Every day, our journalism dismantles barriers and shines a light on the critical overlooked and under-reported issues important to all North Carolinians.
Before you go …
If you like what you are reading and believe in independent, nonprofit, nonpartisan journalism like ours—journalism the way it should be—please contribute to keep us going. Reporting like this isn’t free to produce and we cannot do this alone. Thank you!
Elections are much more complicated than just which candidate received the most votes, though that count does determine who claims victory. Voters may be individuals, but often tend to behave in blocs that can help in understanding their behavior in an election. During the 2016 presidential race in North Carolina, different candidates’ messages resonated with people who shared different things in common – party affiliation and ideological identity, religious and cultural values, ethnicity, socio-economic status, gender and sexual identity, or educational level.
But as Carolina Public Press analyzed the results in Western North Carolina precincts, one divide emerged as more important than any of the rest: urban vs. small town and rural. That finding comes in part because the most readily accessible data was for geography. Components of other things, such as wealth, education and race, will have some overlap with this divide based on population density. But the divide is so pronounced for this geographic divide in county after county that it’s difficult to discount.
Even when Democrats carried or competed closely in urban centers, Republicans dominated by wider margins in more thinly populated outlying areas of smaller towns and rural communities. Only the Asheville/Woodfin/Black Mountain and Boone Democratic votes were strong enough to outnumber the Republican votes in their overall counties.
Another trend that emerged was the strength of the Democratic vote in college areas, including around Appalachian State, Western Carolina, the University of North Carolina at Asheville and Brevard College. This may correlate to a combination of liberal professors, young millennial students and ethnically diverse student and staff populations. Or it could be seen as voters further removed from educational institutions rejecting “elite” political ideals.
An exception to the rural/urban divide came in two counties with large Native American populations in rural precincts, where Clinton either led or was very competitive.
To better understand what happened throughout the region, CPP has broken down the counties into three population groups:
Three counties of 75,000+ population: Buncombe, Henderson and Burke;
Eight counties of 25,000 to 75,000 population: Rutherford, Haywood, Watauga, McDowell, Jackson, Macon, Transylvania and Cherokee;
Eight counties of less than 25,000 population: Madison, Polk, Avery, Mitchell, Yancey, Swain, Clay and Graham.
Before examining the trends in each of these counties, an important caveat demands attention. The numbers available at this point for precinct-level analysis do not include early votes, mail-in absentees and provisional ballots. The totals for each county and the state do include early votes and mail-in absentees that arrived before Election Day. Late absentees and provisional ballots will be totaled later this week when the county boards of election canvass the results. It’s due to those ballots that several other statewide races, including governor, remain too close to call.
In some other recent previous elections, the early vote leaned heavily Democratic because emphasis on the early vote has been an important party strategy. For the same reason, in some locations the Election Day vote might skew Republican. As a result, the precinct-level discussions that follow cannot take the early vote into account.
It’s certainly possible that some precincts that currently appear to have gone narrowly for the Republican candidate actually went for the Democratic candidate, which will become clear some time later when the state and local election boards can assign those votes to precincts. Although it would run counter to recent election trends, it’s also equally possible that some precincts that narrowly went for the Democrat actually went for the Republican. So the analysis that follows is the based on the best information that’s available, but is necessarily incomplete.
Counties with 75,000+ population
In Buncombe County, Hillary Clinton claimed a 54 percent majority of votes, thanks primarily to heavily Democratic Asheville, the region’s largest city.
But this narrow majority was a strong showing without being the landslide she needed from urban counties to have been more competitive statewide. Clinton’s problem in Buncombe was that beyond Asheville, she carried few of the county’s small-towns and rural precincts.
Clinton did have an edge in the western portion of Black Mountain and the town of Woodfin. But elsewhere Donald Trump had strong support, including the incorporated towns of Montreat, Weaverville and Biltmore Forest. Unincorporated communities like Candler, Enka, Leicester, Swannanoa, Barnardsville, Arden, Fairview and Avery Creek all went for Trump.
The demographics of those places don’t tell a simple story. Some areas like Candler are home to many lower-income low-education white voters who were expected to back Trump. But Biltmore Forest, where the wealthy and highly educated voter pool didn’t match up as clearly with the supposed Trump demographic, also went solidly for the Republican candidate.
While the overall outcome was very different in Western North Carolina’s second-most populous county, Henderson, where Trump won more than 60 percent of the vote, the same divide between the Democratic core city and rural outlying areas applied.
Hendersonville’s precincts went for Clinton. But the rest of Henderson County was Trump country, from small towns like Laurel Park, Fletcher, Mills River, Flat Rock and Bat Cave, to the county’s many unincorporated communities.
Burke County followed precisely the same script. Trump took 67 percent of the vote, but Clinton carried the precincts in the largest city, Morganton. In all other small towns and rural areas of Burke, Trump was dominant.
Counties with 25,000 to 75,000 population
Eight of the 19 counties in Western North Carolina have populations between 25,000 and 75,000, based on population estimates the U.S. Census Bureau released in September.
With one exception, these counties were favorable to Trump and undoubtedly contributed significantly to his statewide edge in North Carolina.
The largest in this group is Rutherford County, which is geographically unique, transitioning from rolling Piedmont in the southeast around Spindale, Ellenboro and Forest City to high foothills near Rutherfordton, Sunshine and Bostic, then rugged mountainous terrain along the eastern slope of the Blue Ridge further northwest near Lake Lure and Chimney Rock State Park. This geographical diversity equates to economic and social diversity, with agricultural areas, mill towns and resort communities all having distinct personalities. Rutherford County is also relatively ethnically diverse for WNC, with a more substantial black portion of its population than in many other counties. Republicans are only slightly more numerous than Democratic registered voters, though many unaffiliated voters may lean Republican.
Even so, Trump enjoyed a landslide in the Rutherford vote, claiming 72 percent to just under 25 percent for Clinton. While Trump carried every precinct, his support was weakest around Spindale, at about 54 percent and the Lake Lure/Chimney Rock area at about 60 percent.
Haywood County has just under 60,000 residents in terrain that’s decidedly mountainous, with a mixed economy of industry, agriculture and tourism. Despite the lack of any substantial ethnic diversity, registered Democrats outnumber Republicans by a substantial amount in Haywood.
It didn’t matter in the presidential contest, as Trump carried Haywood with more than 60 percent of the vote. Clinton carried only the Central Waynesville precinct, at the heart of the county seat.
Watauga County, home to Appalachian State University, was a bright spot for Democrats in Western North Carolina, with Clinton edging past trump by a few hundred votes. But she only claimed a 47.2 percent total to Trump’s 45.7 percent in Watauga, doing little to make up his much stronger advantage through most of the region.
The urban/rural divide was as pronounced as ever in Watauga, with Clinton racking up modest victories in Boone precincts, but Trump scoring well in the less densely populated areas.
In McDowell County, Trump led Clinton by nearly 10,000 votes, comparable to President Obama’s statewide margin of victory in 2008. Clinton received less than a quarter of the vote in McDowell and failed to carry any precincts. She was strongest in one central Marion precinct where Clinton and third-party candidates combined to deny the Republican an outright majority, though he still led with a 30-vote plurality.
Jackson County, home to Western Carolina University, went for Trump, though he emerged with just under 53 percent of the vote, less than in most of the region’s other mid-sized counties. Unlike the rural-urban divide that appeared elsewhere, the educational center of Cullowhee was the strongest location for Clinton here, where she narrowly carried a majority of the precinct. While Trump led in other precincts, the outcome was closely contested in the county’s northernmost area, which has a substantial Native American population.
More than two-thirds of Macon County voters chose Trump, who carried every precinct in the county along the state’s Southern border with Georgia. Clinton’s percentages were best in the county seat of Franklin, but even then she received less than a third of any precinct’s votes.
Although Trump carried Transylvania County, like neighboring Jackson the result was closer than in many other WNC counties, with the president-elect claiming 59 percent of the votes to about 37 percent for Clinton. The Democratic nominee carried a majority in three precincts in the largest town, Brevard, home to Brevard College, but she trailed badly in most other parts of the county.
Trump was especially dominant in Cherokee County where he had nearly 77 percent of the vote, while Clinton claimed only about 20 percent. Trump’s support fell below 70 percent only in the county’s solitary appreciable population center, the small town of Murphy.
Counties below 25,000 population
In the region’s eight smallest counties, with virtually no sizable towns, the Trump vote was pervasive. These victories may be small because the populations don’t add up to much in comparison to the margins in bigger counties. But when Clinton carried a large county like Buncombe by 19,000 votes with a 54 percent to 40 percent age, the margins in these small counties where Trump generally won more than 60 to 78 percent of the vote, start to add up and surpass Clinton’s weaker margins in the areas where she was strongest.
Trump carried all precincts in Madison County with 60 percent of the countywide vote. However, he led Clinton by just a slim margin in the precinct around Hot Springs. The only other precinct where Clinton did better than 40 percent of the vote was in the largest town, Mars Hill, home to Mars Hill University.
Trump had nearly 62 percent of the vote in Polk County, on the South Carolina line. However, the town of Tryon went narrowly for Clinton.
Yancey County on the Tennessee line saw Trump claim 64 percent to just 32 percent for Clinton. However, the county’s southernmost precinct around the town of Burnsville gave Trump only a narrow majority of about 54 percent.
Just to the northeast in Avery County, Trump was much more dominant with about 76 percent of the vote, claiming more than 90 percent in several precincts along the state line. Only in three precincts near Newland did Clinton score better than 20 percent.
While the peak of Mt. Mitchell is actually in Yancey County, one of Trump’s highest points may have come in nearby Mitchell County where he garnered almost 78 percent of the vote versus less than 20 percent for Clinton. She performed moderately better in the southern portions of the county, but wasn’t competitive anywhere in Mitchell.
Swain County, home to the majority of the Eastern Band of Cherokee Indians, is unique among western counties as the only one with a substantial ethnic minority – Native Americans make up about 29 percent of the population. Although Trump won the county easily with about 58 percent of the vote, Clinton led in the eastern precinct around Cherokee, with a 48 percent plurality.
Clay County may not have many voters, but they overwhelmingly backed Trump, who scored about 74 percent of the vote to just under 23 percent for Clinton. Clinton’s best showing was in the precinct that includes the southern portion of Hayesville, but even there she could claim only about 27 percent of the vote.
Graham is the region’s smallest, but it was the most heavily lopsided in favor of any candidate, with Trump taking almost 79 percent of the vote to just 18 percent for Clinton, who did not break 20 percent in any precincts.
Some early analysis of the vote in WNC speculated that Libertarian candidate Gary Johnson might pick up some sizable vote totals from voters who had previously backed libertarian-leaning Republicans, like Kentucky Sen. Rand Paul. That didn’t happen. Johnson did relatively well in Watauga County with more than 4 percent of the vote and nearly that level of support in two counties with large Native American populations, Swain and Jackson. While still miniscule, that was well above the 2.7 percent support Johnson received statewide.
Green Party candidate Jill Stein didn’t make the ballot in North Carolina and was forced to run as a write-in, potentially blunting her ability to grab votes in places like Buncombe, which has been a haven of green activism in the Southeast.
The usual plethora of write-in protest votes also could be expected, with perpetual also-ran Mickey Mouse vying with Harambe the gorilla and others vying for throw-away votes in what was billed as one of the most consequential votes in U.S. history.
Why the West mattered
Most of North Carolina’s voters do not reside in the 19 counties that CPP defines as Western North Carolina, but in the other 81 counties, especially in the great Piedmont Crescent of densely populated counties that stretches along Interstate 85 from Charlotte to Greensboro to Raleigh.
Those populous areas mostly went blue, or bluish-purple. Rural suburban counties like Cabarrus, Alamance and Lee went for Trump, but not by the large margins seen in the West. Coastal areas, like the mountains, were red. The heavily African American counties in the northeastern corner of the state and along the central southern border remained their traditional blue.
Trump won the state with a roughly 177,000 vote advantage. About half of that margin came from the 19 WNC counties, with the rest coming from the rural interior counties or those along the coast.
The Republican candidate’s ability to mobilize pockets of rural voters into a winning statewide majority, as well as the Democratic candidate’s inability to excite sufficient numbers of urban voters, shaped the outcome in this election, one in which North Carolina was a pivotal state.
Whether one loves or hates the outcome, the mountains loomed large and refused to be forgotten.