Mutual/Hedge Funds

Should investors believe that the financial industry can offer low-cost, liquid funds that reliably mimic Commodity Trading Advisor (CTA) hedge funds? In their June 2016 paper entitled “Just a One Trick Pony? An Analysis of CTA Risk and Return”, Jason Foran, Mark Hutchinson, David McCarthy and John O’Brien identify and examine performances of CTA-like hedge funds across eight distinct categories defined via iterative correlation clustering. Their goal is to determine whether category performance is amenable to modeling (cloning) via liquid exposures to four futures risk factor premiums:

1. Value – long (short) high-value (low-value) futures, with “value” based on book-to-market ratios for stock index futures and 5-year change in yields/spot prices/purchasing power for government bonds/commodities/currency forwards.
2. Carry – long (short) futures with high (low) roll returns.
3. Time series momentum – long (short) futures with positive (negative) 12-month past returns.
4. Options-based trend following – from Fung and Hsieh, correlated with trends shorter than time series momentum.

They estimate these premiums from monthly returns of rolling nearest contracts for: 12 global equity index futures series; eight global 10-year government bond synthetic futures series; 22 commodity futures series; and, nine global currency forward series versus the U.S. dollar. They employ a hedge fund screening process that suppresses backfill bias (lucky starts). Using monthly net returns and assets under management (AUM) for specific (not fund-of-funds) and distinct CTA funds with at least 12 months of returns denominated in U.S. dollars and monthly data required to estimate futures risk factor premiums as available during January 1987 through July 2015, they find that: Keep Reading

How well do mutual funds and exchange-traded funds (ETF) designed to track hedge fund indexes work? In their October 2015 paper entitled “Synthetic Hedge Funds”, Mario Fischer, Matthias Hanauer and Robert Heigermoser examine the performance of synthetic hedge funds, defined as open-end mutual funds and ETFs that explicitly employ hedge fund indexes as their primary benchmarks. They assess replication success: (1) based on both return distribution shapes and risk-adjusted performance; and, (2) overall, for mutual funds and ETFs separately as groups, and by specific hedge fund strategy (when enough synthetic funds exist for a strategy). They group funds via value-weighted portfolios. Using monthly returns for 72 synthetic hedge funds (52 mutual funds and 20 ETFs) and associated Credit Suisse hedge fund index benchmarks during January 2009 through December 2013, they find that: Keep Reading

Is selecting mutual funds based on strong performance over the last three years helpful (discovering fund manager skill) or harmful (signaling imminent fund strategy mean reversion)? In the February 2016 version of their paper entitled “The Harm in Selecting Funds that Have Recently Outperformed”, Bradford Cornell, Jason Hsu and David Nanigian investigate future mutual fund performance based on recent past performance relative to stated benchmarks. They focus on a past performance interval of three years because: institutional consultants cite this measurement as one of the most important criterion for fund selection; and, Morningstar’s rating algorithm emphasizes three-year past performance. Specifically, every three years they:

1. Rank funds by expense ratio and exclude the highest tenth as likely poor choices.
2. Define Winner, Median and Loser funds as the tenths of the rest with the highest, middle (centered on the 50th percentile) and lowest benchmark-adjusted returns the past three years.
3. Track the performance of the equally weighted and monthly rebalanced Winner, Median and Loser groups over the next three years.

Using benchmark-adjusted returns for actively managed U.S. equity mutual funds during January 1994 through December 2015, they find that: Keep Reading

Mebane Faber states in the first chapter of his 2016 book Invest with the House: Hacking the Top Hedge Funds: “We make two assumptions…: 1. There are active managers that can beat the market… 2. Superior active managers can be identified. …There is a general feeling that the market can’t be beat, and it is tough to get past that belief. A big challenge is separating luck from skill. But would anyone deny that some people are better than others at stock picking? Just like any other profession, the investment field has top experts who are paid handsomely for what they do. …You have access to the stock picks made by fund managers who often spend millions of dollars and every waking moment thinking and obsessing about the financial markets. …The best ones know everything there is to know about a company before they invest. …You can then build a stable of these managers and use them…for stock ideas to research and possibly implement in your own portfolio.” Based on prior research/experience and performances of the top ten (long) holdings from quarterly Form 13F filings for selected fund managers during January 2000 through December 2014, he concludes that: Keep Reading

Who are the givers and who are the takers among mutual funds and hedge funds? In their January 2016 paper entitled “Style and Skill: Hedge Funds, Mutual Funds, and Momentum”, Mark Grinblatt, Gergana Jostova, Lubomir Petrasek and Alexander Philipov analyze quarter-to-quarter changes in Form 13F stock holdings to assess investment styles and sources of performance for hedge funds and mutual funds. They focus on the interaction between portfolio weight changes and future stock returns to measure investing skill. They calculate fund alpha via adjustments for stock size, book-to-market ratio and (when appropriate) momentum. Using quarterly 13F filings of 589 mutual funds and 1,342 hedge funds during 1998 to 2012, they find that: Keep Reading

What are the principal themes of research on hedge funds published in top journals over the past decade? In their August 2015 paper entitled “Hedge Funds: A Survey of the Academic Literature”, Vikas Agarwal, Kevin Mullally and Narayan Naik summarize 121 papers on hedge funds and commodity trading advisors from four leading finance journals. They focus on the 105 papers published since 2005. They organize this research into five categories:

1. Fund performance over time and by type, including return drivers, risks and assessment of manager skill.
2. Relationships between fund characteristics (such as contractual terms, size, age and manager background) and fund performance.
3. Investor risks, including manager incentives and capital flows.
4. Role of hedge funds in the financial system.
5. Biases in and limitations of data.

Based on this review, they conclude that: Keep Reading

What is the state of the hedge fund industry? In the July 2015 draft of their paper entitled “Hedge Funds: A Dynamic Industry In Transition”, Mila Getmansky, Peter Lee and Andrew Lo review recent academic research on hedge funds and update industry performance statistics. They surmise that  hedge fund data from 10 years ago may be unrepresentative of today’s environment, especially in the aftermath of the 2007-2009 financial crisis. Their review considers four perspectives: investor, portfolio manager, regulator and academician. Based on this review and self-reported hedge fund performance data during January 1996 through December 2014, they conclude that: Keep Reading

Should investors adopt a mutual fund for the long term, or should they occasionally switch to funds with fresh ideas and energy? In the July 2015 draft of their paper entitled “Milk or Wine: Mutual Funds’ (Dis)economies of Life”, Laura Dahm and Christoph Sorhage investigate whether mutual fund performance tends to decline or improve with age. They measure fund performance via four alphas: Jensen’s; three-factor (market, size, book-to-market); four-factor (adding momentum) alpha; and, five-factor (adding liquidity). All alpha calculations employ 36-month rolling window regressions of net fund returns in excess of the risk-free rate. The regression methodology allows measurement of the performance difference between a mature fund and its younger self. Using returns, characteristics and holdings data for 3,489 actively managed U.S. domestic equity funds during 1991 through 2014, they find that: Keep Reading

“Measuring the Level and Persistence of Active Fund Management” (pro) and “Fund Activeness Predicts Performance?” (con) summarize debate on the ability of Active Share, how much portfolio holdings differ from a benchmark index, to predict mutual fund performance. The authors of the con paper summarized in the latter (principals of AQR Capital Management) assert that “neither theory nor data justify the expectation that Active Share might help investors improve their returns.” In his June 2015 paper entitled “AQR in Wonderland: Down the Rabbit Hole of ‘Deactivating Active Share’ (and Back Out Again?)”, Martijn Cremers rejoins the debate by examining the methodology and motives of the con paper. Using data on active U.S. equity mutual funds from the original research, and holdings/performance data for seven AQR Capital Management funds offered to retail investors that concentrate in U.S. stocks as available through December 2014, he finds that: Keep Reading

Do any mutual funds reliably generate significant alpha and, if so, do fund investors receive this alpha? In their June 2015 paper entitled “Active Managers Are Skilled”, Jonathan Berk and Jules Van Binsbergen examine interactions among equity mutual fund gross alpha, assets under management, fees and net alpha. To measure a practical gross alpha, they benchmark active mutual fund gross performance against an historical best-fit linear combination of net returns from contemporaneously available Vanguard funds. To account for the effects of mutual fund size, they measure monthly dollar value added by the fund manager as gross alpha times assets under management. This approach accounts for competition among funds, whereby investors chase an outperforming fund until its alpha drops to zero. They then estimate fund manager skill as average monthly valued added divided by the standard error of the monthly value added series. Using gross monthly returns and fees for a broad survivorship bias-free sample of of active equity mutual funds and net monthly returns for Vanguard mutual funds during January 1977 through March 2011, they find that: Keep Reading