Arrow's impossibility theorem

Arrow's impossibility theorem. In social choice theory, Arrow's impossibility theorem is a landmark result demonstrating the fundamental difficulty of designing a fair voting system. Formally proven by economist Kenneth Arrow in his 1951 book Social Choice and Individual Values, the theorem shows that no rank-order voting method can simultaneously satisfy a set of seemingly reasonable criteria when there are three or more alternatives. This finding has profoundly influenced the fields of political science, economic theory, and philosophy, challenging notions of collective rationality and democratic decision-making.

● Statement of

the theorem The theorem states that for an election with at least three distinct candidates, it is impossible for any preference aggregation rule to convert the ranked preferences of individuals into a complete and transitive social welfare ordering while simultaneously meeting all of a set of specified conditions. These conditions are designed to encapsulate principles of fairness, non-dictatorship, and responsiveness to individual preferences. The impossibility holds regardless of the specific voting rule, such as plurality voting, instant-runoff voting, or the Borda count, that might be employed. The formal proof relies on the logical inconsistency of the required axioms within the framework of ordinal utility.

● Conditions and definitions

Arrow's theorem is framed around several precise conditions. First, the condition of **Unrestricted Domain (U)** requires that the voting system must produce a social ordering for every possible configuration of individual preference orderings. Second, the **Pareto Efficiency (P)** condition, also called unanimity, mandates that if every voter prefers alternative X over Y, then the social ordering must also rank X above Y. Third, the condition of **Independence of Irrelevant Alternatives (IIA)** stipulates that the social preference between X and Y should depend only on the individual preferences between X and Y, not on preferences for a third alternative Z. Finally, the **Non-dictatorship (D)** condition asserts that there is no single voter whose preferences automatically determine the social ordering, regardless of the preferences of others.

● Proofs and intuition

Arrow's original proof, presented in Social Choice and Individual Values, used a method of constructing pivotal voters to demonstrate the logical conflict between the conditions. A common intuitive understanding is that in a field of three or more options, any system satisfying Unrestricted Domain, Pareto Efficiency, and IIA will inevitably concentrate decisive power in a single individual, thereby violating Non-dictatorship. Subsequent proofs have employed different approaches, including the use of the **Field Expansion Lemma** and the **Contraction Lemma**, which analyze how decisive coalitions behave. The work of scholars like Amartya Sen and John Geanakoplos has provided simplified and more accessible proofs, reinforcing the theorem's robustness across various logical frameworks.

● Implications and interpretations

The theorem has had far-reaching implications, suggesting a fundamental trade-off in democratic design between fairness, consistency, and responsiveness. In political philosophy, it has fueled debates about the very possibility of a "common good" derived from individual values, influencing thinkers like John Rawls. In practical politics, it implies that all real-world voting systems, from elections for the United States House of Representatives to procedures in the United Nations Security Council, must necessarily violate at least one of Arrow's conditions, often IIA. This has led to the analysis of specific systems, such as the susceptibility of plurality voting to Duverger's law and the potential for strategic voting in methods like the single transferable vote.

● Variants and related theorems

Several important results extend or relate to Arrow's foundational work. The **Gibbard–Satterthwaite theorem**, independently proven by Allan Gibbard and Mark Satterthwaite, demonstrates the susceptibility of ranked voting systems to tactical voting. The **May's theorem** characterizes majority rule for two alternatives. Research into relaxing Arrow's conditions has explored possibilities, such as weakening IIA, as in the approach of Duncan Black and his analysis of single-peaked preferences. Other related concepts include the **Condorcet paradox**, which illustrates cyclical social preferences, and the study of **approval voting**, which operates outside the strict ranked-preference framework of Arrow's theorem. Investigations into judgment aggregation and the work of the Nobel laureate Elinor Ostrom on collective action also engage with its central challenges.

Category:Social choice theory Category:Mathematical economics Category:Theorems in discrete mathematics Category:Voting theory