Podcast Icon

David Brady on Labor Unions and U.S. Poverty

  • David Brady
  • October 11 2022
  • PC119-2022

David Brady
David Brady

Labor unions receive relatively little attention in U.S. poverty research and our guest for this episode, Professor David Brady, says that this is an unfortunate omission. His research in a paper with Tom VanHeuvelen finds that being in a household with a union member and even being in a state with higher rates of union membership are both correlated with a lower likelihood of being in poverty. Brady says we should pay attention to the role of unions because “political power drives the policies you get, and the policies you get drive the amount of poverty you have in society.”

David Brady is Professor of Public Policy at the University of California, Riverside.

View Transcript

Dave Chancellor [00:00:04] Hello and thanks for joining us for the Poverty Research and Policy podcast from the Institute for Research on Poverty at the University of Wisconsin–Madison. I’m Dave Chancellor. For this episode, I talked with David Brady about labor unions and how they matter for poverty in the United States. Brady is a professor in the School of Public Policy at the University of California, Riverside, where he directed the Blum Initiative on Global and Regional Poverty for six years. He’s also a research professor in the Inequality and Social Policy Department at the WZB Berlin Social Science Center. I really enjoyed our conversation, so let’s turn to that now. Professor Brady, thanks so much for being here. And we are talking today about labor unions and poverty in the United States. But before we get to that, I’m hoping you can tell me just a little bit about yourself and the sort of things that you’re interested in as a scholar. What do you what do you do?

David Brady [00:01:00] Yeah, I study social policy, inequality, health politics. But mostly the most important thing I study to me is the work I do on poverty. And one of the things that’s motivated me to think about poverty is the focus in how we traditionally say poverty has been on contrasting individual poor people against individual non-poor people. We tend to as American poverty researchers, we tend to highlight the individual, demographic and labor market risks that are more common among the poor. So, we emphasize the problematic choices, the cultures, and especially the behaviors of the poor compared to the non-poor. So, scholars routinely ask, why did the poor fail to get married? Why did they not complete their education? Why did they not work? And implicitly, we say the poor are poor because of bad behavior, bad cultures or bad traits. Now, I would argue, and what I’ve been trying to do for the past few decades in poverty research is that this distracts us away from the systemic problem of high poverty in the United States. And the reality is focusing on individuals doesn’t get us to answers of the really big questions and the really important questions about American poverty. And the reality is the US has always had far higher poverty in other rich democracies. This has been stable at least since the 1970s. And poverty is high even among the people with the right behaviors that people that play by the rules. So, if poverty is so pervasive, it’s so endemic in American society at such a high level, focusing on individual poor people isn’t going to get us to the answers we need to understand poverty in America. And so, I have argued in a couple of places that the problem is poverty, not the poor. And instead of looking at the individual behaviors or cultures or traits of poor people, I think we should focus on political explanations of poverty rather than behavioral explanations of the poor. And in that sense, this gets us away from individualism. And concretely, this means I’m always interested in how power, politics, institutions and policies drive those systemically high levels of poverty in America.

Dave Chancellor [00:03:19] In your 2021 paper, Labor Unions and American Poverty, that you were with Tom VanHeuvelen, you start off by kind of pointing out that U.S. poverty research does not devote a lot of attention to labor unions. This rings true to me from my work here at the Institute for Research on Poverty. We don’t do a lot that specifically focused on labor unions. And I actually am searching back through our archives, and I found a 1981 working paper that also said that US poverty research basically does not do a lot on labor unions.

David Brady [00:03:50] Who wrote that working paper and can I get my hands on it?

Dave Chancellor [00:03:52] That was Mike Podgerski. Okay. And he cites a couple of other scholars, 1976, in the 1960, who also more or less imply that we don’t do a lot on labor unions. I mean, certainly there are other things that have been written on it. But I think I think you’re right that there is not a lot covering that kind of. My question is, why do you think that poverty research does not really cover unions a lot?

David Brady [00:04:16] That’s a great question. I just wanted to reaffirm what you said, Dave, and say that’s really compelling to me that the only thing you could find was a working paper from 1981. So over 40 years ago, one working paper, the IRP on this.

Dave Chancellor [00:04:30] And there are other things, but it was like number three in the search results. Right.

David Brady [00:04:35] But it is true. I mean, like, look at all of your favorite books on poverty. And I do I can I voraciously consume the poverty literature. I edited a big Oxford handbook on the social science of poverty. I read all the IRP volumes on poverty. There’s just no attention to unionization. Go read your handfuls of annual review essays on poverty literature, especially in sociology and including in recent years. And they completely neglect delivery. And so that neglect is real. And I would really substantiate that. And we talk about that at the beginning of the article and also say it’s bad because unions are really critical and they’re a very powerful influence on the amount of poverty. So, it’s both are true that we have an omission, and it’s also true that this omission is unfortunate. So, I have a couple of theories about this. One theory I have is that, look, American poverty research, like anything else, reflects American ideology. And Americans are the most individualistic people in the world. There’s a lot of psychology and anthropology research that shows this. So, it’s not surprising that our scholarship reflects our culture, our belief system. On some level, we mimic some of the values. Of beliefs and orientations of American culture. So, Americans are prevailing with their individualism. And therefore, poverty research reflects that individualism. And we have a prevailing individualist explanation. So that’s one reason. And that applies to a lot of things, why we don’t talk enough about policy, we talk too much about behavior or culture or what have you. Okay. A second thing I think that’s really important is that American poverty research implicitly draws an artificial line between the poor and workers. We think of the poor is different from workers, and this is just flat wrong. The reality is the overwhelming majority of poor people in America are working poor people. So, for every homeless person, you see as much of a tragedy that that is. And here in Southern California, we have a very large homeless population. I always tell my students that they should imagine there are 30 to 50 more Latinas working at Wal-Mart for every homeless person you see. And so, the sheer vast size of work in poverty dwarfs all other categories of poverty. But Americans, I don’t think we think of it this way. I don’t think American poverty research think of it this way. They think work is the answer, which to a certain extent, of course, is better to have a job. But still, most poor people are working, or they live in households where somebody is working. And so, we have to sort of break down this myth that there’s a distinction between workers and the poor. The fact is the poor are workers and many workers are poor. So that’s the second thing I think is in play. And then I would say that I think that implicitly there’s something important about that. In the 1970s, as labor unions became more diverse, they became more racially and ethnically heterogeneous. They had a larger share of black workers, larger share of Latino workers, larger women. At that time, we became just interested in that. In the social sciences, so many scholars of inequality in economics, for example, will downplay or diminish the role of labor unions, when, in fact, I think there’s always been very compelling research that unions remain one of the most critical influences on earnings inequality, on socioeconomic inequality in general. And it’s sort of sad that as unions look less white and less male, you know, we’ve lost interest in those institutions. I think there’s something really problematic there.

Dave Chancellor [00:07:56] In the context of thinking about poverty and reducing poverty, what is the case for unions? I mean, you’ve spoken about this a little bit, but I mean, spell it out more clearly for us.

David Brady [00:08:06] Yeah. Yeah. I mean, we often forget unions have been so effective for so long, we almost forget how effective they are. They feel like sort of like your grandparents’ egalitarian institution. And in fact, they still remain extremely vital, extremely relevant. I’ll just say there’s overwhelming research that shows that unions are good for workers. I just don’t think a serious person could debate that fact. this is the case for a variety of outcomes. We could look at pay earnings, job security, autonomy, fringe benefits, dignity, safety and health on all of these outcomes. Unions protect workers. They treat workers better. This is also the case for a wide variety of workers. It’s the case that men benefit from unions. But it’s also the case that women benefit from unions. Whites benefit from unions. But there’s strong evidence that ethnic racial minorities may even benefit more from unions, immigrants benefit from unions, young workers, older workers, what have you. And what’s really important is that it’s not just that individual workers benefit if they’re unionized. If you are a not unionized worker and you go to a firm that is now unionized, you will benefit. You’ll get better pay, you have better compensation, what have you. But there’s also evidence that firms that are unionized are better than firms that are not unionized. You’d rather work at a firm. So even if it’s a question of whether your own firm unionized, they’re better. There’s also evidence that states that have higher unionization, such as Minnesota or Hawaii or New York in some ways are better places to work than states that have low unionization like Mississippi or South Carolina. There’s evidence that countries that have higher levels of unionization are better places for workers than places that have lower levels of U.S. And the real reason is, is just generally unionization has this powerful diffuse effect on inequality. It reduces that equality. And especially important is that it spills over even to people that are members of. Unions are a political force for what we call their what we call a power resource that pushes the state to institute social policies that take care of a broad swath of the population. So, you know, because we have strong unions at various points in history or certain countries have it, that leads to better health care systems, that leads to better pension systems, it leads to effect to policies that embrace equality for a broader population and not just the individual workers.

Dave Chancellor [00:10:32] I’m curious about what the skeptical take here is. And I mean, even that 1981 paper I referred to in that author said that he didn’t see a lot of movement in terms of poverty numbers because of unionization, because it’s his perspective that a lot of people that were in poverty just didn’t have sort of access to some of the benefits that would be gained from unions. But what do you find? What do you see and how do you address those critiques?

David Brady [00:10:58] Yeah, it’s a really good question. And I think it is important to always entertain the critiques and to anticipate and think through the good reasons that unions might not be good for the poor, specifically the poor. You know, one argument is they’re too small and they’re too irrelevant. And there’s something to this. I mean, the United States, as Jacob Rosenthal just said very nicely, it’s like, you know, the levels of unionization are the lowest we’ve seen in more than a century, and it doesn’t look good going forward. So, there’s this real challenge for unions that they’re too small and irrelevant. We have places like North Carolina that have something like 1% of their private sector workers that are unionized. And it’s just really hard to believe unions can really move the needle when they’re so small and so irrelevant. So that’s a fair question. Second, kids are going to be that it’s just selection effects look. So, if you’re a savvy, skilled worker that knows how to improve your situation, you’re going to go find that unionized firm, go find that unionized workplace and go find a way to be nice. And so maybe it’s just the savvy workers that are able to exploit the system that’s already in place. And it’s not the unions per se, but it’s just the savvy workers that hunt down. So, it could be selection effects and there even could be adverse spillover effects. It could be that unions crowd out the nonunion workers because union workers get to be more expensive is to create insider outsider demand mix, where the union members on the inside are protected, but those in the outside are treated poorly. And it can also create some sort of rigidity, just like minimum wages. So, when people say, oh, if you raise the minimum wage, firms are going to hire fewer workers. The same sort of reasoning could apply to unions. Now, as we say in this paper, we don’t find these rates, but I think there are plausible arguments that one should entertain for why they might not be beneficial to workers.

Dave Chancellor [00:12:48] So I want to talk about this paper and in it you use the panel study of income dynamics or PSID, and you look at both household union membership and state level union density. And you write that to the best of your knowledge, this is the first time that a study like this has used individual level panel data. And so just to make sure that we’re all on the same page, what is individual panel data and how is it helpful in this case? What’s what does it let you do?

David Brady [00:13:16] Okay. So, the individual panel data just means you take a sample of people and you follow those exact same people over time. Okay. The PSID is sort of the granddaddy of panel data sets has been around since about 1969, 1970. And there’s people we can follow over a 30, 40 year period. We follow their children over decades as well. So, you can observe this same exact person as the world changes around them. Okay. And that gives you a lot of leverage in order to sort out the effect of unionization from those sorts of selection effects that one might worry about. So, for instance, we can make sure that it’s not just the individual person who maybe has a propensity to get a better paid job, but instead see as that person, as their job changes over time or the unionization in their firm or their household changes over time, or for that matter, the unionization in their state changes over time. You can see what happens to their probability of being poor. We can exploit that within person variation, which is a more credible argument that it’s really the effect of unionization. And again, as you know in the paper, we also look within states, and we also look within time. We’re trying to use these what we call three way fixed effects models to really, really pinpoint and precisely tease out the effect of unions separate from the people, the places and the times.

Dave Chancellor [00:14:41] So we’re talking about the effect of unions on poverty rates. And so, I just wanted to sort of establish like how you are measuring poverty in this paper, because I think that’s important.

David Brady [00:14:51] I would argue that the best measure is a relative poverty measure that’s based on what we can call post fisc equalized income. And by post fisc I mean that you factor in the taxes that people pay, the tax credits they receive, and the transfers they receive by equalize. I mean, you got to adjust for the size of a household. So, if you have four kids, it needs to be your household income needs to be higher than if you have no kids. And we calibrate according to household size, and I would argue this is the best way to measure poverty. This reflects what I would argue is an emerging international consensus on poverty measurement, a consensus that in many ways was led by your colleague Tim Smeeding, amongst other people. And I grew up as a poverty scholar reading people like Tim Smeeding and thoroughly convinced that this this international approach to poverty measurement is the best way to measure poverty. And there’s a real problem that American poverty research tends to neglect that international consensus. Now, relatedly, I’d say one of the common ways poverty is measured in the US is to use the official poverty measure. But as I keep saying in paper after paper, the opium poverty measure is absurd. It doesn’t make sense, it’s not defensible and it’s not a scientific measure. The thresholds are too low, the household size adjustments are incoherent. And the whole thing, the whole enterprise, the official poverty measure was never based on rigorous science. There was never any real scientific basis for it. It’s become unreliable over time. So, we can’t compare the official poverty measure to the past because the way we measure income and the official poverty measure makes no sense. And I’ll give you a concrete example. The official poverty measure counts the Temporary Assistance to Needy Families. The Temporary Assistance to Needy Families would affect the sorts of people we’re studying in this study. Right. Because we’re looking at the working poor and we’re looking at all working age people. Well, the problem is Temporary Assistance for Needy Families has nosedived and has become this trivial government program. We’ve spent maybe seven, $8 billion on poor people in America in actual cash assistance. So, it’s a very, very small program. But at the same time, the official measure, the official poverty measure ignores tax credits. So, we have the earned income tax credit, the child tax credit, the Biden tax credit. All of these have grown substantially over time. I mean, the EITC is probably over $100 billion, and the CTC is probably over $100 billion, and it’s ignored by the official measure. So TANF counts, but that’s declined. They become irrelevant. The EITC and CTC are ignored, but those have become dramatically more important. Similarly, the official poverty measure ignores near cash transfers, and the reality is we use the snap or what we knew colloquially as the food stamp program, as one of our principal family assistance policies these days. So, the policies have changed, but the official poverty measure hasn’t changed to reflect that. And I’d finally add that most of what we did during the pandemic and most of what President Biden has done to fight poverty would be ignored by the official measures. The official measure is just terrible, and it shouldn’t be used at all. Now, on the. Questions. So, we have we have three questions we want to know, does being in a unionized household, you know, we don’t care if it’s you that’s unionized, the regional worker, we care somebody, your household. So, it could be your spouse, for instance, that’s unionized. And of course, kids are working, but we care if their parents are in there. So that’s one question. Second is, we want to know, is that does the what’s the effect of unions in the state around. Working in Hawaii, New York versus Mississippi, South Carolina, those sorts of contrasts. And then we want to know the combinations of those two. If you’re a unionized household member and you’re in a high unionized state, or conversely, you’re a non-unionized household, but you’re in a high unionized state and all that all the potential of those four possibilities. So that’s really what we want to know.

Dave Chancellor [00:18:46] So in your paper, you ask three main research questions, and I’m wondering if you can go through those for us to sort of give us a sense of what you were looking for and what you learned.

David Brady [00:18:56] Okay. On the first question of do does household union membership influence working and working age poverty? The answer is definitely yes. So being in a household that’s unionized reduces your probability of being poor by about five percentage points. So, it’s not small. It’s a big effect because in working age households, you’re looking at poverty rates of like 10 to 15%, depending on how you look at it. It’s substantially moving the needle to have a unionized household. Now we find that’s the case for both workers. What we’re interested in working poverty, but we’re also just generally interested in working age poverty, which includes the unemployed, the disabled and so forth. And so, we find effects on both these groups. So being anti household really does move the needle. On the second question of does state level unionization reduce poverty? We find the answer is also yes. Okay. In addition to individual household membership being a member of a union, if the state is unionized at a higher level, this reduces poverty for everybody as well. Okay. So that’s really powerful. And again, we find that higher unionization in the state reduces poverty for union households, nonunion households, working non-working households, what have you. And then the third issue of like do these interact or do they offset each other? Well, we find that they augment each other. So being both in a union household and in a high state, high unionization state, that really reduces your probability of being poor. Poverty is quite on probable if you’re in a union household in the state. So, in some ways, in the places in the United States and for the workers in the United States that look like our European social democracies, their poverty rates are really low. So those institutions work here in the United States, just like the work elsewhere. The problem is we just don’t have enough unionization, enough in U.S. households. Okay. Moreover, we find that being in a non-unionized state and a nonunion household, your probability of being poor is very, very high. So, there’s this interaction, this chemistry that happens where if you’re in a union household and in a union state, you get really low poverty. And if you lack both of those, you’re really highly probable to be power poor. And what I think is really particularly compelling is that high unionized states benefit the nonunion households. So, if you’re sort of swimming in a sea where there’s just more egalitarianism and lower levels of inequality, you’re going to benefit even if you’re not a union member.

Dave Chancellor [00:21:31] Earlier you mentioned that unions are relatively weak in the US right now. I mean unionization rates are pretty low in most places. You know, in some ways, I mean, I think that sort of undergirds some of your results because like it’s like even if you’ve seen this despite those conditions. But how do you how do you think about that?

David Brady [00:21:49] Yeah. I mean, in some ways that we find such strong results for unions, it explains some might say, well, how can it be that poverty is fairly stable, but unions are declined if unions are so important? Would you expect poverty increase? Well, the reality if you sort of, you know, lift up the hood and look underneath the hood, what’s going on is that unions declining has basically prevented poverty from falling further. America has actually made some real progress in fighting poverty. We have a lot less low educated households in most households. The vast majority of households are headed by people that have high school degrees. That was not the case 30, 40 years ago. So, we’ve had a rise of educational attainment and this curbs poverty. We have fewer households today that are headed by young people. People wait longer to become parents and to head households than they did in the past. And that certainly has reduced poverty as well. And like I said, most Americans work. I mean, we’ve got to be the hardest working country in the rich democracies. We’re obsessed with works. I mean, just ask any of your colleagues were terrible at taking vacations. Right. And Americans work. We’re a working culture. We have a crazy work ethic. And our people work, our poor work. And all of those things would have produced lower levels of poverty. But unions weakening, along with the weakening of certain social policies, has maintained that level of poverty. So sometimes when you see stability, there’s countervailing forces that are working against each other and you end up with stability. And I think that’s the case, the poverty. But all that said, we need an even better understanding of why unions are struggling. So, I often say, you know, if unions are so great, why are they declining everywhere and they’re declining in the United States, as we all know. They’re declining in Western Europe, even in countries that historically had very, very high levels of unions. And unions are struggling in most places. We have some wins in America. There are ways in which unions have been successful recently, but they still face a real uphill battle and there’s still a lot of loss going on. So, we need to understand why unions are struggling. If unions are so good, why do why are so many workers still reluctant to join unions and that those are real things. We need to understand why there’s a lack of solidarity among workers in general in America. But let’s not lose sight of the real big picture issues, and that’s that employers have become extremely effective at resisting and undermining unionization efforts. And to be frank, it’s a fact. It’s just a simple fact that employers violate the law routinely to undermine U.S. and the government has not done a good job preserving workers’ rights to unionize. So, if workers have a legal right to unionize that law, that rights should be protected. But the courts and the government have been incredibly weak in protecting that right, and employers have exploited that fact in order to routinely violate the stipulated statutory laws and make it very, very hard to have a unionization effort. What you see all the time is you see that workers started unionization drive and it barely loses. It doesn’t get unionized. Right. And you’re like, oh, that happened. Don’t workers want unions? But then, sure enough, there’s all this evidence that comes out that the employers are breaking the law, or the employers were skirting the edges of the law to do everything possible to undermine the decision. And you can look at examples like when Starbucks unionized or Amazon unionized, this obvious evidence that the employer is violating, restricting the edge of the law to undermine utilization. And that’s the real story, is we don’t reserve the right to unionize in America. It is true that Americans have been more reluctant to unionize than other countries, but that’s because partly we haven’t had the momentum of strong unionization and employers have all this political and economic power to discourage unionization.

Dave Chancellor [00:25:40] So a couple of years ago you wrote an article for the annual Review of Sociology about theories of the causes of poverty. And at the end, you advise readers of research to sort of envision the authors of what they’re reading as having a magic wand. And so, if they could only change a few things, what would they do to reduce poverty? And in the context of unionization about what we’ve been talking about today, you’ve got a magic wand. What are you going to do?

David Brady [00:26:09] Yeah. Thank you. I love this question, Dave, and I need to tell the story of where that comes from. I wrote this annual review of sociology essay on the theories of the causes of poverty. I presented it at the IRP at Wisconsin and Maria Cancian, former IRP director and very influential party scholar, an icon in the field. She and I were talking, and she had this great insight. She’s like, you know, really what you’re asking people to do is to say, you know, if the author had a magic wand. So, this is Maria’s idea. And I thought it was such a genius way to articulate what I was trying to say. But struggling to say was that we should diagnose the books, we read the articles, we read it, and just try to get a sense of what is the theory here? What do they think really moves poverty? And that does two things. One is that it strips away all the labeling and the banners we wave about what kind of poverty is all we are. But secondly, it makes us ask questions about like, if you think that’s most important, if you had a magic wand and could change one thing, you know, do you think that’s what’s most important? That’s what you should be studying. That’s where you think the action is. That’s where the power is to move the needle on poverty. And I thought that was such a clarifying and brilliant insight by Maria Cancian. I want to give her full credit for that. Okay. So, asking me what’s a magic wand? Of course, I would say strong U.S. with help, and that would help because it leads to not only better workplaces and better households in terms of fighting poverty, but it also leads to social policies. So perhaps the biggest influence unions have is their what we call a power resource. They push the political coalitions to get more generous social policies and more egalitarianism in society. And they do it indirectly through political parties. They do it in coalitions with political parties. But the ultimate goal is they’re a political actor. They’re a collective political actor that pushes towards egalitarianism. And so, there’s really few power resources or collective political actors that have been as successful as labor unions in the rich democracies to accommodate poverty. We can also say, look at any country that’s ever accomplished low levels of poverty. And there’s been some for at least some period of time. They’ve always had strong labor movements and they’ve always had big governments. No country in the history of the world has accomplished low child poverty without an above average government size. It just can’t be done. So that would be my in my initial emphasis. Beyond that, concretely, I would push to lots of things that we know work. What we failed to do. And so, I think that we should have much higher taxes on the rich and we should use those taxes to pay for more generous social policies. I think we should have a universal child benefit. The child Biden tax credit worked very well. I would have it be bigger. I would make it fully refundable. I might use Social Security system instead of the tax system. But these are tweaking the fundamental sort of policy that did work. And the reason we don’t have the policy has nothing to do with it being effective or not. It’s 100% due to politics, which proves the point that power, resources and collective political actors matter. If we had a strong labor movement, there be a lot more political pressure on Joe Manchin and Kirsten Sinema to switch over and support the extension of child benefit. It’s that simple. Political power drives the policies you get and the policies you get drive the amount of poverty you have in society.

Dave Chancellor [00:29:29] Professor Brady, thanks so much.

David Brady [00:29:30] Thank you so much, Dave.

Dave Chancellor [00:29:32] Thanks again to Professor David Brady for talking about this work. We’ll have a link to that paper in the show notes. And you can also find Dr. Brady on Twitter at @DaveBrady72. The production of this podcast was supported in part by funding from the US Department of Health and Human Services Office of the Assistant Secretary for Planning and Evaluation. But its contents don’t necessarily represent the opinions or policies of that office. Any other agency, the Federal Government or the Institute for Research on Poverty. Music for the Episodes by Martin de Boer. Thanks for listening.


Economic Support, Economic Support General, Employment, Labor Market, Low-Wage Work