Many institutional investors rely on market indicators such as beta to manage the systematic risk in their portfolios. Beta shows to what extent a portfolio moves in line with the market. But how well do investors really understand this relationship? New experimental research suggests: not as well as we think.
Even professional investors systematically misjudge their market exposure, often in their own favor.
In a series of large-scale online experiments, researchers Christophe Merkle and Michael Ungeheuer asked investors about their expectations for portfolio returns under different market scenarios. Some participants were allowed to construct their own portfolio from Dow Jones Index stocks; others were assigned a random portfolio. The question was simple: what do you expect your portfolio to do if the market rises or falls by, say, 10 percent?
The results are striking and important for anyone investing based on risk factors such as beta. On average, investors estimated their beta at just 0.7, significantly lower than the expected value of 1. For negative market scenarios (downside beta), the average estimate was even lower, around 0.6. In other words, investors believe their portfolios will fall less than the market during downturns. But in positive market scenarios (upside beta), they estimated a slightly higher beta, suggesting they do expect to fully participate in rising markets.
Most notable: this pattern of upside participation and downside protection was stronger among those who built their own portfolio. Clearly, we place more faith in our own stock-picking than in random diversification, even when that belief does not match reality.
In a follow-up experiment, the researchers repeated the study with more than 800 financial professionals. Even this relatively well-informed group significantly underestimated downside beta. Active choice led to overconfidence regardless of knowledge level. Professionals were somewhat better at estimating their overall beta, but were just as prone to error when it came to the asymmetry between market gains and losses.
Why this matters
For institutional investors, this is not a trivial point. If the average market participant underestimates systematic risk, that has direct implications for how risk is priced in the market. It also undermines the effectiveness of quantitative models that assume rational expectations for beta and market exposure.
The research suggests that investors not only overestimate their portfolios’ return potential, but also underestimate their risk. That is a recipe for overconfidence, think overly concentrated positions, under-diversification, or blindly sticking with favored stocks.
To avoid these pitfalls, it is essential to challenge your own assumptions. Even professional investors are vulnerable to the illusion of control. Let model outputs, not gut feelings, drive decisions. Incorporate stress tests that explicitly account for asymmetric scenarios. If your team consistently expects milder losses in downturns than gains in upturns, that may point to a persistent underestimation of risk. Awareness and training can help, for example through conditional thinking exercises, explicitly asking what would happen to the portfolio if the market fell by 15 percent, to break ingrained patterns. Regular external reviews by risk managers or data teams are also invaluable. This is especially true in active management, where they help expose blind spots and curb overconfidence.
Conclusion
Overconfidence is stubborn. Even with access to data, experience, and models, investors tend to overrate their own decisions. This research shows the problem is not a lack of knowledge, but a human tendency to view our own choices more favorably than they deserve. For institutional investors, this is a wake-up call: true risk management requires not just robust models, but also self-reflection.
Gertjan Verdickt is Assistant Professor of Finance at the University of Auckland and a columnist for Investment Officer.