The Washington PostDemocracy Dies in Darkness

Young people say they plan on voting in November — but they usually don’t

Students wait in long lines to register to vote in April 2016 in Eau Claire, Wis. (Dan Reiland/Eau Claire Leader-Telegram/AP)

Harvard’s Institute of Politics regularly surveys young Americans on their political views. They’ve been doing this for a while. So it’s little surprise that those responsible for the release of the spring 2018 iteration Tuesday knew exactly what was likely to attract the most attention.

“A new national poll of America’s 18- to 29-year-olds by Harvard’s Institute of Politics (IOP), located at the Kennedy School of Government, finds a marked increase in the number of young Americans who indicate that they will ‘definitely be voting’ in the upcoming midterm Congressional elections,” the report reads. “Young Democrats are driving nearly all of the increase in enthusiasm; a majority (51%) report that they will ‘definitely’ vote in November, which represents a 9-percentage point increase since November 2017 and is significantly larger than the 36 percent of Republicans who say the same.”

That bit of data overlaps with a number of other reports, both anecdotal and substantive. In poll after poll, Democrats report being more energized about the midterm elections. In report after report, young people have described being energized, many as a result of gun-control activism that emerged after the mass shooting in Parkland, Fla., in February. In California, more than 100,000 youths under 18 have preregistered to vote, allowing them to be official voters as soon as they turn 18.

It is not our intent to damp the enthusiasm of any voter in a country where midterm elections typically draw well under half of those eligible to the polls. Hopefully registered voters will vote! That’s the point. But we do have to note that one should not assume that the details in the paragraph above necessarily mean an exceptional flood of young Democrats to the polls. Nor does the Harvard IOP poll.

As we noted, these polls are released regularly, so we can compare how often young people said they definitely planned to vote with the percentage who actually did. The Census Bureau compiles data on voter turnout by age, but we generally use data from the United States Elections Project which corrects for statistical quirks in the data.

In each year since 2004 for which the Harvard IOP poll asked about likelihood of voting, fewer young people actually went to the polls. In 2004 and 2008, the percent who turned out was far lower than those who said they definitely would. In recent years, the difference between the two has been about seven percentage points. It’s the highest percentage reporting that they plan to vote of recent midterm election years — six points higher than the value in 2010 and 14 points higher than four years ago.

Actual turnout among those under 30 has been about 26 points lower than the turnout rate of those 60 and older in presidential election years — and about 38 points lower in midterm years.

The net result is that those 60 and older make up much more of the electorate than those under 30. While turnout among younger voters is substantially higher in presidential election years, the shift in the percent of the electorate that’s under 30 is more subtle.

It’s important if young Democrats are more motivated to come out and vote than young Republicans, of course. But what does history tell us about how many young people will actually vote this year?

If the Harvard IOP data continues to be about seven percentage points higher than actual turnout, that means that turnout among those under 30 will be about 30 percent. That’s somewhere in the middle of typical midterm and presidential turnout numbers. It suggests that about 14 to 15 percent of the electorate would be under 30 — or about one out of every seven voters.

Not insignificant. But in every election since 1992, those 45 and older have made up more than half of the electorate. In recent elections, they’ve typically voted more heavily Republican. And 30 percent turnout isn’t good. It means, necessarily, that about 70 percent of young people will end up staying home.

A surge in young voters would shift the 2018 results. But the surge suggested by that Harvard IOP poll may be more modest than it seems at the outset.