Asset Owners and Investing in Diversity: Intention versus ...
Unconstrained Multi-Asset Investing
Transcript of Unconstrained Multi-Asset Investing
Unconstrained Multi-Asset InvestingSeeking diversification through efficient portfolio construction using
cash-based and derivative instruments
22.01.13
“Uncertainty is the only certainty there is, and knowing how
to live with insecurity is the only security”
John Allen PaulosProfessor of Mathematics at Temple University, Philadelphia, USA
Background
• Working party paper prepared in 2012– Presented at Risk and Investment Management conference, June 2012
• This paper formed core part of sessional paper to be presented to Institute and Faculty audiences in London and Edinburgh in 2013
• Themes– Measurement of portfolio diversification– Diversification benefits of unconstrained multi‐asset investment– Use of derivatives to implement investment strategies– Risk measurement of an unconstrained multi‐asset portfolio
• Question– Where does the actuarial profession believe it has most to add in the realm
of portfolio construction?
Traditional approach fails to deliver consistent returns
Annualised rolling 3 year returns
-15.0%
-10.0%
-5.0%
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
1989 1992 1995 1998 2001 2004 2007 2010
Ann
ualis
ed %
retu
rn
Typical return expectation
Annualised rolling 3 year returns
-15.0%
-10.0%
-5.0%
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
1989 1992 1995 1998 2001 2004 2007 2010
Ann
ualis
ed %
retu
rn
Typical return expectation
Annualised rolling 3 year returns – Standard Life Balanced Fund
Uncertainty is the only certainty there isSource: Standard Life Investments
Source: IPD UK Monthly Property Index, All Property; Federal Reserve Trade-Weighted Exchange Value of US Dollar vs 6 Countries; Dow Jones UBS – Commodity Index; Barclays Capital Global Corporate Index, Excess Returns; Barclays Capital Emerging Markets Index, Excess Returns; Barclays Capital US High Yield Index, Excess Returns; Standard Life Investments, 31 December 2011
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Property Currency Basket Commodities
IG Corporate
Bonds
EmergingMarket Bonds
High Yield Bonds
2000-2007 2008-2011
Correlation of Global Equities with other market sectors
Finding diversity is increasingly difficult
Portfolio diversification measures
• No single measure of diversification
• Growing academic literature on measures of systemic risk
• Need for some definite measures of diversification:1. Assess current state of the world2. Observe how it is changing through time
• We want to consider measures for:1. Diversification in a given portfolio2. Diversification potential in an investment universe
Introduction to principal components analysis
•Analysis is based on the weighted covariance matrix:1.Covariance matrix of asset returns: 2. Portfolio weights
Introduction: Eigenvalues
• For every linear combination of assets, represented by a vector v, we can compute the size of the projected variance
• The first eigenvalue is the projected variance associated to the vector for which this projection is maximised. This is typically interpreted as the market
• Later eigenvalues are associated with vectors that maximise this expression under the constraint that they are orthogonal to vectors computed earlier
• This process creates an ensemble of uncorrelated strategies called principal portfolios. This process is called principal component analysis
Effective Number of Assets
• We define the effective number of assetsbased on the entropy of the eigenvalues of the weighted covariance matrix
• Entropy H(p) is a measure of the information in the probability distribution p. It has been used widely in physics, information theory, signal processing, etc
• Hence the effective number of assets is a measure of the concentration of eigenvalues
• It takes values between 1 and N
Effective number of assets:
Diversification within a given portfolio
Diversification Potential
• The diversification potential is the maximum over the effective number of assets as a function of the portfolio weights
• Well defined optimisation problem since the measure varies smoothly with the portfolio weights w
• Because of symmetries, we are looking for the local minimum near the weights that define an equal weighting in terms of risk
• Can be solved with a local optimiser (Newton–Raphson type)
Diversification potential:
The maximum amount of effective assets that any given portfolio universe could have
Effective number of stocks in FTSE 100Market cap weighted
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
Dec 00 Apr 02 Sep 03 Jan 05 Jun 06 Oct 07 Mar 09 Jul 10 Nov 11
Effe
ctiv
e nu
mbe
r of s
tock
s
3000
3500
4000
4500
5000
5500
6000
6500
7000
FTSE
100
inde
x
FTSE100 index Effective number of stocks; cap weighted (lhs)
Effective number of stocks is low compared to universe
Effective number of stocks in FTSE 100Equally weighted and maximum potential
0
10
20
30
40
50
60
Dec 00
Apr 02
Sep 03
Jan 05
Jun 06
Oct 07
Mar 09
Jul 10
Nov 11
Effe
ctiv
e nu
mbe
r of s
tock
s
3000
3500
4000
4500
5000
5500
6000
6500
7000
FTS
E 1
00 in
dex
FTSE100 index Effective number of stocks; cap weighted (lhs)
Effective number of stocks; equally weighted (lhs) Diversification Potential (lhs)
Maximum potential is three times that of a market cap weighted approach
Effective number of assets in UK pension portfolio
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
Mar 04 Aug 05 Dec 06 May 08 Sep 09 Jan 11 Jun 12
Effe
ctiv
e st
rate
gies
100
110
120
130
140
150
160
170
180
190
Bala
nced
Fun
d to
tal r
etur
n in
dex
Total return index Effective number of strategies (lhs) Diversification Potential (lhs)
Low number of effective assets reflects dominance of equity and interest rate risk
Defining the investment challenge
• “How can I build a better diversified investment portfolio but without impacting long term return expectations?”
• Allow a broader global investment universe currencies asset classes and sectors yield curves across different geographies volatility
• A number of these strategies (such as specific currency pairs) will necessitate the use of derivative contracts
• Allow unconstrained dynamic asset allocation offers access to the full diversification potential of the chosen investment universe
Money making strategies and diversification
Strategy correlation with global equity
-0.60
-0.40
-0.20
-
0.20
0.40
0.60
0.80
1.00
Cor
rela
tion
coef
ficie
nt
Least diversification
Most diversification
UK EquityHigh yield creditRussian Equity
Mexican 10 year bonds
Global inflation-linked bondsUK corporate bonds
Europe bond yield steepenerUS Dollar v EuroUS equity large v small cap
Strategy correlation with global equity
-0.60
-0.40
-0.20
-
0.20
0.40
0.60
0.80
1.00
Cor
rela
tion
coef
ficie
nt
Least diversification
Most diversification
UK EquityHigh yield creditRussian Equity
Mexican 10 year bonds
Global inflation-linked bondsUK corporate bonds
Europe bond yield steepenerUS Dollar v EuroUS equity large v small cap
Derivative-based strategies can offer the most diversification potentialSource: Standard Life Investments, July 2012
Portfolio Themes – choice of implementations
• Profit opportunities in areas of growth Implementation choices• US equity Physical or equity futures/options• US technology equity vs US small cap equity Physical and / or equity futures/options • Korean equity Physical or equity futures/options
• Corporate balance sheet strength • US large cap equity vs US small cap equity Physical and / or equity futures/options• European high dividend‐yield equity Physical or equity futures/options• Investment grade and high yield credit Physical or index CDS contracts
• Fiscal advantages of resource‐based economies • Mexican government bonds Physical bonds• Brazilian Real vs Australian Dollar Currency forward• Russian equity Physical• US Dollar vs Canadian Dollar Currency forward
• Challenging Eurozone resolution• US Dollar vs Euro Currency forward
Wide range of strategies that can make money
Effective factors for unconstrained multi‐strategy portfolio
Portfolio’s greater number of ‘moving parts’ allows for more return consistency
0
2
4
6
8
10
12
14
Mar 04 Aug 05 Dec 06 May 08 Sep 09 Jan 11 Jun 12
Effe
ctiv
e st
rate
gies
110
120
130
140
150
160
170
180
190
200
210
220
Glo
bal e
quity
inde
x
Global Equity index Effective number of strategies (lhs) UK balanced
Portfolio risk managementCombining physical and derivative holdings
• Risk‐based portfolio construction Size of each position Volatility of each position
• Ensures consistency Some assets physically invested others by derivatives Different asset classes have different volatility
• Stand‐alone riskmeasures the risk any of any one investment strategy US Equity has volatility: 23.2% Portfolio allocates 10% investment to US equities Stand‐alone risk of US equities to portfolio is 2.3%
Source: Standard Life Investments, 31 March 2012
Absolute Return investing Risk‐based portfolio construction
0%5%
10%15%20%25%
30%35%40%
45%50%
Nominal holding Strategy volatility
0%5%
10%15%20%25%
30%35%40%
45%50%
Nominal holding Strategy volatility
3%
Rus
sian
E
quity
0%
1%
1%
2%
2%
Stan
d-al
one
risk
US
equ
ity
Glo
bal h
igh
yiel
d cr
edit
Kor
ea E
quity
vs
Eur
opea
n E
quity
US
Dol
lar v
s E
uro
Glo
bal i
ndex
-lin
ked
bond
s
UK
cor
pora
te
bond
s
Aus
tralia
n sh
ort-t
erm
in
tere
st ra
tes
Sw
edis
h vs
G
erm
an s
hort-
term
rate
s
3%
Rus
sian
E
quity
0%
1%
1%
2%
2%
Stan
d-al
one
risk
US
equ
ity
Glo
bal h
igh
yiel
d cr
edit
Kor
ea E
quity
vs
Eur
opea
n E
quity
US
Dol
lar v
s E
uro
Glo
bal i
ndex
-lin
ked
bond
s
UK
cor
pora
te
bond
s
Aus
tralia
n sh
ort-t
erm
in
tere
st ra
tes
Sw
edis
h vs
G
erm
an s
hort-
term
rate
s
0%
1%
1%
2%
2%
Stan
d-al
one
risk
US
equ
ity
Glo
bal h
igh
yiel
d cr
edit
Kor
ea E
quity
vs
Eur
opea
n E
quity
US
Dol
lar v
s E
uro
Glo
bal i
ndex
-lin
ked
bond
s
UK
cor
pora
te
bond
s
Aus
tralia
n sh
ort-t
erm
in
tere
st ra
tes
Sw
edis
h vs
G
erm
an s
hort-
term
rate
s
Stand-alone risk aligns size of position with its volatilitySource: Standard Life Investments, sample absolute return portfolio, March 2012
Risk‐based portfolio constructionSeek a number of different risks
6.5%
15.9%
0%
5%
10%
15%
20%
25%
US
Equ
ity
Rus
sian
Equ
ity
US
Lon
g B
ond
Yie
lds
(For
war
d S
tart)
Rel
ativ
e V
aria
nce
Inco
me
Hig
h Y
ield
Cre
dit
Fina
ncia
l Sec
tor v
s B
road
Cre
dit
Kor
ean
Equ
ity v
s E
urop
ean
Equ
ity
Long
US
Dol
lar v
s E
uro
Mex
ican
Gov
ernm
ent B
onds
vs
Eur
o
Bro
ad v
s Fi
nanc
ial S
ecto
r Equ
ity
US
Equ
ity L
arge
vs
US
Equ
ity S
mal
l Cap
Glo
bal E
quity
Eur
opea
n E
quity
Glo
bal I
nfla
tion-
Link
ed B
onds
Long
US
Dol
lar v
s C
anad
ian
Dol
lar
US
Equ
ity T
echn
olog
y S
ecto
r vs
US
Equ
ity S
mal
l Cap
Long
US
Dol
lar v
s Ja
pane
se Y
en
UK
Inve
stm
ent G
rade
Cre
dit
UK
Infla
tion
Sec
urity
Sel
ectio
n - a
lpha
Long
Nor
weg
ian
Kro
ne v
s E
uro
FX H
edgi
ng
Eur
opea
n In
vest
men
t Gra
de C
redi
t
Chi
na E
quity
vs
UK
Equ
ity V
olat
ility
Aus
tralia
n S
hort-
term
Inte
rest
Rat
es
Sw
edis
h vs
Ger
man
Sho
rt-te
rm In
tere
st R
ates
UK
Equ
ity
Eur
opea
n B
ond
Yie
ld S
teep
ener
Div
ersi
ficat
ion
Exp
ecte
d V
olat
ility
Vol
atili
ty %
Diversification
Expected volatility
Source: Standard Life Investments, APT system, March 2012
Total aggregated risk 22.4% (total nominal exposure 220%)
Leverage and portfolio risk
Portfolio Leverage factor Strategy volatility Portfolio risk
US equity 1.0 23.2% 23.2%
UK corporate bonds 3.0 7.7% 23.1%
Australian short-term interest rates 10.0 0.9% 9.0%
Portfolio Leverage factor Strategy volatility Portfolio risk
US equity 1.0 23.2% 23.2%
UK corporate bonds 3.0 7.7% 23.1%
Australian short-term interest rates 10.0 0.9% 9.0%
• In theory lowest risk portfolio is the most leveraged
• However, actual outcome is driven by the behaviour of only ONE factor
• Leveraged credit is a strategy that failed in the GFC
Historical scenario analysis
-28 -26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18
Bank Meltdown 2008 ( September 12 - October 15, 2008)
Subprime Debacle 2007 (July 15 - August 15, 2007)
Emerging Market Sell-Off 2006 (May 1 - June 8,2006)
Bond Sell-Off (June 14 - July 31, 2003)
Bond Rally (May 1 - June 13, 2003)
Gulf War 2 (March 1-23,2003)
Equity Rally (October 10 - November 27,2002)
Equity Sell-Off (August 23 - October 9, 2002)
Sept 11th
Tech Wreck (April 7-14, 2000)
Russian/LTCM
Asian Crisis 1997
Mexican Crisis 1995
Rate Rise 94
Gulf War 1990
Black Monday 1987
% Move
Multi-asset Portfolio MSCI World (GBP) move over same period
-28 -26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18
Bank Meltdown 2008 ( September 12 - October 15, 2008)
Subprime Debacle 2007 (July 15 - August 15, 2007)
Emerging Market Sell-Off 2006 (May 1 - June 8,2006)
Bond Sell-Off (June 14 - July 31, 2003)
Bond Rally (May 1 - June 13, 2003)
Gulf War 2 (March 1-23,2003)
Equity Rally (October 10 - November 27,2002)
Equity Sell-Off (August 23 - October 9, 2002)
Sept 11th
Tech Wreck (April 7-14, 2000)
Russian/LTCM
Asian Crisis 1997
Mexican Crisis 1995
Rate Rise 94
Gulf War 1990
Black Monday 1987
% Move
Multi-asset Portfolio MSCI World (GBP) move over same period
Source: UK £ GARS Regulated Unit Trust, RiskMetrics, 30/09/2012
Analyse actual return experience
0
20
40
60
80
100
120
140
160
Weekly % return
No
of o
bser
vatio
ns
Multi-asset portfolioMSCI World
WORSTthree equity weeks
Equity mulit-asset-16.6% -5.5%-9.3% -3.1%-8.3% -3.0%
BESTthree equity weeks
Equity multi-asset+9.9% +4.3%+8.3% +3.1%+7.6% +3.1%
-7 -6 -5 -4 -3 -2 -1 2 3 4 5 6 7 810 9 10 1-8-9-100
20
40
60
80
100
120
140
160
Weekly % return
No
of o
bser
vatio
ns
Multi-asset portfolioMSCI World
WORSTthree equity weeks
Equity mulit-asset-16.6% -5.5%-9.3% -3.1%-8.3% -3.0%
BESTthree equity weeks
Equity multi-asset+9.9% +4.3%+8.3% +3.1%+7.6% +3.1%
-7 -6 -5 -4 -3 -2 -1 2 3 4 5 6 7 810 9 10 1-8-9-10
Distribution of weekly returns of unconstrained multi-asset portfolio v MSCI Global Equities
Source: Standard Life Investments, net performance from 12/06/2006 to 30/11/2012Portfolio performance is based on the £, institutional pooled pension portfolio* Source: Thomson Datastream, MSCI World (£) (net of tracker fund fee)
Summary
• Traditional multi‐asset funds offer insufficient diversification opportunities
• Derivatives significantly broaden the range of investment strategies that a portfolio can utilise
• Combining traditional assets with derivative‐implemented strategies produces more risk efficient portfolios increase the likelihood of ‘good’ outcomes for your portfolio
• Actuaries well placed to add value in understanding and communicating investment diversification
Appendices
Absorption Ratio: related measures
• Several related indicators based on correlations and principal components exist: Billio et. al. (2010): Measuring Systemic Risk in the Finance and Insurance Sectors Golts et. al. (2010): Diversification, Volatilities, and the Supercurrency Fenn et. al. (2010): Temporal Evolution of Financial Market Correlations Meucci (2009): Managing diversification
• There are several ways of measuring the concentration of eigenvalues: Absorption Ratio Herfindahl‐Hirschman index (HHI) Gini coefficient Entropy
• But which one is mathematically justified? What properties do they satisfy?
Properties of ‘effective factors’
• Invariantmeasure• Depends continuously on the portfolio weights• High level of stability• Varies intuitively under addition of extra assets• Interpretable as the number of independent risk factors required to capture the risks
of the underlying asset positions without losing information• Rigorous mathematical interpretation:
Known in the signal processing literature as the effective rank Satisfies a property that is analogous to Shannon’s noisy coding theorem
Risk‐based portfolio guidelines
• Limit the amount of risk allocated to any one broad asset group Eg equities, interest rates, credit etc.
• Limit the total aggregate risk that can be taken Risk models underestimate volatility and overestimate diversification benefits in
stressed markets
• Continuous monitoring of portfolio’s actual risk behaviour versus what is expected
The information shown relates to the past. Past performance is not a guide to the future. The value of investment can go down as well as up.
Any data contained herein which is attributed to a third party ("Third Party Data") is the property of (a) third party supplier(s) (the “Owner”) and is licensed for use byStandard Life**. Third Party Data may not be copied or distributed. Third Party Data is provided “as is” and is not warranted to be accurate, complete or timely. To the extentpermitted by applicable law, none of the Owner Standard Life** or any other third party (including any third party involved in providing and/or compiling Third Party Data)shall have any liability for Third Party Data or for any use made of Third Party Data. Past performance is no guarantee of future results. Neither the Owner nor any other thirdparty sponsors, endorses or promotes the fund or product to which Third Party Data relates.
**Standard Life means the relevant member of the Standard Life group, being Standard Life plc together with its subsidiaries, subsidiary undertakings and associatedcompanies (whether direct or indirect) from time to time."
Standard Life Investments Limited is registered in Scotland (SC123321) at 1 George Street, Edinburgh EH2 2LL.The Standard Life Investments group includes Standard Life Investments (Mutual Funds) Limited, SLTM Limited, Standard Life Investments(Corporate Funds) Limited, SL Capital Partners LLP and AIDA Capital Limited.
Standard Life Investments Limited is authorised and regulated by the Financial Services Authority.Calls may be monitored and/or recorded to protect both you and us and help with our training.
www.standardlifeinvestments.com
© 2013 Standard Life, images reproduced under licence