Style-based feedback trading—fund managers’ attempt to outperform by adjusting their exposure to styles across the growth/value and market-capitalization continua—is highlighted as a common practice among US domestic equity mutual funds. The authors describe various methods of implementing such trading across fund types and estimate related performance effects.
How Is This Research Useful to Practitioners?
As an objective checkup of the baseline behavior of active professional investors, the authors find strong evidence of style-based feedback trading. Fully 77% of the authors’ sample of US domestic equity mutual fund managers are shown to engage in style switching, the opportunistic shifting of asset exposure along the dimensions of growth versus value and small-cap versus large-cap categories. “Twin-style” switching—the concurrent reconfiguration of capital across growth/value and small-cap/large-cap funds—is evident. Distinct trading niches are identified, with growth funds driven by positive feedback (momentum) and value funds taking the negative-feedback (contrarian) track.
The authors also examine the operational effects of this switching behavior and its consequences for performance. Their findings are helpful for a range of actors—from external wealth managers assessing potential allocations to funds at different points in their life cycles to the fund sponsors themselves examining possible product-line extensions or judging the effectiveness of their portfolio managers.
Operationally, the funds that most aggressively pursue style switching tend to be younger and have higher total expense ratios. Although the extent of switching can vary significantly over time, it is definitely persistent at the fund level. Finally, in terms of risk-adjusted performance, the type of feedback trading is directly related to performance, with positive feedback resulting in positive alpha and the contrarian pattern negatively affecting returns. Twin-style switching, however, shows no performance benefits.
How Did the Authors Conduct This Research?
The authors claim that their study’s important contributions are creating a new model to directly test style-based feedback trading and the related detailing of various forms of style-oriented behavior. By empirically extending the concepts of Barberis and Shleifer (Journal of Financial Economics 2003), the authors combine a long postulated discrete choice model with an adaptive rational equilibrium framework that compares individual fund returns against their time-varying exposure caused by switches across four benchmark portfolios (growth and value, segregated by small- and large-cap size). The authors cover a lengthy trading period (December 1961–September 2010), using the CRSP mutual fund database to form a sample of 2,044 unique US domestic equity funds. These examined funds are classified under the Lipper fund classification codes.
The authors acknowledge a potential modeling misestimation issue for mutual fund research—namely, that of funds misclassifying themselves. This issue may be further complicated by the lengthy study period, over which the definitions of what could be considered growth versus value and small cap versus large cap may be unstable. Also, there are 12 Lipper classifications. The authors try to simplify by conducting special consistency checks and focusing on the individual fund portfolios of interest (small/large growth; small/large value) rather than amalgamated typical style factors (small versus big; high versus low). They also test their model against a randomly selected fund from a group known to engage in significant switching; the resulting goodness of fit appears satisfactory.
Abstractor’s Viewpoint
One of the oldest propositions in investment research on the workings of markets, shown by experience to have ongoing validity, is that active management is a loser’s game. Even investors willing to assume potentially undercompensated risk by forgoing the baseline opportunity to index passively can be exposed to the possibility of not getting what they paid for in the form of “style drift.” Such volatility can make rebalancing an even more essential safeguard for attaining objectives, made more challenging by intentionally moving targets.
The authors offer a fairly persuasive mix of “good news and bad news” regarding the behavior of asset managers. Their proclivity to practice style-based trading can be reassuring with respect to their motivations to outperform. But there is a downside to doubling down and adding greater risk in pursuit of returns. For example, fund managers who apply feedback trading rules “correctly” (momentum trading in the short run and contrarian trading in the long run) are shown to be rewarded initially with extra gains, whereas “incorrect” executions (using the opposite trading techniques) are punished in terms of performance. Unfortunately, there is no clear signal for when the music stops.
Good research can raise additional questions. The following are a few the authors leave unasked: Is switching a matter of pushing the accelerator (exaggerating one’s established style) or swerving across lanes (shape shifting across capitalizations or growth/value for style transformation)? Are changes attributable to new portfolio managers or background market conditions? Does feedback have a sweet spot of an appropriate “lookback” period, or is the interval changing unpredictably to frustrate course corrections?