1. 1 The Role of Information in Driving FDI: Theory and Evidence Ashoka Mody, IMFAshoka Mody, IMF Assaf Razin, Tel-Aviv University and Cornell UniversityAssaf Razin, Tel-Aviv University and Cornell University Efraim Sadka, Tel-Aviv UniversityEfraim Sadka, Tel-Aviv University

2. 2 Foreign direct investment (FDI) has been growing faster than world GDP, and is becoming a major component of foreign investment.Foreign direct investment (FDI) has been growing faster than world GDP, and is becoming a major component of foreign investment. We identify from empirical data two main categories of variables that significantly explain FDI inflows:We identify from empirical data two main categories of variables that significantly explain FDI inflows: 1)Positive correlation between the industry specialization in the source countries and FDI flows into the host countries is shown to exist; 2)Countries with higher quality of corporate transparencies and stronger capital market institutions attract less FDI flows.

3. 3 In this paper we develop a simple information-based model, that is consistent with empirical findings. We interpret the industry specialization in the source country as providing a comparative advantage to the potential foreign direct investors, in eliciting good investment opportunities in the host country, relative to domestic investors in the latter country: A lower cost of cream-skimming (of high-productivity firms) on the part of foreign direct investors. This advantage of FDI investors in their cream-skimming skills is less pronounced when corporate transparencies and capital market institutions are of high quality in which case FDI inflows are less abundant.

4. 4 Our model suggests that the gains from FDI are reflected in a more efficient size of the stock of domestic capital and its allocation across firms. Domestic firms that are controlled by FDI investors are typically the “cream” (high-productivity firms). The magnitude of the non-traditional gains from trade that arise in our model depends crucially (and inversely) on the degree of competition among potential FDI investors over the domestic firms. These gains can shrink to zero if there is no such competition altogether. Also, FDI inflows enlarge the size of the aggregate stock of domestic capital (under plausible assumptions).Our model suggests that the gains from FDI are reflected in a more efficient size of the stock of domestic capital and its allocation across firms. Domestic firms that are controlled by FDI investors are typically the “cream” (high-productivity firms). The magnitude of the non-traditional gains from trade that arise in our model depends crucially (and inversely) on the degree of competition among potential FDI investors over the domestic firms. These gains can shrink to zero if there is no such competition altogether. Also, FDI inflows enlarge the size of the aggregate stock of domestic capital (under plausible assumptions).

5. 5 N – Large number of ex-ante identical domestic firms. K – Capital stock (in the first period). - Rate of depreciation. - Rate of depreciation. - Output (in the second period). - Productivity factor. - Productivity factor. G – Cumulative distribution function of. A common knowledge signal about. - A common knowledge signal about.

6. 6 Cumulative distribution function of, conditional on Cumulative distribution function of, conditional on : max expected market value of the firm, conditional on : max expected market value of the firm, conditional on : K 0 – Initial stock of capital r – world rate of interest.

7. 7 Therefore:where:

8. 8 Suppose that after an FDI investor acquires and gains control of the firm, she can apply at a cost C F (lower than for domestic investors) a screening technique that elicit the true of the firm. Therefore, the bid price of an FDI investor for a firm with a signal is: Where:

9. 9 The ask price of the domestic owners is. Therefore, there is a cutoff level of the signal, denoted by, and defined by such that

10. 10 That is: all firms with signals above are purchased by FDI investors; all other firms (with ) are purchased by domestic or foreign portfolio investors. ) are purchased by domestic or foreign portfolio investors. In the absence of FDI, the cutoff signal, denoted by, is defined by: where C D is the screening cost of domestic investors (C D >C F ). We then have

11. 11 Stock of Capital 1 In the absence of FDI with FDI Note:

12. 12 Results: (1) Stock of capital is higher with FDI than in its absence. (2) Gains from FDI: (a) Efficiency gain – (b) Cost gain –

13. 13

14. 14

15. 15

16. 16

17. 17

18. 18 Brief Description of Each Round Round 1: The three original tables from the paper. Gravity Equations for Trade, FDI, and Equity with trade residual not including host country creditor rights. Round 1: The three original tables from the paper. Gravity Equations for Trade, FDI, and Equity with trade residual not including host country creditor rights. Round 2: Regressed trade including host country creditor rights. The trade residual from this regression is then used in the regression from round 1. Round 2: Regressed trade including host country creditor rights. The trade residual from this regression is then used in the regression from round 1. Round 3: Replaced Trade Residuals with the actual value of Trade. Then ran the same regression as in round 1. Round 3: Replaced Trade Residuals with the actual value of Trade. Then ran the same regression as in round 1.

19. 19 Round 4: Replaced the dependent variable to FDI/Trade and Equity/Trade. Then ran the same regression as Round 3. Round 5: Regressed the gravity equation for trade and using the fitted values of trade, then ran the same regression as in round 4. Round 5: Regressed the gravity equation for trade and using the fitted values of trade, then ran the same regression as in round 4. Round 6: The same regression as Round 4 including time dummy variables. Round 6: The same regression as Round 4 including time dummy variables. Round 7: Replaced the dependent variable to FDI/Equity and ran the same regression as round 2. Round 7: Replaced the dependent variable to FDI/Equity and ran the same regression as round 2.

20. 20 Round 8: Replaced the dependent variable with the volatility of FDI and ran the same regression as round 2. Round 9: Instrumented the host country Debt- Equity ratio with Host country GDP per capita, host country creditor rights, and host country dummies. Then ran the same regression as round 4. Round 9: Instrumented the host country Debt- Equity ratio with Host country GDP per capita, host country creditor rights, and host country dummies. Then ran the same regression as round 4.

21. 21

22. 22

23. 23

24. 24

25. 25

26. 26

27. 27

28. 28

29. 29

30. 30

31. 31

32. 32

33. 33

34. 34

35. 35

36. 36

37. 37