Trump and the Rust Belt

Can Donald Trump win the Great Lakes region, which is pretty much his only hope of winning the general election? Almost certainly not. The shift in white voters away from Obama would have to be gargantuan.

— In Michigan, where Romney beat Obama by 52-46 among white voters overall, Trump would have to win among them by 58-40, an improvement of 12 points.

— In Wisconsin, where Romney beat Obama by 50-49 among white voters overall, Trump would have to win among them by 54-45, an improvement over Romney of eight points.

— In Pennsylvania, where Romney beat Obama 54-44 among white voters overall, Trump would have to win among them by 58-40, an improvement of eight points.

— In Ohio, where Romney beat Obama by 56-42 among white voters overall, Trump would have to win among them by 58-40, an improvement of four points. (This seems doable, but again, this presumes Trump makes inroads among college educated whites and that the nonwhite spread remains the same.)

“It seems very unlikely that Trump can do so much better than Romney among whites and particularly working class whites in these states,” Teixeira tells me. “The swings are just too big.”

Now, Trump backers might argue that he will also drive up turnout among white voters relative to nonwhites, meaning he would not have to win among them by these margins to prevail. But here is where a demographic trap intrudes: All of the things that Trump might say and do to drive up white turnout — particularly working class white turnout — would also likely drive up nonwhite turnout. So there’s no reason to expect a major boost in turnout from one group and not the other, Teixeira says.

On top of this, each of these states has become less white since 2012. The white percentage of the electorate in these states is expected to fall by about 1 percent and the working-class white percentage by 2 percent. There’s almost no way Trump can win these states.