Abortion-rights ballot measures may not help Democrats as much as they think

In the 2022 midterm elections, Democrats defied pundits' predictions of a "red wave" (though the polls didn't really support those predictions) thanks in large part to the motivating power of the issue of abortion. Not only were reproductive rights popular, but they also became more important to many voters in the wake of the Supreme Court's decision in Dobbs v. Jackson Women's Health Organization.

Two years later, Democrats are once again hoping that abortion will be their secret weapon at the ballot box. Specifically, many believe that pro-abortion-rights ballot measures can boost Democrats' chances of winning a handful of key Senate races and presidential swing states.

This November, at least five states — including Florida and Nevada — will vote on proposed constitutional amendments to enshrine abortion rights into their state constitutions. And as of July 15, efforts were actively afoot to get such measures on the ballot in five other states, including Arizona, where advocates recently submitted more than twice the number of signatures required to qualify for the ballot.

Most, if not all, of the measures that make the ballot are likely to pass; abortion rights are popular, and liberals have won all seven abortion-rights ballot measure elections that have taken place since Dobbs. But there's no guarantee that their success will rub off on Democratic candidates. In 2022, plenty of people voted for both abortion rights and Republican politicians. Nor is there strong evidence for the theory that abortion-rights ballot measures will turn out new Democratic voters. And while these plebiscites could help Democrats by putting abortion at the front of voters' minds when they fill out their ballots, that's something Democratic campaigns will probably try to do anyway, regardless of whether there's a measure on the ballot.

First, it should be obvious that not everyone who votes for an abortion-rights ballot measure will automatically vote for Democrats elsewhere on the ballot. Polls indicate that a sizable chunk of Republicans support abortion rights, at least in some cases, and of course, voters weigh several different issues when deciding who to vote for. Sure enough, every time abortion-related ballot questions shared the ballot with other races in 2022, the pro-abortion-rights side of the measure got more support than the Democratic candidate at the top of the ticket.

But that doesn't mean these abortion measures didn't inspire more people to vote for Democrats than otherwise would have. Proponents of this theory often make the "reverse coattails" argument: that abortion-rights ballot measures increase turnout among liberal voters, who will vote for Democratic candidates elsewhere on the ballot while they're at it.

Many Republicans believe this phenomenon helped President George W. Bush win reelection in 2004, when 11 states also voted on ballot measures banning same-sex marriage. The theory is that those ballot measures juiced turnout among socially conservative voters, who then also supported Bush. However, research into that election has generally found that those ballot measures did not increase turnout, either overall or among conservatives specifically. For example, a University of Florida study found that anti-same-sex-marriage ballot measures in Ohio and Michigan did not have a significant impact on turnout in heavily evangelical counties; in fact, in Michigan, support for the ballot measure was actually associated with slightly lower turnout.

Of course, it's possible that, 20 years later and with a different issue, things would be different. As Aaron Blake pointed out in The Washington Post, same-sex marriage was illegal in all those states in 2004, so these ballot measures didn't actually change the law. By contrast, the pending ballot measures in Arizona and Florida, as well as Arkansas, Missouri, Nebraska and South Dakota, would actively expand abortion rights. (However, equally noteworthy, the proposals in Nevada, Maryland and Montana would likewise just reinforce the status quo.*)

But while interest in these measures will undoubtedly be high this November, there just isn't much evidence that they'll turn out many voters who weren't already going to vote anyway. If abortion-related ballot measures were the main thing on the ballot driving turnout, we'd probably expect the number of votes cast in those races to be greater than the number cast in other races on the same ballot. But in every state with an abortion-related ballot measure in 2022, more votes were cast for the office at the top of the ticket than for the ballot measure.

"But maybe all the new voters just filled out their whole ballot!" you might argue. It's possible, but if that were the case, we should see a spike in overall turnout in these states compared with their turnout in previous elections. And that did happen in some of these states in 2022 … but not in others.

Midterm elections always have lower turnout than presidential elections, so to make for a fair comparison, I looked at how much higher or lower each state's turnout rate** was compared with the national average in 2022 versus in 2020 and 2018. The results were really mixed. Kentucky's turnout rate, for example, consistently matched the national average all three years. Montana's 2022 turnout rate was a bit higher than 2020's (relatively speaking), but it was significantly lower than 2018's, when the state had a hotly contested Senate race. And California's relative turnout rate actually fell in 2022 compared with the two prior elections.

The two states that had significantly higher turnout in 2022 than the historical norm were Michigan and Vermont; while both states tend to have above-average turnout even under normal circumstances, they had some of the best turnout rates in the whole country in 2022. However, this may not have been because of their abortion-rights amendments — or at least not only because of them. Both states also made it significantly easier to vote ahead of the 2020 election: Vermont started automatically mailing ballots to all voters, while Michigan enacted automatic voter registration and same-day voter registration. (It's interesting that there was no corresponding turnout spike in either state in 2020, but it's possible that the impact of these reforms was blunted by historically high interest and turnout in that election.)

Basically, the data on whether these abortion-rights ballot measures goosed turnout is inconclusive. It seems plausible that it genuinely helped in Michigan, whose initiative also happened to be the most impactful; it resolved an active legal dispute over the legality of abortion in the state. But we can't ignore other factors, especially when Vermont — whose pro-abortion-rights amendment did nothing but reinforce the status quo — saw a similar turnout bump and a state with the exact same type of amendment (California) didn't. The common link seems to be voting reforms, not the abortion questions.

It's hard to chalk it up to some ballot measures having more active campaigns behind them, either. There is only a weak relationship between the relative turnout increase in each state and how much money was spent (per capita) on its abortion-ballot-measure campaign. Michigan saw the biggest increase with the highest amount of spending ($8.34 per registered voter), but Vermont wasn't far behind despite only $1.30 getting spent per registered voter there.

And in 2024, there's an additional complication: It's a presidential election year. Presidential races always have higher turnout than midterm elections (nationwide, turnout was 20 percentage points higher in 2020 than it was in 2022), and if abortion-related ballot measures couldn't attract more votes than gubernatorial and congressional races in 2022, then interest in them in 2024 will almost certainly be dwarfed by interest in the presidential race. More money was spent on Michigan's abortion-rights proposal than in the governor's race in 2022 ($68.1 million to $43.5 million), but it's unlikely that will be the case with any state hosting a competitive presidential race this year, like Arizona or Nevada (though it probably will be in states like Arkansas or South Dakota). Finally, even if some voters in 2022 didn't care about the other races on the ballot but turned out anyway because of abortion-related ballot measures, there will probably be fewer of that type of voter in 2024, since only the most stubborn nonvoters sit out presidential elections.

At the end of the day, it's most likely that, if these ballot measures do help Democrats, it will be the old-fashioned way: through persuasion. Even if it's not as much as the presidential candidates spend, the campaigns for these ballot measures will probably drop millions of dollars on ads that highlight abortion — raising the salience of an issue that is good for Democrats. Poll after poll has found that Americans trust Democrats more than Republicans on abortion and reproductive rights.

And if Democratic candidates in these states lean into the issue, there is evidence they could reap the benefits. Going back to the 2004 election, the University of Florida study did find that Bush performed unusually well in counties with large evangelical populations in Ohio, where he campaigned hard against same-sex marriage, but not in Michigan, where he didn't. On the other hand, abortion-rights ballot measures are a no-win proposition for Republican candidates. They can try to avoid talking about the measures, but they may eventually be forced to make a difficult choice: either take an unpopular anti-abortion-rights position or tick off their base by supporting the measure.

Of course, given their strength on the issue, Democrats would be wise to spend the fall campaign talking about abortion regardless of whether they are sharing the ballot with a measure on it — and they probably will. In 2023, Democrats spent millions of dollars on TV ads about abortion in states like Kentucky and Virginia where it wasn't directly on the ballot. And in the first four months of 2024, a majority of their ad spending on broadcast TV in Michigan, Arizona and Georgia was about abortion.

Abortion-related ballot measures can certainly help make reproductive rights an even bigger part of the conversation. But Democrats shouldn't overestimate their impact. Based on the evidence we have, it doesn't look like they are going to motivate many new voters, and not every voter who supports abortion rights is automatically going to vote Democratic. If Democrats want to ride the wave of these measures in 2024, they'll have to put in the work.

Footnotes

*Abortion is also already legal in Colorado with no gestational limit, but the ballot initiative there would allow the use of state funds for abortion coverage, which is currently illegal.

**Based on the voting-eligible population.