PurposeThe purpose of this paper is to uncover the direction of causality between foreign exchange market and stock market in Bangladesh, where financial markets are yet in their early development stage.Design/methodology/approachThe paper employs the Granger causality tests using monthly data spanning over two decades. In order to study possible existence of causality in the data, sub‐samples are constructed in addition to the full sample.FindingsThe overall results indicate absence of any causality running between foreign exchange market and stock market in the full sample and in the sub‐sample created around the stock market crash.Originality/valueThis study would be the first of its kind using Granger causality approach to test whether change in exchange rates lead to changes in the stock market in Bangladesh, and vice‐versa. The paper also offers some implications of the findings which could be of significant value to policy makers.