Median voter theorem

The median voter theorem states that in a voting system with single-peaked preferences arranged along a single dimension, the outcome selected by majority rule will be the preference of the median voter — the voter whose ideal point is exactly in the middle of the distribution. The theorem is due to Duncan Black (1948) and was later generalized and extended by Anthony Downs in his 1957 book An Economic Theory of Democracy.

The theorem is significant because it provides an escape from Arrow's impossibility: when preferences are single-peaked, majority rule produces a consistent social ordering that satisfies all of Arrow's conditions except universal domain. The restriction to single-peaked preferences is not arbitrary. Many political conflicts — taxation levels, defense spending, environmental regulation — can be mapped onto a single left-right dimension where each voter has a most-preferred outcome and preferences decline monotonically as policy moves away from that ideal point.

The theorem's predictions are testable and often confirmed: politicians in two-party systems tend to converge toward the center, platforms in proportional systems reflect the distribution of median voters across districts, and policy outcomes in direct democracy track median preferences. But the theorem also reveals the fragility of democratic aggregation: introduce a second dimension, allow non-single-peaked preferences, or relax the assumption that all voters participate, and the coherent center dissolves into cyclical majorities and strategic manipulation.