Resumen de Spatial Competition Between Two Candidates of Different Quality

Enriqueta Aragonés, Thomas Palfrey

• This paper examines competition in the standard one-dimensional Downsian model of two-candidate elections, but where one candidate (A) enjoys an advantage over the other candidate (D). Voters� preferences are Euclidean, but any voter will vote for candidate A over candidate D unless D is closer to her ideal point by some fixed distance d. The location of the median voter�s ideal point is uncertain, and its distribution is commonly known by both candidates.

  The candidates simultaneously choose locations to maximize the probability of victory. Pure strategy equilibria often fails to exist in this model, except under special conditions about d and the distribution of the median ideal point.

  We solve for the essentially unique symmetric mixed equilibrium with no-gaps, show that candidate A adopts more moderate policies than candidate D, and obtain some comparative statics results about the probability of victory and the expected distance between the two candidates� policies. We find that both players� equilibrium strategies converge to the expected median voter as A�s advantage shrinks to zero.

  We test the predictions of this model using laboratory experiments, and find strong support: the better candidate adopts more centrist policies than the worse candidate; the equilibrium is statistical, in the sense that it predicts a probability distribution of outcomes rather than a single degenerate outcome;

  and the equilibrium varies systematically with the level of uncertainty about the location of the median voter. We also show that there is a �centrist bias�, in the sense that both candidates adopt centrist positions significantly more than predicted by the theory. This bias is explained by Quantal Response Equilibrium theory.