Americans took to the polls in record numbers in the 2018 midterms, shifting party control of the House of Representatives and sending a clear message of disapproval to President Donald Trump and congressional Republicans. Although the president and his party gained ground in the U.S. Senate, primarily in states Trump won handily, they failed to capitalize on the low unemployment rate or overall positive sentiments about the economy. The signature GOP legislative achievement of the first two years—the $1.5 trillion tax cut that passed last year —failed to boost Republicans’ chances overall and hurt candidates in several seats. Subsequently, they lost in major suburban and urban districts across the country and also lost ground in some rural areas. The president’s gamble of nationalizing the election around his personality and his administration’s harsh immigration policies ultimately cost Republicans their House majority and failed to persuade voters outside of already conservative or rural counties and states to stick with the GOP. Likewise, health care dominated voters’ minds this year according to both pre-election and Election Day polls, with Democrats benefitting from their commitment to protect and expand Americans’ health care and House Republicans suffering for their repeated attempts to repeal the Affordable Care Act (ACA).
Democrats are on pace for around a 34-seat gain in the House (outstanding races are still to be called in a few places), with a record number of women candidates winning overall. This is slightly above historical average gains for opposition parties in midterm elections but below the massive 2010 shift of 63 seats during the Obama presidency. Democrats gained seven governors’ seats, including in important presidential battleground states such as Michigan and Wisconsin and flipped six state legislative bodies, with about 330 state legislative seats gained across the country. Ballot initiatives to expand Medicaid won in three red states, while several measures to increase the minimum wage, legalize marijuana, and expand voting rights also emerged victorious. And, although definitive turnout data will not be available for a while, preliminary estimates suggest a massive increase in voter participation, with likely more than 110 million votes cast for the House—far above 2014 levels.
Midterm elections typically unfold on fleeting political terms and local issues that cannot be applied easily to future elections. But, given the highly polarized nature of U.S. politics under President Trump and the partisan divisions in control of the House and Senate and in key states, a few trends should be noted. The first two favor Democrats going into 2020, and the second two favor Republicans.
- First, President Trump has not expanded his support in any significant manner and his policy agenda has dragged down his popularity, particularly with independents. The president’s disapproval ratings have been higher than his approval ratings for his entire tenure. Strong disapproval of Trump has hovered around 40 percent for the past two years, with less than one-quarter of voters strongly approving of his presidency. This is unlikely to shift, particularly since the president could not improve his numbers even with strong economic indicators. President Trump’s signature policy achievement, the 2017 tax bill, failed to produce gains, and in fact served as a liability for some Republicans, as most Americans concluded that it did little to help them while primarily benefitting corporations and the wealthy. Likewise, Trump’s immigration agenda remains potent among his base supporters, but it is not moving the needle his way in large swaths of suburban and urban America. The repeated attempt to undermine national health care policy remains a significant drag on his support among independents and moderates across the nation. Unless the president changes course toward a more popular national agenda, it is unclear how he plans to both secure and expand his 2016 vote base. Exit polls indicate that Republicans lost significant ground with independents in 2018, relative to Trump’s decent showing among them in 2016.
- Second, Trump’s standing in the Electoral College is uncertain. Outside of Florida and Ohio, where Republicans appear to have held off an opposition surge this year, Democrats made substantial statewide gains in key states in the 2018 midterms, controlling the governorships of Pennsylvania, Michigan, and Wisconsin and picking up Obama-Trump districts in the state of Iowa. Trump’s hold on the Electoral College, which was weak in 2016 with only a narrow victory of around 80,000 votes in three states, is therefore tenuous—at best. Democrats only need to add Pennsylvania and Michigan back to their column plus either Wisconsin or Iowa to win in 2020. This assumes Florida and Ohio remain for Trump and no changes occur in emerging battlegrounds such as North Carolina, Arizona, or Georgia, neither of which are given.
- Third, Democrats have demographic and geographic challenges in relation to both the Senate and some key Electoral College states. Despite advantages that emerged in the 2018 midterms and overall favorable demographic trends, the Democratic Party base of support is still shaky in terms of potential turnout and support rates going into 2020, relative to the geographic structure of the Electoral College and the Senate. Youth turnout and turnout from voters of color looks very strong for 2018, but even minor dips in key Electoral College states can shift the terrain toward Trump. Likewise, the concentration of Democratic support in metropolitan areas diffuses overall demographic advantages and increases the chances that large turnout and support from Trump’s base, primarily white noncollege-educated voters in more rural and working class states, can once again lead to a narrow victory.
- Fourth, Democrats need to develop a strong and popular vision to counter Trump’s nationalist message. Despite Trump’s manifest political deficiencies, he does benefit from very strong support and fervent backing from most Republicans. In contrast, the Democrats successful, mostly nonideological “big tent” strategy for 2018 will not hold for 2020. In order to ensure strong Democratic base turnout, encourage new voters, and persuade Obama-Trump and more white noncollege voters to return to the party, Democrats will need a very clear and compelling vision that convinces voters that they are on their side on economics and social policy and are willing to make significant changes to a political system that is viewed as corrupt and often unresponsive to voter needs.
With this overall context, here is a concise overview of what we know so far about who voted and how they voted in 2018 and what it may mean going forward.
Who voted in 2018?
Perhaps the shortest answer to this question is “everyone.” This was an exceptionally high turnout for a midterm election. It would therefore be surprising if the turnout of most demographic groups did not go up. However, that does not mean that the share of voters attributable to these various groups necessarily went up. That would only be true if turnout of a given group went up more than the average among eligible voters and/or if the eligible voter share of a given group went up.
With that in mind, some patterns can provisionally be seen in the results currently available. We emphasize provisional since the exit polls are typically an unreliable guide to turnout patterns and need to be supplemented with other survey data and modeling that integrates actual election returns. Those supplementary resources are not yet available.
The National Election Pool (NEP) exit polls, compiled by Edison Research for a consortium of news organizations, indicate that the share of white voters fell from 75 percent in 2014 to 72 percent this election. Note that this does not necessarily tell us much about the turnout of white voters, since some of this declining vote share (if real) was attributable to the declining white share of eligible voters.
The same could be said about the rising share of nonwhite (black, Hispanic, Asian, and other race) voters, which increased from 25 percent in 2014 to 28 percent in the 2018 NEP exit polls. Some of this increased vote share, if real, was due to the increasing nonwhite share of eligible voters, rather than unusually high nonwhite turnout.
Within the broad group of nonwhite voters, these exit polls indicate that Hispanic vote share went up from 8 percent in 2014 to 11 percent in 2018, while black vote share declined slightly from 12 to 11 percent over the same time period. (The AP-NORC VoteCast voter survey, which is new and has a different methodology than the NEP exit polls, reports the Hispanic share as 9 percent and the black share at 12 percent; since the survey is new, there are no comparisons available to 2014.)
Within the overall group of white voters, it has become important to look at education divisions. However, while it seems likely that both white college graduate and white noncollege voters increased their turnout levels this election, it is impossible to say how much their vote shares really changed. Because the NEP exit polls changed their methodology around education, they now show a much more realistic distribution of voters between white college and white noncollege: 31 percent white college and 41 percent white noncollege. But this methodology change invalidates any comparison to their previous exit poll voter shares among this demographic.
Voter share among 18- to 29-year-olds remained stable at 13 percent between the 2014 and 2018 elections, according to these exit polls. This implies that young voter turnout did indeed go up in this election but by no more than the average turnout increase.
How did they vote?
The most interesting changes in this election, in comparison to 2014, took place in the realm of voter preference. As we might expect, there were big shifts toward the Democrats among many voter groups, given that the overall House popular vote shifted from +6 Republican in 2014 to the current estimate of +7 for the Democrats this election.
Notably, women went heavily for Democrats, with a +19 margin in the NEP exit polls, while Republicans had a +4 margin among men. Comparable figures for 2014 were +4 for Democrats among women and +16 for Republicans among men. The gender gap is alive and well.
Turning to the white vote, these exit polls indicate that Democrats lost the white congressional vote by 10 points this election, a substantial improvement over their 22-point loss in 2014. Among nonwhites, Democrats improved their margin among Hispanics from +26 to +40 across the two elections, from +79 to +81 among blacks overall (with black women at +85 in 2018), and from a mere +1 among Asians to +54 this election. Again, we await further data to evaluate these changes, but this is the story told by the NEP exit polls.
While possibly affected by changes in methodology, these exit polls indicate a very strong pro-Democratic shift among white college voters, improving from a 16-point deficit in 2014 to an 8-point advantage in this election. Democratic performance also improved among white noncollege voters but only modestly, moving from a 30-point deficit in 2014 to 24 points in 2018. Other data indicate that Democrats did particularly poorly among white noncollege voters in the South.
White college women were particularly good for the Democrats, supporting their candidates by a 20-point margin; white college men gave Republicans a 4-point advantage. White noncollege men were the worst for the Democrats—they lost this group by a whopping 34 points. Democrats did better among white noncollege women, losing them by a comparatively modest 14 points.
Young voters may not have increased their share of voters but, according to the NEP exit polls, they were very pro-Democratic this year, increasing their support for Democratic candidates from +11 in 2014 to +35 this year. Notably, the 18- to 24-year-old group, which now includes a healthy share of post-Millennials—the pro-Democratic Millennials’ successor generation—actually voted more Democratic (+37) than the 25- to 29-year-old group (+33).
What does this mean for 2020?
America remains a deeply divided nation in its politics and partisan preferences. The 2018 midterms clearly did not settle these divisions. If anything, it appears to have solidified them. Republicans under Trump maintain a strong hold on white noncollege voters in more rural and exurban counties and states across the country. However, they are rapidly losing support among college educated whites in many suburban and metropolitan areas and face particular challenges with women and voters of color.
Democrats, on the other hand, have broader but less intense support that can manifest itself in strong majorities with the right leadership, as in 2008 and 2012 under former President Barack Obama, or fall just short when facing intense partisan support from Trump and his base. Democrats lack both intense bases of support and the ability to reach into voting blocs that are deeply upset with the political status quo. These challenges may be amplified by Trump’s geographic advantages in some key Electoral College states, as seen in 2016.
President Trump appears committed to his strategy of base mobilization built on maximizing conflict on cultural and racial grounds. If he wants to succeed in 2020, he will need to broaden his outreach, adopt a more accommodating style, and focus on the economy more or else risk serious blowback across his narrow Electoral College pathway. Democrats, in turn, must quickly solidify an inclusive and forward-looking vision and agenda that offers voters turned off by Trump’s style, corruption, and ethno-nationalism a real reason to vote for their party. They cannot afford to remain in constant locked-horn battle with Trump on his terms and must stay focused on building consensus behind principled and pragmatic progressive policies on jobs, wages, health care, and democratic participation.
Whichever side figures out how to both maximize its partisan advantages and make inroads with voters who do not fit their ideological profile will most likely emerge victorious in 2020.
John Halpin and Ruy Teixeira are senior fellows at the Center for American Progress.