The objective of this thesis is to empirically identify the determinants of the formation of an international financial centre (IFC). To achieve this objective, we firstly identify factors that determine the concentration of international financial activities, and subsequently estimate the relative importance of these factors using a dynamic panel econometric model. As part of the research, a large panel data set covering a number of macroeconomic variables for 71 countries over a time period of 21 years, from 1982 to 2002 is compiled. This data set is used in assessing the significance of factors such as agglomeration effect, economic size, GDP per capita growth, macroeconomic soundness, infrastructure, and time zone effects. An innovation of this thesis is the construction of a time zone matrix that can be used to measure the effects of time zone. Although extensive research on IFCs has been conducted, little empirical work has been done. The majority of previous studies on IFCs are descriptive; the remaining limited empirical studies typically focused on a handful of already established IFCs. Given the specification of the model, which is a dynamic panel model with random effects, this thesis makes use of the methodology based on the Generalized Method of Moments (GMM) approach proposed in Arellano and Bover (1995) and discussed in Arellano (2003). The results indicate that economic size effect by far is the largest followed by agglomeration. The growth rate of an economy, the stage of development of an economy, and the time zone effect individually explains much smaller variations. Based on the econometric results, a simulation exercise is conducted to investigate the prospect of China and India in becoming an IFC in 5 to 10 years time. The simulation results indicate that while China has plenty of potential in establishing an IFC, the prospect for India is not as promising.