Journal of Financial Planning; November 2014
David Nanigian, Ph.D., is an associate professor of investments at The American College. He is internationally known for his research on mutual funds and teaches a course on mutual funds in The American College’s Master of Science in Financial Services program. Learn more about his research at ssrn.com/author=843211.
In the mutual funds literature, it has become a virtually stylized fact that, on average, mutual funds with greater operating expenses generate lower net-of-expense returns to their shareholders. Such findings have spurred the trend toward low-cost passive investing. The purpose of this month’s column is not to advocate one particular portfolio management strategy over another, but rather to suggest some additional ways to think about which strategy to choose.
Consider mutual funds to be players or teams in a big game. We will call this the Money Management Game. In this game, one’s decision of whether or not to bet on a particular player or team should depend on how strong he or she is at the sport and on how strong his or her opponents are. Let’s summarize the practical implications of selected research studies on how to evaluate the strength of the players and their opponents in the Money Management Game.
Evaluating the Strength of the Players
In a 2013 paper (“Active Share and Mutual Fund Performance” published in the Financial Analysts Journal) Antti Petajisto shows that “closet indexing,” in which a fund claims to adhere to an active portfolio management strategy but tracks its benchmark index instead while raking in the higher fees of active management, has become increasingly prevalent among mutual funds. Petajisto shows that around one-third of the money invested in mutual funds is now closet indexed.
Engagement in closet indexing may occur because a fund has become too large to take concentrated positions in individual securities without incurring overwhelming transactions costs, a challenge that I allude to in the Investment Management column in the May 2014 issue of the Journal of Financial Planning. It may also occur because fund managers face tracking error volatility constraints (tracking error volatility refers to the volatility of benchmark-adjusted returns over time). Richard Roll illuminated this issue in a seminal 1992 Journal of Portfolio Management paper, “A Mean/Variance Analysis of Tracking Error,” yet it continues to be a problem in delegated portfolio management. Alternatively, closet indexing may occur due to managerial slack arising from ineffective monitoring by fund shareholders. I developed a theory on this in the 2012 Financial Services Review paper, “Why Do Mutual Fund Expenses Matter.”
Practitioners can use metrics such as active share or the R-squared statistic from an asset pricing model regression to gauge just how actively a fund is managing its portfolio of securities. Active share denotes the percentage of a fund’s portfolio holdings that do not overlap with those in its benchmark index. R-squared indicates the percentage of the variation in a fund’s returns over time that can be attributed to variation in the returns on its benchmark index over time. A low active share or a high R-squared value is indicative of closet indexing. In contrast, a high active share or a low R-squared value indicates that a fund is taking a more active approach to portfolio management. Financial planners should pay close attention to these measures, because recent research shows that more actively managed funds outperform their counterparts.
In a 2009 paper by Martijn Cremers and Antti Petajisto (“How Active Is Your Fund Manager? A New Measure That Predicts Performance” published in The Review of Financial Studies), the authors sort funds into groups based on their quintile rank of active share and find that active share has a positive impact on performance. For example, the average benchmark-adjusted net-of-expense return is monotonically increasing from –1.42 percent per year to 1.13 percent per year across the five portfolios of funds.
Similarly, in a more recent paper (“Mutual Fund’s R² as Predictor of Performance” published in 2013 in The Review of Financial Studies) Yakov Amihud and Ruslan Goyenko sort funds into portfolios based on their quintile rank of R-squared from a Carhart four-factor model regression and find that fund performance is monotonically decreasing in R-squared. These recent research studies suggest that financial planners should look beyond expense ratios when evaluating mutual funds and consider just how actively they are managing their portfolios
Evaluating the Strength of the Opponents
In the seminal 1980 The American Economic Review paper, “On the Impossibility of Informationally Efficient Markets,” Sanford Grossman and Joseph Stiglitz show that the relative performance of active versus passive portfolio management strategies depends on the popularity of each. If nearly all money is actively managed, then there are greater returns to passive portfolio management. Money should then, rationally, flow into passive strategies. However, as money flows into passive strategies then the performance of active strategies will improve. This will stop, if not reverse, the flow of money from active to passive strategies.
To appreciate the practical relevance of Grossman and Stiglitz’s theory, consider two economies: a “99 percenter” economy in which 99 percent of market participants actively manage their portfolios (and the remainder passively manage their portfolios) and a “1 percenter” economy in which only 1 percent of market participants actively manage their portfolios. For the purpose of simplicity, assume that each market participant is endowed with the same amount of wealth.
In the “99 percenter” economy, the competition to acquire new information on future corporate cash flows is incredibly intense. Consider a “superstar” company in this economy that has been making spectacular improvements to its cash flow-generating ability, above and beyond those of its counterparts. Nearly all investors would recognize its accomplishments and its stock price would quickly increase upon any news about its phenomenal cash flow-generating ability. However, because nearly all investors in this economy quickly act upon the same set of information, they would all enjoy nearly similar returns. In this “99 percenter” economy, one is certainly better off being among the 1 percent of the population that passively manages their portfolio, because they will enjoy the fruits of the labor of the other 99 percent, yet not bear any of the cost of acquiring information.
Now consider what would happen if “Superstar” trades in a “1 percenter” economy in which only 1 percent of the population actively manages their portfolios. Let us also consider two different possible dividend policies that Superstar might employ. First, consider a policy where all of its free cash flows are simply paid out each quarter to its shareholders in the form of a dividend. The 1 percent of investors in this economy that do recognize Superstar’s accomplishments through buying its stock would enjoy larger gains to their personal wealth, in the form of dividend income, than the other 99 percent who ignore Superstar’s fine accomplishments. Now let us assume that Superstar does not pay a dividend to its shareholders, but rather reinvests all of its free cash flows back into its business. Eventually another company will recognize that its stock is trading for far less than the discounted present value of the cash flows that it is expected to generate. That company will buy Superstar out at a premium to its trading price. The “1 percenters” that recognize Superstar’s accomplishments through buying its stock will enjoy larger enhancements to their personal wealth, in the form of the takeover premium, than the 99 percent of the population that is agnostic to Superstar’s remarkable cash flow-generating ability.
If we were in a “99 percenter” economy, market participants would begin to passively manage their portfolios, because the cost of information acquisition would exceed its benefit. But if we became a “1 percenter” economy, then everyone else would want to be a “1 percenter” too. Money would then flow back into active strategies.
In reality, the shifts in the popularity of active portfolio management strategies may not be as large as those provided in this simple example. The main point is that the popularity of active strategies varies over time depending on the ferocity of the competition for acquiring information. Moreover, both the popularity of active strategies and the ferocity of the competition for acquiring information vary across investment categories.
In the Financial Planning Association’s 2014 Trends in Investing Survey, study participants were asked whether actively or passively managed funds provide the best overall investment performance, after taking into account the costs associated with the portfolio management strategy. Eighteen percent responded that actively managed funds provide the best performance, and 25 percent said passively managed funds are the best. Both responses are incorrect. The correct response, although not an option on the survey, is “it depends.” It depends on just how actively managed a fund is, and on the intensity of the competition for acquiring information. Fifty-seven percent of survey participants responded that “a blend of active and passive management provides the best overall investment performance.” Maybe I would give partial credit if one of my students provided this response on an exam.
In each of the FPA Trends in Investing Surveys from 2006 through 2014, mutual funds ranked as the most popular investment vehicle used by financial advisers. Surely advisers have more to say to their clients about mutual fund selection than “just buy a cheap passively managed fund.” Here I’ve provided a framework of how to think about choosing a portfolio management strategy. However, I’ll refrain from suggesting what strategies are currently the best for particular investment categories, because if everyone followed my advice then it would prove to be wrong.