Measuring the Vibecession (Neon Liberalism #72, with Matt Darling)

Samantha is joined by Matt Darling to discuss the state of the economy. Macroeconomic indicators are good, but the vibes are rancid.

People feel like America is not on the right track. The divergence between the vibes and the macro is the vibecession—and it can be measured empirically.

But what's causing it?

Economic precarity that isn't properly measured? Phones and social media and doomscrolling? The revolt of the upper middle class against Bidenomics?

Join us as we discuss what we know, what we don't, and why being right about the economy isn't as easy as it looks.

Note: Early in the discussion, I say "Biden" when I ought to have said "Harris." Apologies.

Neon Liberalism can be heard on Spotify, on Apple, on YouTube, on Amazon, and elsewhere via its RSS feed.

Transcript

**Samantha Hancox-Li** [00:10]
Welcome back to Neon Liberalism. I'm Samantha Hancox-Li. In October of 2024, The Economist ran an issue of their magazine, and the cover image is a bundle of $100 bills with rockets on the back of it shooting into the stratosphere, and the tagline is "The Envy of the World." It's about the American economy and how the American economy is totally outperforming its peers. Yeah, you can compare America and say, oh, look, America's growth isn't as high as China's growth, or America's growth isn't as high as this or that developing country, but if you look at our peer countries — countries that are rich, that are out there at the edge of the technological frontier — we seem to have been doing pretty well, right, on most of the big macroeconomic metrics, whether that's economic growth, wage growth, inflation. We're doing pretty all right compared to Britain or Japan, right? If you had to pick one of those two economies to have, by most metrics, you would say pick America in a heartbeat.

And a month — two months — after The Economist runs this magazine, Joe Biden loses the election to Donald Trump, in no small part because people are mad about the economy. And many people, right, if you told them that the American economy is the best in the world, they'd be like, "You're crazy, what are you talking about?" Right, it's all doom and gloom. Nothing's going right. The economy is broken. The economy has been broken for a while — for a year or years or a decade — and we need to fix it.

And this contrast between certain macroeconomic indicators about the economy and the vibes, for lack of a better word, is what in technical land we call the "vibecession." And so I actually think that the vibecession is one of the most important things to understand in our politics today, that what's going on here impacts so many other debates we are having about policy, about what happened in the last four years and what we should do about it in the next four years.

And so I'm really excited to have on Matt Darling — the very best trousers, senior research associate at MEF Associates, fellow at the Searchlight Institute. Don't get too mad, he's just writing a paper, he's very insistent. It's just a paper, we'll come back to that — or probably not, actually. And overall erudite economic commentator. So yeah, Matt, thanks so much for coming on the podcast.

So yeah, I mean, I guess — easy question, easy question — what's up with the vibecession? I was looking back into this, and from what I understand — you can correct me in the comments if I'm wrong — the first published article about the vibecession is 2022, it's Kyla Scanlon, something about — I think it just has "vibecession" in the article title, right? And she's picking up on something that has been happening for a while already. And at the time, people sometimes cast it as this fear, like, oh, inflation is really high, Federal Reserve is going to come down on this like a hammer, it's going to cause a recession. There were constant predictions of a recession being right around the corner in this time period, and by all the standard metrics, it just never materializes, right? It just doesn't happen.

**Matt Darling** [03:00]
Thanks so much, Samantha. I'm so happy to be on. Yeah, so when we say the vibecession, the way I think about it is — there's the economic data, which you mentioned, right? There's a whole bunch of economic data which has been fairly good, and is fairly good right now, and has been fairly good for a while, right? In terms of the things you said — the unemployment rate, the inflation, GDP growth, anything like that. Obviously, that hasn't been the case consistently. Obviously, in 2020 we had a huge spike in unemployment, in 2021 huge inflation. But since, let's say, 2023, we've been sort of like a normal to good economy by most metrics, right?

When I say metrics, I mean the hard metrics — stuff coming out of administrative data, things coming out of surveys where you ask people, are you employed, anything like that. At the same time, when you ask about what we'd call maybe soft metrics, where it's saying, hey, how do you feel about the economy — people will just say, oh, it's really, really bad. And when I say really, really bad — it's one thing to say, oh yeah, people are mad that inflation was high a few years ago, and prices are still too high. If you said, oh yeah, people were thinking it's kind of bad, that would be very reasonable. But it's almost like the level which is sort of crazy, right? It's not that it's bad — people are saying this is effectively worse than it's ever been, right?

So the main metric that people use when talking about this is the University of Michigan Consumer Sentiment Index, which is basically a series of questions — is your family better off or worse off than a year ago, do you think you'll be better off a year from now, things like that. And that is right now, I think — or maybe last month — at the lowest point ever recorded. And when you think about that, you say, well, think about the late 70s, early 80s, where you have high inflation, high unemployment, high interest rates — really everything's going badly that could be going badly. And then people kind of think it's worse now than they did then, even though most of the metrics are looking really good.

And I think one thing that's really important is this has been a long-running discussion, right? Some of the first articles highlighting this were in 2022, and it's been four years now. And one of the interesting things is it didn't change during the shift in administrations. Yes, Democrats' sentiment overall went down, Republicans went up, but in aggregate, people still think the economy is really bad. And the difference between how you would expect them to feel about it, given the good economic data, and how they are reporting how they feel about it, has actually kind of increased over time. So it's not some — it's gotten worse. Even in 2024 you'd say, "Oh yeah, people are mad about inflation," and you'd think, "Oh, that will go away eventually," and that just has not happened. It just keeps continuing to drop.

**Samantha Hancox-Li** [06:42]
Yeah, I mean, I absolutely agree with that, right. And there's the whole "no one wants to work anymore," even as people are working all over the place.

**Matt Darling** [06:52]
Yeah, she definitely coined the term, yeah. But it was fascinating too, because you don't even talk about that as much right now. Obviously, people are thinking about this with the recent tariff stuff, but everybody was poised for a recession in 2022, 2023 — never happened, economy just kept on growing, labor market kept on growing at a fairly decent pace.

One of the things that bothered me the most about this time period, which I think is really related, was this "no one wants to work anymore" narrative, or the "quiet quitting" narrative. And they were just both sort of insane in that — listen, we're currently experiencing the fastest employment growth of all time, right? Partly because things went down so badly during COVID, but we hit — I forget where we are right now, but we've sort of been around almost hitting 81% of prime-age workers, people between 25 and 54, working, which is as high as it's been in the 21st century. The only time it's higher is like 1999, where it was like 81.9%. And so you've had all these sorts of narratives about this that a lot of them just don't add up but have become sort of part of the way that people think about it. When you ask people, they say these things, and then it's like, "No, I'm working," or, "No, that's not happening."

**Samantha Hancox-Li** [09:00]
There's one narrative that it's all just inflation, but then there's this weird gap, where it's like, okay, we have this big inflation spike, right, where there is genuinely a lot of inflation that happens on a month-by-month basis over this period, and then it kind of comes back down, right, inflation comes back under control. And long after inflation has returned to totally normal-ish rates of inflation, people are still super mad about inflation, right? They're still telling surveys inflation is out of control, even after inflation is under control. And I think this leads to a lot of the puzzlement that people express about the vibecession.

**Matt Darling** [09:38]
Oh yeah, and it's interesting — one of the Michigan survey questions I mentioned earlier, which is one of the — there's a couple different sentiment surveys, that's the one I'm most familiar with — but it does ask a question, which is, if you are worse off than you were last year, why? People very consistently say, oh, because of prices. And again, that made total sense in 2022, 2023 even, but now you're like — prices have not had that big of a jump in the last year. Again, any given amount you can sort of quibble at, but on average it's around a normal amount. But it's still something that's just so salient in people's heads that that's what they always say in terms of what's getting them upset.

**Samantha Hancox-Li** [10:25]
Yeah, there's so much discourse about this, and people, I think, either saying stuff that is obviously false or talking past each other, getting very heated about it. And you have a recent essay on the subject, and I really liked what you wrote, because you kind of show that the vibecession is not just vibes, it's not just some made-up term, but it's measurable. We can measure in a pretty precise way what the vibecession is, and I think that really can help clarify the debate here. So, how do we measure the vibecession?

**Matt Darling** [11:22]
Yeah, so one way of doing it — and I'll use some statistics methodology terms here, but I'm going to try to make it fairly light — basically, like I was saying, you could take the sentiment data that we're talking about, right, and then you can feed in other economic data. So I think the first person who really did this to quantify it was a guy named — quasi ironically — Quant, right? So he went and made this really good model, put a whole bunch of stuff in there to see that.

And it's sort of funny because I was like, oh, I should run my own model. I'm just going to do three very basic things, right, which is just looking at unemployment rate, inflation year over year, and the federal funds rate — the interest rate that the Federal Reserve charges for loaning money. You could add stuff to that, but the nice thing about those three data points is that they're all sort of anti-correlated with each other, in general. Higher inflation and unemployment will tend to be correlated against each other, so you sort of have these three indexes that you can use, and then plop that into a spreadsheet and say, okay, given these numbers — given what unemployment is, given what inflation and interest rates are — what would you predict historically the data would be.

And again, this can get very complicated. There's a paper that came out looking at, oh, we changed some of the ways that inflation was calculated in 1983 and we have to sort of adjust for that. But the version I'm using is the simplest possible version of it, right?

So the Michigan survey is on a scale that is sort of indexed to 100 — 100 is where the economy was in 1966, which is a fairly good time. It almost ends up being like a grade point — 100 is a good economy. If things are going really good, like in 2019 or 1999, you get some extra credit, you go to 105 or something like that. But a good economy is about at 100, an okay one is at 90, and then things sort of fall down after that. And I think right now — I can't remember the exact number — but it's at like 50 or something, it's like this failing grade.

But what I think is really interesting is the difference between the predicted sentiment and what we would call the residual — the difference between those — and that's been sort of just increasing over time. That to me is — you can define the vibecession lots of different ways, but every once in a while you'll have someone say, "Hey, of course people are upset, tariff stuff is happening right now, and some people are upset about that." But you're like, okay, no, people can be upset about that, it's reasonable to be upset about that. What's interesting is this gap between the predicted level and what you'd expect. And again, it's noisy data, but at some level it's almost this straight line down, which has gotten worse and worse since around January 2022 or so.

**Samantha Hancox-Li** [14:19]
Yeah, so right, you can — we have all this data that we've been collecting for years and years, right, on a national level, about how people feel and how the economy is doing, these nice big data sets. And you can take your model and it'll be very predictive, right, for the 1990s, it'll be predictive for the 2000s, and then you get to our time period. And where do you think the breakpoint is? Is it just COVID, or is it after COVID?

**Matt Darling** [14:34]
Yeah, until now, yeah. I mean, COVID is weird, right? Because COVID is weird in that — I wouldn't, for very obvious reasons, I would expect the COVID residual to be kind of high, right? COVID was a weird time, and especially you have things like unemployment went up really high, but also everybody who was unemployed was getting an extra $600 a week in their UI checks. So that's not — obviously you'd expect to see a breakdown in that sort of thing because of the policy level.

But honestly, what's sort of interesting to me is you don't actually see, except for a really fast split in March and April and May of 2020 — but through the second half of 2020 and most of 2021, you're like, oh yeah, this is fairly normal. The predicted and the actual sentiment — the economy was by most respects bad in a lot of ways, right? Unemployment was elevated, and then as soon as unemployment wasn't elevated, inflation was elevated. So you're like, okay, these are normal things.

And what I think is really interesting is that when you would expect sentiment to start — when the expected sentiment sort of started to recover, right, when you started to see inflation and everything else normalized, which I think is basically you could almost date from the year 2022, right — the actual sentiment just didn't recover, or recovered much more slowly, such that the gap between the two of them kept increasing.

And there's so — it's interesting, because the COVID and the immediate post-COVID economies were just really weird in a lot of ways, right? Such that you'd expect to see some of that difference continuing. But now it's a little bit more normalized, so anyway, yeah.

**Samantha Hancox-Li** [17:22]
One analyst said — well, the reason that everybody's wrong about this is that we had all of these government relief programs during COVID, right? We had the child tax credit, we had the Super Dole, we had the CARES Act, we were shotgunning money at the economy. And the economy starts to recover exactly as all of these taper off, and so if you run the numbers, actually people's incomes are going down, so it makes sense that they would be mad about the economy. I'm curious what you think about that style of explanation.

**Matt Darling** [17:22]
Yeah, so there's a couple of people who have said this, right. So the reason why I think you're drawing a little bit of a blank there — there's a really good paper, and I'm drawing a blank on the author of it — but that just came out, actually looking at this a little bit more in depth, specifically looking at the child tax credit. So I'll say — on one level, yes, when people have a lot of money in their bank accounts, and they don't usually have money in their bank accounts, yeah, you'd expect sentiment to go up.

I think one thing that is a little bit tricky about this explanation — there's two things that are tricky. One is the timeline doesn't quite work right. So if you think about when stimulus payments went out — I think the last one went out like March 2021, right? That's the extra $2,000 after the Georgia election. So there's a couple cycles of that. And then unemployment — specifically, most states ended unemployment around May 2021, all states ended it I think September 2021. So the vibes are getting worse during 2022 — the timing doesn't quite work out. And you could tell a story where you're like, okay, it works out, but it's lagged, right? It's not that people didn't care about the money, but they care about their bank accounts, and as that sort of declines...

The other thing that I think doesn't work out — the Michigan data does give you a breakout by income, right? So you can say, okay, what's going on with low-income households, what's going on with middle-income households, what's going on with high-income households. And I think one of the fascinating things for me is that — and this is less the case now, but was very much the case in 2023, 2024 — it's actually the upper-income households that saw the steepest decline, right? Upper-income households are normally a little bit more comfortable with the economy, and lower income — their sentiment is higher. But when you look at the change in sentiment over time, it's actually the upper and middle incomes that went down the most during the bulk of the Biden years, right?

So again, it's a bit noisier in the last couple of months, but 2023, 2024, it is mostly like they saw the furthest decline. And so I think that makes the story of "oh, this is all due to the drawbacks of the welfare state" a little bit harder. One thing I'll say is, as a person who spent most of that time period being like, "we need to extend the child tax credit, we need to reform UI" — I would love to be able to say that these are the things that are causing it. And so I really get that. But it's just, I don't think it's consistent with the data when you decompose it over the population.

**Samantha Hancox-Li** [20:20]
Yeah, I think these compositional effects that you're talking about are really interesting, because one of the most common responses you will get on social media — which is always a hive of reasonable discourse — is people get, like, if you talk about the vibecession, if you even say the word vibecession, people will get mad at you. They will say there was no vibecession, there is a real economic crisis, and you are making light of my suffering, you are making light of the precarity of the working class. More Americans are living paycheck to paycheck than ever before. I mentioned this because I know that you have had an exchange with Matt Bruenig about this paycheck-to-paycheck kind of statistic. So I'm curious what you think about that — about this argument, like, okay, oh yeah, GDP is growing or whatever, but I am suffering. There is something that has gone wrong in my life, or in the lives of working-class Americans, poor Americans, and we just need to find a different metric to capture what that is.

**Matt Darling** [21:30]
Yeah, so there's so much going on with this question, and I'm going to try to tackle it — I'm going to miss things, so feel free to come back to anything.

So one thing is that, of course, lots of people are suffering, even when the economy is good by any measure, right? There have always been people who are poor, there's always people who have lost their job. One thing I always emphasize is that people lose their jobs a lot — 2% of Americans are laid off every single month, and that's been sort of consistent since we started data collection on that in 2000. And then a lot of economic data is flawed in some ways. Now, I think people sort of over-index on thinking how flawed it is. This is actually, when you go and read about how it was collected — some data, you read about the methods and you're like, oh my gosh, this is awful. But a lot of the Bureau of Labor Statistics data, you read the methods, you're like, oh wow, they've thought of everything, right? There are 200 econ PhDs that are spending 40 hours a week thinking about how to improve data methods. It is very unlikely that a random person on Twitter or Bluesky is going to come up with something that they did not think of.

The paycheck-to-paycheck discourse is, as you mentioned, something that bugs me a lot, because when people talk about that, they're almost always referring to this survey that was run by basically a payday lending firm, which is saying, oh, Americans need — they're living paycheck to paycheck, clearly we need to stop regulating payday lending. And it's sort of funny to see a lot of leftist-type voices accidentally pick that up and start running with it.

Matt Bruenig wrote a great thing about this, which I want to be very clear about — I agree with 99% of the blog post that he wrote. I keep meaning to write a response to it since I've started actually writing Substacks. But I basically agree with everything he said, right? 90% of what he's saying there is, hey, paycheck-to-paycheck is just kind of a silly metric, right? It's not that meaningful. And he has other metrics that are good at looking at things like financial precarity, and I'm like, oh yeah, I would love it if we started looking at that.

One thing I always try to emphasize is, almost every time people start talking about paycheck-to-paycheck, I'm like, you should just use the poverty rate. The poverty rate did increase, it's a meaningful concept, it's very regular. Use that instead of these random one-shot surveys that don't tell you what their questions were, or how many people they talked to, or anything like that.

But there is an issue, and America has, compared to other OECD countries, just a not-great welfare state in a lot of different ways — more piecemeal, a lot more difficult to access, and everything like that. These are real issues that I do think we should be dedicating a lot of our political will to solving. But I think it's hard to make the case that they're responsible for the vibecession insofar as they existed in 2017 and 2014 and 2009. And when we're talking about the vibecession, we're very much talking about this phenomenon that is four or five years old. You can maybe tell the story where it's like, oh, people had an awakening that just happened to happen post-COVID, and I find that a little bit implausible, because you don't see these signs of this awakening anywhere else, really, right? It's not like people's politics have suddenly shifted. A lot of things are normal, except how we're sort of talking about the economy.

**Samantha Hancox-Li** [25:30]
So I agree with that as well. This is kind of why I like the quantitative measure of vibecession that you've offered here, because yeah, I mean, there are things that I think aren't great with the American economy — whether that's how we do housing in this country, or how we structure healthcare, or unemployment insurance — but none of that really changed that much over the course of the period when this enormous gap between predicted and actual consumer sentiment emerges, right? There are long-running problems that shouldn't be responsible for an acute crisis.

The income breakdown that you've mentioned is very striking — that the vibes are worse as you get richer, which is again not at all consistent with the idea that this is primarily about precarity, because somehow the most precarious people feel the least bad about it. I think these are why it's good to start with some actual quantitative data here.

And I guess this brings me to one of the others — the big one — this brings me back, as always, to Will Stancil, who cannot be escaped.

**Matt Darling** [26:30]
You can't escape the gravity well of Will Stancil.

**Samantha Hancox-Li** [26:35]
That's right, the gravity well of Will Stancil is enormous. But Will Stancil — if you don't know who that is, he's done a lot of things, he's been kind of involved in anti-ICE resistance in Minneapolis — but he's also known for this thesis that actually a lot of our discourse, and in particular consumer sentiment, is detached from material reality. That discourse has its own kind of self-standing logic to it that is driving how people feel about the world, and it's only at best lightly connected with material conditions.

And so people will say — to put the most plausible case on it — yeah, the media spent a year telling Americans that recession was around the corner, and Americans believed it, right? Americans got on their own phones and talked about how their lives were terrible, and people read these posts going viral and were like, "You're right, my life is terrible too. I'm going to talk about how terrible my life is." And this creates this kind of self-sustaining doom loop of social media. And that is responsible for the vibecession. Why are the vibes off? It's because the vibes are rancid, right? The vibes have their own internal logic to them, so to speak.

**Matt Darling** [28:00]
Yeah, and so there's clearly something to that, right? So I think one of the interesting things to me is the amount of things that people believe that isn't really true. And a lot of times you can talk about it in this sort of way of, there's a story that goes viral, and the reason it goes viral is it has this incredibly negative framing, and then I'm like, wait, but also the information in it is inaccurate, right?

So I'm a labor market guy, so one thing that I find fascinating is every time a company has layoffs, right? So I think Facebook slash Meta just announced that they're laying off some workers, and it goes viral, and people are like, oh my gosh, layoffs are so high. But 2% of Americans are laid off about every month, which is an enormous flow, a huge flow. It means that yes, you probably know someone who was laid off recently if you know a representative sample of the population. But it also means that any month you can go and be like, "Aha, we're going to put these things on the front page of the newspaper," or not. You can just decide that.

And so it's interesting, because people do have these misperceptions about some of the raw data, almost. And I do think there's something to that. But it is still a little bit under-theorized, right — which I want to be clear, I feel like every explanation of the vibecession is a little bit under-theorized, to some extent. Almost everything has to be true, because the gap is so large, right?

But in the same way that you, if you say, "Oh, it must be the phones," you're like, "Okay, but why? Why 2022?" I think Will himself initially was like, "This is because right-wing media has started controlling the narrative very effectively," and that made absolute sense until January 2025, where the right wing became empowered, and the vibes continued their steady decline.

I think one explanation that does make sense is this sort of change in how the media works — both happening slowly over time, but I think also taking off during COVID — this sort of lack of centralized authority, right? You no longer have Tom Brokaw going and saying, "Hey, this is what's happening today." You now have a million people on TikTok doing it. There's a well-known bias — negative stuff goes more viral than positive stuff — and so that sort of continues to happen.

And you can see that in some of the data. There's a good report, I think by Brookings — Ben Harris and Aaron Sojourner — where they're like, yeah, the media has been in general reporting more negative news compared to sentiment over time. And that's just the regular media looking at headlines. I'm sure if you went and looked at TikTok and Bluesky and Twitter, there'd be even more of that.

So I do think there's something to that. Again, the fun thing about the vibecession is that even though we have tons of data on it — but tons of data means 50 years of data, right? They report it every month, and we've got it going back to the 1960s or so, which is both a lot of data but also not much where we can go and say, aha, here's what actually makes the difference. It's one country over 60 years where you can say, aha, some things go up, some things go down. So it can fit a lot of different theories, and it's basically impossible to falsify anything exactly.

**Samantha Hancox-Li** [33:00]
If you've done any work with this at all, you know that more data just makes for easier p-hacking.

**Matt Darling** [33:06]
Yeah, so this is — I think the original article that went viral was by Derek Thompson, he wrote something for The Atlantic about this. And again, 2022 when a lot of these things started, but it has been continuous since then.

So I think he's looking at the SHED poll, which is run by the Federal Reserve, and they ask that exact question — how are your personal finances, on a four-point scale, bad, poor, good, very good, something like that. And they also ask you to evaluate both your local and the national economy.

And like you said, when you ask people about their personal finances, that's fairly consistent over time, right? I can't remember the exact numbers, but most people — I think 60, 70, 80% — are like, yeah, my personal finances are fine, right? At some level, they're saying one or two on the scale, one and two being the better ones. And then when you ask them, okay, how's the economy doing, they're like, oh no, it's quite bad.

I think one of the things that I think is really fascinating about this — and this goes to some of the social media stuff we were talking about earlier — is that sometimes people are sort of like, okay, things are good for me, but I also know other people out there are suffering, and so the economy is bad. But again, if there is this sort of misperception happening, where people are saying, well, I'm going to speak for the people who are doing much worse...

One thing that I think is really interesting, and goes into some of these things we're talking about, is one thing that has happened really over the last 10 years — not even just the pandemic itself — is we've actually seen a fairly rapid reduction in income inequality, right? Both before the pandemic and during and after the pandemic, we actually saw this really large increase in wages for the people who've had low wages at baseline. Those have increased quite rapidly, mostly as a product of having high employment rates — the higher employment is, the lower unemployment is, the more firms have to bid up wages and everything like that. And that affects low-wage people the most.

And there's this sort of thing where — I'm like 80th-percentile income, not absurdly high but not absurdly low or anything like that — but everyone's like, oh man, things are bad. And there are certain things in the last few years where I'm like, oh man, here's this bad thing that happened, maybe the economy is bad. But then my actual analysis kicks in — like, wait a second, that's not what's happening.

So I'll give an example. Summer 2019 — I have a daughter, she's 10 now, she was four or five back then — I remember basically being like, oh yeah, every day the ice cream truck comes by at 5 o'clock and we go outside, we spend $2 and get an ice cream. And then the ice cream truck stopped coming in 2022, and it's like — "because no one wants to work anymore." But actually what was happening is you could get better pay somewhere else besides running an ice cream truck. So basically, you saw — and similarly, I like to go to work at coffee shops every once in a while, and I'm like, oh, all the coffee shops used to be open till six, and now they close at three.

And there's this sort of thing that feels to me like not necessarily economic stress, but I'm like, oh, that's annoying, that's a bit frustrating. There's all these things that used to make my life a little bit easier, a little bit more pleasant, and I'm like, oh well, if it's a little bit less pleasant for me, there must be much worse for people who are making half as much money. But then you're like — wait, no, the people making half as much money are the people who have gotten the better jobs and are now not providing services to me.

And this can get flattened a lot of ways. I think the way people talk about it on social media is like DoorDash discourse — the burrito taxi. And it is this thing where sometimes people who are like — there's one way of looking at this which is like, oh my gosh, people don't care about anything except the burrito taxi, and I think that's actually a little bit too cynical, right? It's one — you say, hey, I'm a white-collar, upper-middle-class person, and things are a little bit more stressful for me than normal. And they're like, no, Matt, you're in the social class that actually has done comparatively worse over the last decade, because the bottom half has actually done better. And then people sort of say, well, things are a little bit hard for me, it must be twice as hard for people who are lower income than me. And it's actually not quite right. It's obviously always worse to be lower income than higher income, but that sort of relative change I think is really important. People miss that.

**Samantha Hancox-Li** [41:22]
Yeah, the burrito taxi — for those in the audience who aren't familiar with the burrito taxi — it's this style of, it's sad to say this, disdainful description of a style of post where people will say, look, the economy is terrible, my Chipotle order costs $50 for a burrito. And then you look at their Chipotle order, and it's like, okay, there's a burrito, there's guacamole, there's a bag of chips, there's a bottle of water — I don't know why the hell you're ordering delivery bottles of water, but you say you are. And what you're asking is like, okay, I'm going to pay somebody to make me a burrito, and then pay somebody else to take that burrito and drive it to my house. Right, this is a level of service that is, in some ways, distinctly luxurious, right? And you're complaining, like, why is the price of that going up? Well, among other things, because the person making the burrito and the person driving the burrito taxi have better options, and they want more money for it.

So that's kind of the burrito taxi situation to me. The way I actually like to describe this is the expat experience, right? If you're an American, you objectively come from one of the wealthiest countries in the world, and if you've traveled abroad — if you've traveled to a country that is doing all right, but it's maybe a middle-income country, a country that is not one of the richest countries in the world — you probably really liked it. I mean, number one, you picked this place to go on vacation because it has a nice beach, or some cool old city, or beautiful scenery. So there's that aspect, you're on vacation. But there's also the aspect where you'll be able to get so many handcrafted goods at such great prices, right? The food will be great, there will be someone who is an exquisite chef who will make you some delightful local food for what you consider to be not very much money. And fundamentally this is about the mismatch between the demand for your labor and the demand for their labor, right? That you come from a very high-wage country, and it feels great to be able to afford not just mass industrially produced goods, but services. Or you might say — servility.

**Matt Darling** [43:11]
Yeah, and even like — every once in a while I see someone go viral, and they're like, "Oh, I moved to Croatia, and the cost of living is so much more reasonable here. I was able to — I have a personal chef." And you're like, what's the personal chef's income? What do they think about the economy? The reason you can hire a personal chef there is because the chef is paid much less than they would be in the United States.

**Samantha Hancox-Li** [44:00]
Yeah, there's the personal chef aspect. The example I go to, possibly for family reasons, is maids. Specifically, if you're familiar with maid culture in certain parts of Asia that are doing quite well, they will hire maids from other parts of Asia that are not doing as well, right? You can hire them for cheap, and then you have a live-in servant who, through various legally enforced means, you can actually control an enormous amount of their life, and you can abuse them pretty badly. You can treat them like dirt. And people — just trust me on this one — people like that. People like having a servant that they can treat like dirt. It makes them feel higher in the world, right?

So there's this aspect where the hot economy — it's got more economic growth, it's got faster technological development, we get more goods out of it, but we get less services out of it. Or less servility. So that's a reading of the vibecession — just, yeah, we ran a hot economy, and it drove up the price of services.

**Matt Darling** [45:05]
Yeah, and then you can think about things like childcare, right? People are like, oh my god, childcare is so expensive. And childcare is really important — it's a fundamental thing that people need to pay for if they want to have a kid and work. But also, the price of childcare is mostly the price of labor, right? It's basically how much money are you willing to pay some 19-year-old to watch your kids. The burrito taxi is a little bit silly, but childcare is very serious — but also the price of childcare is driven by the same dynamic.

And then I think your point about the expat experience is really good. One thing that adds to that too is — if you remember, in the beginning of COVID, people were like, oh, we need to order out food because restaurants are closed. And so I remember at one point I was like, okay, we'll order out food once a week, partly just to kind of help out our local restaurants and keep them afloat. And I remember at one point I'm like — I really shouldn't be doing this anymore, because it doesn't actually make sense anymore. There's this really nice sit-down pizza place that I was like, oh, I'll order from them every couple weeks or so. And then I'm like, wait, I should not have this as a habit now that restaurants are open again. Yes, this is a nice pizza place, I like their pizzas, but I should only really go there for a nice sit-down dinner. If I'm just getting pizza, I should just order Domino's or something like that.

One thing that I think is interesting about the 2020, 2021 economy is it was such a weird economy in so many different ways. Like, yes, unemployment was high, but then you're like, oh, but also I can order out from restaurants and stuff like that. And I think one thing that's interesting is — our baselines maybe changed, because it's like, oh, I want the ease of finding a job of 2021, the excess money of 2020, and the inflation of 2019. And that's a great economy, but you can't have all three of those things at the same time, probably.

**Samantha Hancox-Li** [47:56]
Yeah, and it made people with money and leisure mad about it. I think we're starting to maybe see why I said at the beginning that vibecession discourse is actually a really important question for politics, because it's like — well, does it tell us that actually the American economy is broken, and the way to make people happy and win back voters is to fix the economy? Or does it tell us something a little bit darker about the American electorate and what they really care about?

**Matt Darling** [47:57]
Let me throw in another wrinkle on that, because the initial findings about the wages going up faster than normal — those came out before the 2022 midterms, right? In 2022 I remember being like, oh yeah, there was an expectation of a red wave or whatever, which is fairly standard, and it was a lot weaker than people were anticipating. The Democrats did slightly better than people thought they were going to do, everyone was really worried, and then things sort of worked out. And I remember thinking, oh, maybe that's because of this income change. But then — so it's sort of funny, because you have to include both of those elections when you think about it, and it's tricky, right?

**Samantha Hancox-Li** [49:00]
It is desperately tricky. There is one last theory that I do want to talk about before moving on to what does this all mean, which is the theory that G. Elliott Morris has argued for — that I think a lot of cold-blooded economists will tell you — which is that actually it's just about prices, right? What is the vibecession? It's just people being mad about the price level.

And Elliott Morris has this kind of — there are technical aspects about the model that maybe you'll talk about, or maybe not — but basically, I'm going to try and summarize what he's saying. People have this estimate of what prices should be in the future, and they make this estimate based on how prices grew in the period between five and fifteen years ago. So this trailing decade is how they form their baseline estimation of what prices should look like in the future. And as actual prices start to deviate from expected prices, people are mad about it.

And he adds in another modifier. If this was the whole model, you would just point to stagflation and be like, this can't be right, because basically the model predicts that any one-time shift in the price level is going to make people super mad for the next 15 years, and that's not really what we observe after the end of stagflation, right? After inflation comes down from its huge peaks in the 70s, people are like, yeah, the economy is working again, I'm happy about that. So he adds this other modifier that you have to divide the gap between prices and the expected prices by the inflation rate of the last five years or something like that. So the model gets a little technical.

**Matt Darling** [50:42]
And this was the point we were raising earlier, about everybody — there's not that much data, you can't falsify — the data can almost support all sorts of things. So let me first off say, I thought that was a really good analysis, right. I went back and forth with him on Bluesky a couple times, and it was great because he went and wrote a second blog post that was mostly responding to some of the critiques I had, and I was mostly satisfied by that.

But so on one level, I think there's definitely something to this, right, in that — especially because this is literally what people say, right? When you ask people what they're mad about, they say prices, right? And I think the price level is still much higher than it would be.

And I'll say, in addition to Elliott Morris, Jared Bernstein wrote something on similar lines — Jared Bernstein being Biden's former chief economic advisor. So there's something to this, absolutely right, and the data works and everything like that. I do think it definitely explains part of it.

I am a little bit skeptical that it explains everything, right? Everyone's like, aha, I found this one thing that explains it. And part of it is — you can make that data work. And again, Elliott Morris is a good statistician, he's doing reasonable things here. But it's also just like — we've had four years of the vibecession at this point, and so I'm a little bit skeptical of whenever anyone goes and says, okay, well, we add these three different ways of organizing the data, now everything works perfectly. Because the more variables you add, the tighter the fit will be.

So there's one level where I'm like, absolutely, this is definitely part of it. Again, people absolutely say — and you even see this, I think we all sort of experienced something where we're going down the grocery store aisle and we're like, wait, that costs that much now? There are things that I'm just still not used to five years later. There are things like going to the movie theater, getting movie theater popcorn, where when I was in my teens, I was going once a month or something like that, and so I'm like, yeah, this is how much movie theater stuff costs. And now I go like once a quarter, and I'm like, it's still expensive, weird.

So there's something to that. At the same time, because to make the explanation work you have to have this fairly complicated model, a lot of different moving parts, and so — how is this going to predict data going forward?

I'll give an example. There's a great Substack called Briefing Book — it's sort of ex-Council of Economic Advisors folks — and they had a piece, I think summer 2024, which was basically the same theory, right? They're saying, okay, look at the deviance, fitting the data, and okay, people will no longer be mad about prices around October 2024. And that was a very reasonable prediction, all the data fit. And then history kept happening, and the data just did not fit going forward. So again, I think it's a good attempt, it's a good look at it. People who are interested in this should definitely read it and think about it. At the same time, I'm like, okay, will we see this fit six months from now?

And again, it's absolutely the case that one of the things people are upset about is the price level. I think the question is — if you look at Elliott Morris's model, that's something that says it explains all of it, and I'm sure it explains 30% of it, or 50% of it. I do think there's a lot of other things going on.

And so, when you have — I would not be surprised, maybe my default answer is this: it's like 40% the Will Stancil theory of it's something about social media and how social media is being reported and how it's changing how we perceive the world. It's 40% the Bernstein-Morris theory, where it's people are just upset about the price level. And maybe 10% the Matt Darling theory about a lot of it being about people upset about low-income workers making more. And then 10% — 30 other random theories that could be out there.

**Samantha Hancox-Li** [53:01]
My Coke Zero is so expensive, yeah. 10% is a lot.

Wouldn't one piece of more data that I thought I would bring up here — and maybe this supports the media thesis, or maybe not, it's kind of an interesting question — it's something that you've highlighted, which is that there's this gap, not just between predicted consumer sentiment and actual consumer sentiment, but there's a gap between how people evaluate their own financial situation and how they evaluate the economy. And from what I recall, people's evaluation of their own financial situation is pretty good and pretty consistent. Which is striking, right? That people are like, yeah, I'm doing pretty well, and I have been for a while, but man — the economy.

So I'm curious what you make of that piece of data here.

**Samantha Hancox-Li** [54:30]
Yeah, because no one wants to work anymore.

**Matt Darling** [54:35]
Yeah, it's real, yeah. The nice thing about this is, what we need is the data in the future, right? To some extent — so again, the Elliott Morris theory, you could go and say, hey, what is this predicting going forward, right? You could say, let's take his model and let's see what happens in 2027, 2028. And I'm sure it'll fit data somewhat. Also, there'll be some source of deviation from it. I think that's what we need for that.

I mean, there's one level at which you could say, okay, the sort of exercise of decomposing it — where you'd say, is it — I gave some random numbers earlier — but you could be like, oh, maybe it's actually 20% the Stancil thesis and 80% the Morris thesis, or 80% Stancil, 20% — we're never going to get that level of it. But I do think that we have a couple of different plausible models that all sort of make logical sense, they all sort of correspond with some dimensions of reality. And we also have a whole bunch of models where I'm like, no, that doesn't make any sense. But the ones we've been mostly discussing — I'm like, okay, these do add up, they do make sense, they do make the frameworks work out.

But the one that is the hardest to measure in this sort of way is the Stancil thesis, right — of, okay, it's something about TikTok and the way social media works. And partly because there's one thing which is like, oh yeah, but everyone has phones now. But there's also sort of a shift in media consumption specifically around 2020, partly because everybody just became more online, or not everybody, but — in the same way that I've been a remote worker for my entire career since like 2005 or so, and it's sort of funny because at one point during COVID I'm like, oh my gosh, no one knows how to speak into microphones, right? You'd have a conference call and I'd be the only one joining remotely, and I couldn't tell what was happening. Now everyone knows how to speak into a microphone — it's a very simple thing that everyone can do. Everyone did sort of start being a little bit more — not everyone, I'm overspeaking — but there is more onlineness post-COVID than there was pre-COVID. And so you can say, hey, these two shocks happen at the same time, maybe they explain each other. Or maybe they just happen to be two shocks at the same time.

And I'll say that was also my initial criticism of Morris — saying, hey, you've got this great theory about the price level thing, but also in your data set there's only one price level shock, and it happens to be the vibecession. So — things about the vibecession, it's like 2020-21 was weird. So much stuff happened that — all of which were strange — and there were theories that made sense.

I'll say, one thing that was very much a concern in 2021 even was like, oh, maybe there's all these quality differences that changed because people had to change things. You go to a hotel and they no longer clean your room every day, they only — you have to put a little sign on the door asking for it. And there were things like that where, yeah, the BLS is not going to pick up those sorts of changes. But that's not the case going forward.

So again, what we always need is just more data in the future. But we also just have to deal with the fact that there are some things in the world that we are never going to be able to solve. One of my — there's an economics textbook called *Mostly Harmless Econometrics*, and its opening chapter is like, there are some questions that are called fundamentally unidentifiable questions, which means — and if you have one of those, you're fucked — F-U-Q-E-D. And this is one of those, right? We are never going to be able to say, aha, here's this one easy-to-identify shock that happened at the same time as COVID. A whole bunch of stuff happened during COVID. So we'll be able to falsify a whole bunch of theories, we'll never be able to prove that one of them was the correct one.

**Samantha Hancox-Li** [1:01:26]
So what I really want to ask you is — what data do we need to resolve the differences between these hypotheses? Right, what could we collect that we aren't collecting, or what could we know that we don't know yet that would actually tell us what's true here?

Oh, you know, I forgot — I've got my book of future past on the bookshelf over there. I should have just looked in there. What was I thinking.

Yeah, no, that's a huge problem.

So that's really rough, because — as you talked a bit earlier about how this was an interesting intellectual problem, and it is an interesting intellectual problem, but as I've been trying to emphasize, there's a problem with very real stakes for the future of democratic policy, right?

**Matt Darling** [1:03:04]
Yeah, but did it maybe happen in 2019?

**Samantha Hancox-Li** [1:03:35]
Like to lay out how I kind of see this — we have to understand Biden, right? Biden comes into office and he has a theory of how he's going to build an enduring popular majority. He is surrounded by a group of economists who look at 2008 and they say, we made a giant missed call as a profession — that we lowballed the stimulus in the aftermath of the Great Financial Crisis, and we had the Great Recession. It lasted for a long time, really did a huge hit to the American economy. We could have fixed it if we had done a bigger stimulus.

And Biden kind of comes in — so there's that aspect kind of in the air. At the same time, there's modern monetary theory. Now, high-church modern monetary theory says all kinds of stuff, but in the popular consciousness, it was saying — we've been hearing all our lives about how austerity is necessary, and we've got to cut back and tighten our belts, and blah blah blah, but that's all bullshit. Actually, we can just spend money and make people's lives better. And that's basically the Biden theory of the case. We're going to have big fiscal outlays from the federal government. It's going to create a hot economy — and then we did very much have a hot economy. It's going to drive wage growth — it did drive wage growth. And this is deliverism, right? We are going to deliver for the American people, and they are going to reward us with votes. And that part didn't happen, right? And in part because of the vibecession. Yeah, as you've noted, maybe it happened once and then didn't happen later. There's some confusion here.

And so I think that is really kind of roiling in the background of Democratic wonk circles, Democratic policy circles at the moment, because at some point — I strongly suspect in 2028 — we're going to retake the White House. We are going to be able to pass legislation, and people are going to ask, should we do it again? Should we go for big fiscal outlays and a hot economy, or is maintaining the price level the overwhelmingly most important thing we can do? We just cannot ever let inflation even blip — like, this is the thing. If you believe in Elliott Morris's theory, even a blip in inflation is going to be disastrous for you, because people are going to be mad about it for a decade, right?

**Matt Darling** [1:04:06]
Yeah, so these are great questions. One thing I want to — because I feel like that's the dominant frame I see about this sort of question, the one that you laid out — but I always feel like it gets a little bit wrong, because I wonder to what extent voters associate Trump more with economic stimulus than they associate Biden, right?

So again, the CARES Act is $2.1 trillion, American Rescue Plan is like $1.9 trillion, right? And there's one way of thinking — who is president in 2020 is also a debatable concept, right? But I do think that people — I feel like there is this real concern about, like, oh man, maybe we shouldn't do Keynesian stimulus anymore, that I think sometimes neglects that both candidates did a lot of Keynesian stimulus, right?

And one way of thinking about it is — you have this big hole in the economy that needs to be filled, what economists would call an output gap. And I want to make up some numbers here, but you could say, Trump filled 60% of the hole, and then Biden comes in, he's like, oh, there's a whole bunch of the hole left, and he fills up the rest of the hole, and then it overspills a little bit — that's the inflation, maybe. You pour 50% of the bucket's capacity into it, and so it has overspilled. And yes, voters are sort of mad about that. But also I'm like, yeah, but it's not something where voters rejected the stimulus candidate against the non-stimulus candidate. People definitely sort of thought of — again, trying to explain exactly how the median voter thinks about this is tricky. But I do think that both candidates were associated with getting these big checks and everything like that in a way that I think is hard to disentangle. And so I get worried when people say, oh, the takeaway of this is to not do that.

The other thing that I think is really important is just trying to figure out how to precisely get legislation done that is at the right level. And I think one thing I find fascinating about this is if you go and read Paul Krugman — pieces from late 2008 — he's like, okay, we're going to pass the big stimulus bill. So he has something in December where he's like, I did some math, and the stimulus bill has to be about $300 billion. And then two months later, he's like, I did some more math, and it actually turns out it needs to be like $1.2 trillion. And that was just two months of new data coming in about how big the shock was. And what I think is interesting — Congress passed about $700-800 billion around that level, and you're like, oh yeah, part of this is just the narrative of how big the budget should be. Congress can't move that fast, right, in terms of absorbing that data.

And then you almost had the exact opposite happen in 2020, where basically after the election, people think, okay, we're going to do a big stimulus thing, we need to do this. And one of the things we need to do is bail out towns, because they're like, oh, towns are going to be losing a bunch of money — towns and cities and states — so we need this big town bailout. And then we got more data on that, it's like, oh, actually, no, it turns out that towns are doing great, they all have extra money. And then we gave them all these grants and money, and the towns are just like, okay, we don't even know what to do with this.

So I think the two points I want to raise on that are: one, making sure we don't say, hey, we have to reject Keynesian stimulus, because both candidates did that — that's not a story the voters were telling us. Two, I think it's about getting it right, right? It's bad to undershoot and it's bad to overshoot. And in both cases I think we undershot 2008 by a lot, and we overshot 2021 by a lot. And I think that's hard, right? We're talking about — can you get a change in the economic, Nobel-Prize-winner consensus that changes in 60 days, can you get that to influence legislation? Maybe not, right? That's just not the timelines that Congress sort of works under.

**Samantha Hancox-Li** [1:08:30]
And those were passed by Barack Obama, right? I'm joking — I'm just saying that the median voter's memory for who passed what is a little...

**Matt Darling** [1:08:35]
And then I think the third point I'll raise — the fact that low-income workers have had their wages increase — I think that's vitally important. One thing, when I bring it up, people are always saying, well, the minimum wage didn't increase. I'm like, well, the minimum wage isn't that important, right? It's important that people are earning and stuff like that, but when people say, "Oh, the vibecession is caused by the minimum wage" — like 1% of workers earn minimum wage. But I do wonder to what extent it's important from the narrative point of view, right?

So again, go to that cynical DoorDash story, right, where it's like, oh, people only care about the burrito taxi. But I think people were saying, oh, well, this went up, but it's all going to corporations. If you did pass a minimum wage at the same time, I think people might have been a bit more willing to accept it. It's partly people being frustrated, but then partly they have a narrative that encourages that cynicism. And so I do wonder if you went back in time and passed a minimum wage — would you, even if the distribution of wages didn't change at all, because the minimum wage went up to $12 even while most workers are being paid more — would that have shifted the narrative enough? I think that's an interesting question.

**Samantha Hancox-Li** [1:09:42]
Yeah, that brings us back to Stancilism, right? That it's a question about messaging as much as it's a question about policy, which takes us in a completely different direction, right? That doesn't say the problem is big fiscal versus little fiscal — the problem is the idea of deliverism, or the execution of it. That you can't just give people stuff in some objective material way — you got to give it to them in a way that captures the narrative.

This is — I don't know. I, like you, don't really have an answer here. Sorry, listeners, I haven't figured this one out yet. It's a tough one.

**Matt Darling** [1:10:15]
Yeah, and I think that is the takeaway that people should have. Like I said, there's just some questions that are really hard to answer. I think the theories that people have put out there — they're all plausible. We're talking about a big impact — a huge impact. They're all sort of right at some margins. There's plenty of room for everyone's pet theories in here. But it's not something where we'll be able to say, aha, we solved it, this is the answer. And so we just have to deal with that — it's always going to be the case.

**Samantha Hancox-Li** [1:11:15]
You know, I was really hoping that would not be the answer. I was hoping you would fix this for us. But yeah, I think this is at least a good note to end on — a certain level of humility and recognition of uncertainty. I think we live in an age where we have so much data, we have so much analysis, we have so much science, we want there to be an answer, and we want there to be a safe answer, right? If you do this, there won't be any risk.

And I think you read a little bit of history, you read about the choices that politicians and leaders face — there's never a safe and obvious answer. They are often confronted with pretty severe uncertainty, and you have to take risks, because risk is unavoidable. There is no choice that isn't going to — as we've talked about here, right? You can make a bet on big fiscal, or you can make a bet on price stability, you can make a bet on narrative, but none of these are actually going to guarantee you, in any kind of sense, that the voters aren't going to be unhappy with you in four years.

There we go. That's a great note to end on, Matt. Thanks so much for coming on the podcast. Yeah, this was very interesting, even if a little bit depressing. So yeah, see everybody next week.

**Matt Darling** [1:11:56]
Thank you so much for having me. I appreciate it.

**Samantha Hancox-Li** [1:12:15]
Bye.