Polarization – divergent policy views – and partisanship – one's loyalty to party – are often used synonymously. However, the two are quite different, and the distinction is important, according to the Wilson School's Nolan McCarty, Susan Dod Brown Professor of Politics and Public Affairs, and Georgetown University Professor Boris Shor.
Using new data from the Measuring American Legislatures project, McCarty and Shor show that while Congress continues to diverge, many state legislatures are actually far more polarized. Their analysis also examines what causes party systems to polarize and what happens to governance when they do.
McCarty discusses his analysis in a Q&A below.
Q. You note the difference between polarization and partisanship. Can you explain the difference, and why you opted to focus on polarization?
McCarty: While polarization and partisanship are often closely related, it is important to distinguish the two concepts. Polarization generally refers to a situation where more and more people hold non-centrist or extreme policy views. The term also is often used to describe a situation where the policy views of members of each party become increasingly divergent. Partisanship, on the other hand, is more general in that it tends to refer to the levels of intra-party cooperation and inter-party conflict. Thus, polarization of policy views could cause an increase in partisanship, but so could many other factors.
I have focused on legislative polarization because I maintain that ideological divergence across the parties is a central component in the increase in legislative partisanship.
Q. Why did you and Professor Shor look at polarization through a state-by-state lens?
McCarty: Much of my previous work on polarization has focused on the national level. But there are limits on what one can learn about the causes and consequences of polarization from looking at just the national trends. Looking at the states provides much more variation with which we can test ideas about what causes party systems to polarize and what happens to governance when they do.
Q. Did you expect your results? What was most surprising, if anything?
McCarty: Prior to our work on the states, many scholars thought polarization was a national level phenomenon caused by regional differences such as those that exist between the so-called “red states” and “blue states.” Our research has shown, however, that most states are polarized internally. This is true of red states such as Texas and blue states such as California. In fact, a majority of state legislatures are more polarized than the U.S. Congress. The polarization measure we used is the average difference in the views of Republican and Democratic legislators. This difference in averages does not depend on whether the legislature is split 50-50 or dominated by one of the parties.
Q. You found that California (a large state) was the most polarized, and Rhode Island (a small state) was the least polarized. Does state size have something to do with these findings?
McCarty: Logically, one might think that there would be a strong correlation between polarization and the size of the state. Generally, larger states are more diverse socially, economically, and racially, and those factors clearly contribute the levels of polarization. But the correlation is not quite as strong as one would expect. Some of the most polarized states are small – like Colorado, New Mexico and Montana. There are some large states where the parties are not that polarized, like New Jersey and Illinois. And remember, the U.S. Congress is less polarized than many state legislatures despite the fact that it represents the entire country.
Q. What policy implications does this research have?
McCarty: It is commonly believed that polarization in Congress and state legislatures is caused by certain deficiencies in our electoral system and laws. Reformers blame the gerrymandering of legislative districts, the rules governing primary elections or features of our private campaign finance system. With data from the states, we are better able to evaluate such claims. States not only vary in terms of their levels of polarization, but they also have different electoral laws in places. Thus, we can assess which reforms may make a difference and which ones won’t.