The Good Fight

Forget Michigan

If Democrats focus on winning back the Obama states Hillary lost, they’re doomed.

The podium stands empty on stage during preparations for the Democratic National Convention
The podium stands empty onstage during preparations for the Democratic National Convention at Time Warner Cable Arena on Sept. 3, 2012, in Charlotte, North Carolina.

Justin Sullivan/Getty Images

Just as generals are always fighting the previous war, campaign strategists always seem to be fighting the previous election. And so every professional analyst and armchair pundit I’ve asked about how Democrats might win in 2020 has immediately offered me an answer to a subtly—but importantly—different question: “Here’s how you win back the white working class.” Which is really just shorthand for: “Here’s how you win back Michigan!”

Hillary Clinton, the standard story goes, lost the 2016 election because of some 80,000 votes in three Rust Belt states: Michigan, Pennsylvania, and Wisconsin. And she did poorly in those states in large part because a lot of white, working-class voters who had supported Barack Obama in the past voted for Donald Trump. The implication seems obvious: To take the White House in 2020, Democrats need to win back white, working-class voters in the crucial states that defected last time around. And so Katrina vanden Heuvel, the publisher of the Nation, has been musing aloud about “How Democrats can win back the working class.” To “get their mojo back,” David Leonhardt argued in the New York Times, “Democrats have to find a way to win more working-class votes.” (The piece makes clear that, while Leonhardt is talking about the working class, he mostly means poor whites.) John Stoehr summarized the emerging consensus most succinctly: “Democrats Need the White Working Class,” he wrote a few weeks ago. “No future Democrat should believe that he can win without them.”

I’ve increasingly become convinced that this thinking is wrong. While Democrats do of course need to win some support among the white working class, the idea that it is the key demographic to target in the coming years is based on a series of implicit assumptions that turn out to be highly implausible once they are made explicit—and therefore risk leading Democrats astray as they try to rebuild a winning coalition.

The first of these assumptions is that the political behavior of different states is relatively stable over time: If the basic political geography of America is pretty much the same at every election, then the route to 270 electoral votes will always involve going to battle over the same old regions. To have a good chance of winning in 2020, the Democratic candidate would have to recapture the states that delivered a previous victory for his or her party.

But a quick look at past presidential maps shows just how quickly the political geography of the United States can change. Would you be able to guess which seven states remained Democratic in Dwight D. Eisenhower’s 1956 blowout victory over liberal champion Adlai Stevenson? Alabama, Arkansas, Georgia, Mississippi, Missouri, North Carolina, and South Carolina. And do you know when, over the past century, California first voted Democrat in a competitive election? 1992.

Political historians often talk about moments of realignment, in which the main political cleavage that structures American politics shifts—and the relative strength that Democrats and Republicans have in different parts of the country rapidly transforms. Given how radically American politics has been changing over the past decade, there is every chance that 2020 will prove to be such a moment of realignment.

But the truth is that, even if the next election turns out to be reasonably ordinary, Democrats could win with a very different geographical configuration. Sure, they could flip Pennsylvania and Wisconsin and Michigan. But they could also win by flipping Florida as well as North Carolina or Arizona or Georgia—states where appealing to the white working class is not as important as increasing turnout among minority groups and appealing to moderate voters in the suburbs. Which path is easier for them? This is a key question for campaign strategists to answer as they plot the easiest path back to the White House, not something for them to assume based on what it would have taken for Hillary Clinton to win an election that is already lost.

But let us suppose, for the sake of argument, that campaign strategists go all out on a plan that has never worked before: winning the exact same configuration of states that delivered their party’s latest victory. The second big assumption is that they should then focus on increasing their support among the kinds of voters who deserted them last time—which is to say, on white, working-class voters in the declining industrial towns of the Rust Belt. Is that a realistic path to winning back those states?

Probably not. In both Ohio and Pennsylvania, white, working-class voters have become about a quarter less likely to vote for the Democratic candidate over the past eight years. This is part of a much larger trend. Across the United States, the share of the white working class that votes for the Democratic candidate has dropped from 40 percent to 29 percent over the same time period. And the working-class migration toward the far right has been even more marked across much of Europe: In France, support for Marine Le Pen’s National Front party among the working class doubled from 16 to 33 percent between 2007 and 2012. In 2017 it nearly doubled again, rising to 63 percent. In Austria’s 2016 presidential election, 86 percent of working-class voters cast their ballots for Norbert Hofer, the candidate of the far right.

Would it be easier for Democrats to turn the clock back by reversing trends? Or would it be easier for them to lean into changes that have been a long time in the making, trying to capitalize on the trends that helped to get Barack Obama elected in the first place?

Put like that, the answer seems obvious. Instead of channeling most their energies into increasing their share of the vote among a shrinking portion of the American population that has long been growing more hostile to them, Democrats

should seek to keep growing their vote share among expanding portions of the American population that are becoming more friendly to them. This would include the “coalition of minorities” to which Barack Obama’s two resounding victories have often been credited. But it also includes the large number of white, educated, suburban voters who are growing increasingly horrified by Donald Trump’s extremism—and its embrace by the GOP.

Once we dispense with the two big misconceptions dominating the Democratic conversation about its future coalition, two important insights become obvious: Democrats should focus on growing, not on shrinking, parts of the population. And they should try to win new states, not just recover old ones.

One key advantage of this new strategy is that it gives Democrats a lot more geographical flexibility. The number of swing states in which the white working class has historically voted Democrat and makes up a large portion of the state is relatively small. So if Democrats aim to assemble a winning coalition by making a play for Michigan but end up falling a few thousand votes short again, they are likely to lose the White House—even if they once again gain a popular majority.

By contrast, there are many states in which elections will, in the coming decades, be decided by a mixture of suburban and minority voters. A forward-looking strategy would thus allow Democrats to make a serious play for Florida, Georgia, and North Carolina—all states that have both a rapidly growing minority population and plenty of political moderates disappointed by Trump’s performance in office. Much sooner than most people realize, this might even put them in a position to turn supposedly unwinnable states like Arizona or Texas blue. And since those same groups are also ascendant in Rust Belt states, this strategy may, paradoxically, even give them a better shot at winning back Michigan, Pennsylvania, and Wisconsin than a single-minded focus on the white working class.

A final point: The trauma of 2016 is in danger of making Democrats focus on the wrong states and the wrong demographic groups. It’s important for them to avoid this easily avoidable misstep. But in talking about this state and that, and contrasting this demographic with that demographic, I fear that I’m encouraging an even bigger mistake. For the real problem with Hillary Clinton’s strategy was not that she focused on the wrong set of voters but rather that she had so little by way of an exciting vision to offer to all voters. And by the same token, the biggest lesson from last year’s election lies not in recognizing the danger of slicing and dicing the electorate in the wrong way—but rather in relying so heavily on slicing and dicing the electorate in the first place.

The best way to be happy, some psychologists have suggested, is to forget about your desire to be happy—and put all your energy into a project about which you’re passionate. I increasingly think that the task of building a winning electoral coalition is rather similar: Trying to predict the exact states that will turn out to be close, and figuring out how to woo the exact demographic groups needed to win those states, is almost certain to fail. And so the most promising way to win 270 electoral votes in 2020 remains perfectly simple: Democrats will have to craft an optimistic, forward-looking message that speaks to a large number of Americans across the nation—white or black, poor or affluent, Floridian, Arizonan, or Michigander.