THE ACTIVE SHARE DEBATE WEBINAR Presented by John Alexander, CFA Billy Welsh
Today s Speakers John Alexander, CFA Solutions Specialist evestment Billy Welsh Client Strategist evestment jalexander@evestment.com bwelsh@evestment.com
Agenda for Today s Call About evestment Introduction to Active Share History Philosophy behind the measure Two Consultant Perspectives on Active Share Billy Welsh: Active Share is not the best metric John Alexander: High Active Share should be emphasized in manager selection Wrapping Up the Debate: Productively Applying Active Share Data for Further Analysis & Discussion Q & A *NOTE: We are not seeking to prove/disprove the predictive value of Active Share when seeking future outperformance
ABOUT EVESTMENT
About evestment Solutions for the investment managers, investors and consultants Greatest depth and breath of data across both traditional and alternative asset classes Full suite of cloud based analytical software solutions to help you make more informed decisions Global client-base of over 2,500 asset managers, consultants and investors
ACTIVE SHARE
Active Share: The Basics Active Share measures the percentage of equity holdings in a portfolio that differ from the index constituents. It is calculated by summing the absolute difference in weight of each holding in the portfolio versus the index and dividing by 2. Active Share = n i=1 w fund(i) w index(i) 2 For each decision to overweight a position in a portfolio, there must be corresponding underweight in another benchmark name. Active Share: measures deviation away from a specific benchmark on a scale of 0-100 Overlap: measures participation in benchmark names Overlap = 100 - Active Share
Active Share: Example Low Active Share High Active Share Stock A Stock B Stock C Stock D Stock E Stock F Stock G Stock H Stock I Stock J Stock K Stock L Stock M Stock N Stock O Stock P Stock Q Stock R Stock S Stock T non-benchmark Cash Stock A Stock B Stock C Stock D Stock E Stock F Stock G Stock H Stock I Stock J Portfolio Stock K Stock L Benchmark Stock M Stock N Stock O Stock P Stock Q Stock R Stock S Stock T non-benchmark Cash 0 5 10 15 20 0 5 10 15 20
Background Collateral
Market Reaction To Active Share Offensive Position Many investment consulting firms published research favoring higher active share Passive investing grants access to benchmark returns cheaply, so consultants favor more active managers over managers who look like the benchmark Many clients are modifying IPS guidelines toward unconstrained/flexible strategies High active share managers also tended to agree with this research Defensive Position Passive managers and firms with lower active share criticized the measure and published reports contradicting the findings mentioned above Many quantitative firms by nature have lower active share due to stricter portfolio construction around the index A lower active share can help maintain lower tracking error Lower tracking error is one quality their current clients expect of them Better measures of portfolio construction exist 10
Background Collateral
Two Camps on Active Share: A Divisive Topic Billy Welsh High active share is not necessarily the answer Higher active share a greater chance performance will differ from the index (positive or negative) Active share is extremely benchmark dependent Better measures are out there to get a more complete picture John Alexander Investors should seek managers with high active share If you want lower tracking error, buy an index (cheaply) AUM growth and high excess returns may incentivize mangers to take less active risk versus the benchmark High active share managers have more opportunity to outperform. 12
ACTIVE SHARE AND ITS LIMITATIONS From a Consultant s Perspective Billy Welsh
Active Share Active Share High Active Share is Not Necessarily the Answer High active share may be import for some investors objectives, but not all Using it as an exclusionary criteria will limit opportunity set of managers who have proven themselves despite having lower active share Looking at US Large Cap Core and Small Cap Core managers suggests as you increase Active Share, your value add (excess return) is not tied to the subsequent increase, but your tracking error is US Large Cap Core US Small Cap Core 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50-2.00 0.00 2.00 4.00 6.00 8.00 10.00 Tracking Error & Excess Return 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 Tracking Error & Excess Return Avg 3 Yr Excess Return Avg 3 Yr Tracking Error Avg 3 Yr Excess Return Avg 3 Yr Tracking Error 3 Year figures over past 12 quarters, ending March 31, 2014 14
Universe Selection Plays a Big Role What is considered high versus low active share? Does it make sense to have the same cut-off points regardless of the universe? Active share above 90 is high Active share below 60 is low Universes have significantly different active share distributions 68% of US Small Cap Core managers have an Active Share above 90 6% of US Large Cap Core managers have an Active Share above 90 15
Active Share Active Share Active Share Active Share Active Share Distribution of Various Universes 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 Large Cap Core Equity 0% 10% 20% 30% 40% Percentage of Managers within Universe Large Cap Core Managers Global Large Cap Core Equity 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 Small Cap Core Equity 0% 10% 20% 30% 40% Percentage of Managers within Universe Small Cap Core Managers Emerging Markets Equity 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 0% 10% 20% 30% 40% Percentage of Managers within Universe Global Large Cap Core Managers 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 Average active share over past two years, ended March 31, 2014 0% 10% 20% 30% 40% Percentage of Managers within Universe Emerging Markets Equity Managers 16
Active Share Noise As it is not held in the index, cash can have a notable impact on a manager s Active Share rank within a universe. Looking at the managers with an Active Share of 100 95 within US Small Cap Core That group of managers held an average weight of 3% in cash The managers holdings in cash helped contribute to approximately 71% of them being in that group Off benchmark holdings Looking outside of the index can be a source of value add When is it considered cheating when compared to the true opportunity set? Comparing oneself against the MSCI EAFE, when the product contains 20-30% in emerging markets (not held in the EAFE), would increase the active share by 10-15% Using an appropriate benchmark The following slide shows the effect of changing the benchmark 17
Active Share Active Share Benchmark Selection Can Change Things The following two charts show the Active Share distributions moving from a managers stated core benchmark to a style specific growth oriented benchmark. Changing from MSCI World to World Growth 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 0% 10% 20% 30% 40% 50% Percentage of Managers within Active Share Group 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 Changing from MSCI ACWI to ACWI Growth 0% 10% 20% 30% 40% 50% Percentage of Managers within Active Share Group vs MSCI World vs MSCI World Growth vs MSCI ACWI vs MSCI ACWI Growth Each manager represented belongs to the evestment Global Growth Equity Universe 18
Appropriate Benchmark? A product with 100% active share means that not one of the portfolio s holdings is in the benchmark, begging the question, is the benchmark valid, is it still an accurate measure to compare against? You can still be different than the benchmark while still holding some or many of the same names. 19
Stock 1 Stock 11 Stock 21 Stock 31 Stock 41 Stock 51 Stock 61 Stock 71 Stock 81 Stock 91 Stock 101 Stock 111 Stock 121 Stock 131 Stock 141 Stock 151 Stock 161 Stock 171 Stock 181 Stock 191 Stock 201 Stock 211 Stock 221 Stock 231 Stock 241 Stock 251 Stock 261 Stock 271 Stock 281 Stock 291 Stock 301 Stock 311 Stock 321 Stock 331 Stock 341 Stock 351 Stock 361 Stock 371 Stock 381 Stock 391 Stock 401 Stock 411 Stock 421 Stock 431 Stock 441 Stock 451 Stock 461 Stock 471 Stock 481 Stock 491 Holding Weight Examples of Low Active Share Adding Value The chart below shows an equal weighted S&P 500 product compared to the S&P 500 (ETF for Holdings and Actual benchmark for performance) Notice in this example the active bets are at the extremes (underweight larger names, overweight smaller ones). Active share is considered low at 44%, yet these are two very different portfolios 3.00 Holdings Comparison of Cap Weighted vs Equal Weighted 2.50 2.00 1.50 1.00 0.50 0.00 Holding S&P 500 S&P 500 (Equal Weight) 20
Examples of Low Active Share Adding Value Despite a low active share, the Equal Weighted S&P 500 provided many periods of a differentiated performance figure. 21
Examples of Low Active Share Adding Value Average Three-Year Tracking Error was 4.6 over the entire time period in this chart, but 3.0 over the last 12 quarters, which is comparable to those in this Active Share Category for Large Cap Core Average TE = 4.6 Average TE = 3.0 22
Better Measures It is better to focus more on: Positions (where a manager is different) and concentration in top holdings Cash position Current number of holdings Effective number of holdings (number of holdings that actually drives absolute performance) Tracking error Information ratio
Stock 1 Stock 12 Stock 23 Stock 34 Stock 45 Stock 56 Stock 67 Stock 78 Stock 89 Stock 100 Stock 111 Stock 122 Stock 133 Stock 144 Stock 155 Stock 166 Stock 177 Stock 188 Stock 199 Stock 210 Stock 221 Stock 232 Stock 243 Stock 254 Stock 265 Stock 276 Stock 287 Stock 298 Stock 309 Stock 320 Stock 331 Stock 342 Stock 353 Stock 364 Stock 375 Stock 386 Stock 397 Stock 408 Stock 419 Stock 430 Stock 441 Stock 452 Stock 463 Stock 474 Stock 485 Stock 496 Holding Weight Stock 1 Stock 12 Stock 23 Stock 34 Stock 45 Stock 56 Stock 67 Stock 78 Stock 89 Stock 100 Stock 111 Stock 122 Stock 133 Stock 144 Stock 155 Stock 166 Stock 177 Stock 188 Stock 199 Stock 210 Stock 221 Stock 232 Stock 243 Stock 254 Stock 265 Stock 276 Stock 287 Stock 298 Stock 309 Stock 320 Stock 331 Stock 342 Stock 353 Stock 364 Stock 375 Stock 386 Stock 397 Stock 408 Stock 419 Stock 430 Stock 441 Stock 452 Stock 463 Stock 474 Stock 485 Stock 496 Holding Weight Same Active Share, Very Different Portfolio 3.00 2.50 Holdings Comparison of Cap Weighted vs S&p 500 Top 20 Replication 2.00 1.50 1.00 0.50 0.00 3.00 S&P 500 Holding S&P 500 - Top 20 Replication Holdings Comparison of Cap Weighted vs Equal Weighted 2.50 2.00 1.50 1.00 0.50 0.00 S&P 500 Holding S&P 500 (Equal Weight)
USING ACTIVE SHARE IN MANAGER SELECTION From a Consultant s Perspective John Alexander, CFA
The Fundamental Argument for High Active Share Active Risks -Higher Fees - Risk of Underperformance - Added responsibility of manager selection Passive Risks -No chance to outperform -Full downside volatility -Less perceived consultant value-add WORST CASE SCENARIO: Hiring an active manager and paying the requisite fees, only to find out later that they are a closet indexer with little to no chance of outperformance (net of fees)
Consultants May Have Reason for Concern Graphic from Lazard s Taking a Closer Look at Active Share, 2013 The rise in index prevalence has resulted in active managers who look more similar to their respective benchmarks than in years past.
Barbell Strategy Passive Core/Active Satellite Rather than getting caught in the middle, consultants have a tendency to barbell the risk of active management towards both ends of the spectrum.
Passive Core/Active Satellite Active #1 Consultants anchor many large mandates with passive strategies without abandoning active management altogether. Active # 2 Passive Active #3 Consultants need to evaluate how their active managers behave in different market environments and how they behave together. Managers need to highlight their differentiating factors and talk about how their strategy is positioned to perform going forward.
Consultants and Investors Reshaping IPS Guidelines Traditional Investment Policy Statement (IPS) Language New Portfolio Tactics at all times manager shall abide by constraints put forth in the Investment Policy Statement at no point shall greater than 5% of the assets be invested in the assets/liabilities of any single issuer the stated mandate of Large Cap Value manager XYZ shall be to outperform the Russell 1000 Value Index over a market cycle (3 to 5 years)
Monitoring Active Share Over Time Behavioral Finance 101: Loss Aversion Human beings strongly prefer avoiding a loss to experiencing an equal gain Portfolio Managers are human, too. Consultants should monitor Active Share over time to evaluate how managers alter portfolio construction through different performance periods and asset flows.
Case Study: Active Share at Work evest Capital: Top performer in All Cap Equity in 2009, 2010 & 2011. Since then, assets have grown tremendously. Active share is currently 73.90. Longer term track record is still very strong. What does the Active Share say about this manager? Next we ll see how to use Active Share data to evaluate this manager further.
What Does the Data Say? 84.00 82.00 80.00 78.00 76.00 74.00 72.00 70.00 68.00 Active Share over time evest Capital evest Capital has significantly decreased its Active Share since its strong performance run in 2009, 2010 and 2011. Its Active Share now ranks in the bottom decile among peer managers in its universe. Consultants should be especially cautious when they observe a trend like we see above. Whether evest Capital is consciously deciding to look more like the index or not, the data points to a major shift in portfolio construction.
PRODUCTIVELY APPLYING ACTIVE SHARE DATA
Portfolio Structuring and Combining Multiple High Active Share Managers The Active Share of the entire portfolio should be considered when looking for a new manager and or structuring a portfolio. High Active share managers (vs. their style specific benchmark), when combined, can result in a lower active share portfolio versus the style neutral benchmark. The following is an Active Share analysis of Four Active Managers. Each has high Active Share vs. their respective Benchmark and vs. each other (overlap = 100 - Active Share) When combined in equal weight and compared to the Russell 1000 the Active Share is 76.2. Manager 1 (Large Value) Manager 2 (Large Value) Manager 3 (Large Growth) Manager 4 (Large Growth) vs R1000 Value vs R1000 Growth Portfolio Weight Active Share vs R1000 Manager 1 (Large Value) Manager 2 (Large Value) Manager 3 (Large Growth) Manager 4 (Large Growth) --- 92 --- 25% 87.6 --- 94.5 --- 25% 93.7 97.2 --- --- 91.7 25% 97.6 100 100 --- --- 94.1 25% Total Portfolio 100% 76.2 35
Active Share Active Share Portfolio Structuring and Combining Multiple High Active Share Managers The Active Share of the entire portfolio should be considered when looking for a new manager and or structuring a portfolio. High Active share managers (vs. their style specific benchmark), when combined, can result in a lower active share portfolio versus the style neutral benchmark. Large Cap Value Equity Large Cap Growth Equity 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 0% 5% 10% 15% 20% 20% 15% 10% 5% 0% Percentage of Managers within Universe Percentage of Managers within Universe Large Cap Value Managers Large Cap Growth Managers Two managers from each of the 95-90% Active Share buckets 36
Active Share Portfolio Structuring and Combining Multiple High Active Share Managers The Active Share of the entire portfolio should be considered when looking for a new manager and or structuring a portfolio. High Active share managers (vs. their style specific benchmark), when combined, can result in a lower active share portfolio versus the style neutral benchmark. Large Cap Core Equity 100-95 95-90 90-85 85-80 80-75 75-70 70-65 65-60 60-55 55-50 <50 The resulting portfolio of combined Large Growth and Large Value mangers results in a total portfolio with an Active Share of 76% when compared to the Russell 1000, which is less than 42% of his peers. 0% 5% 10% 15% 20% Percentage of Managers within Universe Large Cap Core Managers 37
Blending Active Share and Excess Returns One way to analyze active share data is in the context of excess returns. Looking at average active share and average excess returns informs us of how efficient a managers has been with bets vs. the benchmark. Active Efficiency = n i=1 R e /AS n How this data can be applied: Consultants should prefer managers who generate the highest excess returns per deviation from the benchmark Managers (especially those with lower active share) should embrace this as a way to demonstrate their skill
Active Share Efficiency Active Share Efficiency Active Share Efficiency Distribution Analysis An Equal Weighted S&P 500 portfolio had an Active Efficiency of 7.1 Large Cap Core Equity Small Cap Core Equity > 5 4 : 5 3 : 4 2 : 3 1 : 2 0 : 1-1 : 0-2 : -1-3 : -2-4 : -3-5 : -4 < -5 > 5 4 : 5 3 : 4 2 : 3 1 : 2 0 : 1-1 : 0-2 : -1-3 : -2-4 : -3-5 : -4 < -5 0% 5% 10% 15% 20% Percentage of Managers within Universe Large Cap Core Managers 0% 5% 10% 15% 20% Percentage of Managers within Universe Small Cap Core Managers Average 3 Year excess return and average active share over past 12 quarters, ending March 31, 2014
Using Peers as the Benchmark In the event that more managers adopt an non benchmark-oriented style, measuring results becomes difficult. evestment has developed Peer Indices, Peer Attribution, and Peer Share as a potential solution for this issue. evestment Peer Index: an index constituted of securities that represent the average weights of active managers in a space Peer Attribution: running a holdings-based Style/Sector/Region attribution analysis versus an evestment Peer Index Peer Share: Applying active share calculation to the Peer Index, thus measuring how unique a manager is vs. peer group
Q & A
Questions and Answer Session Please enter your questions into the panel on your screen Thank you for attending this webinar If you are interested in learning more about how evestment s solutions can help you make more intelligent decisions, email us at info@evestment.com www.evestment.com