Democrats Learned the Wrong Lessons From Losing Virginia

The question persists: Why did Glenn Youngkin win in a blue state? Photo: Andrew Harnik/AP/Shutterstock

Understandably, the loss of three statewide contests in previously blue-trending Virginia last November was shocking to Democrats. Interpretations of what went wrong there are having a heavy effect on how the party perceives its weaknesses heading into the 2022 midterms, so it’s pretty important to get it right — but it doesn’t seem that’s what’s happening.

The prevailing conventional wisdom has been that Republican Glenn Youngkin won and Democrat Terry McAuliffe lost because suburban swing voters upset about education and, to a lesser extent, economic issues switched from voting Democrat to Republican between 2020 and 2021. Indeed, a lot of influential focus-group work on the election began with the assumption that these voters made the difference and tried to interpret why they swung rather than how far they swung and how much it mattered. And the more that analysts dwelled on education issues as crucial, the more they agreed that school closures during the COVID-19 pandemic may have damaged Democrats in Virginia and New Jersey (where Democratic governor Phil Murphy had a surprisingly narrow winning margin) even more than Republican attacks on the alleged teaching of “critical race theory” or other culture-war topics related to schools.

Now comes a new study from the data-analysis firm TargetSmart that calls this narrative into question even more than past dissents. As TargetSmart CEO Tom Bonier notes, comparing the Virginia results to data on school closures calls into question the idea that the latter affected the former:

Of the top 10 counties in Virginia ranked by days with in-person education during the 2020-2021 school year, 6 of the 10 saw a larger swing towards Republicans than the state average swing of 5.3%, while the remaining 4 counties saw a slightly below average GOP swing. In fact, the biggest swings towards Republicans occurred in southwestern Virginia, where schools were open for in person instruction for most of the year.

Conversely, those counties that conducted virtual learning for most of the 2020-2021 school year saw a smaller shift towards Republicans than the state average — the top 10 counties for days spent in virtual learning in 2020-2021 saw a 3.8% average swing towards Republicans, well below the statewide average of 5.3%.

More generally, TargetSmart took a look at the voter-file information recently made available by Virginia and drew attention to some rather dramatic turnout numbers that seemed to suggest parents of school-age children in the Washington and Richmond suburbs weren’t the keys to this election:

Turnout among voters age 75 or older increased by 59%, relative to 2017 while turnout among voters under age 30 only increased by just 18%. Notably, turnout of all other age groups combined (18-74), which would likely include parents of school-aged children, only increased by 9% compared to 2017.

These are massive changes in the electorate in an election that was far from a blowout: Youngkin won by just 2%.

It’s common for seniors to turn out to vote significantly more than younger cohorts in non-presidential elections. But the figures for Virginia in 2021 were unusually large:

Voters age 65 and older are an estimated 15.9% of Virginia’s population according to the census, yet accounted for 31.9% of all ballots cast in 2021.

348,314 more seniors (ages 65+) voted in Virginia’s 2021 gubernatorial election than in the 2016 presidential election.

TargetSmart calls it a “silver surge.” Whatever you call it, it seems to suggest that variable turnout patterns rather than swing voting was the biggest deal in Youngkin’s win. It’s also what a December analysis in FiveThirtyEight of precinct-level data showed, indicating that the big net gains by Youngkin were in Democratic- and Republican-base areas, not in highly competitive swing areas. And for that matter, that’s what the much-discounted exit polls suggested, as Ron Brownstein pointed out right after the election:

Compared with the 2017 governor’s race, or the 2020 presidential contest in the state, the electorate Tuesday was older, whiter, less college-educated, and more Republican, the exit polls found. Census figures show that voters of color have increased as a share of the state’s eligible voter population since 2017, but in the exit polls nonwhite voters plummeted from about one-third of the electorate in both 2020 and 2017 to only a little over one-fourth this year. Voters under 30 fell from 20 percent of the vote in 2020 and 14 percent in 2017 to just 10 percent Tuesday. College graduates shrank from nearly three-in-five voters in 2017 to just under half. And although Democrats outnumbered Republicans by 11 percentage points in the 2017 electorate, the exit polls found that GOP voters almost exactly equaled them this year.

None of this is to say that Youngkin’s victory over McAuliffe was some sort of aberration or that it shouldn’t alarm Democrats. But what it takes to boost turnout by Democrats without further boosting turnout by Republicans is not the same as what it takes to persuade a narrowly defined suburban swing vote upset principally about schools. Narratives matter, and Democrats should take care to ensure they aren’t telling themselves the wrong story.