Three empirical essays on foreign direct investment, research and development, and insurance

This dissertation consists of three independent essays, all of which are empirical treatmentsof different determinants of economic growth.The first essay, which is in Chapter 2, evaluates the role economic freedom playsin mediating the effect of foreign direct investment (FDI) on growth. It tests whethercountries with sufficiently high level of economic freedom can exploit FDI more efficiently. Ituses cross-country observations from 84 countries for the 1976-2005 period. It applies athreshold regression which is flexible enough to accommodate the possibility that theimpact of FDI on growth ‘kicks in’ only when the level of economic freedom exceeds someunknown threshold. The results show that FDI has no direct (linear) effect on output growth.Instead, its impact is conditional on the level of economic freedom in the host countries.Only countries whose level of economic freedom has exceeded the threshold level ofeconomic freedom benefited from FDI inflows. In countries below the threshold level, FDIdeliver no beneficial effects. The findings are robust to several sensitivity checks andconsideration of endogeneity.The second essay (Chapter 3) tests the channels and magnitude of R&D spilloversfrom developed countries to East Asian countries (China, Korea, Malaysia, Singapore, andThailand). It examines three possible spillover channels - imports, inward FDI, and outwardFDI - using panel data for the period 1984-2005. It uses a novel panel estimator whichallows for cross-sectional dependence and provides country-specific estimates of R&Deffects. There are several important conclusions emerge. First, both domestic and foreignR&D are important for productivity improvements. Second, imports are the most importantchannel of spillovers while spillover effects via FDI in uncertain. Third, there is someevidence that domestic R&D helps to increase the incidence of R&D spillovers, especiallyvia import channel. Fourth, the U.S. is a relatively stronger provider of spillovers thanJapan.Chapter 4, which is the final essay, examines the impact of insurance sectordevelopment on output growth, capital accumulation and productivity improvement. It usespanel data from 52 countries for the period 1981-2005, and applies a recent generalizedmethod-of moments (GMM) dynamic panel estimator. The results show that thedevelopment of insurance sector is important for long-run output growth, capitalaccumulation and productivity growth. For developing countries, insurance affects growthpredominantly through capital accumulation while in developed countries it enhancesproductivity growth. The findings are robust to biases introduced by unobserved countryspecificeffects, simultaneity, weak or numerous instruments. It remains valid even aftercontrolling for bank and stock market developments