As the 2014 midterm election campaigns headed into their last weeks, a number of journalists and pundits suddenly realized that for all that was at stake (control of the U.S. Senate) and for the billions of dollars spent, it was really “a campaign about nothing,” as MSNBC’s Perry Bacon Jr. put it. Others have taken to calling it the “Seinfeld election.” Almost all the campaigns have been about gaffes and personality, such as whether Senate candidate Bruce Braley (D-Iowa) threatened to sue a neighbor over runaway chickens, rather than issues or ideas.

But why? Is it just the superficiality of life in the Kardashian era? Is it that politicians have run out of ideas? No – as with so much else in modern politics, it’s deeply and directly related to the fact that the two political parties are now completely divided, with little ideological overlap either in Congress or among voters. When politicians are polarized and voters are polarized, issues, too, become either Democratic or Republican issues. There’s little payoff to being a Republican who favors health insurance or a Democrat who supports, say, Paul Ryan’s plan for changes to Medicare. It doesn’t get you any new voters, and it risks losing the voters you already have.

When politicians talk about issues or ideas, they’re really talking about what kind of person they are. The purpose of ideas is not to indicate what they are going to do, but usually how they’re different. Bill Clinton, for example, promised to “end welfare as we know it” — not because it was the issue he most cared about, but because it was a way of telling the voters then known as “Reagan Democrats” that he was different from other Democrats. George W. Bush talked about improving education and his record in Texas to show that he was not the sort of heartless, budget-slashing Republican of the Gingrich era. For John McCain, his support of campaign-finance reform marked him as an independent, free from the big-money influences on other Republicans.

Not anymore. Now there are just two parties, lining up their votes, their money and their (non-overlapping) agendas. Putting a new issue on the table is risky and unlikely to pay off.

We’ve known for a while that swing voters are disappearing. Increasingly, they are an exotic species seen only in television focus groups. Whatever swings we do observe between early polls and the election results are almost certainly better explained by changes in mobilization, turnout and enthusiasm than actual undecided voters wrestling with two competing choices. This is especially true in midterm elections, when the actually “independent” voters don’t pay attention at all. Most “independents” are actually partisans; true independents are almost all unpredictable low-information voters, who turn out in very low numbers in midterms.

Historically, parties and candidates have sought political advantages by trying to raise the salience of particular issues that might cut across party lines. The assumption was that voters might support Democrats on some issues, while they might agree with Republicans on other issues. So, for example, a pro-choice Republican might be able to win over enough Democratic voters to emerge victorious in a Democratic district, or the reverse. (In Pennsylvania, for example, it was long observed that a pro-choice Republican such as the late senator Arlen Specter or a pro-life Democrat, such as current Sen. Robert Casey, would have a better chance than a candidate whose position on the issue aligned with his or her party.)

As party politics has become at once more homogenous and polarized, it’s not only harder to reach a compromise at the end – such as on immigration reform or the federal budget. There’s also less room for ideas and fresh approaches to issues at the beginning. The “political issue space,” once wide open and full of opportunities to form new coalitions, is now narrow and closed.

Political scientists Thomas Carsey and Geoffrey Layman describe the process as “conflict extension.” Forty years ago, a wide range of economic, social and cultural issues stood outside of the conflict between the parties. In some cases, there was broad agreement and, in other cases, regional differences or local interests were more significant than party. In some other cases, such as health reform, the parties had different views about a solution but agreed that there was a problem to be solved, creating an incentive to introduce a fresh idea that might give each party some of what it wanted.

But over time, party leaders picked sides on these issues, and often even on whether a problem existed; voters’ loyalties followed. If you were a Democrat, you couldn’t oppose abortion; if you were a Republican, you couldn’t approve of government-supported health care. Similarly, Ashley E. Jochim and Bryan D. Jones have shown that on a wide range of issues, voting patterns in Congress have collapsed to a single dimension. There are few cross-cutting issues left.

What’s a campaign to do? To introduce a new cross-cutting issue in today’s polarized environment runs a serious risk with an unlikely payoff. The loyal base of your party knows what it thinks on all the major issues, and any violation of the party platform risks alienating donors and key supporters. The voters of the other party will be skeptical, because they have the opposing partisan worldview, which increasingly sees the other party as entirely untrustworthy. And whatever possible “middle” exists is actually not a middle at all, but rather a mix of low-information voters whose positions just haven’t fallen in line with the party platforms and tend to be all over the place. They are not moderates. Even the primaries, where one might expect a marketplace of ideas, have become issue-free contests over partisan bona fides.

In other words, it makes perfect sense for the candidates to not raise any new issues, or any issues at all. Just keep to the old turn-out-the-base chestnuts, raise the stakes of the other side winning to existential threat level, and do your best to demoralize your opponent.

Party polarization is not going away, and there’s every sign that the more fluid and overlapping party politics of the second half of the 20th century was the exception, not the norm. But there may be ways to restore some of the incentives to explore new political territory or coalitions: Although few voters are true independents, the two sitting senators and several candidates in Senate and gubernatorial races in 2014 who ally themselves with neither party might be able to maneuver around the rigid lines and put some fresh ideas on the table. Reforms that make it easier for third-party candidates or independents to run – such as fusion, which allows a party to either endorse candidates from another party or run its own, open primaries, or instant runoff voting – might create more room from candidates with new ideas. Campaign finance reforms could be structured not to limit money but to make it easier for candidates to be heard without first appealing to established donors.

Michael Grunwald of Time recently argued that the lack of ideas is explained by the fact that the two parties have achieved their biggest goals – low tax rates for conservatives and universal health insurance for liberals – and, therefore, there’s not much left to fight about. It’s true that both sides have won important victories. But in a healthy marketplace of ideas, there would be incentives for two competing parties to raise new issues. Our society and our world do not lack for potential problems to solve. Yet, in our system, the two sides continue to fight the same fights, over and over again.

Lee Drutman is a senior fellow in Political Reform Program at New America. Mark Schmitt is director of the Political Reform Program at New America and a former editor of the American Prospect.