Arrow's theorem

Arrow's theorem
Name	Arrow's theorem
Born	1951
Known for	Social choice theory

Arrow's theorem is a foundational result in social choice theory that demonstrates constraints on aggregating individual preferences into a collective decision. Formulated by Kenneth Arrow and presented in his work connected to The General Theory of Employment, Interest and Money debates and later summarized in Social Choice and Individual Values, the theorem shows that no voting system can convert ranked preferences of individuals into a community-wide ranking while simultaneously meeting a set of reasonable conditions. It has influenced debates in United Nations, European Union, United States Supreme Court, World Bank, and academic contexts such as Harvard University, Princeton University, Massachusetts Institute of Technology, London School of Economics, and Stockholm School of Economics.

Statement of the theorem

Arrow proved that when there are at least three alternatives, no social welfare function can satisfy a combination of plausible conditions: unrestricted domain, Pareto efficiency, independence of irrelevant alternatives, and non-dictatorship. The result connects to earlier work by Marquis de Condorcet, the development of voting theory in 18th century France, and the mathematical foundations advanced at institutions like Cambridge University, University of Chicago, and Birkbeck, University of London. Arrow’s framing influenced economists and philosophers at Columbia University, Yale University, University of Oxford, University of Cambridge, and research programs associated with Nobel Memorial Prize in Economic Sciences, John Bates Clark Medal, and policy studies at RAND Corporation.

Assumptions and axioms

The theorem uses axioms: unrestricted domain (all individual orderings allowed), Pareto efficiency (if everyone prefers A to B then society prefers A to B), independence of irrelevant alternatives (IIA), and non-dictatorship (no single individual always determines outcomes). These axioms were debated by scholars at Princeton, Stanford University, University of Michigan, and in correspondence with figures connected to Cowles Commission and Mont Pelerin Society. Discussions referenced traditions reaching back to Jean-Jacques Rousseau, John Stuart Mill, Alexis de Tocqueville, and formal methods from David Hilbert and Émile Borel that shaped modern axiomatization.

Proof outline and variations

Arrow’s original proof used lattice-theoretic and functional methods influenced by work associated with John von Neumann, Oskar Morgenstern, and mathematical logic from Alfred Tarski. Subsequent proofs and simplifications were developed by scholars at University of California, Berkeley, University of Michigan, Cornell University, Brown University, and in journals linked to American Economic Association and Econometrica. Variations include quantitative formulations by researchers connected to Bell Labs, algebraic treatments by mathematicians working with Institute for Advanced Study, and combinatorial approaches favored at Princeton. Related formal results include the Gibbard–Satterthwaite theorem (strategic manipulation), the work of Condorcet and Borda, and impossibility results discussed by theorists at Carnegie Mellon University, University of Pennsylvania, Duke University, and University of California, Los Angeles.

Examples and implications

Classic examples illustrating the theorem involve cyclical majorities noted by Condorcet and scenarios like three-candidate elections similar to contests in United States presidential election, party systems such as those in United Kingdom general election and coalition bargaining in Germany, and ranking problems encountered by advisory bodies like Council of the European Union and committees at World Health Organization. Implications extend to institutional design debates at International Monetary Fund, European Central Bank, Bank of England, and policy-making bodies in Japan, Brazil, India, and Canada. Arrow’s conclusions influenced mechanism design in programs at Institute of Economic Research, electoral reform campaigns including proposals in New Zealand, and constitutional theory discussed in contexts like Federalist Papers and debates around Separation of powers.

Extensions and generalizations

Researchers extended Arrow’s framework to probabilistic social choice explored at Bellagio Conference and algorithmic social choice developed at Microsoft Research, Google Research, and labs at Facebook AI Research. Generalizations include weakening IIA, considering single-peaked preferences traced to models by Anthony Downs and analyses used in Median voter theorem, and stochastic approaches related to work at Santa Fe Institute. Connections were drawn to cooperative game theory from Lloyd Shapley and Robert Aumann, to matching theory as in Gale–Shapley algorithm, and to fair division studied at Carnegie Mellon University and Massachusetts Institute of Technology.

Criticisms and interpretations

Critics and interpreters include philosophers and economists from Stanford University, Yale University, Princeton University, University of Chicago, and London School of Economics, who argue about the realism of axioms, normative import, and applicability in institutional contexts such as United Nations Security Council voting and corporate governance at firms like Goldman Sachs and JP Morgan Chase. Alternative normative frameworks draw on work by Amartya Sen, John Rawls, Kenneth Arrow himself in later writings, and debates framed within traditions linked to Welfare economics, Public choice theory, Behavioral economics, and empirical electoral analysis from Pew Research Center and International Institute for Democracy and Electoral Assistance.

Category:Social choice theory