One of the most important factors in every election is turn-out. Predicting who turns out, of course, is tricky. History tells us that angry people vote. The more enraged a voter is about issues/the political environment/the candidates, the more engaged in the election they will be. There’s also been a pretty consistent pattern in recent years of who does and who doesn’t turn out in midterm years. Voters who are white, older, and more highly educated turn out more consistently than those who are younger, non-white or non-college educated. The drop off among younger/diverse voters has been problematic for Democrats in recent midterm elections (and, drop off in African-American voters was devastating for the Clinton campaign in 2016 as well).

Data analytics has added an air of authority and predictability to the unpredictable patterns of turn-out. But, of course, 2016 taught us that analytics aren’t infallible. Then there’s the always important question: are elections decided by turn-out or persuasion? Is it more important to simply get your “tribe” out to vote, or is it more important to win over swing voters? As fewer and fewer voters split their tickets, the harder it is for a party to try and “flip” a seat by winning disaffected voters from the other party. The Trump candidacy and presidency has also raised the question of whether personality–not policy–is what really drives voter engagement. Voters aren’t attached to the ideological constructs that many in Washington are, especially since they have little to no faith in the institutions of government themselves. This gives candidates with a personality “brand” an ability to hold multiple positions (a thing we used to call “flip-flopping” or “breaking a promise”), without losing support from their base.

At this early stage in the process, there are conflicting data points about what turnout could look like next fall.

Three reasons why Democrats should be feeling optimistic:

  • Exhibit A: Special elections. Democrats lost all four House special elections. That’s good news for Republicans who can only afford to lose 24 House seats in 2018 if they are to keep their majority status. But, as my colleague David Wasserman has written, the Democratic candidates in these races “have outperformed their ‘generic’ share of the vote significantly in every contest,” by an average of eight points. If they were able to replicate this over-performance in every district in the country in 2018, Democrats would pick up 80 seats. That’s not going to happen, of course, writes Wasserman, “because Republican incumbents will be tougher to dislodge than special election nominees. But these results fit a pattern that should still worry GOP incumbents everywhere, regardless of Trump's national approval rating and the outcome of the healthcare debate in Congress.” 
  • Exhibit B: Intensity. Those who strongly disapprove of the job Trump is doing as president outnumber those who strongly approve his job performance by a 2-1 margin. For example, the latest SurveyMonkey poll finds Trump’s overall job rating at 42 percent approve and 57 percent disapprove. But, while almost a quarter (24 percent) “strongly” approve of the job the president is doing, almost half (47 percent), strongly disapprove. Among partisans, most Democrats (79 percent) “strongly” disapprove while just over half of Republicans (55 percent) “strongly” approve. If enraged really does equal engaged, that should be good news for Democrats.
  • Exhibit C: Independents. They are not happy with Trump’s performance as president. And, historically, their opinion matters. The last three mid-term elections were terrible for the party in the White House. In each case, independent voters broke decisively for the ‘out’ party.

    Independent Vote in Mid-Term Election
    Independent Vote*
    House Results
    2006 D+18 D+30
    2010 R+19 R+63
    2014 R+12 R+13
    * Exit Poll data

    Polls taken right before those three midterm elections, show independent voters lukewarm to downright cold on the sitting president. The most recent Gallup poll finds Trump’s approval among independents at 31 percent. That’s about where George W. Bush was sitting in October of 2006. On Election Day, Republicans lost 30 seats in the House and the majority.
    Pres. Job Approval
    Pres. Job Approval Among Independents
    10/20-22/06 GW Bush 37% 28%
    10/25-31/10 Obama 45% 41%
    10/20-26/14 Obama 42% 37%
    7/17-23/17 Trump 37% 31%

Reasons Republicans should be feeling optimistic:

  • Exhibit A: Limits of Enthusiasm. There was a lot in the recent ABC/Washington Post poll to make Republicans cringe. The President’s approval rating dropped to 36 percent. By a 52 to 38 percent margin, voters said they’d like to see Democrats control Congress next year. But, it also highlighted the limits of the “enthusiasm” advantage currently held by Democrats. Among those who said they “strongly” disapproved of the job Trump was doing, 61 percent said they were certain to vote in 2018. But, among those who “strongly approved” of Trump, 72 percent said they were going to vote in the upcoming midterms. While that can change over the next few months, it should serve as a healthy reminder to Democrats that rallies and marches aren’t always indicative of voting behavior.
  • Exhibit B: Trump’s Not Losing His Base. Talk to GOP consultants and strategists and they’ll tell you that despite all the tumult and trouble in DC, Trump remains very popular with his base. One strategist told me: “Trump is more popular than any Republican president I’ve seen since Bush post 9/11.” The success of Republican Karen Handel in GA-06 suggests that Trump and the Republicans in Congress have a longer rope with traditional Republican voters than many had expected.

    In places where Trump outperformed previous GOP nominees–the so-called “Trump Counties”–registered voters actually like him more today than they did in 2016. Earlier in July, NBC/Wall Street Journal poll did a deep dive into parts of the country that either: 1) flipped from voting for Obama in 2012 to Trump in 2016, or 2) where Trump performed a net 20 points higher than Romney did in 2012. The interviews were conducted in 16 swing states: Colorado, Florida, Georgia, Indiana, Iowa, Maine, Michigan, Minnesota, Missouri, Nevada, New Hampshire, North Carolina, Ohio, Pennsylvania, Virginia and Wisconsin.

    In 2016, Trump was personally disliked in these areas–just 35 percent had a positive feelings about him while 54 percent viewed him negatively. But, by July, 47 percent said they felt positively about him, while 38 percent viewed him negatively.

    And, despite the friction between the president and his party in Congress–on Russia, on health care, on his tweeting habits–voters in the Trump Counties are as willing to support GOP congressional candidates today as they were back in 2016. Back in 2016, Republicans had a ten-point advantage on the congressional ballot test (48 to 38 percent). This July, Republicans still held a 10-point advantage (47 percent to 37 percent).

    What we don’t know, of course, is Trump’s support in his flip/surge counties is equal to–or more significant–than the desire to oust him in other parts of the country. Moreover, we don’t have an equivalent “Clinton County” metric to compare these numbers to.
  • Exhibit C: The Missing Obama Coalition. The coalition that helped put–and keep–Barack Obama in office has been unreliable for Democrats. A recent study by the Voter Participation Center–a 501c3 group “dedicated to increasing the share of unmarried women and other historically under-represented groups in the electorate,” warned that participation among the so-called Rising American Electorate or RAE (non-white, younger, and unmarried women), will drop off more significantly in 2018 than non-RAE voters. They predict that “35.1% of those who voted in 2016, or 25.4 million RAE voters, will stay home. The predicted drop-off among non-RAE voters is only 22.1% or 14.4 million voters. In fact, of the nearly 40 million Americans predicted to drop-off from 2016, two-thirds will come from the RAE.” The challenge for Democrats, says the Center’s pollster, Josh Ulibarri of the Democratic polling firm Lake Research Partners, is that Democrats “have to persuade our voters to turn-out.” This is going to require both a message that resonates with these voters, says Ulibarri, as well as a serious – and early – effort to register new voters and engage with them throughout 2017-18, not just in October of the mid-term year.

    Ulibarri’s message is echoed by Steve Phillips, a senior fellow at the Center for American Progress and the founder of Democracy in Color. In a New York Times Op-ed, Phillips argues that Democrats are focused too much on trying to woo “Trump country” voters to support Democrats and not enough time motivating and turning out voters of color who skipped the 2016 campaign.

    The Democratic Party committees and its allies are likely to spend more than $750 million on the 2018 midterms. Will they spend it fruitlessly trying to lure Trump voters, or will they give uninspired black Democrats a reason to vote and offer disaffected Obama-Johnstein voters a reason to return to the fold?
    Steve Phillips

    A Washington Post poll of DC residents taken earlier this summer found an activism gap between white and non-white district voters. According to the Post poll, “53 percent of white residents participated in a march or demonstration in opposition to Trump’s policies since the start of the year, compared with 16 percent of African Americans and 36 percent of Hispanics and those of other racial and ethnic groups.” While many white residents of DC are wringing their hands and/or shaking their fists, African-Americans see little incentive to taking to the streets to protest the president. “Justina Jackson, 28, a black assistant pastry chef in Brookland who voted for Hillary Clinton, said that she is accustomed to being disappointed in government, no matter which party is in the White House, and that Trump is no different,” wrote the Post’s Paul Schwartzman and Emily Guskin. “I’m used to stuff not going my way,” she said. “There’s always some kind of obstacle I have to overcome just because I’m a young African American female. I don’t have a lot of sympathy for people who are upset about Trump.”

    And, the lack of interest in protesting also suggests a lack of interest in voting next fall. A SurveyMonkey poll taken this June found that while Trump had dismal approval ratings among non-white voters (17 percent among African-Americans and 28 percent among Hispanics), these voters were less committed to voting next fall than white voters who gave him a 50 percent job approval.

Bottom Line:

Both Democrats and Republicans should resist the temptation to cherry pick from the data in front of us. It’s also very early–too early to find a clear pattern. Still, the challenges for the two parties are becoming pretty clear. Republicans should worry about bleeding support from independent voters and an enthusiasm gap between their voters and the Democratic base. Democrats, however, can’t count on intense dislike for Trump to automatically translate into voter turnout. Moreover, Democrats’ difficulty in motivating and turning out the Obama coalition in years when Obama’s not on the ticket is showing no sign of improvement.


More from the Cook Political Report

Trump Biden
2024 Swing State Project
First Person
Cook Politcal Logo
Larger logo
2024 Swing State Project
Cook Politcal Logo