Italian Equity Funds : Efficiency and Performance Persistence

Are Italian Mutual Funds able to generate extra-return? Are some of them able to persistently beat the competitors? In this paper we address these questions and provide a detailed and systematic performance and return persistence analysis of the Italian equity mutual funds. The implications of this empirical work do not affect only mutual funds market but also the entire Italian capital market and its efficiency. In our work we show that, in general, fund managers are not able to score extra-performances and only few managers have stock picking ability or market timing ability. To illustrate these results the performance analysis has been carried out, analysing: (i) total returns, (ii) risk adjusted returns (Sharpe Ratio, Treynor Ratio, Sortino Ratio), (iii) stock picking ability measures (Jensen's Alpha) and (iv) market timing ability measure (Treynor-Mazuy and Henriksson-Merton models). Performance measures have been evaluated by means of a significance test over the whole sample period, and stability test with other time-series tests have been performed. An interesting result of this paper is the bi-periods and multi-periods persistence analysis; we find: (i) absence of long-run persistence on total returns and on risk adjusted returns; (ii) a weak evidence of reversal effect, and (iii) evidence of a hot-hand effect on risk adjusted returns on four-month intervals (short run persistence). We use a general approach to investigate performance persistence paying attention to the relation between the persistence measure and other variables such as the lag of evaluation, the performance measure, the number of performing class and the statistical evaluation test. A rank order correlation test is considered in order to assess the effect of the different performance measures on the persistence results. The following statistics are curried out on a bi-periods analysis with chi-square test, Cross Product Ratio, Spearman's coefficient and Cross Sectional Regressions. The multi-periods analysis has been performed with Kolmogorov-Smirnov test