Behavioral economics

Behavioral economics studies the ways actual investors and consumers depart from the rational-agent assumptions traditionally used in economics, and what those departures mean for markets and collective decisions. Its significance in this account lies in refusing a simple equation between irrational individuals and irrational outcomes: individual biases can be costly, yet aggregation may still produce sound judgments when errors do not systematically point in the same direction. When incentives or shared tendencies align those errors, however, collective mistakes can emerge, making the design of decision rules consequential.

From Rational Agents to Behavioral Evidence

For much of the discipline’s history, economists treated people as fundamentally rational in their economic lives, or at least as behaving as if they were rational. They knew that actual consumers did not perfectly match an ideal model, but without a clear account of how behavior deviated, it was difficult to develop rigorous claims about how markets worked. That position has since changed: economists have devoted substantial attention to the psychology and behavior of investors and consumers, documenting departures from rationality that occur across significant groups of people.

The documented tendencies include herding: investors may find the company of others safer than making independent choices. They can give excessive weight to recent, prominent news while overlooking longer-lasting trends or less dramatic information. They may mistake a money manager’s recent success for reliable evidence of skill, even when randomness could explain it. Losses can feel roughly twice as painful as gains feel pleasurable, encouraging investors to hold losing stocks rather than acknowledge a loss. Overconfidence can lead people to trade far more than is prudent, turning confidence in one’s judgment into a cost.

Evidence from individual investors illustrates that cost in concrete terms. A study by Brad M. Barber and Terrance Odean examined the stocks bought and sold by 66,000 individual investors between 1991 and 1996. The average investor turned over 75 percent of a portfolio each year; the most aggressive traders turned over 250 percent. While the market returned 17.9 percent annually over the period, active investors earned 11.4 percent, and the average investor earned 16.4 percent. On these figures, frequent trading diminished returns: investors would generally have done better by trading less.

Individuals and the Collective Judgment

Professional experience does not remove these vulnerabilities. Professional investors also herd, grow overconfident, underestimate randomness, and may explain good results as skill while attributing bad ones to luck. The fact that most money managers perform worse than the market makes it difficult to treat professionals as exemplars of rationality. Yet recognizing individual irrationality does not, by itself, establish that markets are irrational. The account’s recurring distinction is between the behavior of a person and the result produced when many people’s judgments are combined.

Overconfidence, for example, explains why people trade so much and why that activity may hurt individual traders. But it does not necessarily push prices in one consistent direction: an overconfident investor might believe a stock just purchased will rise, while another might be equally certain that a stock sold short will fall. Unless overconfidence is systematically associated with a particular view—such as a shared dislike of technology stocks—it need not skew market prices uniformly. Likewise, investors may overvalue recent news without all interpreting that news in the same way. When biases vary, their effects can offset one another rather than accumulate into a single market error.

This distinction complements the account’s broader description of wise crowds. A crowd is more likely to judge well when its members have diverse opinions, make independent judgments, draw on decentralized or local knowledge, and have a way to aggregate what they know. Averaging estimates can cancel individual errors, leaving information that no one person possesses in full. But cancellation alone is not enough: the judgments must contain some relevant information in the first place. The point is not that a crowd is invariably wise, but that its collective verdict may contain more usable information than the imperfect reasoning of any one member suggests.

Decision Rules and the Value of Independence

Behavioral evidence also bears on how a group should make decisions. In an experiment by Angela Hung and Charles Plott, students had to identify which of two urns was being used. One urn held twice as many light marbles as dark ones; the other held twice as many dark marbles as light ones. Volunteers drew marbles in sequence, kept the color of each draw private, and announced only their guesses. A correct answer earned a small payment. Because a participant could see earlier guesses but not earlier draws, the sequence created a conflict between private evidence and the apparent judgment of the group.

In most cases, a student faced with earlier guesses favoring one urn followed them, even when a privately drawn marble suggested the other. Information cascades began in 78 percent of trials under these rules. Hung and Plott then changed the reward: instead of being paid for an individually correct answer, students were paid according to whether the group’s majority answer was correct. The individual still had imperfect private evidence, but the incentive now made the group’s accuracy the objective.

Under the group-based reward, students paid more attention to their private information and less to the guesses they had observed. An individual who relied on a single marble could be more likely to guess incorrectly than one who followed a cascade, yet the group became significantly more accurate when its members were encouraged to contribute their separate evidence. The experiment thus demonstrates a design principle: reducing the weight of earlier choices can strengthen the independence needed for aggregation. Organizations can do this more readily than a whole economy, for example by collecting judgments simultaneously or nearly so rather than allowing the first opinions to shape all those that follow.

Incentives, Cooperation, and Organizational Information

The same attention to rules applies to cooperation. In the public-goods experiments run by Ernst Fehr and Simon Gächter, four players each received 20 tokens for a game lasting four rounds. Each could contribute tokens to a shared pot or keep them. Contributing a token cost the contributor one token, while each of the other three players received 0.4 tokens; the group gained 1.6 tokens in total for every token invested. If everyone kept their money, each ended with 20 tokens; if everyone contributed all they had, each ended with 32. An individual could benefit by free riding on others’ contributions, but universal free riding left the group with no contributions.

Most participants initially contributed about half their tokens. As the rounds passed and players observed free riding, contributions declined; by the end, 70 to 80 percent of players were free riding, and the group as a whole was poorer than it could have been. The pattern shows how choices respond to expectations about other people: willingness to contribute can erode when participants see cooperation going unrewarded. In economic behavior, therefore, the outcome depends not only on private preferences but also on the incentives and expectations built into the interaction.

Organizational incentives can similarly corrupt the information a company needs. In the early 1950s, sociologist Donald Roy worked as a lathe operator in a machine shop where workers were paid by a piece-rate system. Their rate rose after they crossed production thresholds, then rose again at a second hurdle before being capped. If workers produced too quickly, the company could raise the hurdles, so they restricted output and concentrated on manipulating the rate rather than maximizing production. Roy called his account “Goldbricking in a Machine Shop.”

The same problem can occur when managers receive bonuses for meeting targets. If rewards depend disproportionately on exceeding a set expectation, managers have reason to make that expectation easier by understating forecasts, then to meet it by whatever means are available. Michael C. Jensen’s analysis describes the resulting incentive as one that effectively pays people to lie; accounting maneuvers can improve current results at the expense of the future. Such cases connect behavioral economics to institutional design: a decision system cannot be judged only by the goal it announces, but also by the behavior its rewards make advantageous.

When Bias Becomes Systematic—and How Design Can Help

The distinction between offsetting errors and systematic bias marks a limit to the reassuring case for aggregation. If deviations from rationality are sufficiently random, they may cancel out. If they all push in the same direction, markets can do a much worse job of finding a good solution. One example is Americans’ tendency to undersave. People often say they should save, yet procrastinate when it comes to doing so, valuing present consumption so heavily that setting aside money for the future is deferred. Retirement consumption then drops substantially, and older people live on considerably less than they did while working.

Here, individual delay can add up to a collective problem: many people may reach retirement without enough savings to live comfortably. The proposed response is not necessarily to impose a mandate, but to alter the choice structure. When enrollment in retirement plans is automatic, with action required to opt out rather than to opt in, people are more likely to remain enrolled. Workers are also more willing to commit a portion of future income than current income. Richard H. Thaler and Shlomo Benartzi used that tendency in a company retirement plan that allowed workers to choose different savings rates for present and future income. Workers adopted substantially higher rates for income months ahead, and their average savings rate doubled within a short time.

These interventions work by making saving easier and spending harder while using preferences people already have. Rather than assuming that individuals will reliably overcome inertia through sheer rational calculation, they reshape the available options and the frame through which a decision is made. The broader lesson is neither that markets automatically correct every human weakness nor that psychological error invalidates collective judgment. Behavioral economics asks which individual tendencies remain dispersed and which become systematic, then examines whether rules can preserve independent information, align incentives, and make more collectively rational behavior easier to achieve.