How Democrats Avoided a Red Wave

The traditional midterm-election dynamic—wherein the president’s party takes a major hit—appears to have failed to materialize.

A split-screen illustration of two ballots being submitted, one black and one white
Getty; The Atlantic

The coalition of voters who turned out to oppose Donald Trump in 2018 and 2020 largely reassembled yesterday, frustrating Republican expectations of a sweeping red wave.

Under the pressure of high inflation and widespread disenchantment with President Joe Biden’s job performance, that coalition of young voters, people of color, college-educated white voters, and women eroded at its edges. And because Democrats began the night with so little margin for error in Congress, that erosion—combined with high Republican turnout—seemed likely to allow the GOP to seize control of the House, and possibly the Senate as well.

But even if the GOP does squeeze out majorities in one or both chambers when the final votes are counted, its margins will be exceedingly narrow, with control of the Senate, once again, possibly turning on another Georgia runoff. Up and down the ballot, Democrats dominated among voters who believe that abortion should remain legal—despite predictions from Republicans and many media analysts that the issue had faded in importance. Democrats held House seats in states including Rhode Island, Virginia, Michigan, and Ohio that Republicans had confidently expected to capture. And with the exception of Georgia, which reelected Governor Brian Kemp, Democrats could win gubernatorial races in each of the five swing states that flipped from Trump to Biden in 2020—a development that would greatly ease Democratic fears of Trump allies trying to rig the vote (and potentially the presidency) in 2024.

The results largely followed the outline of what I’ve called a “double negative” election. On balance, voter dissatisfaction with Biden’s performance meant that Democrats faced more losses, but the continuing unease about the Republican Party lowered the ceiling on GOP gains well below what the party might have expected.

These relatively positive results for Democrats were so striking because the findings of the national exit poll conducted by Edison Research for a consortium of media organizations, like virtually all preelection polling, showed deeply pessimistic attitudes that typically spell doom for the sitting president’s party. More than three-fourths of voters, Edison found, described the economy as only “fair” or “poor.” Four-fifths of voters said inflation had caused them either severe or moderate hardship. Fifty-five percent of respondents said they disapproved of Biden’s job performance as president. His approval stood even lower in many of the key Senate battleground states: 43 percent in Nevada and Arizona, 42 percent in New Hampshire, just 41 percent in Georgia.

Exit polls suggested that unhappiness over the economy could doom the most embattled Democratic Senate incumbent, Catherine Cortez Masto in Nevada, though that race remains on a knife’s edge awaiting the counting of the last mail ballots. Across a wide array of other battleground states, Republicans carried significant majorities of voters who expressed negative views on the economy.

But Republicans did not win those economically pessimistic voters by quite as big a margin as midterm precedents had suggested. Usually, the party out of power has dominated voters with those views: Democrats, for instance, in 2018 won about 85 percent of those who described the economy as either not so good or poor. This year, Republicans slightly exceeded that result among those who called the economy “poor,” the most negative designation. But among those who gave the equivocal verdict of “not so good,” Republicans won only 62 percent, way down from the Democrats’ total four years ago.

The relationship between presidential-approval ratings and the midterm vote was similar. Biden’s national job-approval rating in the exit poll (44 percent positive, 55 percent negative) resembled Trump’s in 2018 (45–54). But, compared with Republicans in 2018, Democrats this year carried slightly more of the voters who disapproved of Biden, as well as slightly more of those who approved of him. Particularly noteworthy: Democrats won almost exactly half of voters who said they “somewhat disapproved” of Biden, whereas about two-thirds of voters who “somewhat disapproved” of both Trump in 2018 and Barack Obama in 2010 voted against their party in House races.

These effects were even more pronounced in several of the battleground states. In 2018, no Republican Senate candidate in a competitive race carried more than 8 percent of the voters who disapproved of Trump, the exit polls found. But Cortez Masto and Raphael Warnock in Georgia carried about 10 percent of them, while Senator Mark Kelly in Arizona and Lieutenant Governor John Fetterman in Pennsylvania reached about 15 percent of support with Biden disapprovers, the exit polls found. In New Hampshire, the exit poll found Senator Maggie Hassan winning a striking one-fifth of voters who disapproved of Biden. Similarly, Warnock won about one-third of voters who described the economy as only fair or poor, while Kelly and Fetterman approached 40 percent with them in the exit polls. All of this may sound like a small difference—but it proved to be the margin between defeat and victory for Democrats in Pennsylvania and New Hampshire, and potentially in Arizona and Georgia.

How did Democrats overperform recent historical trends with voters dissatisfied with the economy or the president? Attitudes about the former president, and the party he has reshaped in his image, may largely explain the difference. In the exit poll, nearly three-fifths of voters said they had an unfavorable view of Trump, and more than three-fourths of them voted Democratic this year. Many of the Republican Senate and gubernatorial candidates he helped propel to their nominations also faced negative assessments from voters. And despite predictions from both Republicans and media analysts that abortion had faded as a galvanizing issue, a clear three-fifths majority of all voters in the national exit poll said they believed that the procedure should remain legal in all or most circumstances—and about three-fourths of them voted Democratic. Democrats also won about three-fourths of the voters who said abortion should remain mostly legal in the key Senate states of Arizona, Georgia, and Pennsylvania, and two-thirds of them in New Hampshire. In Michigan, Governor Gretchen Whitmer won a stunning four-fifths of the voters who said abortion should remain legal.

These concerns about Trump and abortion rights didn’t completely erase voter discontent over the economy and inflation. Inflation still ranked highest when the exit polls asked voters what issues most concerned them (with abortion a very close second). And Republicans still won most of the voters who expressed the purest “double negative” views—those with unfavorable opinions of both Biden and Trump. But it’s hardly a surprise that the party out of the White House might win most voters who express an unfavorable view of the sitting president, no matter what other attitudes they hold. The notable part was that the exit poll found Democrats holding 40 percent of those double-negative voters—a number that helped them apparently avoid a titanic red wave.

In the past, when midterms have turned decisively against the sitting president’s party, one reason is a backlash among independent voters, who are the most likely to shift allegiance based on current conditions in the country. Each time the president’s party suffered especially large losses in a midterm since the mid-1980s (a list of electoral calamities that includes 1986, 2006, and 2018 for Republicans and 1994, 2010, and 2014 for Democrats), independents have voted by a double-digit margin for House candidates from the other party, according to exit polls. But yesterday’s exit polls showed the two parties splitting independent voters about evenly on a national basis and Democrats winning among them in the Arizona, Georgia, New Hampshire, and Pennsylvania Senate races.

The other ingredient in decisive midterm losses has been what political strategists call “differential turnout.” Almost always in American history, the party out of the White House has shown more urgency about voting in midterms than the side in power, but when midterms get really bad, that disparity becomes especially pronounced.

A complete picture of this midterm won’t be available for months. But the early indications are that this year’s electorate leaned more toward the GOP than the past few campaigns. In 2020 and 2018, the exit polls found that self-identified Democrats made up slightly more of the voters than Republicans. But the exit polls yesterday showed Republicans with a slight edge.

Young people gave Democrats preponderant margins in most races, but likely made up slightly less of the electorate than they did in 2018. Among voters of color, the story was similar—some erosion in support for Democrats, but not a catastrophic decline. The exit polls showed Democrats winning about 60 percent of Latino voters and 85 percent of Black voters. That was down just slightly from their level in 2020, though it represented a bigger fall from the party’s support with those voters in 2018. Republicans in the coming days will likely trumpet the continuing gains—though Democrats can fairly rebut that they have a clear opportunity to rebound if and when the economy recovers.

Before Election Day, conservative pundits speculated rampantly about a sweeping shift toward the GOP among nonwhite voters without a college degree—what Axios breathlessly declared “a political realignment in real time.” But Democrats nationally carried about two-thirds of those non-college-educated voters of color, almost exactly their share among minorities with degrees; the picture was similar in the heavily diverse states across the Sun Belt, the exit polls found. Among white voters, the familiar educational divides held: The national exit poll showed Democrats slightly underperforming expectations among college-educated whites (winning only about half of them) but still showing much better with them than among non-college-educated whites, who once again broke about two-to-one for the GOP. (College-educated white voters did provide more resounding margins for Kelly, Hassan, and Fetterman, the polls found.)

The full results won’t be known for days, and control of the Senate may not be settled until another runoff election in Georgia. But the 2024 presidential contest will likely kick into motion almost immediately. Trump has repeatedly hinted that he may announce a 2024 candidacy as soon as next week—and the GOP’s gains, even if less than the party anticipated, will only encourage him.

Throughout American history, midterm results have had little relationship to the results in the next presidential contest. Presidents Jimmy Carter and George H. W. Bush had relatively good first-term midterm results in 1978 and 1990, and then lost for reelection two years later. Harry Truman, Ronald Reagan, Bill Clinton, and Barack Obama were all shellacked in their first midterm and then won reelection.

Could Biden follow those precedents and recover in time for 2024? Much will depend on the economy. Doug Sosnik, a senior White House adviser to President Clinton during his recovery after the 1994 midterm, pointed out that the period from fall of the third year to spring of the fourth year is when voters really lock in their judgment about a first-term president. That doesn’t leave Biden much runway to dispel the economic pessimism that weighed so heavily on Democrats yesterday. Many economists believe that the Federal Reserve Board’s actions will trigger at least a mild recession before squeezing out inflation, potentially by late next year.

Given the doubts many voters have expressed about Biden’s age, it’s not clear that a rising economic tide would lift his prospects as much as it did for Reagan, Clinton, and Obama. Many Republicans (and even some Democrats) believe that the loss of the House, and possibly still the Senate, when all of this year’s votes are counted will increase pressure on Biden to step aside in 2024. In the exit polls, two-thirds of voters said they did not want to see Biden run again.

Yet the GOP may be saddled with a 2024 nominee carrying even more baggage. Trump will inevitably interpret any GOP gains as a demand for his return. But even in a Republican-leaning electorate, the exit polls still registered enormous resistance to him.

One of the night’s clearest winners was Trump’s most serious competitor for the next GOP nomination, Governor Ron DeSantis of Florida, who won a convincing victory that included breakthrough results in heavily Latino Miami-Dade County. His success will likely embolden the Republicans urging the party to turn the page from Trump—though Trump has already signaled his willingness to bludgeon DeSantis to secure the nomination, the way he did Marco Rubio and Ted Cruz in 2016.

For Biden, the situation will likely be more equivocal: The results for Democrats probably won’t prove good enough to completely quiet the chatter about replacing him, but nor will they likely prove so bad as to significantly amplify it. After this double-negative election produced something of a standoff between the parties in 2022, it remains entirely possible that the nation may find itself plunged into the same grueling trench warfare between Trump and Biden again two years from now.

Ronald Brownstein is a senior editor at The Atlantic and a senior political analyst for CNN.