Bounded Rationality

Bounded rationality is the idea that people make decisions with limited information, limited time and limited mental capacity, so they aim for choices that are good enough rather than mathematically best. The term was introduced by the American economist and cognitive scientist Herbert Simon in the 1950s as a challenge to the "economic man" of standard theory, who was assumed to know every option, weigh every outcome and always pick the one that maximizes his utility.

The concept sits at the root of behavioral economics, organization theory and modern decision science. It is often contrasted with full rationality in neoclassical economics and with rational expectations in macroeconomics. It is not the same as irrationality: people still try to do well, but with shortcuts. Simon received the Nobel Memorial Prize in Economic Sciences in 1978 for his research on decision-making inside organizations.

Think about choosing a restaurant in a city you have never visited. A perfectly rational diner would read every menu, check every review and compare every price before sitting down. Nobody does that. Most people walk until they find a place that looks clean, has a few free tables and serves food they like, and then they stop looking.

That is bounded rationality. The search ends once something meets a personal threshold, because searching longer costs time and effort. The choice may not be the best restaurant in town, but it is a sensible one given what the diner knew and how hungry they were.

Simon set out the idea in papers such as "A Behavioral Model of Rational Choice" (1955) and gathered it in the book Models of Man (1957). He argued that the classical model ignored two things: the limits of the human mind (memory, attention, computing power) and the structure of the environment in which choices are made. He compared rational behavior to a pair of scissors whose two blades are the mind and the environment; a theory that looks at only one blade cannot explain the cut.

From this he drew the notion of satisficing, a blend of "satisfy" and "suffice". Instead of maximizing, a decision-maker sets an aspiration level and accepts the first option that clears it. If options are plentiful the aspiration rises; if they are scarce it falls. Simon applied this to firms, arguing that managers pursue targets for sales, market share or profit rather than a single profit maximum, an idea that fed into the behavioral theory of the firm developed by Richard Cyert and James March.

Researchers describe several typical patterns that follow from limited attention and information:

• Heuristics. Rules of thumb, such as "buy the brand I bought last time", replace full calculation. Daniel Kahneman and Amos Tversky documented how such shortcuts produce systematic errors, while Gerd Gigerenzer argued that simple heuristics can be "ecologically rational" and sometimes beat complex models.
• Defaults and inertia. When a choice is complicated, many people keep whatever option is preset, which is why automatic enrollment raises pension saving.
• Narrow framing. Decisions are judged one at a time rather than as part of a whole portfolio, which links to the framing effect and loss aversion.
• Rules and routines in organizations. Firms rely on standard procedures, budgets and checklists because no manager can re-optimize every decision every day.

Information costs matter as much as mental limits. Under asymmetric information, one side of a deal knows more than the other, and gathering the missing facts takes time and money, so stopping early can be a reasonable choice rather than a mistake.

Bounded rationality helps explain why retail savers often under-diversify, chase recent performance or hold too much cash, and why complex products sold to households can produce poor outcomes. It also underlies the design of "choice architecture": regulators and employers simplify forms, set sensible defaults and require plain-language disclosure. In the United Kingdom, workplace pension auto-enrolment, phased in from October 2012, uses inertia to raise participation; in the United States, the Pension Protection Act of 2006 made automatic enrollment easier for 401(k) plans.

In macroeconomics, models with "rational inattention" (associated with Christopher Sims) or adaptive learning assume households and firms update slowly, which can help explain sticky prices and gradual responses to monetary policy. In market analysis, bounded rationality is one of the arguments raised against a strict reading of the efficient market hypothesis, and it is used to explain episodes of herding such as an economic bubble. Related ideas, including Knightian uncertainty, stress that some risks cannot be measured at all, which makes rules of thumb unavoidable.

Game theorists also use the concept. Classical game theory assumes every player can compute a Nash equilibrium, but laboratory experiments show that players often reason only one or two steps ahead, and "level-k" and "quantal response" models were built to capture that.

Supporters of standard models argue that bounded rationality is hard to turn into precise predictions: if people satisfice, the theory must still say where the aspiration level comes from, and almost any behavior can be explained after the fact. Some economists also argue that competition, arbitrage and learning push markets toward outcomes that look rational even when individuals are not, so the idea matters less in aggregate than in single decisions.

From the other direction, some psychologists say the behavioral literature focuses too much on "biases" and too little on how well simple rules work in real environments. Policy critics add that using bounded rationality to justify nudges can be paternalistic, because officials designing defaults are boundedly rational too, a point also made in public choice theory. Despite these debates, the idea that decision-makers face real limits on information and attention is now part of mainstream economics.

• Nobel Prize: Herbert A. Simon, facts
• Stanford Encyclopedia of Philosophy: Bounded Rationality