The Geography of Trump

By Neil Cosgrove

One local, hard-to-miss feature of the weeks leading up to the November 8th election was the proliferation of Hillary Clinton lawn signs within the city of Pittsburgh, and the equally striking plethora of Donald Trump signs the further from Pittsburgh one traveled. The rueful surprise is that lawn signs appear to have more accurately predicted the election results than did the majority of polls, or the musings of a self-appointed punditry.

The stark, geographically determined contrasts in support for the two major-party candidates may be the most striking symptom of the country’s condition in the fall of 2016. But that contrast represents a trend closely followed by researchers for years. As Dr. Mindy Fullilove observed at the recent Pittsburgh Housing Summit, the growth of so-called “landslide counties,” in which there is a 20% or more difference in voting in favor of either the Democratic or the Republican presidential candidate, has been steady since 1992, when such counties made up 38% of the total. In 2012, that figure had grown to 50%, and in 2016, it will no doubt be higher still.

In Philadelphia County, the figures were 82.4% for Clinton, 15.5% for Trump; in Butler County, two-thirds of voters preferred Trump, and only 29% voted for Clinton. Such massive pluralities suggest that within those counties it may be difficult to discover, through daily contact and casual conversation, why anyone would choose to vote for either Trump or Clinton. Stereotypes fill the void. Trump’s support leaves Pittsburghers mystified, and inclined to attribute his pluralities to atavistic views on race and gender, while people in Butler marvel at clueless big city-dwellers who seem ignorant of how recent public policy has negatively impacted their lives within hollowed-out small cities and towns.

Considering the exceptions that disprove the facile generalizations is a good way to begin chipping away at such distrust.  Did every “white working class” citizen vote for Trump?  No, even in the most rural of counties.  Are the minorities in that same working class who voted for Clinton animated by completely different concerns than their white brethren?  Probably not.   On the other hand, are the overwhelming number of Pennsylvania counties who gave landslides to Trump all alike? Comparing just two—Butler and Lawrence—suggests differently.

When I first came to western Pennsylvania in 1987, Butler County was as solidly Republican as it is today, predominantly populated by small-government as well as cultural conservatives. During the ensuing 30 years, Butler has steadily grown in population and prosperity. Its median household income of $59,365 is nearly $20,000 higher than Pittsburgh’s, and less than 9% of its citizens live below the poverty line. Kansas journalist Sarah Smarsh cited both extensive polling by Gallup and primary exit polls while arguing that “Trump voters were … more affluent than most Americans,” with a distinguishing “penchant for authoritarianism.”

Still, Smarsh’s analysis doesn’t explain Lawrence County, which went for Trump over Clinton by a margin of 62% to 34%, or the significant number of infrequent voters who appeared to give Trump victory in Pennsylvania, Wisconsin, North Carolina and Florida. In that year of 1987 we purchased a home in New Castle, Lawrence’s county seat, and a decidedly Democratic small city. Since that time, New Castle has lost more than 20% of its population, while the county itself is down more than 10%. Median household income is more than $15,000 less than in Butler and the poverty level is 5 and a half points higher.

In 1991, Shenango China ceased manufacturing and New Castle lost 275 jobs. Later that same year, Rockwell International decided to close its 530-employee axle plant, with much of the operation landing in Mexico. And starting around the turn of this century, Lawrence County voters began to noticeably abandon Democratic candidates, although to this day there are still 10,000 more registered Democrats than Republicans. In 2008, McCain beat Obama by slightly more than 2,000 votes; in 2012 the Republican margin had grown to over 3,500, and in 2016 Trump won by more than 11,000.

Despite the large registration gap, only 800 more Democrats than Republicans voted in last April’s primary, indicating that many who didn’t participate voted for Trump in November.

Lawrence County becoming a 2016 “landslide county” indicates both alienation from the Democrats and an embrace of someone quite different from the usual Republican presidential candidate. To acknowledge deep economic frustration as a motive does not deny the strong presence of racial animus, sexism, xenophobia and status anxiety. It does suggest Republicans in a place like Lawrence County may be as vulnerable as 1990s Democrats were, especially given the more than likely outcome that neither Trump’s nor the congressional majority’s policies will make life easier for its shrinking number of residents over the next two to four years.


Neil Cosgrove is a member of The NewPeople Editorial Collective and the Merton Center Board.

Photo taken from:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s