This paper revisits the institutional determinants of foreign direct investment (FDI) using a comprehensive new data set on the regulations that govern FDI in more than 80 countries. It exploits the presence of confirmed zero investment flows between countries to estimate productivity cut-offs of firms that invest abroad profitably. This approach corrects likely biases arising from firm heterogeneity and country selection in a theoretically derived gravity-type model. The analysis finds inward FDI to be highly responsive to cross-country variation in specific institutional provisions, such as arbitration of disputes and legal procedures to establish foreign subsidiaries.
The importance of FDI-specific provisions stands out even after controlling for the general quality of institutions. Statutory openness to FDI, however, has no association with actual inflow of investment. These results are found to be robust to different specifications.
1.1 introduction
Foreign Direct Investment (FDI) remains one of the most important forms of crossborder capital flow into developing countries: in 2009, FDI inflow amounted to more than US$510 billion, exceeding inward remittance (US$307 billion) and development aid (US$91 billion). As shown in Figure 1, however, just nine countries have accounted for about 60 percent of FDI inflow into developing countries over the past decade. In an era when almost all countries in the world welcome FDI, allowing an average foreign equity ownership of 90 percent across all sectors (Table 22), this cross-country asymmetry deserves explanation that is beyond the obvious such as countries’ sizes and growth prospects. In this paper, I focus on the policy and institutional determinants of FDI using a new cross-country data set, Investing Across Borders (IAB) 2010, drawn from World Bank Group (2010). The data set consists of indicators of FDI regulation that specifically measure each country’s i) openness to foreign investment by sector; ii) quality of institutions related to resolving investment disputes; and iii) time, procedures and rules required to set up wholly foreign-owned subsidiaries. These FDI-specific indicators are the most comprehensive to date in terms of topics and countries, and they obviate the need to rely on proxy indicators for policy openness or the quality of institutions.
I also adopt a new methodological approach that corrects for two major biases prevalent in standard gravity models of FDI. The first one arises when limiting the sample to only those countries that actually have an investment relationship with each other and excluding those that do not. This is a problem of country selection induced by zero bilateral flows. The second bias arises when firms are not differentiated by their ability to meet the fixed costs of investing abroad. This is the problem of firm heterogeneity. The two biases are linked when zero flows are caused by the fixed cost of investing abroad, and only the more productive firms meet such costs.
I build on the insight of the new trade literature that firms are heterogenous within industries in terms of productivity, size, use of inputs, and wages.4 This translates into distinct decisions by firms (that co-exist within a narrowly-defined industry) on whether or not to export, or undertake FDI, or just serve the domestic market (Greenaway & Kneller 2007; World Trade Organization 2008; Helpman 2011). When profits are a function of varying productivity and differing fixed costs, Helpman et al. (2004) show that there is a natural sorting of firms, with the most productive self-selecting to undertake FDI. The next tier serves the foreign market through export, and the least productive serve the domestic market only (Appendix B).
The earlier generation of “new” trade models that integrated economies of scale and monopolistic competition was a breakthrough in understanding a new source of comparative advantage (Helpman 2011). However, they addressed neither heterogeneity nor country selection. In particular, the assumption of symmetry in firm size and productivity leading to a prediction that all firms export to all countries is not supported by evidence at the firm level.
The paper proceeds as follows. Section 2 describes the relevant literature on the determinants of FDI, highlighting the institutional drivers on which the paper builds. Section 3 sketches the theoretical derivation of a gravity-like model for FDI and its empirical extension. Section 4 introduces the data. Section 5 explains the econometric method used to incorporate firm heterogeneity and redress country selection bias. Section 6 shows the main results by comparing benchmark estimates with those obtained after correcting for biases. Section 7 uses alternative dependent and explanatory variables to check for robustness of results.The final section concludes.
World Bank.Author: Wagle, Swarnim.Document Date: 2011/12/01.Document Type: Policy Research Working Paper.Report Number: WPS5914.Volume No: 1 of 1
x
