John Keats once famously noted, "Nothing ever becomes real till it is experienced."
Modern finance theory usually assumes that people are fully rational, incorporating all available information instantaneously into their expectations. Yet researchers have discovered that our personal experiences disproportionately impact our expectations about the future and risk taking behavior. In other words, we often overweight our own personal experiences at the expense of broader and often more objective and accurate information.
Below, I examine the evidence behind, and implications of, this bad investing behavior, and what investors can do to try to overcome it.
Q: What is the evidence on how personal experiences affect our investment decisions?
A: Researchers have found that our experiences have a disproportionate impact on how much we save for retirement, our willingness to repurchase a given stock, our propensity to participate in initial public offerings and our portfolio diversification decisions.
Based on data from a large benefits recording firm, investors who have experienced high average returns or low volatility in their 401(k) accounts tend to save more. Meanwhile, as evidenced by trades during the 1990s at a large U.S. discount brokerage and a large retail brokerage, investors shun stocks that they previously sold for a loss and stocks that have risen in price after being sold. This is presumably because investors are trying to distance themselves from the associated negative feelings of regret and disappointment.
Our personal observations about the macroeconomic environment also impact our return and volatility expectations, risk taking and financing decisions. For instance, based on data from the Michigan Survey of Consumer Attitudes, individuals expect higher stock market returns and lower volatility in times of economic expansion than during recessions. In addition, corporate managers who experienced the Great Depression subsequently shied away from external financing and chose a more conservative capital structure with less leverage.
Finally, based on 1960 to 2007 data from the Survey of Consumer Finances, individuals who have experienced low stock market returns during their lifetimes tend to be more pessimistic about future returns, more risk averse, less likely to participate in the stock market and generally invest a lower fraction of their liquid wealth in stocks. The same is true for bond investors who have experienced low returns.
Q: What are the portfolio implications of this bad investing behavior?
A: By simply repeating investing behaviors that resulted in good outcomes for us in the past, and avoiding those that resulted in poor outcomes, we're potentially eliminating important information that could help future investment performance. This is the primary issue with relying on our personal observations and experiences to form return and risk expectations and building portfolios.
For example, an individual who has only experienced poor stock market returns may cut down his retirement contributions to the point that he ultimately falls short of his retirement goals. Today's younger generations, who over the past 15 years have seen two big stock market crashes, may be especially susceptible to underinvestment in risky assets if they base their investing decisions solely on their own personal investment experience.
Q: What steps can investors take to mitigate the potential adverse impact of this bias on their portfolio positioning and performance?
A: Key here is broadening one's horizons, starting with an awareness of our tendency toward the personal experiences bias. More specifically, individual investors and advisors would want to consider deliberately lengthening the time period on which they are basing their return and risk assessments to include different macroeconomic and market scenarios.
In addition, given the dangers of spot estimates, portfolios should be scenario and stress tested for different kinds of environments, and the data sources used to make assessments should incorporate a broad array of information, ranging from macroeconomic to company specific data.
I also advocate using a rules-based or systematic investment methodology, whether simple quantitative screens or full investment models built to predict fair values, to help construct a portfolio. This approach can potentially draw attention to, and help weed out, behavioral investment biases.