Past Performance A Guide to the Future Leon Beukes Finance

1
Past Performance A Guide to the Future?Leon
BeukesFinance Investment Conference
abcd

• 22-24 June 2003
• The Caledonian Hilton Hotel, Edinburgh

2
Outline

• Putting past performance in context
• Difficulties when looking for persistence
• Literature review
• Results of some recent studies
• Concluding thoughts

3
Past performance in the correct context

• 1. Help answer questions like
• Is the managers performance consistent with
  their stated style (eg Value)?
• Do the controls result in the desired dispersion
  of returns?
• 2. Help validate qualitative analysis
• should never select a manager SOLEY on past
  performance

4
Difficulties in looking for performance
persistence

• Survivorship bias
• Allowing for expenses
• Allowing for Risk levels
• Which Benchmarks to use?
• Allowing for manager style
• Lack of relevant performance history
• What is significant?

5
High signal-to-noise ratio

• Source Hewitt Bacon Woodrow Assuming a
  Tracking Error of 2.5

6
Manager Skill Cycle
7
Literature Review

• Many studies done on the persistence of past
  performance specially in the US
• Most credible studies found no conclusive
  evidence for the persistence of risk-adjusted
  outperformance
• some empirical evidence that underperformance
  could persist.

8
Jain Wu (2000) Pre-advertisement Performance
9
Jain Wu (2000) Post-advertisement Performance
10
Rhodes (2000) 1981 quintile performance
11
Rhodes (2000) 1987 quintile performance
12
Rhodes (2000) Case 1

• Number of funds 186

13
Rhodes (2000) Case 2

• Number of funds 380

14
Rhodes (2000) Case 3

• Number of funds 368

15
Finding performance based rules

• Source Hewitt Bacon Woodrow

16
Concluding thoughts

• Need to educate investors that past performance
  cannot be the sole reason for selecting a manager
• Less focus on short term performance likely to be
  better for fund managers
• Some evidence for persistence in
  underperformance, but it is important to
  understand the reasons behind it!