سرمایه گذاری مستقیم خارجی و اثر سرریز صادرات: مدارک و شواهد از ویتنام

The existing IB literature suggests that the presence of foreign firms in a country can benefit domestic firms through the formation of inter-firm linkages. These linkages can take various forms. By making use of firm level data from Vietnam's manufacturing sector, this paper examines the impact of horizontal and vertical (backward and forward) linkages between domestic and foreign firms on (i) the decision of domestic firms to export and (ii) the export share of domestic firms. This paper considers only transactional linkages. The empirical analysis is based on Heckman's two-step estimator in selection models. It is shown that the presence of foreign firms in Vietnam, through horizontal and forward linkages, significantly affects the decision of domestic firms to export as well as their export share. This result continues to hold when we take into account factors such as the (a) level of technology of domestic firms, (b) ownership structure of domestic firms, (c) orientation of foreign firms and (d) geographical proximity to foreign firms.

Foreign direct investment (FDI) affects the economic performance of host countries through direct as well as indirect channels. The indirect effect (also known as the spillover effect) of FDI arises from a number of sources including the linkages that are formed between domestic and foreign firms and increased competition in the domestic market that leads to better allocation of resources. The linkages between domestic and foreign firms can also facilitate technology and knowledge transfer. Recent studies (for example, Blomström and Kokko, 2003, Görg and Greenaway, 2004, Greenaway and Kneller, 2004, Kneller and Pisu, 2007, Wagner, 2007 and Sun, 2009) have suggested that linkages between domestic and foreign firms can also affect the export performance of domestic firms, which provides yet another explanation for increased competition for FDI among host country governments.1 While a large number of existing studies (for example Blomström, 1986, Blomström and Kokko, 2001, Liu and Wang, 2003, Görg and Hijzen, 2004, Wei and Liu, 2006, Duanmu and Fai, 2007, Beugelsdijk et al., 2008, Grima et al., 2008, Liu, 2008, Suyanto et al., 2009 and Barbosa and Eiriz, 2009; and Blalock & Simon, 2009) have considered the impact of FDI and FDI-linked spillovers on productivity and technology transfer, relatively few empirical studies have considered the impact of FDI-related industrial linkages on the export performance of host countries. Furthermore, drawing on the early theoretical work of Rodriguez-Clare (1996), it is clear that FDI can also affect the export activities of domestic firms in upstream and downstream industries through vertical linkages, yet the studies that have considered the impact of FDI-related industrial linkages on export performance (also known as export spillovers) have mainly focused on horizontal linkages.2 Some empirical studies (for example Kokko et al., 2001 and Alvarez and López, 2008) have found the impact of horizontal linkages on export performance to be positive and statistically significant which suggests that the presence of foreign firms promotes the export activities of domestic firms in the same sector. On the other hand, some studies (for example Aitken and Harrison, 1999, Djankov and Hoekman, 2000, Lutz et al., 2003 and Greenaway et al., 2004) have found the impact on export performance to be zero or negative. In other words, the empirical evidence is mixed, which is not surprising since the impact of FDI related industrial linkages on export performance depends on the characteristics of domestic firms, industries and indeed the host country. Some of these characteristics are categorised as the absorptive capacity which includes variables such as the size of a country's stock of human capital, the level of its financial market development and the technology gap between domestic and foreign firms. Based on the existing studies, it can be argued that the presence of foreign firms in a country does not always increase the probability of exporting or the export performance of domestic firms (for example, see Wagner, 2007 and Alvarez and López, 2008 and the references therein). Alvarez and López (2008) have argued that the presence of sunk-entry costs to export markets tends to diminish the overall size of the export spillover effect. In other words, the net effect on export performance will be positive only if the export spillover effects can more than compensate the sunk-entry cost. In a recent study Harris and Li (2009) have argued that absorptive capacity can lower the barriers to export market entry. The presence of mixed evidence warrants further research. The impact of FDI in any host country varies from industry to industry and hence results presented for one country cannot be applied to another. This paper focuses on Vietnam, a country that has attracted significant FDI since the late 1980s. At the end of the War in Vietnam, the country had two economic systems, a centrally planned system in the north and a market based system in the south. However, after national reunification, from 1975 to 1989, Vietnam remained a centrally planned economy. The introduction of the reform policy known as Doi Moi in 1986 marked the beginning of the opening up of the Vietnamese economy which resulted in a significant inflow of foreign investment. FDI inflows have played an important role in the rapid economic growth experienced by Vietnam. Vietnam is also an interesting case to consider because FDI in Vietnam is mainly concentrated in designated key economic regions while remote regions are unable to attract significant FDI. This factor influences the magnitude of horizontal as well as vertical linkages between domestic and foreign firms in different regions. Vietnam promulgated the Law of Foreign Investment in 1987. This law, which has been subsequently amended, is aimed at (a) increasing FDI inflows into Vietnam, (b) promoting technology transfer through FDI inflows and (c) enhancing export activities of domestic firms. Investigations of the effect of FDI on the Vietnamese economy to date are scarce, mainly due to data limitations. While some existing studies have considered the impact of FDI on productivity, economic growth and exports, none of the available studies has considered the impact of FDI-related industrial linkages on export spillovers in Vietnam.3 An analysis of the indirect effect of FDI on Vietnam's export performance allows one to assess the effectiveness of government policies. This paper examines the impact of horizontal and vertical linkages between domestic and foreign firms on (i) the decision of domestic firms to export and (b) the export share of domestic firms. The empirical analysis, which is conducted by means of Heckman's two-step estimator in selection models, reveals that the presence of foreign firms in Vietnam's manufacturing sector has a strong positive effect on domestic manufacturing firms’ decision to export and their export share, mainly through horizontal and forward linkages. We also consider the impact on the decision to export and export share after taking into account factors such as the differences in technology level of domestic firms (i.e., low versus medium and high technology firms), the ownership structure of domestic firms (i.e., private versus state ownership), whether or not foreign firms are export-oriented and the geographical proximity of domestic firms to foreign firms. The rest of this paper is organised as follows. Section 2 provides a theoretical perspective on FDI and FDI-generated linkages between domestic and foreign firms and the impact of these linkages on economic growth in host countries. Section 3 contains a review of some empirical studies that have considered the impact of transactional linkages between domestic and foreign firms on the export performance of domestic firms in host countries. Sections 4 and 5, respectively, contain data description and empirical analysis and Section 6 contains concluding remarks.

This paper contributes to the existing literature that deals with the impact of spillovers that arise from FDI-generated linkages between domestic and foreign firms on productivity and exports. FDI is regarded as a vehicle for the development of new technologies and technology transfer. Technology transfer can boost productivity and hence contribute to a larger increase in production. It has been suggested that the presence of foreign firms can also affect the decision of domestic firms to export as well as their export share. Most existing empirical studies have focused only on the impact of horizontal linkages between domestic and foreign firms on the export performance of domestic firms. Recent studies, such as Kneller and Pisu (2007), have suggested that the export performance of domestic firms can also be affected by vertical (forward and backward) linkages between domestic and foreign firms. This paper focuses on Vietnam, a country that has attracted significant FDI since the late 1980s, yet none of the existing studies has considered the impact of the presence of foreign firms on export performance of the Vietnamese firms. An analysis of export spillovers in Vietnam is also useful in that the lessons learned from Vietnam's experience can be of considerable value to other regional and transitional economies. In order to examine the impact of the horizontal and vertical linkages between domestic and foreign firms on export performance of domestic firms in Vietnam, this paper makes use of a Heckman type selection model. The model is estimated by means of Heckman's two-step estimator in selection models. The empirical analysis, which is based on firm level data from the Vietnamese manufacturing sector, reveals that the presence of foreign firms in Vietnam has a positive and significant effect on (i) the decision of domestic firms to export and (ii) the export share of domestic firms only through horizontal and forward linkages. In other words, export spillovers experienced by Vietnam's manufacturing firms can be mainly attributed to FDI-related horizontal and forward linkages. Our empirical analysis suggests that backward linkages have contributed to a significant reduction in exports of domestic firms. The existing theoretical literature suggests that the entry of foreign firms can increase domestic firms’ sunk-entry cost of export activities. Based on the empirical evidence provided in this paper, it can therefore be argued that the presence of foreign firms in Vietnam has contributed to an increase in the sunk-entry costs for a significant number of domestic firms and hence the overall effect on export activity through backward linkages with foreign firms is negative Further empirical analysis shows that horizontal linkages have resulted in a positive and significant export spillover effect to both low and medium/high technology domestic firms. However, forward linkages have resulted in a positive and significant export spillover effect only to low technology domestic firms. We also consider the export spillovers by splitting the sample into state-owned and private sector firms. The empirical analysis suggests that horizontal linkages with foreign firms lead to a significant positive export spillover effects to state-owned firms, whereas forward linkages with foreign firms lead to a significant positive spillover effect to private firms. In an attempt to improve the competitiveness of the state-owned enterprises (SOEs) and to attract additional FDI, Vietnam has undertaken a program to semi-privatise some SOEs that are not involved in the provision of public goods and services. However, so far only a very small number of small and medium-sized SOEs have been semi- privatised and these SOEs remain under government control. By 2005, more than 3000 SOEs were partially privatised (See Sjoholm, 2006). It has been suggested that export spillovers are stronger if foreign firms located in host countries are export-oriented (see Grima et al., 2008). Our empirical investigation suggests that horizontal linkages with export-oriented foreign firms do lead to significant positive export spillovers to Vietnamese firms. Finally, we consider the impact of the geographical location of foreign firms on export spillovers to Vietnamese manufacturing firms. The empirical analysis shows that an increase in the geographical concentration of foreign firms operating in a region increases (a) the probability that domestic firms located in the same region will be involved in exporting and (b) the export share of domestic firms located in the same region. We also found that the presence of export-oriented foreign firms in Vietnam significantly increases the probability that domestic firms in the same region will start exporting. The empirical results presented in this paper are consistent with the theory that linkages with foreign firms can have a significant impact on export activities of domestic firms. However, given the structure of the Vietnamese economy, its stock of human capital and government policies, not all linkages have a positive effect on the export activities of domestic firms. In other words, the empirical results presented in this paper are dependent upon the state of the economy and government policies. The empirical results presented in this paper suggest that, among other things, there is a need for improved export promotion programs in Vietnam. Firms that enter an industry face competition from existing firms. The level of competition is higher when foreign firms are present. New firms not only face stiff competition but they have to also incur significant sunk costs which can discourage export activities. For example, the cost of product promotion, the cost of establishing contacts with potential clients and, of course, the cost of product development. The presence of negative and statistically insignificant spillover effects appears to suggest that the sunk cost of exporting is too high and the quality of linkages with foreign firms is relatively poor. A decrease in sunk costs could make Vietnamese firms more competitive at the international level. By making use of Irish manufacturing sector data, Görg, Henry, and Strobl (2008) have shown that government support in the form of capital grants, training grants, R&D grants, rent subsidies, technology acquisition grants and loan guarantees can improve the performance of exporters. It can therefore be argued that there is a need for government support in Vietnam. Such support can serve to indirectly reduce the sunk cost associated with exporting. Government support may be lowered over time since new firms can learn from the experiences of the pioneering firms. In other words, government action can help to create, among other things, information externalities which can help to reduce the cost of finding a new market and/or to reduce the cost of export growth in existing markets. FDI has contributed to significant GDP growth in Vietnam but this is mainly due to increased employment and wages growth. An increase in net exports can also make significant contribution to the overall GDP growth. While FDI has contributed to export growth in Vietnam, Anwar and Nguyen (2011) have shown that FDI has also led to a significant increase in imports since a number of foreign firms located in Vietnam utilise imported inputs. Hence the positive effect of export growth is significantly offset by the negative effect of import growth. Government steps to reduce the sunk cost of exporting can help increase Vietnam's net exports and this is likely to enhance the overall rate of economic growth. Based on the results presented in this paper, it can also be argued that there is a need for increased spending on advanced education and training in Vietnam. A recent World Bank report indicates that the share of GDP spent on education in Vietnam increased from 3.5% in 1997 to 4.6% in 2004 and that the education sector is now given a higher priority. Spending on education as a percentage of total public spending has increased from 14% in 1997 to 18.6% in 2005. Despite an increase in spending on education, Vietnam's higher education sector still faces mismatches in fulfilling the needs for industrialisation (World Bank, 2008). A further increase in spending on education along with education sector reform is likely to increase the stock of human capital that will help not only to decrease the technology gap between domestic and foreign firms but also help Vietnamese firms increase their competitive advantage in the production of high technology products over time. An increase in the stock of human capital can also contribute to a decrease in reliance on imported inputs. From a business management point of view, Vietnamese firms need to budget for sunk cost associated with exporting since the long-term benefits will more than offset the short-term cost. Domestic firms need to be proactive in establishing and exploiting their contacts with MNCs. Such contacts can improve the quality of technological spillovers. In addition, government can also facilitate collaboration between domestic and foreign firms, thereby enhancing the quality of both forward and backward linkages. This is particularly important for private sector firms. An important implication for Vietnamese firms that are contemplating entry into the manufacturing sector is that they should consider their location decision very carefully; location in a region where foreign firms are concentrated can result in significant spillover effects. In addition, being export-oriented from the start can be very helpful. Due to the unavailability of data, this study is based only on cross sectional data. The limitations of studies based on cross sectional data are well known. It will be useful to repeat this exercise when panel data becomes available. This research can be further extended in several directions. For example, it may also be useful to examine the impact of FDI-generated spillovers on capital structure of domestic firms empirically. Entry of new foreign firms in a country can affect the R&D spending of domestic firms as well as the spending of existing foreign firms. The level of R&D spending affects among other things new product development and hence it will be interesting to examine the impact of spillovers on R&D spending empirically.