Outward Foreign Direct Investment, Interindustry Networks, and U.S. Trade Politics

Abstract

This article investigates how outward foreign direct investment by U.S. multinational corporations influences industry lobbying for trade protection in the United States, focusing on interindustry structure of goods sales networks between upstream and downstream sectors and also on the multinationals’ input procurement patterns. If foreign affiliates of U.S. multinationals switch input sources from U.S. to host-country suppliers, U.S. suppliers should receive a negative demand shock, ceteris paribus. An empirical test finds that those U.S. upstream sectors that are highly dependent upon U.S. multinationals for goods sales tend to lobby more as the multinationals’ overseas production and sales increase.

• foreign direct investment
• multinational corporations
• trade politics
• industry lobbying

Notes

¹ According to the Bureau of Economic Analysis, “Input–Output (I-O) tables show the commodity inputs that are used by each industry to produce its output, the commodities produced by each industry, and the use of commodities by final consumers” (Horowitz & Planting 2006, p. 2).

² Previous studies of trade politics show that a sector model generally performs better than an equivalent factor model (e.g., Magee et al. 1989; Frieden 1988; Fordham & McKeown 2003).

³ Various sources determine the degree of capital mobility across borders. They include interest rates, labor costs, strength of the home economy, market size of the host economy, government regulations and policies, distance and culture, firm-specific advantage factors, and others (Huang 1997).

⁴ Related to factors behind outward FDI, researchers found that host-country market size and home-country policies on capital liberalization and trade openness affect outward FDI (Goh & Wong 2011). Also, capital-intensive multinationals with a higher level of productivity invest in foreign markets where the firms have been exporting (Damijan et al. 2007). This is an example of horizontal FDI.

⁵ In a case of intrafirm trade by vertically integrated multinationals, the firms may not have been heavily reliant upon home-country upstream suppliers for procurement of intermediate input goods, and therefore the suppliers will not receive any significant demand shock from the multinationals’ outward FDI.

⁶ Domestic content requirements have usually been imposed by developing countries to encourage industrialization. Yet Japanese automobile companies’ practice of using Japanese parts when they produce cars overseas brought the issue to Europe and the United States (Krugman & Obstfeld 2003).

⁷ Because portfolio investment usually takes place in a secondary securities market in which most transactions are not initial public offerings, I assume that these transactions do not change the quantity or quality of the goods sold or purchased by the relevant firms.

⁸ Data for other periods are available, but the cost of mapping it into SIC categories and coding it is very high. The industry-level campaign contribution data were compiled by Fordham and McKeown (2003).

⁹ The industries covered include: food and kindred products (SIC code 20), textile mill products (22), apparel (23), lumber and wood products (24), furniture and fixtures (25), paper (26), printing and publishing (27), chemicals (28), petroleum and coal (29), rubber and plastics (30), leather (31), stone, clay, and glass products (32), primary metals (33), fabricated metals (34), industrial machinery (35), electrical machinery (36), transportation equipment (37), and instruments (38).

¹⁰ For the primary sectors (agriculture, forestry, fishing), Input–Output Benchmark data cannot be converted to SIC two-digit codes. According to the BEA, sales to the federal government and industry concentration ratios are not available for the primary sectors. The service and construction sectors are not included, as they were not tradable in the 1980s.

¹¹ A separate article by the author measures industry lobbying for trade protection by petitions to U.S. International Trade Commission (ITC). The results are not reported in this article.

¹² The list of protectionist bills used to create the dependent variable is provided in Appendix I.

¹³ The BEA collects data on FDI by means of mandatory surveys of the U.S. affiliates of foreign firms and foreign affiliates of U.S. firms. Assets are measured at gross book value of property, plant, equipment, and so on. See Quijano (1990).

¹⁴ Background information for the I-O data is provided in Appendix II.

¹⁵ These measures implicitly ignore the elasticity of demand and supply, an assumption generally consistent with input–output analysis.

¹⁶ Appendix III provides data sources of variables.

¹⁷ CLARIFY software by Tomz et al. (2003) is used for calculating the marginal effects.

¹⁸ The variable’s positive impact on industry lobbying is not empirically supported in Table 5 though.

¹⁹ The original House vote record of the bill (H.R.4328) for passage was 424–0 (yeas and nays), and the vote to override a presidential veto of the bill was 275–152 (yeas and nays). The latter voting record was chosen to measure protectionist House of Representatives because it is the more conservative measure between the two voting records.

²⁰ The 1977 version is employed to interpolate data for 1981. Likewise, the 1992 version is used to calculate data from 1988 to 1990.