Moreover, the expansion of production overseas by the transaction partners does not necessarily have to. have a negative effect on local suppliers. Thus, how the expansion of overseas production affects non-internationalized domestic suppliers is a purely empirical question. However, the effects of expanding overseas activities on domestic non-international supplier firms are quite complex and depend on the type and characteristics of the MNEs and supplier firms involved.
For example, in some cases the expansion of overseas production by Japanese MNEs has been associated with an increase in exports from Japan (see e.g. Nishitateno 2013 for the case of intermediate goods and Belderbos et al. The effect of the expansion of overseas production on domestic economic activities has long been a hotly debated issue in many developed countries.In short, the effect of an expansion of overseas activities on domestic activities is not entirely straightforward and depends on exactly what one focuses on.
But it is also possible that multinational companies' expansion of overseas production increases the purchase of parts and components from their domestic suppliers. Several of the above-mentioned studies show that multinational companies' overseas production and home country exports are complementary.
Domestic and Overseas Operations of Japanese Manufacturing Firms 1 Data
Overview of the Firm-Level Dataset
More specifically, column (6) in Table 2 shows the foreign employment ratio of the five largest customers of non-MNEs. Similarly, columns (10) and (12) show the employment ratios of customers abroad when considering only the three largest customers. However, while some companies provide information on all five top customers, others only provide the name of the top customer or, for example, the three largest customers.
Therefore, we use the extent of overseas activities of customer firms for each supplier, which we measure as the average of the overseas employment ratios of the top five or top three customers. It should be noted that when we focus, for example, on the top five customers, but a firm has only two customers, the average is calculated using the information on these two customers. Similarly, when focusing on the top three customers, if the firm has only one customer, we use information about that customer.
In the following empirical analysis, we mainly use the average value for the top five customers; however, we also use the average value for the top three customers to check the robustness of our estimation results. In terms of the number of firms, the latter group is much larger than the former.
Empirical Analysis 1 Empirical Specification
Empirical Results
Looking at the results in columns (3) and (4), it turns out that the growth coefficients of the employment rate of the five largest foreign buyers are insignificant. We begin by estimating equation (5) using the foreign employment ratio of the top three rather than the top five buyers. This increase in the coefficient means that the expansion of the three largest buyers abroad has a greater positive impact on the growth of domestic employment of suppliers.
First, the coefficients for top buyers' overseas expansion are not statistically significant in the second-stage estimation when focusing on the top five buyers, as shown in columns (1) and (2). Second, the coefficients are positive and significant when we use the overseas employment ratio of the top. Another possible criticism of the analysis here is that the overseas employment ratio does not fully capture the extent of the expansion of buyers' overseas and domestic operations.
To account for such possible criticisms, we first estimate comparisons that instead of the overseas employment ratio include the absolute number of workers employed by the top five buyers' foreign affiliates and that of domestic workers employed by the top five buyers as separate variables. When we include both the top five buyers' domestic employment and overseas employment, the estimated coefficient on the overseas employment variable is not significant, while that on the domestic employment variable is positive and significant. Second, to control for changes in the size of domestic operations, we split our sample into two groups of firms depending on the growth rate of the top five buyers' domestic employment.
We first calculate the average number of domestic workers employed by the top five buyers for each firm and year and then calculate the growth rate of the top five buyers. When we focus on the top five buyers, the coefficient on the overseas employment ratio of buyers is insignificant but positive in both cases, implying that overseas expansion by customers does not have a negative effect on non-SMEs. Furthermore, the results in Appendix Table 4 show no significant difference in the estimated coefficients on the overseas employment ratio of the top five buyers for each group of firms.
Specifically, the positive effect of the foreign expansion of the three largest buyers is much larger and statistically significant for domestic suppliers whose domestic employment grew more than the average growth rate. We add a dummy variable for exporters and its interaction term with the foreign employment ratio of the top five buyers to the basic specification shown in Table 4. Finally, if we include firm-level total factor productivity (TFP) and its interaction term with the top five If the employment ratio of buyers abroad is compared to the base specification in Table 4, the estimated coefficients are not statistically significant for both the TFP variable and its interaction term, indicating that TFP does not have a significant impact on employment.14.
Conclusion and Policy Implications
As part of the robustness tests of our results, we estimated the labor demand function using a Heckman selection model to account for the possible bias resulting from the death of supplier firms. The death of supplier firms, as well as changes in transactional relationships, are issues that deserve further investigation in the future in order to gain a better understanding of the heterogeneous impact of downstream firms' overseas expansion on supplier firms. The third, closely related reason why our results appear to be at odds with the observed decline in manufacturing employment is that our measure of overseas expansion may not fully capture the dynamic changes in the overseas and domestic production of the main buyers.
For example, our current measure does not sufficiently take into account the frequency of changes in customers and the strength of transaction relationships (ie, duration of transaction relationships and/or transaction volume). Zou (2012) "Business Groups, Foreign Direct Investment and Capital Goods Trade: The Import Behavior of Japanese Subsidiaries", Journal of the Japanese and International Economies. Schott (2006) "Survival of the fittest: Exposure to low-wage countries and the (uneven) growth of US and Swedish direct investment and exports", in Robert E. Baldwin, ed., Trade Policy Issues and Analysis Empirical, Chapter 9, p. 257-302, National Bureau of Economic Research.
Piscitello (2008) "The impact of external investment on parent company employment and skill mix: Evidence from the Italian case". Weiss (1984) "Foreign production and exports by individual firms". 2013) "Global Production Sharing and the FDI-Trade Nexus: New Evidence from the Japanese Automobile Industry," Journal of the Japanese and International Economies 27: 64-80. 1985) "Sweden," in John Dunning, ed., Multinational Enterprises, Economic Structure, and International Competitiveness, p. 2012) "The Effects of FDI on Domestic Employment and Workforce Composition," RIETI Discussion Paper Series 12-E-069, The Research Institute for Economics, Trade and Industry.
2003) "The Impact of Import Competition on Japanese Manufacturing Employment", Journal of the Japanese and International Economies. Tanaka (2008) “The Internationalization of Japanese Firms: New Findings Based on Firm-Level Data”, RIETI Discussion Paper 08-E-036, Research Institute of Economics, Trade and Industry. We therefore supplement the BSBSA data with information on smaller firms from the Census of Manufactures (also compiled by METI), which covers firms with 4 or more employees.
Manufacturing companies are defined here as companies with production departments or facilities in Japan based on the information reported in the BBSSA. All employees All employees All employees All employees. top five buyers top three buyers top five buyers. Sources: Companies with 50 or more employees and production companies abroad: Authors. calculation based on the micro data underlying the BBSSA and the BSOBA.
Companies with 49 or fewer employees: data collected by the Ministry of Economy, Trade and Industry based on the Census of Manufactures. Number of observations of non-MNEs by industry: Non-MNE firms with production units or branches in Japan (2007).
