CHAPTER 2
Behavioral Biases

Behavioral finance research is driven by observations suggesting that individuals' decisions can be irrational and different from what previous theories assume. In this chapter, we will see that individuals' decisions can be systematically wrong because people's decisions are driven by emotions or misunderstandings or because people use inappropriate rules of thumb, also called heuristics, to handle information and make decisions. Certainly, financial markets are very complex so that optimization can lead to fragile results and good heuristics are preferable.1 But what is typically observed is that people apply successful heuristics from other domains without properly assessing their effect in the investment domain. One example for the latter is adaptive learning, which is very successful in many day‐to‐day situations like choosing food: One tries out a new wine. If one likes it, one buys it again. However, in finance it leads to buying assets when they are expensive and selling them when they are cheap, as the roller coaster in Figure 2.01 illustrates.

FIGURE 2.01 Market dynamics and decision behavior of a typical investor

To more deeply understand why we may observe such behavior, we consider a typical decision‐making process and discuss how each stage of the process can be biased. First, decision makers select the information that appears to be relevant for their decisions. Then, they process the selected information to form beliefs and to compare alternatives. After deciding, individuals receive new information as a feedback. This feedback influences, in return, the way the decision makers search for more data, that is, the loop is closed.

The chapter provides evidence that certain mistakes can occur in each of these steps. It discusses the relevance of these mistakes for investors and suggests strategies to avoid the mistakes.

2.1 INFORMATION SELECTION BIASES

When confronted with information, individuals need to judge how relevant it is for the task they need to handle. Thereby individuals seem to consider only particular information while disregarding other that might be relevant as well. For investment decisions, such information filtering can be dangerous since there is uncertainty about the relative importance of economic factors for the future—investment rules that have worked in the past do not always work in the future. So, are there any patterns in the way people select relevant information, and why should we expect that their impact is systematic?

2.1.1 Attention Bias

The first observation on individuals' selection of information is that it can be biased due to a specific task. People gather information that they think is relevant for dealing with the problem and disregard others, which they would otherwise notice. This is demonstrated in an experiment, where participants have been asked to watch a video with two basketball teams: one team wearing black shirts and another one wearing white shirts (Simons & Chabris, 1999). The task was to count the passes of the white team. Afterward, participants have been asked whether they have observed something unusual. Some participants spotted that there was a second ball. But only a few noticed a big black gorilla walking slowly through the picture, stopping in the middle, winking, and passing slowly away. The reason for not seeing the gorilla is the attention bias. Due to the limited attention that people have, they can get only the information they consider important for solving a specific task. All other information remains disregarded, independent of how extreme it is. Hence, when people focus too much on one task, something unexpected can happen that they might not notice. Moreover, related experiments show that even when people know that something unexpected might happen (e.g., that a gorilla would appear), this doesn't help them notice otherunexpected things.

Relevance for Investors and Moderation    The attention bias is relevant for investors because all investors use media to inform themselves. But the media process follows certain patterns. Some media set the agenda, other media follow, and for some time all media report the same story. In these times, other investment relevant information is not seen—like the gorilla in the experiment just mentioned. For example, in summer 2011 we observed a global stock market downturn: From the end of July to the end of August, the DJIA fell from 12,700 to 10,700, the Euro Stoxxs 50 fell from 2,800 to 2,200, and the Nikkei from 10,000 to 8,750 (i.e., stock markets plunged by 16%, 21% and 12.5%, respectively). Looking at the words Internet users searched in Google2 during summer 2011, we see that the public attention mainly focused on the US debt ceiling debate that was positively resolved by August 1st. So why did stocks decline after the showdown in the US Congress was resolved? One explanation is that the gorilla “US recession” was not seen in July, so the attention for a possible recession in the United States was hidden behind the budget ceiling debate while after that debate was over the recession attracted the attention of the public. Indeed, the search for the words “US debt” peaked in July 2011 while the words “US recession” peaked in August 2011. And indeed, the US business cycle slowed down considerably during the summer of 2011.

The best moderation of the attention bias is to agree on certain key information (e.g., macroeconomics, politics, valuation levels, sentiment of the market) that one always discusses with the investors irrespectively of whether it is topical or not.

2.1.2 Selective Perception Based on Experience

Perception of information is, by its nature, always selective. But in many situations people might not be able to see things just because they do not expect them to occur given their experience. This has been demonstrated in an experiment with playing cards (Bruner & Postman, 1949). Participants were shown five playing cards and asked what they have seen. What researchers were testing is whether the participants would recognize doctored cards (e.g., a black three of a heart). They found out that, on average, participants needed four times longer to recognize a doctored card than a normal card. Most of the people were very sure that the doctored card was a normal card. Even when participants recognized that something was wrong, they sometimes misperceived the incongruity (e.g., people who were shown a black four of hearts declared that the spades were “turned the wrong way”). This experiment shows that experience can influence the way people look at new evidence. When people have enough experience with a specific situation, they often see what they expect to see based on their experience. Hence, in some cases, experience may lower performance.

Relevance for Investors and Moderation    To give an example of how selective perception can affect investments, recall the stock market crash in the years 2007–2008. From the summer of 2007 to the beginning of 2009, the DJIA fell from 14,100 to 6,525, the Euro Stoxxs 50 from 4,500 to 1,800 and the Nikkei from 18,250 to 7,125—that is, stock markets plunged between 50% and 60% around the world. Unfortunately, none of the standard indicators could predict this decline. The P/E ratios and the Fed measure that could predict for example the crash of the dot‐com bubble signaled no risk during the summer of 2007. Investors who used those risk measures because of the positive experience with them were caught by surprise during the stock market crash of 2007–2008. Indeed, that stock market crash did not come from overvaluation of stocks but from a bubble in the housing market in the United States, the United Kingdom, and Spain. This housing bubble resulted in a financial crisis, which then slowed down the global economy. Thus, experience with some indicators might seduce investors to stop thinking transversally.

The best way to deal with the selective perception bias is to ask yourself: What is my motivation to see things in a certain way? What expectations did I bring into the situation? Why do others not share my view?

2.1.3 Confirmation Bias

Previous experience influences the way we perceive information that we face, but it also affects the way we search for information. People tend to search for information that confirms one's beliefs or hypotheses, while they give disproportionately less consideration to alternative possibilities. This bias in information selection is known as the confirmation bias. It has been first discovered by Wason (1960). In his experiment, participants were asked to identify a rule applied to triples of numbers (e.g., 2, 4, and 6). To discover the rule, participants could decide on their own triples and receive a feedback on whether their numbers conform to the rule or not. While the true rule was “three numbers of increasing order of magnitude,” most participants tested a specific hypothesis as for example “increasing by 2.” However, those who test their rule can never discover that their rule is wrong because all examples that fit their rule fit also the true rule. Thus, to test the rule “increasing by 2,” it is critical to try, for example, 2, 4, and 7.

Although there are circumstances where searching for confirmatory evidence can be useful...