Political polarization in the United States of America
In the past few decades, the U.S. has experienced a dramatic increase in political polarization. That is, compared to the policy preferences of the median voter, the Democratic Party’s and Republican Party’s policy platforms have become increasingly liberal and conservative, respectively. This growing divide between both political camps, according to scholars and pundits, has detrimental effects on society in terms of tension between citizens, income inequality, economic welfare, and so on.
Various causes of contemporary U.S. political polarization have been discussed in the scientific community, some of which have found empirical support while others still demand further theoretical and empirical investigation. For example, research on the impact of policymakers and the media on political polarization has turned out inconclusive, which is often due to the difficulty in establishing a causal link between the actions of these groups and the polarization trend. What most scholars can agree upon, however, is that policymakers have on average more extreme policy tastes relative to the preferences of the general public.
Modeling political polarization in elections
A model of political polarization in elections should take into account, and ideally predict, the empirical observation that U.S. policymakers are more polarized than the citizens they represent. To this end, my coauthor, Prof. Tom Palfrey from Caltech, and I have developed a game theoretical model of endogenous candidate entry in plurality elections where this observation is an equilibrium, or steady state, phenomenon even if most citizens have moderate policy tastes close to the median voter (Großer and Palfrey 2009, 2014). Stated as an “antimedian voter theorem,” we show that in large elections only the most extreme citizens choose to run as candidates, irrespective of the distribution of voter ideal policies (i.e., their most preferred policies). And, in a recent publication (Großer and Palfrey 2019) we report the results of a laboratory experiment that support our model’s specific mechanisms that lead to political polarization.
The essential assumptions driving polarization are that political candidates cannot commit, if elected, to implementing a public policy other than their own ideal policy and that citizens are at most informed about the candidates’ political leanings Left or Right from the median voter’s ideal policy (albeit the probability distribution of voter ideal points is public information). The latter assumption is consistent with the empirical finding that citizens often know little about the exact preferences of candidates and oftentimes only their party affiliation.
The laboratory results support our comparative statics predictions on polarization. In particular, relative to independent candidates, political parties are even more polarized but yield more efficient elections since the majority party is more likely to win due to vote coordination of its supporters. Other predictions are in line with casual observation of historical trends in U.S. politics. For example, the snowballing costs of mounting a successful campaign for national office in the past decades (e.g. due to greater costs of television advertisement and the relaxation of contribution limitations) should increase polarization, which has indeed been observed. Finally, the number of senators and representatives in the U.S. Congress (100 and 435, respectively) has been constant since 1963, while at the same time the U.S. population has grown by about eighty percent since 1960. The model predicts that an increase in the electorate size increases polarization.
Dr. Jens Großer is an Associate Professor in the Departments of Political Science and Economics. His co-author, Thomas R. Palfrey is a Professor in the Division of the Humanities and Social Sciences at California Institute of Technology.
The featured image is from the Pew Research Center.