Populists in both parties tell the story of a stagnant American economy where little progress has been made in reducing poverty. In today’s America, they claim, it’s almost impossible to raise a family on a single income. But is that really true? In this episode of Political Economy, my AEI colleague Scott Winship returns to explain what’s really going on.

Scott is a senior fellow and Director of Poverty Studies here at AEI. He’s also author of the new report, “Bringing Home the Bacon: Have Trends in Men’s Pay Weakened the Traditional Family?“

What follows is a lightly edited transcript of our conversation. You can download the episode here, and don’t forget to subscribe to my podcast on iTunes or Stitcher. Tell your friends, leave a review.

Pethokoukis: I would think that if there’s one thing that we could very easily measure and define, it would be poverty in America. We can just measure the number of poor people and figure out a poverty rate. First of all, is it that easy? Because sometimes I hear about multiple, different kinds of percentages. When we’re talking about poverty America, what are we really talking about? And how do you think it’s best for us to define it?

Winship: You’d think that would be super easy, but we could take your listeners down a road of tedium the likes of which they maybe have never experienced on your show before. . .

Which we are not going to do. But we’ll take maybe a few steps down that road. Maybe a half block.

Good clarification. In the United States, we do have an official poverty measure. Lots of times you hear people ask, what’s the official definition of middle class? We don’t have anything like an official definition of a middle class, but we do have an official poverty measure. And that has been around since the 1960s. It essentially takes the cash income that people have, including any safety net benefits that take the form of cash, and it compares it against the poverty line, which differs depending on how big your family is and a few other things. If you’re below the line, you’re poor. If you’re above the line, you’re not poor. Easy peasy. The problem with that is that much of our anti-poverty policy in the United States takes the form of benefits that are not cash: think of food stamps, think of housing benefits, think of school lunches. The other problem is that the official poverty measure is a pre-tax measure. If tax rates go down and people are left with more take-home pay, that doesn’t show up as lower poverty at all. If we create a big program like the earned income tax credit or the child tax credit that puts money in people’s pockets even if they don’t owe taxes, that doesn’t get counted as reducing poverty at all. Most of the ways over the last 25 years that we’ve actually made life easier for lower-income people don’t show up in the official poverty measures. . .

That’s the measure we probably hear the most about when it comes out every year. If you’re on one side of the line, you’re counted as poor, and other side. . . How is that line determined?

It’s fairly arbitrary. If you go back in time, in the mid-‘60s, there was an economist in the US Department of Agriculture who was working on food budgets, essentially how much money people had to spend for a nutritionally minimally adequate diet. They had these estimates. There was some research suggesting that about a third of people’s budgets was spent on food. So you take these food budgets, you multiply them by three, and that was essentially the original poverty line. Pretty arbitrary. Over time, it gets adjusted every year for the cost of living—although using an inflation measure that overstates the rise in the cost of living so we basically make it more difficult every year to get over this line, because it’s actually rising faster than prices are. It’s basically an arbitrary line—that’s a real point of contention between some people that are really anal retentive about this stuff and prefer other measures versus a bunch of us who say, “Draw a line. Keep it consistent over time. Then look at how many people are above or below that line.”

If I was someone who was just below that line in 1970, and if I’m someone who’s just below that line today, my lifestyle is pretty different, right?

In theory it shouldn’t be. The theory behind the poverty line is that if you are a dollar below the poverty line today versus a dollar below the poverty line in 1970, that you should really be at the same standard of living. That gets super complicated because that gets into, how much would people in 1970 pay to have an iPhone back then? How do you value improved environmental quality today versus in 1970? But the theory is that line should be the same over time, except for the increase in the cost of living. It probably, as I said, understates the extent to which things have improved over time.

If I’m looking at the cash-only poverty share versus one that takes into account all that we do but isn’t a check, how do those two differ? And what do they both say about how we’re doing fighting and reducing poverty?

They show up as different amounts of poverty. And I should have those numbers right on the tip of my tongue, but I don’t. I think it’s something like the difference between, I don’t know, 12 percent poverty and 9 percent poverty, roughly—with the official measure being higher. The 9 percent is using something called the Supplemental Poverty Measure that does a much better job taking into account all these other things. There are problems with that, too, that I won’t go into.

And are those rates better than they were in 1965?

Yes. That’s where they really make a big difference, which one you use. If you use the official poverty measure, then poverty is actually at an all-time low, or it was last year. But the progress over time hasn’t been all that strong. And in particular, if you look at kids, for instance, you get this finding that over 40 or 50 years poverty rates didn’t change that much. When you hear people say we fought a war on poverty and poverty won, they’re using the official poverty measure, which doesn’t show much progress over time. It has all these problems with it. This thing called the Supplemental Poverty Measure is better. It only goes back officially about 15 years, so you can’t trace it back to the 1960s. There are researchers who have tried to extend it back to the 1960s, and you do find over longer periods of time much more impressive declines in poverty. A lot of us don’t like the Supplemental Poverty Measure either. We like that it includes a lot of these benefits that aren’t cash and don’t get count in the official poverty measure, but it also has this aspect of being a relative measure. If the middle class is getting richer and the poor are getting richer, but the poor are not getting richer as fast as the middle class, then the Supplemental Poverty Measure can actually show that poverty has increased even though everybody’s better off. It conflates poverty trends within inequality trends.

If you use a measure that includes all the good stuff that’s in the Supplemental Poverty Measure but doesn’t have this feature of being complicated by inequality, then you get really huge declines in poverty over time. And I think that’s the most accurate reflection of what’s happened. Rich Burkhauser, our colleague at AEI, has a paper with several people, including Kevin Corinth, our former colleague at AEI, where they basically show that if you start with Lyndon Johnson’s stat of one in five Americans being poor in 1963, and you hold everything constant, you measure income as comprehensively as you can, you look at that over time, what would that poverty rate be today? And it turns out it’s like 3 percent. It used to be 20 percent, now it’s down to 3 percent. Similarly, you find big declines in child poverty over time if you do it this way. But that is very different than if you use the official poverty measures and find very little progress in poverty reduction over time, except among the elderly.

It seems to me that if I was a professional anti-poverty activist, I could come up with a number that had some real empirical basis and research behind it, and I could make whatever case I wanted, if I wasn’t really pressed or if I was giving a quick quote to a journalist. I think the result of that is people probably think that this country either has made no progress on poverty or a huge share of this country is poor. And I would also guess that people have no idea what we spend per year on reducing poverty. Is that your sense of it at all?

Yeah, that’s all exactly right. It’s sort of a conspiracy from both sides a lot of times to understate how much progress we’ve made. Obviously, if you are super progressive, you don’t want to tell the world that we’ve reduced poverty to 3 percent of the population, because that’s going to undermine all your attempts to pass a child allowance, universal basic income, what have you. So lots of folks on the left, there’s a tendency for them to want to shy away from this fact. On the right, of course, it’s what I mentioned earlier: You want to argue that we fought a war on poverty and poverty won. Why did we spend all these trillions of dollars for no reason? But, when you measure things the right way, we did reduce poverty, and the safety net expansion was part of it. It’s not the main part of the story or the whole story. But, as you say, you come away with this impression that we’ve made no progress over time. You come away with this impression that poverty is just rampant out there. I was on a panel recently where somebody was saying one-third of the population was “near poor.” (I’m making air quotes.) I’m not sure how he got that number. If you look at the official poverty line, then I think only 26 or 27 percent of the population is under two times the poverty line. If you count being “near poor” as having twice as much income as poor people, that still doesn’t get you up to a third. But there are these claims that are floating around out there.

You said that all these programs is not the whole story. So what’s the rest of the story?

For the trend, I would say part of the story is we’ve expanded the safety net dramatically. I have a paper with Angela Rachidi and Matt Weidinger, where we show that even since the mid-‘90s we’ve doubled federal anti-poverty spending—even if you don’t count Medicaid, which is the biggest program. So there have been big increases over time. A lot of it has gone towards working families. And the second part of the story is that the safety net itself has, over the last 25 years, incentivized work much better than it used to. Before the mid-1990s, it was essentially: give people cash, give people non-cash benefits, but without really much in the way of expectations. You could be on these programs for a long time. You didn’t have to work, didn’t have to look for work. In the mid-‘90s with welfare reform, we made it a lot tougher to stay on the biggest program that provided cash benefits without working. We imposed lifetime limits. We imposed work requirements. The result was this unprecedented increase in work among sick parents, and poverty fell.

It was this great set of outcomes where we had fewer people dependent on government programs, more people working, and less child poverty. There were also these tax credit programs, the earned income tax credit, child tax credit, which were sort of the carrots that paired with the sticks. They kind of helped people that were working and low income. And then the third part of the story is, as you say, economic growth: just the strength of the American labor market and economy over time at creating jobs, providing work for people. In the mid-‘90s, everybody said, “We can’t expect all these folks to work. There just aren’t jobs out there for them.” Turns out that was completely wrong. I think the last 25 years we’ve hit on a great formula for reducing poverty. Unfortunately, there are a lot of folks on the left who disagree with that. We are still fighting battles around things like the child tax credit and the general shape of anti-poverty policy.

Your description of changes in welfare policy in the ‘90s, is that generally accepted by people on the left and right? Is it mostly accepted, but you have people who say, “No, you’re figuring it out all wrong”? How controversial does that now, 25 or 30 years later, remain?

It remains, unfortunately, pretty controversial, even though I don’t think it should. There was a report that came out in September by the group Child Trends that had two headline findings. The first was child poverty was at an all-time low, which was actually something that a few of us had been reporting for a long time. I had a report out in 2016 that showed that by 2014 child poverty was at an all-time low. And others have written about this too. But a center-left group coming out with this report makes it safe to say. And the way they made it safe to say was their second big result, which was that the reason that child poverty was at an all-time low is mostly due to expansions of the safety net.And they had some analyses that were deeply flawed. Their own numbers, if you dug into it, indicated that what they were saying wasn’t actually true. I showed that if you just look at the incomes of single mothers with the lowest formal education levels and how that changed over time, then almost the entire story was earnings and other private income rather than safety net benefits.

It’s still a very complicated story. A lot of folks on the left think welfare reform in particular was a huge failure. There was a book that came out in, I believe 2017, called $2 a Day by Kathy Edin and Luke Shaefer. They argued that there were a bunch of people in the United States living on $2 a day, that that number had gone up a lot since the mid-‘90s, and it was welfare reform that was the cause of it. And each of those claims have been totally demolished since the book came out. But it was very influential on the left. People became convinced that tying conditions to safety net benefits, like work expectations, was a horrible thing. And that if you want to reduce poverty, the way to do it is you just give people money. What could go wrong?

How much of that is based on looking at economic research and people’s behavior and drawing conclusions, and how much of that is based on an ideological belief that, if you’re a human being, you should be able to live a decent life because you are a human, not because you are a working human? I think this is part of the argument behind universal basic income, not necessarily that robots are going to take all the jobs, but that a certain level of income is a human right. And to put any attachments on it is sort of denying who we are as human beings.

I think there’s an element to both. I think at its root, it’s the second thing that you’re talking about. I think there is a widespread belief on the left that there are just things that are rights: You have a right to a universal basic income, you have a right to have as many children as you desire and have taxpayers and the federal government subsidize the cost of it. It would be a much more honest debate if that’s how the debate proceeded. But I do think there is a real belief that welfare reform hurt single parents and their kids. It’s not consistent with the evidence.

There are endless debates about it, because we didn’t have a randomized controlled trial when we rolled out national welfare reform so we can’t say the people that experienced it had these outcomes and the people that didn’t had these other outcomes. Just the pattern of trends is pretty unambiguous to my reading. You just look at things like the employment rates of the single mothers with the lowest education levels, and they jumped like 18 percentage points over the course of four years around the time of welfare reform. You look at poverty trends and they went down quickly. Obviously the second half of the 1990s, was a very strong economy. A lot of folks on the left will say that’s why poverty went down. But poverty never went back up to the levels of before welfare reform—even when we experienced things like the Great Recession. Poverty rose less during the Great Recession than it had in any of the previous recessions before that. It’s because we have a safety net that acts as a safety net amid hard times. But we have a program that much better supports people during good times and encourages them to take advantage of increased economic growth.

You have a new paper investigating the claim from populists on the left and right that raising a family on a single income has become harder over the decades. And those declining economic prospects for men, these populists claim, are making men less “marriageable” and contributing to rising single parenthood. How do you evaluate the marriageability of men, and what do you find in the paper?

As you said, I had noticed that a lot of the claims on the populist right look a lot like the populist claims on the left: that the economy is failing men, and that’s leading to these changes in family life. I decided to take a look at whether there was any truth there. What I decided is that what we could do is set a threshold for marriageability. It’s kind of like a poverty line: If there’s some level above which men are marriageable but below the line they’re not marriageable, let’s at least draw a line and see how the percent of men that are marriageable changes over time.

I draw a couple lines. I look at sole-earning married dads in 1979, for reasons we’ll talk about in the paper. I look at what the 25th percentile of those guys was for earnings. (Twenty-fifth percentile meaning 75 percent of sole-earner married dads made more than that, 25 percent made less than that.) That was my first threshold. That ends up being almost $34,000 in today’s dollars. And then the second threshold was the median: What does the middle sole-earning married father make? That’s $47,500 in today’s dollars. You hold that line constant; you look at how that changes over time among all young men.

And the answer is that by the lower bar, marriageability was 63 percent in 1969. That drops by just five percentage points to 58 percent in 2019. The drop is pretty much all in the 1970s. The 2019 level is still higher than in the early to mid-1960s when people nostalgically think about Ozzie and Harriet. If you use the higher bar for marriageability, then the marriageability rate actually increases from 30 percent to 40 percent, 40 percent being an all-time high. By that metric, men are more marriageable than they’ve ever been. I looked at less educated men in particular to see what happens with their marriageability. Basically, you get the same story there.

I looked at African-American men. If you go back to William Julius Wilson in the 1980s, he had this story about rising single parenthood in the African-American community being about men being less marriageable. It turns out that among young black men, marriageability went from about 8 percent in 1969 to 25 percent today. These stories just don’t hold up. The magnitude of the changes we’re talking about in terms of single parenthood, in terms of how many young men are sole breadwinners, are just huge changes against changes in marriageability that just don’t move that much at all, or they move in completely the wrong direction.

Built into the premise is this notion that raising a family on one income has become, I think most people would say, not just harder or a lot harder, but virtually possible. What is your perspective on that?

That’s exactly what I’m trying to answer with these numbers, and it suggests that it’s just not true. The reason for the gap is that essentially people don’t acknowledge the trade-offs that we’ve made over time. If a lot of people who would prefer to raise a family on one income were willing to live at 1970 living standards, that’s no harder than it’s ever been. The issue is that we don’t want to live at 1970 living standards. That’s why more wives work. That’s why marriage has been pushed back into the late 20s and 30s. That’s why we have fewer kids today than we did in the past. Because what we want has changed. We want more stuff. Men and women both have voted with their wallets for more stuff. You can’t have more stuff and try to raise a family on one income. The problem is we just haven’t acknowledged this trade-off that we’ve all chosen. Well, not all of us.

Scott, thanks for coming back on the podcast. I appreciate it.

My pleasure. Thanks for having me, Jim.

The post Did We Lose the War on Poverty? My Long-Read Q&A with Scott Winship appeared first on American Enterprise Institute – AEI.