Pennsylvania matters. The Trump and Clinton campaigns will spend millions in this state to nudge the needle a few percentage points one way or the other. But whatever they do, geography and demographics are pushing hard toward the left.
No Republican presidential candidate has carried the state since 1988, but there are three ways Donald Trump could hope to turn Pennsylvania red this year. He could make inroads in the core counties of metro Philadelphia and well-educated Allegheny County. He could make gains with a group other than working-class white men. Or he could elicit a historic surge in turnout for working-class whites.
He’ll certainly try. Without Pennsylvania, an emerging consensus says, Trump has no realistic path to the White House.
In 2012, a late surge in some polls for Mitt Romney gave hope to Republicans that they could win the state. Those votes failed to materialize in that election, and are unlikely to in 2016.
As in most other states, there's a large partisan difference between Pennsylvania’s major metropolitan area, Philadelphia, and the rest of the state. As urban areas have leaned more Democratic and rural areas have become more Republican, metro Philadelphia has become the Democrats' firewall in the state. The Democratic margin of victory in Philadelphia's five core counties -- Bucks, Chester, Delaware, Montgomery and Philadelphia -- has expanded since 1992. That year, Democrats carried the five counties by 304,000 votes. So far, the Democratic high-water mark for the metro area was in 2008, when Barack Obama carried it by 682,000 votes. That margin fell a bit in 2012, when he carried it by 615,000 votes.
The rest of the state has been a different story: Outside the Philadelphia area, Pennsylvania is a true swing state. Looking only at the counties outside the Philadelphia area, Democrats won in 1992 and 1996, but Republicans have won every election since. George W. Bush won the rest of the state by 355,000 votes in 2004, John McCain won it by 62,000 votes in 2008, and Romney expanded the margin back to 306,000 votes in 2012.
But Democrats have one other firewall in the state: the increasingly well-educated Pittsburgh area, home to Carnegie Mellon University and a thriving post-industrial economy. Roughly 45 percent of people over age 25 in Allegheny County have at least a bachelor's degree, compared with 35 percent for Pennsylvania as a whole. Starting after 1992, when Ross Perot did well in the county, Allegheny has produced a consistent margin for Democrats of 80,000 to 100,000 votes.
Between metro Philadelphia and Allegheny County, Republicans start with around a 400,000 vote deficit -- more votes than they are likely to scrounge up in the rest of the state. Using 2012 demographics and turnout, and holding the metro Philadelphia and Allegheny County totals the same, Trump would have to get over 61 percent of the vote in the rest of the state, about 5 points higher than Romney got in 2012 or Bush got in 2004.
Just looking at geography, maybe that 5 percentage points seems within the realm of possibility. But a look at the demographics changes that. As of the 2015 Census estimate, Pennsylvania's population is 77.4 percent white non-Hispanic, down from 78.6 percent in 2012. Roughly 29 percent of the adult white population in the state has at least a bachelor's degree. And let's assume an electorate that will be 47 percent male. Putting those three together gets us to an approximate share of the voters who will be white males without a college degree: about 25 percent.
Why does this matter? Because the Democratic advantage in Pennsylvania in 2012 was 5.4 percentage points. And because the only demographic group with which Trump is doing significantly better than Romney did is white men with no degree. So for Trump to win Pennsylvania, given current demographic trends, he probably needs to get more votes outside of metro Philadelphia and Allegheny County than Romney did, more than he can hope to get just from white men with no degree. To squeeze that margin out of 25 percent of the electorate, he needs working-class white men in Pennsylvania to move not just a 14 points towards Republicans, as indicated in current national polling, but closer to 25 points.
This doesn't even take into account other demographic headwinds working against Republicans. Since 2012, the five core counties of Philadelphia have added 40,000 people, while the rest of the state has lost 10,000. The white, non-Hispanic population has shrunk by 135,000 while the Hispanic or non-white population has grown by 165,000.
If he is to rely on his white, working-class male coalition, as appears to be the case, the only way Trump can carry Pennsylvania is if there's a surge in white voter turnout, a vindication of the "missing white voters" theory. However, as David Wasserman shows, the decline in white voter turnout since 1992 has mostly been in noncompetitive states. There's been virtually no drop in turnout in Pennsylvania. So they’re hardly missing.
Even if this election year brings out many more working-class white voters, it’s likely to motivate the rest of the electorate as well, eliminating any advantage for the Republican Party.
On or before Nov. 8, Trump is likely to come to the same conclusion that his Republican predecessors have: Pennsylvania is unattainable.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story:
Conor Sen at firstname.lastname@example.org
To contact the editor responsible for this story:
Philip Gray at email@example.com