November 2016 WHITE PAPER A Snapshot of Active Share With the rise of index and hedge funds over the past three decades, many investors have been debating about the value of active management. The introduction of style analysis by Sharpe and others has fundamentally changed the way managers performance is assessed. But in recent years, a new metric, active share, has emerged as another popular measurement. In this paper, we discuss: what is active share; active share and fund performance; how it compares to tracking error; limitations of active share; and how active share is used in our investment process. What Is Active Share? As defined by Martijn Cremers and Antti Petajisto 1, active share represents the share of portfolio holdings that differs from benchmark index holdings. Active share emphasizes how individual stocks are weighted differently in a portfolio compared with a benchmark. It is calculated as half the sum of the absolute active weights of all securities in a portfolio: NN AAAAAAAAAAAA SShaaaaaa = 1 2 WW ffffffff,ii WW IIIIIIIIII,ii ii=1 where WW ff,ii = weight of the i-th security in the fund; WW IIIIIIIIII,ii = weight of the i-th security in the index; and NN = the number of all the securities in either the index or the fund. A fund with high active share has little overlap with the benchmark and thus is truly actively managed. For long-only, fully invested portfolios, active share is always between 0% and 100%. More specifically, Cremers and Petajisto believe funds with an active share less than 20% are pure index funds; those with an active share between 20% and 60% are what they call closet indexers nonindex funds with relatively low active share; and those with active share above 60% and below 100% are generally actively managed. However, we believe active share thresholds are guidelines only and should be adjusted based on the benchmarks chosen. Also, if the portfolio holds short positions, their active share could be greater than 100%. Active Share and Fund Performance Previous literature has shown that, on average, the long-term net performance of actively managed mutual funds is similar to the performance of passively managed funds. In his 1991 paper, 2 Dr. William Sharpe concluded that before costs, the return on the average actively managed dollar will equal the return on the average passively managed dollar and after costs, the return on the average actively managed dollar will be less than the return on the average passively managed dollar. However, some authors 3 argue that a subset of actively managed funds are able to consistently outperform on average over a fairly long period of time. In our opinion, one key conclusion from the Cremers and Petajisto paper is that, as a practical matter, having 1 K. J. Martijn Cremers and Antti Petajisto, How Active is Your Fund Manager? A New Measure that Predicts Performance, Review of Financial Studies 22, no. 9 (2009): 3329-65. 2 William F. Sharpe, The Arithmetic of Active Management, The Financial Analysts Journal 47, no. 1 (January/February 1991): 7-9. 3 These authors would include R.B. Cohen, J.D. Coval, and L. Pastor, Judging Fund Managers by the Company They Keep, Journal of Finance 60, no. 3 (2005): 1057 96; M. Kacperczyk, C. Sialm, and L. Zheng, On the Industry Concentration of Actively Managed Equity Mutual Funds, Journal of Finance 60, no. 4 (2005): 1983 2011; H. Mamaysky, M. Spiegel, and H. Zhang, Estimating the Dynamics of Mutual Fund Alphas and Betas, Review of Financial Studies 21 (2008): 233 64; M. Kacperczyk and A. Seru, Fund Manager Use of Public Information: New Evidence on Managerial Skills, Journal of Finance 62 (2007): 485 528; K.J.M. Cremers and A. Petajisto, How Active Is Your Fund Manager? A New Measure that Predicts Performance, Review of Financial Studies 22, no. 9 (2009): 3329 65.
a large active share is a necessary condition for long-term outperformance in actively managed portfolios. In their study, funds with active share above 90% outperformed their benchmarks by about 1% a year, net of fees. In addition, Petajisto s 2013 study 4 shows that among U.S. equity mutual funds, those with active share below 60% fall into the lowest-performing quintile, and among global and international equity funds, those with active share below 70% fall into the lowest quintile. High active share funds are well positioned to outperform in large part because it is necessary to be different from the benchmark in order to outperform. There are two main ways for an active manager to add value: stock selection and factor timing. Stock selection involves taking active bets on individual stocks. Factor timing, also known as tactical asset allocation, involves taking time-varying bets on broader factor portfolios. Active share focuses on the stock selection perspective. Although high active share is a necessary condition for outperformance, we believe it is not a sufficient one. Having high active share alone does not guarantee good performance, as some authors have indicated. 5 Other factors should also be considered. For example, Cremers and Pareek (2014) 6 found that in the long run, high active share portfolios outperform consistently if they use patient investment strategies, that is, if they hold stocks for a relatively long period. Their paper shows that for the 27-year period from 1986 to 2012, stocks held by high active share and patient institutional portfolios outperform by 2.22% annually on average, with a strong t-statistic of 4.10. When compounded over the entire period, this would generate cumulative outperformance of 80%. To emphasize their point, they argue that holding stocks for a relatively long time alone does not generate outperformance on average. Rather, active share matters most among those that have long holding durations. Their study further suggests that the clear majority of outperformance of the patient and active mutual fund managers seems due to their picking safe, high value, and quality stocks, and then holding on to these for a long period of time, which almost sounds to us like a Warren Buffet investment philosophy. Together, the two characteristics help distinguish managers pursuing short-term mispricing from those who stick with their convictions of long-term value stocks. Chart 1 Median Active Share, Fund Holding Duration, Fund Turnover Ratio of Mutual Funds Active Share, Percent 82 80 78 76 74 72 70 68 Active Share Fund Turnover Ratio Source: Cremers and Pareek (2014), PNC Fund Duration 2000 2002 2004 2006 2008 2010 2012 Using the data provided by Cremers and Pareek (2014), we plot Chart 1 showing the trend of median fund holding durations, median turnover ratios, and median active share of U.S. mutual funds. The chart shows that from 2010 to 2013, the median active share among mutual funds has been slowly increasing, while the median turnover ratio declined and the median fund duration holding time trended up. 1.9 1.7 1.5 1.3 1.1 0.9 0.7 0.5 Fund Duration, Turnover Ratio 4 Antti Petajisto, Active Share and Mutual Fund Performance, Financial Analysts Journal 69, no. 4 (July/August 2013): 73-93. 5 Todd Schlanger, Christopher B. Philips, and Karin Peterson LaBarge, The Search for Outperformance: Evaluating Active Share, Vanguard Research (May 2012). 6 K. J. Martijn Cremers and Ankur Pareek, Patient Capital Outperformance: The Investment Skill of High Active Share Managers Who Trade Infrequently (Working Paper, 2014). 2
Chart 2 Two Dimensions of Active Management Source: Cremers and Petajisto (2009), PNC Active Share and Tracking Error Active share is frequently compared with another traditional measure of active management, realized tracking error, which is the standard deviation of the difference between the return on the investment portfolio and the return on the benchmark. Both active share and tracking error measure how different a portfolio is from a perspective benchmark, but the similarity ends there. Some may argue that active share measures how active or passive a portfolio is regardless of risk, while tracking error measures how distinctive the portfolio risk is regardless of return. 7 One wants to be as different as possible from the benchmark from a return standpoint, but as similar as possible to the benchmark from a risk standpoint. Another difference between the two measures is that tracking error is based on a historical return series, while active share is a snapshot at a given point in time. A large tracking error can often be seen as an alarm for potential risk. Funds that take large and concentrated bets tend to exhibit larger tracking errors and are more volatile in general. However, we think being different from the benchmark does not always lead to large tracking errors: for example, portfolios that take many active bets (likely having a large active share) may have small tracking errors if they are well diversified. In our view, this is mainly due to tracking errors incorporating the covariance matrix of returns. In this regard, Cremers and Petajisto takes active share and tracking error as two key dimensions of active management (Chart 2). For example, they believe that a systematic factor betting strategy would generate large tracking error even without large deviation from index holdings, yet concentrated stock pickers take positions in individual stocks as well as systematic factors, thus generate both high active share and high tracking error. What Cremers and Petajisto mean by systematic factors here are factors defined in the Carhart four-factor model. The Carhart four-factor model is an extension of the Fama-French three factor model, adding a momentum factor to the size, value, and market sensitivity factors. It can be expressed as: EXR t = α + β mkt EXMKT t + β HML HML t + β SMB SMB t + β UMD UMD t + ε t EXR t is the portfolio s return in excess over the risk-free rate; α is the four-factor alpha; EXMKT t is market index return in excess over the risk-free rate; HML t is the premium of the book-to-market factor; SMB t is the premium of the size factor; and UMD t is the premium on winners minus losers, the momentum factor. Limitations of Active Share Active share also has its limitations and we believe can be meaningless if used the wrong way. 7 Laton Spahr, Forbes, http://www.forbes.com/sites/oppenheimerfunds/2013/11/20/rethinking-active-share-and-tracking-error/. 3
First, active share can be artificially skewed if: an inappropriate benchmark is used. For example, it makes no sense to calculate the active share of a domestic equity fund against an international equity index. Even though this would produce a large active share, because the two universes do not really overlap it does not say anything about the manager s skills. there is a mismatch of multiple security types that represent a single company. For example, if the benchmark holds Berkshire Hathaway class A and the portfolio holds Berkshire Hathaway class B, then treating them as two different holdings would increase active share, which is not a result of active management. the index is too broad and includes too many constituents, for instance, a world index. In this case, a global fund compared with the world index likely may find it easier to reach a high active share than a domestic fund compared with a domestic index. Consequently, the dividing line between closet indexers and actively managed funds should be shifted upward for the global fund. All in all, we believe the 20% and 60% thresholds defined by Cremers and Petajisto are basic guidelines only; when different types of managers are considered and different indexes are used, such thresholds should be adjusted accordingly. Second, when comparing active shares across funds, it only makes sense to compare peer funds and compare them to the same benchmark. It is important to note that even if funds are matched up with their appropriate benchmarks, the comparison of active share is only meaningful if the same benchmark is used. This is because even for indexes designed for the same style investing, the differences in their calculation methodologies would result in wide differences of stock concentrations and number of constituents. In turn, this would affect active share values. Table 1 compares the top 20 stock concentration of 12 indexes, in other words, the total weight of the 20 heaviest weighted stocks in each index. It is evident to us that indexes within the same style category, for example S&P 600 Growth and Russell 2000 Growth, can differ widely in terms of their concentration and the number of securities held. All those factors have heavy influence on active share Table 1 Benchmark Concentration Summary As of April 8, 2015 Benchmark Weight in Largest Number of 20 Holding Sectors* Constituents MSCI EM Latin America 54.75% 119 S&P 500 29.02% 505 Russell 1000 25.98% 1,001 S&P 600 9.33% 601 Russell 2000 4.73% 1,967 S&P 600 Growth 16.36% 352 Russell 2000 Growth 8.55% 1,163 S&P MidCap 11.25% 400 Russell MidCap 8.69% 799 MSCI EM Index 28.96% 834 MSCI EAFE 18.75% 927 MSCI ACWI 14.64% 2,471 *sorted Source: MSCI; Standard & Poor s; Russell via FactSet Research Systems Inc.; PNC 4
calculations. Thus, the comparison of funds active share is only meaningful if the same benchmark is used. Also, because we have to make sure the appropriate benchmark is used for each fund, such a comparison is only plausible for peer funds. Third, active share does not capture every aspect of active management, and we believe it should be used together with other measures. For example, active share does not distinguish between many small active bets and a few big active bets. In addition, it does not take into consideration variances and covariances. Active Share and the PNC Investment Process PNC believes investors should employ an approach that evaluates both quantitative and qualitative factors of managers skills. Our Investment Advisor Research (IAR) group has identified five different areas of an organization that it uses in selecting high-quality investment management firms: investment professionals; investment process; strategy performance; business and operational structure; and legal and compliance. Once a manager has been selected, it is then necessary to monitor that manager s performance. Part of this monitoring might include a review of both shorter- and longer-term performance trends relative to benchmarks and peers or a review to determine whether any major changes have been made to the management team or investment process. This process will help investors uncover any potential issues from a performance perspective sooner rather than later. For more information regarding PNC s due diligence process and some of the qualities that an investor should consider when examining a manager, please see the IAR January 2014 white paper, Selecting the Managers: Research and Due Diligence Process. As for the quantitative part of our investment process, we have been using sophisticated risk management software to forecast ex ante tracking error for our recommended asset allocation profiles. In addition, active share is often considered directly or indirectly (via our proprietary optimization methodology) as part of our portfolio construction process. PNC Investment Advisor Research group s Equity Quarterly Review tracks active shares of equity mutual funds and separately managed accounts that are available on the PNC IAR platform. Yet we again stress that active share is only one way of monitoring active management, and it is not a perfect indicator. It should be used in combination with other risk management/active management measurements to attain a thorough understanding of managers investment strategies. Conclusion In our opinion, active share is a useful addition to the toolkits for active management evaluations. It gives investors another way to compare portfolios against their benchmarks. In fact, some studies have shown that high active share and patient investing portfolios tend to outperform over the long term on average. In our view, an appropriate level of active share is a necessary but not sufficient condition to choose a manager. For better usage of active share, an investor should confirm that appropriate benchmarks are chosen and holdings are defined on the level of immediate issuers. When comparing funds active shares, we think it is only meaningful to do so if peer funds are compared and the same benchmark is used. It follows that the active share ranges as defined by Cremers and Petajisto for pure indexer and closet indexer do not hold across all types of managers. Overall, active share captures only one aspect of active management and has its limitations. A comprehensive evaluation of active management should rely on a broad selection of quantitative and qualitative measurements, as practiced by PNC Investment Advisor Research. 5
Bill Stone, CFA, CMT Chief Investment Strategist Chen He, CFA Senior Portfolio Strategist Paul White, PhD, CAIA Director of Portfolio Strategy The PNC Financial Services Group, Inc. ( PNC ) uses the marketing name PNC Wealth Management to provide investment, wealth management, and fiduciary services through its subsidiary, PNC Bank, National Association ( PNC Bank ), which is a Member FDIC, and to provide specific fiduciary and agency services through its subsidiary, PNC Delaware Trust Company or PNC Ohio Trust Company. Standalone custody, escrow, and directed trustee services; FDIC-insured banking products and services; and lending of funds are also provided through PNC Bank. This report is furnished for the use of PNC and its clients and does not constitute the provision of investment advice to any person. It is not prepared with respect to the specific investment objectives, financial situation, or particular needs of any specific person. Use of this report is dependent upon the judgment and analysis applied by duly authorized investment personnel who consider a client s individual account circumstances. Persons reading this report should consult with their PNC account representative regarding the appropriateness of investing in any securities or adopting any investment strategies discussed or recommended in this report and should understand that statements regarding future prospects may not be realized. The information contained in this report was obtained from sources deemed reliable. Such information is not guaranteed as to its accuracy, timeliness, or completeness by PNC. The information contained in this report and the opinions expressed herein are subject to change without notice. Past performance is no guarantee of future results. Neither the information in this report nor any opinion expressed herein constitutes an offer to buy or sell, nor a recommendation to buy or sell, any security or financial instrument. Accounts managed by PNC and its affiliates may take positions from time to time in securities recommended and followed by PNC affiliates. PNC does not provide legal, tax, or accounting advice unless, with respect to tax advice, PNC Bank has entered into a written tax services agreement. PNC does not provide services in any jurisdiction in which it is not authorized to conduct business. PNC Bank is not registered as a municipal advisor under the Dodd-Frank Wall Street Reform and Consumer Protection Act ( Act ). Investment management and related products and services provided to a municipal entity or obligated person regarding proceeds of municipal securities (as such terms are defined in the Act) will be provided by PNC Capital Advisors. Securities are not bank deposits, nor are they backed or guaranteed by PNC or any of its affiliates, and are not issued by, insured by, guaranteed by, or obligations of the FDIC, the Federal Reserve Board, or any government agency. Securities involve investment risks, including possible loss of principal. PNC Wealth Management is a registered service mark of The PNC Financial Services Group, Inc. 2016 The PNC Financial Services Group, Inc. All rights reserved. 6