Recent evidence on the persistence of fund performance—A note

Abstract

This note highlights recent evidence on the persistence of fund performance in South Africa, throwing new light on the 1985–1995 study conducted by Meyer (1997). Our analysis on the persistence of South African fund performance is based on the General Equity category of unit trusts over the period 1995-August 2001. Examining performance persistence across consecutive years, we found evidence of fund performance persistence. This latest evidence seems to be primarily driven by the consistently superior performance of some top performers (rather than persistent poor performers). It seems that there are managers who have significant skill, to the extent that they have systematically been able to outperform their peers. Interestingly this evidence is similar to evidence forthcoming from studies conducted in the USA.

As a confirming exercise, we considered the impact of investing in historical top quartile performers, second quartile performers etc., rebalancing annually and accumulating returns during the subsequent year of “unseen” performance data. The results are indeed surprising and compelling. Contrary to popular local belief, selecting historical top quartile performers yielded vastly superior performance results to a wide range of alternate combinations. Interestingly the second best strategy was that of a passive peergoup benchmarking approach.

KEY WORDS:

• Performance persistence
• peergroup benchmark
• rank correlation