The manager beauty contest: Do the figures matter?
- PDF / 145,262 Bytes
- 12 Pages / 650 x 822 pts Page_size
- 79 Downloads / 157 Views
Stan Beckers* is Global Head of Investments at WestAM and Visiting Professor at KULeuven (Belgium). His research interests cover risk modelling and forecasting, investment management skill, capital market integration and multiple management strategies. *WestAM, 33/36 Gracechurch Street, London EC3V 0AX. Tel: ⫹44(0)20 7 444 6601; Fax: ⫹44 (0)20 7 444 6691; e-mail: stan [email protected]
Abstract This paper investigates persistence in the performance of UK institutional pension fund managers. In particular, it questions to what extent managers succeed in beating their benchmark over longer time periods and, hence, whether ‘winners’ can be identified on the basis of past performance. This paper does not generate the ultimate formula that will lead to the selection of the perfect manager, but it does provide important insights into the length of the period over which managers can be evaluated, the forecasting accuracy that can be achieved and the additional information that can be relied upon to improve this forecasting accuracy. Keywords: performance persistence; manager selection; manager evaluation; information ratios
Introduction Distinguishing luck from skill is one of the more vexing tasks in asset management. All investment management marketing literature now clearly states that ‘past performance is no guide to the future’. Still, we know that not all managers are equally good and that their quality must somehow be reflected in their performance. It is therefore quite natural to try to separate the good from the bad, based on fund returns. This is especially so since past performance is one of the few quantifiable, and therefore objective, measures for manager or fund selection. This paper empirically investigates to what extent historical performance numbers can be meaningfully used to separate the wheat from the chaff. In the process, we
䉷 Henry Stewart Publications 1470-8272 (2000)
address the following questions: — Should performance or risk-adjusted performance numbers be used? — What is the ideal time period over which a manager should be evaluated? — How is performance persistence measured? — How are ‘good’ and ‘bad’ defined? It is futile to look for the ultimate recipe that uses past performance ingredients to forecast future league table rankings optimally. This paper is more concerned with setting realistic expectations, applying good housekeeping rules and avoiding mistakes. It will help put track records in the right perspective and unearth whatever (little) information they contain.
Vol. 1, 1, 7–18
Journal of Asset Management
7
Beckers
The sample This study uses the quarterly UK pooled fund returns for the period Q1 1989 to Q3 1999 (a total of 43 quarters). These pooled funds can be considered as the managers’ flagship portfolios, since they are the typical investment vehicle for the smaller pension funds whose investable assets are too small to justify a segregated account. These pooled funds are managed on a full discretionary basis and are therefore not hampered by potential client-i
Data Loading...
