Market Efficiency vs. Behavioral Finance: Which Strategy Delivers Better Returns?
I’m the most important person in behavioral finance, because most of the behavioral finance is just the criticism of efficient markets. So, without me what do they got?
Eugene Fama
Gene has it all wrong. If it were not for Behavioral Finance, he and French would have had nothing to do for the past 25 years. He owes me everything.
Richard Thaler
After reading these quotes from Fama and Thaler, you may conclude that they are bitter rivals. But this is far from the case. Fama and Thaler are business school professors at the University of Chicago and well-documented golf buddies. But despite sharing the occasional 18 holes, there is very real underlying tension between the two. Fama is captain of Team Efficient Markets and Thaler is captain of Team Behavioral Finance. Each represents conflicting academic market philosophies that have been warring for years. It’s the academic equivalent of Lakers vs. Celtics.
Team Efficient Markets believes that market prices reflect all available information and are therefore efficient. Its strongest proponents believe that risk-adjusted performance over long-time horizons isn’t possible. Over time, the philosophy expanded to include risk factors. Investors can be compensated by tilting their portfolios toward risk factors to achieve higher returns. This team believes that because these factor tilts represent increased risk, risk-adjusted performance over long-time periods isn’t possible.
Market efficiency proponents argue that if empirical evidence shows long-term risk-adjusted performance was achieved, investors didn’t achieve it due to skill but by tilting their portfolios toward a previously unidentified risk factor, or by dumb luck. “Buffett’s Alpha” deconstructed Warren Buffet’s phenomenal track record at Berkshire Hathaway into different explanatory factors. The paper won the Graham and Dodd Award for best paper in 2018. The award recognizes excellence in research and financial writing in the Financial Analysts Journal. Although the authors conceded that Buffett’s track record was not due to luck, it’s hard to read the paper without coming away with the feeling that its purpose was to knock Buffett’s performance down a peg.
Team Behavioral Finance, on the other hand, believes market prices reflect all available information most of the time, but that market participants are also influenced by behavioral biases. This behavior leads to market inefficiencies that can be exploited to achieve superior risk-adjusted performance, even over long-time horizons. Regarding factor investing, the behavioral camp believes that ‘risk factors’ represent price/value gaps due to behavioral biases rather than an increase in risk taking. As it pertains to Buffett, this camp is more likely to believe that his track record is due to his even-headed decision-making skill and access to unique information sources.
Unfortunately, many issues arise when debating market anomalies. The main two issues stem from hypothesis testing difficulties (e.g., how would you test for behavioral biases?) and the subjective interpretation required when a market anomaly is discovered (e.g., increased risk, behavioral inefficiency, or spurious correlation).
But fortunately, Fama and Thaler’s respective philosophies heavily influence two major asset management firms, Dimensional Fund Advisors (DFA) and Fuller & Thaler Asset Management (FullerThaler).
DFA’s founder David Booth served as a research assistant under Fama while attending the University of Chicago in 1969. The firm’s investment underpinnings heavily rely on Fama’s academic research, leading it to tilt their portfolios toward small, cheap companies with higher-than-average profitability. Fama also serves as a director and consultant at DFA.
As the name implies, Thaler co-founded FullerThaler with Russell Fuller. The firm seeks to exploit behavioral biases to outperform markets. Like DFA, the firm also tilts its portfolios toward value and size factors. Unlike DFA, the firm seeks to exploit the loss-aversion bias, believing that investors overreact to bad news and losses and underreact to good news. As the name implies, Thaler co-founded FullerThaler with Russell Fuller. The firm seeks to exploit behavioral biases to outperform markets. Like DFA, the firm also tilts its portfolios toward value and size factors. Unlike DFA, the firm seeks to exploit behavioral biases, believing that investors overreact to bad news and losses and underreact to good news.
Both firms have an investment fund with a long track record and the same benchmark, The Russell 2000 Value Index. Figure 1 pits the competing philosophies against each other and the funds’ benchmark.
Figure 1. DFA’s U.S. Small Cap Value Portfolio (DFSVX), FullerThaler’s Undiscovered Managers Behavioral Value Fund (UBVLX), and The Russell 2000 Value Index.
Team Behavioral Finance outperformed Team Efficient Markets by an annualized 0.91% between December 1998 and July 25, 2024. But many readers may disagree that this proves Team Behavioral Finance’s victory, because the results don’t account for risk taken. Fair enough. To test this, I applied Jensen’s Alpha (Alpha) and only use The Russell 2000 Value Index as a benchmark. For the risk-free rate, I de-annualized the three-month treasury rate.
Figure 2.
After accounting for risk, Team Behavior still comes out on top. This is nearly confirmed unanimously throughout all risk-adjusted return metrics as shown below, apart from the Information Ratio.
Despite the results implying that investors can exploit behavioral biases, even over long-time horizons, strong market efficiency believers may be hesitant to change their minds. If so, I encourage these individuals to check their own behavioral biases to ensure they exhibit the same rational traits that the market efficiency hypothesis assumes are true.
If you liked this post, don’t forget to subscribe to the Enterprising Investor.
All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.
Image credit: ©Getty Images / Ascent / PKS Media Inc.
Professional Learning for CFA Institute Members
CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker.
As the displayed chart of the “Team Behavior” looks very similar to the one of the “Team Efficiency” and the index, it can only contain marginal alpha. However, the time span of 25 years is much too short for presumed alpha to be statistically significant. Thus, this comparison is hardly relevant for the porpose.
Why didn’t you compare pure equity beta of MSCI World or S&P500 with definitely statistically significant pure alpha, e.g., of the SG Trend Index? Its risk adjusted return is about the same or slightly better than that of equity beta, but uncorrelated. During financial crises it is even mostly negatively correlated.
Thus, their combination in one portfolio provides significantly higher risk adjusted returns than pure beta or alpha alone. The Harvard Finance Professor John Lintner, who also contributed to the CAPM along with William Sharpe, discovered this already more than 40 years ago and published it in his legendary “Lintner paper” in 1983:
“The combined portfolios of stocks (or stocks and bonds) after including judicious investments in appropriately selected sub-portfolios of investments in managed futures accounts (or funds) show substantially less risk at every possible level of expected return than portfolios of stock (or stocks and bonds) alone.”
This holds true until today as proven by the SG Trend Index. Thus, why not take such proven and dependable uncorrelated pure alpha for your analysis instead of questionable alpha, highly correlated with equity beta?
Hey Norbert – thanks for your comment!
My personal philosophy as to what ultimately matters is maximizing geometric returns, as many of the assumptions underpinning many risk-adjusted measures are unrealistic for most portfolio managers (e.g., you can borrow at the risk-free rate to lever a superior Sharpe portfolio, etc.). As the saying goes, “You can’t eat Sharpe.”
Your critique is fair as it pertains to statistical significant alpha when directly measuring Team Behavior against Team Efficiency. If you regress one return stream against the other, there is no statistical significance in the intercept. But, I’ll push back and caution against the need for such a long-time horizon for measurement. I’ll note that for Figure 2 I used daily returns in the analysis and annualized the results, so there were 6,435 data points for the measurement and the five year rolling metric displayed contained a rolling 1,260 data points for 5,175 calculations.
To the dangers of requiring very long time horizons, it’s my opinion that markets are ever evolving complex adaptive systems and by requiring such a long time horizon for data analysis, you’re implying that future market dynamics will be the same as they’ve been in the past (i.e., markets are stationary). This can introduce significant bias in models and model conclusions as many of the data points are “stale.”
Many market efficiency proponents and factor investors have learned this hard way, since many of the original academic findings from these early papers have performed poorly. For instance, the Russell 1000 Growth TR Index has recently overtaken the Russell 2000 Value TR Index in performance (since 12/31/1978). If you’re a risk-premium market efficiency believer, you’re in trouble. If markets are efficient and higher returns should be compensated with higher risk taking, how does this occur?
Well, you’ll notice that they’re bolting on more factors to their model portfolios and have gotten away from the concept of risk-premiums altogether (e.g., How are more profitable companies riskier?). I’d also argue that by publishing their findings, many market participants believing these were inefficiencies and not risk-premiums, corrected their mistake. Markets adapted.
At the end of the day, it all comes down to your investment philosophy and how you believe markets work. I personally try to blend the underlying theory and logic from each camp. Believing that markets are efficient most of the time, but investors can be irrational some of the time causing inefficiencies, all while keeping in mind that “past performance isn’t indicative of future results” since market dynamics change and evolve.
I’d be more than happy to review your findings if you choose to do the research you’re suggesting!
All the best – Scott
here is little doubt the stock market is more behavioral than informationally efficient. Fuller Thaler pursues a combined behavioral/fundamental stock methodology, so not a behavioral pure play. It is not clear which part of this combined strategy delivers FT’s outperformance. My guess is that both contribute.
Enough of the “risk adjusted return” game. CAPM beta has no evidence supporting it as a measure of risk, while even Fama and French admit they do not know whether so called “factors” represent risk or opportunity.
So, to penalize investment performance for what may be an opportunity and not a risk makes no sense. You can see the perverse logic here: the strategy outperformed so it must be due to risk!
Until we have a legitimate, agreed upon measure of risk then “risk adjusted returns” make no sense.
At AthenaInvest we have been managing two pure play behavioral strategies, one a value strategy for 22 years and another a market exposure strategy for 14 years. Each has dramatically outperformed since inception.
We are strong believers in Behavioral Portfolio Managment.
Hey Tom – Thanks for the comment!
I mostly agree with you but would probably disagree to the degree at which you believe in market inefficiency. I believe it’s more informationally efficient most of the time but behavioral inefficiencies can be the primary driver over shorter time horizons. For instance, the meme stock and unprofitable tech mania in 2021.
To you’re point about a legitimate risk-adjusted return measure, I believe geometric performance is the most relevant measure. It’s the return you actually “eat” and accounts for volatility, skewness, and kurtosis when looking at historical performance.
Congrats with the success at your firm!