01 / 05
American Poverty Is a Measurement Problem | Podcast Highlights

Blog Post | Poverty Rates

American Poverty Is a Measurement Problem | Podcast Highlights

Marian Tupy interviews Scott Winship about how bad measurement choices distort the picture of poverty and inequality in the United States.

Listen to the podcast or read the full transcript here.

Let’s start with the broader picture. It is my sense that popular narratives about the state of the American worker are much darker than the data support.

Am I terribly wrong in this assessment?

There are surveys that look at economic anxiety or insecurity among Americans. And if you look at a question that has been asked for 25 years or so, about how people feel about their own personal finances, about half the population says their finances are “excellent or good.” We might wish that number were higher, but the main thing is it’s not any lower than it was 25 years ago. It’s been pretty steady over time.

However, if you ask people how they think the American economy is doing, the share of people who say “excellent or good” is really low. So, there’s this misconception about how other people are doing, but if you ask people how they’re doing, they aren’t especially worried.

That is a finding in psychological literature that repeats itself time and time again. It’s called the optimism gap. When people are asked to reflect on their own lives, they are invariably much more optimistic than when they are asked about the situation in the country. The explanation for this phenomenon, according to psychologists, is that people are much better judges of what is happening in their own personal lives as opposed to what is happening to the country as a whole. In the latter case, their opinion is also swayed by the media, which is very negative.

There is also this myth that the American worker has not really seen real progress since the 1970s. What would you say to that?

I was born in 1973, so I don’t have a lot of memories in the 1970s, but it was a period of high inflation, worse than we’ve had in the last five years. And for a longer period of time, there was very high unemployment. There was a lot of terrorism and other violence. There were a lot of drug overdoses. Not a great decade, I think, by anybody’s standards.

People often claim that earnings have stagnated since the 1970s, particularly men’s earnings, but the numbers that I’ve published in the last year suggest that since 1973, earnings among men are up by something like 45 percent, and earnings among women are up by around 120 percent. Hourly wages, annual incomes, and family incomes are all either at all-time highs now or have been at some point in the last five years.

Another major part of your work has to do with mobility and opportunity in the United States. We are told that society is fundamentally stagnant: if you are born into unlucky circumstances, then you are stuck there.

So, how would you summarize your research on economic and social mobility in the United States?

There are two main ways that people talk about intergenerational mobility. One is comparing adult kids to their parents. The way to think about that is “if you start in the bottom fifth, are you able to make it to the middle fifth by the time you’re an adult?” That hasn’t gotten worse over time, but you’d be hard-pressed to find people who say it’s gotten better, either.

The other big way that people think about mobility is “Do you make more money at the same age than your parents did?” The conventional wisdom there is based on the work of Raj Chetty and his colleagues, who found that, if you were born in 1940, you had a 90 percent chance of ending up better off than your parents. For kids born in 1980, that had dropped to about a 50 percent chance. So, a big decline over time.

My colleagues and I are investigating that evidence right now. Preliminarily, it looks to us like if you use a better inflation adjustment, and if you take into account the fact that families have become smaller, it looks to us like in the United States, 70 percent of recent waves of adults are better off than their parents were, down from 90 percent.

Now, everything I said about mobility was comparing individual people to their parents. If you’re just asking how well new generations are doing compared to previous generations, the evidence is that Millennials and Gen Z are already better off at the same age than previous generations in terms of earnings and wealth.

However, student debt levels are higher in younger generations because college graduation rates are higher than they used to be. But for most people, that investment is going to pay off down the road. And Homeownership is lower. Now, I think the reason that homeownership is down for recent generations is that marriage rates have plummeted, and single young adults have never had high homeownership rates. There’s been a case of reverse causality there, where people say, “nobody’s getting married because they can’t afford a home,” but 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 after some time, they become homeowners.

What did you find out about income inequality in the United States?

Ten years ago, I was writing a ton on this. I was trying to push back on Thomas Piketty and Emmanuel Saez, who were claiming that there was an incredible increase in the share of income that was being captured by the top 1 percent.

There were a number of problems with their analysis. They used pre-tax and pre-transfer incomes, which miss the effect of progressive taxation and the social safety net. There was a problem where a lot of teenagers and young adults who were living with parents with a summer job were classified as low-income Americans. There were also issues with how they treated capital gains. People strategically time when they receive these gains based on tax law and the state of the economy, so you might have 20 years of gains that show up in the data as one year, which, as you can imagine, tends to inflate the incomes at the top. They also only included taxable gains, and the main way that the middle class gets wealth is through homeownership, which didn’t show up in the data.

Eventually, Saez, Piketty, and Zucman improved on the earlier estimates and 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 redistribution through progressive taxation, there hasn’t been much of an increase in inequality since the 1960s.

You’re a mild-mannered scholar and might not want to endorse what I’m about to say, but reporting pre-tax and pre-transfer statistics seems like intellectual deceit. What does it matter what your pre-tax income is if the government ends up taking 40 to 50 percent of it? And what’s the point of talking about Americans at the very bottom of the income ladder not earning anything if they are getting tens of thousands of dollars in transfers?

Yeah, I agree. Folks on the left would point to the Piketty and Saez numbers and say, “Look how bad inequality is, we need more redistribution,” but they weren’t even counting the redistribution we currently do. If you don’t include redistribution, inequality wouldn’t fall even if we leveled incomes.

Even the official poverty statistics that the US government releases don’t include most of the ways that we have tried to reduce poverty over time. It doesn’t count food stamps, Medicaid, housing subsidies, or refundable tax credits, which are the major ways that we’ve tried to reduce poverty over the last 20 years.

That’s extraordinary.

It reminds me of the finding in the Gramm and Boudreaux book, The Triumph of Economic Freedom: Debunking the Seven Great Myths of American Capitalism. They found that once you account for taxes and transfers, the difference between the top quintile and the bottom quintile of American earners decreases from 16 to one to four to one.

One last thing on poverty. We have a relative poverty measure that tends to bump around between 12 percent and 8 percent since the 1960s. But I have seen a couple of papers suggesting that once you measure poverty by consumption, it falls to about 2 to 2.5 percent.

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?

I think a couple of things are going on.

First, you’re absolutely right that people underreport their incomes. If you look at the bottom fifth of families, for instance, people are spending 20 or 30 percent more than the incomes they report. And I’ve talked to folks on the left who say, “Oh, that’s because they’re going into debt.” 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 both at the bottom and at the very top, and that underreporting has gotten worse over time.

The second issue is that most of the poverty measures out there, including the official ones, simply don’t count a bunch of sources of income. And they overstate inflation, so the poverty line becomes a more and more difficult threshold to get over. When you measure incomes more comprehensively and when you use a better price index, the income poverty trend tends to look a lot like the consumption poverty trend.

You could argue that poverty lines are ultimately pretty arbitrary. You can set them so that 2 percent of the population is poor, or you can set them so that 10 percent of the population is poor. The important thing is that you hold them constant over time. Rich Burkhauser, Kevin Corinth, and Jeff Larrimore have a paper where they say, All right, let’s take seriously when Lyndon Johnson said in 1963 that 20 percent 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 percent.

Now, you run into people who say, “How can you believe poverty is at 2 percent? That’s completely unrealistic.” To which I say it’s arbitrary. If you prefer to start today with the official poverty rate of around 10 percent, we can go back to 1963 and see how many people lived under that line. It turns out that was 70 percent.

Good god.

Let me just repeat that for our listeners. If you decided that the poverty rate today in the United States is 10 percent, then, by that standard, 70 percent of Americans were poor in the 1960s?

That’s right.

Michael Green, who I believe is an investor of some kind, posted on his Substack that the real poverty line in the United States today should be $140,000. For reasons that are mysterious to me, The Free Press decided to republish that article on its website, which of course got everybody very excited. And then you stepped in. So, what does he get wrong?

He gets to this number two different ways.

The first thing he did was misinterpret the official poverty line. In the early 1960s, Lyndon Johnson wanted to start the war on poverty, and he wanted to say that a fifth of the population was poor. And there were a number of different researchers who had arrived at a poverty line of around $3,000 at the time. One of those researchers was a woman named Mollie Orshansky, who had gotten there by noting that nationally, Americans at the time spent about a third of their incomes on food. The US Department of Agriculture had this minimally adequate food budget, 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. Eventually, in 1969, they said, let’s just go with Mollie Orshansky’s numbers, except we’re going to adjust them for inflation moving forward. So, we’re not going to get into how much of people’s income they spend on food. We’re not going to change what an adequate diet is. We’re just going to take her line and adjust it for the cost of living over time. And that’s still the official poverty line today.

Green thought he understood how the poverty line was initially developed, and he said, “Okay, let’s look at how much Americans spend on food today.” And it turns out Americans today spend around 5 or 6 percent of their income on food. From there, Green said, “Okay, so let’s not multiply the original food budget by three. Instead, we should be multiplying it by 17. And clearly, if you multiply this number by 17 instead of 3, you get a much higher threshold.

It’s a ludicrous way of calculating poverty. We spend a smaller share of our incomes on food because we are richer, but Green has used that to argue we are poorer. It makes no sense.

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. He got these from this living wage calculator that someone has created online, which also had a bunch of problems with it. Maybe the biggest one was that he was using Essex County, New Jersey, to represent the United States. Turns out Essex County is one of the four or five richest counties in the country.

The key thing for Americans to understand is that, in general, wages are increasing faster than prices. However, certain parts of our spending, primarily education and healthcare, are becoming more expensive relative to wages.

So let’s finish by talking a little bit about the Baumol effect, which is basically that, even in industries where there is no growth in productivity, we still have to pay people higher wages because of productivity growth in other industries. Basically, nurses and teachers might not be getting much more productive over time, but we still need to pay them more, or we won’t have any nurses or teachers.

However, in the book that I co-wrote with Gale Pooley, Superabundance, we found that plastic surgery prices are dropping like a rock relative to income. So, how much of the inflation in healthcare is thanks to the Baumol effect as opposed to government subsidies? Would the Baumol effect be lessened if we had proper competition?

It’s a great question, and I don’t think there’s been enough research done on it. But clearly, government intervention has been incredibly important.

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 low probability of ever having to pay. That’s why we have car insurance. There’s a small chance that you’re going to get in a big car accident, and rather than risking bankruptcy if that accident happens, you pay into an insurance policy that will take care of it.

In healthcare, largely because of government mandates, it’s not like that at all. Health insurance covers annual checkups, which are completely predictable. By including a bunch of things like that in health insurance coverage, you incentivize people to overconsume health care, which pushes up costs.

The analogy I make is imagine if the government mandated that car insurance had to cover paint jobs. Well, if I’m paying for insurance that includes an annual paint job and I’m not taking advantage of it, then I’m a sucker. Other people are getting these fancy paint jobs with their insurance coverage, so I will too. That’s going to increase the cost of car insurance, and it’s going to increase the cost of paint jobs. That’s what we’re getting in the healthcare sector.

World Bank | Income Inequality

High-Inequality Economies Have Declined by More Than One-Third

“A measure that captures prosperity should account for both the level of welfare and level of inequality. A measure that captures inequality, such as the Gini index, is an example of a measure of inclusivity. The World Bank reports the number of economies with high levels of inequality, defined as those with a Gini index greater than 40. Based on this, the number of economies with high inequality has declined by more than a third in the past three decades.”

From World Bank.

World Bank | Income & Inequality

Global Progress Lifts Six Nations Into Higher Income Classifications

“Since 1987, the global income classification landscape has changed significantly, with the share of economies classified as low-income declining from 30% to 11%. These shifts have not been evenly distributed, however, with some countries moving across income groups much faster than others. This year’s update to the World Bank Group Country Income Classifications shows six countries moved to a higher income category, but each has a different story to tell: a country emerging from economic crisis, an export powerhouse outpacing its peers, and an economy that turned out to be 10% larger than previously thought are just three of them…

This year, none of the countries assessed moved down. Five moved from lower-middle to upper-middle income: Jordan, Micronesia, the Philippines, Sri Lanka, and Viet Nam. One moved from low to lower-middle income: Togo. They reached the same thresholds through very different paths.”

From World Bank.

Blog Post | Income & Inequality

Was COVID Also an Inequality Pandemic?

COVID slowed but couldn’t stop the fall in global inequality.

Summary: Recent debates have framed global inequality as rapidly worsening, particularly in response to the COVID-19 pandemic. Evidence from the Inequality of Human Progress Index indicates that, despite temporary setbacks, long-term declines in inequality across multiple dimensions of wellbeing have largely persisted, with global disparities remaining well below 1990s levels.


Affordability fears, talk of a “K-shaped” economy, and claims of a new Gilded Age have pushed inequality to the center of today’s policy debates. Calls for a worldwide wealth tax and other unprecedented measures are not treated as radical but as inevitable—across academianon-profitsthe press, and international organizations, including the United Nations.

The COVID-19 pandemic seemed to clinch the case. As economies contracted and progress in poorer countries stalled, it was easy to assume that decades of convergence between developed and developing countries had come to an end. The authors of one Oxfam paper, for example, proclaimed during the pandemic that “unparalleled action [is] needed to combat unprecedented inequality in the wake of COVID-19.”

New research suggests a more nuanced reality. The updated Inequality of Human Progress Index assesses how the pandemic affected progress toward a more prosperous and equal world.

The pandemic clearly slowed improvement in global living standards and interrupted the pace at which countries were becoming more equal. It did not, however, cancel out the long-term, positive trends. Even under the strain of COVID-19, its attendant lockdowns, and other forceful policy responses, global inequality across key measures of well-being remained lower than it was a generation ago.

The index looks beyond income alone. It measures inequality across eight dimensions that shape everyday life, including lifespan, child survival, nutrition, education, internet access, environmental safety, income, and political freedom. The index, which I co-authored with George Mason University economist Vincent Geloso, seeks to offer a fuller view of gaps in global development, taking into account more aspects of human well-being than any prior index of inequality.

The data show a substantial decline in global inequality over the past three decades as rising prosperity allowed poor countries to narrow gaps with rich ones. That pattern held through 2019. During the pandemic years of 2020 and 2021, progress slowed sharply and, in some areas, stalled or modestly reversed. Yet the earlier gains were not erased.

This distinction is important. COVID-19 was a severe shock. Life expectancy fell worldwide. School closures disrupted education. Economic activity and international trade declined, with especially devastating effects on low-income countries. The index reflects these setbacks. Inequality stopped falling at its earlier pace and, in some measures, edged upward slightly after years of progress. Still, the overall level of global inequality remained far below where it stood in the 1990s.

In a few areas, improvement continued even during the crisis. Internet access expanded rapidly, especially in poorer countries, reducing inequality in access to information to its lowest level on record. Faster regulatory approvals amid the pandemic helped bring more people online. In Kenya, for example, Alphabet’s high-altitude internet balloons were finally cleared in 2020, allowing rural areas to gain internet access for the first time. The project had been stalled in regulatory review for nearly two years before the crisis prompted action.

Not all the data were encouraging. Inequality in political liberty ticked up during the pandemic as many countries took a turn toward greater authoritarianism. Even with the long-term shift toward electoral democracy intact, the setback shows the importance of protecting political liberty during emergencies.

For all the turmoil, the damage across different measures of well-being was thankfully limited.

These findings complicate popular claims that the world is experiencing a runaway increase in inequality. Calls for a global wealth tax, massive new aid commitments, or other significant expansions of state redistribution often rest on the premise that trade and free enterprise have failed to deliver shared gains. The data suggest otherwise.

If anything, the pandemic highlighted how sensitive progress can be to disruptions in markets. Countries with greater economic freedom generally proved more resilient. In contrast, prolonged lockdowns and restrictions often imposed heavy costs on poorer populations, particularly in countries where remote work and online schooling were not viable options for most people.

The broader lesson is that global convergence is neither automatic nor guaranteed, but instead depends on certain conditions such as undisturbed markets, even as long-term progress has proven more robust than critics often assume.

Mistaken narratives about global inequality have real consequences. They shape public opinion and influence policymakers to embrace sweeping interventions. A more accurate assessment of recent history suggests a need for caution.

COVID-19 tested the global economy in ways few events in modern history have. It slowed human progress and exposed vulnerabilities. At the same time, it demonstrated the durability of the long-term trend toward lower global inequality. Preserving and strengthening the policies and institutions that made that progress possible, including economic and political freedoms, remains a better bet than assuming they have already failed. The gains of recent decades have left the world both better off and more equal.

This article was published in the Orange County Register on 2/1/2026.

Demography | Personal Income

Generational Progress on Income Growth Continues in US

“Whether each generation of Americans continues to economically surpass the previous one has recently been called into question. We construct a posttax, posttransfer income measure from 1963 to 2023 based on the Current Population Survey Annual Social and Economic Supplement that allows us to consistently compare the economic well-being of five generations of Americans at ages 36–40. We find that Millennials had a real median household income that was 20% higher than that of the previous generation, a slowdown from the growth rate of the Silent Generation (36%) and Baby Boomers (26%), but similar to that of Generation X (16%). The slowdown for younger generations largely resulted from stalled growth in work hours among women. Progress for Millennials younger than 30 has also remained robust, though largely due to greater reliance on their parents. Additionally, lifetime income gains for younger generations far outweigh their higher educational costs.”

From Demography.