Fox, Trump and Millennial Movement

After a big news event, I almost always get asked about whether the latest polls show an earth-shattering shift in public opinion. They almost never do, and the handful of polls since the Trump indictment appear to be no exception.

But just because an event doesn’t yield a huge swing in the polls doesn’t mean that the event can’t or won’t matter. The indictment might ultimately fall into this category. For now, the way to tell whether it could eventually make a difference may not be to watch the polls, but to watch Fox News instead.

Even more than his loyal base of popular support, Donald J. Trump is protected by a wall of elites — conservative media commentators and politicians who forcefully defend the former president, attack his opposition and deter his rivals from going on offense. In some cases, they can even force Mr. Trump’s rivals to come to his aid!

To fundamentally change the race, the indictment would need to be so serious that it gradually corrodes these walls of elite support. It might start with his usual defenders acting a little less outraged. It might become acceptable to lightly critique Mr. Trump’s conduct. Maybe the former president’s most vocal defenders would concede his conduct was imperfect. Bit by bit, his skeptics might become more emboldened. Perhaps in time, his opponents would feel comfortable expressing outrage about Mr. Trump’s conduct — even if it’s outrage that he allowed himself to fall prey to the odious Deep State.

This wouldn’t fulfill the liberal or Never Trump fantasy of Mr. Trump’s imminent collapse, but it would create the conditions for a more competitive race.

I’ve been watching plenty of Fox News over the last few days, and I could make a case that something like this process could be underway. On the night Mr. Trump announced his indictment, the wagons were circled. There was outrage and talk of Hunter Biden, Hillary Clinton, Burisma, emails, political prosecution. Multiple Trump lawyers and staffers defended him. There was no room for criticism.

But after the indictment was released, the conversation became more mixed. Trump-friendly legal minds like Andrew McCarthy and Jonathan Turley, as well as Bill Barr, acknowledged the seriousness of the charges. Nikki Haley, who has generally been reluctant to critique Mr. Trump so far this cycle, said his conduct was “incredibly reckless” if true.

I don’t think any of this will show up in the polls anytime soon. I’m not predicting this trend will continue, either; for one thing, Fox concerns over viewer defections could mute criticism of Mr. Trump.

But if this trend did continue, the Republican primary electorate would begin to receive a much more mixed set of information. In time, it might begin to change their views on whether they want to see more Donald Trump or whether they’re ready to move on. They might decide they like what they’re hearing about another candidate a little bit better. Of course, they might also stick by Mr. Trump’s side until the bitter end — Mr. Trump has a commanding lead in the polls, and his support runs deep. But either way, you would expect to see erosion in Mr. Trump’s support on Fox News before you saw it in the polls.

More on millennials

While we wait for more information on the Republican side, I want to return to the topic from a recent post on millennials. If you missed it, it reported evidence that millennials — especially older millennials — seem to have shifted toward the right since 2012. I got a lot of questions about the article, and I thought they were interesting enough that they might interest a broader audience, so let’s dig in:

How did Biden do so well in 2020 if he’s doing worse among millennials?

It’s an understandable question. We showed many age cohorts shifting markedly toward the right after 2012, even though Joe Biden won by more than Barack Obama did. How can those two facts be squared?

Because of the scope of the article, we didn’t show every age cohort, so there are two main factors.

One is the advent of the new generation of voters who weren’t eligible to vote in 2012: everyone who, today, is age 29 or younger. They’re even more liberal than the millennials.

A second factor is that Mr. Biden fared much better than Mr. Obama among voters born before 1960 — those who were at least 60 years old in 2020 or 52 in 2012. These cohorts lurched to the left between 2012 and 2016 and yet again between 2016 and 2020.

In the end, these two factors — the ascent of the Zoomers and a leftward shift among older cohorts — slightly canceled out Republican gains among millennials and Gen X, with Mr. Biden ultimately winning the election by 4.4 points compared with Mr. Obama’s 3.9-point victory.

Is it just about turnout?

One common question was whether all of these shifts might just be about turnout, with few really changing their minds. After all, Mr. Obama was thought to have benefited from a strong turnout among young voters.

The data we reported — a compilation of tens of thousands of interviews from high-quality surveys archived at the Roper Center — suggests that people really did change their minds. Nearly all such data was collected among all adults, not just registered or likely voters. In theory, the survey results among all adults ought to offer a clean look at how cohorts shifted, without any effect from turnout. And at least in this data, about 25 percent of the move toward Republicans is attributable to shifting turnout. The rest would be attributable to changed minds.

Most of that shift in turnout is among Black voters. In these surveys, the Black share of likely voters among millennials eligible to vote in 2012 fell to 10 percent from 14 percent between 2012 and 2020. This would have reduced Mr. Obama’s support among millennials by about 1.5 points. The effect is more muted for Mr. Biden since he won Black voters by a smaller margin than Mr. Obama did.

Is it just about Obama?

Mr. Obama didn’t just benefit from a strong turnout, of course. He was beloved by younger voters in a way that was absolutely not the case for Mr. Biden. Is it possible that the only thing that’s changed over the last decade is that Democrats went from nominating someone beloved by young people to someone with relative strength among older Americans?

Yes, that’s a possibility. With this data, it’s hard to know whether the Democrats’ showing with younger voters reflects a strength that was unsustainable or a growing underlying weakness. It’s also not possible to untangle the effect of the Republicans’ nomination of Mr. Trump compared with Mitt Romney.

But there’s one other opportunity for benchmarking shifts among millennials: John Kerry in 2004. According to this data, Mr. Kerry won 57 percent of those oldest millennials who were eligible to vote in that election, then ages 18 to 23. Mr. Obama then won that cohort with 61 percent in 2012, a clear improvement, but not quite a sign of truly special appeal. By 2020, Mr. Biden won just 54 percent of these once-young voters.

Is it just your data?

The figures from the last newsletter come close to the average among the various election studies with data back to 2008 or 2012: the American National Election Study, the Cooperative Election Study and the Catalist estimates.

It’s a little hard to make a simple, exact comparison between the data I’ve used and the other election studies, because the other sources don’t always offer the same age categories. But the simplest way is by looking at the categories they do have in common: those 18 to 29 in 2012, and 30 to 44 in 2020. No, they’re not the same groups of people. In fact, nearly half of the 30-to-44 group in 2020 was already older than those 18 to 29 in 2012.

But on average, the studies show nearly the same findings. The other studies show Mr. Obama winning 62 percent of those 18 to 29 in 2012, while Mr. Biden won 56 percent of those 30 to 44 in 2020. In our recent newsletter, we showed 63 percent for Mr. Obama and 56 percent for Mr. Biden for their respective groups. And the various more detailed cuts — when available — continue to show close similarity. This suggests they would probably also show similar findings for the more exact age cohorts in question.

Of the various studies, the exit polls show the largest shift (they actually show Mr. Trump coming close to outright winning those 30 to 44 in 2020). On the other hand, the C.E.S. barely shows a shift at all (in fact, it shows Mr. Biden outrunning Mr. Obama among young millennials). The spread in the data is a good reason to be cautious about what’s going on, even if the balance of evidence tells a fairly clear story.