The Dangerous Illusion of a Presidential Third Party in 2024
It isn’t the first time and it certainly won’t be the last, but American voters are frustrated with the state of our politics, and approval ratings for both the Democratic and Republican parties are low. Data showing this discontent and interest in alternatives have led some to consider running a third-party candidate for president in 2024. Their reasons might sound high-minded, ripe for the moment, and a needed shakeup for a stale two-party system, but the reality is that a well-financed third-party candidacy will most likely benefit the presumptive Republican nominee Donald Trump or a Trump acolyte in 2024.
Our two-party system has evolved in such a way that makes it close to impossible for a third-party candidate to actually win the election outright. Instead, they would act as a spoiler—and anything but a staunchly conservative third-party candidate would be far more likely to pull support from the party in power in the White House, the Democratic ticket rather than the Republican. Given that recently discussed third-party candidates are either moderates or from the far-left, for the purpose of this report we will not focus on the possibility of a candidate running as more far-right than Trump.
To be sure, we support efforts to thoughtfully study and implement reforms that could make our democracy more representative and less polarized. But running a third-party candidate at the Presidential level in 2024 would do nothing to advance those goals. Below we lay out the historical case and empirical data indicating that a strong moderate or left-wing third-party candidacy would help only Donald Trump or a Trumpian Republican nominee.
Third-Party Candidates Don’t Win
History indicates that third-party candidates do not win presidential elections, despite often showing promise early on. Every presidential election cycle includes third-party candidates, some major and some minor. Without exception they consistently underperform expectations and do not come close to winning.
Presidential third-party candidates’ tendency to underperform is due in part to an oversimplification of the American electorate when it comes to interpreting voters’ true meaning when describing their own party affiliation. As FiveThirtyEight pointed out in 2020, polls show that a large number of voters identify as political Independents, but that self-described label doesn’t necessarily indicate a voter is in search of an Independent candidate when it comes time to vote, particularly in presidential elections where underperformance of Independents is so common. According to Gallup, 39% of Americans say they are Independents, while 31% say they are Democrats and 29% say they are Republicans. But with the occasional exception in state and local races, these self-identified Independents generally have a partisan leaning. For example, when asked which way they lean, 44% identify as a Republican or Republican leaner and 47% identify as a Democrat or Democratic leaner. While those “leaners” may be more politically fluid over time than those who identify as strong partisans, they tend to vote for the party to which they are leaning in a single election cycle at a very high rate. Researchers posit that this leaves 9% of Americans who are truly unattached to either political party. Polling conducted by the Wall Street Journal also indicates that in recent elections, 10% of voters cast a ballot for a different party in successive races.
Nine percent is not small (in fact, it represented about 14 million voters last cycle), and it is certainly large enough to influence national election results, but it is not enough to compete with the major parties at the Presidential level. Some argue that this is a unique moment in American politics, in which American voters are rejecting the parties in a new way. Yet the number of Independent voters has ticked up only four points since 1994.
Americans’ attachment to the two-party system and reluctance to support third-party candidates has played out historically. Going back to Theodore Roosevelt’s Bull Moose Party run in 1912, third-party candidates have won a grand total of 186 Electoral Votes, over 27 presidential elections. Eighty-eight of those Electoral Votes were won by Roosevelt, who had already been president. During that time, seven different third-party candidates either won an Electoral Vote or won over 5% of the national popular vote.
In some races, people took the third-party candidates seriously, and the candidates looked promising at points in the race. John Anderson polled as high as 26% in one Gallup poll. Ross Perot actually led the three-way race in a June 1992 Washington Post poll. In 2016, Gary Johnson polled at 10% in an August 2016 Pew Research survey. In the era before widespread public opinion polling, Strom Thurmond was expected to win a large number of southern states.
These candidates’ momentary strength shows that they are temporarily able to harness dissatisfaction with the state of politics and the two parties. But when it comes time for Americans to vote, these third-party candidates consistently underperform their peaks. Across these 27 elections, from 1912 to 2020, third-party candidates did not win enough Electoral Votes in sum to win a single election. Even Theodore Roosevelt, an immensely popular former president, won just over a quarter of the national vote. These drop-offs are evidence that in a two-party system, voters may kick the tires on third-party candidates but prefer not to waste their votes on them, recognizing that only the major-party candidates stand a real chance. Our two-party system is powerful, and history proves it.
Trump’s Supporters are Die-Hards. Democrats’ are…not.
Joe Biden’s victory in the 2020 election was not due to any erosion in support for Donald Trump. Rather, not only did Trump’s raw vote total increase, but in the key states of Pennsylvania, Wisconsin, and Michigan, Trump’s share of the vote actually increased. The change that made the difference between Hillary Clinton’s loss in 2016 and Biden’s victory in 2020 was the decrease in support for third-party candidates. As a result, Biden’s vote shares were higher in battleground states than Clinton’s, even when Trump’s vote share increased as well.
Given that Trump’s base of support actually expanded after four tumultuous years of his presidency, there is no reason to believe that some portion of his base would abandon him in 2024 for a third-party candidate. Rather, a third-party candidate would be more likely to divide the anti-Trump vote, as it did in 2016.
We can think about Trump support as a slice of pie. Despite the fact that Trump won in 2016 and lost in 2020, his portion did not shrink between 2016 and 2020; it actually grew. So, what changed to flip the results? In 2016, there was a third slice eating into Clinton’s support. In 2020, that third party segment shrank, so that Biden’s support eclipsed Trump’s.
We cannot say for certain that all of Gary Johnson and Jill Stein’s voters would have supported Clinton in 2016 given a forced choice. Third-party voters are unpredictable, and do not necessarily vote in ideologically coherent ways. But exit poll data from 2020 indicates that those who voted third-party in 2016 did support Biden in 2020 by a whopping 35-point margin, and AP Votecast data indicates that Biden won 2016 third-party voters by a 30-point margin. Whether these voters determined the outcome in 2016 is up for debate, but it is clear that these mercurial voters joined Biden’s coalition in 2020, helping solidify his victory.
The details of Trump’s approval ratings and favorability illustrate his fans’ devotion as well. Among his supporters, a vast majority strongly approved or viewed him very favorably. According to AP Votecast, at the time of the 2020 election, 50% of voters viewed Biden favorably, compared to 47% who viewed Trump favorably. But of those who viewed Trump favorably, 69% viewed him very favorably. Of those who viewed Biden favorably, 56% viewed him very favorably. This gap in strong enthusiasm has grown over the course of Biden’s presidency, as his approval rating among Democrats has softened. According to Pew, 41% of voters approve of Biden’s job performance, with 21% saying they strongly approve and 20% saying they somewhat approve. By contrast, Pew data at the end of his presidency showed Trump with 29% strong approval and just 9% somewhat approving.
Obviously, voters who strongly approve of a candidate are more likely to vote for that candidate than those who only somewhat approve. According to AP Votecast data, 96% of voters who said they strongly approved of Trump’s performance in office voted for him, compared to 83% of those who said they somewhat approved. Similarly, 95% of voters who said they viewed Biden very favorably supported him in 2020, compared to 85% of those who said they viewed him somewhat favorably. Biden’s voters are less committed to him and are less likely to remain at his side if offered the option of a third-party candidate.
Double Haters Lean D
Third-party voters tend to be those who do not like either of the major party candidates. These voters are often referred to as “double haters,” as they have unfavorable views of both candidates. In 2016, Hillary Clinton performed poorly with these “double haters,” losing them by 17 points, according to exit polls. In 2020, such voters backed Biden by a 15-point margin, 47%-32%, according to AP Votecast data.1 And an analysis of double haters’ demographics and issue stances indicates that this is a Democratic-leaning group that would tend to support Democrats in a forced choice. Double haters skew younger, with 65% under 50. On specific issues, double haters were 5 points more likely (82%-77%) to say racism is a serious problem in our society, compared to all voters. They were also 7 points more likely (56%-49%) to say Covid was not at all under control in November of 2020.
Double haters’ voting history, demographics, and stances on key issues in the 2020 election indicate that these voters are more inclined to support Democrats than Republicans. But given their dissatisfaction with both major-party candidates, they are ideal targets for a credible third-party candidate. Since this Democratic-leaning voter bloc would be more likely to support that candidate, Republicans would gain a real advantage.
Denying 270 to Both Candidates Means Republicans are Favored
The last third-party candidate to win even a single Electoral Vote was George Wallace in 1968. The most likely scenario is that even a well-funded third-party candidate would get zero, or a small number of Electoral College voters. But if a third-party candidate blew past historic precedent and managed to win a significant number of Electoral College votes – enough to keep any candidate from getting 270 – the outcome would be decided in the House of Representatives, where Donald Trump would prevail. While this scenario is unlikely, it is worth considering as the best-case scenario for a third-party candidate.
We should start with the hypothesis that the deepest blue and red states are off the table; there is not a scenario where Democrats lose California and New York, for example, or Republicans lose Wyoming and Mississippi. No third-party candidate could stand a chance in a state in which either Biden or Trump won by over 15 points in 2020. This takes the following states off the table:
Alabama, Arkansas, California, Connecticut, Washington DC, Delaware, Hawaii, Idaho, Illinois, Indiana, Kansas, Kentucky, Louisiana, Massachusetts, Maryland, Maine’s First District, Mississippi, Missouri, North Dakota, New Jersey, New York, Nebraska’s Third Congressional District, Oklahoma, Oregon, Rhode Island, South Dakota, Tennessee, Utah, Vermont, West Virginia, Washington, Wyoming.
These safe blue and red states together take 289 Electoral Votes off the table for a third-party candidate. These states are loyal to the Democratic or Republican parties. This means that only 249 Electoral Votes could even conceivably be in play for a third-party candidate.
While a third-party candidate cannot win, a performance in which they won a significant share of Electoral Votes would have a huge impact. It could ensure that no candidate reaches 270 votes, and the election would be decided by the House of Representatives.
When the House decides an election, each state delegation receives one vote. After the 2022 midterms, Republicans control 25 out of the 50 state delegations, while Democrats control 23, with two split evenly. To win the presidency in the House of Representatives, a candidate would need the support of 26 state delegations. It is therefore likely that if an election was thrown to the House, we would plunge into a constitutional crisis, with neither candidate able to reach the threshold to win. And because Republicans control more of the state delegations, a Republican candidate would be at an advantage.
It is common in politics to believe that the current moment is entirely unprecedented and an aberration from any previous time in our history. This attitude feeds into the belief that even though third-party candidates have never won in the past, now is the time that one might prevail.
This belief is false and misguided. The vast majority of American voters still lean towards either the Democratic Party or the Republican Party. While some like the idea of a third party, people want a third party that matches their particular belief set – some are leftists, some libertarians, some moderate. The most successful third-party candidacy possible would deny any candidate 270 Electoral Votes and would likely throw the election to House Republicans.
Pundits spent years waiting for Trump’s base to leave him, either by abandoning the GOP and supporting Democrats, or by staying home on Election Day. But they won’t. The Trump base is far more loyal than Biden’s coalition.
Anyone backing a third-party candidate should be clear eyed: they are not establishing a new political faction, because their candidate is not going to win. Rather, they are creating a spoiler who will help elect Donald Trump.
It should be noted that 2020 exit polls show Biden losing the double haters. However, the 2020 exit polls were widely considered to be historically inaccurate, and we therefore relied on AP Votecast data for our 2020 analysis.