The economics literature provides ample evidence that higher corruption levels discourage FDI inflows. In this paper we address, for the first time, the exact reverse link, i.e., the empirical effect of higher FDI inflows on corruption. Our dataset covers a wide set of countries in the period between 1980 and 2000 and we confront the issue of causality by constructing a new set of instrumental variables that rely on geographical and cultural distance between FDI emitting and receiving countries. FDI as a share of GDP significantly decreases country corruption and the quantitative impact is stronger when we instrument for FDI. The results are extremely robust to the inclusion of other determinants of openness such as trade intensity and average tariff level and the impact of FDI on corruption is of the same order of magnitude as that on per capita income.
Available at: http://works.bepress.com/josetavares/17/