Democrats’ Real Liability in the House

A cartoon of a donkey behind a visual representation of the political-party makeup of the House of Representatives.
Getty / The Atlantic

Joe Biden marked an unexpected and unwanted milestone this month when he won a clear popular-vote majority in the presidential election but saw his party suffer substantial losses in the House of Representatives.

That unusual combination of results—the first time it’s happened in more than 120 years—crystallizes the core challenge Democrats face in translating their consistent victories in the popular vote into congressional power. In geographic terms, their coalition is deep but narrow. The party has consolidated its hold on the nation’s largest metropolitan centers, which allows it to amass substantial popular-vote victories, but it has systematically declined in the smaller places beyond them—a dynamic that’s intensified during Donald Trump’s polarizing presidency.

The distortions created by this geographic sorting have been most apparent in the Senate. There, the GOP’s dominance of less populated, heavily rural states has allowed it to control the upper chamber more than half of the time since 1980, even though Republicans have represented a majority of the nation’s population for only one two-year span during that period. Democratic senators are guaranteed again to represent a majority of the nation’s population next year, whether or not the party wins the two Georgia runoff elections in January, which would allow it to control the chamber.

Recommended Reading

While Democrats will still run the House, Republicans’ unanticipated gains there underscore how the growing concentration of the Democrats’ political support into a few large places threatens their position in that chamber as well. With three House seats still to be decided (one each in California, Iowa, and New York), Republicans have substantially narrowed the majority that the Democrats amassed in their 2018 sweep. So far, Democrats will control 222 House seats and Republicans 210; immediately before the election, the balance was 232 Democrats to 197 Republicans, with one libertarian and five vacancies.

These results closely follow the outcome of the presidential race. While Biden is on track to win the popular vote by well over 6 million, the best estimates are that he may carry only about 223 House districts. With very few districts backing a presidential candidate of one party and a House candidate of the other, that left Democrats with very little margin for error in their search for 218 seats.

It’s unusual for a president’s or president-elect’s party to lose House seats while he wins, but it’s hardly unprecedented. Based on official House statistics, Woodrow Wilson, John F. Kennedy, Bill Clinton, and Grover Cleveland (twice) won elections while losing ground in the House. But each of those Democratic presidents won with less than half of the popular vote, well below the 51 percent Biden has captured at last count. Republicans Ulysses S. Grant and William Howard Taft won a majority of the presidential popular vote but lost a handful of House seats (two and four, respectively). The most recent president to win a majority of the popular vote and lose a substantial number of House seats was Republican William McKinley in 1896. (His 48-seat loss came after a landslide two years earlier in which the GOP won nearly three-fourths of the House.)

The juxtaposition between Biden’s substantial popular-vote win and the GOP’s substantial House gains captures the geographic sorting that is reshaping American politics. Growing advantages in the biggest places are the key reason Democrats have now won the popular vote in seven of the past eight presidential elections, something no party had done since the formation of the modern party system in 1828. Based on the latest data, Biden won fully 91 of the nation’s 100 largest counties. That’s more than the 87 Hillary Clinton carried in 2016, and far beyond the 69 that Bill Clinton took in 1992.

But even as Democrats have improved their position inside the nation’s largest and most economically vibrant metropolitan areas, their support in exurban, small-town, and rural regions has collapsed. While Bill Clinton twice won about 1,500 counties (roughly half the counties in America), Hillary Clinton carried just less than 500 (roughly one-sixth). Though Biden won the popular vote by at least 3 million more votes than she did, he only slightly expanded her geographic reach: So far, he’s carried 509 counties, based on the latest count from the Brookings Institution’s Metropolitan Policy Program.

When measured in House districts, the performance of Democratic presidential candidates has similarly narrowed. According to data collected in Brookings’s Vital Statistics on Congress, Bill Clinton carried a clear majority of House districts in his two victories, while winning only a plurality of the popular vote in both three-way races. But Al Gore lost most House districts while winning the popular vote in 2000. And both Barack Obama in 2012 and Hillary Clinton in 2016 lost most House districts while winning the popular vote (2008 was an exception to this pattern: Obama won 242 House districts while winning almost 53 percent of the popular vote).

In one sense, Biden’s success in winning more districts (about 223) than Obama in 2012 (209) and Hillary Clinton in 2016 (205) constitutes clear progress for Democrats, reflecting the expansion of the party’s support in white-collar suburbs. But it still highlights the constraints on the Democrats’ reach: Trump won more House seats (230) in 2016 while losing the popular vote, and George W. Bush in 2004 won many more (256) while winning slightly less of the national vote than Biden did this year.

Those disparities explain why many analysts in both parties believe Democrats face a natural disadvantage in the House, even before factoring in gerrymandering.

“If you apportion the House in a fair drawing, it favors Republicans, because Democrats live in these urban enclaves that are 80 percent [Democratic] and they waste a lot of votes,” Tom Davis, a former Republican representative from Northern Virginia who chaired the National Republican Congressional Committee, told me.

The consequences of this imbalance are growing more significant because, as in Senate races, it’s getting harder for either party to win House seats in areas that vote the other way at the presidential level—especially in a presidential-election year.

Through the late 20th century, it was common for a large number of districts to support House candidates from one party and presidential nominees from the other. About 190 districts split their votes during landslides for Richard Nixon and Ronald Reagan, largely because many conservative southerners still voted Democratic for the House even as they backed those GOP presidential candidates.

But as more voters have treated congressional elections as choices between competing parties rather than competing individuals, the number of split districts has dwindled, reaching a modern low of 26 in the 2012 election and rebounding only slightly to 35 in 2016.

This year could set a new record for the fewest split-ticket House seats. Depending on the final vote tallies, it’s possible that each party will win only about 10 seats that voted for the other side’s presidential candidate. Democrats have reelected 10 members in seats that voted for Trump (though Biden may win some of those districts as the final votes are counted, removing them from the split-ticket category). Republicans in turn reelected three GOP incumbents in seats that Biden carried, beat four Democratic incumbents in Biden districts, and won one open seat that he took.

Even so, the bulk of the Democrats’ House losses came in districts that Trump carried in both 2016 and 2020; the party has lost eight such seats already, with two still undecided. Democrats expected to lose some of those districts with Trump himself on the ballot. But they were surprised—and disappointed—by two trends.

One was the victories by Republican House candidates in several urban and suburban districts that Biden carried, including seats around Miami; Omaha, Nebraska; Dallas; Philadelphia; and Orange County, California. The clear implication of those results is that some college-educated, suburban voters who rejected Trump supported Republicans for the House, perhaps because they did not want to give Democrats a free hand to advance their agenda. (The share of college graduates exceeds the national average in almost all of the Biden districts that elected Republicans to the House.) “We have voters who didn’t want to vote for Trump but wanted to be able to support the kind of Republican House candidates who they traditionally supported,” Molly Reynolds, a senior fellow in governance studies at the Brookings Institution, told me.

The second disappointment for Democrats is that Biden did not win more Republican-held suburban seats that voted for Trump last time. Biden did win three Republican-held seats that Democrats captured this election (two redrawn districts in North Carolina and a suburban Atlanta seat); he also flipped about 10 seats that Trump won in 2016, and that House Democrats took in their 2018 sweep. But Biden fell short in many of the party’s new targets for 2020, including suburban Republican-held districts around St. Louis; Cincinnati; Indianapolis; Phoenix, Arizona; Austin, Texas; and Houston. Those were also places where Democrats expected to offset any losses in the Trump districts they won in 2018. But against the headwind of Trump’s continuing strength in these new targets, Democrats could not capture any of them in the House contests. “Trump was sneaky strong—not enough to win, but he was not the albatross that we expected him to be” for down-ballot Republicans, the GOP communications consultant Liam Donovan told me.

In one sense, those seats were always tough targets for Democrats: They are almost all much more Republican in their core political DNA than the Trump districts the party won in 2018 House races. But the inability of either Biden or Democratic House candidates to capture them reinforces the party’s worry that Democrats can’t hold very much more than 218 House seats on a lasting basis. The most traditionally Republican-leaning suburbs still resist them (particularly in Midwest states, where those suburbs remain predominantly white). Meanwhile, the GOP’s red curtain over rural America looks almost impenetrable at this point.

Adding to the Democratic challenge, Republicans this month maintained control of legislatures in states such as Texas, Florida, North Carolina, and Georgia, which will give the GOP the upper hand in the redrawing of congressional-district maps next year, following the 2020 census. The Democratic position in the competition for the House is still stronger than it was during the redistricting process 10 years ago, though. The party’s advances in white-collar suburbs, particularly around Sun Belt cities such as Atlanta, Houston, Dallas, and Phoenix, means it is now competing on a wider House battlefield than it was then. But in states where Republicans control redistricting—and they will control the redistricting of more congressional seats than Democrats—the GOP may be able to draw Republican-leaning seats that submerge those newly blue suburban areas into immense tracts of red rural terrain.

For Democrats, the surest way to defend their House majority may be to rebuild their capacity to compete in at least a few more small-town and rural districts. That proved impossible with Trump polarizing the electorate so sharply along cultural lines. The future of the House Democratic majority may depend on whether Biden succeeds in his uphill quest to lower the temperature of partisan conflict and narrow the nation’s gaping political divides.