Conservatives are calling on Democrats to win back Trump voters with centrist policies. But there's a much more effective way for Democrats to defeat Trump
- Conservative commentators like David Brooks and Peggy Noonan have recently argued that the 2020 Democratic field is too far to the left to have a chance at winning back Obama-Trump voters.
- But an analysis of data from the 2016 election shows that Democrats don't necessarily need to win back Obama-Trump voters, or Trump voters at all, to win in 2020.
- Multiple analyses have found that those who didn't vote at all in 2016 are much more similar to the base of the Democratic party than they are to Obama-Trump voters.
- This is especially true for those who voted for Obama in 2012 but stayed home in 2016.
- An analysis from Data For Progress and three political scientists shows that Obama-Trump voters don't just dislike Medicare For All, but over 70% opposed the Affordable Care Act and less than half support a path to citizenship for undocumented immigrants - putting them out of step with even centrist Democrats.
- Visit Business Insider's homepage for more stories.
Over the past week, conservative commentators including The New York Times' Bret Stephens and David Brooks wrote columns in which they expressed concern over what they called a sharp turn to the left among 2020 Democratic presidential candidates.
Stephens wrote that he was concerned that the high price tag of some Democratic policy proposals would scare away "ordinary voters (e.g., people who voted for Obama and later Trump)," with Brooks worrying that "the party is moving toward all sorts of positions that drive away moderates and make it more likely the nominee will be unelectable."
Naval War College professor and self-described "Never Trumper" Tom Nichols wrote a similar USA Today op-ed warning that Democrats' "lurch left" will alienate Obama-Trump voters in swing states. And former Bush speechwriter Peggy Noonan similarly wondered in the Wall Street Journal what Democrats "are offering voters who backed Trump in 2016."
All of these commentators argued that by embracing policies including Medicare For All, government healthcare coverage for undocumented immigrants, and decriminalization of unauthorized border crossings, Democrats are alienating swing voters who choose Trump in 2016, throwing away their chances of ever winning those voters back, and therefore, losing the election altogether.
While conservatives like Stephens, Brooks, and McCain may boast hundreds of thousands of Twitter followers and big platforms, they don't represent the Democratic base or the Democratic primary.
A close examination of electoral data from 2016 shows that Democrats don't actually need to win back a substantial number of Trump voters to win the election in 2020 - and that focusing on turning out non-voters may be a more promising strategy.
There may not be huge swaths of moderate swing voters
Brooks' column mainly argued that Democrats are alienating the 35% of Americans who identify as political moderates, according to Gallup polling.
Conservative Washington Post columnist Max Boot echoed a very similar sentiment, arguing that "Trump doesn't need his own agenda if he can terrify independent voters in swing states about what would happen if the Democratic agenda is implemented."
A Pew Research Center study from earlier this year, however, dispelled the myth that the American electorate is full of up-for-grabs swing voters who are equally likely to vote for either party.
Pew instead found that the vast majority of political independents prefer one party over the other, and just 7% of Americans - including self-described independents and moderates - report having no political lean at all. They also found serious policy disagreements among independents on many key issues, making it "misleading to look at 'independents' as a single bloc."
Furthermore, the dynamics of the 2020 election are quite different from 2016 where voters had never seen what either candidate but would be like as president, taking a chance on the candidate of their choice.
Trump has been in office for nearly two and a half years, giving the electorate a much better sense of what another four years of his presidency would entail.
While there are certainly undecided voters across the ideological spectrum, it seems unlikely that voters who still fervently support Trump will change their minds now, or be swayed by Democrats slightly moving to the center on certain policy issues - especially with Republicans accusing all Democrats, including moderate ones like former Vice President Joe Biden, of "advocating a socialist agenda."
Obama-Trump voters are a smaller segment of the electorate than nonvoters
An in-depth analysis of validated voter data from the Cooperative Congressional Election Survey (CCES) conducted by Data For Progress co-founder Sean McElwee and political scientists Brian Schaffner, Jesse Rhodes, and Bernard Fraga estimated that out of the 66 million people who voted for Obama in 2012, 80%, or 53 million, voted for Clinton in 2016, 9%, or 6 million people, voted for Trump in 2016, 2.3% voted for third parties, and 7% - or 4.4 million - stayed home altogether.
Because the 207 counties that flipped from Obama to Trump are largely concentrated in Upper Midwest swing states including Iowa, Michigan, and Wisconsin, they played a pivotal role in Trump victory and have received significant media attention since.
But according to a 2018 analysis from the Pew Research Center, the approximately 38 million people, or 42% of eligible voters, who did not cast a ballot in 2016 also substantially contributed to Trump's 2016 election by staying home, largely because many expressed political leanings and belonged to demographic groups that lean Democratic.
Of all nonvoters, 55% identified as Democrats or Democratic-leaning, and 37% expressed a preference for Clinton compared to 30% who preferred Trump, according to Pew.
The Pew Research study and the aforementioned CCES analysis further showed that nonvoters are more similar to the Democratic base in many respects:
- 52% of all nonvoters and 51% of Obama-to-nonvoters were non-white, in comparison to 16% of Obama-Trump voters. In 2016, Hillary Clinton overwhelmingly won among African-Americans, and carried around two-thirds of both Latinx and Asian-American voters.
- 29% of nonvoters and 23% of Obama-to-nonvoters were under the age of 30 - an age group Clinton won - compared to 11% of Obama-Trump voters.
- 56% of nonvoters and 60% of Obama-to-nonvoters made less than $50,000 per year - a group that went to Clinton by 53% to 41% - compared to 52% of Obama-Trump voters.
There are a myriad of mathematical ways that Democrats could win in 2020 without winning back any of the estimated 6 million Obama-Trump voters by turning out the 65 million voters who cast ballots for Clinton, combined with a share of those who voted for third parties in 2016, and/or making inroads among the 42% of the electorate that didn't cast ballots at all in the 2016 race.
The Washington Post's Dave Weigel pointed out in a Saturday tweet, for example, that "Democrats could win in 2020 by taking the Clinton 2016 voter and 2/3 of people who voted third party in 2016 - and not a single Trump voter. Trump's base won't be as static, but they don't *need* Trump voters."
Obama to non-voters are more in line with Democrats on policy than Obama-Trump voters
Not only are there numerically more nonvoters than Obama-Trump voters for Democrats to win back, but the policy positions of the former group are already more aligned with the Democratic party, according to the McElwee, Rhodes, Schaffner, and Fraga study of 2012 Obama voters.
Brooks', Noonan's, and Nichols' case for Democrats to move to the right on healthcare and immigration to win back those voters makes even less sense given what Obama-Trump voters actually believe.
Obama-Trump voters don't just dislike Medicare for All, but over 70% wanted to scrap the Affordable Care Act altogether in 2016, a law that even the most moderate Democrats in the 2020 field say they want to preserve and expand.
When it comes to immigration too, just under 50% of Obama-Trump voters supported a path to citizenship for undocumented immigrants - a mainstream position in the Democratic party- compared to the 70% of Obama to non-voters who support such a path.
The study showed that Obama to non-voters are much closer to the consistently Democratic voters who voted for Obama in 2012 and Clinton in 2016 on every issue from healthcare to abortion and the environment.
Noonan argued the embrace of policies like Medicare for All are "too extreme for America, and too extreme for the big parts of its old base that the Democrats forgot in 2016" - but it's not even clear that Obama-Trump voters could be considered solidly part of the Democratic base in the first place.
As the authors of the CCES analysis note: "less than one third of Obama-to-Trump voters supported Democrats down-ballot in 2016, and only 37 percent identify as Democrats."
It's too early to really know who is most 'electable'
Washington Post data columnist David Byler wrote in February that "electability" is often falsely conflated with winning over the stereotypical image of a rust-belt, white working class voter in the Upper Midwest.
He argues that while working-class whites are an incredibly important voting bloc for both major parties, low turnout among voters of color in crucial swing states arguably cost Clinton in 2016, and will be essential for the eventual 2020 Democratic nominee to turn out.
Byler noted that Democrats can solidify their position by restoring African-American voter turnout in battleground states like Michigan and Wisconsin, which plateaued in 2016 after reaching an all-time high in 2008 and 2012.
Citing a study from demographers Ruy Teixiera and Keith Griffin, Byler noted that Clinton would have won the electorate college had she replicated Obama's performance among African-American voters in both Midwestern and Southern swing states.
Asian-American and Latino voters further represent two other segments of the electorate among which Democrats currently have a slight advantage, but turn out to vote at relatively low rates.
Such voters are mostly concentrated in safe blue states, but Byler notes that concerted outreach and voter registration efforts in those communities could help Democrats flip Arizona - where the Democratic party is gaining ground - or make inroads in Florida, where Republicans have recently been prevailing in statewide races.