Evaluating Managers on an After-Tax Basis Brian La Bore Senior Manager Research Analyst Head of Traditional Research Greycourt & Co., Inc. March 25 th, 2009
Is Your Alpha Big Enough to Cover Its Taxes? 1993 paper by Arnott Most manager s alphas are not big enough to cover their taxes. Arnott and Jeffrey s paper caused a flood of analysis and discussion on tax efficient investment strategies and after-tax performance measurement. Little attention has been given to the assessment of the Required Excess Return of active management.
Required Excess Return 10.00% Market Return Market Return Required Return Less Passive Management Costs Market Return Less Active Management Costs Market Return ETF Return Active Manager Return Required Excess Return: Index Fund Forecast Return Manager Forecast Return
How Have Active Managers Done Historically? 4.50 4.00 3.50 Median Core Manager Gross Excess Return (%) Ending December 31, 2008 3.90 3.00 2.50 2.38 2.00 1.50 1.00 0.50 0.95 1.50 Emerging Markets International Equity Domestic Large Cap Domestic Small Cap Source: Custom Universe - Zephyr 15 Years
Universe Results: Survivorship Bias and Cyclicality 5-Year gross excess performance for median small cap core manger = 1.27% 5.00% Median Small Cap Core Gross Excess Return 4.00% 3.00% 3.38% 3.48% 3.90% 2.00% 1.00% 0.33% 0.93% 0.79% 1.27% 0.00% 1.00% 0.32% 1 year 2 years 3 years 4 years 5 years 8years 10 years 15 years Source: Custom Universe Zephyr Small Core
Cost of Active Management: Management Fees and Taxes Hypothetical: Cost of Active Management $20.00 $18.00 Market Forecast: $19.58 $16.00 $14.00 $12.00 $10.00 $8.00 $6.00 $4.00 $2.00 $0.00 Year 1...Year 10 Market Forecast Manager: Net Mgt Fee (0.75%), Net Taxes (40% Turnover) ETF: Net Mgt Fee (0.25%), Net Taxes (7% Turnover) ETF: $14.70 Manager: $10.02 Required Return
Cost of Active Management: Variance Drain $25.00 Forecasting Arithmetic Returns $20.00 $15.00 $10.00 $20.91 $12.28 Variance Drain $5.00 $0.00 Manager A Manager B Annualized Annualized Annualized Arithmetic Average Standard Deviation Geometric Average Manager A 17.11% 24.12% 13.92% Manager B 17.11% 0.00% 17.11%
Required Excess Return Is your estimated excess return big enough to cover its costs versus a passive ETF? A reasonable approach: Step 1: Determine the Net-of-Fee After-Tax return for Index ETF. Management Fees Use ETF fees as proxy Asset Turnover Generator of Capital Gains Taxes Yield Income Taxes Volatility.Variance Drain Step 2: Calculate Required Excess Return. Step 3: Examine Magnitude and Stability of Excess Return.
Step 1: Determine the After-Tax Return for the Index. Key Assumptions: 10 Year Time Horizon No Liquidation Maximum Federal Tax Rates Assumptions: (1) Management Fees (2) Asset Turnover (3) Volatility Small Cap Index Fund 0.25% 35.00% 19.00% Large Cap Index Fund 0.15% 7.00% 16.00% Small Cap Index Fund Large Cap Index Fund Return Summary Projected Average Return 10.00% 9.00% Projected Annual Compound Return 8.19% 7.72% Loss due to Ordinary Gains Taxes 0.82% 0.18% Loss due to Income Taxes 0.21% 0.40% Loss due to Management Fees 0.25% 0.15% Total Reductions 1.28% 0.73% Net After Tax Return 6.85% 6.95%
Step 2: Determine Required Excess Return for Manager What is the manager s required excess return to match index fund? Assumptions: (1) Management Fees (2) Asset Turnover (3) Volatility Example: Small Cap Manager Example: Large Cap Manager Manager vs. Index Manager vs. Index 1.00% vs. 0.25% 0.75% vs. 0.15% 60% vs. 35% 40% vs. 7.0% 20% vs. 19% 17% vs. 16% Small Cap Large Cap Return Summary Manager Benchmark Manager Benchmark Projected Average Return 11.54% 10.00% 10.36% 9.00% Projected Annual Compound Return 9.54% 8.19% 8.92% 7.72% Loss due to Ordinary Gains Taxes 1.46% 0.82% 0.82% 0.18% Loss due to Income Taxes 0.11% 0.21% 0.32% 0.40% Loss due to Management Fees 1.00% 0.25% 0.75% 0.15% Total Reductions 2.57% 1.28% 1.89% 0.73% Net After tax Return 6.85% 6.85% 6.95% 6.95% Required Excess Return 1.54% 1.36%
Greycourt s Cassandra Model: Forecasting Capital Gains In Greek mythology, Cassandra was the daughter of King Priam and Queen Hecuba of Troy. Her beauty caused Apollo to grant her the gift of prophecy. However, when she did not return his love, Apollo placed a curse on her so that no one would ever believe her predictions. - Wikipedia After-Tax Model Inputs Combined PortfolioOutputs Investment Alternatives: Y/N Return Summary Manager 1. Invest Cash y Arithmetic Annual Gross Return 9.00% 3. Sell & Reinvest Existing Portfolio n Geometric Annual Gross Return 7.38% 4. Transfer Existing Portfolio n Loss due to Ordinary Gains Taxes 1.07% Settings: 3/4/2009 Loss due to Rebalancing Gains Taxes 0.00% Portfolio Time Horizon (Years) 10 Loss due to Income Taxes 0.22% Estate Liquidation (Y/N)? n Loss due to Management Fees 0.19% Allowable Asset Class Variance (%) 0% Impact of Cash Inflows/Outflows 0.00% Worst Case Scenario (Y/N)? n Impact of Portfolio Liquidation 0.00% # of Standard Dev (worst case) 2 Tax Data: Rate Total Reductions 1.48% S-T Capital Gains - Federal 35.00% L-T Capital Gains- Federal 15.00% Income- State 0.00% Net After-tax Return 5.82% Capital Gains- State 0.00% Net After-Tax, After-Cashflow Return 5.82% Dynasty Defective Trusts Active (Y/N) n Estate Tax Rate 55.00% Cash Inflows/Requirements: Assumed Inflation Rate 3.0% Annual Cash Contributions/(Needs) ($) $ - Adjust Cashflows by Inflation (Y/N)? n AA Model Data Portfolio #: 0 (Initial) to 100 0 Partial after-tax return N/A Beginning Portfolio Market Value $ 1,000 Portfolio Risk 18.00% Final Portfolio Market Value $ 1,760 Asset Class Expected Expected Mgt. Expected Expected Use Special Assumptions Total Return Yield Fee Turnover Risk (St.Dev.) Turnover? INPUT ASSUMPTIONS 9.00% 1.75% 0.20% 60.00% 18.00% N
Step 3: Calculate Coverage Ratio Has the manager s historical performance covered management fees and taxes? Small Cap Large Cap Return Summary Manager Benchmark Manager Benchmark Projected Average Return 11.54% 10.00% 10.36% 9.00% Projected Annual Compound Return 9.54% 8.19% 8.92% 7.72% Loss due to Ordinary Gains Taxes 1.46% 0.82% 0.82% 0.18% Loss due to Income Taxes 0.11% 0.21% 0.32% 0.40% Loss due to Management Fees 1.00% 0.25% 0.75% 0.15% Total Reductions 2.57% 1.28% 1.89% 0.73% Net After tax Return 6.85% 6.85% 6.95% 6.95% Required Excess Return 1.54% 1.36% Historical Excess Return 9.10% vs. Russell 2000 Index 8.37% vs. Russell 1000 Index Coverage Ratio 5.91 x 6.15 x The Coverage Ratio: = Manager s Observed Excess Return / Manager s Required Return
Examine the Coverage Ratio Large Cap Manager Analyze persistence and pattern of historical performance. 14.00 Large Cap Manager Excess Return (%) vs. Russell 1000 Index 60 Month Rolling Window, Computed Monthly 12.00 10.00 8.00 6.00 4.00 2.00 0.00 12/1/2003 1/2/2005 1/2/2006 1/2/2007 1/2/2008 Excess Return vs. Russell 1000 Average Excess Return Overall persistence of excess return is low, but strong in down markets. Historical excess return has decreased since 2003.
Examine the Coverage Ratio Analyze size and style biases 15 Excess Return (%) Versus Russell 1000 Index 20 Quarter Moving Windows, Computed Quarterly 10 5 0 5 10 12/1/1997 1/2/2000 1/2/2002 1/2/2004 1/2/2006 1/2/2008 Large Cap Manager Russell 1000 Growth Russell 1000 Value Manager has large cap value style bias Manager s historical excess return is highly correlated to value
Examine the Coverage Ratio Analyze Style Benchmark 16.00 Large Cap Manager Example Excess Return(%) 36 Month Moving Window, Computed Monthly 14.00 12.00 10.00 8.00 6.00 4.00 2.00 0.00 2.00 12/1/2001 1/2/2003 1/2/2004 1/2/2005 1/2/2006 1/2/2007 1/2/2008 Excess Return vs. Style Benchmark January 1999 December 2008. Single Computation Annualized Excess Return (%) vs. Style Benchmark Explained Variance (%) Annualized Excess Return (%) vs. Russell 1000 Explained Variance (%) Large Cap Manager Example 5.65 73.8 8.37 67.8
Examine the Coverage Ratio Up market versus down market performance 120 Upside Downside (%): Large Cap Manager January 1993 December 2008 100 Y Axis: Upside (%) 80 60 40 20 0 0 10 20 30 40 50 60 70 X Axis: Downside (%) January 1993 December 2008. Single Computation # of Quarters Average Return (%) Average Return (%) vs. Market Market Benchmark (%) Up Down Up Down Up Market Down Market Up Capture Down Capture R Squared Large Cap Manager 47 17 6 5.74 6.13 4.27 98.7 61.8 82.19 Russell 1000 44 20 6.22 7.35 6.22 7.35 100 100 100 Manager s excess return was generated in down markets.
Adjust the Coverage Ratio Adjust expected excess return given following: Scenario Positive return environment Growth outperforms value Adjustment Downward given manager s upside capture Downward given manager s value bias Large cap outperforms small cap stocks Upward given manager s size bias Adjust manager s expected excess return downward and recalculate Coverage Ratio
Weighing the Cost of Active Management 14.00 Gross Excess Return (%): Average Large Cap Core Manager versus Russell 1000 Index 12 Month Rolling Window, Computed Monthly After Tax Apathy 12.00 10.00 8.00 6.00 4.00 2.00 0.00 2.00 4.00 Today Arnott s Paper 1/1/2000 1/1/2001 1/1/2002 1/1/2003 1/1/2004 1/1/2005 1/1/2006 1/1/2007 1/1/2008 Excess Return vs. Russell 1000 Average
References: Is Your Alpha Big Enough to Cover Its Taxes? Robert D. Arnott THE JOURNAL OF PORTFOLIO MANAGEMENT SPRING 1993 Disclaimers: The statistical information presented in this report has been obtained from independent sources as noted. While Greycourt believes these sources to be reliable, Greycourt has not independently verified this information. Past performance does not guarantee future results. Greycourt maintains a proprietary LTACF which it updates periodically. This document explains in detail Greycourt s view of potential returns by asset class for use in one-on-one presentations only and is available on that basis upon request. Greycourt s LTACF model does not include the deduction of advisory fees or other expenses that a client may have to pay. It does assume the reinvestment of interest and dividends. As with all models, there are inherent limitations to the model particularly the fact that results may not reflect the impact that material economic or market factors may have on actual advisor decision making. Custom Universe Disclosure Purpose: The custom universe is designed by Greycourt and uses publicly-available information to compare a group of managers with similar investment styles ( peer groups ). Peer groups are constructed to provide an objective comparison of a specific, client manager s performance against other managers that invest in a similar style. Peer groups are not intended to be used as a buy list and will include some managers that would not be recommended by Greycourt to its clients. Methodology: A peer group is constructed by first identifying a particular market benchmark. For example, a peer group for core, domestic equity managers is generally benchmarked to the S&P 500 Index. Greycourt screens a large number of managers to determine those managers that have investment characteristics similar to those of the benchmark. Characteristics include R-squared, growth/value orientation, capitalization and product description. Returns are not considered in the screening process so good and poor performing managers are included in the peer groups. Whenever possible, Greycourt used published, widely used manager databases to provide the most diverse group of managers possible. Plan Sponsor Network (separate accounts) and Morningstar (mutual funds) are currently used. Plan Sponsor Network returns are presented gross of fees while Morningstar is net of fees. Peer groups for hedge fund of fund managers are constructed by specific identification of managers that invest in a similar manner. Greycourt has not identified a reliable public database for this group of managers.
Disclaimers Russell 1000 Index measures the performance of the 1,000 largest companies in the Russell 3000 Index, which represents approximately 92% of the total market capitalization of the Russell 3000 Index. As of the latest reconstitution, the average market capitalization was approximately $13.0 billion; the median market capitalization was approximately $4.6 billion. The smallest company in the index had an approximate market capitalization of $1.8 billion. Russell 2000 Index measures the performance of the 2,000 smallest companies in the Russell 3000 Index, which represents approximately 8% of the total market capitalization of the Russell 3000 Index. As of the latest reconstitution, the average market capitalization was approximately $664.9 million; the median market capitalization was approximately $539.5 million. The largest company in the index had an approximate market capitalization of $1.8 billion. The MSCI EAFE Index (Europe, Australasia, Far East) The MSCI EAFE Index (Europe, Australasia, Far East) is a free float-adjusted market capitalization index that is designed to measure developed market equity performance, excluding the US & Canada. As of April 2002 the MSCI EAFE Index consisted of the following 21 developed market country indices: Australia, Austria, Belgium, Denmark, Finland, France, Germany, Greece, Hong Kong, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Portugal, Singapore, Spain, Sweden, Switzerland and the United Kingdom. The MSCI EM (Emerging Markets) Europe, Middle East and Africa Index The MSCI EM (Emerging Markets) Europe, Middle East and Africa Index is a free float-adjusted market capitalization index that is designed to measure equity market performance in the emerging market countries of Europe, the Middle East & Africa. As of August 2002, the MSCI EM EMEA Index consisted of the following 10 emerging market country indices: Czech Republic, Hungary, Poland, Russia, Turkey, Israel, Jordan, Egypt, Morocco and South Africa.
Contact Information Brian La Bore Senior Manager Research Analyst Head of Traditional Research Greycourt & Co., Inc. (503) 226-0468 (503) 226-0471 blabore@greycourt.com www.greycourt.com