The seemingly imminent government shutdown has brought out a familiar argument from both Democrats and Republicans: “The last election proves we’re right.” Democrats think Obama’s victory in 2012 settled issues like health-care reform.  Jonathan Chait sums up this view when describing congressional Republicans’ proposed conditions for raising the debt ceiling:

The fact that a major party could even propose anything like this is a display of astonishing contempt for democratic norms. Republicans ran on this plan and lost by 5 million votes.

Unsurprisingly, Republicans see the election differently.  Only a week after the election, Paul Ryan dismissed the idea of that Obama got any mandate, noting that Republicans were returned to the House as well.  More recently, former Sen. Jim DeMint took an even stronger view.  As Joshua Green describes it:

DeMint, Cruz, and all those trying to defund Obamacare drew precisely the opposite lesson from the last election than just about everyone else did. “Republicans were told, ‘Don’t do anything. Don’t be the issue. Don’t stand for anything. Make it about Obama,’ ” DeMint says. “What happened in 2012 was that there was a void of any inspiration, any attempt to lead. It certainly wasn’t because the party was too conservative—it was because there was no conservative leadership at all!”

Both sides are wrong.  Elections don’t convey policy mandates because most voters don’t vote on policy.  Instead, they vote based on longstanding loyalties to one party or the other.  As political scientist Gabriel Lenz shows, rather than using policy to choose a candidate, voters more often choose the candidate first, and then mold their policy views to fit those of the candidate they support.  Party comes before policy.

But politicians and their allies never learn this lesson.  As Lynn Vavreck and I show in our new book on the 2012 election, The Gamble, commentators were as quick to over-interpret this election as they were the 2008 and 2004 elections.  (Remember George W. Bush’s “political capital”?)  In reality, Vavreck and I demonstrate that in 2012 voters tended to perceive Romney as ideologically closer to them than was Obama.  Moreover, with the exception of a few issues like same-sex marriage, public opinion under Obama has largely trended in a conservative direction.  Obama appeared to win in spite of ideology and policy, not because of it.

Consistent with the idea that the 2012 election didn’t convey a mandate, public opinion about the shutdown battle doesn’t suggest a clear-cut victory for either side.  Partisan loyalties dominate: the vast majority of Republicans say they’ll blame Obama and congressional Democrats for a shutdown, and the vast majority of Democrats say they’ll blame congressional Republicans.  Independents are only a bit more likely to say they’ll blame Republicans (41 percent) instead of Obama (32 percent) or both parties (23 percent).  And Americans seem ambivalent about the relevant policy debates: the majority doesn’t want a shutdown but doesn’t want to raise the debt ceiling without some conditions.  The majority disapproves of the Affordable Care Act, but does not want to cut off funding for it.

All of this complicates any prediction about which party will benefit politically from this battle.  At the moment, the GOP is forecast to gain seats in the House and Senate.  Democrats hope for a replay of 1998 — when in the wake of the government shutdown and the impeachment of President Bill Clinton — the party gained seats in the 1998 midterm.  Republicans would be foolish to dismiss that experience (as DeMint wants them to do): research (pdf) by political scientist Benjamin Highton suggests that views of then-Speaker Newt Gingrich, which were more negative than was approval of Clinton, cost the GOP votes.  But at the same time, there is no Republican leader in Congress that was as polarizing as Gingrich was.  On Election Day in 1998, 56 percent of voters had an unfavorable view of Gingrich, 34 percent had a favorable view, and 11 percent had no opinion.  In a March CNN poll, views of Boehner were more evenly split (34 percent favorable vs. 39 percent unfavorable) and a larger fraction (27 percent) had no opinion.  2014 may not shape up like 1998.

Politicians have a love-hate relationship with public opinion.  None of them will admit taking cues from the polls, but they’re quick to claim that “Americans” agree with them.  It’s more accurate to say that elections and public opinion polls send weak signals about policy.  In fact, one politician seemed to realize that after the election — calling it only a “a mandate to work together to do what’s in the best interest of our country.” Interestingly enough, that politician was John Boehner.