Decision making
Amos Tversky (1937–96)
Daniel Kahneman (1934–)
1940s US economist Herbert Simon argues that rational decision alone does not explain human decision making.
1953 French economist Maurice Allais criticizes expected utility theory, saying that real-life decisions are not always taken rationally.
1990 Economists Andrei Shleifer and Lawrence Summers show that irrational decisions can affect prices.
2008 US psychologist and economist Dan Ariely publishes Predictably Irrational, showing irrationality has a pattern.
Until the 1980s standard economic theory was dominated by the idea of “rational economic man”. Individuals were understood to be agents who look at all decisions rationally, weighing the costs and benefits to themselves and making a decision that will give them the best outcome. Economists thought that this was how people behaved in situations of both certainty and uncertainty, and they formalized the idea of rational decision making in expected utility theory. In reality, however, people often make irrational decisions that don’t give them the highest payoffs and may even hurt their own prospects.
Early studies of these quirks of behavior were made in 1979 by two Israeli-American psychologists, Amos Tversky and Daniel Kahneman. They looked at the psychology involved in decision making and backed up their hypotheses with empirical examples. Their key paper, Prospect Theory: An Analysis of Decision under Risk, outlined a theory that marked the start of a new branch of study known as behavioral economics. This aimed to make economists’ theories about decision making more psychologically realistic.
"One may discover that the relative attractiveness of options varies when the same decision problem is framed in different ways."
Amos Tversky, Daniel Kahneman
Tversky and Kahneman found that people commonly violate economists’ standard assumptions about behavior, particularly when consequences are uncertain. Far from acting with rational self-interest, people were found to be affected by the way a decision is presented and responded in ways that violate standard theory.
Economists had long understood that people are often “risk-averse.” For example, if given a choice between definitely receiving $1,000 or having a 50 percent chance of receiving $2,500, people are more likely to choose the guaranteed $1,000—despite the fact that the average expectation of the second, uncertain, option is higher, at $1,250. The psychologists constructed the opposite situation, giving the same people the choice of either definitely losing $1,000, or having a 50 percent chance of no loss and a 50 percent chance of losing $2,500. In this situation, people who chose the safe option in the previous example now chose the riskier alternative of the gamble between no loss and a large loss. This is known as risk-seeking behavior.
The standard economic approach to decision making under uncertainty assumed that any one individual was risk-averse, risk-loving, or didn’t mind either way. These risk preferences would apply whether the individual was facing risks that involved gains or losses. However, Tversky and Kahneman found that individuals are risk-averse when facing gains but risk-loving when facing losses: the nature of individual preference seems to change. Their work showed that people are “loss averse,” and so are willing to take risks to avoid losses, where they would not be willing to take risks to gain something. For example, the loss in utility from losing $10 appears to be greater than the gain in utility from gaining $10.
These quirks in behavior show that the way that choices are presented influences people’s decisions, even if the ultimate outcomes are the same. For example, consider a situation where a disease is projected to kill 600 people. Two programs exist to counter the disease: A, which saves 200; and B, which offers a one-third chance that 600 people will be saved versus a two-thirds chance that no one will be saved. When the problem is explained to them in this way, the majority of people show themselves to be risk-averse—they opt for the certainty of saving 200 people. If the question is restated, however, with the choice being between program C, which guarantees the death of 400 people, or program D which offers a one-third chance that nobody will die versus a two-thirds chance that 600 people will die, most people will pick the risky program D.
The ultimate outcomes of the pairs of choices are the same: in both A and C we definitely end up with 400 dead, while with B and D, there is an expected outcome of 400 dead. Yet now people prefer the option that is more of a gamble. People are more willing to take risks to prevent lives being lost (a loss) than they are to save lives (a gain). We place more subjective value on losing something than gaining something—losing $10 feels worse, apparently, than gaining $10 feels good.
This tendency toward loss aversion means that, when choices for change are framed in such a way that the consequences are seen as negative, people are more likely to perceive the change as a problem. Knowing this can be used to influence people. For instance, if a government wants to encourage people to adopt something, it is more likely to be successful if it emphasizes the positive gains involved in making that decision. If, on the other hand, it wants people to reject something, it should focus on what they stand to lose.
A government wishing to persuade people to be vaccinated should stress the increased probability of death if they are not vaccinated. People hate losing more than they love winning.
Kahneman and Tversky also showed that the process by which decisions are made can affect choices even when the process doesn’t affect the final payoffs.
For example, imagine a game of two stages in which a player is given a choice of two options at the second stage if they make it that far. However, they must make their choice before the first stage. An example of such a game is laid out on the opposite page.
In this two-stage game, most people choose the guaranteed $3,000 option. However, when the decision is shown as a straight choice between a lower chance of winning $4,000 or a higher chance of $3,000, most people choose the lower chance of winning more money. Why the change?
In the two-stage process people ignore the first stage because it is common to both outcomes. They see the options as a choice between a guaranteed win and merely the chance of a win, even though the probabilities are altered by the first stage. This contradicts standard economic rationality in which decisions are only influenced by final outcomes.
People’s choices in multistage games vary according to how questions are framed. If they are directed to ignore factors that both choices have in common, such as Stage 1 in this example, they may make inconsistent choices.
The key insights to this work—that we hate to lose more than we like to gain, and that we interpret losses and gains in terms of context—have helped illuminate why people make decisions that are not consistent with utility theory or the idea of “rational economic man.” The theory is a founding pillar of behavioral economics, and has also had wide-ranging influence on marketing and advertising. By understanding the way we make decisions, marketers are able to market their products much more effectively. A good example of this is in-store promotions, which offer “huge discounts” on items with initially inflated prices.
Prospect theory has implications for many kinds of common economic decisions. For example, the theory explains why people may travel to a different part of town in order to save $5 off a $15 DVD, but they are unlikely to make the same trip in order to save $5 off a $400 TV, even though their net wealth is impacted by the same amount in each case. Loss aversion also explains what is known as the endowment effect: people tend to place a higher value on an object when they own it—and do not want to lose it—than before they own it, when it is only a “potential gain.”
Behavioral economics is vital to our understanding of the economy and has introduced psychological realism into modern economics. Prospect theory was the first to suggest that people are not simply 100 percent rational machines. The implications of this realization—for economic theories and government policies—are wide-ranging. For example, giving people a sense of ownership may affect how well they look after something.
A scalper sells a sports ticket for cash. The amount that seller and buyer value the ticket depends not just on its perceived utility, but also on factors such as the way the seller obtained it.
The new field of behavioral economics has provided firms with new ways to drive their businesses. In 2006, a group of economists devised an experiment for a bank in South Africa that wanted to grant more loans. Traditional economists would have advised the bank to lower its interest rate to stimulate demand. Instead, the bank allowed the economists to experiment with various options to find out which might be most profitable for the bank. They sent out 50,000 letters offering different interest rates—some high, some low. The letters also featured photos of employees, and a simple or complicated table showing the different chances of winning a prize if the letter was replied to.
By tracking which customers responded, it was possible to quantify the effect of psychological factors against the purely economic factor of the interest rate. The experiment discovered that the interest rate was only the third most important factor in stimulating demand, and including a photo of a female employee in its marketing had an effect equal to dropping the interest rate by five points. This is a groundbreaking result: identifying psychological factors to stimulate demand can be a lot cheaper than lowering the interest rate.