Economics Rules

Dani Rodrik argues that economics is best understood not as a unified science with a single framework but as a diverse collection of models, each suited to particular circumstances. The discipline's greatest strength lies in this multiplicity, and its most damaging failures occur when practitioners treat one model as universally applicable.

Rodrik opens with three examples of economics at its best. At the 1944 Bretton Woods Conference, economists John Maynard Keynes and Harry Dexter White designed the postwar international economic order through fixed exchange rates and limited capital flows. In 1952, economist William Vickrey proposed congestion pricing (charging higher fees during peak demand to reduce crowding) for the New York City subway, an idea Singapore successfully implemented beginning in 1975. In 1997, economist Santiago Levy designed Mexico's Progresa program, which provided direct cash grants to poor mothers on the condition that their children attend school and receive health care, revolutionizing antipoverty policy worldwide. From these successes, Rodrik draws his central claim: Models, the abstract and typically mathematical frameworks economists use, are both the discipline's core contribution and its vulnerability.

Rodrik defines models as simplifications that isolate specific causal mechanisms. He presents three contrasting examples to show how different assumptions lead to different conclusions: The perfectly competitive market model produces efficiency; the prisoners' dilemma shows how self-interest traps competing firms in a wasteful outcome; and a coordination model demonstrates how interdependent actors get stuck when neither invests. The correct answer to almost any question in economics, Rodrik argues, is that it depends on which assumptions hold.

To explain why simplicity is a feature rather than a flaw, Rodrik draws two analogies. Models resemble fables: Both are spare and yield clear morals, yet because many fables offer contradictory lessons, judgment is needed to pick the right one. Models also resemble laboratory experiments: Both isolate cause and effect, though models do so by manipulating assumptions rather than physical environments. Both face the challenge of external validity, the question of whether conclusions from a controlled setting apply in the real world. Rodrik engages economist Milton Friedman's 1953 argument that the realism of assumptions is irrelevant so long as predictions are correct. Drawing on Stanford economist Paul Pfleiderer, he counters that critical assumptions, those whose modification would change a model's conclusions, must approximate reality.

Rodrik argues that models make economics a science, though not in the manner of physics. Models clarify hypotheses, reveal counterintuitive possibilities, enable knowledge accumulation, and provide shared professional standards. He illustrates with two landmark results. The First Fundamental Theorem of Welfare Economics, proven by economists Kenneth Arrow and Gerard Debreu in the early 1950s, demonstrates that a competitive market economy is efficient only under precise conditions, including no monopolies and no externalities (costs or benefits imposed on parties outside a transaction). Economist David Ricardo's 1817 Principle of Comparative Advantage showed that even a country less efficient at producing everything benefits from trade by exporting what it produces relatively less badly.

Rodrik explores counterintuitive outcomes that models reveal. General-equilibrium interactions—feedback effects that ripple across multiple interconnected markets—explain why the 1980 Mariel boatlift, in which Cuban immigrants increased Miami's labor force by seven percent, had virtually no effect on local wages. The General Theory of Second Best, formulated by economists James Meade, Richard Lipsey, and Kelvin Lancaster, shows that freeing up some markets when others remain restricted can backfire. The concept of time-inconsistent preferences captures the paradox that reducing one's freedom of action can be beneficial, as when a government delegates monetary policy to an independent central bank to resist short-term temptation to inflate.

Economic science, Rodrik argues, advances horizontally by adding new models to its library rather than vertically by replacing older ones. He traces this expansion from perfectly competitive models to imperfect competition, to asymmetric information models (which examine situations where one party to a transaction knows more than the other) by economists Michael Spence, Joseph Stiglitz, and George Akerlof (joint Nobel Prize, 2001), to behavioral economics, which integrates findings about loss aversion (the tendency to value potential losses more heavily than equivalent gains), overconfidence, and bounded rationality (decision-making constrained by limited information and cognitive capacity). Each generation expands the range of insights without rendering earlier models obsolete.

Rodrik then addresses the profession's most neglected skill: selecting the right model for a given situation. He describes the growth diagnostics framework he and colleagues developed for advising developing countries. Rather than recommending a laundry list of reforms, they built decision trees asking whether investment constraints were on the supply side or demand side, with empirical tests at each node. In South Africa, the framework dismissed conventional explanations such as skill shortages and instead identified the high cost of unskilled labor. In El Salvador, the diagnosis pointed to coordination failures: Pineapple canneries could not operate without frequent air cargo service, which itself was not profitable without many existing exporters. Rodrik generalizes four verification strategies for model selection: checking whether critical assumptions fit the setting, whether causal mechanisms operate, whether predictions match outcomes, and whether incidental implications hold.

Turning to grand theories, Rodrik argues that frameworks such as the neoclassical theory of value (the framework explaining how prices and wages are determined through supply, demand, and marginal productivity) and Keynesian macroeconomics function as scaffolding rather than standalone explanations. The marginalist revolution of the late 19th century showed that wages reflect labor's marginal productivity, but the theory cannot account for the growing gap between US labor productivity and compensation since 2000. After the 2008 financial crisis, new classical models could not explain the recession's depth, while Keynesian models had struggled with 1970s stagflation (simultaneous stagnation and inflation). Rodrik uses the rise in US income inequality since the mid-1970s, during which the Gini coefficient (a standard measure of income inequality) rose from 0.40 to 0.48, to illustrate how competing explanations, including globalization, skill-biased technological change, and weakened unions, each illuminate certain channels without providing a complete account.

Rodrik examines how economists go wrong when they mistake a model for the model. He identifies errors of omission, as in the 2008 financial crisis: Economists had models for every relevant element, including bubbles, bank runs, and principal-agent problems (situations where managers' incentives diverge from shareholders' interests), but placed excessive faith in the efficient-markets hypothesis. Formulated by economist Eugene Fama, this hypothesis held that market prices reflect all available information and that intervention is more likely to distort than correct. He also identifies errors of commission, as in the Washington Consensus of the late 1980s and 1990s, when international institutions pushed developing countries to stabilize, privatize, and liberalize without accounting for weak institutional foundations. In Latin America and Africa, trade liberalization destroyed protected manufacturing without generating new export activities, while Asian countries pursuing second-best strategies achieved far better outcomes. The push for free capital movement, championed by officials like Stanley Fischer at the International Monetary Fund (IMF), produced financial crises across developing nations.

Rodrik attributes these failures partly to the sociology of the profession: Economists with strong convictions dominate public debate while cautious colleagues remain silent, and a proprietary attitude toward markets leads economists to downplay outside concerns. He invokes philosopher Isaiah Berlin's distinction between hedgehogs, captivated by a single big idea, and foxes, who hold many views, arguing that economics needs more foxes.

In his final chapter, Rodrik addresses external criticisms, arguing that most lose force once economics is understood as a diverse collection of models. He engages with Harvard philosopher Michael Sandel's argument that markets breed corrosive values, responding that the only value economics can properly claim for markets is efficiency, a consideration worth weighing but not the sole relevant one. He acknowledges that economists transgress their expertise when making categorical policy recommendations without weighing justice and distribution, and he points to three areas of genuine transformation: behavioral economics, stimulated by psychologist Daniel Kahneman's Nobel Prize-winning work; randomized controlled trials borrowed from medicine; and institutional economics influenced by comparative political history.

Rodrik concludes with 20 commandments, 10 for economists and 10 for noneconomists. For economists: Economics is a collection of models whose diversity should be cherished, every model is a model rather than the model, the world is almost always second best, and substituting personal values for the public's is an abuse of expertise. For noneconomists: Economics has no predetermined conclusions, models should be challenged by asking how results change under more realistic assumptions, and when an economist recommends a policy, the right question is what makes the underlying model applicable.