Why Americans Feel Worse Than the Economy Looks

[00:00:00] Speaker 1: I was about to say, they ask each other if there's work at NPR or the local kale factory. The local kale factory. Does kale not come from factories? I owe all Brooklynites an apology. Next time you're at the co-op meeting, could you tell the board and the commissar that I apologised? [00:00:19] Speaker 2: All right. Inflation looks decent. Unemployment looks decent. But Americans are very gloomy about the economy. Consumer sentiment this summer sat at 52, which is about where it was in May 1980. At that time, unemployment was nearly twice as high as it is now, and inflation was running at about 14%. Justin, you just made a piece where you estimated that consumer sentiment should be around 82. You explained that 30-point gap by saying that it has to do with how consumers rank their confidence in the government's economic management. Today, I'm going to bring you some follow-up questions about that from our audience. I'm Augusta, and I'm a producer here at Platypus Economics. [00:01:05] Speaker 1: And I'm Justin Wolfers. I'm an economics professor. Think of this as office hours where Augusta brings me your questions. Office hours are open. Augusta, I'm going to tell you, I am so excited about this piece. It's been one of the big puzzles, and it's not often I think I crack the case, but I think we have a pretty plausible answer here, mate. [00:01:28] Speaker 2: Well, good. I'm going to ask you a bunch of follow-up questions, try to poke holes in it. [00:01:32] Speaker 1: That's right. And I'm going to take it the way an economist is meant to, which is criticism and poking holes in an argument is how you show respect and love for an argument. So thank you in advance to all of our audience who want to poke holes. [00:01:46] Speaker 2: Okay. So the first question is, which way does the causality run here, right? People are gloomy about the government, or do we think maybe people just are gloomy and then they blame whoever is in charge for that? [00:01:58] Speaker 1: Okay. So a really good question. The following is true. People have a great deal of fear about the quality of our economic leadership. They respond to many surveys, not just the one that I answered. I think I might do more work on this soon, actually. So stay tuned. Saying that they really do not trust that economic policy is being well run at the moment. I haven't, I think that also makes sense because I've seen how economic policy is being run at the moment. So we know that that's high and we know that at the current moment, consumer surplus is, consumer surplus, my goodness, that's what happens when you teach economics 101, consumer sentiment. Surplus is a triangle, sentiment is an index. Okay, there. Oh, no, there. Anyway, no, we're not going to, don't distract me with finger graphs. Sorry. You actually did the whole student thing. I'm sorry. [00:03:01] Speaker 2: I pressed the raise hand button. [00:03:03] Speaker 1: Yes. Um, okay. We know that people don't have faith in economic leadership right now. We also know that their consumer sentiment is quite low and it's far lower than you would expect given economic conditions. A very natural interpretation. I'm not trying to pretend I'm doing causal inference here, but a natural interpretation is when you feel bad about leadership, you feel bad about the future. [00:03:25] Speaker 2: Right. [00:03:26] Speaker 1: Just makes sense. You could have a view, something happened and everyone got grumpy and they got grumpy about everything. If that's the case, first of all, I'd say what? [00:03:35] Speaker 2: Right. [00:03:37] Speaker 1: To which, by the way, I got grumpy, but I got grumpy because I felt grumpy that we have bad economic leadership. So, which case I'd take that for my side, but maybe it is that. But if that were the case, what you would see is pervasive grumpiness across all the different survey measures. Or another way of saying this is the claim that I make is that consumer sentiment is falling short of economic fundamentals by exactly the level you would expect given people's distrust of our economic leadership. If this were just people were feeling generically grumpy across the board, then you should see people's distrust of our economic leadership is just as strongly correlated with other measures of economic sentiment. [00:04:21] Speaker 2: That makes sense. [00:04:23] Speaker 1: And there are questions like, is now a good time to buy a house? And it turns out that's not as closely correlated. So basically questions in which you're asking people to look into their economic future, they're quite closely correlated with how do you feel about our economic leadership when you're looking at a specific market or the current moment in time, much less strongly correlated. So that's the sort of pattern that you would expect if my story were right, but not if you thought this was pervasive grumpiness. [00:04:53] Speaker 2: Does sentiment even matter here then? Like if people say they're miserable, but they keep spending, is this maybe just not a useful measure of how people are feeling? Is this a meaningful survey? [00:05:05] Speaker 1: I'm going to give you the full economist on the one hand, on the other hand, because I'm a teach the controversy guy, not a here's what I think. So one view is who cares about consumer sentiment anyway. These are just words, people saying words, who cares? Sticks and stones may break my economy, but may break my bones, but words will never hurt the economy. Many economists are deeply suspicious of asking people things. They like to watch what they do instead. We call that revealed preference. So that is one view. Now, let me come back and I want to give two defenses for why this is still interesting. So first, there is some evidence. It's not wildly compelling, but it's probably on balance convincing that consumer sentiment is useful for helping to forecast the future path of the economy. Which I think is another way of saying, at the very least, it contains information content and more likely people's feelings about the future actually affect the choices that they make. I wanted to say the evidence on that's not strong, not because it's necessarily false, but because we don't have a lot of data sets where we ask people what they think and observe what they do. We have surveys that ask people what we think. We have different surveys that ask people look at what they do. So if consumer sentiment is actually a driver of economic activity, and remember Keynes talked about animal spirits as being one of the most important drivers of economic activity. Animal spirits, I think if Keynes were alive today and attended the University of Michigan, he would say, by which I mean the index of consumer sentiment. So there's certainly a tradition that says this is deeply important. There's a tradition of sort of people doing public opinion research and economic research, which actually came out of the University of Michigan economic psychology that says that this is important. So the second answer is, yeah, maybe it really does matter that your lack of trust in the leadership is undermining your willingness to spend, to invest, and so on. Third is let's forget about the measurement altogether and just think. If you think that economic policy over the next five years, next two and a half years is going to be a mess, do you want to build a factory now or do you want to wait and see what that mess is and maybe think about building it in two years' time? So that's the sense in which sentiment directly matters, whether we're measuring it or not. And I think that's sort of obviously true. Whether it moves up and down a lot is a much harder question. The surveys seem to show it does. I want to give you a final answer. Boy, I gave a lot of them. One of them better be right. And it's just to say there's a puzzle out there. This is the part that got me excited. The economy looks good. People sound miserable. Feels like both things can't be right. And so then do you throw away the hard data that we have or do you throw away the survey data? Or do you say there's actually something that connects the two of them? And my view is there's something that connects the two of them. When we look at the economic data, we learn what people are doing today. When we look at the survey data, we're figuring out how they feel about the future. And when we zero in on, and we discover they feel miserable, and when we zero in on possible causes, we suddenly realize that if you don't trust the economic leadership of your country, that leads you to feel miserable. And that also, by the way, where I'm going to keep going, that feeds in to this broader question of do presidents affect the economy? If what they do is every once in a generation is a president who is perceived to be a chaos monkey that makes it hard for businesses to plan, then presidents matter, at least to the extent that they can get the general public to either believe or not believe in them. [00:09:04] Speaker 2: Can you talk for a little bit about the actual hard data? You were talking about, you know, should we believe the survey? But I've seen a lot of people say that they think that actually the economics data is either wrong, which we've talked a lot about why we shouldn't believe this data, or incomplete. It's not capturing something that really is happening in the economy. So can you talk a little bit about the idea that this really is the right data and the thing that's missing is just the connecting thread? [00:09:32] Speaker 1: Let me go back a step and talk about how ideological bias can inflect the way we see the world. Many people, particularly on a university campus, so I meet a lot of these people, have no faith in the current American president. And therefore, that leads them to look for evidence that confirms their view that the world must be terrible under Trump. And when they go looking in the economic statistics, they really have to squint. It's been a disappointment to a certain class of people that the economy's actually been okay, even under an idiosyncratic president. But I want to be really clear. It kind of has been okay. Not for everyone and not in every way. And by the way, okay is the wrong question. The right question for us always to be asking is, could it be better? I genuinely, profoundly believe with better, sounder, smarter economic policy, our economy could be in a better place and people could enjoy higher paying jobs, more work, greater prosperity. But is the economy okay? Well, a 4.1 or 4.2% now unemployment rate is low compared to America's history. We're measuring it the same way we always did. It's low compared to other countries. It is good. Or at the very least, less bad than everyone else. An inflation rate of 3.5%, not ideal. But there was a time when inflation was 10%. And there was a time when it would be 5% one year, 10% the other, 7% the next. In the late 70s and early 80s. So it's not terrible. So I think those are the reality. Now, lots of people would say, well, Justin, you can't go and defend current economic statistics. No, no, those things are true. Some other things are true too. Inequality is greater than it's been in a long time. One thing we've talked a lot about, Augusta, is the slice of our national economic pie that's going to workers is the smallest that it's ever been. And so that can help you understand how the pie can be growing, but not everyone's slice looks like it's big enough. So I think our statistics do describe an important reality, and they're a great check on our ideological biases, and they force us to test our theories against reality. I think no set of statistics ever tells the full truth. There's no good statistic that says, how do you feel every morning? Do you wake up full of hope? No standard economic statistic, but we could go out there and have surveys and ask people. So we can turn these into statistics. And so that's basically the idea of what the index of consumer sentiment's trying to do. It says there's something out there, let's call it a feeling in the pitier stomach, which is not picked up by asking me whether I have a job and not picked up by asking me whether prices are rising and is not picked up by gross domestic product, but still is either intrinsically important or important because it shapes the rest of my economic behaviour. I don't think that's controversial to say. I don't think there's an economist alive who thinks that the statistics created by the government is the full human story. And so, you know, I am deeply in favour of any effort to paint more of the human story on our statistical canvas. I'm not sure I answered your question, did I? [00:12:50] Speaker 2: Yeah, you did. And it brought up another question, which is really nerdy, but I think people in the audience will enjoy it, which is, can you talk a little bit about survey research and its role in economics and how that's changed over time? Like what's the consensus in the field or what are the different thoughts about it? [00:13:07] Speaker 1: Yeah, great question. So I grew up in a world in which survey research was regarded with suspicion. And it's because economists really had this methodological obsession with watching what people do rather than what they say. And I started work on happiness. So I've actually done a great deal of research on asking people how happy they are, how satisfied they are with their life, how often they smile, things like that. And when I started doing that research, people would say to me, come on, Justin, why are you looking at this data that is just the answers to questions? What sort of idiot answers, looks at answers to survey questions? You're a goose. When I was really defensive, I would say, well, my friends in psychology, that's what they do for a living. But then I got sophisticated and defensive. It turns out every economic statistic comes from asking people questions. [00:14:00] Speaker 2: Right. [00:14:02] Speaker 1: How do we measure the unemployment rate? When I used to study the unemployment rate, no one would ever say, Justin, what are you doing? Studying survey responses. They didn't say, of course, people are just judging their unemployment relative to other people, but they used to say that about happiness. But at a very literal level, the way we measure the unemployment rate is a survey in which we say, last week, did you work for more than one hour? And then if you say, no, we say, were you willing to work? Were you available to work? And we put you through a bunch of questions. We say, did you look for work? And so it's literally, anytime someone's analysed the unemployment rate, it's literally answers to surveys. And you might say, yeah, yeah, but that's different. That's asking about an objective truth. No, it's not. I might say to you, did you search for work last week? And maybe you logged on to Craigslist for 15 minutes. But the truth is, there's not many jobs on Craigslist. And they're not really good jobs for people with your qualifications or Custer, but you did it. So now the question is, did you look for work last week? It's actually, that's a social construct. Whether we call that joblessness or helplessness, that is literally a social construct in the same way that, you know, everything else that my friends in the humanities say, everything's a social construct these days. It's socially conditioned. There is no way that I can track you and say, how many phone calls did you make? Partly because no one calls for jobs anymore. That would have been a perfectly reasonable question in 1983, right? But the question is, how many clicks and on which sites count as job search? What if you went out to a hipster Brooklyn bar and you asked all your friends if there was work? Does that count? [00:15:53] Speaker 2: I don't appreciate the stereotyping here. [00:15:56] Speaker 1: Really? I was about to say, and they ask each other if there's work at NPR or the local kale factory. The local kale factory. Does kale not come from factories? I'm going to work so much harder on not stereotyping. No, don't worry about it. No, I'm sorry. I owe all Brooklynites an apology. Next time you're at the co-op meeting, could you tell the board and the commissar that I apologized? [00:16:24] Speaker 2: All right. [00:16:25] Speaker 1: Sorry, that didn't last long, did it? Um, it's actually my insecurity. If you're from Brooklyn, you're cooler than me and that's okay. [00:16:32] Speaker 2: I'm from Seattle. I just live here. [00:16:35] Speaker 1: That's also cooler than me. So that's also okay. Um, okay. How do people feel about survey research was actually the question, not how do they feel about coarse stereotyping by a middle-aged man. Again, I apologize. So it's, so people are like, oh, okay, fine. Maybe the labor market statistics, their survey data. But industrial production is not. But actually, how do you collect data on industrial production? It's called the survey of industrial production. You call factories and you say, how many chunks of steel came out the other end? It's still a question. You know, it's all steel is an objective thing. Well, a diamond's objective because they used to only come from the ground. Now they can be grown in labs. Like, you know, everything, once you start looking at it, everything's a survey because everything has to come from asking someone something or having something monitor it in some way. Basically, all of GDP is surveys, by the way. Retail sales is surveys. Almost every economic statistic is a survey. So that was my sophisticated defensive answer. Didn't always work. What has happened, though, is that economists have become much more open to different forms of evidence. And this has partly happened as economists have, on the one hand, opened ourselves up to looking at more domains of human behavior. So in my career, I've studied the death penalty and I've studied divorce and I've studied happiness and I've studied gambling and so on. And if you want to try to get into any of those issues, I've studied elections. You want to be open to different forms of evidence. You can't just call the Bureau of Labor Statistics and ask them for the latest evidence on why couples get divorced. So that's one. Our question, the issues that we study has gotten broader. We're no longer just trying to predict M2 and industrial production. And I think also our openness to other methods and to other disciplines. So we've had the behavioral economics revolution, which also faced a lot of these same issues. Early on, these were questionnaire surveys asking people if they'd make dopey decisions. It didn't have a big impact on economists. But when they documented people actually making dopey decisions, it had a much bigger impact. I would like to argue that my field has become enormously more open-minded. And if there's a survey researcher watching right now, they might say, yes, but you guys are not all the way there yet. And I would simply concede the point. [00:19:03] Speaker 2: You could also make the point that there's other fields that could use an injection of more analytical methods. [00:19:12] Speaker 1: You're inviting me to agree and therefore get drummed out of the next faculty meeting I go to. The stakes that I have in this fight are far greater than yours, Augusta. I have dear friends who are sociologists and psychologists and historians and political scientists. And I don't want to start any fights. [00:19:36] Speaker 2: Let the record show I didn't name any of those fields. [00:19:43] Speaker 1: You are merely hinting other fields might exist. They have no names. There's no people attached to any of them. And you love them all like one would love each of their children equally. [00:19:54] Speaker 2: Exactly. If you run the record back, that's what I said. [00:19:56] Speaker 1: Okay. Well, now is when I will confess. In my very first class of Economics 101, I tell the students, the president has a council of economic advisors. He does not have a council of sociological advisors or a council of psychological advisors or a council of historical advisors. And there's a newspaper called The Economist, but there's not a newspaper called The Sociologist. And I continue this way because I want to keep enrollments up. And if I happen to be a booster for the field I love, I plead guilty. [00:20:31] Speaker 2: Okay, I'm going to ask you one final question. And it's that in your piece, you said that sentiment has been weak since 2020 under both parties and that that's a slightly different puzzle than what we're talking about. What's your best guess at that one? How do you think that today's gloom is just whatever's happening there versus the other bigger story? [00:20:54] Speaker 1: Right. So, I think this very sharp decline in consumer sentiment over the last year and a half probably can be largely described by the fact we had a very sharp deterioration in our trust that economic policy is being run well. Prior to that, there'd been something of a gap. The year you just mentioned was 2020. [00:21:15] Speaker 2: Mm-hmm. [00:21:16] Speaker 1: 2020 was a very interesting year in all of our lives or alternatively, a very boring year. Lots of things broke that year. And I think in ways that we still haven't fully come to terms with, we see ECAs. I see them when I teach students. I think we see them through fertility statistics. And I think we see them in a whole lot more places than people are necessarily thinking. So, that would be the most obvious one. I want to give you a boring one. My colleagues at the University of Michigan who collected these data, they changed from telephone interviews to web interviews. And it turns out asking people about how their feelings about the future, actually, those mode changes end up mattering more than you realize. In the analysis that we did, we corrected for those. But there's certainly some part of the story. Nate Silver argues that they're a particularly large part of the story. I think they're a notable part of the story. I wouldn't go further than that. [00:22:13] Speaker 2: Could the decline in consumer sentiment and the general gloom be the phones? [00:22:19] Speaker 1: Yes. [00:22:20] Speaker ?: Yeah. [00:22:20] Speaker 1: Could it be something else? Yes. Right. The problem with could it be the phones is it basically says at some point over the last many years, things started getting worse. But the, it's the phone story doesn't give you a lot of other, doesn't tell you when, it doesn't tell you which countries, which age groups, which, you know, it's a very, very hard story to test. Obviously, there are people like Jonathan Haidt working very hard on that problem. My sense of the, is it the phones literature? Um, sorry, to deal people in, this is the view that somehow, now that we spend all of our life, we go to restaurants and we just do this in front of each other. In fact, Augusta, maybe we could act out what it's like to be young today at a high school lunch cafeteria. [00:23:05] Speaker 2: You just need to swipe. [00:23:07] Speaker 1: Oh, I swipe. Did I forget to swipe because of the TikTok? Sorry. Yeah. Okay, great. Wasn't that amazing? And so we feel connected now, or some people would argue we feel less connected. Maybe we feel connected because I just saw your TikTok and I can't believe you have a dancing dog. That's incredible, Augusta. [00:23:22] Speaker 2: Thank you. [00:23:23] Speaker 1: I love that. You must have done a lot of work together. Um, look, that could be the story. Um, of course, phones weren't invented in 2020. [00:23:34] Speaker 2: Okay. [00:23:35] Speaker 1: Can I ask you one question? [00:23:37] Speaker 2: Yeah. [00:23:38] Speaker 1: I read in the YouTube comments, fear that you are in fact an AI. How would I know? [00:23:48] Speaker 2: I don't know what proves it. I mean, these are my hands. [00:23:52] Speaker 1: I see 11 fingers. [00:23:53] Speaker 2: Oh, well, I have a birth defect. You just make me look so bad. You have 10 and hold them back up. I'm not AI and I have 10 fingers. Um, this time it's nine. Yeah. I don't know how to prove it to people, but I'm not AI. This is not the first time I've received those accusations. Um, and I'll continue to defend myself. [00:24:16] Speaker 1: Should I hold up the newspaper? You spoke with a lot of M dashes just then. You can hear them? Yeah. I heard the M dashes. [00:24:24] Speaker 2: Put me through Pangram. Put a picture of me through Pangram. See what happens. [00:24:27] Speaker 1: That is not a weight bearing assertion. Is that what it says? That's a standard one as well. Anyway. All right. That was my question. You've had your chance to prove yourself. You didn't hold up a newspaper. But the fact that you sort of moved on past that, I think tells me everything I need to know. [00:24:43] Speaker 2: I don't have a newspaper. [00:24:43] Speaker 1: If you were an AI, that's exactly what you would have said. [00:24:48] Speaker 2: Sorry. I didn't quite catch that. That's it for our show. If you liked what you saw, if you learned something new, please like and subscribe to Platypus Economics. You can follow us on YouTube. You can follow us on Substack. You can follow us on social media. And don't forget to download the Platypus Economics podcast wherever you get your podcasts. [00:25:10] Speaker 1: Stay curious. Yes.