Chasing Tail

Chasing Our Tails Through Past Performance

Chasing Tail

For those who have been investing for some time, you should be very familiar with the following statement: “past performance is no guarantee of future results.” This SEC mandated disclaimer when representing any past performance data is not without cause. Although there have been many studies refuting the idea that past performance is indicative of future results, many investors still use past performance to make their investment decisions, often selling their recently poor performing mutual funds to buy recent winners.

There is a very rational explanation for this behavior. First, it seems like common sense, right? These recent winners have obviously proven that they can do well amongst their peers and should be a good investment going forward. Second, many investors and even investment consultants rely on Morningstar to be able to efficiently sort through the over 5,000 open-end mutual funds in any given year and give a recommendation, often over-relying on their star rating system.  Lastly, the now $13 trillion mutual fund industry is a marketing machine, flashing their recent performance in television commercials and money magazines.

Unfortunately, proven track records end up being random outcomes, 5 star-rated mutual funds will fall from grace, and advertisements are nothing more than selfish aims, not evidence backed advice that is in your best interest. Academic research over the last two decades has reinforced the conclusion.

Let’s review some of the groundbreaking research papers that have been done:

  • Mark Carhart from the University of Chicago wrote his graduate thesis on the lack of persistence in mutual fund performance once common risk factors (size, relative price), as well as survivorship bias inherit in most mutual fund databases, are accounted for. In other words, the persistence that had been well documented in previous studies [(Hendricks, Patel, and Zeckhauser, 1993), (Goetzmann and Ibbotson, 1994), (Brown and Goetzmann, 1995, and Wermers (1996)], has not been due to any particular skill exhibited by professional mutual fund managers. They just happened to be in the right place of the market at the right time. As soon as that part of the market starts to lag, so does the manager’s performance.
  • Amit Goyal & Sunil Wahal supported Dr. Carhart’s conclusions in their paper, The Selection and Termination of Investment Management Firms by Plan Sponsors. They examined 8,755 hiring decisions made by 3,400 plan sponsors from 1994 to 2003. The most recent three year performance of managers who were hired by plan sponsors showed an average annualized excess return of their benchmark by 2.91%. Subsequently, after they were hired, they ended up underperforming their benchmark by an average of -0.47% per year. See chart below.

To add insult to injury, the researches also analyzed 660 round-trip decisions between 1996 and 2003 and found that before the hiring decision the hired outperformed the fired by 5.87% per year for the three years leading up to hiring decisions. In the subsequent 3-year time period, the recently hired managers underperformed their predecessors by 0.95% per year, indicating once again that the short-term performance figures by managers are random at best. See the chart below:

  • More recently, Vanguard submitted an article to Seeking Alpha in which the researchers attempted to simulate an investment strategy based on performance chasing versus a simple buy and hold strategy. They chose from the universe of U.S. Equity mutual funds available in any of the nine equity style boxes in Morningstar’s database during the ten years ending December 31, 2013. You can read the article to find the trading/rebalancing assumptions used to simulate an actual strategy that an investor could have deployed. Although it is not actual results, it does highlight the pitfalls of using defined parameters based on past performance when making investment decisions. Here is a summary of their findings with a subsequent figure below:
    • Annualized returns were higher for the buy and hold strategy across all 9 Morningstar style boxes by an average of 2.77% per year for the 10 years ending 12/31/2013.
    • Sharpe ratios, a measure of return once adjusted for risk, was much higher for a buy and hold strategy versus a performance chasing strategy by an average of 0.12% per year. In other words, investors received 0.12% in return per year for the risk taken.
    • Both of these metrics exclude the impact of transaction costs and taxes, which would only further exaggerate the outperformance of the buy and hold strategy.

As investors, we are constantly trying to navigate through the constant financial noise that bombards us everyday. Track records can be very indicative of future value in a lot of ways. When hiring our next employee, we look for past performance results to extrapolate how we predict they are going to add value to our business. Colleges admit students almost solely based on past performance. Although it seems like common sense when it comes to thinking about investing, the analogy does not hold. Whether it be actual or simulated results, investors are better served to buy, hold, and rebalance their portfolio.