توسعه در خارج از کشور و کار در منزل: شواهدی از شرکت های چند ملیتی ژاپنی

This paper examines the so-called ‘exporting job’ hypothesis – that expansion of overseas operations of manufacturing multinational enterprises (MNEs) reduces home employment – using a newly constructed matched parent-affiliate panel dataset of Japanese MNEs over the period 1991–2002. The results do not support the widely held view that overseas operations of MNEs lower home employment. On the contrary, there is some evidence that expansion of overseas operations may have helped to maintain the level of home employment.

The debate over the possible adverse effects of overseas production by multinational enterprises (MNEs) on home employment (‘exporting jobs’) first emerged in the US in the late-1960s (Kravis and Lipsey, 1988). It has gained increased attention in policy circles of industrial countries in recent year with the growing importance of international fragmentation of production (Lipsey, 1995 and Harrison and McMillan, 2006). It had also become the subject of heated policy debate in Japan under the label of ‘manufacturing hollowing-out’ following the spread of production bases of Japanese MNEs to low cost countries in East Asia from the mid-1980s. Given its policy importance, there has been a proliferation of empirical investigations of the exporting jobs hypothesis, using matched parent-affiliates datasets for MNEs of various national: Brainard and Riker (1997), Hanson et al. (2003) and Desai et al. (2009), for the US MNEs, Braconier and Ekholm (2000), and Fors and Kokko (2001) for Swedish MNEs, Navaretti and Casellani (2004) for Italian MNEs, Becker et al. (2005) and Marlin (2006) for German MNEs and Konings and Murphy (2006) for European MNEs. However, to our best knowledge, so far no similar study has been undertaken for Japanese MNEs. This paper fills that gap. The available Japanese evidence on this subject is mainly drawn from the readily available industry-level FDI data (Fukao, 1995, Fukao and Amano, 1998 and Fukao and Yuan, 2001). Since the FDI decision is made at the firm-level rather than the industry-level, the firm-level dataset allows us to undertake the first analysis for properly assessing the exporting jobs hypothesis for Japanese MNEs. Head and Ries (2002) uses the firm-level data compiled from the Toyo Keizai which only includes the listed firms in the Japanese stock market in Japan. Instead, the dataset compiled for our paper cover divers scale of Japanese parent firms. This paper uses a newly constructed firm-level matched parent-affiliates panel dataset over the period 1991–2002. The data is compiled from the unpublished returns to two annual enterprises surveys of Japanese MNEs parent firms and their foreign affiliates, the Basic Survey of Business Structure and Activity and the Basic Survey of Overseas Japanese Business Activity, collected by Japan Ministry of the Economy, Trade and Industry (METI) over the period 1991–2002. 1 The METI database is well known for their high quality and reliability ( Nishimura et al., 2005, Kimura and Kiyota, 2006 and Hijzen et al., 2010). The analysis of this paper is based on estimating labour demand of parent firms controlling for the firm attributes and specific regional characteristics of overseas operation of MNEs. The findings of this paper do not support the ‘exporting job hypothesis’. Instead, there is some evidence of complementary relationship between employment in foreign affiliates of MNEs and their home employment; a 10 percent increase of foreign affiliate employment leads to a 0.2 percent increase in home employment. This finding is robust to alternative model specifications that appropriately allow for locations-specific characteristics of foreign affiliates. Thus, this paper alerts to the possibility that, as the globalisation process continues, policy initiative driven by the public fear of exporting jobs could have perverse effect of constraining MNE's ability to avert domestic employment contraction by outsourcing some segments of the production process. The next section summarises the theoretical discussion and the existing empirical evidence on the impacts of overseas operation on MNEs’ home economic activity. Sections 4, 5, 5.1 and 6 introduce the econometric specification, the description of the data and the estimation method, followed by interpretation on the results in Section 7. Section 8 concludes.

This paper has examined the hypothesis that expansion of overseas operations of Japanese manufacturing MNEs reduces their home employment. A standard labour demand equation of parent firms was estimated based on a newly constructed panel dataset that has information on both home and foreign affiliates’ operations. In addition, the study took account of the geographic locations of foreign affiliates to control for specific regional characteristics of MNEs. Despite widely held concerns about the adverse effects of outward FDI on the home employment, the results do not provide any evidence that outward FDI reduces employment at home. On the contrary, the findings provide some weak evidence that during the study period overseas operations may have actually helped to maintain the level of home employment in Japanese manufacturing. Overall, the evidence suggests that expanded overseas operations by MNEs not only help firms to enhance their competitiveness and profitability but may also have a positive impact on home MNE employment by generating higher demand for more technology and skill intensive activities where the Japanese home base is likely to have a comparative cost advantage. This finding challenges the popular perception of MNEs ‘exporting jobs’ when they expand overseas operations. Indeed, when the positive general equilibrium effects of profit remittances on overall home employment is also taken into account, there appears to be no case for any government action to restrain overseas expansion of Japanese MNEs. Eventually this process will help upgrade the skill intensity of the home production. However, there will be a transitional period during which there is likely to be significant adjustment pressures on the MNE workforce at home as MNE home production is restructured to focus on production of more technology and capital intensive production process. It is important that research is undertaken to investigate the firm-level effects of this process to identify the adjustment problems faced by the existing MNE workforce and to address related assistance and labour retraining requirements.