Micromotives and Macrobehavior

Thomas C. Schelling, a Nobel Prize-winning economist, examines a deceptively simple question: How do the choices of individuals, each pursuing their own interests and responding to their immediate surroundings, combine to produce large-scale social patterns that no one intended or even foresaw? Originally published in 1978 as part of the Fels Lectures on Public Policy Analysis at the University of Pennsylvania, the book spans seven analytical chapters plus a Nobel Prize acceptance lecture added to the 2006 edition.

Schelling opens with an anecdote about arriving to give a lecture and finding 800 audience members packed into the rear rows, leaving the first dozen rows empty. He proposes competing hypotheses: perhaps everyone prefers the rear, or each person simply avoids the very front occupied row, or early arrivals sit toward the back and latecomers cluster nearby. The point is not to solve the puzzle but to illustrate that aggregate patterns do not straightforwardly reveal individual motives. This gap between individual behavior and collective outcomes is the book's central concern. Schelling distinguishes simple cases, where behavior scales directly (all drivers turn on headlights at sundown), from complex ones, where people respond to an environment composed of other people responding in kind, creating feedback loops that yield surprising results. He introduces the concept of "vicarious problem solving" (18), anticipating what someone will do by analyzing the problem as that person sees it, while cautioning that people can pursue misguided goals. Evaluation, he argues, must focus on aggregate outcomes: how well each person adapts to a social environment is not the same as how satisfactory a social environment people collectively create.

Economics, Schelling contends, is the social science most aligned with this analysis. Markets achieve remarkable coordination because economic transactions are mainly voluntary, involve knowledgeable participants, and affect primarily the parties involved. But Schelling warns against generalizing this result to all social behavior, using the exchange of Christmas cards as a counterexample: people feel trapped by accumulated obligations, yet there is no mechanism to produce a satisfactory collective outcome. He clarifies that equilibrium analysis is descriptive, not evaluative; an equilibrium is simply the state that remains after adjustments have settled, and there is nothing inherently good about one.

Chapter 2 introduces propositions that hold in the aggregate regardless of how individuals behave. Within any closed telephone system, the total calls placed must exactly equal the total calls received. Schelling extends this to Christmas cards and bicycle theft, establishing that many social propositions are logically true of the aggregate but not of any individual. These are not empirical discoveries but logical consequences of how transactions are structured, yet failing to recognize them leads to confused reasoning. Schelling applies these structural constraints to demography: in a monogamous population, the difference between unmarried women and unmarried men is determined by the overall sex difference, which reflects differential life expectancies and marriage ages. He introduces the acceleration principle, showing that modest changes in a desired stock require disproportionate changes in the flow. If housing is to grow by an extra 2.5 percent per year, the construction industry must nearly double and then contract sharply once the target is reached. He also notes that if people refuse to be in the bottom 10 percent of a group, the group unravels as successive departures redefine who occupies that position.

Chapter 3 surveys recurring families of models. The home thermostat illustrates cyclical processes: time lags between the furnace heating water, radiators heating air, and the thermostat switching off cause the temperature to overshoot and undershoot. Schelling extends this to measles vaccination in poor countries, where success breeds complacency, which allows epidemics, which renew vaccination efforts. Critical mass describes activities that become self-sustaining only once participation passes a minimum threshold: dying seminars, pedestrians crossing against traffic, applause that either builds or sputters. He emphasizes that different individuals have different thresholds and that an activity's collapse does not reveal how close it was to viability. Related models include the "lemons" phenomenon, named by economist George Akerlof after the used-car market, where asymmetric information drives the best products off the market; tipping, where the entry of a minority into a neighborhood prompts majority departure, opening space for more minority entry; and the commons, drawn from ecologist Garrett Hardin's 1968 article in Science. In the commons model, individually rational behavior produces collectively inferior outcomes because costs are shared while benefits are private. Schelling also examines self-fulfilling prophecies, self-enforcing conventions such as daylight saving and the typewriter keyboard, and the social contract problem, where collective restraint is essential but individual restraint is negligible in effect.

Chapters 4 and 5 apply this framework to sorting and mixing. The self-forming neighborhood model, using pennies and dimes on a checkerboard, produces the book's most famous demonstration. Each coin requires some minimum fraction of like-colored neighbors. Starting from a nearly integrated pattern where only 9 of 45 coins are dissatisfied, the chain reaction of moves produces strikingly segregated outcomes, with the ratio of like to unlike neighbors reaching two to three times the minimum anyone demanded. Even a preference for congregation rather than avoidance of the other group produces visually indistinguishable segregation. This is a central finding: Different micro-motives can generate identical macro-patterns, making it impossible to infer individual preferences from observed aggregate segregation. The bounded-neighborhood model examines a single area where people stay or leave based on the overall color ratio. With illustrative distributions, the only stable outcomes are entirely one group or the other, even though compatible mixtures exist. Crucially, the same results apply whether people prefer segregation or prefer integration but have limits on minority status. For continuous variables like age, Schelling shows that departures change a group's average, triggering further departures that can leave only the oldest members. He derives conditions for stable divisions and demonstrates that equilibrium divisions are typically not optimal.

Chapter 6 explores the hypothetical possibility of parents choosing their children's chromosomes, selecting among the more than 60 trillion genetically distinct children a couple could naturally conceive. Sex is the first and most culturally legitimate choice. Schelling calculates that if every family chose a boy first and left subsequent children to chance, 70 percent of births would be male. For other characteristics such as body size, IQ, or handedness, he identifies a central danger: competitive choices could produce inflationary spirals as each family hedges against others' expected choices, and minor characteristics could become stigmatized through cascading avoidance.

Chapter 7 develops formal theory for binary choices with externalities, situations where each person makes an either-or decision and the payoff depends on how many others choose each alternative. Using the hockey helmet problem as the opening example, Schelling constructs a framework in which two payoff curves are plotted against the number choosing one alternative. The multi-person prisoner's dilemma arises when everyone has the same unconditionally preferred choice, yet all would be better off if enough people chose the opposite. Beyond this, intersecting curves produce stable but inefficient equilibria, dual equilibria create situations where the superior outcome requires coordinated action, and curves sloping in opposite directions reproduce the commons structure.

The final chapter, Schelling's 2005 Nobel Prize acceptance lecture, treats the 60-year tradition of nuclear non-use as a case study in the evolution of international behavioral norms. He calls this "THE MOST SPECTACULAR event of the past half-century... one that did not occur" (247) and credits much of it to a taboo that Secretary of State John Foster Dulles perceived as early as 1953. Schelling traces the history from Dulles's campaign to treat nuclear weapons "as available for use as other munitions" (250) through the dramatic reversal under John F. Kennedy and Lyndon B. Johnson, when Johnson declared in 1964 that nuclear use would be "a political decision of the highest order" (251). He identifies two arguments for the taboo: an intuitive one holding that nuclear weapons are axiomatically different and an analytical one based on bright lines and slippery slopes, analogous to "one little drink" for a person recovering from an alcohol addiction (252). The Soviet Union's heavy investment in conventional forces and its abstention from nuclear use in Afghanistan demonstrated that the taboo crossed cultural boundaries. Schelling concludes that the tradition of non-use is a treasure to be cherished and that the Comprehensive Test Ban Treaty would powerfully reinforce it.