Managerial Turnover, Fund Family Tournament and Investors Learning in the UK Fund Market

Abstract

This research is constructed to provide a comprehensive analysis of key factors that influence fund performance, and the methods of evaluating such performance. Our examination is from both the managerial and the investor perspectives in the context of the UK fund industry.

We begin by analysing the managerial response. We examine the effectiveness of the internal monitoring and control system in UK funds, focusing on the relation between top manager turnover and the performance of UK unit trusts and OEICs. We apply a series of methods, including performance evaluation based on the factor models, percentile ranking, and sample matching tests. Funds that have experienced replacement of top managers are examined for their performance in the pre and post-replacement periods. A variety of methods are applied to measure factors that affect the probability of replacement in the pre-replacement period. Bootstrapping simulations are implemented to further examine whether UK fund companies can distinguish between poorly skilled managers and unlucky ones. It is found that many fund companies are not captivated by the ‘lucky’ managers’ extreme performance and are willing to give ‘unlucky’ managers another chance. Moreover, underperforming managers are more likely to be replaced when fund inflows are declining than are outperforming managers. Managers' adjustment of portfolio compositions exerts mixed influences on the probability of replacement.

Next, we apply tournament analysis to the UK fund market. We find supportive evidence of significant risk shifting in the family tournament; i.e. interim winning managers tend to increase their levels of risk exposure more than the losing managers do. Our results also show that the risk-adjusted returns of the winners outperform those of the losers following the risk taking. As such, risk altering can be regarded as an indication of managers’ superior ability. However, the tournament behaviour can still be a costly strategy for investors, since in this case winners would be seen as beating the losers in terms of the observed returns due to deterioration in the performance of their major portfolio holdings.

Finally, the thesis further examines the cross-fund learning among investors. A linear hierarchical model is constructed to consider cross-learning of the funds within a fund family in performance evaluation. We apply a full Bayesian treatment of all factors of the pricing model and allow both the fund family and individual managers to have dependent prior information regarding the funds’ alphas. The simulation results suggest that returns from peer funds within the family significantly affect investors’ updating on fund alphas, since the posterior distribution on fund alphas exhibits a faster shrinkage than is reported in the previous literature. The model is also simulated with various prior beliefs on different factors of the pricing model, i.e. fund alphas, betas and factor loadings of each pricing benchmark, to better address the learning process.