University of St.Gallen, School of Economics and Political Science, Swiss Institute for International Economics and Applied Economics Research
Suggested Citation: Klodt, Henning; Christensen, Björn (2007) : HomeMarketEffectsofForeignDirectInvestment: TheCaseofGermany, Aussenwirtschaft, ISSN 0004-8216, Universität St.Gallen, Schweizerisches Institut für Aussenwirtschaft und Angewandte Wirtschaftsforschung (SIAW-HSG), St.Gallen, Vol. 62, Iss. 1, pp. 63-76
start turning positive after the 40% mark. Hence, this result indicates that low proportions offoreign ownership in a cluster are not optimal from the perspective of treated (i.e., foreign owned) firms. Instead, it seems to be thecase that a “critical mass” offoreign owned firms in a cluster (with the critical mass point being in the region of 40%) is necessary in order to stimulate productivity in foreign owned firms. This evidence thus suggests that foreign firms thrive in clusters with a high presence of other foreign firms. This is in line with arguments by Lee et al. (2013), Chen (2009) and Henderson (2003) who find that agglomerations of firms (in particular foreign owned) can be important for firm level productivity and development. Our evidence, however, shows that this is not necessarily a monotonically positive relationship, but that the positive effects only kick in after reaching a certain level offoreign presence in a cluster.
All labor market institutions reveal a negative sign, and only a few of them are in- significant. Using a fixed effects estimator in column (I) we find that the coefficients for wage distortion, union density, and employment protection are negative but not significant. Random effects in column (II) reveal the same sign pattern, but the co- efficients are now significant for wage distortion and employment protection. Union density reveals the right sign but the effect is not significant and thus zero. The Haus- man test strongly favors the random effects estimation. This result supports our theory by indicating that countries with lower labor market institutions seem to attract more FDI inflows than countries that have a tendency to protect their workers. Moreover, addressing cross panel heteroscedasticity by running FGLS yields significant and nega- tive coefficients for all labor market institutional variables. Even union density has the right sign and is significant for FGLS. Running IV regressions and instrumenting lags ofthe variable wage distortion and employment protection as instruments confirm the findings in column (1) and even the magnitude oftheeffects do not vary by much. Par- tial R-squares in all regressions range from 0.6 to 0.8 indicating that the instruments are valid. In columns (5) - (8) we redo the whole procedure with FDI-flows instead of stocks, which support the findings discussed so far.
strong evidence for negative spillovers (indirect effect), the total effect ofthe treatment (i.e., foreign ownership) is generally positive.
The non-monotonic relationship between the proportion offoreign firms in a cluster and the total treatment effect is further illustrated when one studies the marginal average treatment effects on the treated. Recall that this gives the marginal change in the potential outcome for treated firms when the proportion of treated firms changes from 0 to p. These marginal effects become increasingly negative up to a threshold of 20 percent. They only start turning positive after the 40% mark. Hence, this result indicates that low proportions offoreign ownership in a cluster are not optimal from the perspective of treated (i.e., foreign owned) firms. Instead, it seems to be thecase that a “critical mass” offoreign owned firms in a cluster (with the critical mass point being in the region of 40%) is necessary in order to stimulate productivity in foreign owned firms. This evidence thus suggests that foreign firms thrive in clusters with a high presence of other foreign firms. This is in line with arguments by Lee et al. (2013), Chen (2009) and Henderson (2003) who find that agglomerations of firms (in particular foreign owned) can be important for firm level productivity and development. Our evidence, however, shows that this is not necessarily a monotonically positive relationship, but that the positive effects only kick in after reaching a certain level offoreign presence in a cluster.
A second possible outcome from FDI is a skill upgrade, which is often the result from the TNC transferring advanced technology. In that case TNCs take advantage ofthe availability of a highly qualified labour force and comparatively low wages. As workers need to learn new production or management techniques, their skills upgrade. Evidence from Hungary clearly supports this argument. A number of TNCs undertake training programmes for their employees in Hungary. IBM, Nokia and Flextronics have opened their own training centres. Other companies, at least during the initial stages of their market entry to Hungary, sent their host-country staff back to thehome country for training. Samsung, which provided training to Hungarian engineers in South Korea, is a case in point. India’s Tata Consultancy Services (TCS), a newcomer to the Hungarian market, plans to train its employees partly in India and partly in Hungary. In addition to training their own employees, multinationals like IBM offer training opportunities to other companies (domestic and multinationals) who have no specialized training programmes of their own.
Park Madison Partners (PMP), a New York–based business leader in the real estate sector, acknowledges Colombian achievements as being due to “sound” macroeconomic
management ofthe economy. According to PMP, fiscal discipline and a successful inflation- targeting monetary policy have contributed towards creating a stable macroeconomic environment together with persistently positive growth rates, even in the wake ofthe most recent worldwide financial crisis; 2 see figure 1 (left-hand side). PMP further notes that Colombian monetary authorities have wisely decided not to obstruct any market-driven appreciation ofthe Colombian peso, in order to reassure foreign investors regarding the political commitment to avoid market distortions and policy-induced exchange rate risks. 3 Thanks to such policies Colombia is now characterized by “a vibrant and developing capital market [...and the above] attractive fundamentals also create significant opportunities in real estate” (PMP 2013, 12). The increasing balance of payments surpluses (figure 1, right-hand side) and mounting capital inflows would seem to confirm Colombia’s bright future.
2. FOREIGNDIRECT INVESTMENTS IN THE WORLD, SOUTHEAST EUROPE AND WESTERN BALKANS
Foreigndirect investments are a way of entering in new foreign markets, through which the company realizes international directinvestment in manufacturing units onto foreignmarket. There are two ways for directinvestment: the company can make acquisitions of other companies on foreignmarket or it can develop its capacities in theforeign country from the beginning and it is called greenfield investment. Acquisitions are a popular way for entrance theforeignmarket, mainly due to the rapid access. In this strategy the risk is lower, unlike greenfield investments, and the results from the acquisition can be assessed more easily and accurately. Greenfield investments represent the establishment of a new company in a foreignmarket which is a daughter company ofthe international company from thehome country. Most often this strategy is complex and expensive, but allows complete control ofthe international company and has the greatest potential for ensuring return on invested capital.
To conclude, it should be stressed that these favourable effectsof FDI on growth found for the Spanish regions would be greatly dependent upon their stability and permanent nature. While the huge affluence of FDI to the Spanish economy following her accession to the EU in 1986, would have led to a positive outcome in terms ofthe evolution of GDP per employee, the picture might be changing since the end ofthe 1990s (i.e., coinciding with the end of our sample period). In fact, last years have witnessed a process offoreign capital divestment, following recent changes in the strategies of MNEs, which has reached significant levels in the Spanish case (FERNÁNDEZ-OTHEO and MYRO, 2004). Accordingly, it would not be unlikely that the results found in this paper should be qualified in the next future. Also, this fact should be borne in mind by those regions seeking to attract FDI as an engine of technology transfer in order to fostering economic growth.
and show how they differ from general FDI in terms of size, age, labour productivity, and their regional and sectoral distribution. Second, we undertake a regression analysis to see whether the subtypes of patenting FDI deviate from general FDI in their response to a set of location determinants at the country level. In doing so, we use a novel and extremely detailed firm-level dataset on the universe of German outward FDI and patenting activity abroad. The dataset was created by merging the MiDi dataset ofthe Deutsche Bundesbank (the German Central Bank), PATSTAT and the Bureau van Dijk Amadeus dataset. It contains 12,631 multinational parent firms and their patenting activity, 42,934 linked foreign affiliates, 202 countries and spans the years 1999-2011. We show that within the same multinational parent, foreign affiliates associated with patenting activity are up to 39% larger, 13% older and have 14% higher labour produc- tivity than the average affiliate. Across different parent firms, the differences are even larger. Patenting FDI is much more focused on developed host economies and the manu- facturing sector. Concerning the dynamic relationship between patents and FDI, we find that affiliates where patenting activity occurred before the FDI entry are much larger and hold higher-quality patents than those affiliates where the timing order is reversed. In addition, our results show that patenting FDI reacts very differently to a whole set of lo- cation determinants: At the extensive margin, patenting FDI is more sensitive to changes in market size, local R&D spending, patent protection and the presence ofhome-country immigrants than general FDI. For distance and local profit taxes, the picture is reversed,
poverty. In other situations, foreign investors may preempt investment opportunities for any local firms, the resulting direct unemployment impact may not be of great value since similar results would have been occurred otherwise. With regard to indirect employment in vertically related entities, including backward (or upstream) linkages like suppliers, subcontractors, service providers and forward (or downstream) linkages like distributors, service agents, FDI’s implication is more complicated. It may raise employment in backward-linkage entities when it purchases raw materials, spare parts, components and services from them helping them extend operations. On the contrary, FDI may have no effect or even negative effects when it relies on imported inputs. Similarly, FDI may have a positive impact on employment in forward-linkage entities when using local distributors or may not have any positive impact otherwise. With regard to indirect employment in horizontally related entities like local enterprises competing in the same industries with foreign affiliates, FDI may have a negative impact when it outcompetes these local entities. This kind of effect is especially significant when foreign affiliates with capital intensive and knowledge intensive technologies replace small, and usually labour intensive, enterprises. This may quite be thecase since foreign investors are supposed to possess a large pool of technology that may grant them a higher productivity compared with their domestic counterparts equipped with poorer technologies. In contrast, FDI may have a positive impact when it helps the domestic enterprises raise the productivity or the quality of products, unintentionally or compulsory by host governments, thereby expanding their access to theforeignmarket for example. Macroeconomic effectsof FDI on employment refer to employment indirectly generated in the host economy as a result of spending of FIEs’ workers or shareholders or
We briefly summarize the literature on experience and ownership choice. Most ofthe studies find a positive impact of experience on the probability that a multi- national firm operates a wholly owned subsidiary instead of less integrated modes such as joint ventures. However, experience does not seem to matter for invest- ments between developed countries. Furthermore, some authors propose a non- linear relationship, yet related empirical evidence is weak. The econometric mod- elling mostly relies on binary or multinomial logit models. Typically, they do not consider firm cluster effects and thus do not allow activities of one firm to be cor- related with each other. Yet, it seems plausible that there are unobserved firm effects. In caseof multinomial outcomes the discussion on the appropriate choice of relevant outcomes tends to be limited. A problem arises if we consider joint ventures and subsidiaries to be similar investment choices compared to trade. In that case these three alternatives should not be considered simultaneously in a multinomial logit model that requires the independence of irrelevant alternatives. We discuss that point in more detail in Section 4.6.2.
4 that FDI is an area for which there is plenty of reliable quantitative evidence on developing countries, but that does not seem thecase. Yet, available data provide stronger support for differentiating the effect of FDI on growth across levels of development rather than in terms of geographic regions. The second set of results relates to the distribution oftheeffectsof FDI. We find that, in the micro studies, 44 percent of these estimates are positive and statistically significant, 44 percent are insignificant and 12 percent are negative and significant (see Figure 1). In thecaseof macro studies, 50 percent ofthe estimates are positive and statistically significant, 39 percent are insignificant and 11 percent are negative and significant. These effects tend to be large in macro than in micro studies. Our results also suggest that publication bias is not particularly severe in this body of evidence, especially when methodological differences are taken into account. Thirdly, regarding the reasons for the observed variation on the estimated effectsof FDI, we show that the choice of econometric method and specification are most important factors. We find evidence that those empirical specifications that control for endogeneity and firm level unobserved heterogeneity tend to report significantly smaller effectsof FDI (in micro studies), and the same for those that take into account the interaction of FDI with R&D expenditures, trade openness, human capital, and financial openness in macro studies. Finally, the FDI effects are larger for backwards than for any other type of linkages.
In our model foreigndirectinvestment has an unambiguously beneﬁ- cial direct eﬀect on the host economy by providing access to more eﬃcient capital markets, while the change in the composition of wages leads to an adverse eﬀect only in caseof sudden capital withdrawal. To keep our anal- ysis tractable and focus on the composition eﬀect, we abstract from other channels such as technological or human capital spill-overs from foreign to domestic ﬁrms, or ﬁercer competition in domestic markets (see e.g. Fosfuri et al., 2001, Markusen and Venables, 1999, among others). These beneﬁts should matter particularly for economies that suﬀer from scarcity of capital and severe capital market imperfections, i.e. relatively backward economies (Findlay, 1978). Empirical ﬁndings on eﬀects offoreigndirectinvestment on growth remain ambiguous; eﬀects seem to be highly dependent on host country characteristics (see the survey by De Mello, 1997). Mayer-Foulkes and Nunnenkamp (2005) ﬁnd that U.S. foreigndirectinvestment contributes to convergence in income only for countries with a relatively high per capita income ex ante, whereas eﬀects on middle and low income economies are ad- verse. The same holds for studies of speciﬁc channels such as productivity spill-overs (see the survey by G¨ org and Greenaway, 2004). Aitken and Harri- son (1999) analyze panel data from Venezuela and ﬁnd a small net impact on plant productivity, which appears to be seized entirely by joint ventures in- volving multinationals. Similarly, Javorcik (2004) reports that productivity spill-overs only occur through joint ownership or vertical linkages between foreign-owned suppliers and domestic ﬁrms in upstream sectors in Lithua- nia. Borensztein et al. (1998) ﬁnd that foreigndirectinvestment seems to be particularly eﬀective when the host country is endowed with suﬃcient human capital; Alfaro et al. (2004) and Hermes and Lensink (2003) ﬁnd this to be thecase when local ﬁnancial markets are suﬃciently developed.
The introduction of a common currency completed the pre-existing common market and advanced the financial integration in Europe, thus it led to the complete elimination of barriers to trade and to the movement of capital. The consequent market integration degraded the motives for market seeking FDI (especially for intra- FDI due to the already low barriers as result ofthe already established European Union). Individual markets are now easier to be served through the conventional trade networks, and import substituting intra-EMU FDI becomes a less attractive option for the expansion of firms in Europe. Import substituting FDI is significant in the cases of Ireland, France, Spain, Italy and Germany, thus any depreciation ofthe motives for such FDI after the introduction ofthe Euro would lead to a negative influence of Euro membership on FDI inflows. Although the motivation for market seeking FDI is now less significant, motives for both rationalized and strategic assets seeking FDI remain strong after the formation ofthe Euro zone. Both FDI types are based on the competitive advantages individual countries have to offer on production cost, agglomeration economies, and technological inputs. Both factors are proved to be positive and statistically significant determinants of intra-EMU FDI, see the 3 The intra EMU-FDI variable has been taken from OECD, Patent applications, exchange rate, GDP
Endogenous growth models allow for a greater impact of FDI on growth. On the one hand, FDI could lead to externalities on the domestic production factors; the effect on growth, however, would be permanent only if the resulting returns to scale over all factors (i.e., including the externality) turn to be increasing. More importantly, the endogenous growth literature has tried to formalize technological innovation, which would emerge as a response to economic incentives, that is, profit opportunities detected by firms that would be influenced by the institutional, legal, and economic environment in which they act (Grossman and Helpman, 1994). And, in turn, this would lead to stress the role of FDI and, in general, the degree of economic integration, on influencing technological progress and consequently growth rates. So, a higher integration would mean an increase in market size, which would lead to greater incentives to R&D and hence higher growth; and it would facilitate the diffusion of knowledge among countries and avoid duplication ofthe research activity (Romer, 1990; Grossman and Helpman, 1991). In particular, integration among relatively similar economies would lead to a higher growth rate in the long run, since it would allow the exploitation at the world level ofthe increasing returns that would exist in the R&D sector (Rivera-Batiz and Romer, 1991).
outward FDI. We also include a firm’s market share and its markup in the control variables. A firm’s market share in its industry and its markup pick up the degree of competition faced by that firm. The higher its market share and markup, the less competitive pressure a firm faces and, as a result, it has less incentive to improve its production technology. Therefore, the sign on the estimated coefficient on firm market share and markup on productivity are both expected to be negative. Finally, in the standard OLS regression the standard errors ofthe estimates will be downward biased if the error terms are correlated within groups in some way, which leads to spurious statistical significance ofthe estimated coefficients. To deal with potential correlations ofthe errors, we allow errors belonging to the same 3-digit NAICS industries to be correlated.
Substituting the values in Table 5 into (36) yields the conditions on α under which the subsidies make each country better oﬀ, apart from the eﬀect that operates through employment creation. Table 6 shows the results. In China, corporate tax and production tax are unambiguously beneficial, apart from the employment-creation eﬀect. This is also thecase for Vietnam if its worldwide equity share is less than 50%, which is plausible. Moreover, from (20) and (28), the employment-creation eﬀects of corporate tax and production tax are positive if these countries’ industries are capital intensive. Therefore, corporate tax and production tax benefit these two countries. The welfare eﬀect of an employment subsidy that operates through employment creation is unambiguously negative. Hence, for these two countries, the total welfare eﬀect of an employment subsidy is ambiguous.
A dynamic panel model 6 has been implemented in this context and all in- dependent variables were used in their logarithmized version. In accordance to Borb´ ely (2006) the term FDI signifies the amount of FDI inflows while LABOR signifies the amount of laborers employed in the respective sector and country. RCA signifies the modified version ofthe RCA indicator. Fi- nally, VA signifies the amount of value added generated by the sector. The value added is included as an additional variable to control for size effectsofthe domestic size ofthe respective sector’s market. Two dummy variables LOWP and HIGHP were added as control variable to account for member- ship in the lowest or the highest price category. The medium price category is thus the reference group. The twelve dummy variables SECT02 to SECT13 were added to control for sector specific effects. The base is here the sector of food products, beverages and tobacco products. The other sector dummies are:
The absence ofdirect positive effect of FDI on TFP is usually explained by low absorption efficiency ofthe economy thus making it impossible for the country to benefit from increase in human capital and technology (Borensztein, 1998). Furthermore the levels of economic freedom, openness ofthe economy and establishment of efficient financial environment also have negative effect. For example, negative relationship between FDI and TFP has also been present in a study by Sadik and Bolbol (2001). For several developing Arab countries (Egypt, Jordan, Morocco, Oman, Saudi Arabia and Tunisia), they investigate whether FDI affects TFP through technology spillovers effects. They find that FDI has actually a “very significant and negative effect” on most ofthe countries included in the study. However, they establish clearly that these effects might be caused by inefficient governmental policies and institutions, lack ofinvestment efficiency and inadequate appreciation and availability of technological innovation.
While pure forms of these types of FDI exist, the decision to set up an operation abroad is often motivated by a combination of factors. For example, a given location may offer both a competitive cost level and a sizeable local market.
In all these forms of FDI, the firm’s decision to internalise cross-border economic activities and so to become a TNC is driven by the economics ofthe boundaries ofthe firm. As suggested by Coase (1937), in principle any economic transaction can be conducted either via ad hoc market interaction between economic agents or internalised within a firm. In between these two extremes there are also many hybrid solutions, such as licensing and other forms of strong contractual arrangements between independently owned firms. Which of these solutions is chosen depends on the relative costs ofthe different alternatives. For example, in thecaseof repeated and relatively complex transactions, information and coordination problems can incur high transaction costs when conducted via themarket. This encourages their internalisation within a firm, where the transaction can be more closely supervised. On the other hand, dispersed hierarchical organisations come with their own information and incentive problems. It may also undermine the incentives for innovation that is normally generated by competition. Effective corporate governance typically becomes more difficult and costly the more distanced the owner is – culturally, functionally and geographically – from the economic activity.