The 2000 presidential election was one of the most significant in American history, from the heated campaign between George W. Bush and Al Gore to the Supreme Court’s controversial affirmation of the decisively-narrow Republican victory in Florida.
At the time, Democrats were trying to hold onto the White House after two terms of Bill Clinton. The party had not won three successive presidential elections since Harry Truman’s surprise triumph in 1948. Despite that success drought, Democrats had a lot of things going for them on paper.
Economic conditions were notably propitious under Clinton, but Gore shied away from stressing that reality, fearing fallout from the Monica Lewinsky scandal. Governor Bush, the unexpected scion of his father’s presidential legacy, led the polls in the early summer and late fall while running on a pro forma conservative platform.
In general, the political environment kept the predicted outcome of the nationwide race within the margin of error. Many Americans appreciated Gore’s connection to the 42nd President and his economic record. An equally-fervent opposition contingent reproached Clinton and Gore as moral reprobates, preferring Bush’s conservative, change-oriented message.
The period of time following election day 2000 was even more hectic than the preceding campaign. On that fateful night, all eyes turned to the swing state of Florida. Over the course of the evening, the major networks called the Sunshine State for Gore and Bush before retracting their projections entirely. It was clear from then on that Florida’s 25 electoral votes would decide the fate of the nation heading into a new century.
After months of legal wrangling on behalf of both parties, the Supreme Court issued a 5-4 ruling in Bush vs. Gore upholding the 537 vote Republican victory in the state. Despite losing the nationwide popular vote to Gore, the constitution ensured that Bush became the 43rd President. The GOP’s controversial success had palpable historical and political implications that affect the U.S. to this day.
But the 2000 election did not merely mark a deviation from the policy agenda of Clinton and the Democrats. Regional coalitions that had clearly proliferated themselves during the 1980s and 90s changed significantly in 2000, often with statewide ramifications. In general, these shifts foreshadowed regional and demographic voting patterns that solidified themselves during the Trump-era.
To better understand how the nation’s electoral past impacted its present politics, Split Ticket decided to look at broad electoral changes that took place in the South, the Midwest (Appalachia), and the West Coast before and after the historic 2000 presidential contest. Additionally, an examination of exit poll data was conducted to analyze how demographics swung and trended from 1996.
Because Ross Perot’s formidable 1992 campaign siphoned votes from both the Democratic and Republican camps, forcing plurality statewide victories on multiple occasions, the comparison data for this piece will primarily come from 1996.
From a modern perspective, Gore’s margins in the South would be considered excellent. At the time, however, they were seen as both below average and indicative of changing electoral winds that benefitted the GOP presidentially.
While the South had more or less devoted itself to Republican presidential nominees from 1972 onward, there were notable exceptions. In 1976, Georgia Governor Jimmy Carter swept Dixieland against President Ford. Four years later, he only narrowly lost the region to Ronald Reagan and the Republicans.
The next two presidential contests witnessed the GOP monopolize the whole South despite entrenched, though gradually weakening, Democratic advantages at the congressional and local levels.
Although much of the South never left the Republican presidential column again post-Carter, there was a marked revival of support in 1992, when Democrats nominated an all-southern ticket in Bill Clinton and Al Gore. In both 1992 and 1996, Clinton carried Arkansas, Louisiana, Missouri, Tennessee, and Kentucky twice. He also carried Georgia and Florida once each in separate elections.
In 2000, the entire southern region swung rightward. Gore lost all seven of the aforementioned Clinton states, including his native Tennessee. He additionally failed to keep the rest of the South close as Clinton had. Despite an evenly-divided national vote, Gore’s performances throughout the south were little better than Michael Dukakis’s in 1988.
Looking at the South by county is more instructive because it shows where pro-Republican shifts struck the hardest. In many cases, counties that did not realign fully in 2000 would do so later in the ensuing decades. For that reason, Gore’s benchmark was predictive of trends that have since borne out. Let’s look at two examples.
The first is rural Foard County in Texas. Prior to 2000, Foard had voted for Republican presidential nominees only twice in its modern existence. From 1988 to 1996, Democrats, including Dukakis, won it by about 30 points. Despite its ancestral lean, Bush carried it by 4. That margin turned into 19 four years later, confirming a realignment that struck many similar counties. Trump won Foard by 63 points in 2020.
The second is Telfair County in Georgia. Located in a part of the state known for its strong English ancestry and historical racial prejudice, Telfair was generally partial to Democrats like Foard. Gore carried it by 2 points in 2000, confirming that the county would realign soon even though it had not immediately done so. Trump won Telfair by 31 in 2020.
Democratic margins here have not suffered as much as they have in counties like Brantley because Telfair has a sizable black population. The countywide shift is entirely attributable to white voters.
2000 also brought change to the Midwest, where President Clinton had won all but one state, Indiana, in 1996. Gore still carried Michigan, Wisconsin, Minnesota, Iowa, and Pennsylvania, but his party lost ground in each. Republican gains were particularly pronounced in Kentucky, Ohio, and West Virginia, where coal Democrats broke traditional voting patterns at unprecedented levels. This regionalized, industry-specific discontent is usually attributed to Gore’s national environmental policies.
Looking at specific sets of counties in the aforementioned states paints a clear picture of a burgeoning realignment. While Gore did indeed perform well enough in KY, OH, and WV to win historically-Democratic regions, his margins in and around these communities were weak enough to cost him statewide victories. Generally speaking, pro-Republican trends in blue counties manifested themselves partially in 2000 before taking full federal effect later on.
Take Elliott County, Kentucky. Gore carried it by 29 points, yet Bush’s 34.7% was, at that time, the best performance for a national Republican countywide. With the exception of 2004, Democratic margins would continue to narrow in Elliott until it flipped hard for the GOP during the Trump-era.
In Ohio, Gore notably underperformed Clinton (and Obama 2012) in the lakefront, south-central, and Ohio River Valley regions. Ancestral ties were only strong enough to deliver him Jefferson, Columbiana, and Belmont by reduced margins. Gore’s only bright spots excluding major metropolitan areas were the Steel-heavy Mahoning Valley counties in which he posted numbers predictable for the era.
Counties throughout West Virginia and Pennsylvania, like Wyoming and Greene, gave Democrats reduced majorities in 2000, foreshadowing a grim electoral future. Change in the Mountaineer State was especially sharp. Clinton had carried almost all of it in 1996, yet Gore only managed to take the most historically-blue, coal-heavy counties. By 2012, every county in West Virginia would be voting Republican presidentially.
In a mirror image of what we saw in the now ruby-red South, the results on the West Coast in this election were astoundingly close by current standards. Republican strength in the suburbs of California, Oregon, and Washington kept all three states relatively close, with Gore winning California by just 12% and scraping by in Oregon and Washington by 0.4% and 5.5% respectively. Considering that in 2020, Biden won California by 29, Oregon by 16, and Washington by 19 points, these margins would probably constitute a Republican dream in the modern era.
Back then, however, Republicans had reason to be disappointed, especially as late polling had suggested that California had suddenly become a battleground worth Republican investment. Bush’s extensive time campaigning in the state courting Hispanics and Ralph Nader’s expected siphoning of some Democratic votes had put the Golden State into play, and both sides felt that Republican efforts with moderates and Hispanics had the potential to make margins somewhat closer in a state Gore had written off as safe early on in the campaign. Ultimately, however, in a foreshadowing of what was to come over the next two decades, sharp Democratic swings left in Los Angeles and the Bay Area wiped out the massive Republican gains in the more rural parts of the state and were enough to carry Gore to victory.
Bush’s relative weakness in the white, suburban, and historically Republican parts of Southern California, such as Orange and San Diego hurt him badly here. In what was to become a general theme over the next twenty years, every one of the Republican strongholds along the southern part of the coast trended towards Democrats, with Gore becoming the first Democrat in 36 years to crack 40% of the vote in ruby-red Orange County. With no other source of votes to counter Gore’s strongholds, Bush’s underperformance in core, historically Republican suburban areas was the nail in the coffin for his chances in the state.
If California was a slight letdown for them, however, the Republicans would have felt bitterly disappointed at the results in Oregon, where large margins in Multnomah County (the most populous in the state) were just enough to insulate Gore against the sharp tilt right that the rest of Oregon took. The sitting Vice President won Oregon by under half a percentage point in the nation’s first all-mail election, with Bush’s strength in the rural east of the state dampened by sharp underperformances in the metro areas of Portland and Eugene.
This theme of educated suburbanites trending against Republicans continued in Washington, where an underwhelming performance in Seattle and its surrounding metro areas neutralized the opening created by Bush’s gains in the rural parts of the state. In general, voters on the more secular, culturally liberal coasts reacted poorly to Bush’s more overtly religious and socially conservative appeals, and extensive time spent campaigning in these states was not enough to prevent Republicans from sliding to losses on the West Coast for the third election in a row. But this was not a theme limited to Bush; Republicans continued free-falling in these states and in their former strongholds, and by 2020, suburban battleground counties such as Oregon’s Washington County and California’s San Diego County had skyrocketed to the left by nearly 30 points.
This fact may have been masked by looking at the swing from 1996 — counties like San Diego and Orange actually swung right from Clinton’s re-election campaign. But the national environment was also 8 points more Republican in 2000 than it was in the prior election, and after adjusting for this, the trends that would define the next two decades on the West Coast became crystal clear. The thermostatic effect (otherwise known as the shifts in the national environment) is one of the more powerful masking factors in American politics, and to truly observe the power of trends and patterns in the electorate, it is important to control for it.
It is difficult to get reliable data on how demographics voted in 2000, but Ruy Teixeira’s book The 2000 Election And The Future of US Politics contains some fascinating VNS exit poll datapoints that give us some insight into what might have happened, which we summarize in the table above and in the paragraphs below. We note, however, that exit poll estimates are noisy and that the numbers here should be treated with wide error bars, as they are only meant to give a rough picture of what happened.
In a sign of what was to come, exit polls suggested that college-educated whites likely trended left, possibly by about 4 points in this election. But interestingly, this movement was driven almost entirely by college-educated white women — Gore actually might have outrun Bill Clinton with this demographic, with VNS exits suggesting he won them by 8 while the former president had won them by 7. Margins with college-educated white men, however, largely stayed static relative to the national environment; Bush won them by 26 while Dole won them by 17, and 2000 was 8 points more Republican nationally than 1996 was.
Instead, Bush’s strength came largely from his tremendous gains with working-class whites and white voters without a college degree. While Bill Clinton might have actually won non-college whites by a point in 1996, Bush won them by a whopping 17 points in 2000, in comparison to his 9 point win with college-educated whites. And while these gains were strongest with non-college white men, who exit polls suggest went from Dole +7 to Bush +29, he also made sizable inroads with non-college white women, who went from Clinton +7 to Bush +7. Both non-college men and women swung and trended sharply right relative to the nation, and this fueled a large portion of Bush’s victory, especially in the South and the Midwest.
At first, this may not seem surprising, especially with current context. Given that in 2020, non-college whites were a Trump +26 group and college whites a Biden +8 group, as per Catalist estimates, educational polarization among white voters now seems the norm in American politics. However, the 2000 election may actually have been the first modern election in which non-college whites were significantly more Republican than college-educated whites. This election saw a reshaping of the GOP coalition, and in doing so, ushered in a new alignment in American politics that would dominate the next two decades.
Another component fueling Bush’s victory were the gains he made with Hispanic voters, who were Clinton +51 in 1996 but swung 24 points to the right in four years, with Gore only winning them by 27 in 2000, as per VNS data gathered by the Roper Center. And while that same dataset suggests Bush lost significant ground with Black voters in this election, losing them by 81 in comparison to Dole’s 73 point loss, his gains with Hispanics likely neutralized this slippage.
It must again be emphasized that the estimates above, like with all exit polls, are quite noisy and should be treated as such. But they are the best source we have for this time period, and they do give us a decent picture regarding what happened on the whole. Sharp swings right with Hispanic and white non-college voters drove Bush’s victory, a fact borne out by a simple examination of county-level data. Meanwhile, both county-level and exit poll data suggest that the traditionally-Republican demographic of college-educated whites did not actually swing nearly as much towards them as one may have expected, and that this demographic actually trended leftwards overall, a shift that would only accelerate over time.
Editor note: To add further context, this article was updated to add 2020 margin estimates with college whites and non-college whites, using Catalist data.
I’m a software engineer and a computer scientist (UC Berkeley class of 2019 BA, class of 2020 MS) who has an interest in machine learning, politics, and electoral data. I’m a partner at Split Ticket, handle our Senate races, and make many kinds of electoral models.
My name is Harrison Lavelle and I am a political analyst studying political science and international studies at the College of New Jersey. As a co-founder and partner at Split Ticket, I coordinate our House coverage. I write about a variety of electoral topics and produce political maps. Besides elections, my hobbies include music, history, language, aviation, and fitness.
Contact me at @HWLavelleMaps or firstname.lastname@example.org