In this paper we empirically investigate the causal link between money and economic growth employing a Markov switching Granger causality analysis. We carry out our investigation using quarterly Turkey real Gross Domestic Product, real M2 and interbank money market rate data which cover the period between 1987: Q4 and 2012:Q1. We find that there are significant changes in the causal relation between money and economic growth over the sample period under investigation. Our results show that M2 growth and interest rate have significant predictive content for real economic activity in the Turkish economy. Furthermore, the causality running from interest rate to output growth seems to be strongly apparent particularly during the periods of economic downturn. We also document that M2 has predictive power in explaining output growth particularly during the mid-1980s and during 1990s.