The Vibes are off. Are we measuring them correctly?
Plus a quick look at where people put MLK and Hitler on a left-right political ideology scale.
Hello friends! Welcome to the first Pulse of the Polis for 2024!
Today, we’ll be looking at some new research by Gallup about levels of flourishing in the United States and other research on how Americans are faring.
I am far from the first person to note that the overall Vibes™ of US adults seem to be Off™. Although we have a lot of messaging hyping-up developments in the economy, it seems that the American people are not accepting this message. I’ve seen some refer to this as “Vibes vs Data” or a “Vibes Paradox”. I’m going to use this research as a springboard to take a stab at what the disconnect is. There are, of course, a few nuggets at the end as well. I hope you enjoy!
New Normal of Lower U.S. Life Ratings Persisted in 2023 | Gallup
Dan Witters and Kayley Bayne for Gallup present the work of Gallup’s quarterly National Health and Well-Being Index. Since 2018, every wave of the survey has been fielded to approximately 5,000 US Adults a dual-mode methodology (phone and online1) and asks people to rank on a 0-10 scale (with 10 being “the best possible life for you” and 0 being “the worst possible life for you”) their current lives as well as what they imagine their future lives will look like in 5 years. If they rank their current lives as a 7 and their future lives as an 8, Gallup has categorized these individuals as “Thriving”; if they rank their current and future lives as a 4 and below they are categorized as “Suffering”; the remainder are categorized as “Struggling.”
The most recent iteration of the Gallup data shows that the percentage of people “Thriving” are up modestly from 48.9% last quarter to 52.2%. 4.7% are classified as “Suffering” compared to 4.8% in 2022 and 3.3 % in 2021. Unsurprisingly, these figures vary based upon one’s political party: the percentage of Independents categorized as “Thriving” has diminished by 2.6 percentage points compared to 2021 and 9 percentage points for Republicans. Democrats have more-or-less remained constant (growth of just 0.4 percentage points).
There are two things that surprises me the about this chart. First is just how consistent it seems to be. I’ll just be focusing on 2020 onwards since that’s when the last major methodological change is reported to have happened. The gap between the lowest point—46.4% “Thriving” in April 2020 with tens of millions unemployed and the emergence of a once-in-a-century public-health emergency with very actionable information2—and the highest point—59.2% in June 2021—is rather narrow: about 13 percentage points. Second is the volume of people “Thriving:” The nation was economically shutting down and nearly half of Americans were “Thriving?” That doesn’t strike me as particularly plausible.
The charts’ Vibes are off about our Vibes
On the one hand, this is probably a function of carrying forward a scoring system, methodology, and public-facing naming convention that had been developed nearly 15 years prior. I doubt that anyone would’ve chosen the term “Thriving” had the scale been deployed for the first time in 2020. Which actually reflects a deeper methodological point about ensuring scale invariance: Are we certain that “Thriving” and “Suffering” means the same in 2008 as it did in 2015? 2020? 2024? I wasn’t able to find any research on the subject3. It’s possible that research on this exists and I simply wasn’t able to find it. But it’s actually not my larger point today. Rather it’s my reaction to seeing numbers this stable and this high. My doubt and suspicion. Because, to me, as with so many others, the Vibes are off—yet these data seem to contradict this.
My incredulity mirrors a lot of what we’re seeing play out in the press. Measures of “wellbeing” appear to suggest that things are pretty good: the inflation rate is down dramatically from its peak, the economy continues to grow and add jobs, real wage growth is up compared to pre-pandemic times (though not necessarily from it’s own pandemic-era peak). But people are pessimistic! They’re anxious about a variety of social, technological, and political developments and they seem to be distrustful of the economic news; they’re waiting for the shoe to drop. Some are stuck on a broken doomsday record: The recession hasn’t happened yet but boy it will, bet your bottom dollar on it.
Are people fine, actually?
One emerging answer to this conundrum is that, actually, people are doing pretty ok if you ask them so outright. This is kind of what the Gallup data would suggest: Most people are actually not “Suffering” or “Struggling”, the preponderance seem to be subjectively ranking themselves as doing pretty good. And people tend to be pretty optimistic about their own lives and finances. Another characteristic survey in this genre, recently reported by Axios, shows that 63% of Americans “rate their current financial situation as being ‘good’” and “66% think that 2024 will be better than 2023.” People are spending more on goods than they were a year ago, including things that are unambiguously not necessities for the vast majority of people such as boats. Holiday spending for 2023 also hit new highs, despite expectations of consumer withdrawal vocalized just weeks prior.
This sort of “covert optimism” might also explain some other weird things going on with consumer behavior. Why on Earth would people continue spending loads and loads of money—reducing their personal savings and dumping increasing amounts into high-interest credit card debt—if they weren’t expecting to be able to offset these present expenditures with future earnings? If people are really expecting the economy to dump into the shitter, why are they burning through money like a 22 year-old on their first trip to Vegas?
There’s a couple things complicating this “people are actually fine” narrative for me though. First and foremost, the narrative itself seems to depend on what survey you’re reading. For example, this Bankrate survey shows that 50% of Americans rate their current economic situation as worse than it was in November 2020. Only 20% of that sample say that it’s improved. This AP-NORC survey shows that the preponderance of respondents (roughly 70%) say that, compared to a year ago, their household expenses have risen, 40% say that they have more debt, 50% say they have fewer savings. And only 23% say that their income has gotten higher to offset any of these challenges.
But the main thing that I think is at the heart of this confusion is a measurement issue. Specifically, I think it’s an issue with the reference times and categories these measures are anchored to compared to what people are anchored to.
What are we actually referencing in this “paradox?”
In the case of the “everything is actually pretty decent” camp, people are frequently making references to the pre-pandemic era. Take this article from American Progress which shows that wages and prices “have grown since the onset of the COVID-19 pandemic, but wages have grown more” comparing prices and wages from Q4 of 2019 to November of 2023. To a certain extent, it makes sense to want to compare things to how they were Pre-Covid because things in Covid got weird4. But doing so neglects the fact that people are actually really good at adapting; weird can quickly normalize and anchor our expectations. And people’s expectations about what their spending should feel like are not from nearly half a decade ago. And, even if it were, the choice of Q4 2019 reflects a blip from the general trend observed from just a few months later on through today: Real median annual household income has been consistently going down since 2020. 2020, 2021, and 2022 all feel like more reasonable places for people to be basing their current situation off of than Q4 of 2019.
In the cases of pointing to increased consumer spend, a lot of the reporting about how people are “spending more” on XYZ appear to be conveniently ignoring that prices have gone up; I have never once seen an “inflation-adjusted” holiday spend metric. And the increase in spending may not reflect secret optimism that people are hiding from the pollsters5, but rather a combination of higher prices and lifestyle creep that happened due to increased spending on goods during the pandemic where services were unavailable or restricted. Lifestyle creep that, remember, people appear to be offloading onto credit cards. And I would bet a sizeable portion of my own debt that people aren’t doing so because they’re actually secretly confident, but because they feel like they have to. And they could be right! The pandemic restructured a lot of lives. I have no doubt that people have purchased luxuries and non-essentials—we are not a nation of ascetic monks—but I also trust that the majority of people are not mindlessly blowing through their cash. I suspect that people feel like these are choices they genuinely felt like they had to make.
There’s also the issue that a lot of economic news discusses increases/decreases in various rates and percentages. But a decrease in a positive inflation rate is still an increase in prices—just a smaller increase compared to previous increases. And though wages are increasing faster than inflation, they haven’t caught up to inflation if we think back to 1-3 years ago. So people may have very good reason to be optimistic about the future, but, in this moment of time, they are still worse off than they were a year or two before. This before we talk about the political and social anxiety that’s happening, you know, everywhere.
Unraveling the paradox
To be glib, it seems to me that the solution to the Vibe paradox is that there is no Vibe paradox! Things have gotten more expensive, people have gotten more into debt, their real incomes have not kept up with the costs of housing, food, and the most basic of creature comforts compared to the time period that people are actually likely to be anchored to. The fact that leading indicators suggest that things will get better may fuel some of the surveyed optimism, but the fact that things aren’t better yet underlies a lot of the currently existing pessimism.
So why are people saying that, actually, their lives are pretty decent (categorizable as “Thriving” even, should one so choose)? Even that AP-NORC survey I mentioned earlier showed that about 55% of people said that their overall financial situation is “Very/somewhat” good, and 75% say that they are at least “somewhat” confident that they can keep up with expenses.
All this optimism after all of the increased debt and static income levels?! What the hell gives, survey respondents?!6
Here too, I think (though admittedly much less confidently), we are dealing with a reference issue. People’s subjective evaluations of themselves are not independent from their evaluations of how the world outside of them is doing. That things are bad, but not catastrophic may be leading people to be “counting their blessings”, as it were. These subjective self-evaluations may be using someone who they know who has been laid off (or the news that they’ve heard about tech company layoffs generally) as a reference category. They could be hearing about people who struggle to afford basic housing and see themselves as relatively better off. Their judgements about what their “ideal” lives could look like (to reference back to the original Gallup survey) are likely constrained by the realities that they’re experiencing now. In short, people may be feeling “ok” because they are doing better than many plausible referents. No one can keep up with the Jones’ right now, so people are perhaps more sanguine about their personal situations than if they were the only ones feeling left behind.
That’s my hypothesis at least, one I hope to get a chance to test out in 2024 (or at least see others test it out). A lot of what we’re seeing is people talking past each other because they’re using different points and means of reference. Some of that is assuredly deliberate: we’re entering an election year. Democratic actors want people to feel like President Biden has been a capable steward of the economy in a turbulent time and Republican actors want people to feel that he’s been sleeping at the helm while the ship scrapes against an iceberg.
But others, I feel, come from people who are struggling to understand why macro-economic indicators appear to have so violently decoupled from consumer sentiment. The impulse is to distrust sentiment because people and their subjective attitudes are squishy and can’t be derived or defended with formal analysis. But we should perhaps be willing to also remember that the macro indicators have always also been squishy. Just in a different way. They’ve always been mere proxies, whose linkages to sentiment has rested upon a series of assumptions about how the world works that have not always been rigorously questioned. These metrics are frequently on temporal scales that are historically strongly correlated with the time spans that people base their economic judgements on. But they may work less well when the reference points of these two squishy things are no longer in sync.
Nuggets
This CBS News Poll of about 2,800 respondents (including an oversample of 800 likely Republican voters) agree with former-President Trump's claim that illegal immigrants "poison the blood" of the nation. 80% of Republicans agree with the statement. This phrase echoes language used in Mien Kampf. What's crazy to me is that, even just 10-15 years ago, this phrase would've been immediately disqualifying for any major political figure. And it's not like politics back then were the halcyon days or anything! But there has definitely been a mainstreaming of dehumanizing rhetoric since 2016. This especially pernicious case not only exemplifies the degradation of critical little-d democratic norms, it furthers it.
Where do Republicans and Democrats place MLK Jr. on the ideological spectrum? Where do they put Adolf Hitler? Samuel Perry posted new research on X that shows that most Republicans tend to see King as a moderate (or even a moderate conservative) whereas most Democrats see King as being left-wing. Hitler, perhaps unsurprisingly (at least if you’ve lived on the web as long as I have), is seen as “whatever the respondents aren’t.” So right-leaning folks see him as “extremely left wing” and left-leaning folks see him as “extremely right wing.” In short, we tend to see people we like as being ideologically proximate to us and people we dislike as either being ideologically distant or emblematic of the “other.”
This NPR article discusses proposed changes to the US Census' American Community Survey (ACS) that affect how disability is measured in the US. Current guidelines have respondents answering whether they have "serious trouble" performing basic tasks and/or functions. This binary approach would be replaced by an ordinal scale which asks respondents whether they have "no difficulty", "some difficulty", "a lot of difficulty", with the tasks or "cannot do [it] at all". Only the top two categories would count for Federal measures, which could potentially decrease the number of residents categorized as "disabled" by as much as 40% according to Census testing. If the top 3 were considered, it would potentially nearly double the nation's estimated disability rates. While I tend to like broader scales versus binary options, the massive measurement differences ought to really give us pause here—especially because the validation testing did not incorporate any comparisons to benchmarks (given that "objective" benchmarks are nigh impossible to produce) nor employed any sort of respondent-level validation (which could have been performed through in-depth interviews with multiple experts categorizing the resulting qualitative data to see which quantitative measures best map to their collective expert judgement).
The online portion has been fielded to Gallup’s proprietary panel since 2020.
I also turned 26, meaning that I was no longer in my “early twenties”—which is clearly the absolute worst part of this time period.
The literature on the Cantril scale (which Gallup adapted for its measure) has been frequently shown to be “reliable” for various cohorts at specific moments in time but I haven’t been able to find any data that suggests that the structure of the scale lends itself validly to comparisons across time. Which, you know, is kind of important if we are to make judgements about groups’ subjective wellbeing across time!
This, you understand, is a super fancy, objective social scientist term.
Seriously, why is the answer to everything “PeOpLe ArE LyInG tO tHe PoLlStErs?” Y’all seem to have a pretty misanthropic view of your neighbors, let alone pollsters who do a lot to focus on survey quality.
This is something that I often yell by the way. Such is the life of a survey researcher…