Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Oct 1.
Published in final edited form as: J Rural Stud. 2017 Oct 1;55:227–236. doi: 10.1016/j.jrurstud.2017.08.010

More than a Rural Revolt: Landscapes of Despair and the 2016 Presidential Election

Shannon M Monnat 1, David L Brown 2
PMCID: PMC5734668  NIHMSID: NIHMS903943  PMID: 29269990

The role of the rural vote was a provocative storyline in the aftermath of the 2016 U.S. Presidential Election. In articles like Rural America and a Silent Majority Powered Trump to a Win (Whitaker 2016), America’s Front-Porch Revolt (Dreher 2016), and Revenge of the Rural Voter (Evich 2016), journalists argued that Trump was victorious due to rural Americans’ frustrations with political insiders after years of neglect of rural economic and social problems. To be sure, Donald Trump received a much larger share of the rural1 vote than Hillary Clinton (63.2 percent vs. 31.3%), and his vote share increased with increasing levels of rurality (Figure 1).2 But Trump’s rural advantage in the 2016 election did not signal a new trend. Republicans have long won larger rural vote shares, particularly in Appalachia, the Great Plains, and parts of the South (McKee and Teigen 2009; Scala and Johnson 2017). Moreover, rural voters account for only about 15 percent of the total U.S. population and a similar share of votes cast in 2016. Therefore, although Trump’s rural advantage certainly contributed to his victory, it was not sufficient to swing the election on its own or to support media rhetoric of a new “rural revolt”. Instead, Trump’s combined rural and small city over-performance (and Clinton’s underperformance), particularly in the Industrial Midwest, was key to Trump’s unanticipated victory.

Figure 1.

Figure 1

Open in a new tab

Percentage of Votes Received by Republican and Democrat Candidates in 2012 and 2016, by Rural-Urban Continuum

Nonetheless, recent media and public attention to rural and other often-ignored places offers an opportunity to more fully consider the role economic and social conditions play on electoral politics in rural and small city America. In this essay, we summarize data from the 2016 Presidential election, identify the regions of the country and the characteristics of places where Trump performed much better than expected, present some in-depth examples of counties that are representative of the patterns we describe, and discuss opportunities for future social science research in this area. As we explain below, the Industrial Midwest was the difference-maker in the 2016 election. Accordingly, we focus on that region, but we also discuss trends more generally and draw attention to other areas of the country where Trump performed particularly well.

Election Numbers: The Predicable and The Unexpected

Hillary Clinton won the U.S. popular vote by nearly 2.9 million votes. Not only did Trump lose the national popular vote, but he also underperformed relative to Mitt Romney nationwide, receiving 45.9 percent of votes in 2016 compared to Mitt Romney’s 47.1 percent in 2012. But, because U.S. presidential elections are based on the Electoral College, some states are more important than others to the outcome (McKee and Teigen 2009; Warf 2009). Therefore, small advantages in key places enabled Trump to accumulate the set of electors needed to claim victory in 2016. Comparing 2012 and 2016 Republican vs. Democrat vote shares illustrates how these instrumental places affected the 2016 election (Figures 2a and 2b). But before discussing the key difference-makers, we briefly discuss some of the trends that were consistent with previous elections.

Figure 2.

Figure 2

Open in a new tab

a. County Vote Shares Received by Obama and Romney, 2012

b. County Vote Shares Received by Clinton and Trump, 2016

Business as Usual

Like Mitt Romney in 2012, Donald Trump garnered large vote shares throughout Appalachia, the rural South, Great Plains, and Mountain West. The Republican stronghold in those areas is not new (McKee and Teigen 2009; Morrill et al. 2011; Scala and Johnson 2017). Indeed, the rural preference for Republicans is consistent with research on rural-urban attitudes toward major economic and social issues, including jobs, immigration, abortion, same sex marriage, gun control, and climate change (Dillon & Savage 2006; Morin 2016; Scala and Johnson 2017). In their recent research, Scala and Johnson (2017) show a consistent gradient of social conservatism on a wide range of social and economic issues moving from large urban cores to the most rural counties.

Of course, there are distinct regional and racial/ethnic differences in rural conservatism and voting patterns (McKee and Teigen 2009), with southern whites being substantially more conservative on social issues. Since over four in ten rural persons live in the South, the concentration of evangelical conservatives in this region has a strong influence on Republican victories. However, the south is also among the most racially diverse regions of the country. Fifty-five percent of all U.S. blacks and virtually all rural blacks live in the South (U.S. Census Bureau 2015). In five deep Southern states: Alabama, Georgia, Mississippi, South Carolina, and Louisiana, Blacks comprise over a quarter of the population, and a larger share of the rural population. Although Democrats consistently win the Black southern vote, including the Black rural vote (Morrill et al. 2011; Scala and Johnson 2017), the concentration of Black voters in particular districts through gerrymandering and other voting constraints like strict voter identification laws systematically diminish Black political power at the state level (Soffen 2016). Moreover, despite the growing Hispanic population in southern states like Georgia, South and North Carolina, and Tennessee (Lichter 2012), large shares of Hispanics in these new destination states are not U.S. citizens (Monnat 2017), and therefore, are not eligible to vote. Hence, in 2016, the “solid South” was solid for Trump.

The Big Blue Wall Came Tumbling Down

What was unexpected was how well Trump performed, and conversely how poorly Hillary Clinton performed, in the Industrial Midwest3. Collectively in the Industrial Midwest, Trump over-performed Romney by 1.4 points, whereas Clinton underperformed Obama by 5.5 points.4 Comparing the 2012 vs. 2016 vote shares in the Industrial Midwest makes clear that it was key to Trump’s victory. Ultimately, Trump’s win came down to a difference of 77,744 votes spread across three states: Michigan (margin of 10,704), Pennsylvania (margin of 44,292), and Wisconsin (margin of 22,748), though he also garnered substantially larger vote shares than Romney in the other Industrial Midwestern states (Ohio, Illinois, Indiana), Appalachia, parts of New England, upstate New York, Minnesota, and Iowa.

Nationally, but especially in these regions, it was not only that Trump won more counties than Romney, but also that Hillary Clinton received far fewer votes and smaller vote shares than Obama, even in counties she won. In many ways, the trends signify a story that is more about Clinton underperformance than Trump over-performance. For example, of the 55 counties Clinton won in the Industrial Midwest, she performed worse than Obama (received a smaller share of votes) in 84 percent of them. Although the Industrial Midwest is home to just 16.2 percent of U.S. counties, 30.6 percent of the 206 “pivot counties” (counties that went for Trump after going for Obama in both 2008 and 2012), are in the Industrial Midwest.5 In nearly all pivot counties, Obama’s victory margin declined between 2008 and 2012, perhaps foreshadowing their shift to a Republican candidate in 2016.6 Importantly, Trump’s advantage in the Industrial Midwest was not confined to rural counties, but included small urban counties like Montgomery County, OH (Dayton) and Luzerne County, PA (Wilkes-Barre) and even larger urban places like Macomb County, MI (located in the Detroit metropolitan area).

Landscapes of Despair

To understand the electoral shift in these and similar “types” of places outside of the Industrial Midwest, it is important to understand the major economic, social, and health declines that have transpired in them over the past three decades. In many of the rural areas and small cities where Trump performed better than expected (or Clinton performed worse than expected), economic distress has been building and social conditions have been breaking down for decades (Monnat 2016).

The places with the largest shifts in 2016 are not all among the poorest places in America (though Appalachia certainly holds that distinction), but they are places that are generally worse off today than they were a generation or two ago. In these places there are now far fewer of the manufacturing and natural resource industry jobs that once provided solid livable wages and benefits to those without a college degree (Monnat 2016). To be sure, deindustrialization is not a new phenomenon in the U.S., but its impacts have been spatially uneven. Manufacturing began to decline as a share of all employment, particularly in rural areas, in the 1970s (Fuguitt, Brown and Beale 1989). In Pennsylvania, the number of workers employed in manufacturing has declined since 1980 in all but one county. Since 1980, median household income has declined in a greater percentage of counties in the Industrial Midwest than in any other region of the country. Nearly 57 percent of counties in the Industrial Midwest, including 69 percent of counties in Michigan and 82 percent of counties in Ohio, have lower median household income today than in 1980.

Although an in-depth analysis of all of the factors that drove the 2016 election outcome is beyond the scope of this essay, it is illustrative to identify some of the county-level characteristics associated with Trump’s over-performance (Figure 3). We define over-performance as the percentage difference in the share of votes received by Trump in 2016 vs. Romney in 2012. We examine over-performance rather than overall vote share because Republicans tend to receive larger vote shares in more economically distressed and sicker places. Accordingly, the question at hand is: what factors predict larger shares of votes going to the Republican candidate above some baseline level (i.e., 2012)?

Figure 3. Mean Trump Over-performance by County Well-Being.

Figure 3

Open in a new tab

Note: N=3,105 U.S. counties

Q1=bottom 25th percentile (highest well-being); Q4=top 25th percentile (worst well-being) Economic distress is a factor-weighted index combining the 2010–14 ACS measures of percent poverty (age 18–64), percentage age 25 to 54 unemployed or not in the labor force, percentage of households with public assistance income, percentage of households within Supplemental Security Income, percentage of families with children that are headed by a single parent, and the percentage of adults age 18–64 without health insurance (alpha=.828).

Poor health is a factor-weighted index that combines the percentages of adults with poor/fair self-rated health (2014), who are obese (2012), current smokers (2014), and with a disability that limits daily activities (2010–14).

Nationally, and across all regions, Trump’s average over-performance was higher in more economically-, socially-, and health-distressed counties. Specifically, Trump performed better in counties with more economic distress, worse health, higher drug, alcohol and suicide mortality rates, lower educational attainment, and higher marital separation/divorce rates. The data in Figure 4 show average Trump over-performance across quartiles of county well-being in the Industrial Midwest, demonstrating that the relationship between place-level well-being and Trump’s over-performance holds for both rural and urban counties. In Figure 5, we show unadjusted relationships between Trump’s over-performance and specific county-level characteristics in the Industrial Midwest7. For illustration purposes, we present the difference in Trump’s average over-performance between counties in the top 25th percentile (Quartile 4) and counties in the bottom 25th percentile (Quartile 1) of each indicator. This allows us to compare and rank the magnitude of each factor. The Industrial Midwest does not have the highest poverty and unemployment rates or the worst health in the U.S., but variation in these conditions within the Industrial Midwest does help to explain Trump’s performance in that region. Although the percentage of residents without a 4-year college degree had the strongest association with Trump over-performance, indicators of despair also help to explain his success there. In particular, economic distress (rates of SSI receipt, poverty, unemployment/not in labor force, uninsured), health distress (rates of disability, obesity, poor/fair self-rated health, smoking, and drug, alcohol, and suicide mortality), and social distress (rates of separation/divorce, single parent families, vacant housing units, persistent population loss) were strong predictors of Trump over-performance. These relationships held even when controlling for metropolitan status.

Figure 4. Mean Trump Over-performance by Measures of County Well-Being, Industrial Midwest.

Figure 4

Open in a new tab

Note: N=504 counties

DAS Mortality =Drug, alcohol, and suicide mortality rate

Urban=USDA ERS RUCC codes 1–3; Rural= USDA ERS RUCC codes 4–9

Q1=bottom 25th percentile (highest well-being); Q4=top 25th percentile (worst well-being)

Figure 5. Model-Estimated Difference in Mean Trump Over-performance by County Characteristic, Industrial Midwest.

Figure 5

Open in a new tab

Note: N=504 counties

Model-estimated differences are from bivariate linear regression models. The bars represent the difference in mean Trump over-performance between counties in the bottom 25th percentile (Q1) vs. top 25th percentile (Q4) for all characteristics except nonmetro county and persistent population loss (both dichotomous). All estimates are from unadjusted models, but all models use clustered standard errors to account for nesting of counties within states.

To be sure, this is not an exhaustive list of factors that likely influenced the election, and many of these factors are strongly correlated, making it difficult to disentangle those with the strongest influence. But these findings make it difficult to argue that localized economic distress was not a salient contributor to the 2016 Presidential election outcome. Of course, this is a place-level analysis so we cannot say who voted for Trump with these data. What remains murky is whether economically-distressed residents themselves supported Trump (or simply did not come out for Clinton) or whether Trump amassed his support among less-distressed community residents who were anxious about and frustrated with the poverty, joblessness, and illness they saw around them. Ultimately, what these descriptive findings suggest is that Trump performed well within these ‘landscapes of despair.’

Below we more fully describe the deteriorating economic, health, and social conditions in a handful of counties that are representative of the findings discussed above.

Luzerne County, Pennsylvania

Trump won Pennsylvania by just over 44,000 votes out of 6.1 million votes cast. Nearly 60% of Trump’s victory margin (over 26,000 votes) in Pennsylvania came from Luzerne County – a mid-size metropolitan county with many outlying rural areas. His 19.3 point victory margin in Luzerne County is remarkable because Luzerne County is a traditionally Democratic stronghold that Obama carried twice (by 8.4 points in 2008 and 4.8 points in 2012), and had not gone for a Republican candidate since 1988. The number of manufacturing jobs there declined from over 42,000 in 1980 to fewer than 19,000 today, and the types of manufacturing jobs there pay less than jobs that were available 40 years ago. The poverty rate has been increasing, median household income has remained stagnant since 1980, over a quarter of prime-age (25–59) residents are either unemployed or not in the labor force at all (U.S. Census Bureau 2015), there is chronic young adult outmigration (Winkler et al. 2013), and over the past 15 years, drug-overdose rates have tripled and suicides have more than doubled (U.S. Centers for Disease Control and Prevention 2015). The Robert Wood Johnson Foundation County Health Rankings (2017) place Luzerne County at 62 out of 67 counties in Pennsylvania.

Macomb County, Michigan

Barack Obama won Macomb County by 8.6 points in 2008 and 4 points in 2012. But in 2016, Trump bested Clinton by 11.6 points (53.6 vs. 42.0). When considering that Trump won Michigan by only 10,704 votes, his victory margin of 48,348 votes in Macomb County is noteworthy. With its population of nearly 850,000, Macomb County has the distinction of being the third largest swing county (by population size) in the U.S. Suffolk County, New York and Pinellas County, Florida are first and second. Located just north of Detroit, Macomb County has been especially hard hit by industrial restructuring. Although it has far from the highest poverty rate in the U.S., or even in Michigan, median household income has declined by nearly $20,000 and manufacturing jobs have declined by 26% in Macomb County since 1980. The prime-age adult (age 25–54) poverty rate, unemployed/not in labor force rate, accidental drug overdose rate, and suicide rate have all increased over the past two decades, as has the percentage of residents who are separated or divorced (U.S. Census Bureau 2015; U.S. Centers for Disease Control and Prevention 2015).

Scioto and Trumball Counties, Ohio

From once-vibrant manufacturing cities to former coal country, labor market devastation has been the norm throughout rural and small city Ohio for over four decades (Quinones 2015; Alexander 2017; Semuels 2017), and its downward spiral only intensified during the Great Recession and weak recovery. In his book Dreamland: the True Tale of America’s Opiate Epidemic, Sam Quinones (2015) described Scioto County’s city of Portsmouth as a place previously known for making things, with a once-thriving manufacturing base anchored by shoe, steel, brickyard, atomic energy, and soda factories. By the 1990s, those factories were long gone, replaced by big-box stores, check-cashing and rent-to-own services, pawn shops, and scrap yards. In today’s Scioto County, incomes are lower than in the 1980s, and poverty, disability, and unemployment rates are high (U.S. Census Bureau 2015). Portsmouth now has the distinction of being the “pill mill capital of America,” with widespread generational opioid addiction, and where the emptying out of factories was followed by the emptying out of people and hope (Arnade 2017). Scioto County tends to go Republican in Presidential elections, with a divergence from that trend only for Bill Clinton in 1992 and 1996. Trump also won Scioto County, but he received a remarkable 33 percent more of the county’s vote share than Romney received in 2012.

Meanwhile, northern Ohio’s Trumball County went squarely for Obama in both 2008 and 2012 (by 22.4 and 23 points, respectively). In 2016, Trump received nearly 35 percent more of the county’s vote share than Romney and beat Clinton by 6.2 points. Since 1980, inflation-adjusted median household income has declined by over 27 percent, manufacturing jobs have declined by 58 percent, and extraction industry jobs have declined by 57 percent. The county suffers from chronic outmigration, the working-age poverty rate has nearly doubled just since 2000, and 27 percent of prime-age residents are unemployed or not in the labor force at all (U.S. Census Bureau 2015). Remarkably, Trumball County has an even higher drug overdose rate than Scioto County, and has the second highest suicide rate in the state (U.S. Centers for Disease Control and Prevention 2015). In places like Scioto and Trumball counties, it is not hyperbolic to suggest that despair is the new normal.

Stories like these unfold across small cities and towns all throughout the Industrial Midwest, Appalachia, upstate New York, and New England. When jobs disappear, more than paychecks go with them. Lay-offs and corporate shut-downs tend to be followed by a declining tax base, resource-disinvestment, and social disintegration (Alexander 2017; Quinones 2015; Sherman 2009). In the year preceding the 2016 election, Wall Street trader-turned-photographer/journalist Chris Arnade traveled to many of these frequently-ignored parts of the U.S. to talk with Trump supporters about their frustrations, fears, and anxieties (Arnade 2016). These places have borne the brunt of declines in manufacturing, mining, and related industries and are now struggling with opioids, disability, and declining health. He heard from folks in communities facing economic and social destruction, just like Katherine Cramer (2016) describes in her research on rural voters in Wisconsin. In these places, good jobs and the dignity of work have been replaced by suffering, hopelessness and despair; the feeling that America is not so great anymore, and the belief that people in power do not care about them or their communities.

Work has historically been about more than a paycheck in the U.S. American identities are wrapped up in our jobs. But the U.S. working-class (people without a college degree, people who work in blue-collar jobs) regularly receive the message that their work is not important and that they are irrelevant and disposable. That message is delivered through stagnant wages, declining health and retirement benefits, government safety-net programs for which they do not qualify but for which they pay taxes, and the seemingly ubiquitous message (mostly from Democrats) that success means graduating from college. Of course, these frustrations are not new, nor are they unique to the U.S.; scholars have long noted disaffection with political elites, perceived threats to a “place-rooted way of life” (Woods et al. 2012:567), and attraction to populist politics among some rural residents (especially among whites) in both the U.S. (Dyer 1998; Isenberg 2017; Kimmel and Ferber 2000; Stock 1997) and abroad (Alba and Foner, forthcoming; Pritchard and McManus 2000; Strijker et al. 2015; Woods et al. 2012). One interpretation is that Trump capitalized on and exploited white working-class frustrations and anxieties in these ‘landscapes of despair.’ His anti-free trade message for example, likely resonated with some voters who saw manufacturing plants shut down and saw low-wage service jobs replace the better paying jobs previously available to their parents and grandparents. Growing racial and ethnic diversity in these same places may have contributed to the perception among working-class whites (however inaccurate) that immigration was at least partly to blame for their woes (Alba and Foner, forthcoming).

But another equally-likely interpretation is that apathy, dislike, or ambivalence about Hillary Clinton in these places, paired with Democrats’ arrogance about the impermeable “big blue wall”, may have led some potential voters who would have supported a different Democratic candidate who emphasized working-class issues to not vote at all. When you’re driving by shuttered factories with boarded up windows, watching nightly news reports about drug overdoses, and seeing more of your neighbors sign up for disability instead of working, the message that “America is great already” simply does not jibe with your own reality.

So what makes these downwardly-mobile places any more likely to swing to Trump than persistently poor and disadvantaged places? After all, persistently disadvantaged and poor racial/ethnic minority communities certainly suffer from frustration, distress, anxiety, and abandonment by political elites, and poverty and unemployment rates are higher in many of these communities than in those that swung to Trump. Certainly, Trump’s racial identity politics played a role. Thanks to both his overtly and implicitly racist messages, Trump was unlikely to garner any significant support from Hispanic or Black voters regardless of long-term economic distress in many majority-Black and majority-Hispanic areas. Reference group theory, developed 60 years ago offers an additional explanation. In a New York Times op-ed, sociologist Andrew Cherlin (2016) applied reference group theory to rising mortality rates among non-Hispanic whites. Perhaps the key to understanding rising white mortality rates in the context of declining mortality for other groups, and likewise the key to understanding Trump’s strong performance/Clinton’s poor performance in many downwardly-mobile places, is considering the reference group by which people (or places) compares themselves. Cherlin makes the case that working-class whites are comparing themselves to a prior generation that had more opportunities, whereas Blacks and Hispanics are comparing themselves to a generation that had fewer opportunities. Likewise, we contend that reference group theory can be used to explain community-level behavior. Residents of once-robust manufacturing and natural resource extraction towns see fewer good employment opportunities for people without a college degree, and out-migration of their best and brightest. Here, race and class interact in important ways. After 40+ years of deindustrialization, automation, globalization, and neoliberal policy regimes (Brown and Swanson 2003; Smith and Tickamyer 2011), the residents of these mostly-white and formerly-strong manufacturing and extraction towns and the rural areas surrounding them feel like they’re doing much worse than previous generations in the same places. The Great Recession and its spatially-uneven recovery only exacerbated these long-term stressors (Bailey et al. 2014). Alba and Foner (forthcoming) find similar trends in Western Europe, noting for example, that the most anti-immigrant places are also places that have experienced significant economic declines and job losses in recent decades. As Arlie Hochschild (2016:221) describes in Strangers in their Own Land, “I see that the scene had been set for Trump’s rise, like kindling before a match is lit. Since 1980, virtually all those I talked with felt on shaky economic ground. It was a story of unfairness and anxiety, stagnation and slippage – a story in which shame was the companion to need.” Just as decades of declines in secure and livable wage jobs, resource-disinvestment, and social decay made some places in the U.S. more vulnerable to the opioid scourge than others, these same forces made some places more vulnerable to quick-fix populist messages from Donald Trump.

Conclusion

That Trump did well in many rural areas of the U.S. is not surprising. Republicans have done well in the rural South, Great Plains, and Mountain West for decades. However, the strong rural and small city vote in the Industrial Midwest signaled an unexpected shift that profoundly affected the election. To be sure, no single factor can explain Trump’s victory, but place-level despair (as defined by several economic, social, and health indicators) appears to have played a major role. Trump’s populist message may have been attractive to many long-term Democratic voters (and previous election abstainers) in these places who felt abandoned by a Democratic party that has failed to articulate a strong pro-working class message, whose agendas often emphasize policies and programs to help the poor at what seems like the expense of the working-class, and who evidently believed it did not have to work very hard to earn votes from behind the “big blue wall.”

Although our analyses show that Trump’s election victory was not primarily due to a new “rural revolt,” the media’s current emphasis on the political impact of rural people and places offers several opportunities for social scientists, especially those with a spatial orientation. First, we encourage researchers to consider the role of growing demographic and economic diversity on politics in rural America. Rural does not automatically equate to white. Racial/ethnic minorities comprise about 20% of the U.S. rural population. The rural Hispanic population in particular is geographically dispersed and is expected to continue growing. This is likely to have implications for future U.S. election outcomes, just as it is now having in Western Europe (Alba and Foner, forthcoming). Rural also does not automatically equate to farming. Although U.S. Republican candidates consistently do well in farming-dependent rural counties, Democrats do well in rural counties dominated by recreation, amenities, and services (Scala and Johnson 2015). These trends suggest potentially growing enclaves of rural democratic support in future elections.

This is also an ideal time to consider the multiple important intersections and interdependencies between rural and urban areas. The rural-urban binary is an outdated concept that never really supported an accurate analysis of the spatial organization of American society. Hence, examining rural vs. urban voting patterns is not a particularly useful way to understand the nation’s changing politics. In the future, we recommend that analyses focus on the urban-rural interface. Rather than a boundary separating rural from urban, the interface is a space of intense social, economic, and environmental relationships between urban, suburban, and rural communities (Lichter and Brown 2011; Brown and Shucksmith 2017; Lichter and Ziliak 2017). An increasing share of the nation’s population and economic activity is located at the interface. Hence, the interrelationships linking urban, suburban, and rural communities will be embedded in future political structures. For example, consider the future of Congressional representation. Will the results of the 2020 Census contribute to Congressional redistricting that reflects how Americans really live in our increasingly integrated society, or will they continue to be used to pack Democratic voters into urban districts, while ceding rural and small city districts to the Republicans? This, of course, is a very high stakes game because it imparts a rural bias on the Electoral College. The Electoral College’s rural bias was intentionally built into the nation’s political system by Jefferson and Madison who feared the concentration of power in urban areas, and preferred an agrarian development trajectory (Badger 2016). Since over 90% of the U.S. population lived in rural areas in the immediate post-revolutionary period, the rural bias might have been more legitimate at that time. Today, however, with 85 percent of the U.S. population living in urban areas, the rural bias is questionable. As Emily Badger (2016:2) has observed, “The Electoral College is just one example of how an increasingly urban country has inherited the political structures of a rural past.” Since governors play a central role in redistricting, future gubernatorial elections will have the knock-on effect of influencing the Electoral College, and hence, the 2020 and 2024 presidential elections.

In conclusion, place matters in politics, and it clearly mattered in the 2016 U.S. presidential election. But rural, suburban, or urban residence per se was not necessarily the causal factor. Rather, the disproportionate distribution of adverse economic, health, and social conditions in some rural towns and small cities is an important key to understanding the 2016 election results.

Annex: Data and Methods

Analyses included 3,112 contiguous U.S. counties. Bedford City, VA and all counties in Alaska were excluded due to the lack of election data for those counties. County-level election data are from Dave Leip’s Atlas of U.S. Presidential Elections (Leip 2017a, 2017b). County-level demographic and economic data came from the 2010–14 American Community Survey (U.S. Census Bureau 2015). Age-adjusted pooled drug, alcohol, and suicide mortality rates (2006–2015) came from the U.S. Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (WONDER) Multiple Cause of Death Files (U.S. Centers for Disease Control and Prevention 2015). For specific causes of death included in the mortality rates, contact the first author. Other county-level health measures came from the Robert Wood Johnson Foundation County Health Rankings for 2017.

The Industrial Midwest region includes Illinois, Indiana, Michigan, Ohio, Pennsylvania, and Wisconsin. Rural-urban continuum categories are defined as follows: Urban Core (counties that host a single city with a population of at least 250,000 in 2010), Other Large Urban (counties with an urban influence code=1 with county population of at least 100,000 people in 2010); Small Urban (all counties with an urban influence code=2 and counties with an urban influence code = 1 with county population <100,000 people in 2010); Nonmetro Adjacent to Metro (counties with rural-urban continuum codes of 4, 6, and 8); Large Nonmetro not Adjacent to Metro (counties with rural-urban continuum codes of 5 and 7); and Remote Rural (counties with rural-urban continuum code of 9).

The economic distress index is a factor-weighted index combining the 2010–14 ACS measures of percent poverty (age 18–64), percentage unemployed or not in the labor force (age 25–54), percentage of households with public assistance income, percentage of households with Supplemental Security Income, percentage of families with children that are headed by a single parent, and the percentage of adults without health insurance (age 18–64) (alpha=.828). The poor health index is a factor-weighted index that combines the percentage of adults (age 18+) with poor/fair self-rated health (2014), percentage who are obese (2012), percentage current smokers (2014), and percentage with a disability that limits daily activities (2010–14).

Figures 3–5 present results from county-level analyses. Although counties are not presidential election units, they are political and economic units that administer health, social, and economic services to their residents. Their boundaries are also stable, allowing for comparisons across elections, whereas electoral district boundaries sometimes change dramatically thanks to both population change and gerrymandering. Model-estimated differences in mean Trump over-performance presented in Figure 5 come from linear regression models with clustered standard errors (to account for clustering of counties within states). The models are unadjusted, but the general findings hold when controlling for metropolitan status. Analyses are unweighted.

Contact the first author for specific questions on data, variables, and analyses reported in this paper.

Highlights.

  • Trump’s rural advantage contributed to his 2016 Presidential election victory, but it was not sufficient to swing the election.

  • Nationally, and especially in the Industrial Midwest, Trump’s average over-performance was higher in more economically-, socially-, and health-distressed counties.

  • Trump performed better in counties with more economic distress, worse health, higher drug, alcohol and suicide mortality rates, lower educational attainment, and higher marital separation/divorce rates.

  • Localized economic distress was a salient contributor to the 2016 Presidential election outcome.

Acknowledgments

Both authors acknowledge support from the USDA Western Association of Agricultural Experiment Station Directors Multi-state Research Project : W-3001: “The Great Recession, Its Aftermath, and Patterns of Rural and Small Town Demographic Change.” Monnat acknowledges support from the Institute on New Economic Thinking.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Rural is defined as counties with USDA ERS rural-urban continuum codes 4–9 (USDA 2015).

2

Moreover, this pattern held across all regions of the U.S. Data are available on request from first author.

3

We define the Industrial Midwest as Illinois, Indiana, Michigan, Ohio, Pennsylvania, and Wisconsin

4

Third-party candidates received 4.1 points more of the vote share in 2016 than in 2012. Data are available from first author upon request.

5

A list of “pivot counties is available from the first author upon request.

6

Interestingly, Bill Clinton won all but six of these counties in 1992 and all but one in 1996.

7

For details about these variables and the regression models, see the Data and Methods Annex. R2 values from the regression models are available from the lead author upon request.

A previous version of this paper was presented at the 2017 Trento Economics Festival and at Hunter College, CUNY.

Contributor Information

Shannon M. Monnat, Lerner Chair of Public Health Promotion and Associate Professor of Sociology, Maxwell School of Citizenship and Public Affairs, Syracuse University, 426 Eggers Hall, Syracuse, NY 13244, smmonnat@maxwell.syr.edu

David L. Brown, Emeritus Professor of Development Sociology, Cornell University, 251B Warren Hall, Ithaca, NY 14850, dlb17@cornell.edu

References

  1. Alba Richard, Foner Nancy. Forthcoming. "Immigration and the Geography of Polarization". City & Community [Google Scholar]
  2. Alexander Brian. Glass House: The 1% Economy and the Shattering of an All-American Town. St. Martin’s Press; 2017. [Google Scholar]
  3. Arnade Chris. What I learned after 100,000 miles on the road talking to Trump supporters. [Accessed Nov 4, 2016];The Guardian. 2016 Nov 3; 2017. via https://www.theguardian.com/society/2016/nov/03/trump-supporters-us-elections.
  4. Arnade Chris. ’The Pill Mill of America’: Where Drugs Mean There are no Good Choices, Only Less Awful Ones. [Accessed May 19, 2017];The Guardian. 2017 May 17; 2017. via https://www.theguardian.com/society/2017/may/17/drugs-opiod-addiction-epidemic-portsmoutho-hio.
  5. Badger Emily. As American as Apple Pie? The Rural Vote’s Disproportionate Slice of Power. New York Times. 2016 Nov 20; Accessed via: https://www.nytimes.com/2016/11/21/upshot/as-american-asapple-pie-the-rural-votes-disproportionate-slice-of-power.html?_r=0.
  6. Bailey Conner, Jensen Leif, Ransom Elizabeth. Rural America in a Globalizing World: Problems and Prospects for the 2010s. Morgantown, WV: West Virginia University Press; 2014. [Google Scholar]
  7. Brown David L, Shucksmith Mark. Reconsidering Territorial Governance to Account for Enhanced Rural-Urban Interdependence in America. Annals of the American Academy of Social and Political Sciences. 2017;672(1):282–301. [Google Scholar]
  8. Brown David L, Swanson Louis E. Challenges for Rural America in the Twenty-First Century. University Park, PA: Pennsylvania State University Press; 2003. [Google Scholar]
  9. Cherlin Andrew. Why Are White Deaths Rising? The New York Times. 2016 Feb 22; Access via: https://www.nytimes.com/2016/02/22/opinion/why-are-white-death-rates-rising.html?_r=0.
  10. Cramer Katherine J. The Politics of Resentment: Rural Consciousness in Wisconsin and the Rise of Scott Walker. Chicago: The University of Chicago Press; 2016. [Google Scholar]
  11. Dillon Michele, Savage Sarah. Issue Brief No. 1. Carsey Institute, University of New Hampshire; 2006. Values and Religion in Rural America: Attitudes Toward Abortion and Same-Sex Relations. [Google Scholar]
  12. Dreher Rod. US Election 2016: America’s Front Porch Revolt. BBC News. 2016 Nov 9; 2016. [Google Scholar]
  13. Dyer Joel. Harvest of Rage: Why Oklahoma City is only the Beginning. Basic Books; 1998. [Google Scholar]
  14. Evich Helena Bottemiller. Revenge of the Rural Voter. Politico. 2016 Nov 13; 2016. [Google Scholar]
  15. Fuguitt Glenn V, Brown David L, Beale Calvin L. Rural and Small Town America. New York: Russell Sage: 1989. [Google Scholar]
  16. Hanrahan Catherine. Trump Won US Election Thanks to the Rural Vote. ABC News. 2016 Nov 10; Accessed via: http://www.abc.net.au/news/2016-11-10/donald-trump-us-election-rural-vote/8012874.
  17. Hochschild Arlie Russell. Strangers in their Own Land: Anger and Mourning on the American Right. The New Press; 2016. [Google Scholar]
  18. Ingraham Christopher. Why Rural Voters Don’t Vote Democratic Anymore. Washington Post. 2016 Accessed via: https://www.washingtonpost.com/news/wonk/wp/2016/11/23/how-republican-gerrymanders-forced-democrats-to-abandon-rural-america/?utm_term=.bcd4a6c88199.
  19. Isenberg Nancy. White Trash: The 400-Year Untold History of Class in America. Penguin Books; 2017. [Google Scholar]
  20. Kimmel Michael, Ferber Abby L. ’White Men are this Nation’: Right-Wing Militias and the Restoration of Rural American Masculinity. Rural Sociology. 2000;65(4):582–604. [Google Scholar]
  21. Kurtzleben Danielle. Rural Voters Played a Big Part in Helping Trump Defeat Clinton. National Public radio; 2016. Accessed via: http://www.npr.org/2016/11/14/501737150/rural-voters-played-a-big-part-in-helping-trump-defeat-clinton. [Google Scholar]
  22. Leip Dave. Dave Leip’s Atlas of US Presidential Elections. [Accessed February 16, 2017];County-Level Presidential General Election Results 2016, version 0.96. 2017a via: http://uselectionatlas.org/
  23. Leip Dave. Dave Leip’s Atlas of US Presidential Elections. [Accessed February 16, 2017];County-Level Presidential General Election Results 2012, version 1.4. 2017b via: http://uselectionatlas.org/
  24. Lichter Daniel T. Immigration and the New Racial Diversity in Rural America. Rural Sociology. 2012;77(1):3–35. doi: 10.1111/j.1549-0831.2012.00070.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lichter Dan, Brown David L. Rural America In An Urban Society: Changing Spatial and Social Boundaries. Annual Review of Sociology. 2011;37:565–592. [Google Scholar]
  26. Lichter Daniel T, Ziliak James P. The Rural-Urban Interface: New Patterns of Spatial Interdependence and Inequality in America. The ANNALS of the American Academy of Political and Social Science. 2017;672(1):6–25. [Google Scholar]
  27. Lobao Linda, Lazarus Adua, Hook Gregory. Privatization, Business Attraction and Social Services across the United States: Local Governments’ Use of Market-Oriented Neoliberal Policies in the Post-2000 Period. Social Problems. 2014;61(4):644–672. [Google Scholar]
  28. McKee Seth C, Teigen Jeremy M. Probing the Reds and Blues: Sectionalism and Voter Location in the 2000 and 2004 U.S. Presidential Elections. Political Geography. 2009;28:484–495. [Google Scholar]
  29. Monnat Shannon M. Deaths of Despair and Support for Trump in the 2016 Election. Penn State University; 2016. [Accessed May 18, 2017]. via: http://aese.psu.edu/directory/smm67/Election16.pdf. [Google Scholar]
  30. Monnat Shannon M. The New Destination Disadvantage: Disparities in Hispanic Health Insurance Coverage Rates in Metropolitan and Nonmetropolitan New and Established Destinations. Rural Sociology. 2017;82(1):3–43. doi: 10.1111/ruso.12116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Morin Rich. Behind Trump’s Win in Rural White America: Women Joined Men in Backing Him. Pew Research Center; 2016. Accessed via: http://www.pewresearch.org/fact-tank/2016/11/17/behind-trumps-win-in-rural-white-america-women-joined-men-in-backing-him/ [Google Scholar]
  32. Morrill R, Knopp L, Brown M. Anomalies in Red and Blue II: Towards an understanding of the roles of setting, values, and demography in the 2004 and 2008 US presidential elections. Political Geography. 2011;30:153–168. [Google Scholar]
  33. Pritchard Bill, McManus Phil., editors. Land of Discontent: The Dynamics of Change in Rural and Regional Australia. Sydney: UNSW Press; 2000. [Google Scholar]
  34. Quinones Sam. Dreamland: The True Tale of America’s Opiate Epidemic. New York, NY: Bloomsbury Press; 2015. [Google Scholar]
  35. Robert Wood Johnson Foundation. County Health Rankings. University of Wisconsin Population Health Institute; Madison, WI: 2017. [Accessed May 18, 2017]. via http://www.countyhealthrankings.org/ [Google Scholar]
  36. Scala Dante, Johnson Ken. National Fact sheet #30. Carsey School of Public Policy. University of New Hamphire; 2015. Red Rural, Blue Rural: Rural Does not Always Equal Republican. [Google Scholar]
  37. Scala Dante, Johnson Ken. Is Rural Red and Urban Blue? Voting and Political Attitudes Along the Urban-Rural Continuum. Annals of the American Academy of Political and Social Science. 2017;672(1):162184. [Google Scholar]
  38. Semuels Alana. All the Men Here are Either on Drugs or Unemployed. The Atlantic: 2017. May 9, [Accessed May 9, 2017]. 2017. via https://www.theatlantic.com/business/archive/2017/05/men-women-rust-belt/525888/ [Google Scholar]
  39. Smith Kristen E, Tickamyer Ann R. Economic Restructuring and Family Well-Being in Rural America. University Park, PA: Pennsylvania State University Press; 2011. [Google Scholar]
  40. Soffen Kim. How Racial Gerrymandering Deprives Black people of Political Power. [Accessed May 22, 2017];Washington Post. 2016 Jun 19; via: https://www.washingtonpost.com/news/wonk/wp/2016/06/09/how-a-widespread-practice-to-politically-empower-african-americans-might-actually-harm-them/
  41. Stock Catherine McNicol. Rural Radicals: Righteous Rage in the American Grain. Ithaca, NY: Cornell University Press; 1996. [Google Scholar]
  42. Strijker D, Voerman G, Terluin I I, editors. Rural Protest Groups and Populist Political Parties. Wageningen: Wageningen Academic Publishers; 2015. [Google Scholar]
  43. U.S. Census Bureau. American Community Survey, 2010–14. Washington, DC: 2015. [Accessed May 18, 2016]. via: https://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml. [Google Scholar]
  44. U.S. Centers for Disease Control and Prevention. Wide-ranging Online Data for Epidemiologic Research (WONDER), Multiple Cause of Death Files, 1999–2015. [Accessed May 18, 2017];2015 via: https://wonder.cdc.gov/
  45. USDA. County Typology Codes. Washington, DC: USDA-ERS; 2015. [Accessed May 18, 2017]. via: https://www.ers.usda.gov/data-products/county-typology-codes.aspx. [Google Scholar]
  46. Warf Barney. The U.S. Electoral College and Spatial Biases in Voter Power. Annals of the Association of American Geographers. 2009;99(1):184–204. [Google Scholar]
  47. Whitaker Marie. Rural America and a Silent Majority Powered Trump to a Win. [Accessed May 22, 2017];NBC News. 2016 Oct 1; 2016. via http://www.nbcnews.com/storyline/2016-election-day/analysis-rural-america-silent-majority-powered-trump-win-n681221.
  48. Winkler R, Johnson Kenneth M, Cheng Cheng, Beaudoin Jim, Voss Paul R, Curtis Katherine J. Age-Specific Net Migration Estimates for U.S. Counties, 1950–2010. Applied Population Laboratory; University of Wisconsin-Madison: 2013. [Accessed July 18, 2017]. via http://www.netmigration.wisc.edu/ [Google Scholar]
  49. Woods Michael, Anderson Jon, Guilbert Steven, Watkin Suzie. ’The Country(side) is Angry': Emotion and Explanation in Protest Mobilization. Social and Cultural Geography. 2012;13:567–587. [Google Scholar]

RESOURCES