外文翻译

原文

Are there positive spillovers from direct foreign investment? Material Source:Journal of Development Economics 42(1993) 51-74

Author: Mona Haddad , Ann Harrison

Many developing countries now actively solicit foreign investment, offering income tax holidays，import duty exemptions?and subsidies to foreign firms. One reason for subsidizing these firms is the positive spillover from transferring technology to domestic firms. This paper employs a unique firm-level dataset to test for such spillovers in the Moroccan manufacturing sector. We find evidence that the dispersion of productivity is smaller in sectors with more foreign firms. However, we reject the hypothesis that foreign presence accelerated productivity growth in domestic firms during the second half of the 1980s. Using detailed information on quotas and tariffs, we also reject the possibility of a downward bias in estimating technology spillovers because foreign investors may be attracted to protected domestic markets.

1. Introductiou

The disappearance of non-equity sources of foreign capital in the 1980s has created a renewed interest in direct foreign investment (DFI). Despite the controversies surrounding the benefits and costs of DFI, a number of developing country governments have now changed their policies from restricting towards promoting foreign Investment. Some countries have actually tilted the balance towards foreign firms by offering special incentives: in Mexico, the maquiladora firms pay no income taxes; in much of the Caribbean, foreign firms receive income tax holidays, import duty exemptions, and subsidies for infrastructure. Are these subsidies justified? One benefit often cited in the literature on the gains from DFI, apart from the capital often inflows and additional employment, is the new technology brought in by foreign firms, it could justify some type of subsidy. This may be the rationable for goverment policies in economices as diverse as Taiwan

and Bulgaria, which target special treatment for foreign firms in high technology sectors.

Technology transfer occurs through many different avenues -new technology is embodied in imported inputs and capital goods, sold directly through licensing agreements, or transmitted to exporters who learn about new techniques from their foreign buyers. In other cases, learning by doing among domestic firms, combined with investments in formal education and on-the-job training, is critical. No individual source of technology is likely to be the best; country experience suggests that the most effective diffusion of new ideas and processes probably involves a combination of all these factors. Foreign investment plays an unusual role in several respects, however.First, new technology may not be commercially available and innovating firms may refuse to sell their technology via licensing agreements. Mansfield and Romeo (1980), for example, found that the technology transferred via multinationals was much newer than the technology sold through licensing agreements. Second, foreign investment may provide the competition necessary to stimulate technology diffusion, particularly if local firms are protected from import competition. Third, foreign investors may provide a form of worker training which cannot be replicated in domestic firms or purchased from abroad. The theoretical literature on foreign investment suggests that foreign investors possess some sort of intangible asset which cannot easily be sold - such as managerial skills. Technology diffusion may occur through labor turnover as domestic employees move from foreign to domestic firms.

Existing case studies of multinational behavior do suggest that foreign investment could be an important source of spillovers. Rhee and Belot(1990) present a number of detailed cases where foreign investors have acted as export catalysts, in some cases fueling domestic export industry where there were no domestic exports at all. Mansfield and Romeo, however, found that only a small share of the 15 multinationals in their survey believed that foreign investment hastened access to process technology for host country competitors. Mansfield and Romeo suggested that more important gains from foreign investment were likely to be through cost savings to downstream users of new products or technology transferred to upstream supplier.

A number of empirical studies have attempted to directly measure the so-called spillovers from foreign investment. In an early study, Cave(1974) tested for the impact of foreign presense on value-added per worker in Australian

domestically-owned manufacturing sectors. Caves found that the diparity between(higher) foreign and domestic value-added disappears as the foreign share of sectoral labor rises, which is consistent with positive spillovers from foreign presence.

Globerman(1979) replicated Cave’s findings(1974) using sector-level, cross-section data for Canadian manufacturing industries in 1972. Globerman, however, was able to control explicity for capital intensity in his estimation of value-added per worker. The results indicate only a weak effect-none of the proxies for foreign presence in the sector are significant at the 5 percent level.

Most of the empirical work on spillovers from foreign investment in developing countries has focused on Mexico, which gathers manufacturing data by ownership type. Blomstrom and Persson (1983) reproduce Globerman’s study using 1970 census data for 215 Mexican manufacturing industries. Controlling for capital intensity, scale effects, and worker quality, Blomstrom and Persson find that labor productivity is significantly higher in sectors where foreign firms employ a higher share of the labor force. Blomstrom (1986) and Wolff(1989) find faster productivity growth and faster convergence of productivity levels in sectors which higher levels of foreign ownership.

This paper, which examines the impact of foreign investment on firms in Morocco’s manufacturing sector from 1985 through 1989, contributes to this existing literature in two respects. This is the first-sepecific attributes such as size. The panel nature of the data (which combine cross-section and time series) allows us to go beyond cross-section analysis comparing partial productivity measures (such as labor productivity) across different firms. Our results suggest that foreign firms exhibit higher levels of total factor productivity, but their rate of productivity growth is lower than that for domestic firms. At first glance, this would appear to support the catch-up hypothesis –domestic firms, at lower initial levels of productivity, are able to increase efficiency at a faster rate. However, our tests on the presence of any spillovers from foreign presence show that although domestic firms exhibit higher levels of productivity in sectors with a larger foreign presence, they do not exhibit higer productivity growth in those sectors.

Second, we are able to use detailed information on the level of quota and tariff protection to test whether the lack of any spilbvers stems from a tendency of foreign firms to move towards protected sectors. We do not find evidence of such positive spillovers in either the protected or unprotected sectors.

Section 2 discusses the trade and foreign investment policies in Morocco before and during regulatory reform in the 1980s. Section 3 examines the relative performance of domestic firms and foreign firms. Section 4 measures the spillovers from foreign presence on the level, growth rate, and dispersion of productivity for domestically-owned firms. This section also extends the analysis to examine whether technology spillovers are related to the degree of import protection. Section 5 concludes with a discussion of the implications of these findings for policies towards multinationals.

2. The regulatory framework: Foreign investment and trade policy 2.1. Foreign investment polocies

The first major action against foreign investment in Morocco took place in 1973, when the government passed the Moroccanization Decree, which restricted foreign ownership of cer6n industrial, commercial, and service activities to no more than 49 perr;nt. The main purpose of this policy was political rather than economic -to reduce the dominant role of French firmsin the Moroccan economy. Activities falling under the Moroccanization law ilycluded textiles, clothing, footwear, leather products, travel goods, toys, fish canriing and preserving, fertilizers, edible oils, vegetable fibers, and processed fruits acd vegetables. The negative impact of this law on foreign investment is evident from the fact that even enterprises not subject to the law voluntarily handed over their capital share to their Moroccan partners.

A major reform uf the investment code was undertaken in 1983. It allowed full foreign ownership of Moroccan companies in certain sectors (especially manufacturing), eased restrictions on the repatriation of capital and divi-dends, and introduced fiscal and other incentives for direct foreign invest-ment. The code guaranteed (i) foreign investment against the risks of nationalization and expropriaaion; (ii) unlimited transfer of dividends and profits to foreign investors; and (iii) the repatriation of foreign investors’ capital and related caoital gains. By 1985, the Moroccan majority-owners restrion no longer applied to any segment in the industrial sector, which meant that foreign firms could have an equity participation fo more than 49 percent. The investment code was further liberalized in 1988, administrative procedures governing the approval of direct foreign investment were simlifide, and rules similar to those granted to nonresident foreigners were extended to nonredident Moroccans. 2.2. Trade polocies

Following independence in 1956, Morocco’s economic development strategy was primarily based on import-substituting industrialization and agricultural self-sufficiency in a ighly protected domestic market. For more than two deeades, trade and in ustrial policies in Morocco were based on high tariffs and on quantitative restrictions in imports. Furthermore, during the 1970s, the Moroccan government expanded growth through high levels of pub!ic spending, financed through foreign borrowing and rising receipts from phosphate exports. This culminated in a major payment crisis in 1983. As a result, the government introduced outward-oriented structural adjust-ment measures designed to eliminate the bias against export activites, liberalized the import regime, and enhanced the allocative role of the financial sector.

The trade reform introduced in 1983 called for the eventual elimination of the Special import Tariff (SIT), a uniform tariff levied on the c.i.f. value of imports, the lowering of the maximum customs duty from 400 percent in 1983 to 60 percent in 1984 and 45 percent in 1985, and a reduction in quantitative restrictions. Changes in the industrial code were also undertaken to promote exports. In January 1988, the SIT and the customs stamp duty were merged into what was called a fiscal levy on imports, set at 12.5 percent. Contrary to the declining maximum tariff trend observed since 1983, the fiscal levy actually exceeded the sum of the two abolished taxes. This was intended to generate additionnal fiscal revenue rather than to provide prtection.