Norges Bank Investment Management
“Exploring Capacity Issues:”
Building a Fund from $30 bn to $324 bn
While Keeping the Alpha Capability
The Q-Group Spring 2007 Seminar,
Georgia, 28 March
Knut N. Kjaer, CEO
Norges Bank Investment Management
WWW.NBIM.NO
1
Norges Bank Investment Management
Agenda
• Background and Facts
• Transition in Two Dimensions:
Strategic: From Oil to International Securities
Financial: From Cash to Equities
• The Alpha Challenge
• Results
• Discussion
The Advantage of Size
Capacity Constraints
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
2
Norges Bank Investment Management
Background and Facts
WWW.NBIM.NO
3
Norges Bank Investment Management
The Ranking of Oil Producers 2005
(mill b/d)
Production
Net export
Saudi Arabia
9.47
Russia
9.44
7.27
USA
Iran
3.91
Saudi Arabia
7.38
6.64
Russia
2.74
Norway
Iran
2.34
Mexico
3.76
Kuwait
2.18
China
3.63
Venezuela
2.12
Canada
3.03
Nigeria
2.09
Norway
2.97
UEA
2.09
Venezuela
2.64
Mexico
Kuwait
2.51
Libya
1.66
1.38
Source: Fact sheet 2006
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
4
Norges Bank Investment Management
The Norwegian Oil Fund and NBIM
• Norway is the world’s 3rd largest exporter of oil
• Since the late 1990’s the bulk of oil revenue has been
saved to the Oil Fund for future generations
• NBIM was established in 1998 and manages the Oil
Fund and the bulk of the Central Bank’s currency
reserves
• The Ministry of Finance owns the Fund and sets the
benchmark
• Asset under management in NBIM increased from
USD 30 bn at inception in 1998 to USD 324 bn at
year end 2006
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
5
Norges Bank Investment Management
Projected Growth of the Government
Pension Fund - Global
500
450
400
350
Bn. USD
300
250
200
150
100
50
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Source: National Budget 2007, October 2006
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
6
Norges Bank Investment Management
A Possible Scenario
Ow nership in E quity Markets
Percent of FTSE Market Cap
1.40 %
1.20 %
1.00 %
Europe
0.80 %
America
Asia
0.60 %
T otal
0.40 %
0.20 %
0.00 %
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
7
Norges Bank Investment Management
Norges Bank Investment
Management (NBIM)
• NBIM was established in 1998 as a separate wing of
Norges Bank. NBIM is organised as a business unit.
Offices in Oslo, London and New York
• Funds managed by NBIM as of 31 Dec 2006 (billion
USD):
The Pension Fund - Global
The Foreign Exchange Reserves
The Petroleum Insurance Fund
Total
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
286
36
2
324
KNK 260307
WWW.NBIM.NO
8
Norges Bank Investment Management
NBIM / Organisation Chart
Executive
Executive Director
Director
Corporate Governance
Knut
Knut N.
N. Kjær
Kjær
Dag
Dag Løtveit
Løtveit
Bjørn
Bjørn Egge
Egge
Risk, Performance &
Accounting
Equity
Equity
Fixed
Fixed Income
Income
Investments
Investments Operations
Operations
Henrik P. Syse
Investments
Investments
Yngve
Yngve Slyngstad
Slyngstad
Operations
Operations
Ilse Bache
Stephen
Stephen A.
A. Hirsch
Hirsch
IT Infrastructure
Ilse Bache
Legal / Finance / HR
Offices in Oslo, London and New York
Marius N. Haug
Bjørn Taraldsen
Kristin S. Kleven
Number of Employees as of 31.12.2006: 132
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
9
Norges Bank Investment Management
Benchmark for the Pension Fund-Global
Strategic benchmark:
FI - Americas
21 %
60 % Fixed Income
EQ - Europe
20 %
FI - Asia / Pacific
3%
40 % Equities
“Smart” rebalancing:
• Monthly inflows
EQ - Asia /
Pacific
6%
EQ - Americas /
Africa
14 %
• Asset class / region
with largest
negative deviation
from benchmark
FI - Europe
36 %
Equity index:
Fixed income index:
FTSE All World Index
Large & Mid Cap
Approx. 2 400 equities
Lehman Brothers Global Aggregate
Government / Agency / Corporate / Securitized
Approx. 9 300 bonds
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
10
Norges Bank Investment Management
Transition in Two Dimensions:
Strategic: Oil to International
Financial Securities
Financial: Cash to Equities
and Fixed Income Securities
WWW.NBIM.NO
11
Norges Bank Investment Management
Petroleum Wealth - Autumn 2006
600
500
USD Bn
400
300
200
100
0
Petroleum In
Ground
Oil fund
Equities
Bonds
Petroleum in the ground: Net present value of the government’s future cash flows from oil activities.
Estimated at NOK 3660 billion at constant 2007 values (Source: National Budget 2007)
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
12
Norges Bank Investment Management
Petroleum Wealth - Value at Risk
200
180
160
USD Bn
140
120
100
80
60
40
20
0
Petroleum In
Ground
Oil fund
Equities
Bonds
Value at Risk is defined as one standard deviation in this context.
The actual values will fluctuate outside the bands in one out of three years
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
13
Norges Bank Investment Management
Real Global Market Returns 1900 – 2005
1000.00
100.00
Equities
Bonds
Money market
Oil
376
10.00
6
3
1.00
18
99
19
06
19
13
19
20
19
27
19
34
19
41
19
48
19
55
19
62
19
69
19
76
19
83
19
90
19
97
20
04
2
0.10
Source: Dimson, Marsh & Staunton, Triumph of the Optimists, 2002, with updates
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
14
Norges Bank Investment Management
Portfolios of Oil and Equities
7.00 %
6.00 %
100% equities
Real Return
5.00 %
4.00 %
3.00 %
2.00 %
1.00 %
100% oil
0.00 %
0.00 % 5.00 % 10.00 % 15.00 % 20.00 % 25.00 % 30.00 % 35.00 %
Standard deviation
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
15
Norges Bank Investment Management
Portfolios of Equities, Oil and Bonds
7.00 %
Real Return
6.00 %
100% equities
5.00 %
4.00 %
3.00 %
2.00 %
1.00 %
100% oil
100% bonds
0.00 %
0.00 % 5.00 % 10.00
%
15.00
%
20.00
%
25.00
%
30.00
%
35.00
%
Standard deviation
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
16
Norges Bank Investment Management
Strategy: Transition From Petroleum
• The Oil Fund’s key function is to diversify the
Petroleum Wealth into a broad portfolio of
international securities
• The transition takes down the expected risk
significantly and increases the expected return
• The Fund makes the income stream from the
non-renewable resources permanent
• The intention is to spend only the (expected) real
return at the annual public budgets, thus
preserving the capital of the fund for all future
generations
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
17
Norges Bank Investment Management
Implementation:Transition From Cash
Inflow LHS
Inflow as % of MV at the beg. of year RHS
50
70
45
60
50
35
30
40
25
30
20
15
20
Inflow in % of MV
Inflow in USD (bill)
40
10
10
5
0
0
1998
1999
2000
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
2001
2002
2003
2004
KNK 260307
2005
WWW.NBIM.NO
2006
18
Norges Bank Investment Management
Trading Efficiency
• Given the expected large inflows and expected large size,
NBIM knew early on that efficient trading was critical to
success
• If trading was not efficient, NBIM would easily spend any
alpha generated and more
• System advances over the last 9 years along with
management willing to focus on trading has enabled NBIM
to build the infrastructure and gain the expertise to
compete in a competitive part of the marketplace
• Data is the key to improving trading efficiency. Tracking
all data components of the trading cycle is extremely
difficult but more than pays for the effort
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
19
Norges Bank Investment Management
Billions
Total Equity Trade Volume 98 - 06
25
20
USD
15
10
5
S
ep
98
D
ec
98
M
ar
99
Ju
n9
S 9
ep
99
D
ec
99
M
ar
00
Ju
n0
0
S
ep
00
D
ec
00
M
ar
01
Ju
n0
S 1
ep
01
D
ec
01
M
ar
02
Ju
n0
S 2
ep
02
D
ec
02
M
ar
03
Ju
n0
S 3
ep
03
D
ec
03
M
ar
04
Ju
n0
S 4
ep
04
D
ec
04
M
ar
05
Ju
n0
S 5
ep
05
D
ec
05
M
ar
06
Ju
n0
S 6
ep
06
D
ec
06
-
Singlestock (phys)
Program
Futures
Equity Swaps
FX
Equity trading volume tripled from the end of 2004 to the summer of 2006
and is expected to increase for the foreseeable future
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
20
Norges Bank Investment Management
How is Volume Distributed (%)
100 %
80 %
60 %
Electronic
Agency
Natural
Capital
40 %
20 %
N
ov
06
Ju
l0
6
S
ep
06
ai
06
M
ar
06
M
N
ov
05
Ja
n0
6
Ju
l0
5
S
ep
05
ay
05
M
ar
05
M
N
ov
04
Ja
n0
5
Ju
l0
4
S
ep
04
ay
04
M
ar
04
M
Ja
n0
4
0%
Natural flows and capital are the sweet spots for broker interaction. This is an area of focus
for us. Tracking costs told us that agency is not optimal relative to other alternatives.
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
21
Norges Bank Investment Management
Electronic Trading Volumes 2004 - 2006
70 %
Billions
6
60 %
5
50 %
4
40 %
Value
Percent
30 %
3
20 %
2
10 %
1
0%
-10 %
Ja
n0
M 4
ar
M 04
ay
04
Ju
l
S 04
ep
0
N 4
ov
0
Ja 4
n0
M 5
ar
M 05
ay
0
Ju 5
l
S 05
ep
0
N 5
ov
0
Ja 5
n0
M 6
ar
M 06
ay
0
Ju 6
l
S 06
ep
0
N 6
ov
06
-
Being nimble is important in the fast changing trading environment. We see electronic
trading as an important tool required to improve efficiency and lower total trading costs.
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
22
Norges Bank Investment Management
Billions
Volume By Instrument 2006
45
40
35
30
25
20
15
10
5
Physical
Swap
Elec. Phys
Elec. Swap
Swaps have there place in an investment strategy, but are expensive to
trade given constraints in execution. Tracking costs enabled us to see
the costs confirming what the traders were telling us.
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
23
Norges Bank Investment Management
Cutting the Trading Cost
Basis points
Total Single Stock Trading Cost
45
40
35
30
25
20
15
10
5
0
2003
Tax&Charges
Commission
Shortfall
2004
2005
2006
Diligent management enabled a significant decrease in
trading costs despite strong growth in volume and average
ticket size
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
24
Norges Bank Investment Management
Trading - Conclusions
• Actively managing and monitoring all trading costs is
important
• Lowering commissions is only a small start
• The ability to influence the trading mix has changed
dramatically over the last 5 years giving the buyside the
tools to actively manage total trading costs
• The key to lowering trading costs is Data and access to
Data. Good Data is important not only for improving
trading costs but should also be extended to improving
portfolio management decisions
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
25
Norges Bank Investment Management
The α Challenge
WWW.NBIM.NO
26
Norges Bank Investment Management
Challenges in Alpha Management
• Create excess return against the benchmark
• targeting 0.25% over rolling 3 year horizons
• with costs less than 0.1% ex performance fees
• and annual inflow at USD 20 – 50 bn / year
Excess Return, Basis Points
140
120
100
80
60
40
20
0
1998
1999
2000
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
2001
2002
2003
2004
KNK 260307
2005
2006
WWW.NBIM.NO
27
Norges Bank Investment Management
Move to More Scalable Alpha-Strategies
May Imply Lower Information Ratio
• NBIM prefers relative value and fundamental
strategies
• Our IR is now 1.22 However, we may be forced to
move to more factor based strategies
Management strategy
Tactical
allocation
Factor-based
strategies
Fundamental
strategies
Relative
value
Analytical ability
+
+
++
+++
Number of independent positions
-
-
+++
++
+++
+
++
+
++
+++
Implementation costs
Experience
-
Expenses
Low
Moderate
High
High
Expected information ratio
Low
Moderate
High
High
http://www.norges-bank.no/english/petroleum_fund/articles/2004/highest_excess_return/
KNK 260307
WWW.NBIM.NO
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
28
Norges Bank Investment Management
The Principles behind Our Active
Management
• Very high respect for the market. Applying financial
theory. Humility and discipline
• Active management only where there may be
some less market efficiency and where we
find/develop managers with unique expertise
• Specialization. More than 100 sub-portfolios
• Extensive delegation. No committee structure
• Incentives linked to performance
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
29
Norges Bank Investment Management
"Financial Market Theory for Cautious
Investors"*
"The possibility of achieving an above average
return lies in being able to identify cases of
incomplete market equilibrium. To be able to
detect what is abnormal, it is necessary to know
what is normal."*
* Jan Mossin, Markedseffisiens (Market efficiency),
TANO, 1986
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
30
Norges Bank Investment Management
Textbook Approach to Active Portfolio
Management
• Active management is a process. Active
management begins with raw information, refines into
forecasts, and then optimally and efficiently
constructs portfolios balancing those forecasts of
return against risk
• Active management is forecasting, and a key to
active management performance is superior
information
• Active managers should forecast as often as possible
(the fundamental law)
• Mathematics cannot overcome ignorance/lack of
information
R.C. Grinold & R.N. Kahn, Active Portfolio Management, 1994, 1999
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
31
Norges Bank Investment Management
”The Fundamental Law of
Active Management”
• Information Ratio = Return / Risk
IR = IC⋅ BR
Information Coefficient (IC) = corr[α, θ]
α = expected (ex ante) return
θ = actual (ex post) return
corr[α, θ] = correlation between α and θ
BReadth (BR) = number of independent positions
• The Challenge: Improve IC (hit ratio)
• and/or increase breadth by taking many independent
positions and trade often
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
32
Norges Bank Investment Management
Neccessary Conditions for Creating Alpa
• Talented people
• Motivated and challenged in a
rational, efficient and supportive
structure
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
33
Norges Bank Investment Management
Rational and Supportive Structure
•
•
•
•
•
•
•
Clear investment philosophy
Various investment strategies
Specialization
Large number of independent positions
Good access to information
Alpha separated from Beta
Low costs in trading and transitions
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
34
Norges Bank Investment Management
Specialization and Independence
• Specialization, focus
Very clearly defined mandates. Specialist
expertise to provide the competitive edge needed
in a near efficient market
• Independence, low correlation
Different types of investment strategies in various
units; a large number of individual investments.
No committee decisions; no predefined
processes
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
35
Norges Bank Investment Management
Example: Risk Allocation Fixed Income
Correlation matrics, profit units
6%
18 %
Average
14 %
Ext. mgrs
RV
GV
12 %
F.I.
QP
EI
Cred
AO
0%
EI
cred b
cred a
rv
gv
qp
ao
ext a
EI
cred b
1
-0.754115
0.152501
0.723803
0.180472
0.457215
0.387236
-0.271266
1
-0.208785
-0.822739
0.44208
-0.328319
-0.560345
-0.38964
cred a
rv
1
0.296542
0.209786
0.136915
0.674421
0.323637
gv
qp
ao
ext a
1
0.2730606
1
0.0839749 0.092456
1
0.6618643 0.036886 -0.063968
1
-0.0025061 -0.612593 -0.359838 0.125951
1
21 %
22 %
1%
0.02
49 %
Oslo c re
RV
14 %
Oslo gov
NYc re
NYgov
Pos.
43 %
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
0 1%
1%%
2%
1%
11%%
10%
1%
1%%
2 %1 %
2%
1%
%
011%
3%
2
%
1%
1%
2%
2%
0%
%
11%
2%
%
001 %
2%
%
0
%
1%
1%
11%
00 %
%
%
2%
3%
0%
2%
2%
11%
%
%
11%
1%
%
01
1%
%
2%
0%
1%
2%
1%
1%
2%
1%
2%
10%%
1%
2%
2%
1%
1%%
0 1%%
1
0 1%%
2%
1%
%
1%
100%%
2%
0%11%
%%
1
0 1%
%%
1
000%%
1
%
1%%
1%
1
%
1%
1%
1 %1 %
0 1%%
KNK 260307
WWW.NBIM.NO
36
Norges Bank Investment Management
Talented People
• NBIM learning from external and internal
management: Alpha creation is mostly about the
property of the people. Only special talents
create alpha.
• The key issues is then: How to recruit, develop,
motivate and retain talents
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
37
Norges Bank Investment Management
Empowerment and Ownership
•
•
Individual investment mandates and considerable freedom to
develop own investment style, methodology and tools
Individual incentive structure
• clearly defined and measured
• ‘managing own money’
•
Ownership in all other processes
• operations included in the business units
• division of responsibilities, defined work task
•
Independence in team structure
•
•
•
•
optimal size for flexibility and communication
accountability and visibility without ‘atomization’
blending competition and support
avoiding ‘group think’
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
38
Norges Bank Investment Management
Performance
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
39
Norges Bank Investment Management
Average Annual Net Real Return Since 31/12 1996
10.0 %
9.0 %
Net Real Return on Actual Portfolio
8.0 %
Net Real Return on Benchmark
7.0 %
6.0 %
• Average annual net
real return (adjusted
for management
costs and inflation) is
4.58 %
5.0 %
• Annual real return on
the benchmark is
4.12 %
4.0 %
3.0 %
2.0 %
1.0 %
0.0 %
1999
2000
2001
2002
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
2003
2004
2005
2006
KNK 260307
WWW.NBIM.NO
40
Norges Bank Investment Management
Excess Return Last Three Years
Monthly excess return
Three years rolling excess return
0.8 %
0.7 %
0.7 %
0.6 %
0.6 %
0.5 %
0.5 %
0.4 %
0.4 %
0.3 %
0.3 %
0.2 %
0.2 %
0.1 %
0.1 %
0.0 %
0.0 %
-0.1 %
-0.1 %
-0.2 %
-0.2 %
Ja
n04
M
ar
-0
4
M
ay
-0
4
Ju
l-0
4
Se
p04
N
ov
-0
4
Ja
n05
M
ar
-0
M 5
ay
-0
5
Ju
l-0
5
Se
p05
N
ov
-0
5
Ja
n06
M
ar
-0
6
M
ay
-0
6
Ju
l-0
6
Se
p06
N
ov
-0
6
0.8 %
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
41
Norges Bank Investment Management
Annualised Contributions to Gross Excess
Return. 2004-2006. Percentage Points
External
Internal
Total Excess return
management management
in each asset
class
Equity management
0.18
0.20
0.38
0.95
Fixed income
management
0.04
0.16
0.20
0.32
Total
0.22
0.36
0.58
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
42
Norges Bank Investment Management
Information Ratios 2004 – 2006
External
Internal
Total
management management
Equity management
0.48
1.44
1.08
Fixed income
management
1.69
2.58
2.77
Total
0.61
2.55
1.60
The IR is a measure of risk-adjusted return and is an indicator of
skills in investment management. It is calculated as the ratio of
excess return to the actual relative market risk to which the portfolio
has been exposed. The IR indicates how much excess return is
achieved for each unit of risk.
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
43
Norges Bank Investment Management
Quarterly Excess Return Since 1998
1.0 %
0.8 %
0.6 %
0.4 %
0.2 %
0.0 %
-0.2 %
Q
3
06
06
Q
1
05
Q
3
05
Q
1
Q
3
04
04
Q
1
Q
3
03
03
Q
1
02
Q
3
02
Q
1
Q
3
01
01
Q
1
Q
3
00
00
Q
1
99
Q
3
99
Q
1
98
3
Q
Q
1
98
-0.4 %
Average annual excess return since 1998: 0.48 percentage points
Cumulative excess return NOK 28,9 bill.
Information ratio 1.22
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
44
Norges Bank Investment Management
Excess Return and Contribution to
Total Risk From Active Management
Excess return and return on the
benchmark portfolio
Benchmark
•
Excess return
Risk contribution due to
active management
Risk due to active management
Benchmark risk
Active management has increased the return with no
increased risk
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
45
Norges Bank Investment Management
NBIM Critical Success Factors
• Time is spent on professional challenges – not marketing
• Empowerment of individuals (mandates and resources)
• Delegated power matters more than placement in org
charts
• No committee governance with unclear responsibilities
• Meritocracy – rewards and positions are based on
performance
• Ability to change leadership and structure when
performance or new challenges requires it
• Incentive level and structure is competitive and fosters the
targeted risk-taking and behavior
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
46
Norges Bank Investment Management
Discussion
The Advantage of Size
Capacity Constraints
WWW.NBIM.NO
47
Norges Bank Investment Management
The Advantage of Size
•
•
•
•
Economies of scale in beta management and
transition
Easy to scale from a system and -competence base
with fixed costs. Can afford large investments in
systems
Low commissions. Our flow has value
Low trading costs gives more opportunity for
applying quant strategies
Size and organizational skills gives power to demand
and build new products with service providers, like:
Custodians, prime brokers, prime custody, backoffice
Size advantage in securities lending
24/7 systems and organization
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
48
Norges Bank Investment Management
The Advantage of Size (2)
•
Access to company information; scale advantage in
some research, easier access to the global talent
pool
•
Some lower fees in external management
•
Top credit rating; balance sheet advantages
•
Some studies indicates a positive relation between
alpha and size (next two slides)
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
49
Norges Bank Investment Management
Large funds did better than small funds.
• On average, the Net Value Added was
higher by 28 bps for a $1 billion fund
Net Value Added versus Size (log10)
compared with a $100 million fund.
Log (Size) Coefficient
1991-2005
All Funds
0.28%
"t" statistic
4.3
"T" statistic values in excess of the absolute values of 2.0, 1.6 and 1.3 are significant at the 95%,
90% and 80% confidence levels, respectively.
• Size predicted Net Value Added since
larger funds have lower costs and have
higher weightings in specialised, high
performing asset classes. The belief is
supported by the fact that the size factor
was reduced significantly when the
following factors were added: Total Cost,
US Small Cap Stock, and Private Equity.
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
Net Value Added vs. % Private Equity, % US Small Cap
Stock, Total Cost (bps) and Size (log10)
1991-2005
All Funds
Coefficient
"t" statistic
Private Equity
4.933%
2.8
US Small Cap Stock
2.700%
3.1
Total Cost
-0.004%
-1.4
Log (Size)
0.155%
1.9
KNK 260307
WWW.NBIM.NO
50
Norges Bank Investment Management
Big is Better?
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
51
Norges Bank Investment Management
Capacity Constraints
• Market impact. (Trading costs may eat to much of the
alpha return. But as explained, we try to build trading into
an advantage)
• May have constraints in external active management
• Lack of scalability in some of the internal alpha strategies.
May be difficult to keep the return targets?
• May a fat and complacent organization follow with
increased size?
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
52
Norges Bank Investment Management
Discussion
• Yes, it is possible to become to big, but we are still not
there
• When there is one owner of the assets; by splitting in
different portfolios, there is a risk for ending up with costly
index management (overlap that average out active
positions)
• Better way of splitting up: Keep the beta and transition
machines, set up different & focused alpha satellites
• Avoiding being fat & complacent: Line structure with real
delegation, results having effect on compensation and
position for everybody, outsourcing of all activities that are
not in the core business, individual measurement
The Institute for Quantitative Research in Finance, Spring 2007 Seminar
KNK 260307
WWW.NBIM.NO
53