Both major parties seem to be embracing unpopular presidential candidates. Republican frontrunner Donald Trump is facing four looming criminal trials while a slew of more traditionally qualified candidates fail to gain traction. And the Democrats seem set to offer America’s oldest-ever president running for a second term despite anemic poll numbers. This time a third party should be able to gain real traction, right?
Enter No Labels, a would-be centrist third party. The group’s organization and funding are murky, creatively structured to avoid campaign finance disclosures. But the effort seems to be well-funded, and it has attracted a fair number of (mostly former) governors, senators, and other notable politicians. Many fear No Labels will be a spoiler tipping the race to Trump, intentionally or not.
Former senator and Democratic vice-presidential nominee Joe Lieberman has acted as a spokesman for No Labels; the organization’s plan is to qualify for ballot access across the country and nominate a centrist ticket. West Virginia’s Sen. Joe Manchin (a nominal Democrat who drives his party crazy) has appeared at No Labels events and has been teasing the possibility he might run. Former Maryland Gov. Larry Hogan has also refused to rule it out. Other possibilities bandied about include Maine’s Sen. Susan Collins and former Utah Gov. Jon Huntsman (those last three decidedly non-MAGA Republicans).
Over the years, I’ve interacted with more than a dozen elected officeholders considering a third-party campaign. This includes seven governors, five members of Congress, and a smattering of others who’ve held lower offices. I was a campaign staffer on the 2016 Libertarian presidential campaign of Gary Johnson, former governor of New Mexico, and his running mate Bill Weld, former governor of Massachusetts. For others, I merely participated in conversations and wrote strategy memos outlining the pros and cons of their prospective candidacies. Most ultimately decided against running.
Some, like Johnson and Weld, were looking at the Libertarian Party, while others considered independent bids or looked at the relative merits of running with a third party versus running in a major-party primary. The world of minor parties and ballot access petitioners is a relatively small one. Word gets around whenever somebody notable gives it a serious look.
In some instances, I was hoping a candidate would run. For others, I was more ambivalent. But I always wanted the potential candidate to know what they were getting into. Those closest to and most eager for a candidate to run can often be persuasive in their enthusiasm. Manchin, Hogan, and anybody else seriously considering a No Labels candidacy need to hear the full truth about what they’d be signing up for.
An Unpromising History
Since World War II, there have been 14 third-party or independent presidential bids by governors, senators, congressmen, and in one case a former vice president. (There have been even more if you count the campaigns that ended before November, or the times a politician let a party put his name on the ballot but did not actively campaign.) Of those, seven failed to garner even 1 percent of the vote. Only one broke double digits. Two won states, but both were segregationists appealing to the Jim Crow South, a dynamic thankfully relegated to the ash heap of history. Taken together, the average result of these campaigns is just 2.33 percent. Some degree of notability is certainly necessary to run a serious campaign for the most powerful office in the world. Necessary, but far from sufficient.
Third parties can benefit from running a credible ex-officeholder instead of an unknown party activist. A governor or senator can attract more support, more attention, and ultimately more votes. But the difference is likely to be just a handful of percentage points, not a real shot at winning.
The plausible best-case scenario for any No Labels candidate with significant experience in high office is not that they make the debates, or win any states, or deadlock the Electoral College, much less actually become president. A realistic best case would be 5 percent of the popular vote. The median likeliest outcome is 2 or 3 percent. Less than 1 percent is a distinct possibility.
Ample resources and media attention might get you out of less-than-1-percent territory, but it won’t move the needle much beyond that. If you aren’t comfortable with a vote percentage in the low single digits, possibly less than a full percent, you shouldn’t run for president as a third party candidate.
The Polling Mirage
No Labels insists they will only put forward a candidate if they see a feasible path to victory. But there is no path to victory. The arguments to the contrary are familiar, and they don’t withstand scrutiny.
For some time, Gallup has been asking voters if they would like to see a third major party. Affirmative answers routinely exceed 60 percent. No Labels has pointed to similar results when voters are asked if they’d consider a generic, unnamed centrist candidate in 2024. Others point to the fact that nearly half of Americans say they are independents, far surpassing those who self-identify as either Republicans or Democrats.
Add this to the prospect of two major-party nominees with historically low approval ratings, and a historic breakout for third parties might seem uniquely possible. But these poll numbers are a mirage.
For one thing, you don’t get to put a generic third party or an unnamed centrist on the ballot. Ask about an actual, real person and the numbers fall off fast. Especially if you ask directly if they’d support the candidate, not merely “consider” supporting them.
It’s easy to project frustrations with Biden and Trump, Democrats and Republicans, onto a blank slate. But specific candidates have actual policy positions and public records, including their prior party affiliation, and nobody pleases everybody. There’s probably not much overlap, for example, between possible No Labels voters and leftists open to voting for Cornel West.
So when you ask voters if they’d back Manchin in a three-man race, you don’t get a lopsided supermajority for the centrist revolution. You get a number more like 16 percent, as one recent Monmouth University poll found.
So You’re Telling Me There’s a Chance?
Polling in the mid-teens may seem like a reasonable starting point. It’s certainly much better than the vast majority of third party candidates have actually done. But history suggests that 16 percent is not what a candidate like Manchin or Hogan would actually get. A quarter of that is far more likely.
Gary Johnson’s 2016 campaign shows the general pattern. At his peak, he was polling into the low teens as he similarly made headway against two historically unpopular opponents, Trump and Hilary Clinton. In the end, the vast majority of those voters “went home” to their preferred major party and the Johnson/Weld ticket wound up with 3.2 percent.
While Hogan or Manchin might think he is simply a better politician than Johnson and Weld were, the same trend holds true for other notable third-party presidential candidates: They almost always poll better than they actually perform on election day. Similar trends can be seen with such candidates as Illinois Rep. John Anderson in 1980 and celebrity activist Ralph Nader in 2000.
Minor-party candidates often complain about being excluded from the polls, but pollsters aren’t just being mean. They know the results won’t be accurate as a predictor of actual votes cast. Asking about a candidate in a poll boosts them beyond what actually happens in the voting booth, where citizens will usually find a much longer list of obscure names. This is especially true when other relatively high-profile independent candidates, such as Cornel West and Robert F. Kennedy, Jr., will be vying for the small number of habitual third-party voters.
There’s also a selection bias that’s hard to weed out. People who answer polls are at least somewhat more interested in politics than the average voter, which means they’re more likely to support a lesser-known candidate.
Pointing to the number of supposed independents is similarly unhelpful. The vast majority of them still “lean” Republican or Democratic and will say so when pressed. The actual number of uncommitted swing voters is 10 percent or less, and no candidate can appeal to all of them.
These genuinely independent voters are not necessarily middle-of-the-road moderates. They often hold an eclectic basket of views that don’t easily fit onto the usual left-right spectrum. Whichever mutually exclusive subset of them you appeal to, you’re already down into the single digits, and that’s before fears of a wasted vote and a flood of anti-spoiler messaging whittle you down further.
Lincoln, the 12th Amendment, and Other Red Herrings
Sometimes those with inflated expectations for a third-party win call on historical precedent: What about Abraham Lincoln’s victory and the rise of the Republican Party—wasn’t that an example of a third party winning the presidency?
Even setting aside the many ways the political system was radically different then, calling Lincoln’s GOP a “third party” misunderstands the situation at the time. The Republicans simply were not the equivalent of a third party in 1860. They had been the runners-up in the previous presidential election. They had already won an effective majority in the House in 1858, and they had control of numerous state legislatures and governorships. They were the minority party in the Senate. And the Democrats’ previous major-party rival—the Whigs—had already collapsed and disbanded; they ran no one for president in 1860.
No alternative party today is anywhere close to that position. No speaker of the House is leading a No Labels majority. The runner-up in 2020 was Donald Trump, not Joe Manchin. Nobody has ever been elected as a No Labels nominee to any office. And both the Republican and Democratic parties are alive and kicking.
Another alleged path to victory for third parties relies on the 12th Amendment’s requirement that candidates must secure an absolute majority in the Electoral College to win. Otherwise, the task of picking the president goes to the House of Representatives in what is known as a contingent election. That hasn’t happened since 1824.
Over the years, many third-party candidates have seen this as a more feasible goal than winning by normal means. But on closer inspection, a contingent election is both unlikely and undesirable.
Even with the House voting by state delegations, an unusual quirk to the 12th Amendment’s rules, there is little reason to think a distant third-place candidate could emerge the victor in a House where the ticket has few if any supporters. How many Republicans and Democrats would feasibly defect to install a candidate neither party had nominated and who had run in third place?
The best that can be hoped for is to deadlock the House, where an absolute majority of 26 state delegation votes are needed to win. But if the House can’t make a decision by January 20, then the Senate’s choice of vice president will take office—and the Senate is allowed to choose only from the top two VP candidates in the Electoral College, with an absolute majority of 51 votes needed. If both houses fail to choose somebody, then the speaker of the House becomes acting president. If the House has successfully chosen a speaker.
If your goal is to break up the two-party system, it’s hard to see how you do that by making a Republican House speaker or Democratic vice president commander-in-chief. And those are both much likelier outcomes than any third-place candidate somehow coming out on top under the 12th Amendment.
Contingent elections are also deeply undesirable on the merits. They are often treated as a fun bit of constitutional trivia, memorably portrayed on the sitcom Veep, but the reality is much more grim. The procedure hasn’t been tested in two centuries. There are a number of ambiguities in the constitutional text, and the House has no established rules for how it would work.
If you thought the disastrous aftermath of the 2020 election was bad, it would pale in comparison to the constitutional crisis and political chaos resulting from a contingent election. It’s worth seriously questioning why anybody should want to see it happen. It could get very ugly. The House could deadlock over ambiguities in the constitutional rules, the Senate could come down to an argument about the vice president breaking a tie, and the whole mess could easily drag on past January 20—at which point we’d open an entirely new can of worms with possible disputes over the Presidential Succession Act. Assuming we got through the thicket of procedural uncertainty, the backlash against any result for a contingent election would be intense. This is what happened in 1824, when Andrew Jackson was rejected by the House but bounced back with a commanding victory four years later, driven in no small part by the perception of a “corrupt bargain” in Congress to lock him out of the presidency.
A third candidate triggering a contingent election is mathematically unlikely anyway. Playing with hypothetical maps, it’s easy to draw a scenario where a single state goes to a third candidate, thus denying anybody a winning majority. But that’s not what actually happens when there’s a significant three-way split in the popular vote. The leading candidate, perhaps with only 40-something percent, wins many more states on pluralities than is usual in a two-way race. The effect is not a deadlock but an Electoral College landslide.
It can also be tempting to think a candidate who has previously won elections in a state will be a kind of favorite son, winning his home state and at least having a shot at throwing the election to the House. Here the historical record is similarly unpromising.
Johnson, a popular two-term ex-governor, got just 9.3 percent in New Mexico in 2016—better than he did elsewhere, but not nearly enough to win the state. In Massachusetts, his similarly positioned running mate boosted the result above the natural average by only about a point. In 1980, Illinois Rep. John Anderson and former Wisconsin Gov. Patrick Lucey likewise saw negligible boosts in their respective home states. In 2008, when the Greens picked former Rep. Cynthia McKinney, she wasn’t even able to get on the ballot in her home state of Georgia. She won just 0.12 percent nationwide.
The home-state advantage is modest and has declined sharply in an era of highly nationalized politics. It might provide a key boost for a major-party nominee in a close race, enough for a crucial point or two. But an extra point or two doesn’t do much good when you’re starting from zero.
None of this is to say that third-party candidates shouldn’t run. They’re part of the ecosystem of American politics, and some have made a genuine impact. And haters of the two-party system are entirely right. It is a terrible system.
The usual anti-spoiler arguments will never be convincing to those who sincerely oppose the two-party system. Most of the attacks on No Labels have put this complaint front and center, alleging that a substantial third-party bid would pull more votes from Biden than Trump. That’s debatable. Democrats were wrong to think Johnson hurt Clinton more than Trump in 2016, for example. But it’s also beside the point.
Whatever the relative merits of the Republican and Democratic nominees in any given election, anybody considering running against both has already decided they’re both unacceptable. Hectoring third-party candidates and supporters for harming either major-party candidate only causes them to dig in their heels. It comes across as whiny, entitled, and undemocratic, and third-party voters are not wrong to perceive it that way.
But third-party presidential campaigns are not the way to smash the duopoly. Even when third-party candidates have won lower offices, as with Gov. Jesse Ventura in Minnesota and Gov. Wally Hickel in Alaska, those wins don’t do long-term damage to the major parties. Republicans and Democrats continue to dominate, winning almost every other election on the ballot. The two-party system doesn’t collapse even in the aftermath of the occasional fluke.
The most successful alternative parties are ones that don’t run their own presidential candidates at all. The Progressive Party in Vermont boasts a number of legislators and even statewide officeholders. The Working Families Party is a real force in New York.
But in these cases, the parties are able to take advantage of laws allowing so-called fusion nominations, where third parties can selectively cross-endorse (or not) major party nominees. The selective use of fusion allows smaller parties to grow and exert influence without being marginalized as spoilers. For example, New York’s Conservatives provided the winning margins for Ronald Reagan in 1980 and George Pataki in 1994. Because fusion allows parties to grow and be relevant, Conservatives were also able to stand on their own to elect James L. Buckley to the Senate, defeating both a Republican and a Democrat.
Fusion offers one way forward to a more representative multi-party system. Proportional representation, increasingly a focus for many electoral reformers, offers another route. Both fusion and proportional representation offer a way around Duverger’s law, the inherent futility of alternative parties in a first-past-the-post electoral system. And though it won’t necessarily mean a huge boost to third parties, the spread of ranked-choice voting offers a chance to do away with the “spoiler” effect.
With such reforms, votes no longer have to be wasted, spoilers aren’t seen as a problem, and everyone can be more accurately represented. We don’t have to be locked into two parties tearing the country apart with alternating bouts of minority rule. We might move beyond what George Washington presciently called “the alternate domination of one faction over another.”
But running for president with wildly unrealistic expectations won’t help. It will just leave people bitterly disappointed.