UK firms invest less abroad if there are other UK rivals that also invest in the same industry and region abroad – i.e. a smaller market share in the presence of rivals reduces the probability of an affiliate. And third, foreign rivals in the UK impede FDI of UK firms in the same industry – i.e. the market share of the UK firm in the home market may be too small to support foreign affiliates. The results of the empirical analysis confirm the first two hypotheses. But the activity of foreign competitors in the UK has no significant impact on foreign affiliate production of UK firms in the same industry.
Henrik Braconier (IUI, Stockholm) and Karolina Ekholm (IUI, Stockholm, and CEPR) presented ‘Multinationals and Wage-Competition Between Different Locations’. The paper explores the short run employment consequences of the actions of multinational firms. In general, the expansion of activity and employment abroad can accompany an expansion or a reduction of activity and employment in a multinational firm in the home country. The paper extends previous work by integrating the effects of employment changes not only of established affiliates, but also of new entries and exits of affiliates. The location decision and the employment decision of a multinational firm are looked at simultaneously. The data are based on a firm-level survey of Swedish multinational firms that starts in 1970 and covers approximately every fourth year until 1994. The results suggest that for the group of affiliates in high-income European countries, a plant is more likely to be set up in a location where the local wage is low, the local market is large and the best alternative wage (in a location without plants) is high. The wage costs in Swedish (or other) locations with plants have no significant effect on the decision to set up a plant in another location. However, there is evidence of a complementary relationship among affiliates in high-income European countries. Low-income European location decisions depend solely on the local market size. High-income non-European countries’ location decisions depend on total factor productivity rather than on the size of the home market and there is some evidence of a substitutionary relationship with similar locations. Marina Papanastassiou asked how the change of industrial structure was taken account of over the time period of the panel dataset. Frank Barry wondered how affiliate employment was effected if Sweden was in a temporary recession or faced a revaluation.
‘Multinational Firms: Easy Come, Easy Go?’ was presented by Jan I. Haaland (Norwegian School of Economics and Business Administration and CEPR) and co-authored by Ian Wooton. The paper examines how entry and exit barriers for multinationals affect the decision to locate in a country and the decision at which scale to operate. Specifically, the paper asks how government policy in the form of subsidies, loans, and redundancy payments affects the firms’ decisions. The model is set up with an affiliate of a multinational firm that operates as a monopolist in an integrated market without transport costs. Within the integrated market the firm can choose different countries which differ by their government policies. The cost function contains some fixed costs, entry costs and exit costs. There are adverse events that can occur at some point within the theoretically infinite life span of the affiliate operation such that the operations will be closed down completely and all workers will have to be laid-off. A government can initially subsidize an investment and impose required redundancy payments in the case of layoffs. The results show that a country with an inflexible labour market distracts investment and reduces the scale of operation and employment, since firms anticipate the redundancy payments in the case of their failure. Redundancy payments protect workers in the future, but reduce the likelihood of attracting foreign affiliates and reduce present labour demand. Countries with low labour market flexibility (i.e. high redundancy payments) will have to pay higher subsidies in order to still attract foreign affiliates. Also, the subsidy payments necessary to attract the affiliate will have to be higher the riskier is the industry. If the host country requires full or partial repayment of the subsidy in case of failure, then investment becomes less attractive. The optimal policy mix involves a negative redundancy payment – i.e. firms should be subsidized for layoffs. In this situation firms face lower costs in the case of failure and are more likely to invest and generate employment. In particular, the policy reduces the need for investment subsidies and raises the employment level of foreign affiliates in the host country.
Alessandro Turrini (Università Bocconi, Milano, and CEPR) and Dieter Urban (Centro Studi Luca d’Agliano, Università Bocconi, Milano) presented ‘For Whom is MAI? A Theoretical Perspective on Multilateral Agreements on Investment’. In 1998 the OECD proposed a multilateral agreement to liberalize FDI, which faced resistance by many of the Least Developed Countries (LDCs). The paper asks why LDCs may have opposed the proposal, even though they were completely free to opt out. The authors present a model where countries face a trade off between extracting rents from multinational firms and deterring FDI through political risk. There are multiple equilibria if MAI is not too strict or too soft: no country would like to enter MAI, all countries enter MAI, and some countries form MAI. World welfare is highest if all countries join MAI, because this minimizes the political risk of rent extraction and stimulates investment. However, countries with few holdings of multinationals may loose relative to a world without MAI because rents are shifted from governments to multinationals. Additionally, countries that opt out of MAI find that some FDI is redirected towards MAI members. Anthony Venables asked for the stability properties of the equilibria. He also remarked that the results are sensitive to the way MAI-membership is modelled and doubted that the asymmetry of countries’ loss of bargaining power was in line with the fact that everybody is equal before the law.
Return to Introduction
Workshops and Conferences