Investors have been migrating in droves to passive management, but is it rational?
Investors tend to pick actively managed mutual funds based on their perception of what was, what is, and what can be. For example, they often idealize successful long-term managers, believing that their success came through unerring short-term performance, despite the reality that slumps occur routinely (“How We Improve Our Odds of Picking Outperformers,” April/May 2011 issue). They shovel assets to category killers, seemingly unaware that peer groups are shape-shifting, making comparisons complex and, in some cases, perilous (“Pitfalls of Peer Groups,” August/September 2011). And they chase performance with abandon, chucking laggards for the next hot thing (“A Strategy That Loves Performance Chasers,” December 2010/ January 2011).
Perception also appears to be playing a key role in another realm: the migration from active to passive management. Investors have increasingly replaced actively managed stock funds with passive vehicles like exchange-traded funds, shrinking its market share over the past decade. Active management seems to have an image problem like never before.
But is it rational for investors to spurn active management in this fashion? Or is this more of the same binge and purge, with investors allowing emotion--in this case, frustration with disappointing active-investing results--to drive their decisions?
Perhaps the answer depends on what investors should reasonably expect from active managers. A basic investing precept holds that it’s a zero-sum game, with winners largely offsetting losers, thereby preventing a clear majority of active managers from succeeding over the long haul. But what about shorter time periods? What does history suggest our odds are of picking a winning active stock manager over a three-year time frame? And how does the recent performance of active equity managers compare with the norm?
In this piece, I explore those questions in detail and offer some possible explanations for shifts in investor attitudes toward active management. We also explore the implications of trends in active-management success on portfolio construction decisions.
To evaluate the success of active funds, we compiled rolling 36-month excess returns for every actively managed U.S. large-, mid-, and small-cap fund for the 15-year period ended Oct. 31, 2011. We defined “excess returns” as the difference between a fund’s returns and the returns of the S&P 500 (for large-cap funds), Russell Mid-Cap (mid-cap funds), and Russell 2000 (small-cap) indexes.
For each of the 144 rolling periods, we counted the number of active funds that had generated positive excess returns (that is, beaten the relevant index) and divided that figure by the total number of active funds. This quotient represented a success, or “beat,” rate. We compared these beat rates over time to derive a clearer sense of patterns in active management’s success. Exhibit 1 shows the beat rates of all U.S.-stock funds (including averages) over the 15-year period we studied.
In examining the trend in beat rates, a few traits jump out. First, there is a high degree of variability, with active managers having their way with the benchmark indexes at times, only to badly lag in other periods. For example, nearly all active large-cap managers lagged the S&P 500 in the late 1990s, when a narrow swath of large, growth-oriented firms propelled the index higher. Yet, their fortunes turned just a few years later when value stocks, which active large-cap managers had tilted toward, held up better than the tech-dominated benchmark.
We also can observe that, far from being a bust, active managers have had more-frequent success than average in recent years. True, active managers took a drubbing during the financial crisis, which punished even measured risk-taking, explaining why so few active funds bested their benchmarks. But many of these funds roared back during the recovery, large-cap managers in particular.
The rejoinder, of course, is that it’s the magnitude of outperformance that really counts--beating the benchmark by a trivial amount counts only as a moral victory, if that. But when we examine the trend in median rolling excess returns, there are few signs that active management has slumped to an unusual extent, at least not when compared with history.
Though relative performance has slipped in recent months, large-cap managers have generally kept pace with, if slightly surpassed, the S&P 500 in most rolling 36-month periods since the depths of the financial crisis. What’s more, the median excess return has been slightly higher than the historical norm in the majority of rolling periods since 2006, a period that’s seen redemptions accelerate.
The Crux of the Problem
One can quarrel with active management for a number of reasons, not least its cost, but active investing appears not to have been any more futile (or, if one prefers, less successful) in recent years than the historical average. So, what’s got investors so bent out of shape?
They haven’t always been so punitive. Take the year 2000, for instance, when investors pumped around $45 billion into actively managed large-cap funds, relatively few of which had topped the S&P in the preceding years. Yet, one obvious and crucial difference is the market’s trajectory--sharply higher then, flat to lower in recent years. Put another way, it’s a lot easier for investors to forgive a manager for gaining 19% when the market’s up 20% than to give that same fund a pass when it’s down 30% versus a 28% loss for the S&P. Loss aversion is powerful in explaining behavior.
Investor behavior likely also plays a role in
explaining the influence of other, more-nuanced
factors in investor attitudes toward active
management. For example, active management’s
quest to top the markets has long
evoked the possible. Those possibilities seem
all the brighter at times when active managers
can be seen smashing their benchmarks.
But that was a much more common sight
a decade or so ago than recently, as shown in
Exhibit 2, which plots the excess returns
of large-cap managers at various percentiles.
As the chart makes plain, the excess
returns at one and two standard deviations
were significantly larger in the past. As
excess returns have narrowed, it’s likely that
the possibility--that is, the payoff--of active management has dimmed in some
investors’ eyes, raising questions about
One other factor is the way we measure active management--beat rates and margins of outperformance can be exaggerated. Why? The most common method of measuring active-manager success is using funds’ most-recent category classification. The problem is that funds change categories from time to time, potentially making such comparisons potentially imprecise.
To illustrate, we compared beat rates calculated under our approach (which accounts for any category changes from one rolling period to the next) with one that considers the most-recent categorization only. As shown in Exhibit 3, the difference is meaningful: Under the “most-recent” method, 52% of active managers appear to beat the S&P, on average, in a given three-year period, versus the 46% beat rate we’d previously calculated. What’s more, the average excess return was 0.54% higher under the “most-recent” method.
These differences arise for two reasons. First, funds leave a better-performing area to join the new peer group, as in the case of mid-cap funds that became large-cap funds in recent years. In these cases, these managers appear to have “beaten” the S&P over many of the rolling periods examined when, in fact, that outperformance is less a matter of skill than style or risk-taking. Second, underperforming funds aren’t captured in the peer group, either because they switched to a different category or were mothballed in earlier periods.
Why does it matter? Disappointment with active management is partly a function of expectations, which the “most-recent” method raises to a potentially unattainable extent. This, in turn, can feed the same emotions that have spurred investors to leave active management for greener pastures.
The Grass Is Greener?
Today’s investors are clearly better attuned to the importance of costs and diversification. They’re placing a premium of importance on portfolio construction and transparency and the market has shifted away from the “closed” distribution model that formerly stacked the deck in favor of actively managed funds. In general, these developments are likely to further cement demand for passively managed strategies. Nevertheless, the recent flight from actively managed equity funds raises questions of whether investors, in their zeal for passive management, have made a fully measured assessment of active managers’ performance. As our research has shown, active management has been no less successful in recent years than the historic norm, whether measured by “beat rate” or excess return generated.
Putting aside questions of whether the pendulum has swung too far, our research finds that active-management beat rates tend to be fairly mean-reverting, with periods of greater success eventually begetting slumps, and vice versa. This has implications for those weighing the role of active and passive management in a portfolio, especially considering the binary, either/or way such decisions are often made. Indeed, our research suggests that trends in the relative success, or futility, of active management should inform such decisions.
For example, in assembling an active/passive portfolio, one would ordinarily consider an asset class’s efficiency when deciding between active and index funds. Typically, investors opt for index funds in areas that are less hospitable to active managers and actively managed funds in less-efficient realms. However, given evidence that active management success tends to be mean-reverting, in aggregate, it could also be useful to assess recent beat rates in various asset classes, adjusting one’s commitment to active managers accordingly (that is, bigger commitment to areas with low beat rates, lower to those with unusually high beat rates).
Of course, it remains difficult to consistently pick successful active managers, even from areas that have experienced unusually low beat rates. For those seeking to exploit mean reversion in active-manager success, the challenge thus becomes making it investable--giving investors the opportunity to “bet” on (or against) the success of a group of active managers, without courting the risk of selection errors (picking a manager who underperforms within a group that outperforms as a whole). While such products do not currently exist, these sorts of innovations could help to reduce the influence that perception and emotion play, leading to sounder decisions on the role and merit of active management and, thus, better outcomes.