Synergy in Spatial Models of Voting: How Critical Cases Show that Proximity, Direction and Discounting are Friends, not Foes
By Laura Sudulich on 2 August 2017
Winner of the JEPOP Best Paper Prize 2015
By Till Weber Baruch College, City University of New York
Do voters support a party because it shares their views, because it is serious about its beliefs, or because it is getting things done? This question summarizes the decades-long controversy between proponents of “proximity,” “directional” and “discounting” theories of voting. Although each theory has its distinctive vision of democratic representation, political scientists have struggled to find an empirical solution. By letting the dust settle on this episode of spatial modeling, I argue, the discipline is losing the race on the home stretch.
The main reason for the empirical impasse lies with the high collinearity of model predictions. The three models happen to make identical predictions in most real-world situations. My analysis published in JEPOP suggests that we can only reap the fruit of the “directional debate” by getting to the root of the problem: Why is it that three distinct theories lead to almost identical predictions? My answer is that collinearity is a nuisance only for scholars but not for voters. Voters in fact love collinearity because it concentrates all spatial virtues in a single party. I find that each of the three criteria for vote choice becomes increasingly valuable as the other two are fulfilled at the same time. Parties then have incentives to defend policy platforms as “collinearity packages.” Supply of and demand for such packages is what electoral competition is all about.
These conclusions derive from an innovative analytical design that isolates the “critical” part of each voter’s preference order. I decompose data from 28 multi-party systems covered in the European Election Study 2009 into about one million party triplets. Contrasting those triplets where model predictions diverge with the bulk of collinear triplets reveals disproportionately strong policy voting when all model predictions agree. In other words, the relationship between the three spatial logics is one of synergy, not one of conflict. Scholarly arbitration has been so difficult because the theories we believed to be foes have been friends all along.