Instead, the authors found that investment losses, combined with loss aversion, contribute to an increase in risk aversion during bear market declines. On the other hand, gains during a bull market lead to the well-documented “house money” effect and a decrease in risk aversion.

The findings from this study— Do Market Returns Influence Risk Tolerance? Evidence from Panel Data by Rui Yao and Angela Curl—suggest that individuals invest greater amounts after periods when market returns are high and withdraw partially or even completely from the market after periods when returns have been poor.

Today, we’ll examine some additional support for Yao and Curl’s conclusions, as well as explore the relationship between loss aversion and investor overconfidence.

To begin with, Yao and Curl’s findings are consistent with those of Michael Guillemette and Michael Finke, authors of the June 2014 paper, Do Large Swings in Equity Values Change Risk Tolerance?

Their study investigated whether the measured risk tolerance of U.S. and Canadian individuals correlated with market movements. Using a dataset provided by FinaMetrica, which creates and distributes a risk tolerance questionnaire widely employed by financial planners, the authors examined average monthly risk tolerance scores (MRTS) during the period from January 2007 through May 2012, a period that spans the financial crisis. A total of 341,782 people were surveyed over the time period. Their objective was to test whether fluctuations in equity returns influence average risk tolerance scores over time. The following is a summary of their findings:

• There was a strong positive correlation (0.70) between the S&P 500 and the MRTS.

• Risk tolerance scores are consistently lower immediately following a market decline. The correlation between the S&P 500 and the MRTS climbed to 0.90 for the period from January 2007 through March 2009, when the market bottomed out. However, the correlation during the recovery was 0. A rising market is seen by some as a buying opportunity, while others remain more risk averse after recent losses.

• The correlation between the MRTS and consumer sentiment was 0.67.

• When consumer sentiment was most negative, investors were the most risk-averse.

• When consumer sentiment was the most positive, respondents were far more risk-tolerant.

Unfortunately for investors, their average monthly risk tolerance was also highly correlated with equity market valuations as measured by price-earnings (P/E) ratios. In fact, Guillemette and Finke found that risk tolerance increases when equity valuations are high (and expected returns are low), and that individuals are most risk-averse when equity valuations are low (and expected returns are high). This change in risk tolerance can lead to a buy high and sell low pattern of trading. As you would expect, this perverse behavior negatively impacts investor returns.

For example, Geoffrey Friesen and Travis Sapp, authors of the 2007 study Mutual Fund Flows and Investor Returns: An Empirical Examination of Fund Investor Timing Ability, found that individual investors lose, on average, 1.56% a year in dollar-weighted returns because they tend to pull money out of equity mutual funds following a significant market decline (when equity valuations are more favorable). Conversely, investors increase equity allocation following recent price increases (when valuations are less favorable).

As yet another example of recency’s impact, one that spanned the period of the financial crisis, the June/July 2011 issue of Morningstar Advisor looked at the behavior gap, the difference between the dollar-weighted returns earned by investors and the time-weighted returns earned by the mutual funds in which they invest, for the one-year and three-year periods ending December 2010. For domestic equity funds, the gap was 2% and 1.3% per year, respectively. For international equity funds, the gap was 0.6% and 0.8% per year, respectively.

Shan Lei and Rui Yao, authors of the 2015 study Factors Related to Making Investment Mistakes in a Down Market, contribute to our understanding of investor behavior. They explored the following hypotheses:

1. Loss aversion positively affects the likelihood of making investment mistakes.
2. Overconfidence positively affects the likelihood of making investment mistakes.

To test their hypotheses, they used data from the 2008 FPA-Ameriprise Financial Value of Financial Planning Research Study. The data was collected online by an independent market research firm between June 27, 2008, and July 18, 2008, in the midst of the financial crisis. The total sample size was 2,792 respondents. The survey asked participants for their reaction to the market changes during the past year. One question asked: “Since the market has changed over the past year, what actions, if any, have you taken?” One possible answer was: “Moving assets into more of a cash position.”

Lei and Yao considered it a mistake if investors moved assets into cash in a down market while having an adequate amount in an emergency fund. Because more loss-averse investors are more likely to react in a down market, those who chose these items as answers were considered to be more loss-averse investors. The following is a summary of the authors’ findings, all of which are consistent with findings already discussed:

• Consistent with the first hypothesis, respondents who were more loss-averse were also more likely to make investment mistakes.

• Consistent with the second hypothesis, respondents who expressed more confidence were 1.4 times as likely to make investment mistakes.

• After controlling for other variables in the model, women were found to be less likely than men to make the mistake of moving to cash during the bear market (with an odds ratio of 0.726). This is consistent with prior research that demonstrates men are more likely to be overconfident, and thus more likely to make mistakes.

• The percentage of respondents making mistakes was generally greater for higher investable asset groups. This is consistent with the idea that investors who are more confident will be less risk-averse and hold riskier assets.

Since misery loves company, it’s nice to know that U.S. investors are not alone in their bad behavior. Thanks to Andrew Clare and Nick Motson—authors of the study Do U.K. Retail Investors Buy at the Top and Sell at the Bottom? have evidence demonstrating that investors in the U.K are equally guilty of bad or irrational behavior.

Clare and Motson examined the impact of timing decisions of both retail and institutional U.K. investors. The authors’ study covered the almost 18-year period from January 1992 through November 2009. The following is a summary of their findings:

• Just like U.S. investors (both individual and institutional), U.K. retail investors (though not institutional investors) are performance chasers as fund flows correlate with prior 12-month returns to the equity market.

• When they examined the prior six-month returns, the correlation between market returns and fund flows increased. Unfortunately, so did the negative correlation between fund flows and future returns, and it was now statistically significant at the 99% level of confidence.

• For retail investors, these correlations were also found to be significant between the 15- and 24-month horizons.

• The performance gap (the difference between a buy-and-hold strategy of a fund and investor returns in that fund) for retail investors was -1.17% per year. For institutional investors there was also a performance gap, but it was smaller at -0.20%.

The bottom line is that, over the 18-year period, the performance-chasing behavior of U.K. retail investors cost them a total return loss of 20%.

Warren Buffett has warned investors that their greatest enemy is looking at them in the mirror. Three of the most common and, unfortunately, most expensive mistakes that individual investors make are: overconfidence in their ability to withstand the stress of bear markets (which may lead to panicked selling), recency, and thinking that they are playing with the “house’s” money. These three are among the 77 mistakes covered in my book, “Investment Mistakes Even Smart Investors Make and How to Avoid Them.”

But individual investors aren’t the only ones impacted by these problems. In his book, “Beyond Greed and Fear: Understanding Behavioral Finance and the Psychology of Investing,” Hersh Shefrin reported that the risk-tolerance levels of both institutional investors and financial advisors were positively correlated with stock market returns.

Having a sound understanding of risk tolerance is not only important for individual investors, but also for their financial advisors. The research shows that while advisors may treat investor risk tolerance as a stable characteristic, it’s clearly important to periodically revisit their clients’ risk tolerance, as risk tolerance changes not only as investors age but with movement in the markets as well. If an investor’s risk tolerance does change in response to market returns, it’s likely that either the investor (or the advisor) overestimated their ability to understand risk and properly assess their individual risk tolerance. And thus a change in the overall financial plan may be required.