Credibility and Flexibility: Political Institutions, Governance, and Foreign Direct Investment

Abstract

This article argues that the relationship between political institutions and foreign direct investment is both nonlinear and conditional upon status quo policies. The empirical analysis demonstrates an inverted U-shaped relationship between political institutions and foreign direct investment in developing countries, with four veto players being the most attractive institutional arrangement. Countries with too few or too many veto players are not favored because of either high policy uncertainty or high policy rigidity. In addition, the benefits and costs of credibility and flexibility vary in good times and hard times. The benefits of maintaining status quo tend to outweigh the costs in countries with good initial regulatory environment. The costs of maintaining status quo tend to outweigh the benefits when countries are more vulnerable to exogenous shocks.

Keywords:

• credible commitment
• foreign direct investment
• investment incentives
• regime type
• veto players

Acknowledgments

I thank Lawrence Broz, Tim Buthe, Peter Cowhey, Roger Gordon, Stephen Haggard, Quan Li, Edmund Malesky, Barry Naughton, Susan Shirk, and three anonymous reviewers for their helpful comments. Replication files are available on the International Interactions dataverse page at http://dvn.iq.harvard.edu./dvn/dv/internationalinteractions.

Notes

¹In 2004, 87% of the 269 FDI-related regulatory changes introduced in 102 countries were designed to make host countries more attractive to FDI (United Nations Conference on Trade & Development [UNCTAD] 2005).

²For a historical review of expropriation of oil and gas investments, see Joffe, Stevens, George, Lux, and Searle (2009).

³Three other measures of FDI have been used in other studies: net FDI inflows (Li & Resnick 2003), log of net FDI inflows (Jakobsen and de Soysa 2006), and log of per capita FDI inflows (for example, Busse and Hefeker 2007). Li (2009a) has discussed the strengths and weaknesses of each measure.

⁴Theoretically, the causality between openness to trade and FDI is unclear because both are endogenous. Empirical studies also show mixed results, suggesting that the direction of trade–investment relationship depends on industries, type of products, or FDI strategies (Markusen 2002).

⁵The LR test of heteroscedasticity across panels indicates very strongly (LR χ²=7277.22, p=.0000) that the variance of the error terms is significantly different between panels (countries). The Durbin-Watson test for autocorrelation in the errors indicates significant first-order correlation of the error terms (F = 5.352, p = .022).

⁶The tipping point is calculated using the formula: X* = β₁/(-2β₂) = 3.203/(-2*(-1.117)) = 1.43.

⁷Shang-jin Wei (2000) also uses PwC's country descriptions on legal and regulatory issues to measure government policies toward FDI. His FDI incentive measure includes two categories: FDI restrictions and FDI incentives, each of which consists of four aspects. The net incentives are the difference between FDI restrictions and FDI incentives.

⁸Note that the tax incentive information in Corporate Taxes and Tax Incentives and Foreign Direct Investment is not identical for some countries. In those cases, I used Corporate Taxes as the baseline to code incentive scores.