Marian Tupy: Hello, and welcome to the newest Human Progress Podcast. Today, I’m very pleased to be speaking with Scott Winship. He’s a senior fellow at the American Enterprise Institute and a director of the center on Opportunity and Social Mobility. Scott is one of the most careful and data driven analysts that I know. And today we will be talking about the state of the American worker. Welcome, Scott.
Scott Winship: Thank you very much, pleasure to be here.
Marian Tupy: So, let’s start with the sort of broader picture. It is my sense that the doom and gloom over the state of the American worker is much more prevalent than the data out there would support. In other words, there is much more gloom than the data would suggest. Am I terribly wrong in this assessment or am I within a ballpark?
Scott Winship: Yeah, I think there is a lot of anxiety out there. There are a lot of pretty wild claims on social media that have gotten a lot of traction that clearly resonate with quite a few people. I do think, you know, as you say, when we stick to the facts, there’s a big disconnect, I think, between a lot of the facts and the impressions online. The other thing that I think that I would say is that, I think we know social media is not necessarily the real world. And so, I think a lot of these impressions of anxiety get overstated just because the people who are concerned about their own lives, but also kind of have these impressions of how everybody else is doing. Their voices tend to be very loud on social media. There are surveys out there that look at how much anxiety or insecurity Americans actually express. And I wrote something on this a few weeks ago, but if you kind of look at a question that has been asked for 25 years or so, about how people feel their own personal finances are doing, about half the population says excellent or good, and we might wish that were higher.
Scott Winship: A very small number say poor. But the main thing is it’s not any lower. The number of people saying that their personal finances are excellent or good is not any lower than it was 25 years ago. It’s pretty, it stayed pretty steady over time. However, if you ask people how they think the American economy is doing, then you get really low levels of people who say excellent or good. And so, in some ways there’s this misimpression about how other people are doing. But if you ask people how they’re doing, it doesn’t it doesn’t seem like that’s especially worrisome today versus in the past.
Marian Tupy: I’m really grateful that you brought up that point because this is a finding in psychological literature that repeats itself time and time again. It’s called the optimism gap, whereby when people are asked to reflect on their own lives, it’s invariably much better than when they are asked about to analyze the situation in the country as a whole. And psychologists provide an explanation for this, is that people are much better judges of what is happening in their own personal lives as opposed to what is happening in the country as a whole, because that’s where the opinion is swayed by the nature of the news cycle, which is much more negative. And obviously, if it bleeds, it leads. People get a sense that the country is going to hell, even though they themselves might be well off. Let’s get into some specifics, and then I want to talk about especially your long intellectual battle with two critics, one being Oren Cass and then more recently, Michael Green. But let’s stick to the generalizations for now. There is this myth that the American worker has not really seen real progress since the 1970s. I think the 1970s features very prominently mostly on the left, whereas the right, it’s probably mostly in the 1950s. Maybe I’m wrong in that, but what would be your answer when people say the golden age, when leftists say golden age was in the 70s, right. To say the golden age was in the 50s? What would you say to that?
Scott Winship: Yeah, it’s interesting. I think the two years you tend to hear about, particularly on the left, are 1973 and sometimes 1979, but other times kind of 1969. And those are all business cycle peaks. 1973 ends up being an important year because the current population survey starts collecting wage data that year. And so, a lot of claims that you see kind of start with 1973, and then they show what happens to the trend over time. And I think you’re right, conservatives increasingly are going even further back and saying, boy, wasn’t life great in the mid 20th century when people could afford a home on one, raising a family on one income, they could afford to buy a home early on, they could go to college for very little, everybody was in a union, everybody had a pension. And those tend to be very impressionistic claims that lots of times just fall down completely when you actually look at the data. But, yeah, the idea that the 1970s was sort of a golden age, I think anybody who lived through it, I was born in 1973. So, I don’t have a lot of memories in the 1970s, but it was a period of inflation worse than we’ve had the last five years.
Scott Winship: And for a longer period of time it was combined with very high unemployment, which is different from what we’ve experienced the last five years. Culturally, there was, there was a ton going on. There was a lot of terrorism, actually, there was a lot of violence, there were a lot of drug overdoses and not a great decade I think by anybody’s standards. Even if you look at figures for the top 1% didn’t do especially well in the 1970s. But what you see often is that people will make these claims that earnings have stagnated since then, particularly men’s earnings. Some will even say that family income has, has not risen much since then. But invariably, these points are made using measures that are especially flawed. Typically they will use an inflation measure whenever you look at wage trends or earnings trends, you need to take into account the increase in the cost. And so, people will use an inflation measure that overstates inflation over time and thereby understates the improvement in living standards.
Scott Winship: That’s really the big error that a lot of people make. There are others as well, but when you dig into the data, the numbers that I’ve published in the last year suggest that since 1973, earnings among men are up by something like 45% and earnings among women are up by over 100%, more like 120%. That’s basically true. If you look at hourly wages, if you, if you don’t like looking at annual earnings, if you look at family incomes all of these things essentially are either at all time highs now or they have been at some point in the last five years. And so, the idea that there’s been stagnation is just way off base.
Marian Tupy: These wage numbers that you have given, are they simply monetary compensation or do they take into account all the non monetary compensation? Because one thing that people underestimate is that back in the day what the company paid you was mostly your paycheck. Right? But there were no transport costs included, holidays were fewer and so on. So, these numbers that you presented and is it just monetary or does take into account non monetary compensation?
Scott Winship: Yeah, so those figures I just gave are just earnings or wages. They don’t take into account non wage benefits.
Marian Tupy: And if you were to add the non wage benefits on top of the monetary, what would happen to that 45% increase? How much bigger would it be?
Scott Winship: Yeah, it’s a great question. It ends up being pretty tricky to answer. I think it’s fairly easy to addm employer contributions to health insurance, for instance, to add their contributions to retirement accounts. And when you do that, it goes up by a little bit more, I would say, maybe for men from 45% to more like 50%. But as you say, there are all these other ways that work has improved. The number of paid holiday, number of people with paid holidays, how many paid holidays they have, things like paid leave. It’s really hard to get long term estimates that incorporate those. There’s I think one federal survey that I know of and it doesn’t go that far back unfortunately. But yeah, and you can even go bigger than that and think of like workplace safety and other kind of workplace accommodations have improved a lot since the 1970s. So yeah, in some ways the numbers I’m giving are kind of like a floor for how to think about how much things….
Marian Tupy: It seems to me that floor…. So, we don’t want to be too optimistic or rather we want to stick to the facts. And some of the data that I have seen suggests that on top of the hourly wage increase that Americans have experienced, you should add another 30% in terms of non monetary benefits child care and dental and eyes and transport. Here at Cato, you’re also at a think tank. You know, people are given travel cards so they can to work and back for free and that sort of thing. Would the 30% on top of the wage data, would that be too optimistic, you think?
Scott Winship: I could believe it. I haven’t seen good numbers myself. That’s something I need to look into a little bit more, but yeah, I could certainly believe that.
Marian Tupy: Another major part of your work has to do with mobility and opportunity in the United States. So, we are told not only that American workers wages had not risen since the 1970s, but also that the society is fundamentally stagnant and broken and if you are born into unlucky circumstances, that you are stuck there. So, how would you summarize your research on the real story of economic and social mobility in the United States and what are the most common misperceptions that are being repeated both on the internet and maybe even by the journalists?
Scott Winship: Yeah, no, that’s a great question. So, there’s usually two main ways that people talk about intergenerational mobility. So, comparing adult kids to their, to their parents. The first one is called relative mobility. And the way to think about that is not are you better off in absolute terms over time, but if you start in the bottom fifth, are you able to make it to the middle fifth by the time you’re an adult? And I think the academic, conventional wisdom on this is sort of, on the one hand, that hasn’t gotten worse over time. Probably you’d be hard pressed to find people that say it’s gotten better over time, but it hasn’t gotten worse, which is kind of contrary to this idea that the American dream is dead and used to be so much easier to make these movements. And if you compare the US to other countries, the conventional wisdom is that the US does poorly relative to other countries. That second interpretation, I think is a misunderstanding of the data. I think it’s true if you’re looking at family income. So, if you start in the bottom fifth of family income, can you make it to the middle fifth of family income?
Scott Winship: That does look worse in the United States. I suspect it has a lot to do with the single parenthood trend in the US versus in European countries where there may technically be a fair amount of single parenthood, but it’s often kind of committed couples that are actually living together, just not married legally. So, if you look at family income, mobility, US does look worse. However, if you look at individual earnings, mobility. So, if my earnings put me, put my parents in the bottom fifth, can I make it to the middle fifth? There we actually look both for men and women, I think about as good as most of our peer countries, Jim Heckman had a really interesting set of papers comparing Denmark to the US that basically found this. And I’ve sort of seen it in comparisons to other countries too. The other big way that people think about mobility is what I alluded to first, which is are you better off in real terms than your parents were? Do you make more money at the same age than your parents did? And I think the conventional wisdom there is based on the work of Raj Chetty and his colleagues who had this big finding that if you were born in 1940, you had a 90% chance of ending up better off than your parents.
Scott Winship: Same age for kids born in 1980, that had dropped to about a 50% chance. So, big decline over time. Got written up in the New York Times, a lot of other places. My colleagues and I are investigating that evidence right now. Preliminarily, it looks to us like if you use a better inflation adjustment, if you take into account the fact that families are smaller today than in the past. And so, $50,000 goes further if it’s being shared among fewer people. When you sort of improve on these measures, it looks to us like it’s more like about 70% of the United States, 70% of recent waves of adults are better off than their parents were. Now that’s down from 90%. On the other hand, you know, I think it’s worth asking, would you rather have been born in 1980 and have a 70% chance of being better off than your parents and, you know, typically having some of the highest earnings in history or being born in the depths of the Great Depression and having an almost certain chance of beating your parents, but again, being born in the depths of the Great Depression. So, we tend, I think, sort of fetishize this absolute mobility idea a little too much. But even there things look better, I think than what even most academics believe.
Marian Tupy: That is a more nuanced answer than I have seen coming out of other scholars. I think that Jean Twenge, also Jeremy Horpedahl, and more recently, there was an article in Financial Times, basically showed the wealth, was it the income or the wealth? You might want to correct me on this, showing that basically the latest generation, which I think is Gen Z, are actually on a better trajectory than the millennials. And Millennials were better off than Gen Xers and Gen Xers were better off than, and so on. What am I getting right there and what am I getting wrong there?
Scott Winship: Yeah, no, that’s right. So, I would contrast everything I said about mobility was kind of comparing individual people to their parents and whether they’re higher or lower. If you’re just asking how well are generations doing, how high are their living standards as a whole versus previous generations, then you’re absolutely right. The evidence is that contrary to what a lot of people think the millennials, Gen Z are on track to, they’re already better off at the same age than previous generations. They’re on track for that to be even more impressive by the time they get to age 58, 60. But yeah, that’s true of earnings, it’s true of income, it’s true of wealth. My colleague Kevin Corinth, I think, has a paper on this too, with let’s invoke the Federal Reserve Board. Jeremy Jeremy Horpedahl has great stuff on it. And yeah, so that’s just contrary to a lot of what you hear, that millennials and Gen Z are just doing a lot worse. The real ways that more recent generations stand out that look more worrisome are in terms of student debt levels, which are higher because college graduation rates are higher than they used to be.
Scott Winship: And presumably for most people, that investment is going to pay off in higher earnings down the road. And homeownership, which is lower as well. Now, I think the homeownership one, I had a report this week that I put out trying to argue that the reason that homeownership is down for recent generations is because marriage has plummeted. And it turns out single young adults have never been able to own a home. It’s never been more than 20% or so. And so, if you have fewer married people, you’re going to have fewer people that can buy a home. That has been, I think, wrongly reinterpreted as sort of the opposite causality. You know, where people say, nobody’s getting married because they can’t afford a home anymore, and there are all these people out there that won’t get married until they can afford to buy a place. But that’s just never been true. It’s never been the case that a majority of young parents have owned a home. They’ve always tended to be renters, and then over time, they become homeowners.
Marian Tupy: To what extent, in your opinion, and this may not be part of your hardcore research, is the student loan problem, student loan debt problem driven by subsidies? In other words, if a university knows that you are able to borrow an extra $50,000 or an $80,000, well, why would they charge you any less than they would be leaving the money on the pavement?
Scott Winship: No, for sure. Yeah. And there’s great research that estimates the fraction of federal financial aid that colleges just capture because they do increase their tuition accordingly. The1 One Big Beautiful Bill act, which I’m not a giant defender of, mainly because I don’t think we can afford the increased deficits that came with it. But it did some really amazing stuff on higher education that a lot of my colleagues, Preston Cooper and Beth Akers and others were advocating for here, where essentially they’re going to hold more schools more accountable for the outcomes of their students. And so, the idea is that if you can’t produce graduates that actually are making more than a high school graduate tends to make, you’ll become ineligible for your students to receive federal student loans, Pell Grants, things like that. And hopefully that will be a way of arresting some of this increased in higher ed costs that really is due, I think, to the increase in federal financial aid.
Marian Tupy: And I wonder how the mechanics of this would work. So, presumably then the universities have an incentive to stop teaching courses that basically produce useless graduates. Would that be one of the ways?
Scott Winship: Yeah, that’s right. And it’s kind of on a program by program basis. So yeah, if a university does really well graduating mechanical engineers, then they’ll still be eligible for all this federal aid. But if it’s dance graduates, you know, don’t do any better than high school graduates do, then you’re not going to be able to qualify for financial aid if you do dance at this school. And there are ways to address that. If we worry about things like if we have enough public school teachers, for instance. But in general, yeah, it’s designed to encourage people like, look, you know, just financially this may not be a good deal for you, so you may want to think about doing this instead of that.
Marian Tupy: One last general question, and that has to do with your research on inequality. So, what did you find out about income inequality in the United States?
Scott Winship: Yeah, so it’s been interesting the extent to which this has died down a little bit in the last five years or so. I spent a lot of time….
Marian Tupy: It was a big deal. I remember the Obama administration. It was considered the biggest policy question in the country.
Scott Winship: Yep, that’s absolutely right. He declared it, I think the most important challenge facing the country. So, 10 years ago I was writing a ton on this and the two things that I was trying to push back on, one is just how much income concentration has increased over time because the initial claims from Thomas Piketty, Emmanuel Saez, later they added Gabriel Zucman to their team. But the initial claims from Piketty and Saez, were just these incredible increases in the share of income that was being captured. Sort of language they use, I think by the top 1%. And even at the time this was the early 2000s, I think that they started writing this. A number of researchers were saying, well, there are a number of problems with this, you’re only using pre tax and transfer incomes. And so, what that means is that you now have a bunch of elderly people who have very low pre tax and transfer incomes because they’re retired. A lot of them depend on Social Security but it also means that you’re not actually looking at the ways that we try to mitigate in inequality through progressive taxation, through the safety net.
Scott Winship: There was a problem with a lot of teenagers and young adults who were basically living with parents, but they get counted as a household or a unit in the data. And so, if they have a work study job or a summer job, they look like really low income Americans, whereas really that’s not how most of us would, would think about them. There were issues in how they treated capital gains, which is the income that the profit essentially that you make when you sell an asset versus what you paid for it. What tended to happen is that people sort of strategically timed when they received these gains based on tax law and based on the state of the economy. And so, you might have 20 years of gains that were sort of slowly building year by year, but they show up in the data in one year, which, as you can imagine, tends to inflate the incomes at the top. There were all kinds of inconsistencies around what sorts of gains are included. So, it’s only taxable gains. Which means, you know, the main way that the middle class has wealth is through homeownership, but for most people, those are not taxed and so that income doesn’t show up in the data and that overstates inequality.
Scott Winship: There were all these problems and over time, a number of researchers, I think the most important ones are kind of Rich Burkhauser, Bruce Meyer at the University of Chicago, Kevin Corinth, my colleague here, kind of tried to see what happens when you improve upon their methods. Eventually, Saez and Piketty and Zucman improved on the earlier estimates and even they found smaller increases in inequality over time. So, the consensus now is that inequality has gone up, but by much less than everybody thought around the time of the financial crisis. And when you take into account the redistribution that we do in progressive taxation… Oh, actually the most important paper I didn’t even mention was by David Splinter and Jerry Otten. And they find that essentially when you take redistribution into account, there hasn’t been much of an increase in inequality at all since even the 1960s. So, it’s really changed I think the conversation actually, it’s sort of ended the conversation. Nobody really….
Marian Tupy: Well, yes, I was aware of that paper, but let me…. You’re a mild mannered scholar, you might not want to endorse what I’m about to say, but I’m sorry, to me, reporting pre tax and pre transfer statistics seems like an intellectual deceit. It seems just wrong on its face. What doesn’t matter what your pre tax income is if the government ends up taking 40% to 50% of it? So, Elon Musk could sell his $400 billion worth of stock, but he will end up with $200 billion. So, what’s the point of reporting the extra 200 billion which we will never see, never enjoy? And in the same way, what’s the point of talking about Americans at the very bottom of the income ladder as not earning anything if they are still getting tens of thousands of dollars in transfer subsidies? When did this become a standard way of talking about inequality? Talking about completely a statistic that is meaningless to its core? I’m sorry, it seems very upsetting to me.
Scott Winship: Yeah, no I agree and I think initially if you were being really generous to Piketty and Saez, they were using IRS tax data and although even then the Congressional Budget Office was putting out inequality figures that were after tax. So, yeah, I agree. And sort of what folks on the left would tend to do is they would point to the Piketty and Saez numbers and say, you look how bad inequality is, we need more redistribution. It’s like, well you know, you’re not even counting like the redistribution we currently do. And so, if you’re, if you’re never going to look at redistribution, it’s never going to be enough. Right? These numbers will not improve even if we leveled incomes. If you look at pre tax income, it’s still going to be a lot of inequality and rising. So it’s nuts. And you’re right and the same thing happen terms of looking at poverty statistics. You know, even the official poverty statistics that the US government releases doesn’t include most of the ways that we have tried to reduce poverty over time. It doesn’t count as income. Food stamps, Medicaid, housing subsidies, refundable tax credits like the earned income tax credit, the child tax credit. These are the major ways over the last 20 years that we’ve tried to reduce poverty. And, and yet we have an official measure that basically ignores them entirely.
Marian Tupy: That’s extraordinary. It reminds me of the finding in the Gramm and Boudreaux book, about the myth of American inequality. I may get it slightly wrong, but I think they found that once you account for transfers, then the difference between the top quintile and the bottom quintile of American earners decreases from 16 to one to four to one. Yeah, it could be quartile, it could be quintile, I don’t know. But it basically shrinks the inequality from 16 to one to four to one. Which is extraordinary that we should have to put up with so much propaganda out there that doesn’t reflect reality.
Scott Winship: Yeah, and even, I think I’ve not seen good numbers on this, but I think if you did even pre tax and transfer inequality, but you restrict it to the working age population you would find much less worrisome numbers than the ones that Piketty and Saez were showing because they include a bunch of senior citizens who don’t have a lot of private income. And that’s….
Marian Tupy: But they are sitting on a pile of wealth that they’ve accumulated.
Scott Winship: Yes, that’s right. Yeah, exactly. They’re doing better than working as Americans in a lot of cases.
Marian Tupy: One last thing on the poverty measures, so you know, we do have relative poverty measure which sort of tends to, tends to bump around between 12% and 8% or something like that since the 1960s. But I have seen a couple of papers suggesting that once you measure poverty by consumption, then poverty really falls to about 2 to 2.5%. Let me ask you about this consumption poverty measure. Why would there be such a huge difference between the official poverty rate and the consumption poverty rate? Are people simply not reporting their incomes? Maybe are they having second jobs? Or are we again, which they don’t declare, or are we simply talking about once again not counting the transfers?
Scott Winship: Yeah. So, I think a couple things are going on. First, you’re absolutely right in speculating that people underreport their incomes. So, we’ve long known from this survey called the Consumer Expenditure Survey. If you look at the bottom fifth of families, for instance, and you ask them how much income they have and then you add up all of their spending, their consumption is a little bit different than spending in ways I’m too complicated to get into, I think. But essentially what you find is that people are spending, I want to say it’s 20 or 30% more than the incomes that they report. And I’ve talked to folks on the left who said, oh, that’s because they’re going into debt to do that extra spending. Bruce Meyer and his colleagues have looked at that and that doesn’t seem to be the case at all. It’s pretty well known that there’s a lot of underreporting at the bottom and at the very top, and that underreporting has gotten worse over time. So, over time you’ve got an increasingly too pessimistic poverty estimate if you’re just looking at income. So, that’s one issue.
Scott Winship: The second issue is that most of the, of the poverty measures that are out there, including the official one, do have this problem of not counting a bunch of sources of income. And lots of times they adjust the poverty lines to account for the rising cost of living over time in ways that overstate inflation. And so essentially the poverty line becomes a more and more difficult threshold to get over. So, when you measure incomes more comprehensively, when you use a better index, the income poverty measure trend and the end level tends to look a lot like the consumption poverty trend and level, which is reassuring. I think if you got, if you kind of measured income as best you could and got prices right as best you could, and you were still getting these very different trends and levels, that would be a little concerning. But they basically both tell the same story, which, as you say, is very low levels of poverty. If you’re using the standard of the 1960s, you could argue that poverty lines are ultimately pretty arbitrary. You can kind of set them so that 2% of the population is poor.
Scott Winship: You can set them so 10% of the population is poor. The important thing is that you hold them constant and then you look and see how much better or worse we’re doing over time, and we’re just doing much better over time. Rich Burkhauser and Kevin Corinth and Jeff Larimore have a paper where they say, all right, let’s take seriously when Lyndon Johnson said in 1963 that 20% of the population was in poverty. Let’s measure everything as best we can and see what that implies about poverty today. And it’s 2%. So, it’s very consistent with the consumption poverty numbers. Now you run into people who say, how can you believe poverty is at 2%? That’s completely unrealistic. To which I say it’s an arbitrary line. If you prefer starting today with the official poverty rate of around 10%. Okay, let’s do that. And then let’s go back to 1963 and say how many people lived under that line in 1963? It turns out it. That was 70%. Seven, zero.
Marian Tupy: Good God.
Scott Winship: Yeah. So either way, kind of pick the level of poverty today that you think is about right, and then look back in time.
Marian Tupy: Let me just repeat that for our listeners. So, if you decided that the poverty line today in the United States income, poverty line, was 10%, then by that standard, in the 1960s, 70% of Americans would be poor.
Scott Winship: That’s absolutely right.
Marian Tupy: Well, speaking about poverty lines, we have to talk about Michael Green. So, Michael Green, who I believe is an investor of some kind of posted on his substack a claim that the real poverty line in the United states today is 140,000. That’s $144,000. That basically, if you are making less than that, you are poor. And for reasons which are mysterious to me to this day, Free Press decided to republish that article on its website, which of course got everybody very excited about it. And then you stepped in and Jeremy stepped in and others. So, let’s talk a little bit about Michael Green. What does he get wrong at the most meta level?
Scott Winship: Yeah, definitely easier to talk about what he gets wrong than to try to talk about what he gets right. So, he released a series of essays, and the first one, as you said claimed pretty directly that the new poverty line was $140,000 for a family of four. Subsequently, people said, well, that’s not really what he meant. And then I had to write a second piece that sort of showed the six to eight direct statements he made that were saying he really thought 140,000 was an appropriate poverty line. The way, so he gets to this number two different ways. And I’m suspicious that his two different ways sort of magically ended up the same place. But the first thing that he did was he kind of misinterpreted what the poverty line today is, the official poverty line. If you look back in the early 1960s, Lyndon Johnson wanted to start the war on poverty. There was a lot of research in his council of economic advisors to try to get an estimate of poverty. And as I said before, it’s sort of arbitrary where you draw this line. I think LBJ kind of wanted to say a fifth of the population was poor.
Scott Winship: And there were a number of different researchers that had kind of arrived at a poverty line for families of around $3,000 at the time. One of those people, those researchers that had arrived at that amount was a woman named Mollie Orshansky, who had gotten there by basically noting that nationally, Americans at the time spent about a third of their income on food. The US Department of Agriculture had this kind of minimally adequate food budget. And so, she just took that and multiplied it by three, and that got you to a little over $3,000 for a family of four. Over the course of the 1960s, there was, you know, there wasn’t clear for a little while how they were going to measure poverty. Eventually in 1969, they said, let’s just go with essentially, Mollie Orshansky’s numbers for 1963, except we’re just going to adjust them for inflation moving forward. So, we’re not going to get into how much of people’s income are they spending on food. We’re not going to get into changing what this adequate diet is. We’re just purely going to take the pretty arbitrary 1963 numbers and we’ll just adjust them for the cost of living over time.
Scott Winship: And that’s still the official poverty line today, oversimplifying a little bit. But Green sort of probably assisted by an LLM, thought he understood how the poverty line was initially developed. And he said, okay, well, let’s look today at how much Americans spend on food. And when you do that, it turns out today Americans don’t spend a third of their income on food. They spend more like 5% or 6% of their income on food. So, dramatically lower. And then from there, Green said, okay, so let’s not multiply the sort of original food budget by three. We instead should be multiplying it by, you know, one divided by 5% or 6%, which ends up being a multiple of like, 17. And so, if you multiply this number by 17 instead of multiplying it by three, you get a much higher threshold. You get a $140,000 threshold. Just a really ludicrous way of doing it, because the problem here is that the fact that we’re only spending 5% or 6% of our income today instead of a third of our income on food is because we’re so much richer over time.
Scott Winship: As societies get richer over time, they’re able to afford all the food they used to buy, plus better food, plus a bunch of other things that they couldn’t afford back in the early 1960s. So, essentially that smaller share that we spend on food is an indicator of how much richer we are. But Green has turned it into an indicator of how much poorer we are. And so it just makes no sense at all. The other way that he got to 140,000 was that he took these estimates of how much “families of four,” need to spend on things like food, childcare, health care, housing, transportation, and some other things. And he got these from this living wage calculator that someone has done online. And when you do that, he also got 240,000 that way. That had a bunch of problems with it. Maybe the biggest one was identified by Jeremy, which was that he was using Essex County, New Jersey, to represent the United States. Turns out Essex county is home to Newark’s western suburbs, which are quite wealthy. And that county is sort of in the four or five richest counties in the United States. So, their incomes are higher. What they spend is correspondingly higher. And if you take what they spend as a needs threshold for the average American, it’s just setting a bar that’s. That’s way too high.
Marian Tupy: It seems to me that one problem that we encounter is the different measures of inflation. A lot of Americans don’t trust the statistics that are being pushed out by the federal government. And then we are also settled with a number of different inflation calculators. If you were to talk to a skeptical American, what would you tell him? What is the best measure of letting how much better off in life you are as opposed to your ancestors?
Scott Winship: Yeah, it’s a great question. It’s a tough project, trying to convince people who are skeptical. These price indexes are very technical. There’s disagreement more than there should be, I think, among even researchers about what the best ones are, among people who really kind of know what they’re doing. There are a certain class of these price indexes that are chained. And what does that mean, to simplify a little bit? It basically means that they take into account the fact that as the price of some things go up, you don’t have to just keep buying the same amount of those things as you used to and just, absorb the price increase. You can actually switch and buy fewer of those things and buy more of something else, and you’re worse off than you were before the price increase. But you’re not as worse off as if you would have had to just keep buying the same amount of this other stuff that’s now more expensive.
Marian Tupy: Could you give me one example to concretize it?
Scott Winship: Yep, for sure. So, red delicious apples, the price of those go up. And so, instead of buying those, you switch to Gala apples. The price of coffee goes up, so you buy more tea, and more generally, the price of gas goes up. And so, you spend more money on subscription services for your TV at home. These are ways that people economize and that you need to take into account when you’re trying to figure out over time how much better or worse off people are. And so, when you look at that class of price indexes, things look much better than even most of our official statistics show, because a lot of our official poverty numbers, for instance, the poverty rate in America uses a price index that does not do any of that, or at least has only done it in recent years. If you sort of go back to the 1970s, it doesn’t take any of that into account. So, even that explanation is probably more technical than, you know, I probably was.
Marian Tupy: No, but it gives you a sense of the kinds of trade offs that, that people do. One last question on this, and that has to do with quality improvements. Do we have any inflation measure that actually takes quality improvements into consideration? So obviously an iPhone 16 or whatever we are now on 18 is a very different thing from iPhone 1. Does anybody account for that? Because, an American may think to himself, oh, I had to pay $500 for an iPhone 20 years ago, and now it’s $700. Therefore I am worse off forgetting that these are two almost completely different things.
Scott Winship: Yeah, that’s right. So, all of these price indexes try to take improved quality into account. And that’s you kind of have to make some effort to do that if you want to do better than the kind of Oren Cass or Michael Green estimates that just take this, that just look at what people spend without bothering to find out whether the things they’re buying are nicer or not. The price indexes do try to take that into account. Some do better than others. They universally understate how much better the stuff that we buy is. And that’s typically because you think of an iPhone in the 19… So, the roots of kind of how an iPhone is captured in these price indexes goes back to rotary telephones sort of mid 20th century. That used to be what people bought for telephone service. Over time you would get cordless phones. And so, as those became more popular and slowly entered into the budgets of more people, the data eventually took into account, oh, we’ve got cordless phones now. But it but it’s quite a while after the time that the sort of first movers bought cordless phones. And in the meantime, the price of cordless phones plummeted a ton before they even ended up in the price data. And that price decline gets missed by the statistics. So, then you move forward.
Marian Tupy: That’s pretty big.
Scott Winship: Yeah. And then you move from cordless phones to, oh, the initial mobile phones that were not smartphones, and you sort of get the same dynamic there. Eventually enough people had flip phones that they did enter into the data. But the huge price decline from the first person that bought a flip phone to the point where everybody was buying flip phones, that gets missed. That’s a price drop that gets missed in the data as well. And then you go to smartphones, and then it becomes super complicated because a smartphone isn’t just replacing a flip phone, it’s replacing a camera, it’s replacing an atlas, it’s replacing an iPod, which was replacing other things before that, it’s replacing so many different things. And you just miss the extent to which we’re better off by being able to buy an iPhone instead of having to buy that whole bundle of other things that we used to have to buy. The other thing, the price indexes universally can’t really handle well, is free stuff. And so you think about the internet and all of the information that we get from the internet, from email to artificial intelligence now to social media, to the extent that that’s free, it doesn’t enter into any of this data except to the extent that the companies are buying advertisements from Google and from other websites. The cost of those ads make it into the data as the price of internet access in some ways. But it doesn’t take into account the fact that we’re getting all this free stuff.
Marian Tupy: And the fact that I can call anybody in the United States from coast to coast essentially for free. But whichever way you look at it, it would be fair to say, I think, that these inflation indices, which Americans are so skeptical about, are actually quite conservative in terms of the almost hyper conservative in terms of underestimating the rate of improvement. Okay, well, you mentioned him, so in the last few minutes, we are going to talk about the Lord Voldemort of this debate. And that’s, of course, Oren Cass from the American Compass. Obviously, a lot of mileage that Orin has gotten from popularizing the notion that American worker is in decline. In the last few minutes of our conversation, how would you summarize the biggest problems with the way that Oren Cass is approaching this subject and the biggest disagreements that you have with him?
Scott Winship: Yeah, so I have a lot of disagreements with him on policy, but sort of sticking with just his methods. I think Oren starts from a belief that Americans are doing worse over time, American workers are doing worse over time. And it’s because of dumb economic elites. It’s because of the China shock, it’s because of the trade deficit, it’s because of all these things that in reality I think are lousy on the merits, haven’t affected incomes all that much. But he starts from that point and then, he sort of constructed ways to back into it. So, for a long time, he was putting out a series on wages in the US and arguing that American wages were stagnant, have been stagnant since the early 1970s. Well, now we get back to the sort of price index debate. You know, he’s using this price index which as I said, doesn’t take into account all of this substitution that Americans can do as the prices of some things go up and relative to other things, it doesn’t take into account quality improvement as well as a bunch of other price indexes do.
Scott Winship: So it’s widely known that if you’re looking at 50 year trends, you don’t use this price index that Orin has used. When you do use that price index, there are measures that show that earning, that wages haven’t increased over time. So, initially that was his big claim. Wages are flat over 50 years, this sounds bad. He has also argued that actually like price indexes are just not the right way to look at improvement over time. And they’re so, flawed that I, Oren Cass, have to come up with this other method that gets around their problems. And this other method is something he called the cost of thriving index. In some ways this is kind of a variation on what Michael Green did. He sort of takes four or five different categories of spending that are important, housing, healthcare, transportation, food. Probably forgetting one, I think.
Marian Tupy: Education, maybe.
Scott Winship: Yes, higher education. Yeah. And he comes up with a standard like this is how much what a, something that middle class people used to be able to afford. That’s how much they used to spend. And now he claims that fewer people can afford that over time. But the problem is, he’s not accounting for all of the ways that these things have gotten better over time. So the typical car that people bought in 1985, was less fuel efficient, less safe, included far fewer amenities than the typical car that people purchase today. And so, if you’re just looking at how much people spend on this typical car versus how much they used to spend on some other typical car, you’re going to say, well, look, that the cost of this car has gone up so much. But really what’s happened is that people could afford to buy the old 1985 car if they wanted, but people don’t want to buy that 1985 car anymore. They want to buy the fancier better version that we have today. And it’s not that prices have gone up. It’s that incomes have gone up and we buy nicer things.
Scott Winship: Some other problems with Oren’s index, he includes the things that, for the most part, the price has gone up faster than wages have gone up. But we sort of. Jeremy Hopperdahl it all, and I wrote a big critique of this. We showed that the things he leaves out are basically everything that has the prices have risen less quickly than wages have. And so, that’s another advantage of price indexes, is they don’t just take into account the big ticket things that have gotten more expensive. They take into account all the things that have gotten less expensive, such as clothes, such as entertainment, food, a bunch of other things. Yeah.
Marian Tupy: So, it seems to me, I mean, the key for Americans to understand, I think, if they listen to this podcast, is that wages are increasing, adjusted for inflation, but also because of productivity gains. The key is what are parts of the spending that are either falling in price or alternatively, are growing at a lower rate than wages? And what are those parts of spending that are growing at a faster rate than than wages? And things that fall in price are things like furnishings, toys, food, electronics, TVs, they’re cheap in spite of their massive sizes. What is becoming more expensive relative to wages would be things like education, health care and education, healthcare primarily. And we already talked about the nefarious effect of subsidies on university education and so forth. But let’s maybe finish talking a little bit about the Baumol effect. The Baumol effect is basically that even in places where there is no productivity, where there are no productivity gains, productivity increases. We still have to pay people a lot of money, otherwise they wouldn’t be actually doing those jobs in the first place. So, they may be inefficient, but we need to pay them.
Marian Tupy: Otherwise that nurse, for example, is not going to be doing that, she’s going to be doing something else. And in the absence of a robotic nurse, you still need to keep her in the job. Okay. However, in our book that I co wrote with Gale Pooley, Superabundance, what we found is that when you look at plastic surgery the prices there are dropping like a rock. In other words, relative to income, cosmetic surgeries are simply decreasing. So, I’m wondering, how much is the Baumol effect at work in education and health care as opposed to government subsidies driving up the price? Would we still get the Baumol effect, or would the Baumol effect be lessened if we had proper competition, in your view?
Scott Winship: Yeah, I think it’s a great question, and I don’t think there’s been enough research done on it. But clearly, I think the government intervention in some of these sectors has been incredibly important. You think about healthcare in every realm except for healthcare, insurance is essentially a tool to pay a little bit in regular amounts to avoid a giant cost that you have a very low probability of ever having to pay. But if you have to pay that giant cost, it would really hurt you. And so, instead you pay a smaller amount as an insurance premium every month, and then if you incur the giant cost, your insurance will pay for it. So, that’s why we have car insurance. It’s the risk that you’re going to get in a big car accident. And rather than being bankrupted because you can’t afford it you’ve been paying into this insurance policy that will take care of it in healthcare, largely because of government mandates about what has to be covered. It’s not that at all. Health insurance covers annual checkups, which are completely predictable.
Scott Winship: They’re not risks of catastrophic costs that you have a low probability of incurring. It’s something that, you know you’re going to get an annual checkup every year, but we include it in health insurance coverage. By including a bunch of things like that, you basically are incentivizing people to over consume health care. And that pushes up the cost of health insurance premiums. Some of that gets captured by healthcare providers in the same way that we talked about colleges, capturing some of the subsidy that the federal government provides. It’s just a recipe for inflating costs. The analogy I kind of try to make is imagine if government mandated that car insurance had to cover paint jobs for cars. Well, what would happen? Well, if I’m paying an insurance premium that includes an annual paint job and I’m not taking advantage of it, then I’m a sucker. Right? So, other people are like getting these fancy paint jobs with their insurance coverage, so I may as well also but that’s going to increase the cost of car insurance. It’s going to increase the cost of paint jobs. And that’s sort of what we’re getting in the healthcare sector because of the way the health that we’ve organized health insurance.
Scott Winship: Same for housing. Housing, we have kind of a different problem, which is that there’s regulation of land use and of zoning. It’s too difficult in a lot of places to build smaller homes. And so, we have a reduced supply that pushes up prices above where they would be in a competitive market. And so, as you say, part of the reason housing costs have gone up so much is because of this government intervention.
Marian Tupy: Well, I wrote down something which I thought was really important when you said that many of the critics of where the American worker is today and where the United States is in general start with a premise that something has gone wrong, and then they try to find the data to fit their preconceived notions. But hopefully this conversation has alleviated some of those concerns or at the very least has put a more realistic spin on what is happening in the United States. So, I’m very grateful to you for the time that you have taken to talk to me. Much appreciated.
Scott Winship: Oh, it’s been a pleasure. I’m a big fan of you and your work and glad you’re doing all the important work you’re doing as well.