Given that we are having public spats over whether our nation was actually founded in 1776, the notion of contributing genuine insight to something just eight months in the rearview mirror seems fanciful. Yet lessons of the 2020 presidential election are not unknowable. And pursuing them reveals important trends in American politics and society.
But if we wish to glean some insight into 2020 that has a chance of standing the test of time, perhaps we might instead start with some narratives or lessons that seem unlikely to hold water. We can start with President Joe Biden’s assertion in Brussels that the Republican Party is “fractured” and “vastly diminished in numbers.”
The truth is that the 2020 election was a particularly close affair — perhaps one of the closest in recent years, depending on how it is examined. Had some 6,000 people changed their minds in both Georgia and Arizona along with 10,000 in Wisconsin, Donald Trump would be president today. Senate control likewise was decided by around 13,000 Georgia voters opting for the Libertarian candidate rather than Sen. David Perdue. The Democrats’ House majority stands at around five seats.
This occurred against a rather grim backdrop for Republicans. In 2020, national gross domestic product dropped a gut-wrenching 33% in the second quarter, the quarter to which many elections forecasters pay the most attention.
The election also occurred against the backdrop of a pandemic that killed hundreds of thousands of people before the election occurred. While some leaders were able to harness the pandemic to improve their lot, often by reassuring the public and providing competent assessments of the pandemic to their voters, Trump seemed at times determined to do the opposite.
Not only that, but Republicans were badly outspent. Sen. Dan Sullivan was outspent 2 to 1 in his bid for a second Senate term from Alaska, as were Sen. Joni Ernst of Iowa and Sen. Thom Tillis of North Carolina. Sara Gideon outraised Sen. Susan Collins by almost 3 to 1 in Maine, while Barbara Bollier outspent Sen. Roger Marshall by almost 5 to 1 in the race for an open Senate seat in Kansas. Republicans were outspent in the presidential race by about $200 million, and overall spending favored Democrats as well.
I march through this not simply to relive the 2020 election, but rather to point out what should be obvious: Republicans had almost everything going against them, yet Democrats basically had to win three coin flips to get their trifecta. This was an election that easily could have gone to Republicans. This is to say: If there is a “big lesson” to be learned from 2020, it probably is not that Trumpism has divided, disrupted, and dispirited the party of Abraham Lincoln.
Eliminating the “Trump destroyed the GOP” lesson is an important step in figuring out what 2020 was about. But another conclusion seems to flow from this, which is perhaps uncomfortable for those who have been appalled by much of the past four years: The appeal of Trump and Trumpism (to the extent those are different phenomena) are genuine. Moreover, among some groups, that appeal expanded from 2016 to 2020.
In an interview with the New Yorker, elections analyst David Shor explained: “Trump didn’t exceed expectations by inspiring higher-than-anticipated Republican turnout. He exceeded them mostly through persuasion. A lot of voters changed their minds between 2016 and 2020.” Who were these voters, and what were they responding to? What lessons could the GOP take away from 2020 that could help it in 2024 and beyond?
It’s tempting to talk about white, working-class voters, but this is more of a well-trod 2016 story than a 2020 story — though Trump did do well among white voters without college degrees, likely saving the GOP from taking a substantial hit in the House and Senate.
More interesting for 2020 purposes is Trump’s performance among Hispanics. This fast-growing segment of the electorate formed a bedrock of the “emerging Democratic majority” theory, which posits in part that as the Hispanic population share of the electorate grew by a steady 1% to 2% each year, the Republican vote share should decline over time, translating to a Democratic advantage.
There were many problems with the theory overall, but a Republican candidate scoring well into the 30s among Hispanics is a particularly large problem for it. More important, a candidate like Trump achieving one of the highest Republican vote shares in the past 50 years represents a challenge to much of the policy analysis about this group.
To understand just how substantial Trump’s gains were, let’s examine the counties in South Texas, including the Rio Grande Valley. This heavily Hispanic section of the Lone Star State has long been a bedrock for the Democratic Party in Texas, with many of these overwhelmingly Hispanic counties providing near-unanimous support for the Democratic Party.
We can see this in this map from the 2008 presidential election, which shows Democratic performances in these counties (the scale is truncated between 34% and 66%, so if anything, the true results are more exaggerated). As we approach the periphery of the Rio Grande Valley, Republican performances increase, but even a famously pro-immigration reform Republican like John McCain was unable to make much headway here.
In 2016, we see much the same thing.
Republican performances here were remarkably stable. Overall, Democrats won almost exactly 62% of the vote in years as varied as 1960, 1976, 2008, 2012, and 2016.
But we actually begin to see the first cracks in the edifice in 2018. Beto O’Rourke mounted one of the strongest overall showings for a Texas Democrat in decades, coming just 2.5% away from unseating Republican Sen. Ted Cruz. Yet South Texas wasn’t quite as blue as it had been in the past.
O’Rourke’s 60% share here was particularly jarring, given that he was running so well statewide. Even Wendy Davis, who famously lost by 20 points in her 2014 gubernatorial bid despite her status as a national media darling, managed to win 77% of the vote in Starr County and 67% of the vote in neighboring Zapata County. (In 2020, Davis lost to Rep. Chip Roy in her attempt to unseat the congressman from Texas’s 21st Congressional District.) O’Rourke’s vote shares? In those counties, he received 76% and 62%, respectively. Those may seem similar, but again, one must remember that O’Rourke was running 10 percentage points ahead of Davis statewide.
Then came 2020.
Trump won 53% of the vote in this region. He won 47% of the vote in Starr County and carried Zapata County with 53% of the vote, becoming the first Republican presidential candidate since Warren Harding to carry the latter county. This performance trickled down the ticket as well. Sen. John Cornyn only lost Zapata County by 10 points, and he lost Starr County by 15. Democratic Rep. Filemon Vela almost lost his reelection campaign in his 83% Hispanic district.
Why? Democrats will point to a lack of Democratic door-knocking campaigns due to the coronavirus and other organizational factors, but turnout was up substantially here and in other areas of the country. Similarly, while Democratic neglect of the Hispanic vote could have played a role, it seems unlikely to have produced swings this large.
The Occam’s razor explanation is that Trump, and perhaps the Republican Party as a whole, simply appealed to these voters in a way that other Republican candidates have not. The reasons for this would require a lengthy research project, but hypotheses abound. Counterintuitively, some emphasize Trump’s strong stance on border security. After all, voters are all citizens, and some may view illegal immigration differently than white liberals expect. Many Border Patrol agents are, in fact, Hispanic.
But it wasn’t just the Texas border that saw this swing. Heavily Hispanic areas of Florida, California, Colorado, Illinois, and New York saw substantial increases in Republican performance. Perhaps the change is attributable to Trump’s strong, “macho” personality, but this also plays a bit too heavily on stereotypes of Hispanic males.
The best explanation is probably twofold. First, the notion that immigration has the same salience for Hispanic voters as civil rights may have for African Americans always stemmed from white liberals attempting to pound a square peg into a round hole. In a late 2020 poll, the Pew Research Center found that Hispanics were concerned primarily about the economy, healthcare, and the coronavirus outbreak, just like other adults. This is consistent with a decade’s worth of findings from Pew and other firms.
The second factor is this: As America becomes more polarized, we should expect similar movement among all voters, including nonwhites. Republicans have agonized for decades over how to get conservative Hispanics and African Americans to vote like conservative, non-Hispanic white voters. Trump may have finally pulled it off. While attributing the shift to particular conservative bete noirs like “defund the police” or “socialism” represents a bit too much of a “just so” story, Shor explained that “what happened in 2020 is that nonwhite conservatives voted for Republicans at higher rates. They started voting more like white conservatives.”
Of course, we must realize that Trump did not appeal to everyone, or even most people. Throughout his presidency, his job approval never exceeded the 47% mark. Even though the Electoral College vote was close, he nevertheless lost the popular vote by 4 percentage points, one of the worst showings for an incumbent on record.
To understand this, we turn briefly to a different group: white voters with college degrees. For years, this group provided the backbone of the Republican Party. Places such as Nassau County, New York, Palm Beach County, Florida, and Orange County, California, helped keep those otherwise Democratic states winnable for the Republican Party. Beginning with Bill Clinton’s “New Democrats” reinvention of the party, many of those suburbs shifted leftward as white people with college degrees warmed to the Democrats’ fiscal moderation and social progressivism. They still leaned Republican, but especially outside of the South, the reduced Republican margins helped push heavily urban states to the Democrats.
Trump’s belligerent attitude and strident stance on “identity” issues likely helped push these areas even further leftward in 2020. Let’s examine Texas again, but this time, we will focus on the major urban areas: Houston and the Interstate 35 corridor that stretches from Dallas down to San Antonio.
In 1960, Texas was a surprisingly close state, and it was Republican dominance in these counties that helped make it so.
Nixon carried these areas with 52% of the vote. By 1980, even though rural Texas was still Democratic, Republican strength in the urban and suburban areas had grown enough that Ronald Reagan piled up almost 60% of the vote here.
Even in the Democratic wave election of 2008, McCain’s margins here held up. In fact, relative to the country as a whole, his 53% vote share was even stronger than Reagan’s.
True, McCain lost Dallas, Bexar (San Antonio), and Harris (Houston) counties, but his overwhelming show of strength in the suburban counties kept the state Republican. Now, compare that with the map from 2020.
Trump still shows substantial strength in the less dense suburban areas, but the core urban counties are now heavily blue. Trump’s vote share in Dallas County (34%) wasn’t that different from his showing in famously liberal Travis County (27%), which includes Austin. Suburban counties including Williamson, Collin, and Denton were also weak, with GOP performance dropping off by over 10 percentage points from 2008.
What do these lessons mean together? How can we synthesize an overall strong Republican performance relative to national conditions, gains among nonwhite voters, and slippage among white voters with college degrees? Can we do so without writing yet another “Trumpism without Trump” piece?
Perhaps not, though this might be about things that go beyond Trump. The fact that we see Republican weakness among white voters with college degrees foreshadowed as early as 1992, or that Democratic slippage among Hispanics in South Texas is evident in a 2018 Senate race, suggests that there are larger forces at work here than just Trump. Demographic and attitudinal changes tend to come slowly, suggesting that 2022 and 2024 are unlikely to look markedly different than the past half-decade.
At the same time, it does illustrate the choice the GOP has to make. One probable lesson is that immigration reform isn’t the issue that Washington insiders make it out to be. To be sure, a Republican who favors it for policy or moral reasons should continue to back it and try to make the case. But the notion that a hard-line stance on illegal immigration is inconsistent with a strong performance among nonwhite voters is just belied by the reality of last year. You can see this tension already resolving with possible 2024 contenders such as former United Nations Ambassador Nikki Haley, who emphasizes her parents’ status as immigrants but also emphasizes that they came here legally.
This shift is tougher for someone like Sen. Marco Rubio, whose name is probably inextricably linked to the ill-fated 2013 immigration reform bill. But a shift in tone and emphasis may make for a position that is more palatable to base Republican voters without further alienating the white people with college degrees who are drifting from the party. Likewise, Trump demonstrates that a conservative stance on other cultural issues isn’t automatically disqualifying, something a dyed-in-the-wool social conservative like Rubio might take hope from.
That, of course, still leaves unanswered questions about other specific would-be Republican candidates. Could former Vice President Mike Pence, who lacks Trump’s forceful personality but holds many of the conservative social views that turn off suburban voters, even hope to match Trump’s showing? Could Ivanka Trump retain her father’s positives while blunting the harder edge that repelled white voters with college degrees? Perhaps most intriguingly, could someone like Florida Gov. Ron DeSantis or Arkansas Sen. Tom Cotton bring the “alpha energy” that motivated at least some of Trump’s supporters without making the outlandish statements that many found off-putting.
The lessons of 2020 are still in the “early draft of the first draft” phase. This is as it always has been. Had the recession of 1981 lasted a few months longer, Reagan might well have lost in 1984, and our understanding of his presidency would be radically different. Likewise, if Biden has a successful first term and wins reelection handily, the history of 2017-2020 will look very different than it does today.
But the lessons of 2020, for now, seem to be this: History and elections are less predictable than we think, and winning electoral coalitions are more flexible than many believe. The range of potentially successful combinations is not limited to those favored by “think tank conservatives” in Washington. That doesn’t mean, however, that the options are limited. Finding the balance between the desires of a party’s base and those of the median voter is a difficult task — which, really, is the lesson of every election.
Sean Trende is a senior elections analyst at RealClearPolitics and a visiting fellow at the American Enterprise Institute. He is a graduate candidate in political science at the Ohio State University.
