Understanding the Principles of Investment Planning Stochastic Modelling/Tactical & Strategic Asset Allocation John Thompson, Vice President & Portfolio Manager London, 11 May 2011
What is Diversification and Asset Allocation?
Why Diversify? A Look at Risk The risk of an asset can be divided into two parts Systematic Risk (also called un-diversifiable risk, market risk, or beta risk) Unsystematic risk (also called diversifiable risk, non-market risk, or idiosyncratic risk)
Stock Diversification Company risk Market risk 1 2 4 6 8 16 30 50 100 1000 Number of stocks in portfolio Past performance is no guarantee of future results. This is for illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index.
Asset Class Diversification Risk 14% 12 10 13.0% 10.8% 9.8% 8 9.2% 8.9% 8.7% 8.5% 8.3% 6 4 2 1 2 3 4 5 6 7 8 Number of asset classes in portfolio Past performance is no guarantee of future results. This is for illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index.
Diversification in Bull Markets (February 2003-October 2007) Pound Sterling Bull market $2,500 2,000 1,500 2 1.8 1.6 1.4 1.2 1 0.8 Stocks 50/50 Stocks portfolio Bonds 50/50 portfolio Gilts B u ll M a rk e t $2,084 $1,653 $1,274 Bear market $1,500 1.83 1,250 1.48 $1,160 1.18 1,000 1,000 0.6 0.4 0.2 $767 750 500 0 Feb-03 Oct Oct Oct Oct Oct Oct Nov Nov 2002 2003 2004 2005 2006 2007 2007 2008 M ay-03 Aug-03 Nov-03 Feb-04 May-04 Aug-04 Nov-04 Past performance is no guarantee of future results. This is for illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index. Feb-05 M ay-05 Aug-05 Nov-0 5 Feb-06 M ay-06 Aug-06 Nov-06 Feb-07 $491 May-07 Aug-07 500
Diversification in Bear Markets (October 2007-March 2009) Bull market $2,500 1.4 Stocks 50/50 Stocks portfolio 1.2 Bonds 50/50 Stocks portfolio Gilts 50/50 portfolio 2,000 Gilts 1 Pound Sterling 0.8 1,500 0.6 0.4 1,0000.2 Bear Market $2,084 $1,653 $1,274 Bear market $1,500 1.83 1.16 1,250 1.48 $1,160 0.83 1.18 1,000 0.58 $767 750 500 0 Oct Oct Oct Oct Oct Oct Nov Nov 2002 2003 2004 2005 2006 2007 2007 2008 Oct-07 Nov-07 Dec-07 Jan-08 Feb-08 Mar-08 Apr-08 May-08 Past performance is no guarantee of future results. This is for illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index. Jun-08 Jul-08 Aug-08 Sep-08 Oct-08 Nov-08 Dec-08 Jan-09 $491 Feb-09 Mar-09 500
Why Correlation Matters?
Correlation and Asset Allocation Correlation measures the similarity or dissimilarity of asset classes relative to one another most asset classes -1 0 1 Range of Correlations Correlations range from -1 to 1, with most assets grouped in the moderately positive range.
Correlation and Asset Allocation 15% B Expected Return 10% Correlation= -1 Correlation= 0 Correlation= 1 5% A 0% 10% 20% 30% Standard Deviation
Asset-Class Correlations 1987 2010 Quarterly UK US Europe Stocks Stocks Stocks Property Past performance is no guarantee of future results. This is for UK stocks 1.00 illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index. The information, US stocks 0.81 1.00 data, analyses, and opinions contained herein do not constitute investment Europe stocks advice 0.91 offered 0.85 by Morningstar 1.00 and are provided solely for informational purposes. 2010 Morningstar. All Rights Reserved. 3/1/2010 Gilts 0.12 0.02 0.00 1.00 Gilts Direct Commodities Inflation Direct Property 0.18 0.12 0.20 0.27 1.00 Commodities 0.08 0.08 0.04 0.32 0.11 1.00 Inflation 0.03 0.03 0.02-0.21 0.24 0.30 1.00 Past performance is no guarantee of future results. Correlation ranges from 1 to 1, with 1 indicating that the returns move perfectly opposite to one another, 0 indicating no relationship, and 1 indicating that the asset classes react exactly the same.
Investor Behavior
According to a study by Alliance Bernstein, 94% of investors surveyed stated that long-term goals are an important factor in their investment decisions, however, most investors admit that they let current market conditions influence their investment decisions. Source: AllianceBernstein Global Literacy Research Program, 2006
Investor Behavior: Short-Term Focus Definition Inappropriately focusing on short-term risk versus long-term risk Implications Many investors talk long term but act short term Overly sensitive to interim volatility regardless of time horizon May tend to behave as though their time horizon is far shorter than it truly is
50% Probability Short-Term Focus: Coping with Near-Term Fluctuations Probability of losing money in the stock market 1990 2010 49% 40 45% 30 41% 32% 20 10 0 Daily Monthly Quarterly Annually Past performance is no guarantee of future results.
Why Not Market Time? Market timing can lead to: High portfolio turnover Exposure to more risk Must believe in past trends repeating Be aware of concentrated returns
Risk of missing the best days in the market 1991 2010 10% Return 8 9.1% 6 4 2 5.4% 3.0% 0.9% 0 2 4 Invested for all 5,043 trading days 1.0% 2.7% 10 best days missed 20 best days missed 30 best days missed 40 best days missed 50 best days missed 10% Return Daily returns for all 5,043 trading days 5 0 5 10 Past performance is no guarantee of future results. This is for illustrative purposes only and not indicative of any investment. An investment cannot be made directly in an index. 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Asset-Class Winners and Losers Past performance is no guarantee of future results. Highest return 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 23.0 33.4 28.6 29.8 21.5 22.8 17.8 60.7 20.7 14.0 26.9 11.6 25.9 32.5 31.3 17.6 22.8 20.3 27.3 5.9 3.8 1.6 39.2 18.4 7.8 16.2 9.9 1.6 28.1 15.1 10.3 15.9 13.1 21.0 0.6 3.7 6.5 28.7 12.0 7.3 15.8 5.5 20.7 26.5 13.6 6.4 15.9 12.2 14.3 3.6 0.8 13.3 24.8 10.9 5.7 12.9 5.4 36.7 14.0 10.1 5.2 5.3 4.9 4.7 9.1 11.9 15.7 1.4 8.5 4.9 4.8 4.7 37.0 0.1 8.2 0.9 2.1 7.3 9.0 14.0 21.2 22.1 1.0 1.2 3.0 1.2 5.2 43.1 14.9 0.1 Lowest return Small stocks Large stocks International stocks Long-term government bonds Treasury bills Diversified portfolio
What is Asset Allocation?
What is Asset Allocation? Asset allocation is the strategic approach of combining asset classes such as stocks, bonds, and cash in a portfolio in order to meet the investor s goals. Stocks Cash Bonds
Tactical Asset Allocation Goal is to take advantage of potential short-term opportunities in the market and deviating from the strategic asset allocation Programs may include valuation techniques or momentum techniques
The Asset Allocation Decision Three factors determine the appropriate asset allocation over an individual s life-cycle: Risk-return tradeoff among asset classes Investor s risk preference (short-term risk aversion) Investor s risk capacity (human capital)
14% Return Risk Versus Return Stocks, gilts, and cash 1976 2010 12 Europe stocks US stocks 10 Gilts 8 Cash UK stocks 6 4 2 Past performance is no guarantee of future results. 0 0% Risk 5 10 15 20 25 30 35
Risk Tolerance Questionnaire Goal Guide clients to an appropriate strategic asset allocation Process Gather information Assess the client s investment profile Identify an investment strategy
Human Capital Human Capital Human capital is the actuarial PV of future labor income Directly affects one s capacity to take on risk with one s financial capital Human capital is used to fund: Current expenses Retirement savings Direct contributions to portfolio Defined contribution (match) Defined benefit State Pensions PV=Present Value
Melting Pot: Portfolio is Greater than the Sum of its Parts Risk and Return Characteristics 1976 2010 US stocks Return: Risk: 12.0% 20.3% UK stocks Return: Risk: 8.5% 15.1% Total portfolio Return: 10.7% Risk: 10.7% Cash (90 Day) Return: Risk: 8.1% 4.1% Europe stocks Return: 12.4% Risk: 18.6% Gilts Return: Risk: 10.1% 9.7% Past performance is no guarantee of future results.
Asset Allocation Best Practices
Developing Asset Class Inputs: Overview of Building Blocks Method What does the market expect today? Incorporates relevant historical data and current expectations How much compensation do investors require for bearing additional risk? Risk premia reflected through supply side analysis What can we learn from historical experience? What historical relationships can we uncover among the asset classes
The Traditional Markowitz Mean-Variance Framework Expected Return MVO MVO Inputs Inputs Mean-Variance MVO Optimizer Optimizer MVO Efficient Frontier Individual Assets Standard Deviation de Finetti / Markowitz Mean-Variance Optimization (de Finetti [1940], Markowitz [1952, 1959]) Capital Market Assumptions Expected Returns Standard Deviations (Risk) Correlations
Mean Conditional Value-at-Risk Optimization Mean-CVaR MVO Inputs Inputs Mean-CVaR MVO Optimizer Optimizer Mean-CVaR Efficient Frontier Capital Market Assumptions Expected Returns Standard Deviations (Risk) Correlations Skewness Kurtosis Expected Return Individual Assets Conditional Value-at-Risk
Building Efficient Portfolios Analyze risk and return characteristics of asset classes Correlation is the key Building the efficient frontier Selecting the appropriate portfolio
Selecting An Appropriate Portfolio Which efficient portfolio is appropriate? Appropriate for investor s risk tolerance Matches investor s behavioral attitude towards risk Consistent with investor s time horizon Use risk tolerance questionnaire and other tools as a guide
Stochastic Modelling
Stochastic Modelling for Investors Stochastic Modelling (aka Monte Carlo analysis): randomly generating data using sampling distributions Two types of data used in financial market studies Real: observed in the real world Simulated: lab generated to simulate the real world
Stochastic Modelling for Investors Illustrates market impact on investment strategy Creates many scenarios Each scenario is a possible future return path The Approach Generate many possible random return scenarios Project the investor s wealth Summarise and present outcomes
Wealth Forecasting Without Cashflows Year Return Wealth Return Wealth 0-1,000,000-1,000,000 1 25% 1,250,000-20% 800,000 2-20% 1,000,000 25% 1,000,000
Wealth Forecasting Without Cashflows Wealth ( ) 4 95th Percentile Expected Value 5th Percentile 1 Dec 2010 Dec 2015 Dec 2020 Dec 2025 Dec 2030 Time
Wealth Forecasting With Cashflows Year Return Wealth Return Wealth 0-1,000,000-1,000,000 1 25% 1,250,000 cashflow - 500,000-20% 800,000-500,000 2-20% 600,000 25% 375,000
Wealth Forecasting With Cashflows: Deterministic Method 3,000,000 2,500,000 Portfolio Income 2,000,000 1,500,000 1,000,000 500,000 Portfolio Balance 0 5 10 15 20 Years
Wealth Forecasting With Cashflows: Deterministic Method 130,000 65,000 Portfolio Income in real terms 0 65 70 75 80 85 90 95 100 Age
Why Stochastic Modelling? Typically used when there is no analytical solution Wealth forecasting with cashflows Individual wealth forecasting & asset-liability analysis Used to personalise or illustrate the uncertainty in the market Flexibility Single-period vs. multi-period Alternative objectives: Probability of short-fall
Case Profile: Couple Age 65 Pre-Retirement Income 100,000/year Target Retirement Income 75,000/year Pension Benefit 25,000/year Portfolio Size 1,000,000 60/40 equity/fixed income 8.8% compound return 13.2% standard deviation 3% inflation
Deterministic Forecast Wealth Today s Pound Sterling 1,000,000 100,000 10,000 Income Today s Pound Sterling 80,000 60,000 40,000 20,000 0 Pensions Longevity Chance of living longer Age 65 70 75 80 85 90 95 100 9 in 10 3 in 4 1 in 2 1 in 4 1 in 10
How Stochastic Modelling Works 1928 1943 1935 1963 1975 1994 1956 1944 1952 1971 2001 1948 1950 1961 1990 1939 1985 1958 1982
1939 1943 1948 1961 1928 1935 1975 1994 1982 1963 2001 1950 1985 1971 1990 1958 1952 1944 1956 1928 2000 1990 1981 1977 1969 1953 1946 1940 1932 1929 1988 1966 1979 1972 1971 1968 1965 1964 1952 1944 1926 1958 1936 1974 1930 2001 1973 1966 1957 1948 1941 1933 1937 1999 1998 1983 1982 1976 1967 1951 1942 How Stochastic Modelling Works
today tomorrow 1939 1943 1948 1961 1928 1935 1975 1994 1982 1963 2001 1950 1985 1971 1990 1958 1952 1944 1956 1928 2000 1990 1981 1977 1969 1953 1946 1940 1932 1929 1988 1966 1979 1972 1971 1968 1965 1964 1952 1944 1926 1958 1936 1974 1930 2001 1973 1966 1957 1948 1941 1933 1937 1999 1998 1983 1982 1976 1967 1951 1942 How Stochastic Modelling Works
today tomorrow 1939 1943 1948 1961 1928 1935 1975 1994 1982 1963 2001 1950 1985 1971 1990 1958 1952 1944 1956 1928 2000 1990 1981 1977 1969 1953 1946 1940 1932 1929 1988 1966 1979 1972 1971 1968 1965 1964 1952 1944 1926 1958 1936 1974 1930 2001 1973 1966 1957 1948 1941 1933 1937 1999 1998 1983 1982 1976 1967 1951 1942 How Stochastic Modelling Works
today tomorrow 1939 1943 1948 1961 1928 1935 1975 1994 1982 1963 2001 1950 1985 1971 1990 1958 1952 1944 1956 1928 2000 1990 1981 1977 1969 1953 1946 1940 1932 1929 1988 1966 1979 1972 1971 1968 1965 1964 1952 1944 1926 1958 1936 1974 1930 2001 1973 1966 1957 1948 1941 1933 1937 1999 1998 1983 1982 1976 1967 1951 1942 How Stochastic Modelling Works
How Stochastic Modelling Works We draw from parameters calculated by the historical data, not from the historical data itself $ today tomorrow 1928 1943 1935 1963 1975 1994 1956 1944 1952 1971 2001 1948 1950 1961 1990 1939 1985 1958 1982
Types of Stochastic Modelling Non-parametric (Bootstrapping) Parametric Simulation based on historical data No distribution parameter assumption made Simulation based on specific distribution assumptions (e.g., normal, lognormal, etc.) Advanced modelling with more parameters (e.g. serial correlation)
Case Profile: Investor at Age 45, Retirement Age 65 Pre-Retirement Income 100,000/year Savings (inflation adjusted) 10,000/year Target Retirement Income 75,000/year Pension Benefit 25,000/year Portfolio Size 300,000 60/40 equity/fixed income 8.8% compound return 13.2% standard deviation 3% inflation
Illustration with Conventional Approach Wealth Today s Pound Sterling Age 45 50 55 60 65 70 75 80 85 1,000,000 616,156 100,000 10,000 Income Today s Pound Sterling 80,000 60,000 40,000 20,000 0 Pensions
Illustration with Stochastic Modelling Wealth Today s Pound Sterling 1,000,000 616,156 1 in 2 100,000 Market Risk Chance of this scenario: 1 in 2 10,000 Income Today s Pound Sterling 80,000 60,000 1 in 2 40,000 20,000 0 Pensions Age 45 50 55 60 65 70 75 80 85
Illustration with Stochastic Modelling Wealth Today s Pound Sterling 1,000,000 616,156 1 in 2 100,000 10,000 1 in 10 Market Risk Chance of this scenario: 1 in 2 1 in 10 Income Today s Pound Sterling 80,000 60,000 1 in 2 40,000 20,000 0 1 in 10 Pensions Age 45 50 55 60 65 70 75 80 85
Illustration with Stochastic Modelling Wealth Today s Pound Sterling 1,000,000 616,156 1 in 2 100,000 10,000 1 in 10 1 in 4 Market Risk Chance of this scenario: 1 in 2 1 in 4 1 in 10 Income Today s Pound Sterling 80,000 60,000 1 in 2 40,000 20,000 0 1 in 10 1 in 4 Pensions Age 45 50 55 60 65 70 75 80 85
Stochastic Modelling General Rules Assumptions Simulations are only as good as what goes into the process Individual Securities Beware of any long-term illustrations that go beyond asset classes
The Interpretive Material Key Points An Investment Analysis Tool Produces simulations and statistical analyses that present the likelihood of various investment outcomes if certain investments are made or certain investment strategies or styles are undertaken Serves as an investor resource to evaluate potential risks and returns of investment choices
The Interpretive Material Key Points Advisor should Describe the methodology and criteria used, including the tool s limitations and key assumptions Explain that results may vary with each use and over time Describe the investment universe considered in the analysis Disclose that the outcomes are hypothetical and do not reflect actual investment results
Summary Mitigate risk by diversification across investment securities and asset classes Utilizing correlation: the portfolio may be greater than the sum of its parts Avoid the pitfalls of market timing and chasing performance If using tactical asset allocation, understand the methodology Select the asset allocation portfolio that suits the investor s risk tolerance and time horizon Asset allocation is the strategic mix of asset classes designed to meet the investor s goals