Financial development and growth: A clustering and causality analysis

This article examines the relationship between financial development and economic growth in a sample of 69 countries. A regime switching panel vector autoregression model is specified to detect directional changes in finance-growth causality and potential time variation of such causality patterns. In addition, a clustering analysis is performed to identify the presence of convergence clubs based on data properties. The findings show that most countries have switching between two states: one way causality from growth to financial development but not the other way round, and coexistence of bi-directional causality. Poorer countries are represented by a system with stable steady state while the clusters of advanced economies tend to exhibit multiple steady states. The clustering results map closely the degree of financial openness, and the cultural and geographical proximities of member countries.