Donald Trump’s 2016 insurgent campaign caused much excitement among some right-leaning intellectuals, who hoped that it might encourage the Republican party to shed its traditional economic policy commitments in favor of a new populist ideology serving a working-class base. Yet, while Trump has focused more on regional interests particular to Rust Belt swing states than previous GOP candidates, in terms of gender, race, and social class, the coalition of voters who turned out to vote for Donald Trump in 2020 was remarkably similar to those who did so for Mitt Romney in 2012.
This seems paradoxical because so much analysis has focused on geographic shifts in support, by examining aggregate county-level or precinct-level vote tallies. But to draw conclusions about individual voters and class voting patterns from community aggregates is to fall prey to the ecological inference fallacy. While Trump clearly received higher support than Romney in low-income counties in rural Ohio and a smaller share of the votes from affluent counties around Atlanta, his surge in support has often been due to middle-income individuals in poorer regions, while he has attracted few votes from low-income residents of the nation’s largest metro areas.
Although more fine-grained surveys of the U.S. population will eventually become available, exit polls offer the best general picture of potentially pollster-averse individual voters who actually turned up to vote. Even if subject to non-response bias (and the topline numbers are in line with the overall results), comparing shifts in exit polls from 2012 to those in 2020 makes possible a general assessment of changes in the electorate within constant demographic categories. While a 3% polling error makes it hard to predict the winner of a close election, it would do little to obscure an inversion of class voting patterns if such a thing had occurred.
Far from transforming the social class alignments of American politics, the 54% share of Americans with household incomes exceeding $100,000 that supported Trump in 2020 was the same as that which supported Romney in 2016. The proportion of Americans with incomes below $50,000 who supported Trump (42%) was only slightly greater than that which supported Romney (38%), while the share of the electorate in that lower-income group fell from 41% to 35%. Trump’s 49% share of voters without a college education was only a touch more than the 47% that supported Romney.
The conventional wisdom is also mistaken on race. Trump’s 57% level of support among white voters in 2020 was lower than the 59% share Romney received. With Obama and his legacy off the ballot, the GOP share of black voters surged from 6% in 2012 to 12% in 2020, while that among Hispanics ticked up from 27% to 32%. Due to the fact that white voters make up a larger share of the electorate, the net impact of these shifts roughly cancelled out. Nor has Trump done much to inflate the gender gap. In fact, the difference between Trump’s 2020 share of male and female voters (49% to 43%) was smaller than that for Romney in 2012 (52% to 44%).
The hope among Christian conservative intellectuals that Trumpism could open the door for a reinvigorated social conservatism, if unshackled by unpopular economic commitments, is similarly misplaced. Both in 2012 and 2020, voters identified the economy as the most important issue, and the perceived contrast between the parties is of primary importance to their vote. Whereas 24% of the electorate in 2020 cast their vote for a presidential candidate they disagreed with on abortion, only 5% voted against their preferred choice on the economy. Trump’s 54% support among married voters is down from the 56% who voted for Romney, while he captured 28% of LGBT voters – up from 22% who voted for Romney.
Trump’s election in 2016 is often spoken of in the same breath as the United Kingdom’s Brexit referendum of the same year. Both results certainly caused a similar amount of surprise, and observers who loved or hated one tended to feel the same about the other. Yet, despite similar atmospherics, it’s worth bearing in mind that a Trump presidency and Brexit were substantively very different things – Trump’s election changed the head of the executive branch, while Brexit represented a sweeping transformation of Britain’s constitutional, economic, and political orders.
Whereas America’s voting behavior in 2020 very much resembled that of 2012, Brexit provoked a complete political realignment. In the 2015 general election, Britain’s Conservative Party received 29% support among voters with incomes below £20,000 and 51% of the vote among those with incomes above £70,000. When the 2019 election was fought over the issue of Brexit, the class gradient was turned on its head: conservative support surged to 45% of the vote among Brits with incomes below £20,000, while plummeting to 40% among those with incomes above £70,000.
While Trump’s 2016 campaign opened deep divisions among right-of-center intellectuals and seemed to represent a major departure from past alignments, below the surface the Republican coalition was largely unperturbed. Trump won the presidency by riding a wave more than by making one. In 2010, furious public opposition to Obamacare swept Republicans to control of the House, while in 2014 it propelled them to a majority in the Senate. In 2012, by nominating Mitt Romney, who had endorsed much of Obamacare, Republicans failed to capitalize – but Trump’s forthright opposition to the Affordable Care Act helped push him over the top as health insurance premiums spiked in the fall of 2016. Trump’s approval rating as President subsequently collapsed in the spring of 2017 as he proved unable to put together a popular replacement for Obamacare, and it never entirely recovered.
Although results for the 2020 election have not yet fully been counted, Trump clearly lags behind the share of the vote received by House and Senate Republicans across the country. In 2016, he trailed the House GOP’s national share of the vote by 3.0 percentage points, whereas Romney was only 0.4% behind it in 2012. It may be true that Trump helped increase turnout, but his presence on the ballot clearly also served to motivate increased turnout among Democrats as well as Republicans.
Yet, while Trump’s coalition of support was very similar to Romney’s in terms of broad demographic categories of class, race, and gender, Trump’s coalition was geographically different – and better-attuned to the requirements of the electoral college. Although many Democrats have damned the electoral college as structurally and irredeemably biased against them, Romney in 2012 received a higher share of the national popular vote (47.2%) than he did in that year’s “tipping point state” of Colorado (46.1%).
Trump’s two-time significant overperformance relative to the popular vote reflects the fact that Trumpism (to the extent it differs from orthodox conservatism) was an ideological grab-bag of opportunistic policies, pragmatically assembled for the dedicated purpose of eking out a victory in the electoral college. Whereas Romney seemed embarrassed to engage in the low-minded politics of pork, pollution, and protectionism in pursuit of swing-state votes, Trump clearly revelled in it. Whereas Romney endorsed market-based policies to reduce emissions from energy production, Trump straightforwardly promised to reopen coal mines and put miners back to work. Though Trumpism’s working class orientation has been greatly exaggerated, his preoccupation with the regional concerns of the Rust Belt was real – and it was on a regional basis, across social classes, that the election swung to him in 2016.
This worked like a charm four years ago. Yet, voters’ desire to punish Democrats for the unpopular policies of the Obama administration receded by 2020, and Trump seems set to carry only the states from the region that Romney won in 2012. Furthermore, a coalition calibrated for the electoral college seems likely to leave the GOP just short of a majority in the House of Representatives, despite the national vote share being similar to that when they carried it in 2012. While they may be able to win the presidency by capturing the Midwest, Republicans will need to take back congressional districts in Blue State suburbs if they want to fully govern.
The GOP can learn from Trump’s success in nudging electoral votes into the Republican column by appealing to swing-state economic interests. But the effectiveness of such pragmatism should not be mistaken for any sweeping ideological transformation. Most Republican voters retain the same primary motivations they have long had: opposition to high taxes and liberal social engineering. Any ambitious politician who imagines that he or she can abandon core conservative commitments because Republican voters have suddenly become “populists,” are in for a rude awakening.
Interested in real economic insights? Want to stay ahead of the competition? Each weekday morning, e21 delivers a short email that includes e21 exclusive commentaries and the latest market news and updates from Washington. Sign up for the e21 Morning eBrief.
Photo by Scott Olson/Getty Images