Assessing the effects of corruption and capital controls on the volume and composition of capital flows to developing countries.

摘要

Foreign, private capital flows to developing countries have proven to be both an engine of growth, as well as a challenge to deal with. While some developing countries have benefited from foreign investment, there are others who have been exacerbated by these flows. The implementation of capital controls has been a remedy in managing the effect of foreign, private investment on the domestic economy. Yet, when making an investment in a foreign country, the international investor can avoid such restrictions.; And much of the literature says that with any policy, there is likely to be some level of corruption associated with it. Indeed, some level of corruption is likely to prevail in virtually every country.; This dissertation presents empirically the effects of corruption and capital controls on the composition and volume of private capital flows to 34 developing countries using random effects, fixed effects and OLS regression techniques. Four types of capital flows are studied: foreign direct investment, bank lending, bonds, and equities. Under the guise of the "efficient grease" theory, this study will test the linearity of the corruption index, as well as the capital control measure.; The results show that corruption is indeed non-linear in all capital flow types, employing each regression technique. With the exception of equity flows, the "efficient grease" theory of corruption applies to private capital flows. Capital controls, which appear to be non-linear, do not behave in an expected manner and do not appear to significantly affect capital flows. Unexpectedly, these results do not lead one to conclude that random effects estimates are significantly different from OLS estimates.; The business implications from this research are that the presence of corruption increases the investment transaction costs for foreign direct investment, bank lending and bonds. Similar implications cannot be drawn for investment policies related to capital controls. In other words, this research shows that because capital controls do not behave in an intuitively non-linear manner no investment policies can be made to account for them.