Investment strategy for the Petroleum Fund

1
Investment strategy for the Petroleum Fund

• Lecture at UiO
• February 14, 2001
• Birger Vikøren
• Governors Staff
• Norges Bank
• www.norges-bank.no

2
Outline

• Background of the Fund
• Portfolio models
• Country allocation
• Equity portion
• Active management

3
Background of the Fund
4
The Government Petroleum Fund

• A buffer which allows greater room for manoeuvre
  in economic policy should, for example, oil
  prices decline
• A fiscal management tool which ensures
  transparency in the use of petroleum revenues and
  the reallocation of petroleum wealth
• A tool for coping with the financial challenges
  connected with an ageing population and declining
  oil revenues

5
Fiscal surplus

6
Pensions expenditures
Petroleum revenues
7
Production of petroleum Mill. Sm3 o.e.
8
Why is the Fund invested abroad?

• Budget concern
• The Petroleum Fund should not be a second
  budget
• Investment concern
• The Fund does not affect international rates of
  return - better returns abroad
• Monetary policy concern
• The petroleum activity yields substantial
  currency incomes
• Accumulation of foreign reserves counteracts
  appreciation of the currency
• Sector balance concern
• Real appreciation would shift resources to
  non-competetive sectors
• The Fund as a buffer
• Drawing on a domestic fund could destabilise the
  economy when activity is low

9
Petroleum Fund
10
Norways national wealthPercentage distribution
11
The investment strategy could be divided into

• Long-term (passive) investment strategy
• Strategic Asset Allocation (SAA)
• reflected in the benchmark
• Short-term (active) investment strategy
• Tactical Asset Allocation (TAA)
• deviation from the benchmark

12
Petroleum Fund - Division of responsibilities

• Owner Ministry of Finance
• Passive investment strategy
• Strategic asset allocation and investment
  universe
• Benchmarks
• Risk limits
• Evaluates manager (uses consultant)
• Reports to the Parliament

• Manager Norges Bank
• Active investment strategy
• Tactical assets allcoation
• Achieve higher return than benchmark given
  investment mandate and restrictions
• Risk control
• Reports to MOF

13
Strategic asset allocation

• Before setting the investment strategy we need to
  define
• purpose of the Fund
• In terms of the Petroleum Fund, it is natural to
  apply a long investment horizon and to recognize
  the importance of preserving the Fund's
  international purchasing power". (Revised
  National Budget 1997)
• owners risk tolerance
• Two implications
• should not focus on short-term fluctuations in
  return
• should not measure return in Norwegian kroner

14
Guidelines for the Petroleum Fund

• Old (until 21.12.97)
• Europe 75
• America 18
• Asia 7
• Equity 0
• Government bond 100
• Number of countries 8
• only industrialized countries

• New (after 1.1.98)
• Europe 50
• America 30
• Asia 20
• Equity 40
• Government bond 60
• 10 can be invested in corporate bonds
• Number of countries 28
• 22 industrialized countries
• 6 emerging markets (included 1.2.2001)

15
Portfolio models
16
Portfolio models

• The portfolio choice is based on expected return,
  variance (risk) and risk tolerance
• The efficient front and indifference curves are
  based on subjective assessments
• Portfolio choice is sensitive to changes in input
• Investment horizon and availability of data

17
Efficient front
18
Risk preferences II
Return
Risk preferences I
Efficient frontier
Risk
19
(No Transcript)
20
(No Transcript)
21
Time-varying correlation and volatility
22
(No Transcript)
23
Country allocation
24
Country allocation

• As always A trade-off between return and risk
• Return
• assume that expected returns in the long-run are
  equal in all countries
• Risk - a compromise between
• Diversification of market risk
• Diversification of national wealth
• Size of the markets could be a restriction

25
Regional weights mainly determined by GDP-weights

Americas Europe Asia
Old guidelines
18
75
7
Import weights
10
81
9
GDP weights
42
38
20
Market weights - bonds
33
47
20
Market weights - equities
55
32
13
New guidelines
30
50
20
26
Country weights within regions

• Equities Market capitalization weights
• Fixed Income GDP weights

27
Petroleum Fund Benchmark 2000
Equities 40
Fixed income 60
America 30
Europe 50
Asia/Oceania 20
28
Equity portion
29
Determinig the equity portion

• What is the return on equity investment (the
  equity premium puzzle)
• How should we assess the risk associated with
  equity investment
• Is the optimal equity portion independent of the
  investment horizon?

30
CAPM
31
Consumption - CAPM
32
How should we calculate average
return?(illustrated using US data for the period
1926-2000)
33
Rullerende 10 årsperioder
34
Rolling fixed window (10 years)
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
Three explanations for current price levels
40
The equity premium puzzle

• Standard economic models utterly fail to produce
  anything like the historical average stock
  return. After 10 years of intense effort, there
  is a range of drastic modifications to standard
  models that can explain the equity premium.
  However, these models are truly drastic
  modifications they fundamentally change the
  description of the source of risk that commands a
  premium in asset markets. Furthermore, they have
  not yet been tested against the broad range of
  experience of the standard models. These facts
  must mean one of two things. Either the standard
  models are wrong and will change drastically or
  the phenomenon is wrong and will disappear.
• Cochrane (1997)

41
Rolling fixed window (10 years)
42
Correlation between bonds and equities
43
Portfolio model Bonds 5 and equities 8
44
Equity portion and investment horizon

• Merton/Samuelsen
• Random walk
• Constant relative risk aversion
• Equity portion independent of investment horizon

• Mean reversion
• Long horizon - more flexibility in labour supply
• Shortfall preferences
• Equity portion depends on investment horizon

45
(No Transcript)
46
Siegel (1998) US data from 1802 to 1997
47
Equities and long-term risk

• We have a sample of 24 markets for which we have
  data in 1931. Out of these, only seven
  experienced no interruption (the US, Canada, the
  UK, Australia, New Zealand, Sweden and
  Switzerland). Seven experienced a temporary
  suspension of trading, less than a year. The ten
  remaining markets suffered a long-term closure.
• Jorion and
  Goetzmann (1998)

48
Active management
49
The benchmark is the starting point for the
operative management

• The benchmark is defined by the Ministry of
  Finance
• Norges Bank has ambition to outperform the
  benchmark within the risk limits set by MoF
• To main alternatives- index management- active
  management

50
Index management

• Among large international pension funds there is
  a tendency towards using index management for a
  large share of the equity portfolio
• Index management is a standard product
• Competitive gains seem to be achieved with an
  increase in the volume under management
  (economies of scale)
• Management costs are very low
• An important assessment criterion for Norges Bank
  was the managers ability to minimize total
  transaction costs

51
Index management

• Low cost and low risk

52
Active management can be carried out in four
different ways

• by changing the country allocation
• by changing the equity portion
• within the equity portfolio by increasing
  investments in sectors or companies that are
  expected to perform better than others
• within the bond portfolio by changing
  interest-rate risk or credit risk.

53
Active management

• Not many managers beat the benchmarks
  consistently over time
• Risk must be controlled

54
Example Large US pension funds
55
Management costs

• 1999 1998
• Equity management 0.l45 0.082
• Fixed income man. 0.048 0.039
• Total 0.087 0.052
• Measured as a percentage of the average portfolio

56
Risk-adjusted return

• Fama and French (1993) three-factor model
• broad stock index
• excess return on a portfolio of small stocks over
  a portfolio of large stocks
• excess return on a portfolio of high
  book-to-market stocks over a portfolio of low
  book-to-markets stocks
• Carhart (1997) augmented this model to include
• a portfolio of stocks with high return over the
  past months

57
Considerable debate about the interpretation of
this results

• Entirely spurious and the result of data-snooping
• Inability of the the broad stock index to proxy
  for the market portfolio return
• Genuine evidence against CAPM, but not against a
  broader model in which there are multiple risk
  factors
• Mistakes that disappear once market participants
  become aware of them
• Enduring psychological biases that lead investors
  to make irrational forecasts

58
Website www.norges-bank.no