Social Preferences
Social preferences are the ways people care about outcomes for others, not just for themselves.
10 frameworks, from 1982 to today · Behavioral Economics · Economics
- Experimental Economics of Social Preferences 1982-Present
- Intention-Based Reciprocity 1993-Present
- Strong Reciprocity 1997-Present
- Evolutionary and Cultural Group Selection Models 1998-Present
- Inequity Aversion 1999-Present
- Developmental Approach 2000-Present
- Conditional Cooperation and Social Preference Types 2001-Present
- Cross-Cultural and Field Experimental Approach 2001-Present
- Neuroeconomics of Social Preferences 2003-Present
- Guilt Aversion 2006-Present
Overview
Social preferences are the ways people care about outcomes for others, not just for themselves1. The standard economic agent is a self-interested maximizer: given a choice between more money and less, they take more, and they do not care who else gains or loses23. Behavioral economics found that this picture fails in systematic, repeatable ways1. People give money away in anonymous one-shot settings, punish strangers at their own cost, and refuse profitable deals they consider unfair. The subfield of social preferences is the attempt to describe, measure, and model those departures: to say what people actually want when other people are involved, and why1.
What is at stake is large. If people have social preferences, then predictions about markets, contracts, public goods, taxation, and organizations built on pure self-interest will be wrong in identifiable ways. The history of the subfield is a history of experiments that made other-regarding behavior visible, models that tried to capture it in a few parameters, and a long argument about whether the underlying motive is outcome-based (I care about how the pie is split) or intention-based (I care whether you meant to treat me well).
The Ultimatum and Dictator Games Make Fairness Visible
The founding experiments came from game theory laboratories in the early 1980s. In the ultimatum game, one player proposes a split of a sum of money and the other accepts or rejects; rejection leaves both with nothing14. A self-interested responder should accept any positive offer, and a self-interested proposer should therefore offer the smallest possible amount1. Werner Güth, Rolf Schmittberger, and Bernd Schwarze reported in 1982 that proposers typically offered around a third to a half of the stake, and responders often rejected low offers, sacrificing money to punish unfairness14.
The dictator game, developed in the 1980s and used widely from the 1990s, removed the responder's power: one player simply divides a sum, and the other must accept. Self-interest predicts giving nothing1. In practice a substantial minority give nothing, but many give something, and average giving in student laboratories typically falls somewhere between a fifth and a third of the stake. The dictator game became the workhorse for measuring "other-regarding" generosity because it strips out strategic motives.
These results were initially read as anomalies. The next two decades turned them into a research program.
Strong Reciprocity and the Puzzle of Costly Punishment
A decisive step was the public goods game, in which players contribute to a shared pool that is multiplied and split equally5. Self-interest predicts zero contributions; in practice contributions start high and decay as free riders exploit contributors. Ernst Fehr and Simon Gächter's work in the late 1990s added a punishment stage: players could pay to reduce the earnings of low contributors. Many did, and when punishment was available, contributions stayed high. This was evidence for strong reciprocity: a willingness to reward fair behavior and punish unfair behavior even when the punisher gains nothing and pays a cost6.
The finding mattered because it could not be explained by repeated-game reputation. Punishment occurred in one-shot and anonymous settings. It suggested a genuine, if conditional, moral motivation. The term "strong reciprocity" was coined in this literature to distinguish it from reciprocal altruism in the evolutionary sense, where helping is ultimately self-serving.
Intention-Based Fairness: Does the Motive Matter?
A parallel line asked whether people respond to outcomes or to intentions. In the trust game, one player sends money that is multiplied, and the second may return some17. Returning more than self-interest dictates can be read as reciprocity—but is it reciprocity toward a kind act, or just a taste for equal splits?
The key experiments manipulated intentions while holding outcomes fixed. In a well-known design by Sally Blount (1995), responders in an ultimatum-like setting received the same low offer whether it was chosen by a human proposer or generated by a random device. Responders rejected human-made low offers far more often than randomly generated ones. The outcome was identical; the inferred intention was not. This supported intention-based reciprocity models, in which people reciprocate kind or hostile intentions, not merely favorable or unfavorable results.
Gary Charness and Matthew Rabin formalized a model of this kind in 2002, building on Rabin's earlier 1993 framework in which players care about the kindness of others' intentions. The models were technically demanding because intentions depend on what a player believes about the other's beliefs—a "psychological game" structure. Their appeal was explanatory: they could account for why the same distribution is accepted from a machine and rejected from a person.
Inequity Aversion: A Model of Outcome-Based Fairness
The competing account was outcome-based. In 1999, Ernst Fehr and Klaus Schmidt published a model of inequity aversion: people dislike unequal outcomes, both when they are behind and when they are ahead, though typically more when behind. A single parameter for disadvantageous inequality and one for advantageous inequality could organize a wide range of game results—ultimatum offers and rejections, public goods contributions, and market behavior.
Gary Bolton and Axel Ockenfels published a related model the following year, in which utility depends on one's relative payoff share. The two models differed in details but shared the claim that fairness is about the distribution of payoffs, not the intentions behind it. The debate between intention-based and outcome-based accounts became one of the central theoretical disputes of the subfield, and it was never fully resolved: both motives appear to operate, and later models combined them.
Conditional Cooperation and Heterogeneous Types
A third strand asked not what people want on average but how they differ. In public goods experiments, a recurring pattern is conditional cooperation: many people contribute more when they believe others contribute more1. This is not unconditional altruism; it is a matching rule. The "strategy method," in which subjects state contributions for every possible average contribution of others, made these response functions measurable.
The result was a taxonomy of social preference types. In a widely cited 2002 study, Urs Fischbacher, Simon Gächter, and Ernst Fehr classified subjects as free riders (about 30 percent in their sample), conditional cooperators (about 50 percent), and a remainder with other patterns. The finding reframed the field's question: not "are people selfish or fair?" but "what mixture of types is present, and how does the mixture shape aggregate outcomes?" Heterogeneity mattered because the same institution could produce very different results depending on the composition of the group.
Guilt Aversion and Promises
A fourth line addressed cooperation sustained by communication. In many experiments, allowing players to make non-binding promises raises cooperation sharply, and people often keep promises even when breaking them is profitable. Gary Charness and Martin Dufwenberg proposed in 2006 a model of guilt aversion: people want to avoid letting others down, and they form beliefs about what others expect. A broken promise creates guilt because the promiser believes the other expected them to keep it. The model explained why cheap talk—costless, unenforceable messages—changes behavior, and why people sometimes act against their material interest to avoid disappointing someone.
This work connected social preferences to belief-dependent emotions and to the psychological-game framework, and it showed that the relevant "other" is not just a payoff recipient but a person with expectations.
Where Do Social Preferences Come From?
The models described behavior; a separate question was its origin. One line of evidence came from cross-cultural and field studies. Joseph Henrich and colleagues reported in 2001 and 2006 on ultimatum and dictator games across small-scale societies, finding substantial variation: some groups made near-self-interested offers, others made generous ones, and market integration and cooperation in daily life predicted higher offers. The finding undercut the idea that fairness tastes are universal constants and supported the view that they are shaped by institutions and experience.
A second line came from developmental studies. Children's behavior in sharing and punishment tasks changes with age, and by the school years many children show costly punishment and conditional sharing. A third line came from neuroeconomics: brain-imaging studies in the 2000s, notably work by Alan Sanfey and colleagues in 2003, found that unfair offers in the ultimatum game activated the insula and dorsolateral prefrontal cortex, and that insula activity predicted rejection. This suggested that fairness judgments engage emotional and self-control systems, and it linked social preferences to the broader neuroeconomic program.
A fourth line was evolutionary and theoretical: models of how other-regarding motives could survive selection, including work on strong reciprocity and on cultural group selection. These arguments remained contested, but they framed social preferences as adaptations or cultural products rather than arbitrary tastes.
The Present: A Plural Toolkit
Today the subfield is not organized around a single model. Experimental work continues to refine measurement: the dictator game and its variants, the trust game, public goods games with and without punishment, and the strategy method are standard instruments, and researchers are increasingly careful about how framing, anonymity, and experimenter demand affect results. Theoretically, outcome-based models (inequity aversion and its relatives) and intention-based models (reciprocity, guilt aversion) coexist, and many researchers use hybrid models that include both distributional and belief-dependent motives. Heterogeneity is now treated as a central fact rather than a nuisance: type-mixture models and finite-mixture estimation are routine.
The open questions are substantive. How stable are social preference types across contexts and over time? How do social preferences interact with markets and institutions—do they get crowded out, or reinforced? How do they develop in childhood, and how much is cultural? And how should policy be designed when the people it targets care about fairness, reciprocity, and the expectations of others? The field's history is a record of experiments that made those questions unavoidable, and of models that made them tractable.
References
- Social preferences (introduction, Experimental evidences, The Ultimatum Game, The Dictator Game, The Trust and Gift-Exchange Games), Wikipedia
- Behavioral economics (History), Wikipedia
- Homo economicus, Wikipedia
- Ultimatum game (introduction, History), Wikipedia
- Public goods game, Wikipedia
- Strong reciprocity, Wikipedia
- Dictator game (Variants), Wikipedia
The frameworks of Social Preferences
Each framework is a way of working the field has used: when it held, and the question it answers.
- 1993-Present
Intention-Based Reciprocity
Do people reciprocate kind intentions or favorable outcomes?
- 1997-Present
- 1998-Present
Evolutionary and Cultural Group Selection Models
How could costly cooperation and punishment evolve?
- 1999-Present
Inequity Aversion
Do people dislike unequal outcomes for themselves and others?
- 2000-Present
Developmental Approach
When do children begin to share and punish fairly?
- 2001-Present
Cross-Cultural and Field Experimental Approach
Do fairness tastes vary across cultures and societies?
- 2006-Present
How the ideas connect
Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.
Intention-Based Reciprocity grew out of Experimental Economics of Social Preferences
The ultimatum and dictator experiments of the early 1980s made fairness behavior visible and posed the question of whether responders react to outcomes or to the proposer's intent. Matthew Rabin's 1993 model of intention-based reciprocity was built directly on that experimental puzzle, formalizing kindness and reciprocity in psychological games.
Carried by Matthew Rabin; Werner Güth; Rolf Schmittberger.
Strong Reciprocity grew out of Experimental Economics of Social Preferences
Fehr and Gächter's public goods experiments with costly punishment grew out of the experimental economics of social preferences established in the 1980s, using the same laboratory game methodology to show that people punish free riders at personal cost.
Carried by Ernst Fehr; Simon Gächter.
Inequity Aversion grew out of Experimental Economics of Social Preferences
Fehr and Schmidt's inequity aversion model was built to organize the ultimatum, dictator, and public goods results that experimental economics had produced since the early 1980s, offering a single outcome-based utility function to explain them.
Carried by Ernst Fehr; Klaus Schmidt.
Conditional Cooperation and Social Preference Types grew out of Experimental Economics of Social Preferences
The strategy-method public goods experiments that produced the taxonomy of social preference types were a direct extension of the experimental economics of social preferences, using the same contribution games to measure individual response functions.
Carried by Urs Fischbacher; Simon Gächter; Ernst Fehr.
Cross-Cultural and Field Experimental Approach grew out of Experimental Economics of Social Preferences
Henrich and colleagues' cross-cultural ultimatum and dictator experiments took the laboratory games developed in the 1980s and ran them in small-scale societies, directly extending the experimental economics of social preferences to field settings.
Carried by Joseph Henrich.
Inequity Aversion rival of Intention-Based Reciprocity
Fehr and Schmidt's inequity aversion model was offered as an outcome-based alternative to the intention-based reciprocity models of Rabin and others, and the two accounts competed to explain the same ultimatum and trust game data.
Carried by Ernst Fehr; Klaus Schmidt; Matthew Rabin.
Neuroeconomics of Social Preferences grew out of Experimental Economics of Social Preferences
Sanfey and colleagues' neuroimaging study of the ultimatum game took the standard laboratory paradigm from experimental economics and paired it with brain imaging, making the neural response to unfair offers the object of study.
Carried by Alan Sanfey.
Guilt Aversion grew out of Intention-Based Reciprocity
Charness and Dufwenberg's guilt aversion model extended the psychological-game framework of intention-based reciprocity, replacing kindness with beliefs about others' expectations and adding guilt as the belief-dependent motive.
Carried by Gary Charness; Martin Dufwenberg.
Evolutionary and Cultural Group Selection Models grew out of Strong Reciprocity
The puzzle of costly punishment in one-shot settings posed by strong reciprocity prompted evolutionary and cultural group selection models that sought to explain how such behavior could arise and persist.
Developmental Approach grew out of Strong Reciprocity
Developmental studies of sharing and costly punishment in children took the strong reciprocity question of whether costly punishment is a basic human tendency and asked when it emerges in ontogeny.
Cross-Cultural and Field Experimental Approach grew out of Strong Reciprocity
Cross-cultural field experiments extended the strong reciprocity program by testing whether costly punishment and fair offers appear across societies, linking variation to market integration and daily cooperation.
Carried by Joseph Henrich.
Conditional Cooperation and Social Preference Types rival of Inequity Aversion
The heterogeneous-types approach to conditional cooperation competed with inequity aversion as an account of public goods behavior, arguing that aggregate patterns reflect a mixture of free riders and conditional cooperators rather than a single aversion to inequality.
Carried by Urs Fischbacher; Simon Gächter; Ernst Fehr.
Guilt Aversion rival of Inequity Aversion
Guilt aversion was proposed as a belief-dependent alternative to outcome-based inequity aversion for explaining promise-keeping and cooperation, keeping the same experimental games but replacing distributional dislike with aversion to disappointing expectations.
Carried by Gary Charness; Martin Dufwenberg.
Beyond the field
Where Social Preferences's ideas came from in other fields, and what they led to there: the strongest 12 of 19 links.
Experimental Economics of Social Preferences led to Cross-cultural experimental economics in Behavioral Game Theory
Cross-cultural experimental economics grew directly out of the experimental economics of social preferences, extending ultimatum, dictator and public-goods games to small-scale societies. The 2001 cross-cultural project by Henrich and colleagues built on the game designs and preference models developed in the earlier laboratory literature.
Carried by Joseph Henrich; Robert Boyd; Colin Camerer.
Evolutionary and Cultural Group Selection Models grew out of Group selection in Social Evolution And Sexual Selection
These models grew out of the revival of group selection in the 1990s, extending it to cultural inheritance and to selection among human groups.
Carried by David Sloan Wilson; Elliott Sober.
Evolutionary and Cultural Group Selection Models grew out of Cultural Group Selection in Social Evolution And Sexual Selection
The evolutionary and cultural group-selection models of the late 1990s were a direct development of cultural group selection theory, formalising how conformist transmission and group-beneficial norms spread.
Carried by Robert Boyd; Peter Richerson.
Inequity Aversion grew out of Social preferences in Experimental Economics
Inequity aversion was proposed as a formal model of the social preferences documented in experiments, fitting distributional concerns into utility functions over relative payoffs.
Carried by Ernst Fehr; Klaus M. Schmidt.
Neuroeconomics of Social Preferences grew out of Neuroeconomics in Behavioral Economics
Neuroeconomics of social preferences grew directly out of neuroeconomics, applying its imaging and lesion methods to the trust, ultimatum and dictator games studied by economists.
Guilt Aversion grew out of Social Preferences in Behavioral Economics
Guilt aversion was formulated as a refinement within the behavioral-economics social-preference models, replacing or supplementing distributional altruism with a belief-dependent motive to avoid letting others down.
Experimental Economics of Social Preferences influenced Beliefs and Preferences over Redistribution in Political Economy
Experimental economics of social preferences supplied evidence that fairness concerns shape preferences over redistribution. Research on beliefs about redistribution drew on dictator and ultimatum game findings showing other-regarding preferences.
Experimental Economics of Social Preferences influenced by Experimental Economics Methodology in Behavioral Game Theory
The induced-value and public-goods experimental methods developed in the 1970s were the laboratory tools the social-preference experiments of the 1980s used to measure other-regarding behaviour.
Carried by Vernon L. Smith.
Experimental Economics of Social Preferences influenced by Behavioral Decision Experiments in Experimental Economics
Kahneman and Tversky's prospect-theory experiments and the heuristics-and-biases programme supplied the experimental method and the framing of deviations from self-interest that the social-preference experiments of the 1980s built on.
Carried by Daniel Kahneman; Amos Tversky; Prospect Theory (1979).
Strong Reciprocity influenced by Gift Exchange and Reciprocity in Personnel Economics
Gift-exchange and reciprocity experiments supplied the evidence of costly rewarding and punishment that strong-reciprocity theory generalised into a disposition to cooperate and punish free riders.
Carried by Ernst Fehr; Simon Gächter.
Strong Reciprocity influenced by Behavioral Game Theory in Game Theory
Behavioural game theory's experimental games, notably public-goods and punishment experiments, provided the data that strong-reciprocity theory was built to explain.
Carried by Ernst Fehr; Simon Gächter.
Strong Reciprocity influenced by Reciprocal Altruism in Social Evolution And Sexual Selection
Strong reciprocity was framed as a distinct, human-specific form of reciprocity contrasted with Trivers's reciprocal altruism, taking over its evolutionary question of how costly helping can persist.
Carried by Robert Trivers.

Social Preferences is a planet in Nousfarer
Its city has a tower for each of these 10 frameworks. Land, walk in, and climb one concept at a time.
Land on Social PreferencesRelated fields
Fields its ideas are linked to
- Behavioral Game Theory 4 links
- Social Evolution And Sexual Selection 4 links
- Experimental Economics 3 links
- Behavioral Economics 2 links
- Game Theory 1 link
- Neoclassical Economics 1 link
- Personnel Economics 1 link
- Ethics 1 link
- Philosophy Of Biology 1 link
- Political Economy 1 link
More in Behavioral Economics
- Behavioral Decision Theory
- Behavioral Economics
- Behavioral Game Theory
- Experimental Economics
- Neuroeconomics
See where Social Preferences sits on the Map of Human Knowledge