1

• Pension Fund Trustee Competence
• By Gordon L Clark,
• Emiko Caerlewy-Smith,
• and John C Marshall
• Sponsored by the NAPF
• University of Oxford
• July 2005

2
1. Introduction

• Anglo-American pension funds are envied by
  continental Europe
• Crucial to many peoples total retirement income
• Trustees are crucial to pension institutions
• Subject to fiduciary duty as well as related
  statutory requirements.

3
2. Trustee competence

• Emphasised in the UK Myners Report
• Questioned by industry analysts re. innovation
• Now considered by The Pensions Regulator with
  respect to trustee codes of practice
• This project is about the nature and scope of
  trustee decision-making (in conjunction with the
  NAPF).

1
4
3. Summary of the project

• Beginning with the education and experience of
  trustees (based on participation in training
  programmes)
• We also assess trustee competence focusing upon
  problems relevant to investment
• Drawing implications for fund governance
• And the relationships of trustees with advisors,
  service providers.

2
5
4. Steps in the project

• Surveys of trustees at NAPF training programmes
• Socio-demographic profiles and attitudes re.
  governance procedures
• Including reference to trustee "knowledge and
  understanding (cf. Myners principles etc)
• With the focus on empirically evaluating and
  measuring trustee competence.

3
6
5. Problem-solving abilities

• Design and execution of problem sets relevant to
  investment decision-making
• Enrolling about 40 trustees, 80 undergraduates
• Two sets of problems, each taking about 1 hour to
  complete (together or separately)
• Drawn from a relatively small group of DB pension
  funds, and Oxford undergraduates.

4
7
6. Problem qualities

• Not intelligence tests (eg. IQ tests etc)
• Not categorical tests - rational v irrational
• Level of education not crucial (although this is
  subject to specific training in finance etc.)
• Basically tests of comprehension and cognition,
  with reference to logic and numerical inference
  in the context of responsibilities.

5
8
7. Methodological Issues

• Problems with recruiting trustees and
  undergraduates
• Limited options for discrimination by
  socio-economic status and education
• Gaps between the problems and the real world
• Expectations of psychology v. the social sciences
  on issues such as representativeness.

9
8. Findings - discounting the future

• Some trustees discount the future, some don't,
  and some are inconsistent over time
• There is a wide range of discount functions only
  some of which appear exponential
• Male undergraduates do discount the future
  (strongly)
• Results are consistent with the psychology
  literature (re. weakness of will etc.).

6
10
9. Findings - my money, your money

• Trustees (as individuals) willingly accept
  moderate risk if asked their attitudes
  (although this varies)
• Trustees are risk-averse regarding others money
  when required to calibrate options
• Undergraduates are more risk tolerant (esp.
  males)
• Consistent (somewhat) with psychological evidence
  although there is little in the experimental
  literature on the effect of commitment to
  otherson risk-taking.

7
11
10. Trustees attitudes to risk

12
11. Trustees risk calibration
13
12. Undergraduates risk calibration
14
13. Findings - probabilistic reasoning

• Neither trustees nor undergraduates are good at
  probabilistic reasoning
• Often unable to recognise or link together the
  elements necessary to assess probability (in a
  Bayesian sense)
• As well, surprising variation in the methods
  used to assess probability
• Consistent with the psychology literature, even
  if probabilities are often involved in discussion
  of investment strategies.

8
15
14. Findings - information processing

• Trustees and undergraduates are inefficient users
  of information
• Undergraduates often use too much information
• Trustees, sometimes, use too little information
• Rules can make a substantial difference to the
  use of information and attaining correct
  solutions (consistent with the psychology
  literature).

9
16
15. Conclusions I

• Should we rely on older men rather than Oxford
  undergraduates to be trustees?
• Many of the trustees sampled have poor
  problem-solving skills relevant to investment
  issues
• Worse, the range of trustee solutions may be
  wide AND rarely correct
• Most trustees are unable to deal with
  probabilities, and most are inefficient
  information processors.

10
17
16. Conclusions II

• Qualifications to findings would include the
  sample size, the nature of problems etc.
• Reasonable concern over the formality of testing
  v. real world settings
• In any event, recent research suggests that
  short-cuts dominate decision-making procedures
• With little regard to how collectivity makes a
  difference.

18
17. Implications I

• Boards carry a diversity of trustee competence
• With important differences in the skills
  appropriate to investment fundamentals
• With profound differences of approach to the
  solving of problems.
• Suggesting the possibility of considerable
  disagreement (or silence on basic issues of
  disagreement).

11
19
18. Implications II

• Fund governance could be important in this
  context
• Leadership and structured decision-making crucial
  in managing ignorance, confusion, and
  disagreement
• Otherwise, there will be high levels of (caution)
  risk aversion (or worse), seeking refuge in
  expert advice, rules and procedures
• With low levels of innovation and slow adaptation
  to changing circumstances.

12
20
19. Implications III

• Can trustee codes of practice make a difference?
• Yes, for basic procedures but no, considering
  the importance of expertise
• Implying the possibility of deference to experts
• And reliance upon advisers to articulate and
  implement coherent investment strategies (in the
  absence of knowledge and understanding).

13
21
20. Future research

• Can organisational structure and coherent board
  decision-making procedures overcome the
  heterogeneity of trustee competence?
• Is the representative model of the UK pension
  trustee system consistent with high-quality,
  efficient decision-making?
• Is the trust institution appropriate for the
  governance of 21st century pension funds given
  the need for financial innovation?
• Has the trust institution come to the end of its
  usefulness?

14
22
21. Further reading

• Clark, G.L. 2000. Pension Fund Capitalism.
  Oxford Oxford University Press
• Clark, G.L. 2004. Pension fund governance
  expertise and organizational form. Journal of
  Pension Economics and Finance. 3 233-53
• Lerner, J. et al. 2005. Smart institutions,
  foolish choices? WP11136. Cambridge MA National
  Bureau of Economic Research
• Merton, R.S. and Z. Bodie 2004. The design of
  financial systems towards a synthesis of
  function and structure. WP 10620. Cambridge MA
  National Bureau of Economic Research.