Patient Capital Outperformance

Transcription

1 Discussion of Mikhail Simutin University of Toronto ICPM Discussion Forum June 9, 2015

2 Cremers and Pareek (2015): Overview Interesting paper that bridges three important areas of institutional money management research 1 Understanding the impact of holding horizons (patience) on fund performance 2 Quantifying managerial activeness and its impact on performance 3 Adjusting fund performance for risk

3 Holding Horizon and Fund Performance How might holding horizon/low turnover/patience impact fund performance? Positively Mispricing may take a long time to reverse Frequent trading is costly My favorite time frame for holding a stock is forever I buy on the assumption that they could close the market the next day and not reopen it for five years. Warren Buffett Negatively If information is quickly incorporated into prices, skilled managers have to trade frequently Prevalence of algorithmic trading

4 Holding Horizon and Fund Performance Early literature on portfolio turnover offers mixed evidence Negative relation: Elton, Gruber, Das, and Hlavka (1993) and Carhart (1997) No relation: Wermers (2000) and Edelen, Evans, and Kadlec (2007) Positive relation: Dahlquist, Engstrom and Soderlind (2000) and Chen, Jagadeesh and Wermers (2001) Current literature applies new definitions of holding horizons / turnover Lan and Wermers (2015) Pastor, Stambaugh, and Taylor (2015) Cremers and Pareek (2015)

5 Cremers and Pareek (2015): Main Results Active Share Fund Duration Uncond Uncond (4.49) (3.97) (3.64) (1.74) (0.45) (2.47) (3.87) (5.52) (4.01) (2.57) (2.13) (2.45) (0.57) (3.20) (4.39) (3.84) (3.37) (1.20) (0.50) (1.64) (2.39) (5.39) (2.47) (3.33) (0.49) (0.37) (1.74) (1.96) (3.17) (3.94) (1.20) (1.07) (1.08) (2.89) (0.98) (4.44) (2.84) (3.09) (0.75) (3.14) (4.78) (3.46) (2.99) (1.83) (0.57) (1.80) (5.18)

6 Managerial Activeness An important concept Activeness can relate to future fund performance Interest among the practitioners to screen funds on activeness Managers get paid active management fees, so we expect them to be active Activeness is typically computed with respect to a benchmark: Industry concentration of Kacperczyk, Sialm, Zheng (2005) Active share of Cremers and Petajisto (2009) R-squared of Amihud and Goyenko (2013) Active weight of Doshi, Elkamhi, and Simutin (2015)

7 Managerial Activeness Every mutual fund manager must make two important decisions when creating a portfolio: 1 Select assets from the universe of suitable investments given a fund s investment objective and benchmark 2 Assign weights to each selected asset Ideally, we want a measure of managerial activeness to capture skill from both decisions Inferring skill from the first decision is empirically challenging: Requires knowledge of the universe of suitable investments Requires knowledge of the actual rather than stated benchmark A third of funds do not follow the objective stated in the prospectus (Sensoy, 2009) Activeness associated with the first decision can be faked

8 Managerial Activeness Doshi, Elkamhi, and Simutin (2015): Propose to infer managerial activeness from the second (weighting) decision A question: If you were a closet indexer, how would you weight stocks in the portfolio? Almost certainly, you would value-weight the portfolio Market indices, exchange-traded funds, and the benchmarks used by funds are almost exclusively value weighted Value weighting is increasingly dominating mutual fund investing, so an active manager needs to deviate from value-weighting to outperform peers (Bhattacharya and Galpin, 2011) We propose computing activeness by comparing portfolio weights with value-weights

9 Active Weight Active weight it = 1 w j 2 j it w jm it After fund expenses Before fund expenses 1.98 Four-factor alphas, percent monthly Low High

10 Advantages of Active Weight Simple to compute: only requires knowledge of fund holdings and their market capitalizations No need to determine benchmarks! Can be easily applied in setting where relying on other measures may be difficult: Pension, endowment, hedge, and other funds that need not disclose returns frequently and that may change their benchmarks periodically International funds for which imputing benchmarks or selecting asset pricing models is particularly difficult Young funds with short return histories Predicts fund flows, fund asset growth, factor-adjusted performance, and value added The predictive ability of active weight is distinct from that of other measures The only measure that predicts performance at longer horizons

11 Adjusting Fund Performance for Risk Cremers and Pareek (2015): Profitability of high active share, patient managers stems in large part from exposure to a betting against beta factor Frazzini, Kabiller and Pedersen (2013): Buffett s alpha stems in large part from this exposure as well Cumulative CAPM Alpha, Percent Low Beta2 Beta3 Beta4 High

12 Betting Against Beta and Institutional Managers Baker, Bradley, and Wurgler (2011) posit benchmarking as a theoretical reason for the persistence of the anomaly Logic: Benchmarked managers tilt to high-beta stocks to try to beat the benchmark, and drive the price of these stocks up They acknowledge that testing this theory is challenging Christoffersen and Simutin (2015): Study equity mutual funds managing retirement money and facing benchmarking pressure from pension plan sponsors Show that funds with a greater proportion of sponsorcontrolled retirement money tilt portfolio to high-beta stocks Provide direct empirical evidence that benchmarking encourages investment in high-beta stocks and limits the appetite for low-beta stocks

13 Conclusion Very interesting paper Bridges several strands of literature Managerial activeness Holding horizons / turnover Risk adjustment / betting against beta Final word from Warren Buffett: According the name investors to institutions that trade actively is like calling someone who repeatedly engages in one-night stands a romantic