11  Cognitive biases

Readings

Required: Bazerman & Moore (2012, ch. 1-2)

In section Section 2.3.3, we discussed how human beings have limited cognitive abilities to arrive at optimal solutions. Behavioral economics, pioneered by Amos Tversky (1937-1996) and 2002 Nobel Prize winner Daniel Kahnemann (*1934), has identified several biases that explain why and when people fail to act perfectly rationally. In the following sections, we will explore some of the most prominent biases that arise from humans relying on heuristics in decision-making. Specifically, we will describe biases result from the use of availability, representative, and confirmation heuristics and can lead to flawed decision-making and negative outcomes for individuals and organizations. By recognizing and accounting for these biases, we can make better decisions.

Exercise 11.1 Shepard Tabletop Illusion

Figure 11.1: Shepard tabletop illusion

Look at the two table shwon in Figure 11.1. Which one is larger, and which one looks more like a square?

11.1 Availability heuristic

The availability heuristic refers to our tendency to make judgments or decisions based on information that is easily retrievable from memory. For example, if someone hears a lot of news about a particular stock or investment, they may be more likely to invest in it, even if there are other better investment options available. This bias can also lead individuals to overestimate the frequency of certain events, such as the likelihood of a market crash, based on recent media coverage.

11.2 Representative heuristic

The representative heuristic refers to our tendency to make judgments based on how similar something is to a stereotype or preconceived notion. For example, an investor might assume that a company with a flashy website and marketing materials is more successful than a company with a more low-key image, even if the latter is actually more profitable. This bias can also lead to assumptions about the performance of certain investment strategies based on their resemblance to other successful strategies.

11.3 Confirmation heuristic

The confirmation heuristic refers to our tendency to seek out information that confirms our existing beliefs and ignore information that contradicts them. For example, an investor who strongly believes in the potential of a particular investment might only read news and analysis that supports their belief and ignore any information that suggests otherwise. This can lead to a failure to consider potential risks and downsides of an investment.

11.4 Investment mistakes

Investing can be a daunting task, but avoiding some common investment mistakes can help set you on the right path to financial success. The following list list shows according to Stammers (2016) the Top 20 common investment mistakes without the explanations provided in the paper:

Stammers, R. (2016). Tips for avoiding the top 20 common investment mistakes (Version 1.0). CFA Institute; Accessed April 24, 2023. https://www.cfainstitute.org/-/media/documents/support/future-finance/avoiding-common-investor-mistakes.pdf
  • Expecting too much or using someone else’s expectations: Nobody can tell you what a reasonable rate of return is without having an understanding of you, your goals, and your current asset allocation.
  • Not having clear investment goals: Too many investors focus on the latest investment fad or on maximizing short-term investment return instead of designing an investment portfolio that has a high probability of achieving their long-term investment objectives.
  • Failing to diversify enough: The best course of action is to find a balance. Seek the advice of a professional adviser.
  • Focusing on the wrong kind of performance: If you find yourself looking short term, refocus.
  • Buying high and selling low: Instead of rational decision making, many investment decisions are motivated by fear or greed.
  • Trading too much and too often: You should always be sure you are on track. Use the impulse to reconfigure your investment portfolio as a prompt to learn more about the assets you hold instead of as a push to trade.
  • Paying too much in fees and commissions: Look for funds that have fees that make sense and make sure you are receiving value for the advisory fees you are paying.
  • Focusing too much on taxes: It is important that the impetus to buy or sell a security is driven by its merits, not its tax consequences.
  • Not reviewing investments regularly: Check in regularly to make sure that your investments still make sense for your situation and that your portfolio doesn’t need rebalancing.
  • Taking too much, too little, or the wrong risk: Make sure that you know your financial and emotional ability to take risks and recognize the investment risks you are taking.
  • Not knowing the true performance of your investments: Many investors do not know how their investments have performed in the context of their portfolio. You must relate the performance of your overall portfolio to your plan to see if you are on track after accounting for costs and inflation.
  • Reacting to the media: Using the news channels as the sole source of investment analysis is a common investor mistake. Successful investors gather information from several independent sources and conduct their own proprietary research and analysis.
  • Chasing yield: High-yielding assets can be seductive, but the highest yields carry the highest risks. Past returns are no indication of future performance. Focus on the whole picture and don’t get distracted while disregarding risk management.
  • Trying to be a market timing genius: Market timing is very difficult and attempting to make a well-timed call can be an investor’s undoing. Consistently contributing to your investment portfolio is often better than trying to trade in and out in an attempt to time the market.
  • Not doing due diligence: Check the training, experience, and ethical standing of the people managing your money. Ask for references and check their work on the investments they recommend. Taking the time to do due diligence can help avoid fraudulent schemes and provide peace of mind.
  • Working with the wrong adviser: An investment adviser should share a similar philosophy about investing and life in general. The benefits of taking extra time to find the right adviser far outweigh the comfort of making a quick decision.
  • Letting emotions get in the way: Investing can bring up significant emotional issues that can impede decision-making. A good adviser can help construct a plan that works no matter what the answers to important financial questions are.
  • Forgetting about inflation: It’s important to focus on real returns after accounting for fees and inflation. Even if the economy is not in a massive inflationary period, some costs will still rise, so it’s important to focus on what you can buy with your assets, rather than their value in dollar terms.
  • Neglecting to start or continue: Investment management requires continual effort and analysis to be successful. It’s important to start investing and continue to invest over time, even if you lack basic knowledge or have experienced investment losses.
  • Not controlling what you can: While you can’t control what the market will bear, you can control how much money you save. Continually investing capital over time can have as much influence on wealth accumulation as the return on investment and increase the probability of reaching your financial goals.

Exercise 11.2 Heuristics can fail

Respond to the following problems which are taken from Bazerman & Moore (2012, p. 15f). In class, we will discuss your answers and how they match with the mathematically correct solutions to these problems.

Problem 1: Please rank the following causes of death in the United States between 1990 and 2000. Place a 1 next to the most common cause, 2 next to the second, and so on. - Tobacco - Poor diet and physical inactivity - Motor vehicle accidents - Firearms (guns) - Illicit drug use

Now, estimate the number of deaths caused by each of these five causes between 1990 and 2000.

According to Mokdad et al. (2004, p. 1240), the answer is the following order:

  • Tobacco
  • Poor diet and physical inactivity
  • Motor vehicle accidents
  • Firearms (guns)
  • Illicit drug use

Problem 2: Estimate the percentage of words in the English language that begin with the letter “a.”

Problem 3: Estimate the percentage of words in the English language that have the letter “a” as their third letter.

Most people estimate that there are more words beginning with “a” than words in which “a” is the third letter. In fact, the latter are more numerous than the former. Words beginning with “a” constitute roughly 6 percent of English words, whereas words with “a” as the third letter make up more than 9 percent of English words.

Problem 4: Lisa is thirty-three and pregnant for the first time. She is worried about birth defects like Down syndrome. Her doctor tells her there is only a 1 in 1,000 chance that a woman of her age will have a baby with Down syndrome. However, Lisa remains anxious and decides to get a test called the Triple Screen, which detects Down syndrome. The test is moderately accurate: when a baby has Down syndrome, the test gives a positive result 86% of the time. However, 5% of babies who don’t have Down syndrome also get a false positive. Lisa takes the test and gets a positive result. What are the chances that her baby has Down syndrome?

  1. 0-20%
  2. 21-40%
  3. 41-60%
  4. 61-80%
  5. 81-100%

Most people think that Lisa has a substantial chance of having a baby with Down syndrome. The test gets it right 86 percent of the time, right? That sounds rather reliable, doesn’t it? Well, it does, but we should not rely on our feelings here. It’s better to do the math, because the correct result would show that there is just a 1.7 percent chance of the baby having Down syndrome. Here is the proof for that small number:

Let \(A\) be the event of the baby having Down syndrome and \(B\) the event of a positive test result. Then,

\[\begin{align*} P(A) &= 0.001 \\ P(B \mid A) &= 0.86 \\ P(B \mid \neg A) &= 0.05 \\ P(B) &= \frac{999 \cdot 0.05}{1000} + \frac{1 \cdot 0.86}{1000} = \frac{50.81}{1000} = 0.05081 \\ P(A \mid B) &= \frac{P(B \mid A) P(A)}{P(B)} = \frac{0.86 \cdot 0.001}{0.05081} = 0.01693 \end{align*}\]

Problem 5: A town is served by two hospitals. In the larger hospital, about 45 babies are born each day. In the smaller hospital, about 15 babies are born daily. About 50% of all babies are boys. For a year, each hospital recorded days when more than 60% of the babies born were boys. Which hospital recorded more such days?

  1. The larger hospital
  2. The smaller hospital
  3. About the same (within 5%)

Most individuals choose C, expecting the two hospitals to record a similar number of days on which 60 percent or more of the babies born are boys. However, statistics tell us that we are much more likely to observe 60 percent of male babies in a smaller sample than in a larger sample. Think about which is more likely: getting more than 60 percent heads in 3 flips of a coin or in 3,000 flips of a coin? Half of the time, 3 flips will produce more than 60 percent heads. But with 3,000 flips, it happens about 0.0001 percent of the time. Most people ignore sample size when judging probabilities.

Problem 6: You and your spouse have had three daughters. Now expecting a fourth child, you wonder about the odds of having a boy. What is the best estimate of your chances of having another girl?

  1. 6.25% (1 in 16), because the odds of getting four girls in a row is 1 in 16
  2. 50% (1 in 2), because there is an equal chance of getting each gender
  3. Somewhere between 6.25% and 50%

Many assume that after having three girls, the probability of having another girl must be lower. However, the gender determination of each baby is independent; the chance remains 50 percent for each child, regardless of previous children.

Problem 7: You manage a Major League Baseball team, and the 2005 season has just ended. Your job is to predict future player performance. You must estimate 2006 batting averages for nine players. Fill in your guesses in the right column:

Player 2005 Batting Average Estimated 2006 Batting Average
1 .215
2 .242
3 .244
4 .258
5 .261
6 .274
7 .276
8 .283
9 .305

Most people predict that a player’s 2006 performance will be almost identical to their 2005 performance. However, statistics show the correlation between Major League Baseball players’ batting averages from one year to the next is only 0.4. This tendency is known as “regression to the mean”—the worst performers tend to improve, and the best tend to decline.

Problem 8: Linda is 31, single, outspoken, and very smart. She majored in philosophy and was deeply concerned with issues of discrimination and social justice. Rank the following descriptions in order of how likely they are to describe Linda:

  1. Linda is a teacher in an elementary school.
  2. Linda works in a bookstore and takes yoga classes.
  3. Linda is active in the feminist movement.
  4. Linda is a psychiatric social worker.
  5. Linda is a member of the League of Women Voters.
  6. Linda is a bank teller.
  7. Linda is an insurance salesperson.
  8. Linda is a bank teller who is active in the feminist movement.

Examine your rank orderings of descriptions C, F, and H. Most people rank C as more likely than H, and H as more likely than F. However, a conjunction (being both a bank teller and a feminist) cannot be more probable than being a bank teller alone. This is a fundamental law of probability, but it’s commonly misunderstood due to representativeness bias.

Problem 9: Take the last three digits of your phone number. Add a “1” to the front to form a four-digit number. Now, estimate whether the Taj Mahal was completed before or after this year.
___ Before ___ After

Now, make your best estimate of the actual year in which the Taj Mahal was completed: ___

Most people are influenced by irrelevant information, such as their phone number. If your phone number resulted in a year like 1978 or 1040, your estimate might change. In reality, the Taj Mahal was completed in 1648, but people with high phone numbers tend to give more recent estimates.

Problem 10: Which of the following instances seems most likely? Which is the second most likely?

  1. Drawing a red marble from a bag containing 50% red marbles and 50% white marbles.
  2. Drawing a red marble seven times in succession (with replacement) from a bag containing 90% red marbles and 10% white marbles.
  3. Drawing at least one red marble in seven tries (with replacement) from a bag containing 10% red marbles and 90% white marbles.

The most common order of likelihood chosen is \(B > A > C\). However, the correct order is C (52 percent), A (50 percent), and B (48 percent). This illustrates a bias where people overestimate the probability of conjunctive events (where multiple events must happen together) and underestimate disjunctive events (where only one of many events needs to happen).

Problem 11: Ten uncertain quantities are listed below. For each, write down your best estimate. Then, put a lower and upper bound around your estimate so you’re 98% confident the range includes the actual value.

Estimate Lower Bound Upper Bound
a. Wal-Mart’s 2006 revenue
b. Microsoft’s 2006 revenue
c. World population (July 2007)
d. Market cap of Best Buy (July 2007)
e. Market cap of Heinz (July 2007)
f. McDonald’s rank in 2006 Fortune 500
g. Nike’s rank in 2006 Fortune 500
h. US motor vehicle fatalities (2005)
i. US national debt (July 2007)
j. US federal budget (FY 2008)

The correct answers are:
(a) $351 billion,
(b) $44 billion,
(c) 6.6 billion people,
(d) $23 billion,
(e) $15 billion,
(f) 108,
(g) 158,
(h) 43,443,
(i) $8.8 trillion,
(j) $2.9 trillion.

Most people are overconfident, estimating too narrow a range for these quantities. Despite claiming a 98 percent confidence, many fail to surround more than 30–70 percent of the actual values.

Problem 12: Which best describes the relationship between a baseball player’s batting average in one season and the next?

  1. Zero correlation
  2. Weak correlation (about .4)
  3. Strong correlation (about .7)
  4. Perfect correlation (1.0)

The correct answer is a correlation of about 0.4 between batting averages from one season to the next.

After answering the 12 questions, please watch this video: https://youtu.be/wEwGBIr_RIw

Exercise 11.3 Twelve cognitive biases

In the textbook of Bazerman & Moore (2012) twelve cognitive biases that are described. Read chapter 2 of the book and summarize the twelve biases in a sentence.

  1. Ease of Recall: Individuals tend to consider events that are more easily remembered to be more frequent, regardless of their actual frequency.

  2. Retrievability: Individuals’ assessments of event frequency are influenced by how easily information can be retrieved from memory.

  3. Insensitivity to Base Rates: Individuals tend to ignore the frequency of events in the general population and focus instead on specific characteristics of the events.

  4. Insensitivity to Sample Size: Individuals often fail to take sample size into account when assessing the reliability of sample information.

  5. Misconceptions of Chance: Individuals expect random processes to produce results that look “random” even when the sample size is too small for statistical validity.

  6. Regression to the Mean: Individuals fail to recognize that extreme events tend to regress to the mean over time.

  7. Conjunction Fallacy: Individuals often judge that the occurrence of two events together is more likely than the occurrence of either event alone.

  8. Confirmation Trap: Individuals tend to seek out information that confirms their existing beliefs and ignore information that contradicts them.

  9. Anchoring: Individuals often make estimates based on initial values, and fail to make sufficient adjustments from those values.

  10. Conjunctive- and Disjunctive-Events Bias: Individuals tend to overestimate the likelihood of conjunctive events (two events occurring together) and underestimate the likelihood of disjunctive events (either of two events occurring).

  11. Overconfidence: Individuals tend to be overconfident in the accuracy of their judgments, particularly when answering difficult questions.

  12. Hindsight and the Curse of Knowledge: After learning the outcome of an event, individuals tend to overestimate their ability to have predicted that outcome. Additionally, individuals often fail to consider the perspective of others when making predictions.

Bazerman, M. H., & Moore, D. A. (2012). Judgement in managerial decision making. In Inc. Hoboken, NJ, USA (8th ed.). John Wiley & Sons.