Purpose – the purpose of this paper is to examine the tournament hypothesis in the uk mutual fund market. Based on a previous us study, fund managers were expected to alter risk-taking behaviour in response to their performance relative to competing fund managers.design/methodology/approach – based on an earlier methodology, contingency tables were used to examine the risk-taking behaviour of fund managers after an interim performance period. The sample consists of 422 uk equity mutual funds with monthly data from 1989 to 2003. To avoid survivorship bias, funds that did not survive the entire sample period were also included. This leads to a total of 3,617 observations.findings – the main conclusions are two-fold. First, using the entire 1989-2003 sample period no consistent evidence for tournament behaviour is found. This is robust to the effects of survivorship bias and window dressing. Second, splitting the sample period into two sub-periods reveals an interesting pattern. During the first part of the sample period, 1989-1996, significant evidence for tournament behaviour is found. During the second part of the sample period, 1997-2003, significant support for strategic behaviour, as described theoretically by taylor has been documented.research limitations/implications – the results suggest that after 1996, managers entered into a strategic game that takes the actions of competing managers into account instead of seeing them as exogenous benchmarks.originality/value – by studying the uk fund market, the us results can be tested to see if they are sample specific or can be carried over to other countries as well. Furthermore, the sample period includes data after 1996, the year of the first publication on the tournament hypothesis. This enabled investigation in to whether managers adapted their strategies.