Trade costs and location of foreign firms in China

Trade Costs and Location of Foreign Firms in China Mary Amiti* International Monetary Fund and CEPR Beata Smarzynska Javorcik# World Bank and CEPR Abstract: This study examines the determinants of entry by foreign firms, using information on 515 Chinese industries at the provincial level during 1998-2001. The analysis, rooted in the new economic geography, focuses on market and supplier access within and outside the province of entry, as well as production and trade costs. The results indicate that market and supplier access are the most important factors affecting foreign entry. Access to markets and suppliers in the province of entry matters more than access to the rest of China, which is consistent with market fragmentation due to underdeveloped transport infrastructure and informal trade barriers. JEL Classifications: F1 F23. Keywords: foreign direct investment, trade costs, market access, supply access. World Bank Policy Research Working Paper 3564, April 2005 The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Policy Research Working Papers are available online at http://econ.worldbank.org. We would like to thank Caroline Freund, Mary Hallward-Driemeier, Will Martin, Stephen Redding, John Romalis, Tony Venables, Shang-Jin Wei, and participants at the Workshop on National Market Integration (World Bank and Development Research Council) in Bejing, September 2003, the Workshop on China's Economic Geography and Regional Development in Hong Kong, December 2003, and the World Bank/IMF seminars for helpful comments and suggestions. We are also grateful to the China National Bureau of Statistics (NBS) and the China Customs Office for supplying us with the data, and to Sourafel Girma for providing some additional data. *The International Monetary Fund, 700 19th Street, Washington DC 20431. Ph (202) 623 7767. Email: mamiti@imf.org. #The World Bank, 1818 H Street, N.W., MSN MC3-303, Washington, DC 20433. Phone: (202) 458-8485. E-mail: bjavorcik@worldbank.org.  1. Introduction Governments all over the world spend large sums of money to entice foreign direct investment (FDI), usually offering generous tax incentives. It is generally expected that foreign firms will generate positive externalities on domestic firms, particularly in developing countries. For example, Javorcik (2004) provides evidence consistent with the existence of positive inter- industry spillovers from foreign firms in Lithuania. However, the evidence on the success of tax incentives in attracting FDI is rather mixed (see Desai et al. 2004), which raises the question of what factors in fact influence where foreign firms locate. The theoretical literature on firm location emphasizes a tension between production costs and access to large final goods markets and input suppliers. Recent work by Krugman and Venables (1995), and Markusen and Venables (1998, 2000) shows that while the market size is an important consideration for firms, the larger the markets the higher the cost of immobile factors. The relative strength of these factors in determining location depends critically on trade costs. The view that both market size as well as access to intermediate inputs affect foreign investors' decisions is supported by anecdotal evidence. For instance, a manager of Salcomp, a Finish mobile company, justified the interest of the company in China by stating that "Our markets are in China, our components are there and wages are much lower..." (Financial Times, May 2004). Building on the predictions of the theoretical literature on economic geography, this study examines the relative importance of market access, supplier access, trade costs and factor costs on the entry of foreign firms into China. While FDI determinants have been 2  analyzed extensively (for example, see Caves 1982; and Markusen 1995), little attention has been paid to the new economic geography aspects of the investment decision. Notable exceptions are studies by Head and Mayer (2004) and Head and Ries (1996). The former study focuses on market access and shows that there exists a positive correlation between entry of Japanese firms into the European Union (EU) and market potential measures, which aggregate demand from multiple EU regions adjusted by distance. The latter study takes into account market and supplier access as determinants of foreign entry into China, but does not incorporate any spatial aspects. That is, their proxies for the availability of inputs are the total number of industrial enterprises and the total value of industrial output within the province of foreign entry in China.1 Our analysis extends the literature in several dimensions. First, we consider the impor- tance of both market and supplier access in determining foreign entry, taking into account spatial aspects. We allow for the possibility that firms purchase inputs not only from within their own province, but also from other provinces within China and from the rest of the world. Second, our measures of market and supplier access take into account the varying degrees of inter-industry linkages. For example, proximity to a steel plant is likely to be more valuable to a car producer than a textile manufacturer. Third, by incorporating all the key factors highlighted in the new economic geography literature, we are able to pro- vide an assessment of the relative importance of production costs and market size effects in attracting new entry. China is a particularly interesting country in which to analyze FDI flows. It was among 1Head and Ries (1996) assume that firms buy all their inputs locally and they do not distinguish in their analysis between various degrees of input availability in different industries. 3  the top FDI recipients in the world during the period under study, receiving US$165 billion of direct investment flows between 1998 and 2001 (World Investment Report 2002, Annex Table B3). Rising inequality between Chinese provinces has been of growing concern to the Chinese government which has introduced a number of policies aimed at mitigating this development.2 With over 90 percent of foreign investment being directed to the coastal regions, the influx of FDI has widened regional disparities between coastal and central regions within China. By providing an assessment of the importance of market access and supplier access relative to production costs, this study provides some guidance on the kinds of policy instruments that would be most successful in attracting FDI to disadvantaged regions. In addition to an intrinsic interest in determinants of FDI flows into China, our study sheds some light on the economic impact of inter-provincial barriers to trade. There exists ev- idence suggesting that in an effort to protect industries from competition, local governments in China are erecting barriers to entry of goods from other provinces. The presence of such barriers was reported by Kumar (1994) and Young (2000) and is consistent with anecdotal evidence. For instance, managers of Chinese firms confirmed that they have indeed expe- rienced some difficulties in accessing markets in other provinces. A manager of a medical manufacturing plant reported that the shipments to other provinces are occasionally stopped by local rail officials for 2 to 4 weeks for no apparent reason. The administrative units of industry and commerce department were reportedly obstructing access to markets through 2Strong economic growth experienced by China during this time did not benefit all provinces equally. For instance, while in 1999 the GDP of coastal provinces increased by 7.2%, central and western provinces experi- enced a growth rate of only 3.8 and 4.7%, respectively. In the same year, coastal provinces accounted for 60% of Chinese GDP and almost three-quarters of national output of manufactured goods. See Amiti and Wen (2001) for a discussion on regional inequality in earlier years and the spatial distribution of manufacturing industries in 1995. 4  audits or local registration requirements.3 Unfortunately, it is not possible to directly mea- sure such barriers. As it is illegal to impose trade restrictions, the measures adopted to protect local industries from competition are usually more subtle than a direct border tax. Thus the only way to assess the significance of such barriers is indirectly, as is the case in our study.4 Our analysis is based on a comprehensive data set provided by the China National Bureau of Statistics (NBS) covering nearly all manufacturing industries, at a highly disaggregated level (515 industries) in 29 Chinese provinces, during the period 1998-2001.5 Using the information on the value of output by industry and province, the national input/output table and inter-provincial distances we construct measures of market access and supplier access. We also create industry-specific measures of tariff rates on imported inputs. We then relate these measures to the change in the number of foreign firms in each province and industry. We also control for a variety of provincial characteristics. Proxies for trade costs at the provincial level include transport infrastructure and openness to international trade. Production costs are proxied by provincial wages and electricity prices. We consider separately market access and supplier access within and outside the province of foreign 3Interviews with firms and government officials were conducted by Amiti in five different provinces in October 2001. 4A number of researchers have tried to estimate the size of these provincial trade barriers using indirect measures (see Poncet 2003, Young 2000, Naughton 1999, Huang and Wei 2003, and Bai et al. 2004), but none of them has considered the consequences of such barriers. None of the studies has ruled out the existence of provincial border barriers and some have found evidence that such barriers have increased over time. 5Other studies on the determinants of FDI in China rely either on information on provincial FDI stocks (Cheng and Kwan, 2000), or on the Almanac of China's Foreign Economic Relations and Trade which lists entry of individual firms (Head and Ries 1996, Dean, Lovely and Wang, 2002). The latter data set is, however, limited in coverage as it includes only about 10 percent of new foreign firms, focuses exclusively on joint ventures and stopped being published in 1996. It is also unclear what criteria were used to select a particular sub-sample of all foreign investors for publication. 5  entry. A lower magnitude of the coefficients pertaining to trade outside the province of entry relative to trade within the province would suggest that the internal trade barriers may be restricting access of foreign investors to suppliers and customers in other regions. The results indicate that market access and supplier access are the most important factors affecting FDI inflows. Doubling either market access or supplier access is associated with a 40% increase in the entry of foreign firms. The presence of customers and suppliers in the province of entry matters much more than market and supplier access to the rest of China, which is consistent with the presence of inter-provincial barriers to trade. Further, our analysis suggests that provinces which are more open to foreign trade attract more foreign firms. Similarly, the availability of infrastructure is positively correlated with foreign entry. Although production costs also play a significant role in determining the location of foreign investment, the magnitude of these effects is around half that of the market and supplier access effects. A doubling of wages or electricity prices reduces entry of foreign firms by 17% and 22%, respectively. Thus, our results suggest that local governments may do well by reducing inter-provincial barriers, and hence increasing the extent of market and supplier access in surrounding provinces, in order to attract foreign investment. The rest of the paper is organized as follows. Section 2 develops the formal model. Section 3 provides background information on China and details of the data sources. Section 4 presents the results, and section 5 concludes. 6  2. Theory We derive our estimating equation from a new economic geography model, based on Krug- man and Venables (1995) and Amiti (2005)6. Firms are assumed to compete in a monop- olistically competitive environment, with each firm producing a differentiated variety. All varieties of final goods enter symmetrically into the consumer's utility function and all va- rieties of intermediate inputs enter symmetrically in the firm's cost function. Profits of a single representative firm in industry i in province p are given by ip = pipxip - wprp Pp u