بررسی دوباره درمورد اثرات سرایتی سرمایه گذاری مستقیم خارجی: فرایند وابسته به رویکرد

Theory suggests that FDI spillovers are a function of the level of foreign presence. We extend this view by developing and testing the premise that variations in spillover effects are the result of differences in the process of foreign entry. Drawing from the concept of time compression diseconomies, we develop the constructs of pace and irregularity of foreign entry, and demonstrate that they negatively moderate the relationship between the level of foreign presence and the productivity of host-country firms. We also explain how and why the moderating role of the process of foreign entry is influenced by the intensity of R&D and technical knowledge in domestic industries. We find evidence that firms operating in low-tech sectors are better able to absorb fast foreign entry.

International business theory suggests that host-country firms can realize substantial productivity gains from the very presence of multinational enterprises (MNEs) (Caves, 1974 and MacDougall, 1960). Despite the appeal of this theoretical proposition, recent empirical work on productivity spillovers casts serious doubt on their net effect. Prior studies have produced mixed results with the estimated spillover impact ranging from positive and high to insignificant or even negative (Aitken and Harrison, 1999, Barbosa and Eiriz, 2009, Buckley et al., 2007a, Liu et al., 2000 and Zhou et al., 2002). The inconclusive evidence indicate that there is no universal relationship between inward foreign direct investment (FDI) and the productivity of host-country firms, implying that FDI may be beneficial in some situations but not in others. Recognizing that this relationship involves important contingencies that are often poorly understood (Spencer, 2008), the international business literature has recently focused on a set of factors, such as the level of competition and export intensity, that may influence spillover effects (Blomström and Sjöholm, 1999, Buckley et al., 2007b and Wang and Yu, 2007). Yet, although industry conditions play a fundamental role in explaining variation in FDI spillovers (Buckley, Wang, & Clegg, 2006), the extant literature is largely silent on the relevance of the building-up process of foreign presence in a given industry. One convention, prevalent in this literature, is to attribute variations in firm productivity to differences in the level of the presence of foreign-invested firms in the same industry. By focusing on the level of foreign presence, however, previous studies fail to conceptualize the process of foreign entry and consider how this might lead to different productivity outcomes. This lack of understanding is important as host-country firms operating in industries with similar levels of foreign presence may experience considerably different spillover effects due to variations in the process-dependent evolution of foreign entry. More importantly, incomplete knowledge of the process of foreign entry may result in suboptimal FDI policies and make it difficult to develop theory about the conditions shaping the spillovers effects of foreign entry. The current study is designed to address this lack of understanding. We advance prior work on FDI spillovers in two important ways. First, we introduce and develop a foreign entry approach that is process-dependent. Our overarching argument is that FDI spillovers are influenced not only by the level of foreign presence in an industry but also by the process of foreign entry which is defined as how quickly and regularly foreign firms enter an industry within a given time frame. To our knowledge, this is the first study that is designed to explicitly capture the process of foreign-entry and conceptualize how it influences the spillovers effects of FDI. Our approach offers a more complete account not only of the direct effects of the process of foreign entry, but also of its moderating role in shaping the relationship between the levels of foreign presence and productivity. Such moderating effects play a significant role in understanding the industrial context under which the magnitude of spillover effects varies. Second, previous analyses provide valuable insights about the impact of industry heterogeneity on FDI spillovers (Buckley et al., 2007a, Kokko, 1994 and Liu et al., 2000), indicating that industry characteristics influence the relationship between the level of foreign presence and the productivity of host-country firms. Our contribution lies in developing and testing the proposition that the strength of the moderating role of the process of foreign entry varies between industries depending on their characteristics. To be specific, we examine how FDI spillovers are influenced by the interaction between the level of foreign presence, the process of foreign entry, and the intensity of R&D and technical knowledge in host-country industries. The process-dependent perspective developed in this study complements the traditional approach to FDI spillovers and has important implications for constructing theory about the mechanisms through which spillover effects occur.

In contrast to prior theorization that focuses on the level of foreign presence, our conceptual framework examines how, and the extent to which, the process of foreign entry moderates the relationship between foreign presence and local firms’ productivity. We employed a comprehensive panel dataset of Chinese manufacturing firms to test our hypotheses and analyze how and when domestic firms benefit from inward FDI. Our analysis offers new empirical evidence showing that the pace and irregularity negatively moderate the productivity enhancing effects of FDI. Our findings demonstrate that FDI spillovers depend not only on the level of foreign presence but also on its interaction with the process of foreign entry. This also helps us understand the reasons for the inconclusive findings of previous studies. By focusing on the level of inward FDI, prior research fails to conceptualize and document productivity variations that may result from the differences in the pace and irregularity of foreign entry. Building on the literature on knowledge flows, we developed and tested the premise that the strength of the moderating role of pace and irregularity depends on the nature and intensity of R&D and technical knowledge in a given industry. The findings support this theoretical prediction, indicating that domestic firms that operate in low-tech industries are on average better able to absorb and react to foreign entry. These findings not only reinforce the argument about the role of knowledge characteristics in shaping spillovers (Martin and Salomon, 2003, Reed and DeFillippi, 1990 and Spencer, 2008), but also extend the literature by incorporating the process of foreign entry in the existing approach. Our analysis has a number of implications for future research on spillovers. Our first contribution lies in explaining how variation in the process of foreign entry can result in different spillover effects and productivity outcomes. To understand how spillover effects arise, we demonstrate that one needs to carefully consider how foreign presence is built up in an industry over time. In this respect, we developed two constructs (pace and irregularity) that capture how quickly and irregularly foreign firms enter an industry within a given time frame. The implication is that how much a local firm benefits from FDI spillovers depends on how quickly and irregularly foreign investors enter the market. Even if two industries have ended up with the same level of foreign presence, different productivity effects may arise due to variations in the process of foreign entry. The findings suggest that low to moderate, constant and regular foreign entry allows local firms to absorb the benefits associated with inward FDI. By contrast, fast and irregular foreign entry leads to time compression diseconomies and makes it difficult for local firms to assimilate and benefit from potential externalities. Second, the findings regarding the role of R&D and technical knowledge imply that some industry-specific characteristics can help firms overcome the negative effects of fast and irregular foreign entry. As high-technology MNEs employ state-of-the-art technology to introduce innovative products, mainstream theory predicts that stronger externalities exist in high-tech environments where technical knowledge is rich and growing. Our analysis shows that this is not always the case. It supports the view that in situations where the pace of foreign entry is high, the less complex nature of technical knowledge enables local firms in technologically lagging industries to learn and benefit more than those firms in technologically leading sectors. From a theoretical point of view, these results emphasize the importance of carefully considering how industry heterogeneity may influence the process of foreign entry and result in different productivity effects. For instance, prior research identifies which characteristics (e.g., knowledge tacitness) influence spillovers to indigenous firms and which are most likely to cause crowding out effects (e.g., Spencer, 2008). Our study contributes to this research stream by adding the dynamic process of foreign entry to contingencies that are often of static nature. For example, although crowding-out effects are often attributed to competitive pressure imposed by foreign entry (Spencer, 2008), our study suggests that crowding-out effects may also arise from fast foreign entry in a short period of time. Furthermore, the literature links externalities to the strategy of the MNE, proposing that spillovers arise from deeper embeddedness and good relationships with local alliance partners (Spencer, 2008). Our study complements this view by showing that the MNEs’ decision pertaining to how fast they enter a foreign market has also a significant impact on spillovers to indigenous firms. Similarly, from a methodological point of view, our study represents a step forward in modeling the mechanisms of spillovers, and offers a refined methodology that measures spillover effects more accurately. Since FDI externalities are likely to occur over years, empirical analyses that ignore that foreign entry is a time-dependent process may provide an incomplete account of FDI spillovers. We developed a model that allows us to test the effects of both the level and dynamic process of foreign entry on the productivity of domestic firms. Our findings demonstrate that the effects associated with the process of foreign entry will not be picked up by the traditional econometric approach that only considers the level of foreign presence. Our study also has important implications for public policy. Our findings challenge the typical view that encouraging the unremitting inflow of FDI is an optimal policy for the development of domestic industry in emerging economies. The study underscores the need for governments to be aware that quick and uncontrolled inflow of FDI can be detrimental for domestic firms’ performance. Government should monitor not only the levels of foreign presence but also the dynamic structure of foreign presence. FDI in each industry should be built up in a modest, regular and controlled fashion. By contrast, a relatively high pace of foreign entry will have negative implications for the productivity of the host nation. This suggests that gradualism is more beneficial compared with a “big bang” approach. Hence, it may even be required that government temporarily restricts inward FDI when the pace of foreign entry is too high. Furthermore, the findings regarding the role of R&D and technical knowledge in overcoming the negative effects of fast foreign entry imply that government can allow for varying processes of foreign entry in different industries. Policy makers should consider and manage both the levels and pace of foreign entry on an industry-by-industry basis. Many governments, for instance, provide strong incentives to attract FDI in high-tech industries. Our analysis suggests that such subsidies may not achieve the expected performance outcomes when the process of foreign entry in these industries is not monitored properly. While our approach extends significantly previous empirical studies, several constraints should be taken into account when interpreting our findings. First, we did not control for a number of firm-specific idiosyncrasies that may affect spillovers due to data constraints. For example, controlling for the motives of FDI can be important. It is possible that the pace and irregularity that allow for the maximum effects of FDI are influenced by different investment motives. Similarly, the moderating effects of pace and irregularity may also vary between different modes of FDI entry. For instance, compared to greenfield investment, mergers and acquisitions (M&As) may allow foreign presence to built up at a greater pace. Unfortunately, our data do not allow us to control for such heterogeneity. Additionally, our analysis may be biased because we only considered the surviving domestic firms and did not take into account the fact that domestic firms may exit each period due to high competition. Future studies may address this by analyzing how the process of foreign entry affects the survival probability of domestic firms.