Could happen...

How Hillary Loses

Donald Trump can actually win if Clinton makes these four mistakes. Spoiler alert: She’s already making all of them.

By David S. Bernstein

It’s a terrifying moment for Democrats:  Hillary Clinton’s double-digit lead in national polls has evaporated and panic is beginning to set in. Polls now show Donald Trump ahead of Clinton, or at worst only a few points behind. During the insanity of the Republican primary, it was easy for them to believe that Trump could never be president—that in a general election, mainstream voters would regard him as an absurdity. But Clinton remains a shaky candidate with historically high negatives, an email scandal that keeps getting worse and a stubborn primary opponent whose supporters may yet become a midsummer nightmare in Philadelphia. Meanwhile, the Republicans, seemingly in all-out civil war just weeks ago, have quickly fallen in line. Democrats are resigning themselves to a tough, ugly, painful and expensive street fight.

The numbers offer some reassurance for Democrats—but also some bad news.

The reassurance is that the recent polls probably don’t mean much. Trump’s current surge is likely driven by Republican voters coalescing around their nominee, and Clinton will almost certainly get a similar bump when Bernie Sanders lets go and Democratic voters return to the fold. Most pundits believe 2016 is still Clinton’s race to lose.

Here’s the bad news: There is now a clear path for her to lose it.

If you drill down enough, it’s clear there are at least four paths to a loss, and any one of them poses a real risk for a candidate likely to follow her usual careful, calculating playbook. The cold math of a potential Clinton defeat is not to be found in national polls, but in the Electoral College—and within each state’s unique demographics and culture. Trump won’t dramatically remake the political map, but he doesn’t need to. He just needs to squeeze a little more out of certain voters in certain states, while Clinton draws a little less.

If Clinton pushes away some of her potential supporters; fails to energize others to vote; and fires up Trump’s base by pandering to her own—well, she just might be able to make the numbers work out for him. If he does pull off the election of the century, Trump’s path to 270 Electoral College votes will begin with 164 practically in the bank, from 21 solid-red states generally considered sure things for the Republican nominee. And here’s how Clinton could push more than enough additional states onto Trump’s side of the ledger—Florida, Ohio, Pennsylvania, New Hampshire, Iowa, Virginia, Colorado, Nevada, Arizona, Wisconsin, Michigan—one mistake at a time.

Step 1: Take Hispanic enthusiasm for granted

It’s been a matter of faith in Democratic circles: Trump’s grotesque demonization of Latin-American immigrants will boost Hispanic turnout and Clinton’s share of their vote. As a result, you’re already hearing a lot less about Housing and Urban Development Secretary Julián Castro, once the odds-on favorite to become Clinton’s vice presidential running mate. Castro was supposed to be part of a big Democratic push for Hispanic votes this year. Now, the thinking seems to be, those votes will take care of themselves.
Early evidence certainly supports that belief. Hispanic-Americans dislike Trump—strongly dislike him—in massive majorities, according to polls. Legal residents are rushing to become citizens, and citizens are registering to vote, just so they can cast a ballot against him in November. That has Clinton supporters believing that she’ll win crucial victories in Florida—where 17 percent of the 2012 vote was Hispanic, according to exit polls—Colorado, Nevada and possibly even Arizona.

But it would be difficult for Trump to keep doing as poorly with Latino voters as he’s done over the past year. And if he’s able to keep his incendiary language to a minimum, there is no guarantee that Clinton’s energy will hold for the many months until the election.

There is also reason to think Clinton’s enthusiasm with Hispanic voters needs stoking. A new Fox Latino poll shows Clinton leading Trump by an impressive-sounding 39 points: 62 to 23. But there’s a problem: That 39-point spread is actually less than the 44 by which Barack Obama beat Mitt Romney in 2012.

Florida, where Democratic confidence is sky-high, carries a critical 29 Electoral College votes. In 2012, according to exit polls, Hispanics made up a larger percentage of the state’s vote than in previous years, and Obama won a higher percentage of them—60 percent—than any Democrat had before. That translated into a 285,600-vote advantage (20 percent) among Hispanic voters for Obama over Romney in the state, which Obama carried by just 73,000 votes overall.

The big question is: Can Clinton sustain that kind of historic lead? All Trump would have to do is roll back the Democratic advantage to 2008 levels, instead of 2012 levels, to reverse the tide. All else being equal, a return to 2008’s numbers—when Hispanics were 14 percent of the vote, and Obama won them by a 15 percent margin rather than 20 percent—would mean Democrats losing 109,200 votes off their advantage. And that could turn Obama’s 73,000-vote Florida victory into a 36,000-vote defeat.

Yes, their numbers are growing. But Hispanics simply don’t like Clinton nearly as much as they like Obama: Her favorable/unfavorable is a net +15 in that Fox Latino poll, while Obama’s is +46. Colorado, where the fast-growing Hispanic population gave 75 percent of its vote to Obama in 2008, is a similar story to Florida. So is Nevada, where all of the major analysts still rate the Senate race between Republican Joe Heck and Democrat Catherine Cortez Masto a tossup—suggesting that they aren’t yet foreseeing a torrent of Democratic-voting Hispanics rush the polls in November.

Oh, and did we mention that Hispanic voters are disproportionately young—a staggering 44 percent of eligible Hispanic voters this year are millennials, compared with 27 percent of non-Hispanic whites, according to Pew—and that Sanders has been pulling large numbers of them away from Clinton, just as much as others their age.

Trump Wins: Arizona (11 electoral votes), Florida (29), and possibly Colorado (9) and Nevada (6)

Running total (in total Electoral College votes): Trump wins between 204 and 219

Step 2: Alienate the young 

Millennials, being both more numerous and less cynical than their Generation X predecessors, once seemed likely to usher in a substantial advantage for Democrats among young voters. Except then they stopped turning out. In 2012, the number of 18-to-29-year-old voters dropped by 1.8 million from the previous presidential election year. “It seems likely that the observed young-adult voting surge of 2004–2008 was temporary,” a U.S. Census study concluded, “and not representative of a permanent shift towards greater young-adult engagement in presidential elections.”
In 2016, this is likely to affect Clinton’s performance in several college-heavy states that have had relatively high turnout—and high Democratic voting rates—among those age groups. That includes swing states such as Iowa, North Carolina, New Hampshire and Virginia. Just look at what happened in the past cycle. In Iowa, for example, 18-to-29-year-olds dropped from 17 percent of the 2008 vote to 15 percent in 2012—and were less likely to vote for Obama. The result was a net loss of close to 30,000 votes, in a contest Romney lost by less than 90,000 votes.

And that was with Obama, who did far better than Clinton with young voters.

During primary season, Sanders took a stunning 84 percent of the under-30 vote in the Iowa caucuses. Many of these young Sanders voters may come around to support Clinton over Trump in the general—just as Clinton’s bitter supporters eventually came to support Obama in 2008. The question is, though: How many?

Maybe not enough. As Jennifer Duffy of the Cook Political Report notes, there is a big difference between 2008 and 2016: Then, Clinton’s so-called PUMA die-hards were mostly middle-age suburban women, with long-standing ties to the Democratic Party. In other words: likely voters. Most young Sanders voters, on the other hand, are not yet regular voters, and certainly not the kind of committed Democrats Clinton can count on; her campaign will need a significant get-out-the-vote effort to persuade them to show up in November. That will be more difficult the more she takes the conservative path, pivoting to the center for the general election, and focusing on messages geared toward her core—older—voters.

To see how much young-voter turnout matters, look at North Carolina. In 2008, Obama had a net advantage there among 18-to-29-year-olds of 368,000 votes—and eked out a 14,000-vote victory overall. In 2012, with dampened enthusiasm, Obama’s advantage in that age group dropped by 120,000, and Romney coasted to a 92,000-vote win. What’s more, those young voters are especially likely to be swayed to third-party alternatives, which—see below—could become more enticing this time around.

Trump wins: Georgia (16), North Carolina (15) and Iowa (6), with a chance at Virginia (13)

Running total: Trump wins between 241 and 269

Step 3: Let establishment Republicans find another place to go

The key to a Clinton landslide is turning red voters blue. And once it became clear that conservatives didn’t have the guts to put forward a #NeverTrump protest candidate, “Republicans for Hillary” was beginning to have a bit of a ring to it. But there’s another possibility: A plausible moderate-right candidate could emerge as a genuine alternative.
Take, for example, former Massachusetts Republican Governor Bill Weld, whose history with Clinton goes back decades. Weld is exactly the type of GOP moderate Clinton has reportedly been wooing as Trump edged closer to the nomination. He endorsed Obama in 2008. He compares Trump’s rhetoric to Nazism. And there are plenty of Republicans like him this year, particularly in the Northeast. Not too many years ago, you could imagine considerable appeal for Clinton among Weld-ian Republicans. But thanks to her primary run, a combination of issues—Sanders pushing her left on economic issues, her unsatisfactory answers on the email scandal—has made that a tougher sell.

And now, instead of endorsing Clinton, Weld has agreed to join Gary Johnson on a potential Libertarian Party ticket, which suddenly looks like a very friendly home for old-fashioned country-club Republicans. It’s not a stretch to imagine Weld’s friend and fellow former Massachusetts Republican Governor Romney, who is clearly disgusted with Trump, could endorse Johnson-Weld. Disgruntled Republican funders loath to back Trump—another demographic being wooed by Clinton—could follow.

Sure, the votes the Libertarian Party siphons off will be primarily those of Republicans. That’s likely to pad Clinton’s lead in some already blue states she doesn’t need to worry about, such as Massachusetts, New Jersey, and New York. However, the disaffected GOP voters it pulls away from Clinton are potentially critical for her in New Hampshire, Pennsylvania and perhaps even Maine, where third-party candidates have a considerable disruptive history. And in the event that, say, John Kasich were to endorse Johnson-Weld, even Ohio could suddenly get shaken up. Sure, that one is a long shot, but even without a big-name endorsement, there’s evidence that a plausible GOP alternative hurts Clinton more than Trump this year. A new Zogby poll of Ohio voters shows that in a four-way race including Johnson (pulling 6 percent) and Green Party nominee Jill Stein (3 percent), Clinton takes the bigger hit, and her lead in the state decreases.

We’ve seen this movie before: It was 1980, and independent candidate John Anderson pulled lots of votes in what we think of as blue states—15 percent in Massachusetts, 13 percent in New Hampshire, 11 percent in Colorado and Washington, 10 percent in Oregon and Maine. Anderson had been a Republican, but the votes he siphoned off would have gone more to Jimmy Carter than to Ronald Reagan—by 49 percent to 37 percent, according to exit polls. Every one of those states went for Reagan. In Maine, the margin was just 3 percentage points.

The emergence of a plausible moderate alternative also threatens to derail a larger piece of Democratic strategy: Clinton hoped to trot out aisle-crossing Republicans to boost her bipartisan credentials. Now, many Republican leaders will have a choice between Trump or Johnson, and not be forced to choose Clinton. Meanwhile, the other big constituency of anti-Trump Republicans—religious conservatives—has given up on finding its own protest candidate. Unlike establishment moderates, these Republicans would never go for Clinton—and will mostly end up boosting Trump’s numbers in states such as Iowa, Georgia, North Carolina and Florida. If Clinton loses in November, her supporters are going to be kicking themselves for not locking in moderate Republicans sooner.

In 2012, Obama won New Hampshire by 40,000 votes, or about 5 percent of the vote. In the Granite State, that was a relative landslide: The margin has been less than 2 percent in three of the past six presidential elections. Should a Libertarian ticket pull an Anderson-like 13 percent of the vote there—and should the bulk of those voters be Never-Trumpers, some of whom would otherwise have gone to Clinton—then the third party could tip a typically close Granite State contest to Trump.

Trump wins: New Hampshire (4), one district of Maine (1) and—if union households desert Clinton over trade (see below)—possibly Pennsylvania (20)

Running total: Trump wins between 246 and 294

Step 4: Fumble on trade

As soon as the votes were tallied in 2012, Richard Trumka of the AFL-CIO and Mary Kay Henry of SEIU were claiming unions had delivered Obama’s victory. They argued, with justification, that Ohio, Wisconsin and Nevada got into the blue column because of a massive turnout effort from labor. But earlier this year, both Trumka and Henry expressed concerns that Trump could flip that script.
“Our members are responding to Trump’s message,” Henry said in one interview. “Donald Trump is tapping into the very real and very understandable anger of working people,” Trumka said in a speech.

It’s not just that these workers are drawn to the raw emotion of Trump’s “you’ve been screwed” rhetoric. Polls show that union households tend to oppose free trade quite strongly. Sanders has made free trade a centerpiece of his primary campaign against Clinton. Trump, hoping to woo Sanders voters, frequently praises his position on that issue.

Union voters largely agree with Trump that trade deals—including those negotiated by Democratic Presidents Obama and Bill Clinton—have taken their jobs away. Hillary Clinton has yet to counter this attack in any meaningful way. Her history on trade has been careful and political, which has left her struggling to articulate a strong argument against Sanders, let alone Trump. She gave measured support at the time to her husband’s controversial NAFTA deal, but later called it a mistake; voted in favor of most but not all trade deals as senator; and flip-flopped unconvincingly on the Trans-Pacific Partnership this year.

It’s not hard to see how quickly this could start costing her Electoral College votes in the Rust Belt, where Trump hopes to improve on past Republican performance. (And where, you may remember, Clinton had to apologize for threatening to put coal companies out of business.) In Ohio, for example, 22 percent of 2012 voters came from union households, and 60 percent of them voted for Obama. In Wisconsin, a similar share of the electorate voted 2-to-1 for Obama over Romney. In 2016, both states went for Sanders over Clinton in their primaries. In Pennsylvania, where Trump is planning a major effort, union households provided Obama more than half his net margin.

Trump wins: Ohio (18) and Wisconsin (10), and maybe Michigan (16)

Running total: Trump wins between 274 and 338

The Final Tally

So there you have it. Trump survives a Latino surge in the South and West; Clinton fails to bring home young voters in the Southeast and Midwest; Libertarians give Trump a foothold in the Northeast; the Rust Belt puts the nail in the coffin—and with somewhere between 274 and 325 electoral votes, Donald J. Trump becomes the 45th president of the United States. Yes, the specifics could vary. But it’s clear Trump can cross the 270 electoral-vote threshold even on the low end, with plenty of cushion on the high end to make up for a state that slips through his fingers here or there.
For what it’s worth, it’s also possible Clinton wins in a landslide, as an increasingly unstable Trump shrinks deeper and deeper into racism, xenophobia and conspiracy theories. But what’s clear is that Democrats can no longer count on a lopsided race that even a problematic candidate running a clumsy campaign can’t lose.

Again: It’s a long way to November, and Trump could always self-destruct. But he probably won’t, and 2016 is shaping up as a contest that a careful Clinton campaign can easily lose, state by state, even as she piles up the popular vote in California and other sure-win places. Demographics are not destiny. In fact, they can be a disaster waiting to happen.