MSc Investment Management

Portfolio Management
Dr Cesario MATEUS cesario.mateus.1@city.ac.uk www.cesariomateus.com

Lecture 1
Introduction to Passive vs. Active Portfolio Management Strategies Strategic Asset Allocation (SAA)

January, 25, 2012

Three forms of market efficiency

Weak form (WF): Today price includes past information Semi strong form (SSF): Today price includes todays public information Strong form (SF): Today price includes todays private information

A market that is strong form efficient is also semi-strong and weak form efficient
3

Two conflicting views

Efficient market type: Rational economists
The market knows best Competition and incentive for financial gains ensure fair pricing and market efficiency Replicate, do not attempt to beat the market

Inefficient market type: Behavioural economists

Behaviouralists know best The irrationality of the herd pushes prices away from fundamentals and allows mispricings to survive Attempt to beat the market by exploiting these mispricings

Strong Form of Efficiency

Since all information is reflected in prices fully and instantaneously it will be useless in predicting future prices (and returns).
The type of information analysts use depends on their belief regarding what information is reflected in market prices Technical analysts Fundamental analysts Technical analysts (chartists) believe weak form inefficiency Its possible to beat the market trading on past price movements and trends Fundamental analysts believe in weak form efficiency Earning abnormal returns/profits requires gathering and analyzing information Forecasting future earnings, dividends and other fundamentals better than other investors increases the chance of earning abnormal returns/profits 5

Fundamental versus Technical Analysis

Fundamental Analysis: involves analysing its income statements, financial statements, its management and competitive advantages and its competitors and markets
The analysis is performed on historical and present data, but with the goal to make financial projections

Technical analysis: the study of market action, primarily through the use

of charts, for the purpose of forecasting future price trends

Investment policies: Passive

Assume that markets are efficient
basing decisions on results of portfolio theory and CAPM

Manage portfolios that are surrogates for the market portfolio or those tailored for particular (not average) clients - index funds
indexation strategy

The aim: to achieve average performance through returns which are equal to the market as a whole
success judged by how small is the difference between funds return and index return - tracking error Morningstar survey found an average of 38 basis points across all index funds.
7

Origins from 1970s: Wells Fargos index portfolio CAPM: concept of the Market portfolio Performance of passive funds: not very competitive in the 70s and 80s

shift towards passive management in the 1990s

increasingly good performance over longer time horizon Important to choose the right benchmark portfolio

Investment policies: Active

Assume that there are mispriced securities or groups of securities: do not believe in market efficiency and have different forecasts than consensus Active portfolio: differences between proportions in the actual and benchmark portfolio
bets are placed on certain securities

The aim: to outperform the benchmark index

success judged by how large is the difference between fund return and index return

Passive v s. active management

Turnover and transaction costs of passive fund are smaller as well as management fees
Active management must generate at least 1.25% of gross additional return to break even with passive

Greater diversification in the passive management Frequent revising of portfolios by active managers in search for the winners The aim of the passive is to achieve returns close to the market returns, while active is trying to outperform
10

Passive vs. active management, cont.

Passive management would be favored by a rational investor if skills are not there, active management would produce high costs and no compensation active portfolio is more risky than passive due to unsystematic risk

Market is said to be zero-sum game: winners & losers

Combination of the two investment policies is possible

11

Do active portfolio managers beat passive portfolio managers?

Percentage of active general equity funds that were beaten by Vanguard (S&P) Index Fund after expenses

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1 Year 5 Years 10 Years

71% 63% 52%

Time period ending 31 December 2001

Source: B. Malkiel (2003)

12

General Approaches to Active Strategies

Security (sector) selection or stock screening
Based on various active quant techniques such as: Momentum investing, earnings surprise, style investing, use of relative valuation techniques, optimisation etc.

Asset allocation (AA)

Allocating the funds within one asset class or between asset classes Strategic AA: based on long-term forecasts and used for determining longer term position in an asset class Tactical AA: based on short-term forecasts and used for determining switching between or within asset classes

Market Timing
Determining when to switch from one asset (class) to another Can be done at individual security level and market level
13

Asset Allocation

14

What is Asset Allocation (AA)?

AA traditionally involves allocation of funds between the following asset classes: equities, bonds, property, cash, etc Within each of these asset classes, separate allocation decisions need to be made: value vs. growth stocks; domestic vs. international stocks; government vs. corporate bonds etc. Practitioners dilemma: traditional vs. alternative asset classes (commodities, private equity, FX, hedge funds)

15

Asset Allocation in practice

In 1980s and 1990s institutional investors allocated most funds to equity
Barclays Capital: 100 invested in 1899 in UK stock market (with income reinvested) would be worth around 25,022 in real terms today; while the same investment in gilts and cash will be worth 323 and 286 respectively.

Asset mix has changed in the last 10 years, particularly since 2008: bonds represent greater proportion of the holdings
Reasons: high equity market returns from 1990s may not be repeated in the future; when markets are in downturn, investors turn to less risky investments

16

AA in practice: change in the UK asset mix

1997: AUM by Asset Type

4.1% 0.4% 1.2%

2009: AUM by Asset Type

1.0% 1.8%
UK Equity & Bond Bond

6.2% 31.4% UK Equity Balanced Bond Global Equity Absolute return Money Market 8.1% Other

32.8% 52.8%

Global Equity Managed Money Market

29.6%

5.7% 2.9%

Other

19.9%

Source: IMA survey, 2009

17

CFA guidelines on investment process

Define objectives: desired investment outcomes (return and risk objectives) Define constraints: regulatory, legal Formulate investment policy statement to include:
Client description, investment horizon, statement of investment goals and objectives/constraints, schedule for review of performance, asset allocation considerations that are accounted for when developing strategic asset allocation and rebalancing guidelines among others

Investment policy is the basis for strategic asset allocation Monitor and rebalance portfolio (if and when needed) Portfolio manager given a mandate: set of instructions detailing his task and how the performance will be evaluated, including the specification of the managers benchmark. Fact: there is no correct way to formulate an investment process 18

To define objectives, start from the client!!

Type of investor
Individual Life insurance Mutual Fund Endowment funds

Drivers of return requirement

Life cycle, education, retirement

Risk tolerance
Variable

Actuarial and business Low risk Variable Depends on when income is needed Variable Usually low risk

Non-life insurance
Pension Fund

Policy risks (short term)

Very low risk

Pension liabilities (long Very low risk term)

19

Top-Down Analysis
Capital Allocation Decision: Choice of proportion of the overall portfolio to place in safe but low-return versus risky but higher-return securities

Asset Allocation Decision: Distribution of risky investments across broad asset classes: Stocks, Bonds, etc

Selection Decision: Choice of which particular securities to hold within each asset class.

20

The next step: Strategic Asset Allocation (SAA)

The efficient optimisation of investment allocation to major asset classes of investment in order:

to meet the overall investment objectives of the institution and

to achieve an acceptable balance between risk and return

21

Strategic Asset Allocation (SAA)

Investment process has been agreed upon The next step is to design the SAA, this is the allocation to asset classes that in the long-term would meet client's needs Involves periodically rebalancing the portfolio in order to maintain a long-term goal for asset allocation. SAA is also known as policy AA (Asset Allocation)

Four main approaches to SAA: Capitalisation-based: i.e., using assets in proportion as in the world portfolioas in CAPM. Follow the median manager, especially among pension managers. No loss of mandate. Maximise managers utility? Mean Variance Optimisation (MVO): how to generate key inputs? Liability driven investment (LDI): asset/liability issues.
22

SAA Following the median manager

Choose SAA to reflect allocation of a median portfolio manager
Popular when performance measured with peer-group benchmark There is incentive NOT to underperform median manager Managers protected from being fired for taking higherthan-average risks Herding bias

23

SAA Mean Variance Optimization (MVO) approach

Best long-term asset mix to achieve investors objectives (5 years) Objective of SAA: identify return requirement of the fund and combine asset classes in such a way that the expected return is achieved with the lowest volatility; i.e. identify efficient frontier for any given group of asset classes Based on long-term view of asset performance and on investors risk profile / investment horizon; so need to know: 1. Long-term expected returns of asset classes 2. Volatility of those asset classes 3. Correlations between those asset classes
24

SAA (MVO): Determining long-term expected returns (building block approach)

To form a view about long-term expected returns, one needs to consider: Expected real return Expected inflation Expected risk premium Expected real return: even if there is economy with no inflation, investors will require real return on investment (as there is opportunity cost: we invest and forego consumption today for future consumption) Expected real return is closely related to the growth rate of the economy, which is quite stable over time (in the UK and major economies, ranging 2-3% over the last 30 years).
25

SAA (MVO): Determining long-term expected returns (building block approach)

Expected inflation: investors require compensation for expected inflation over time, in addition to the long term growth rate of the economy Target inflation: in UK Bank of England (BoE) target is 2%1%; ECB target is below 2%, similar to USA, Australia, Canada etc. Expected real return + Expected inflation = required return on cash deposit over time Also known as neutral policy rate (interest rates over time average around this rate) or riskless rate, used as a basis for return expected on risky investments Add the risk premium: credit risk, liquidity risk, equity risk premium 26 etc.

SAA (MVO): Determining long-term volatility and correlations

Estimates of future volatility are based on historical values BUT both volatility and correlations are time varying
Correlations have risen over time due to globalisation (many companies operating in foreign markets) - is there any benefit in international diversification and global asset allocation then? YES!

27

SAA: choosing the strategic portfolio

Asset allocator produces mean-variance frontier which can enable them to select a strategic investment portfolio:
Expected return: established via 'building block approach'
B Efficient frontier A

E D

Individual asset classes C F

Standard deviation, risk: established using historic estimates of volatilities and correlations

28

MVO based SAA portfolio weights

Stable weights over time (Fixed weight asset allocation) However, weights do not really have to remain fixed all the time optimal strategic asset mix changes if there is: Revision of estimates for expected returns, risks and correlations Change in the clients risk profile

Accept current market valuations; i.e., consistent with passive portfolio management and market efficiency

29

MVO based SAA what are the problems?

Michaud (1989) The Markowitz Optimisation Enigma: Is Optimized Optimal?, Financial Analysts Journal, Jan/Feb 1989:
Assumes normal distribution and requires long history of data Meaningless optimal portfolio (only a few assets in very high weightings) Based on risk and return which are both subject to estimation error Liquidity of assets is often ignored introducing liquidity as a constraint results in less return maximisation and less risk reduction, moving efficient frontier to the lower right corner on the mean/standard deviation graph Existence of optimally equivalent portfolios: portfolios that have statistically identical risk return profile but very different composition Small changes in inputs in MVO result in large changes to optimal portfolio

Despite these problems, optimisation is used in solving asset allocation problems and index tracking
30

SAA and optimisation problems potential solution

Version of Global optimisation combining CAPM, investor expectations and traditional MVO analysis is given by Black and Litterman model from Goldman Sachs. Model provides the flexibility of combining the market equilibrium with additional market views of the investor For detailed discussion refer to Black F. and Litterman, R. (1992) Global Portfolio Optimisation, Financial Analysts Journal, September/October 1992.

31

Liability Driven Investment (LDI) based SAA

Popular with pension funds Objective of LDI: return on the assets should at least match the payout on liabilities Objective until mid 1990s: beat the peer-group benchmark Example of LDI objective: match the change in liabilities plus outperformance of x % p.a (focuses on the liabilities first and then addresses the desired level of outperformance over the liabilities, subject to various risk constraints). LDI objective should be viewed as a refinement of existing investment objectives rather than a completely new approach Before LDI approach was introduced, the focus was purely on assets. Whereas in LDI exposures to various asset classes, can be translated to an expectation of performance relative to liabilities
32

LDI: 4 Step Process

Step 1- Creating Liability Matching Portfolio: Liability matching portfolio: lowest risk portfolio which is a combination of assets having similar sensitivity to inflation, interest rates and other variables as the liabilities
Forecasts of cash-flows and their sensitivity to inflation, interest rates etc..

Typical liability matching portfolio would include:

Index linked gilts, Corporate index linked bonds, Corporate bonds (to match liabilities with shorter duration) and swaps (interest rate, inflation and credit default - to synthetically match longer duration liabilities). Considered as the reference portfolio in the LDI strategy

33

LDI: 4 Step Process

Step 2 Risk Budget and Benchmark
Determine the overall risk constraints of the portfolio, the aggregate target outperformance and how much of this will come from market exposure (passive management) and how much from active management. For example a target of 0.5% or 1% outperformance over liabilities which is coming from market exposure (beta) can be shown below:

34

LDI: 4 Step Process

Step 3 Active manager outperformance
A target of 2% outperformance over liabilities might be adopted with 1% coming from market exposure (beta) and 1% from active management (alpha). Active portfolio composition example: 55% government and corporate bonds (liability matching portfolio) and 45% other risky assets (e.g. 25% active domestic and international equity, 10% property, 5% hedge funds and 5% commodities)

35

LDI: 4 Step Process

Step 4 - Implementation
Be pragmatic, there is more than one way to apply
Recall the structuring of fixed income portfolios to meet known liabilitiesclassical immunisation, cashflow matching, etc. What if liabilities are uncertain (i.e. stochastic)? You can choose from a wide range of assets, e.g. equities, hedge funds - these are also stochastic. Asset risk: asset allocation will produce different returns, volatilities, correlations, etc: should it be static or dynamic asset mix? Around half of the UK pension funds have adopted LDI approach
36

Source: IMA survey 2009

37

Reading list (Lectures 1&2)

Baca, S., Garbe B and R. Weiss, The rise of sector effects in major equity markets, Financial Analysts Journal, Sep / Oct 2000, 34-40 Bodie, Kane and Marcus, Essentials of Investments, pp 593-598 on Performance Attribution Black F. and Litterman, R. (1992) Global Portfolio Optimisation, Financial Analysts Journal, September/October 1992 Brinson, G., Hood R. and G. Beebower (1986), Determinants of Portfolio Performance, Financial Analysts Journal, July/August 1986. Brinson, G. Singer B and G. Beebower (1991), Determinants of Portfolio Performance II: An Update, Financial Analysts Journal, May/June 1991. Dahlquist, M. and C. R. Harvey (2001), Global Tactical Asset Allocation, The Journal of Global Capital Markets, Spring 2001.

Elton, E, Gruber, M, Brown, S and W. Goetzmann, Chapter 10 on International Diversification in 8th edition Modern Portfolio Theory and Investment Analysis, Wiley
Hood, R. (2005), Determinants of Portfolio Performance 20 years later, Financial Analysts Journal, September/October 2005. Idzorek, T(2010), Asset Allocation is King, Morningstar Advisor, April/May 2010. 38