FPA proudly thanks our Retreat 2006 sponsor for their support 1 C. Thomas Howard, PhD President, Athena Investment Services and University of Denver In Search of Alpha: Are you looking in the wrong box? Financial Planning Association Annual Retreat May 6, 2006 Scottsdale, Arizona In Search of Alpha Are you looking for alpha in a characteristic box? 3
Characteristic Boxes Value Blend Growth Large-Cap Mid-Cap Small-Cap 4 Russ Wermers A Matter of Style: The Causes and Consequences of Style Drift in Institutional Portfolios Source: University of Maryland working paper, July 2002 5 Results - Drift & Performance Managers with high levels of drift also deliver better future portfolio performance as a result of their trades." Source: Wermers, University of Maryland working paper, July 2002. 6
Results - Drift & Excess Returns Top 10 percentile drifters: 3.23% Alpha Bottom 10 percentile drifters: 0.27% Alpha 2.9% annual return benefit to drifting Risk does not explain performance differences Source: Wermers, University of Maryland working paper, July 2002. 7 Wermers Conclusion "Managers with the best stock picking talents often tend to implement strategies that involve a significant amount of equity style (characteristic) drift." Source: Wermers, University of Maryland working paper, July 2002 8 Art Lutschanunig, President and CIO Morningstar Investment Services The Performance of Drifters vs. Index Huggers 9
Lutschanunig Used 2002 through 2005 mutual fund data Sorted funds into those that hugged the Morningstar box index vs. those that drifted Drifters outperform huggers by roughly 300 basis points annually For international funds, the drifter advantage was a much larger 600 basis points Risk does not explain performance differences 10 Generating Alpha Characteristic Drift Superior Stock Picking Alpha 11 Characteristic Boxes Inhibit Performance Value Blend Growth Large-Cap Mid-Cap Small-Cap 12
In Search of Alpha You need to move outside the Box to find Alpha 13 In Search of Alpha Why can t you find alpha in a characteristic box? 14 In Search of Alpha Because keeping a manager in a characteristic box forces the manager to style drift. 15
Style and Portfolio Characteristics are not synonymous! Characteristics: Style: - Value, Blend, Growth - Small, Mid, Large Method of Investing 16 Equity Manager Style Portfolio Stock Universe Screened Set Style Portfolio 17 Howard and Callahan The Problematic Style Grid Journal of Investment Consulting Winter 2005-06 (IMCA) with Craig T. Callahan President ICON Advisers 18
We tested four Styles representing a range of value/growth and size Benjamin Graham John Neff William O Neil T. Rowe Price 19 Our Stock Universe Stock information for years 1995 2003 S&P1500 SuperComposite each year No look-ahead bias Extreme values eliminated 20 Necessary Conditions for Boxed Investing to Make Sense 1. Screened Stocks - single box 2. Screened Stocks - no drift to other boxes 3. Multi-specialists outperform drifters 21
Benjamin Graham Style Estimated year-end stock value: V = EPS * (8.5 + 2*G EPS ) * 4.4/AAA Ranked stock universe by ratio of P/V Each year 1995 2003 invested in the best 20 stocks and recorded the return 22 Necessary Conditions for Boxed Investing to Make Sense 1. Screened Stocks - single box 2. Screened Stocks - no drift to other boxes 3. Multi-specialists outperform drifters 23 Graham Characteristic Box Composition (Best 20 stocks each year 1995 2003) Value Blend Growth Large-Cap 6% 3% 1% Mid-Cap 21% 4% 1% Small-Cap 53% 9% 3% 24
Necessary Conditions for Boxed Investing to Make Sense 1. Screened Stocks - single box - No 2. Screened Stocks - no drift to other boxes 3. Multi-specialists outperform drifters 25 Small-Cap Drift Using Graham Style (Best 20 stocks each year 1995-2003) 90 80 % of Portfolio 70 60 50 40 30 1995 1996 1997 1998 1999 2000 2001 2002 2003 26 Value Drift using Graham Style (Best 20 stocks each year 1995-2003) % of Portfolio 95 90 85 80 75 70 65 60 55 1995 1996 1997 1998 1999 2000 2001 2002 2003 27
Styles Don t Fit into Boxes Value Blend Growth Large-Cap Mid-Cap Small-Cap 28 Necessary Conditions for Boxed Investing to Make Sense 1. Screened Stocks - single box - No 2. Screened Stocks - no drift to other boxes - No 3. Multi-specialists outperform drifters 29 Graham vs. S&P 1500 Annual Return 60 50 40 30 20 10 0-10 -20-30 1995 1997 1999 2001 2003 Average Graham S&P 1500 30
Average Graham Rank By Characteristic Box (Best 20 stocks in each CB each year 1995 2003) Value Blend Growth Large-Cap 213 175 419 Mid-Cap 61 116 289 Small-Cap 21 79 266 31 Now returning to all four styles Benjamin Graham John Neff William O Neil T. Rowe Price 32 Combined Style Performance Results Style Managers equally weighted average, four styles, 20 stocks each Excess Return vs. Universe 3.73% Risk Adjusted Alpha 4.13% Multi-Specialist Approach: 9 Different CB Managers CB Managers vs. Style Managers -3.39% Rank slope -2.7bp 33
Alpha as a Function of Rank (Average of Graham, Neff, O Neil, T. Rowe Price styles) Alpha (%) 5.0 4.0 3.0 2.0 1.0 0.0-1.0-2.0-3.0-4.0-5.0 Rank Slope = -2.7bp x 20 = - 54bp 1-20 21-40 41-60 61-80 81-100 101-120 121-140 Style Rank 34 And the answers are... 1. Screened Stocks - single box - No 2. Screened Stocks - no drift to other boxes - No 3. Multi-specialists outperform drifters - No 35 The Investing Circle Characteristic Drift Equity Style Consistency Alpha 36
Four Box Fallacies 1 Boxes are part of Boxes are 2 Modern asset Portfolio classes Theory Managers must be Boxes are a constrained to necessary risk avoid control tool undesirable 3 portfolio tilts 4 37 Box Fallacy 1 1 Boxes are part of Modern Portfolio Theory 38 1: Boxes are part of MPT MPT rests on Markowitz optimization, CAPM, and Efficient Markets Eliminating boxes does not impact any of these Boxes confused with size/pe research Box use evolved out of convenience, not based on careful research Source: The Characteristic Grid is not Part of MPT by Howard and Callahan November 2005 White Paper available on www.athenainvest.com 39
Box Fallacy 2 Boxes are asset classes 2 40 2: Boxes are Asset Classes Highly correlated Not compositionally unique Unstable over time 10% risk reduction at most Source: Boxes are not Classes by Callahan and Howard Investment Advisor, January 2006 41 Box Fallacy 3 3 Managers must be constrained to avoid undesirable portfolio tilts 42
Invest in these Four Style Managers Benjamin Graham John Neff William O Neil T. Rowe Price 43 Value (Growth) Tilts over Time 1 0.8 Value (Growth) Tilt (Standard Deviation) 0.6 0.4 0.2 0-0.2-0.4-0.6-0.8-1 1995 1996 1997 1998 1999 2000 2001 2002 2003 Graham Neff O'Neil T. Rowe Price 44 Portfolio: equal weighting of the four styles 1 0.8 0.6 Value (Growth) Tilt (Standard Deviation) 0.4 0.2 0-0.2-0.4-0.6-0.8-1 1995 1996 1997 1998 1999 2000 2001 2002 2003 Portfolio 45
Size Tilts over Time 1 0.8 Size Tilt (Small to Large) (Standard Deviation) 0.6 0.4 0.2 0-0.2-0.4-0.6-0.8 Graham Neff O'Neil T. Rowe Price -1 1995 1996 1997 1998 1999 2000 2001 2002 2003 46 Portfolio: equal weighting of the four styles 1 Size Tilt (Small to Large) (Standard Deviation) 0.8 0.6 0.4 0.2 0-0.2-0.4-0.6-0.8-1 1995 1996 1997 1998 1999 2000 2001 2002 2003 Portfolio 47 Box Fallacy 4 Boxes are a necessary risk control tool 4 48
4: Necessary risk management tool Random selection of mutual funds performs as well as knowing the mutual fund s box Even selecting all funds from the same box has good risk features A consistent style, without regard to the box, performs the best Source: Risky Business by Callahan and Howard Investment Advisor, February 2006 49 Our Conclusion Boxes wipe out alpha and provide no risk reduction benefits. 50 The Problem and the Answer The Problem is the Box while the Answer is the Circle Characteristic Drift Equity Style Consistency Alpha & Better Risk Control 51
StylePlus SM Managers define and pursue their own unique equity style Style is regularly audited Advisors and consultants have tools to control risk and characteristics at the portfolio level Investors able to earn the best returns at an acceptable level of risk 52 The Win, Win, Win Circle Risk Control Tools Manager Style Consistency Investor Results 53