Economics•Chapter 7•5 min read•Updated September 24, 2026

Behavioral Economics — Social Preferences: Fairness, Reciprocity and Trust

O
OIYO EditorialContributor
7/10

Behavioral Economics — People Care Not Only About Their Own Share but About the Gap with Others

Agents in the standard model maximize only their own payoff. This assumption predicts many outcomes in competitive markets well, but its predictions go badly astray where people face each other: bargaining, team production, paying taxes and giving. This chapter measures those departures with experimental games and organizes them with models.

1. Four experimental games

Games that measure social preferences
GameStructureSelfish predictionTypical results
UltimatumA proposer splits 100; the responder accepts or rejects (rejection leaves both with 0)Offer the minimum; the responder accepts anythingOffers of 40–50% are common; about half of offers below 20% are rejected
DictatorThe proposer splits unilaterallyGive 0On average 20–30% is given
Public goodsContributions are multiplied and shared equallyContribute 040–60% contributed at first, declining with repetition
TrustMoney sent by an investor triples and a trustee returns someInvest 0, return 0Most invest, and on average get back around the amount invested

A responder in the ultimatum game who rejects a positive amount is throwing money away. A proposer who gives close to half may do so out of fairness or out of fear of rejection. The dictator game removes the risk of rejection and so measures pure altruism separately. The difference between the two games (less is given in the dictator game) shows that a large part of ultimatum offers is strategic.

2. The inequity aversion model

Fehr and Schmidt modelled people as disliking not only low payoffs but also differences from others.

Fehr-Schmidt utility
Ui=xi−αimax⁡(xj−xi, 0)−βimax⁡(xi−xj, 0),αi≥βi≥0U_i = x_i - \alpha_i \max(x_j - x_i,\,0) - \beta_i \max(x_i - x_j,\,0),\quad \alpha_i \ge \beta_i \ge 0
α is the discomfort when others have more (envy) and β the discomfort when I have more (guilt). Disadvantageous inequity is disliked more (α ≥ β). With α = β = 0, it is the standard model.

In the ultimatum game, a responder offered ss out of 100 gets s−α(100−2s)s - \alpha(100-2s) by accepting and 0 by rejecting. The condition for accepting is

s≥α1+2α×100s \ge \frac{\alpha}{1+2\alpha}\times 100

With α=1\alpha=1, offers below 33.3 are rejected; with α=0\alpha=0, even 1 won is accepted. On the proposer’s side, if β≥0.5\beta\ge 0.5 the discomfort of increasing inequality by 2 outweighs the gain of increasing one’s own share by 1, so the proposer offers 50:50.

A single model explains the different results across games. It also explains why, where one person cannot change inequality — as in competitive markets — inequity aversion does not change behaviour and the outcome is the same as in the standard model.

3. Reciprocity: responding to intentions

Inequity aversion looks only at outcomes. Experiments show that people also respond to others’ intentions. The same 80:20 split is rejected more often when the proposer could have chosen 50:50 but chose 80:20, and less often when the only options were 80:20 and 90:10. Rabin’s fairness model captures preferences for repaying kindness with kindness and hostility with hostility.

Gift exchange in labour markets applies reciprocity. When an employer pays more than the market wage, workers reciprocate with more effort. This has been confirmed repeatedly in the laboratory, but in field experiments the effort effect of a pay rise has sometimes vanished after the first few hours, so its size and persistence are debated.

4. Punishment sustains cooperation

Contributions in the public goods game decline because of conditional cooperation: “if others don’t pay, neither will I.” Fehr and Gächter (2000) showed that letting players spend their own money to punish others after each round kept contributions at a high level. Punishment is costly to the punisher and brings no gain, so a selfish person would not do it. This altruistic punishment is interpreted as a basis for the enforcement of social norms.

5. When monetary incentives crowd out motivation

Gneezy and Rustichini (2000) introduced a small fine for late pick-ups by parents at six of ten Israeli day-care centres. Late pick-ups did not fall but rose, and did not return to the original level even after the fine was removed. The interpretation is that the fine turned “something to feel sorry about towards the teachers” into “a service you can buy.” It is evidence that monetary incentives can crowd out intrinsic and social motivation. This is also why the design of Pigouvian taxes in public finance must ask whether the tax weakens norms.

Check your understanding

Below what offer out of 100 does a responder with α = 0.5 reject? s≥0.5/(1+1)×100=25s \ge 0.5/(1+1)\times 100 = 25, so offers below 25 are rejected. If the proposer does not know the distribution of responders’ α, even a selfish proposer has an incentive to offer comfortably more than 25 to reduce the risk of rejection. This strategic motive is why ultimatum offers exceed the shares given in the dictator game.

References

  • Ernst Fehr and Klaus Schmidt, “A Theory of Fairness, Competition, and Cooperation,” Quarterly Journal of Economics (1999)
  • Ernst Fehr and Simon Gächter, “Cooperation and Punishment in Public Goods Experiments,” American Economic Review (2000)
  • Uri Gneezy and Aldo Rustichini, “A Fine Is a Price,” Journal of Legal Studies (2000)
  • Colin Camerer, Behavioral Game Theory (2003), ch. 2
O

OIYO Editorial

Editorial Desk

The OIYO editorial desk researches money, law, lifestyle, and self-understanding topics against primary sources and public statistics. Every piece carries source notes and is reviewed on a regular cycle for accuracy and usefulness.