Explaining party ideological stances

We apply the iterative algorithm by Merrill and Adams (Political Analysis 9:347–361, 2001) to compute the optimal positions of vote-maximizing parties along a one-dimensional space for a large number of elections. This allows us to systematically evaluate factors that push parties toward more centrifugal locations in equilibrium. Our findings largely corroborate expectations suggested by the theoretical literature, albeit with one notable exception: the relative size of a partisan constituency reduces, rather than enhances, the likelihood of a party having an extreme optimal position. We then compare derived equilibrium positions with actual party positions, with the aim of understanding the nature of electoral incentives and offering insights about party-system competition. In particular, we show that the age of a party, the party family to which it belongs, post-election expectations, and internal rules for candidate and leader selection, are all factors that increase the probability of a party approaching the position predicted by pure vote-maximizing considerations.