The purpose of this paper is to examine the level of success of actively managed investment strategies in US equity markets relative to index or passively managed investment strategies. Throughout this examination, we'll demonstrate why our philosophy utilizes a default position on US equity strategies that is built on indexing. For an original formatted copy of this whitepaper, please visit our website.
Indexing refers to an investment methodology that attempts to track a market index as closely as possible after taking into account all expenses to implement the strategy. The goal of the investment manager for an index fund is to replicate the returns of the index exactly after expenses and without incurring any tracking error or deviations from the index's return. Many indices weigh their stock membership by market capitalization so that the largest stocks make up the largest percentage of the index's value. The argument for passive management is that it provides near-market returns, easy diversification and a much lower cost than active strategies. Additionally, those who point to the success of passive strategies believe in market efficiency. The concept developed by William Sharpe explains that the market is a zero-sum game where every investor's holdings are represented in aggregate. Because all investors are represented in total, if one investor outperforms by one dollar, then theoretically another investor has underperformed by one dollar. Thus the dollar-weighted performance of all investors must equal the performance of the market.
Active management suggests that the investment manager's goal is to outperform their appropriately assigned index or benchmark after accounting for all expenses to perform the strategy. The rationale for selecting active managers is the potential for higher returns with lower comparable risk to the index. Skilled active managers create an investment process that find market inefficiencies around the price of individual securities and exploit that inefficiency to the benefit of investors. The key commonality of active managers is that they contain some version of a price/valuation consideration, and will gravitate away from securities that they believe are overpriced and towards securities that have greater appreciation potential. Outperformance for an active strategy is measured by alpha or excess return relative to the strategy's benchmark.
Expanding on the Zero-Sum Game
William Sharpe's theory of the zero sum game implies that one would expect to see investment strategies half successful and half unsuccessful at any point. Unfortunately, investors in strategies are exposed to commissions, management fees, bid-ask spreads, administrative costs, taxes, liquidity constraints, and other costs. In 2008, Professor Kenneth French published a study titled, "The Cost of Active Investing" that analyzed the total cost to investors who have hired active managers to be more than $100 billion. Professor French estimated this number has grown from nearly $7 billion in 1980.
(click to enlarge)
The above chart from Vanguard displays the importance of minimizing costs as the layering of additional costs on an investment strategy greatly limits the potential universe of investments that will ultimately end up outperforming a benchmark. Of note, Financial Research Corporation (2002) performed a study where they evaluated the predictive value of different fund metrics including a fund's past performance, Morningstar rating, alpha and beta. The findings of the study pointed to the fund's expense ratio being the most reliable predictor of its future performance with low-cost funds delivering above-average performance in all of the periods examined.
Another surprising study came from Morningstar's FundInvestor periodical in August 2010 titled, "How Expense Ratios and Star Ratings Predict Success." Morningstar's study supported the Financial Research Corporation's study that in each time period and data point tested, low cost funds beat high cost funds, and that expense ratios were more predictive than star ratings in 23 out of 40 observations (58%.)
Examining the Track Record of Active Managers
The table above comes from Vanguard where they analyze underperformance of US equity funds relative to their style benchmark. Benchmark selection is critical in this process because many times managers choose benchmarks that don't necessarily match their style, and that they believe they can outperform. Another key consideration is the inclusion of "dead" funds to any study. Surviving funds generally outperform funds that have been liquidated or merged, meaning only the best performing funds are in today's fund universe. The key takeaway from this table is the difficulty that investment managers have in beating their style benchmarks over multiple time periods.
Another perspective to analyze is the ideology that actively managed funds perform best during segments of market cycles such as bull and bear markets. For example, a common theory is that actively managed funds will outperform their benchmark in bear markets because they will foresee difficulties in the market and sidestep trouble spots by becoming defensive through avoiding particular sectors or holding more of their assets in cash. The results of active managers during market cycles show a similar story as our previous time period study. Avoidance of a bear market or inclusion into a bull market creates significant issue for the active manager as they are required to time the market exactly correct twice - during the start and end of each cycle. The proceeding illustrative chart from Vanguard highlights these results.
But Not All Active Managers Are The Same
Antti Petajisto's article "Active Share and Mutual Fund Performance" accepts the premise that the average actively managed strategy loses to a lower-cost index funds after all fees and expenses. However, this article examines active management and subdivides the category by using a measure of "active share" or the percentage of the portfolio that differs from the passive benchmark index and "tracking error" or the standard deviation of the fund's return and the benchmark's return. Active management in this case was subdivided into Closet Indexers, Diversified Stock Picks, Concentrated Stock Picks and Factor Bets. By examining these categories closely, we think it will shed further light on why most active strategies are not successful.
Closet Indexers follow the practice of staying close to their benchmark while claiming to be an active manager and thus charging active management fees. In the Petajisto's article, this group is defined as having less than 60% active share and a low tracking error. The problem with this group is that they are expensive relative to their offerings and have a high degree of difficulty to outperform their benchmark. Unfortunately, even when closet indexers outperform their benchmark, they do not show much persistence in continuing outperformance. The article estimates that Closet Indexing is more popular than actual indexing, whereas one third of all assets are in this category in comparison with approximately 20% in actual index products.
Diversified Stock Pickers tend to have a high percentage of Active Share but a low tracking error. This can only happen when the fund's sector weights are similar to the benchmark but focused on finding underpriced securities within each sector. An example of this type of fund is the FMI Large Cap Fund that has an Active Share of 95% and a tracking error of 5.4% relative to the S&P 500. The fund generally has only 24 stock positions, but is diversified in weights very closely to that of the index.
Concentrated Stock Pickers combine active stock selection with factor bets. Many of these active managers will have as few as 10 to 20 stocks in their portfolio and may weight some of these positions up to 10% of the strategy. An example of this strategy is the Sequoia Fund with an Active Share of 97% and a tracking error of 14.1%. This fund holds 22 stocks in total. An example in the small cap space is Longleaf Partners Small-Cap Fund with only 19 stocks in the portfolio, and an Active Share of 99% and a tracking error of 14.4% relative to the Russell 2000.
Lastly, funds that focus on Factor Bets tend to have a low active share but high tracking error. These funds tend to use more timing strategies of such factors as industries, sector, size and value. It may be common to see these funds keeping assets in cash or high quality debt instead of trying to minimize risk relative to the benchmark. Petajisto cites an example of this category to be the GMO Quality Fund with an Active Share of 65%, but a tracking error of 12.9%, and the AIM Constellation Fund with an Active Share of 66% and tracking error of 9.7%. These numbers generally reflect the significant sector bets and the application of cash during bear markets.
The study concluded that for each category, returns after fees and transaction costs found Factor Bet strategies to be poor performers, returning -1.28% per year relative to the benchmark. Closet Indexers lost -0.91%, Concentrated Funds were equal to their benchmarks, and Diversified Stock Pickers outperformed their benchmarks by 1.26% (Note this time period evaluated was from 1980 - 2009.) Of interest, there were no statistically significant differences between large cap versus small cap or during the financial crisis of 2008-2009. Please note that Active Share can be a good check on change in strategy, but shouldn't be used exclusively to determine outperformance ability.
The Persistence of Alpha
Unfortunately, once an investor has successfully located a winning strategy and manager, there is no guarantee the strategy will continue to perform as previously indicated.
(click to enlarge)
Above, Dimensional Fund Advisors (DFA,) examines the persistence of funds with an outperforming or winning strategy ending 2009.
Of recent note, the Vanguard Investment Strategy Group in May 2013 ranked all active US equity funds covering the nine Morningstar style categories based on their excess returns relative to their stated benchmark during the five-year period as of 2007 (The table is included below.) There were 5,763 total funds and 1,168 that fell into the top excess quartile as of 2007. Of those top quartile funds as of 2007, only 14.9% remained in the top quartile until 2012. Interestingly, 24% of top quintile performers fell to the bottom quartile, and 16.8% of those funds were liquidated or merged. The chance of a manager starting in the top 20% of all managers and remaining in the top 20% over a full ten-year period was demonstrated to be 3% (20% x 14.9%.) The top quintile strategy had an excess return of 2.02 percentage points during the subsequent five-year period. The next logical question is, "What happened to the bottom quintile funds? Did they become top quartile funds?" Only 13.4% of the bottom quintile became top quintile in the next five-year period. 50.4% of those bottom quintile funds were liquidated or merged within the next five-year period. This study is also being continually tracked by SPIVA, or S&P Indices versus Active Funds Scorecard. The results of SPIVA are consistent with Vanguard's findings.
(click to enlarge)
The Vanguard Investment Strategy Group also examined the effect of Morningstar rankings on performance (using Morningstar data as of December 31, 2012.) They found that upon a fund achieving a 5-star ranking, the fund had excess returns relative to their benchmark of -1.36% over the next 3-year period. Upon receiving a ranking by Morningstar, every star ranking had negative excess returns with 5-stars worse than 4-stars, which was worse than 3-stars and so on. The public's reliance on Morningstar star rankings to evaluate fund performance going forward is seen to be a faulty assumption.
Due to this public assumption around Morningstar star rankings, investors generally invest their money into 5 and 4 star ranking funds and remove their money from 3, 2, and 1 star funds. The asset bloat of previously outperforming funds tends to change the investment philosophy and/or ability to execute on the strategy and thus cause underperformance. The cash flows out of average to below average funds tend to cause a spiral of forced liquidation of positions to fund redemptions and further underperformance. Much of this behavior is attributed to herd instinct where individuals gravitate to the same or similar investments based almost solely on the fact that many others are investing in those positions. The fear of regret of missing out on a good investment is often a driving force behind herd instinct.
Of course, active managers are aware of these studies and their results. Like all businesses, investment management survives to create profits, and active managers have an incentive to convince the public of the merits of their unique investment strategy and to pay a high fee for the strategy. For example, a recent investor performance by Janus Capital, "Ten Timely Reasons For Active Equity Today" highlights Antti Petajisto's study, "Active Share and Mutual Fund Performance" that mutual funds in the top quintile of active share beat their benchmark by an average of 1.265% a year from 1990-2009. Janus shows the compounding of the top quintile's outperformance would produce an excess profit of $71,489 relative to the index based on a $100,000 investment. The conclusion of this slide implicates the ability of outperformance for active managers. This representation presents some misleading insinuations as it ignores the persistence issue of fund managers, assumes an investor is qualified to research, able to find and invests in all of the top quintile active managers, and behaviorally the investor would continue to stick with all of the managers over the twenty year time period.
Institutional investors seem to be catching on to the advantages of indexing over active management. Anecdotally, our firm has seen much of the institutional space considering the merits of indexing their US equity exposure entirely. Earlier in 2013, the $256 billion California Public Employees' Retirement System (CalPERS) hired a consultant that demonstrated at any given time, only about one-quarter of the fund's external active managers are outperforming their benchmarks. Further, the results of the winning managers may not be high enough to cancel out the underperformance by the losing managers. He also noted that winning active managers change over time that complicates the selection process.
Guidance Point's investment philosophy is built on independent consulting that acts always in a fiduciary fashion or in the interest of our clients before our own interests. Based upon the discussion of this paper, our recommendations for investment managers generally start with an index approach. We do caution that not all index funds are created equal, and not all index funds will perform similarly. Also, we do know that indexed strategies will not outperform active strategies in all time periods. However, we strongly feel that investors should not overlook the advantages presented by indexing strategies in this paper when considering their portfolio construction process.
Additional disclosure: Guidance Point Retirement Services' clients benefit from our consultants being truly independent and objective. In our recommendations to clients we use several index related products that may or may not be referenced in this article. Any position suggested is not intended to constitute legal, tax, securities, or investment advice or a recommended course of action in any given situation. Additionally, our partnership with Seeking Alpha is based upon reach of audience and not financial consideration. Any compensation paid from Seeking Alpha to Guidance Point for our publications will be donated to a local charity of Guidance Point's choosing.