Most people in the field of finance assume that people act rationally. But research and life experience suggest otherwise, at least some of the time. What are some recent research findings on how our hearts cloud our heads? How can we minimize the irrational parts of our decision making in finance?
By Mark D. Harris
Introduction
The Efficient Markets Hypothesis (EMH) suggests that all the available information about a publicly traded company that is pertinent to investing in that company is contained in the stock price at any point in time (Vasileiou, 2020). Insofar as this is true, investors are rational actors who make investment decisions solely on rational grounds. However, company stock prices sometimes are higher or lower than one would expect based on purely objective valuations. This fact suggests that something besides rationality is present in company stock prices.
To explain market behavior beyond the purely rational, researchers turn to behavioral finance. Growing out of Adam Smith’s Theory of Moral Sentiments, one of behavioral finances’ primary observations is that “investors (and people in general) make decisions on imprecise impressions and beliefs rather than rational analysis.” Further, “the way a question or problem is framed to an investor will influence the decision he/she ultimately makes.” The article concludes, “These two observations largely explain market inefficiencies; that is, behavior finance holds that markets are sometimes inefficient because people are not mathematical equations” (Behavioral finance, 2019).
Investment irrationality has been present in all areas of life for millennia. In the early 17th century, shortly after investing in futures became possible, investors in Europe began dumping money into the latest craze… Dutch tulips. A bubble developed. Fortunes were made and fortunes were lost. The South Sea Bubble (1720), the Mississippi company (1720), the Tech Crisis (2000), and the Subprime Mortgage Recession (2008) were later bubbles formed by “irrational exuberance,” to use Alan Greenspan’s famous phrase (Federal Reserve, 2018).
Using the tools of modern psychology, medicine, and finance, researchers in behavioral finance have tried to identify what causes irrational actions in finance, and how to address them. Human cognitive biases play a role, such as the following from the Corporate Finance Institute (Cognitive Bias – Examples, List of Top 10 Types of Biases, 2019).
- Overconfidence: This includes overevaluation of one’s own skills, and a belief that something will happen because you want it to (desirability effect).
- Self-serving bias: Investors attribute positive events to personal skill and negative events to bad luck.
- Herd mentality: This includes self-deception, heuristic simplification (using mental shortcuts that oversimplify complex issues), emotion, and social bias.
- Loss aversion: People would rather avoid a loss of $1,000 than enjoy a gain of $2,000.
- Framing cognitive bias: Investors make decisions based on how an issue is framed to them rather than on pertinent facts.
- Narrative fallacy: People like stories and are more likely to be moved to action, such as investing, by stories.
- Anchoring bias: Humans use preexisting data as a reference point for subsequent data.
- Confirmation bias: Mankind often seeks out information that confirms existing biases and rejects information that does not.
- Hindsight bias: Humans who predict a correct outcome think that they “knew it all along.”
- Representativeness Heuristic: The belief that if two objects are similar, they must be correlated with each other.
Other human biases also affect investor’s behavior and help to account for the irrationality in stock prices and the stock markets.
Current Research
Having recounted many of the human frailties that impact investment decisions above, it is well to evaluate academic research about behavioral finance. COVID-19 has been one of the biggest events in the past twenty years and has had a powerful impact on investing (Vasileiou, 2020). Vasileiou investigated US stock market behavior from 2 Jan 20 to 30 Oct 20. The market maintained slow but steady growth and the risk estimate remained small in January and February 2020. The market may have underestimated the risk of COVID. By late March 20, returns plummeted and risks as calculated by standard deviation (SD) soared. Vasileiou suggested that the market seemed to finally have reacted to the disease by late March 2020, commensurate with the mandatory US nationwide shutdowns. Market performance improved from April to October but variance remained high, reflecting risk. Data from the Coronavirus Fear Index demonstrated that the greatest fear correlated with the worst S&P 500 performance during the pandemic (Vasileiou, 2020). He concluded that once people realized the danger, fear became a major player in market performance.
Markets have long recognized that information, whether true or false, has an important effect on the behavior of investors. Bryan Fong wrote that “intentional disinformation, or ‘fake news’, is defined as “false stories that appear to be news, spread on the internet or using other media, usually created to influence political views or as a joke” (Fong, 2021). Under the efficient market hypothesis (EMH), fake news should have no impact because it is fake and rational investors should discount it. In reality, not only does fake news impact markets, but because its tone and content are usually more extreme than real news, fake news “diffused significantly farther, faster, deeper, and more broadly than the truth in all categories of information” (Fong, 2021). Fake news renders consumers more skeptical of all news…even the truth. Fake news is usually debunked after 3-4 days, but the fake news leads to initial overreaction and debunking to underreaction, so the overall effect persists longer.
Along the same theme of information in market behavior, Statman surveyed financial advertising over nearly a century to identify how companies convinced investors to invest. He took issue with the idea that financial advertising is misleading or manipulative by design. Rather, the author argued that people seek three types of benefits from their investments. The first is utilitarian, the desire to make more money. The second is expressive, the desire to communicate values, tastes, and social status to themselves and others. The third is emotional, the desire to make themselves feel good.
Common financial thought stresses the utilitarian benefits of investing and eschews or even derides the other benefits, but expressive and emotional benefits are often more important to people. In one example, “prospective workers submitted 44% lower wage bids for the same job after learning about the employer’s high social responsibility” (Statman, 2017). Of seven wants that people hope to fulfill by investing, only two (financial security and paying less taxes) are primarily financial. The other five (nurturing children and families, playing games and winning, staying true to our beliefs, attaining high social status, and promoting fairness) are not.
Readers may be surprised to know which information is considered valuable to investment decisions and which is not. Coibion et al., (2018) found that business managers did not consider broad conditions such as inflation to be of significance in making such decisions. This is problematic because central authorities such as the US Federal Reserve try to inform companies and investors of general economic information in the hopes that the firms will respond more intelligently to market conditions. If businesses are not paying attention, the Fed’s efforts are for naught. However, firms in highly competitive markets and those making important business decisions in the near future did prioritize broader economic knowledge (Coibion, 2018).
Information gained personally, within the investor’s own experience, heavily influences investor behavior. Using the Survey of Consumer Expectations database, Kuchler & Zafar, (2019) discovered that respondents expected US national housing price volatility to be similar to housing price volatility in their area. So, if house prices fell by 5% in Houston, people in Houston expected that house prices would fall by a similar degree in Chicago and throughout the county. Likewise, a respondent who had recently lost his job had a more pessimistic view of the US job market as a whole (Kuchler & Zafar, 2019). In both areas, local uncertainty bred national uncertainty. Such expectations are dangerous, because personal experience is a common but often unreliable guide to future action.
Investors consider second order information important. First order beliefs are what a person believes about something. Egan et al., (2014) stated that second order information is beliefs about beliefs, in this case, “beliefs about what stock market returns other investors expect.” If Investor A believes that Company A will do well, or poorly, that is a first order belief. If Investor A believes that Investor B believes that Company A will do well, or poorly, that is a second order belief. If all investors were perfectly informed, first and second order beliefs would be the same. Unlike weather, which does its own thing regardless of what humans do, individual and corporate beliefs, as manifested by words and actions can dramatically affect the stock market. For example, when an investor believes that others are optimistic about a stock or about the market in general, he will invest more. Investors suffer from a “false consensus effect,” in which they assume that other people believe like them, and a “bias blind spot,” in which they believe that those who disagree with them are biased, but they are not. Notably, “second-order beliefs are a positive and significant predictor of future stock market returns while first-order beliefs are not” (Egan et al., 2014).
The articles noted above emphasized the importance of information in investment decisions, but Kariofyllas et al., (2017) expanded the theme noted in the introduction – that cognitive biases impair our ability to sort out the best information, interpret it properly, and act on it. “Representativeness is the tendency of investors to classify firms into discrete groups based on similar characteristics” (Kariofyllas et al., 2017). A company may be in terrible shape today, but if it was successful in the past, especially the recent past, investors may favor it. The converse is also true. Similarly, “conservatism is an individual’s tendency to update their beliefs slowly, based on a Bayesian rule, in which they overweight their prior beliefs and underweight the sample evidence” (Kariofyllas et al., 2017). Compounding both biases is the effect of confidence, which overestimates the veracity of personal perceptions and information and underestimates broad, publicly available information.
Kliger et al., (2014) summarized important studies in behavioral finance. Medical factors such as activity of the monoamine oxidase gene seems to be linked to credit card usage. Consumers tended to save less and consume more the more they are aware of others’ consumption. Many people maintained high-cost consumer debt and low yield liquid assets simultaneously, which is irrational from a rate of return standpoint. Lack of self-control, not financial illiteracy, seemed to underlie this behavior.
Technical analysis is a practice in which an investor uses statistical patterns in market data, such as price movement and volume of investments, to predict the future performance of any security (stocks, futures, commodities, currency). It can be contrasted with fundamentals analysis, in which investors evaluate a company’s financials, as well as other key indicators (physical systems, leadership, workforce, products and services, etc.). Technical analysis assumes that all pertinent information from fundamentals analysis is already revealed in the price of the security. Kliger et al., (2014) noted that investors that use technical analysis face high opportunity costs: the time it takes to follow markets. They also trade more frequently, and lose 50 basis points of raw returns per month.
Upswings and downswings both influence investors to increase the frequency of monitoring their portfolio. Gains and losses also increase investors’ appetite for risk. Investors overvalue prepaid money, which is consistent with the well documented human tendency to throw good money after bad, hence refusing to acknowledge sunk costs. Finally, Kliger et al., (2014) revealed that simple diversification as a panacea to risk management is practiced by both novice and experienced investors.
Two studies revealed specifics on some of the factors involved in investing in a particular place and provide useful local information. Yurttadur & Ozcelik, (2019) sent over 1,000 surveys to individual investors in Istanbul, Turkey. They discovered that traditional financial theories such as the Efficient Markets Hypothesis (EMH) cannot properly account for investor behavior. A majority of investors preferred gold, domestic currency, and foreign currency over other investments. Investors in Istanbul found most of their investment information on the internet, television, and the newspaper. Middle age, low education, and low income were more often associated with overconfidence, and middle age was also associated with more fear of investment regret. The authors noted extreme optimism as a characteristic of large families with low income. Yurttadur & Ozcelik, (2019) focus on financial training as a useful tool to improve investing.
Kiymaz et al., (2016) identified four biases, similar to what was described above, as significant factors in Turkish investors. The first bias, overconfidence, increases trading volume, liquidity, and volatility, and decreases diversification. Nonetheless, some have higher than expected returns. Investors may trade for entertainment as much as they do for money. The second bias, the home bias, occurs when investors only or preferentially invest in their home country. Geographic bias is the third bias. It is found when investors only or preferentially invest in companies whose headquarters is close to their place of residence. The final bias is emotional bias, which can occur with “worthy causes” and heavy media influence. Technical analysis came up earlier and seemed to produce inferior results to fundamentals analysis. However, Kiymaz et al., (2016) identified concerns about fundamentals analysis also. He cited a study in which fundamentals analysis was associated with high risk, high trading volumes, and overconfidence.
The weakness of EMH in predicting investor behavior is not limited to any one group. De La O & Myers, (2021) stated that “Using subjective expectations based on survey data, we find that changes in subjective cash flow growth expectations account for the vast majority of movements in both the price-dividend ratio and the price-earnings ratio for the S&P 500.” Restated, subjectivity, more than objectivity, moves markets everywhere. Human nature is the same the world over.
Further Questions
Several research questions were implicit or explicit in the studies reviewed. First, what factors motivate investor behavior? How much does each factor motivate each individual? Do such motivational factors arise from advertiser manipulation or from bona fide felt needs in the investors? If advertising is to blame for manipulating investors, how do regulators respond? If human nature is to blame, what can be done about it? Overconfidence, the home bias, and the geographic bias are strongly rooted in everyone, but how can the danger of unnecessary risk be mitigated?
Second, how can governments get economic actors, both business leaders and individual investors, to act rationally based on broad economic data? Government entities need to signal their actions ahead of time to allow the market to act. For example, the US Federal Reserve Bank typically announces interest rate changes, quantitative easing, and other interventions well before the Fed implements them. The purpose of such signaling is to get major and minor economic players in the US and abroad to change their actions to maximize value in the changing conditions. Along the same lines, how can we minimize the human-generated damage caused by fake news? Governmental organizations value stability, and fake news brings instability into the market. Fake news can also harm companies, such as when Toyota, Taco Bell, McDonalds, Wendy’s, and Procter and Gamble faced allegations which were later proved to be false.
Third, how can we minimize the damage of natural events such as the COVID-19 pandemic? Such damage is present not only in worldwide events, but in regional events, such as earthquakes.
Fourth, what role do genetics and medicine, including gene expression and hormonal balances, play in investor decisions? What are the links between investing behavior and clinical conditions, whether mental health conditions such as depression or other medical conditions such as heart disease?
Conclusion
Humans are rational and irrational. Lies from insiders and wishful thinking from investors allowed Arif Nakvi of Abraaj and Elizabeth Holmes of Theranos to bilk investors such as Bill Gates and Rupert Murdoch out of billions of dollars. On a smaller scale, each of us suffers from biases which cloud our thinking about investing. Simultaneously, most research has been done assuming that investors act to maximize financial return. Investors with other priorities, such as supporting causes important to them, and feeling good about themselves, should and will behave differently. Educating the public at all levels, implementing the right regulatory framework, and continuing research will improve investing, and consequently economic conditions, for everyone.
References
Behavioral finance. (2019). TheFreeDictionary.com. https://financial-dictionary.thefreedictionary.com/Behavioral+finance
Coibion, O., Gorodnichenko, Y., & Kumar, S. (2018). How Do Firms Form Their Expectations? New Survey Evidence. American Economic Review, 108(9), 2671–2713. https://doi.org/10.1257/aer.20151299
De La O, R., & Myers, S. (2021). Subjective Cash Flow and Discount Rate Expectations. The Journal of Finance, 76(3), 1339–1387. https://doi.org/10.1111/jofi.13016
Egan, D., Merkle, C., & Weber, M. (2014). Second-order beliefs and the individual investor. Journal of Economic Behavior & Organization, 107, 652–666. https://doi.org/10.1016/j.jebo.2014.04.001
Federal Reserve. (2018). FRB: Speech, Greenspan — Central banking in a democratic society — December 5, 1996. Federalreserve.gov. https://www.federalreserve.gov/boarddocs/speeches/1996/19961205.htm
Fong, B. (2021). Analysing the behavioural finance impact of “fake news” phenomena on financial markets: a representative agent model and empirical validation. Financial Innovation, 7(1). https://doi.org/10.1186/s40854-021-00271-z
Kariofyllas, S., Philippas, D., & Siriopoulos, C. (2017). Cognitive biases in investors’ behaviour under stress: Evidence from the London Stock Exchange. International Review of Financial Analysis, 54, 54–62. https://doi.org/10.1016/j.irfa.2017.09.003
Kiymaz, H., Öztürkkal, B., & Akkemik, K. A. (2016). Behavioral biases of finance professionals: Turkish evidence. Journal of Behavioral and Experimental Finance, 12, 101–111. https://doi.org/10.1016/j.jbef.2016.10.001
Kliger, D., van den Assem, M. J., & Zwinkels, R. C. J. (2014). Empirical behavioral finance. Journal of Economic Behavior & Organization, 107, 421–427. https://doi.org/10.1016/j.jebo.2014.10.012
Kuchler, T., & Zafar, B. (2019). Personal Experiences and Expectations about Aggregate Outcomes. The Journal of Finance, 74(5), 2491–2542. https://doi.org/10.1111/jofi.12819
Statman, M. (2017). Financial Advertising in the Second Generation of Behavioral Finance. Journal of Behavioral Finance, 18(4), 470–477. https://doi.org/10.1080/15427560.2017.1365236
Vasileiou, E. (2020). Behavioral finance and market efficiency in the time of the COVID-19 pandemic: does fear drive the market? International Review of Applied Economics, 1–18. https://doi.org/10.1080/02692171.2020.1864301
Yurttadur, M., & Ozcelik, H. (2019). Evaluation of the Financial Investment Preferences of Individual Investors from Behavioral Finance: The Case of Istanbul. Procedia Computer Science, 158, 761–765. https://doi.org/10.1016/j.procs.2019.09.112
