Rise of the super-cooperators

Tuesday, March 24, 2015 - 1:12pm

By Kathy Cappelli

A few years ago, The Department of Geography’s newest faculty member, Clio Andris, wondered if location was related to bipartisan cooperation in the U.S. House of Representatives. “I was really interested in seeing if there were regional ties that existed, if we could find a coalition of some sort between Democrat and Republican representatives from certain states or regions. For instance, if they shared a common boundary or physical feature, like a river, maybe they would go across party lines to agree on certain things,” Andris explained.


What she found was a resounding “No.” Regardless of shared boundaries, there was very little across-the-aisle cooperation. Andris and the team she was working with found—just like other political scientists had—that partisanship is at an all-time high. And then, Andris noticed another phenomenon: the emergence of representatives that she termed “super-cooperators.”



Read the research article on Plos One: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0123507

Additional news coverage:



These few congressmen are the swing votes of the House, working with both Republicans and Democrats depending on the issue at hand. In a House where representatives are governed mainly by their party’s platform rather than their own ideology, these few super-cooperators are depended on for bipartisan cooperation. A single super-cooperator can account for up to 48 percent of all cooperation in the House—an unprecedented amount. Additionally, they tend to be Democrats from the Midwest and South, or Republicans from wealthy suburbs in the Northeast—New York, Maryland, and New Jersey. “These are suburbs that tend to have progressive views on social issues but are fiscally conservative,” Andris explained.


To perform this research, Andris and her colleagues at MIT and the Santa Fe Institute paired up all Representatives from the 81st Congress in 1949 to the 112th Congress in 2012. Each Congressman was paired with every other, and instances of cooperation were based on how many times the two agreed in the roll call votes for that session of Congress. For example, if both voted ”yay” on a bill, it would be counted as an agreement but if one voted ”yay” and the other voted ”nay,” then they did not demonstrate cooperation. To count bipartisan agreement, each pair was classified as a same-party or cross-party pair, and the instances of their agreement were recorded. The team found that in the early years, cross-party pairs frequently agreed. Over time, the number of cooperating cross-party pairs has decreased until the burden of bipartisan cooperation rests with a small number of Congressmen.

cooperators in 1949

cooperators in 2011

Images show cooperators in 1949 versus 2011. Republican representatives are in red

and Democrat representatives are in blue.  Image by Mauro Martino.


“This decrease in cooperation can be attributed to a number of historical and current events,” said Andris, citing among other things the realignment of Southern Democrats, change in Congressional procedures, and changes in media. “We have our own hypotheses, however,” she said. “I think that perhaps another part of the divergence is that representatives, whether by circumstance or the norm, used to ‘hang out’ more. They all lived in D.C. and would go out after work. Now we see more and more that they are going home on weekends—or sleeping in their offices. In addition, there are now stricter laws governing lobbyist-Congressman interaction; in the past where lobbyists might have been able to take representatives out to dinner and bring together different parties to talk, they are now restricted. As a result, “they are all plugged in elsewhere, which also means ‘directly to the party' via email, etc."


Despite the identification of certain representatives as super-cooperators, Andris said that their motivations are still not well understood. “We don’t really know day to day why they’re different,” she said. “We just know that they’re the only ones agreeing on both sides." These representatives can go to so many people on both sides of the aisle, she said, though they didn’t use to have to; there were just normal people cooperating on both sides.


Andris emphasized that the research was just a statistical method to find out who was bridging party lines and to find out what constituencies were supporting this. In general, partisanship is studied by experts in political science, mostly qualitatively. “Researchers on the state and party don’t provide conclusive evidence on who will cooperate, though. This is not a last step, it’s a first step.”


So why do we keep electing representatives who are not going to cooperate? That’s a good question, Andris laughed. “If you can frame it the right way, it makes for a beautiful mathematical paradox. It seems that Congress thinks like they are in a massive game of tug of war; if you add 3 people to one side it’ll fix everything, but in this case it makes it worse." People see nothing getting done, so they seem to be electing more and more extreme thinkers hoping to pull Congress in their direction, and when that doesn’t work, they elect more radical representatives. “They’re afraid of electing moderates to dilute their tug of war team,” Andris said. “There’s a perception that it’s not their problem, it’s the country’s problem. But we have to start somewhere." According to Andris, it’s not the number of bills passed that has an effect on cooperation, but rather the number of bills brought to the table. Additionally, tumultuous events tend to increase cooperation, and in the meantime Andris said that America’s getting better. “We still have inequality, but we’re getting better. Nutrition, race relations, access to schools, car safety; all these things are improving, and at some point going in the direction we’re going, we may run out of innovative legislation . . . but not just yet.”