The Return Gap
Russ Kinnel of Morningstar has frequently observed that the buy-and-hold or time-weighted return is typically much higher than the dollar-weighted return. In addition, Hsu, Myers, and Whitby (2014) have robustly documented this phenomenon. The writers attribute the return gap to investors' poor market timing decisions as they reallocate assets among funds on the basis of recent performance.
Investors are so spectacularly bad at market timing that they routinely wipe out all, or more than all, of the outperformance produced by value-oriented managers. An investor who spurned value strategies after the bludgeoning they received during the tech rally of the 1990s lost out on the value premium's 94% run between July 2000 and June 2002. An investor who exited value funds in early 2009, after the collapse of banking stocks killed value returns, missed the 27% surge from March 2009 to April 2010. These examples are admittedly cherry-picked, but they vividly describe the average investor's behavior. Trend chasing typically results in forgone profits or outright losses
What can we conclude from the observed return gap between the outperformance of the buy-and-hold value strategy and the underperformance of the average value investor?
Mean-Reverting Value Premium
First, this result tends to corroborate the finding, documented in Hsu (2014), that the value premium is mean-reverting. Apart from needlessly incurring transaction costs, the investor's trend-chasing allocation would not be harmful if the value premium were constant over time. But the mean-reverting value premium has had a whiplash effect on the average value investor, whose philosophical commitment to value investing is belied by trend-chasing allocations.
Value Strategy Capacity
Second, value investors have not earned a positive premium. This observation has very deep implications. If average investors have not extracted positive dollar alpha from value strategies, then it is specious to claim that investors on the other side of the value trade are being systematically exploited and will ultimately be eliminated. Indeed, given that the average value investor's dollar alpha is negative, at least some of their counterparties must be making a handsome profit! This reasoning challenges the prediction that the free lunch from value investing might already have been arbitraged away by the significant allocation to value funds. Quite the contrary, fund flow data show that trend-chasing value investors far outweigh buy-and-hold value investors. Thus it would appear that, on average, value investors are supplying a premium to other market participants rather than collecting one. The value strategy may have far more unused capacity than we suspected.
A Pyrrhic Victory
It is small consolation that growth investors' dollar-weighted returns are even worse. In fact, large or small, value or growth, investors' dollar-weighted returns are overwhelmingly lower than the fund managers' buy-and-hold or time-weighted returns. Table 2 shows the dollar-weighted return, the buy-and-hold return, and the gap between them for different types of funds.
For a larger view, please click on the image above.
Across all funds, investors earned an average dollar-weighted return of only 6.87%, 194 bps less than the 8.81% that managers achieved on a time-weighted basis. To be clear, if investors bought mutual funds and held them throughout the measurement period, they, too, would have earned 8.81%. The 1.94% shortfall is due to poor timing on the part of investors (not managers).
Value investors did better; they underperformed their respective funds "only" by 1.31%. Growth investors, on the other hand, were ravaged by their trading behavior, losing a whopping 3.16% on top of growth strategies' general underperformance relative to value (and, in fact, relative to the S&P 500). It's not a pretty picture. All of the differences between dollar-weighted and buy-and-hold returns in Table 2 are highly significant, both statistically and economically.