Eempirical studies have not documented an unambiguous relationship between financial development and volatility of economic growth. Existing evidence of contemporaneous association is also insucient to establish the direction of causality. This paper studies the time series relation between financial development, economic growth and growth volatility in one unified framework. Relying on a parsimonious panel VAR framework with a panel of 81 countries between 1962 and 2000, I find significant positive Granger causality from financial development to growth and negative Granger causality from financial development to volatility of growth. Financial development accounts for 2.6% of negative forecast error covariance between growth and growth volatility. These empirical regularities, however, are detected only in emerging countries and not in advanced economies. My findings are consistent with the idea that financial development is especially beneficial to countries at the early stage of industrialization. Further, country specific economic environment and structural factors, such as banking concentration degree, proportion of privately held banks, and legal environments, determine the likelihood of both Granger causality from financial development to growth and volatility, and the fraction of covariance between growth and volatility due to financial development.