A really staggering share of the replies (*on twitter*, though conspicuously not on any other channel) essentially just assert that the media is singularly responsible for Biden's low approval rating
As best I can tell, this is founded on two basic arguments: 1) Many economic indicators (growth, unemployment) are good, inflation isn't especially bad, therefore perceptions of a bad economy must be driven by poor reporting and an unfair narrative
2) Biden's policies are popular, voters do/should judge policies based on policy, and therefore Biden's inability to claim credit for his popular policies reflects a media failure to give sufficient attention to the substance of his agenda
(I'll stuff Afghanistan-related commentary under #2 as well, since it was a popular policy goal that became unpopular, purportedly due to unfair media coverage)
I think some of these arguments are more plausible than others. And even some of the plausible ones are beyond anyone's ability to prove or disprove (how much of Biden's problem in Afghanistan is media v. a chaotic reality? I have no idea how to know)
From the standpoint of a political analyst, I think argument 1 is a lot more interesting and important than 2.
That's mainly because I think a core assumption of 2--that people mainly vote on views of the details of policy--is basically wrong.
Most people just don't think about policy like you all do, and I have no reason to expect the median voter to swing on a policy initiative if they think the economy is bad.
There's a reason why 'I have a plan for that' Warren was basically DOA beyond liberal college grads.
#1, otoh, is an important hypothesis. The role of the economy in politics is a really central finding in political science. If the media environment and/or polarization has severed the connection between economic reality and public perception, that's really important
And the case that the connection has been frayed is at least plausible. Most economic indicators support the perception of a fairly strong economy. And it's not the first time we've seen something odd in this area: approval ratings were less responsive to the economy from 08-20
But from 08-20, I do think public perceptions of the economy were fairly credibly tied to economic reality. It would be a step further if voters now reached a fairly erroneous conclusion about the economy
As an aside, it's worth noting that we had several months where the economic numbers undershot reality and required a significant upward correction. It's possible the perception-reality gap is based in biased coverage, but that bias stems from biased data not biased journalism
Another possibility is that this is a case where voter attitudes are catching *real* things that the numbers don't capture so easily, like supply chain issues, delays and worker shortages
"Business can't hire" is a real and very visible economic problem. So is the supply chain issue. I see no reason why it couldn't/shouldn't affect economic perceptions. But it's not one that's worked into the usual statistics; it's not an issue we know how to think about
Inflation is also a real issue, even if it's transitory. A household with 40k annual expenditures loses 2k in real dollars due to 5% inflation. There goes the stimulus check that you think should drive Biden's approval ratings up
And while some people get raises that keep up with inflation or change jobs, many people do not. Inflation almost invariably creates losers v. low inflation, even if real per capita income keeps up
So I'm very agnostic about whether voters are 'wrong' about the economy today--whether their perceptions of the economy are very different than what we ought to expect given the facts. I just don't know. This is obviously a highly unusual moment, coming out of COVID etc
Public opinion can also move fairly slowly. It's possible that, today, the economy is clearly back on track but the bad news from August-October continues to exert a big effect on the numbers. It will for a while.
Ultimately, I think the case that public perceptions of the economy are being driven more by weird coverage than weird conditions would be a lot better if... conditions weren't obviously so weird. The key will be whether 'normal' conditions ultimately yield normal numbers
• • •
Missing some Tweet in this thread? You can try to
force a refresh
A few thoughts on the disconnect between Biden's popular policies and his personal unpopularity nytimes.com/2021/11/27/us/…
TLDR... I think it is worth fully internalizing the century-long pattern of voters a) rewarding parties for presiding over peace/prosperity; b) punishing presidents for enacting ambitious agenda.
The obvious implication: a popular, ambitious policy agenda doesn't do you much good if there isn't normalcy/peace/prosperity, especially if that agenda is not seen as attempting to respond to immediate challenges at hand
I'm late to this @jbouie piece, but I do think it's worth coming back to a week later to reiterate how odd it is that HR1/FPA, for all their gangly ambition, don't include a republican government-based attack on state legislative gerrymandering nytimes.com/2021/11/12/opi…
A guarantee-clause based attack might be very different than what you usually hear about on this website.
It would narrowly establish that republican gvt = majority rule = can't draw maps that would thwart majority rule
Only one metric would logically follow from a republican government clause attack: the mean-median gap.
A lot of the usual anti-gerrymandering would be irrelevant in this framework, like commissions, efficiency gap, compactness, communities of interest, proportionality, etc.
We did a poll of Wisconsin just after the unrest in Kenosha. Biden led by 5, 48-43
It had some pretty interesting findings on the issue was playing out in the race, with a mixed bag for both sides int.nyt.com/data/documentt…
BLM fav/unfav: 53-42
Who handles protests best? Biden+8
Who handles race relations best? Biden+19
Who handles crime best? Even
Who handles law and order best? Even
What's more important: covid or law and order? COVID +1
On defund, voters opposed it by a huge margin (19% said defund, 40% maintain, 36% increase)
And voters thought Biden supported defund, 45-38 (including 22% of Dems and 38% of nonwhite voters)
They also thought Biden hasn't done enough to condemn rioting, 31% enough 56% not enough
It will be a while until we have authoritative data on turnout in Virginia, but at the moment I think it's fair to say two things, judged against 2020:
--Black/non-white turnout was weak for Democrats
--Otherwise, Democratic turnout was probably fine
For clarity, I've assigned the turnout among early voters to each precinct using voter file data, allowing for a direct comparison to 2020 turnout using vote history data
In overwhelmingly Black precincts, turnout was just over half of 2020 levels. It was at about 75% of 2020 levels in areas with no Black voters.
I said this to a few people on the phone so it may be worth adding publicly too: from the standpoint of electoral implications, one of the most important things about CRT (either IRL or caricature) is that it's a critique of liberalism from the left.
The implication: it lets certain GOPers be relatively liberal on race.
If CRT is raised to sufficient salience, GOPers don't have to rev up the base with outright conservative views--like anti-immigration or denial of police brutality--to polarize along racial attitudes
Of course, many Republicans will instead emphasize outright conservative views that alienate more voters. I'd guess one might have lost VA.
But for more moderate GOPers, CRT is a gift. They can bash the left and earn cred by merely sounding like... Obama '08
There are only four precincts where Hispanics make up a majority in Virginia, but they did pretty notably underperform our baseline for a tied election--perhaps suggesting continuing, disproportionate Democratic weakness with the group
It also seems like the turnout in majority Black and Hispanic precincts was relatively weak (see right hand column)
I know it's kind of annoying to judge compared to some vague 'expectations,' but it's sort of necessary in Virginia, where all of the absentee/early vote is reported by county, while precincts are all Election Day precincts