The aim of this paper is to investigate the long run relationship between the development of banks and stock markets and economic growth. We make use of a Johansen-based panel cointegration methodology allowing for cross-country dependence to test the number of cointegrating vectors among these three variables for 5 developing countries. In addition, we test the direction of potential causality between financial and economic development. Our results conclude to the existence of a single cointegrating vector between financial development and growth and of causality going from financial development to economic growth. We find little evidence of reverse causation as well as bi-directional causality. We interpret this as evidence supporting the significance of financial development for economic development although banks and stock markets may have different effects depending on the level of economic development.