The Washington PostDemocracy Dies in Darkness

Our politics may be polarized. But that’s nothing new.

David W. Brady is the Bowen H. and Janice Arthur McCoy Professor of Political Science and Leadership Values at Stanford University and the Deputy Director of the Hoover Institution. Hahrie Han is an Associate Professor of Political Science at Wellesley College and the author of two forthcoming books: How Organizations Develop Activists: Civic Associations and Leadership in the 21st Century and Groundbreakers: How Obama’s 2.2 Million Volunteers Transformed Field Campaigns in America.

A historical perspective on polarization helps us better understand both its causes and consequences. We agree with the excellent points Nolan McCarty and Frances Lee have already made in their posts—that Congress is as polarized now as it ever has been, and that we are in a historically unusual period of sustained partisan competition. At the same time, looking at the data another way shows that the present period of polarization is not necessarily a historical anomaly.

Let’s say we counted the number of Democrats who were more conservative than the 10th percent, 25th percent, and 50th percent most liberal Republican in the House and Senate, and the number of Republicans who were more liberal than the 10th percent, 25th percent, and 50th percent most conservative Democrat. (Here, we’ll use DW-NOMINATE scores, which political scientists commonly use to measure ideology, although almost any other measure will work.) Those conservative Democrats and liberal Republicans are the ones who lie in what we call the “overlap” region—the region where members from both parties overlap with each other. In periods of higher ideological polarization, there will be fewer—or no—members in the overlap region, and more members in that region during periods of lower polarization.

If we simply count the number of legislators in the overlap region in each year, what becomes evident is that the present period of polarization is akin to the polarization we had for much of the nineteenth and early twentieth century. The graph below, taken from a previously published paper (ungated, gated, and discussed further here), makes this point by tracking the percentage of overlapping Democrats and Republicans over time.

The graph only shows the House from 1867-2003, but the story would not change if we extended it to the present Congress or looked at the Senate (details about the methods are in our paper). Basically, there are no legislators in the overlap region until after World War II. In the post-WWII period, the number of legislators from each party in the overlap region spiked upwards and persisted until the early 1970s, when the numbers began to decline. (The pattern in the Senate is remarkably similar with one exception: overlapping voting in the Senate begins in the post-World War I era at a low level, when farm coalition senators seeking support for agriculture voted across party lines).

Obviously, part of this pattern can be explained by the fact that there were a number of very conservative Southern Democrats in Congress at the time. Yet, the historical pattern persists even if we account for that. By the 1980s, however, we are back to a flat line. There are no Democrats or Republicans in the overlap region.

These data are another way of making the same point other scholars make about the high levels of polarization we have today—but it highlights more clearly the fact that today’s polarization is a return to the historic norm of polarization. From the Civil War to World War II, there was no meaningful overlap in Congress. Taking this view, we realize that it is the immediate post-WWII era that is really unusual.

What are the implications for analysis of contemporary polarization? Looking at this data shows that congressional polarization cannot alone be the cause of our present gridlock (though, as Sarah Binder points out, they are certainly related). Ideological polarization was common during a long period of American history in which our nation transitioned from an agricultural to an industrial economy and then to the modern welfare system. Even when polarized voting was the norm, government was able to achieve progress on significant economic and social issues.

So, simple claims about polarization per se being problematic or polarization itself being the explanation for Congress’s inability to solve problems oversimplify the historical reality. Although many contemporary political observers decry polarization and yearn for an era of bipartisanship, the opposite was true in the immediate post-war era. During that time, which was the most bipartisan era of our nation’s history, political observers wrung their hands over the inability of parties to present clear alternatives to voters. In fact, the American Political Science Association convened a task force of top political scientists in 1950 to put together a report assessing the party system of that era. Their report decried the bipartisan parties of 1950, and, in fact, called for more polarization, arguing that more ideologically polarized parties would be more “responsible,” and present clearer alternatives to voters. The grass is always greener.

Further Reading

This is the latest post in our ongoing series on political polarization.  The previous posts include:

What we do know and don’t know about our polarized politics.

American politics is more competitive than ever.  That’s making partisanship worse.

Polarization we can live with.  Partisan warfare is the problem.

How political polarization creates stalemate and undermines lawmaking.

Electing more women to Congress isn’t a solution for polarization.

How U.S. state legislatures are polarized and getting more polarized (in 2 graphs).

How ideological activists constructed our polarized politics.