Partisan conflict over funding the Department of Homeland Security this winter was just the latest in a series of battles on Capitol Hill that illuminate record-high levels of polarization. The conventional wisdom is that the widening ideological divide in Congress is intractable, and that “there is no common ground anymore.”

However, my new book – “Is Bipartisanship Dead? Policy Agreement and Agenda-Setting in the House of Representatives” – challenges this view. A key reason voting in Congress is so polarized in recent years is simply that the majority party allows only a few bills come to the floor for a vote, and these are usually the measures that divide the two parties sharply. But once you look beyond these bills, you can see a surprising stability in bipartisanship over time in the House.

The increasing partisanship in how members of Congress vote is clear. In 2013, 70 percent of all recorded votes in the House pitted a majority of Democrats against a majority of Republicans. Only 30 percent of votes had bipartisan support. In contrast, in 1973, less than 40 percent of votes split the parties, while more than 60 percent of votes garnered bipartisan support.

But partisanship in voting is only part of the story, because members vote only on bills that are priorities of the majority party, and the majority leadership mostly determines whether bills get voted on at all. In other words, roll call votes essentially capture how much majority party leaders pursue bipartisanship, but do not capture the underlying potential for bipartisan agreement among all members.

An alternative way to gauge patterns of bipartisanship is to look at co-sponsors of all measures. Members decide whether to co-sponsor a bill before the majority party decides whether a bill will come to a vote, and members form these co-sponsorship coalitions on many more bills than actually come to a vote. Co-sponsorship patterns on legislation can thus illuminate how much underlying bipartisanship may actually exist.

Analyzing these co-sponsorship networks shows that bipartisan agreement — defined as at least 20 percent of the co-sponsors being from the opposite party as the bill’s original sponsor — persisted to a surprising degree from 1973 to 2006. Bipartisanship in bill co-sponsorship coalitions declined by less than 20 percent over this period:

By contrast, bipartisanship in actual voting coalitions declined more. This is because party leaders in the House do not have to pursue bipartisan bills, and they often don’t. Of all co-sponsored bills that received roll call votes, those with bipartisan coalitions declined from 84 percent in the 93rd Congress (1973-1994) to 51 percent in the 104th Congress (1995-1996). Leaders’ pursuit of bipartisan legislation has been similarly low in the 111th and 112th Congresses (2009-2012).

Partisan agendas have at least one other important consequence: They make lawmakers less responsive to constituent interests – even though co-sponsorship patterns actually reveal increasing responsiveness over this period.

Why wouldn’t members who have endorsed bipartisan bills object when those bills never get a vote? One reason is that strong partisanship poses less of an electoral risk. Most members represent districts whose political complexion lines up with their party’s positions. Few members actually face a significant tension between what their constituents want and what their party demands. In these cases, the consequences of being out-of-step are typically too small to cost a member his or her seat.

Moreover, a partisan agenda can help members by building their party’s brand and emphasizing the differences between the two parties. As a result, there is little that would constrain the pursuit of partisan legislative agendas.

The primary analyses in my book end in the mid-2000s. Since then, and especially since the election of Barack Obama in 2008, the potential for bipartisanship has declined somewhat, as measured by co-sponsorship patterns. But leaders’ pursuit of partisan legislation on the agenda still magnifies the apparent partisan divide in Congress.

Any effort to reduce this divide thus confronts a big challenge. Democrats and Republicans do not oppose one another on votes merely because they have different preferences. Their party leaders also structure the legislative agenda to highlight partisan conflict and play down bipartisan agreement. This strategy makes sense when both parties want to emphasize their differences and individual lawmakers worry more about primary rather than general election threats.

In short, House members and their leaders have few incentives to facilitate bipartisan agreement — even if bipartisanship is not dead.

Laurel Harbridge is an assistant professor in the Department of Political Science and a faculty fellow at the Institute for Policy Research at Northwestern University.