Testing the Vibes Hypothesis

Some suggestions for how to slice the polls

The New York Times has a big new poll out of the electorate with some very sobering data on public perception of the economy. They polled voters in six battleground states—Arizona, Georgia, Michigan, Nevada, Pennsylvania and Wisconsin—and along with asking them about their views of President Biden and former President Trump, their votes in 2020 and their preferences for 2024, and their views on various issues, they asked what their impression was of the health of the American economy. As you’d expect from prior surveys of this sort, those impressions are significantly out of line not only with aggregate economic data—strong growth, low unemployment, significant wage gains and dropping inflation—but with people’s assessments of their personal economic situations and with their objective behavior. The data suggests the economy is quite good, people say their own situation is good, and in general people are spending and investing as if the economy were strong. But when asked, they say it’s at best fair, maybe downright poor.

I wrote on this topic a month ago, and gave a whole bunch of possible explanations for the divergence we see. I won’t rehash those here; feel free to read that piece of you want my analysis.

The latest poll, though, gives us some new data. First of all, it reveals just how badly Biden voters specifically perceive the economy to be doing. Only 5% of those surveyed who voted for Biden in 2020 think the economy is currently excellent, and another 33% think it is good. That compares with 36% who think it is only fair and 25% who think it is downright poor. If those were the numbers for the electorate as a whole, it would be a warning sign for Biden. That those are the numbers for his own voters in battleground states is vastly more alarming.

But breaking the poll down by various demographic factors is revealing in other ways:

The age gap in perception is by far the most striking fact to jump out of the data. Biden voters under 45 have overwhelmingly negative views of the Biden economy; voters over 45 are considerably more sanguine, even positive. The racial gap is also notable, with White Biden voters much more positive than Black or Hispanic voters. The gap between lower-income voters and higher-income voters is probably the least surprising; even in good times—and wage gains have been especially strong at the lower end of the income spectrum lately—things are always tougher for those earning less, and inflation takes the biggest percentage bite out of the smallest incomes. But the gender gap, though relatively small, is still noticeable and probably the hardest to explain based on fundamental data. Why would the economy be worse for women than for men?

One explanation for the gaps in perception is the fallacy of composition. Looking at the electorate as a whole in those battleground states, rather than just at Biden voters, perceptions of the economy do not vary significantly by race, sex or income. Most White voters picked Trump in 2020, though, while most female voters picked Biden. The gaps in perception between different demographic components of the Democratic coalition, then, may simply reflect those groups’ differential propensity to vote Democratic regardless of their views of the economy, rather than an actual difference in perception between different groups.

That explanation isn’t adequate for the age gap, however. The youngest voters are by far the most negative on the economy, and the oldest voters are by far the most positive, both among Biden 2020 voters and the electorate at large. It’s a genuine and profound divide, and far too large to be explained by young voters’ greater propensity to vote for Democrats in general. What can explain it?

I can think of two broad explanations, one fundamentals-based and one vibes-based.

The fundamentals-based explanation is that it is all about housing and interest rates. Seniors living on a fixed income should be very vulnerable to inflation eroding their standard of living, so it’s hard for me to believe that they’d be more sanguine, in aggregate, about inflation than young people earning rising wages are. But older folks are also more likely to be homeowners and less likely to be borrowers, and housing costs are a much larger percentage of most people’s budgets than other expenses. Housing prices have held up very strongly overall, but the more dramatic driver of higher costs has been the backup in mortgage rates. Average 30-year mortgage rates have gone from under 3% in early 2021 to nearly 8% recently. That kind of spike quite plausibly prices many people out of the housing market entirely, or forces them to look at much less-attractive housing options. I wouldn’t be surprised at all if that had a profound impact on at affected people’s views of the economy.

To be clear: current mortgage rates are far from unprecedented. They’ve backed up to roughly where they were in 1999, and the economy in 1999 was very good. But the housing market is much more expensive now than it was in 1999, and, probably more important, the backup in rates was very substantial and very rapid. In percentage terms, it’s a bigger move than from the 1977 low to the 1980 high, and it happened faster as well.

To test the above, I’d love to see polling data on the economy sliced by homeownership status, to see whether there’s a notable difference in perceptions of the economy on that score within each demographic bucket. I doubt the data is available, but if it is it might shed some light. Also, it might be worth polling people to ask what the biggest problem in the economy is right now, and see how many choose housing costs. I’ve seen data saying that inflation is voters’ biggest concern, but the usual gloss on that is that voters are concerned about gasoline and groceries, not the overall price level. Housing is handled very peculiarly by the inflation statistics, though, so it might be worth asking about housing specifically to figure out if that’s what really has people concerned.

The vibes-based explanation, meanwhile, is that Biden is suffering from rising expectations from within the core of his coalition. Republicans’ perceptions of the economy are clearly affected by political polarization—they believe things are going better when they are in charge and worse when the other team is charge. Moreover, the disconnect between their behavior and their stated perceptions can be reconciled to some extent if you presume that Republicans, when they talk about the economy, are really talking about their expectations for the economy. Democrats, historically, may have had a similar and offsetting effect—their perceptions of the economy certainly dropped suddenly when Trump was elected in 2016, and rebounded quickly when Biden was elected in 2020. But it’s possible this symmetry may have broken, inasmuch as Democrats had specifically outsized expectations for what the Biden administration could and would achieve that have not been met. If that’s the case, then Democratic perceptions of the economy could easily have been affected by that disappointment in much the same way that Republican perceptions are affected by their negative feelings about the inevitable consequences of Democratic governance.

I’m reluctant to believe that this is true, because it would imply that Democrats have become negatively polarized against themselves, which would be politically disastrous if true. But it is suggestive that the core elements of the Democratic coalition—women, non-Whites, young people, lower earners—have a more negative view of the economy than the less-core segments. It’s also reasonable to hypothesize that younger voters would have more outsized expectations than older, more experienced voters. So it doesn’t strike me as impossible that something like this is happening.

I can think of a couple of ways to test the hypothesis. One would be to screen by ideology as well as vote. Do strong liberals/progressives have more negative views than moderates, or the other way around? If so, that would be suggestive of the rising expectations/disappointment hypothesis; if not, that would be contrary evidence. Another would be to ask voters whether they blame Biden for not doing enough or for doing the wrong things. Republicans, for example, think that the American Rescue Plan is a major culprit in having caused inflation. Do Democrats with negative views of the economy agree? Or do they think the opposite, that the problem is that Biden hasn’t done enough to boost their incomes or control prices? I’m not expecting the typical voter to have coherent or sensible views on economic policy—but the difference between “didn’t do enough” or “did the wrong stuff” could help confirm or disconfirm my “rising expectations” hypothesis anyway.

Those are my thoughts. If anyone knows whether the data is available, I’d love to hear about it.