Dwindling productivity in Congress linked to vanishing cooperation

"We can't say for sure that the decline in cooperation is the sole reason that there are fewer bills being introduced or passed by Congress, but we do know the two are statistically correlated, and both have been dropping steadily over the past 60 years," said Clio Andris, lead author and assistant professor of geography at Penn State.

The researchers tracked roll call vote data -- the record of whether Congressional representatives abstained or voted "yay" or "nay" on motions or bills -- on every vote, beginning in 1949 -- the start of the 81st Congress -- and continuing through 2012 -- the end of the 112th Congress. The researchers then identified all times when a representative voted with members of the opposite party, referred to as a cross-party pair, or with members of the same party, referred to as a same-party pair.

The findings show that the number of cross-party votes decreased at an exponentially higher rate than same-party votes. From 1967 to 1979, there were more than 10,000 instances of representatives voting across party lines. From 2001 to 2010, there were fewer than 1,500 of these cooperative votes.

The findings also show that far fewer representatives today engage in cross-party voting than in the past. "If you look at past data, it was uncharacteristic that one representative would be involved in even 1 percent of the total cross-party voting, because people were more likely to vote against party lines. Today, several people account for upwards of 50 percent of the total cross-party votes, and these are the people we refer to as 'super-cooperators' because they account for a bulk of the cross-aisle voting," said Andris.

In the 112th Congress, 7 of the 444 members accounted for 98.3 percent of cross-party pairs, which indicates that the majority of representatives vote with their party members only. This study was the first to analyze congressional role call voting using a network model, and this approach allowed the researchers to identify patterns at the individual level.

"Our original goal was to try to identify 'hidden partnerships' among Congressional representatives and to understand whether geography played a role in any of those bonds. Surprisingly, we found that geography had very little influence over how representatives voted. You might think that issues around land resource management or tourism might be affected by local geographies, but that's not the case," said Andris. "What we're seeing now is that, more and more, party platforms are the determining factor for how the majority of representatives are voting."