Homo economicus, or “economic man,” is a theoretical model of human behavior used in economics. It portrays people as perfectly rational agents who always act in their own self-interest, have access to all relevant information, and consistently make choices that maximize their personal satisfaction. No real person behaves this way, and economists have always known that. The concept exists as a simplifying assumption, a starting point for building economic models and predictions.
Where the Term Came From
The idea traces back to John Stuart Mill’s 1836 essay “On the Definition of Political Economy,” where he sketched out an artificial version of a human being who made decisions to gain the most wealth in the most convenient fashion, with the least effort and least self-denial. Mill gave birth to the concept but never named it. The phrase “economic man” was introduced by economist Francis A. Walker in the 1870s, after Mill’s death.
The Latin version, “homo economicus,” first appeared under the pen of French Catholic economist Claudio Jannet in 1878. For decades, the coinage was attributed to Italian economist Vilfredo Pareto in 1906, but earlier usage by his Italian colleague Maffeo Pantaleoni in 1889 has since been found. The exact first use may never be pinned down with certainty, but the concept solidified throughout the late 19th century as economics became more formalized and mathematical.
The Core Assumptions
Homo economicus rests on a handful of key assumptions about how people make decisions:
- Perfect rationality. The agent can evaluate every available option and rank them consistently. Preferences don’t contradict each other, and choices always follow logically from those preferences.
- Complete information. The agent knows everything relevant to the decision: all available options, their costs, their consequences, and the probabilities involved.
- Self-interest. Every decision aims to maximize the agent’s own well-being or “utility.” Altruism, fairness, spite, and loyalty don’t factor in.
- Optimization. Given all of the above, the agent always selects the single best option. Not a good-enough option, not a convenient one. The best one.
In economic models, this translates into utility maximization: the process of comparing the satisfaction you’d get from every possible choice and picking whichever option delivers the highest return given your resources. If you have a fixed income and a set of goods to choose from, homo economicus will allocate every dollar in the mathematically optimal way.
Why Economists Use It
The model isn’t meant to describe how you actually decide what to eat for lunch. It’s a deliberate abstraction, similar to how physicists sometimes model objects as frictionless spheres. By assuming people behave rationally, economists can build mathematical models that generate testable predictions about markets, prices, and trade. If the predictions hold up reasonably well, the simplification earns its keep. When the predictions fail, that failure points researchers toward what’s missing from the model.
Standard microeconomics textbooks still use homo economicus as a baseline. Students learn supply and demand, market equilibrium, and consumer choice theory all built on the assumption of rational, self-interested agents. The model remains the default starting point for much of mainstream economic theory, even though virtually every economist acknowledges its limitations.
How Real People Differ
The gap between homo economicus and actual human behavior is enormous, and decades of research have catalogued exactly how. Daniel Kahneman and Amos Tversky published work in the late 1970s and early 1980s that became the foundation of behavioral economics. Their landmark paper has been cited over 50,000 times and is considered one of the most significant challenges to the rational agent model.
Several of their findings directly contradict the assumptions of homo economicus:
- Loss aversion. Losing $100 feels roughly twice as painful as gaining $100 feels good. A rational agent would weigh them equally. This asymmetry drives people to make choices that no optimizing agent would make, like holding onto losing investments far too long.
- The framing effect. People make different choices depending on how options are described, even when the actual outcomes are identical. A rational agent would respond to outcomes, not descriptions. This is considered the most severe challenge to the model because it violates “invariance,” the basic assumption that your choice depends on what you get, not how someone words the offer.
- The endowment effect. People value things more simply because they own them. Kahneman described the example of a wine collector who would refuse to sell a bottle for $200 but wouldn’t pay even $100 to replace it. Homo economicus would assign the bottle one consistent value.
- The certainty effect. People treat the difference between a 90% chance and a 100% chance as far more significant than the difference between 30% and 40%, even though both gaps are ten percentage points. Certainty has a special psychological weight that pure probability calculations don’t capture.
Kahneman also identified the hindsight bias, our tendency to view past events as having been inevitable once we know what happened, and the law of small numbers, where people draw sweeping conclusions from tiny samples. None of these patterns fit a model of cool, consistent rationality.
Bounded Rationality and Satisficing
Even before Kahneman and Tversky, economist Herbert Simon argued in the 1950s that the whole framework was unrealistic. His alternative concept, bounded rationality, starts from a simple observation: real people don’t have complete information, unlimited time, or the computational power to solve complex optimization problems. We work with what we have.
Simon proposed that instead of optimizing, people “satisfice.” That means you consider options until you find one that meets a minimum threshold of acceptability, then you stop looking. You don’t compare every apartment in the city to find the mathematically best one. You set standards for price, location, and size, then take the first place that clears all three bars. Simon originally described this as a limitation, writing that humans “satisfice because they have not the wits to maximize.” But researchers have since found that satisficing strategies often perform remarkably well in complex, real-world problems, and they show up frequently in fields like machine learning and artificial intelligence.
What Happens in Game Theory
Game theory exposes some of the model’s sharpest failures. In the Prisoner’s Dilemma, two players each choose whether to cooperate or betray the other. A homo economicus agent, acting purely in self-interest, always betrays. If both players are rational and self-interested, they both betray, and both end up worse off than if they had cooperated. The theory predicts mutual betrayal as the inevitable outcome.
Experimental economics tells a different story. In dilemma games, researchers observe a significant level of cooperative behavior, contradicting the conventional predictions of game theory. People cooperate even when betrayal would maximize their individual payoff. In the Ultimatum Game, where one player proposes how to split a sum of money and the other can accept or reject, homo economicus would accept any offer above zero, since something is better than nothing. Real people regularly reject offers they perceive as unfair, willingly losing money to punish greed. Fairness, trust, and social norms drive real decisions in ways the rational agent model simply doesn’t account for.
Its Place in Economics Today
Homo economicus occupies an unusual position. It remains the default model in mainstream economics education and in much of formal economic theory. At the same time, behavioral economics has grown into a major field, and its insights are increasingly integrated into policy, finance, and organizational design. The model hasn’t been replaced so much as supplemented. Economists use it where it works well enough (large markets with many participants, where individual quirks average out) and turn to behavioral models where it doesn’t (individual decision-making, savings behavior, health choices).
There’s also evidence that studying the model changes how people behave. Research has found that training in mainstream economics correlates with less cooperative behavior in experimental games, making students act closer to what game theory predicts and what common sense considers selfish. Whether economics attracts self-interested people or teaches them to be that way remains debated, but the pattern is consistent across multiple studies. The model, in other words, isn’t just a description of behavior. It may partly shape it.