New Perspectives on Foreign Direct Investment

Transcription

1 New Perspectives on Foreign Direct Investment Elhanan Helpman Harvard University January, 2008 (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

2 Background International trade and foreign direct investment have been among the fastest growing economic activities around the world. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

3 Background (continued) Trade and FDI are concentrated in the industrial countries: (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

4 Background (continued) (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

5 Background (continued) Trends in trade and FDI are related, because multinational corporations account for 40% of world trade. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

6 Background (continued) Trends in trade and FDI are related, because multinational corporations account for 40% of world trade. A remarkable feature of this growth has been an unprecedented expansion of trade and FDI in services. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

7 Background (continued) Trends in trade and FDI are related, because multinational corporations account for 40% of world trade. A remarkable feature of this growth has been an unprecedented expansion of trade and FDI in services. Moreover, the nature of trade and FDI have changed dramatically: there is growing trade in intermediate inputs, including services, and the growth in input trade takes place within and outside the boundaries of the rm. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

8 Multinational Corporations We care about multinational rms, because they play a major role in the global economy. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

9 Multinational Corporations We care about multinational rms, because they play a major role in the global economy. The extent of their activity is comparable in size to the U.S. economy. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

10 Multinational Corporations We care about multinational rms, because they play a major role in the global economy. The extent of their activity is comparable in size to the U.S. economy. One-third of the volume of world trade is intra rm. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

11 Multinational Corporations We care about multinational rms, because they play a major role in the global economy. The extent of their activity is comparable in size to the U.S. economy. One-third of the volume of world trade is intra rm. A third of the volume of world trade is accounted for by transactions in which multinational rms are in one of the two sides of the exchange (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

12 Types Three types of FDI: (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

13 Types Three types of FDI: Horizontal: Exporting vs. replication of the production process in a foreign market. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

14 Types Three types of FDI: Horizontal: Exporting vs. replication of the production process in a foreign market. Vertical: Fragmentation of production in the presence of factor price di erences across countries. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

15 Types Three types of FDI: Horizontal: Exporting vs. replication of the production process in a foreign market. Vertical: Fragmentation of production in the presence of factor price di erences across countries. Complex: Assembly and components production can generate interdependence between horizontal and vertical FDI, as well as third market e ects. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

16 Types Three types of FDI: Horizontal: Exporting vs. replication of the production process in a foreign market. Vertical: Fragmentation of production in the presence of factor price di erences across countries. Complex: Assembly and components production can generate interdependence between horizontal and vertical FDI, as well as third market e ects. MNEs decide on location and the extent of control: internalization versus outsourcing. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

17 Horizontal FDI: Proximity-Concentration Tradeo Emphasizes the tradeo between: (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

18 Horizontal FDI: Proximity-Concentration Tradeo Emphasizes the tradeo between: Concentration; saves plant-level xed costs. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

19 Horizontal FDI: Proximity-Concentration Tradeo Emphasizes the tradeo between: Concentration; saves plant-level xed costs. Proximity; saves trading costs. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

20 Horizontal FDI: Proximity-Concentration Tradeo Emphasizes the tradeo between: Concentration; saves plant-level xed costs. Proximity; saves trading costs. Moreover, in the data there is a lot of within industry heterogeneity, with multinationals being larger and more productive than all other rms. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

21 Sorting into Exporting and FDI profits domestic FDI export 0 productivity (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

22 (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

23 Predictions The ratio of exports to FDI sales is: (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

24 Predictions The ratio of exports to FDI sales is: decreasing in transport costs; (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

25 Predictions The ratio of exports to FDI sales is: decreasing in transport costs; increasing in plant-level economies of scale; (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

26 Predictions The ratio of exports to FDI sales is: decreasing in transport costs; increasing in plant-level economies of scale; decreasing in productivity dispersion. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

27 Results Dispersion is as important as trade costs and plant scale economies (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

28 Contractual Approach Early views of vertical FDI were based on factor price di erences across countries. Together with the horizontal models they generated a theory in which the emergence of MNEs is determined by some combination of: (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

29 Contractual Approach Early views of vertical FDI were based on factor price di erences across countries. Together with the horizontal models they generated a theory in which the emergence of MNEs is determined by some combination of: lactational advantages (distance, factor prices); (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

30 Contractual Approach Early views of vertical FDI were based on factor price di erences across countries. Together with the horizontal models they generated a theory in which the emergence of MNEs is determined by some combination of: lactational advantages (distance, factor prices); technological factors ( rm vs. plant-level economies of scale). (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

31 Contractual Approach Early views of vertical FDI were based on factor price di erences across countries. Together with the horizontal models they generated a theory in which the emergence of MNEs is determined by some combination of: lactational advantages (distance, factor prices); technological factors ( rm vs. plant-level economies of scale). These theories enhance our understanding of trade and FDI ows, but they share a common failure to account for internalization. Why is fragmentation occurring within rm boundaries? Or across rm boundaries? (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

32 Contractual Approach Early views of vertical FDI were based on factor price di erences across countries. Together with the horizontal models they generated a theory in which the emergence of MNEs is determined by some combination of: lactational advantages (distance, factor prices); technological factors ( rm vs. plant-level economies of scale). These theories enhance our understanding of trade and FDI ows, but they share a common failure to account for internalization. Why is fragmentation occurring within rm boundaries? Or across rm boundaries? More recent theories, based on contractual frictions, yield new insights. They predict: (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

33 Contractual Approach Early views of vertical FDI were based on factor price di erences across countries. Together with the horizontal models they generated a theory in which the emergence of MNEs is determined by some combination of: lactational advantages (distance, factor prices); technological factors ( rm vs. plant-level economies of scale). These theories enhance our understanding of trade and FDI ows, but they share a common failure to account for internalization. Why is fragmentation occurring within rm boundaries? Or across rm boundaries? More recent theories, based on contractual frictions, yield new insights. They predict: Larger shares of intra rm trade in more headquarter-intensive sectors. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

34 Contractual Approach Early views of vertical FDI were based on factor price di erences across countries. Together with the horizontal models they generated a theory in which the emergence of MNEs is determined by some combination of: lactational advantages (distance, factor prices); technological factors ( rm vs. plant-level economies of scale). These theories enhance our understanding of trade and FDI ows, but they share a common failure to account for internalization. Why is fragmentation occurring within rm boundaries? Or across rm boundaries? More recent theories, based on contractual frictions, yield new insights. They predict: Larger shares of intra rm trade in more headquarter-intensive sectors. Larger shares of intra rm imports from countries that are well endowed with inputs that are intensively used in headquarter services. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

35 Sorting Pattern (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

36 Evidence 1 0 dru log of (Mif / M) -1.5 ele cle com veh rub aud oel ima foo fme ins sto tra pla och pap bev che -3 lum pri tex oma y = x (1.02) (0.24) R 2 = log of (Capital / Employment) Notes: The Y-axis corresponds to the logarithm of the share of intrafirm imports in total U.S. imports for 23 manufacturing industries averaged over 4 years: 1987, 1989, 1992, The X-axis measures the average log of that industry s ratio of capital stock to total employment, using U.S. data. See Table A.1. for industry codes and Appendix A.4. for data sources. Share of Intra rm U.S. Imports and Relative Factor Intensities (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

37 Evidence 2 0 log of (Mif / M) -2 PHL MEX PAN MYS BRA HKG ARG COL OAN CHE IRL SGP NDL CAN GBR DEU BEL FRA ESPSWE JPN AUS ISR ITA -4 IDN CHL VEN -6 EGY y = x (2.55) (0.29) R 2 = log of Capital-Labor Ratio Notes: The Y-axis corresponds to the logarithm of the share of intrafirm imports in total U.S. imports for 28 exporting countries in The X-axis measures the log of the exporting country s physical capital stock divided by its total number of workers. See Table A.2. for country codes and Appendix A.4. for details on data sources. Share of Intra rm Imports and Relative Factor Endowments (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

38 Evidence 3 The above sorting pattern holds also in Japan. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

39 Evidence 3 The above sorting pattern holds also in Japan. In U.S. data intra rm trade is larger: (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

40 Evidence 3 The above sorting pattern holds also in Japan. In U.S. data intra rm trade is larger: 1 The larger the share of headquarter services (R&D intensity). (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

41 Evidence 3 The above sorting pattern holds also in Japan. In U.S. data intra rm trade is larger: 1 The larger the share of headquarter services (R&D intensity). 2 The larger productivity dispersion. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

42 Evidence 3 The above sorting pattern holds also in Japan. In U.S. data intra rm trade is larger: 1 The larger the share of headquarter services (R&D intensity). 2 The larger productivity dispersion. 3 Intra rm trade is largest where headquarter inputs are important and productivity is high. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17

43 Evidence 3 The above sorting pattern holds also in Japan. In U.S. data intra rm trade is larger: 1 The larger the share of headquarter services (R&D intensity). 2 The larger productivity dispersion. 3 Intra rm trade is largest where headquarter inputs are important and productivity is high. 4 Internalization rises with improved contractibility of the supplier s inputs. (Harvard University) New Perspectives on Foreign Direct Investment January, / 17