DescriptionChapter 2 of the dissertation analyzes the relationship between de jure financial openness and de facto financial openness. When a country changes its policy towards foreign financial transactions, do actual capital flows respond? I use the Chinn-Ito index to proxy a country's legal regime regarding external transactions, and examine its relationship to realized international capital flows disaggregated by type and direction. Panel data estimation methods are used to explicitly assess the interaction of country development characteristics with policy. I find that in general the relation between legal openness and realized international financial flows is weak. However looking at the decomposition, I find policy does affect Foreign Direct Investment and debt outflows. Furthermore, I find that country attributes play a key role in the efficacy of the policy stance, in particular a country's level of domestic financial development. Turning to time series methods, I then estimate a VEC model and analyze the variance decompositions of five Asian economies' bank lending rates. I find that the experience of Japan, South Korea and Indonesia support the conclusions arising from the panel analysis that greater financial development enhances realized financial openness. Focusing on determinants of financial development, Chapter 3 and 4 test for empirical evidence of experience effects in banking. The hypothesis is that firm-specific or sector-wide learning, via knowledge spillovers, improves bank cost efficiency. Chapter 3 constructs a bank cost function extended to include firm-specific experience. Using a sample of US banks and applying a two-step correction procedure to my bank cost function, I correct for endogeneity as well as selection biases. I find that experience is associated with reduced costs: the experience effect is decreasing and fades after around 10 years. Chapter 4 extends the analysis to a sample of international banks and tests for knowledge spillovers. First a simple learning curve model is estimated. Next, the international sample is used to estimate the bank-specific cost function developed in Chapter 3. The estimated experience effects vary depending on the experience proxy and econometric model used, potentially due to data limitations. Nevertheless, the results suggest some evidence of firm-specific learning by doing and international knowledge spillovers in banking.