Why individual rankings can clash
Imagine three voters choosing among A, B, and C. Their rankings can combine so that a majority prefers A to B, another majority prefers B to C, and yet another prefers C to A. This is a majority cycle: the group’s choices loop even though each voter’s own ranking is orderly. So a voting rule cannot always produce a collective ranking that simply mirrors every pairwise majority.
What Arrow’s theorem rules out
Economist Kenneth Arrow formalized this problem for ranked voting systems. His theorem asks whether a system can turn people’s individual rankings into one group ranking while meeting several appealing conditions at once. These include allowing voters’ rankings to be considered in any combination, respecting unanimous agreement between two options, and avoiding a rule that lets one voter dictate the outcome. Another condition requires the group’s ordering of two options to depend only on how voters rank those two—not on the presence or ranking of other choices.
For three or more options, no system can satisfy all those conditions in every possible case. The result is not that voting is useless or that every election produces a cycle. It means a rule that always returns a ranking must give up at least one of these fairness demands, or use a different way to define what counts as fair.
The tradeoff is built into the rules
Different systems make different choices. A method may focus on pairwise contests, while another uses points or transfers to reward broader support. Each can behave well in many elections, but Arrow’s result says none is a perfect all-purpose solution under the theorem’s assumptions. The practical question is therefore not whether a voting system is flawless, but which tradeoffs its rules make—and whether those tradeoffs fit the decision being made.
