Electoral Politics as a Redistributive Game

Spatial models of electoral competition typically simplify the analysis by ignoring the question of internal constituency politics: constituencies are modeled simply as a distribution of ideal points along a set of issue dimensions. Matters related to the stability of divergent electoral coalitions have rarely been addressed. We explicitly take into account how differential rates of support by various groups in a constituency will influence candidates' campaign promises and the likelihood that stable electoral coalitions will be forged. Viewing campaign platforms as promised redistributions of welfare, we argue that the optimal strategy for risk-averse candidates will be to promise redistributions first and foremost to their reelection constituency and thereby to maintain existing political coalitions. We use evidence from the urban services literature to support our propositions.