What Medicaid and other safety net cuts could mean for US poverty

House Republicans are currently negotiating a budget reconciliation bill that packages extensive tax breaks with deep spending cuts to anti-poverty programs like Medicaid and SNAP. Robert Greenstein explains how the U.S. safety net has cut the poverty rate nearly in half in recent decades and the impacts of the cuts now being considered.  

PITA: You’re listening to The Current, part of the Brookings Podcast Network. I’m your host, Adrianna Pita. This month, congressional Republicans are turning their attention to a large-scale budget reconciliation bill that will include extensive tax cuts, increased spending on defense and immigration priorities, and a large share of spending cuts with the goal of having a complete bill by the end of May.

Here with us today to talk about some of the budget cuts that are being considered, specifically those to Medicaid and SNAP, the Supplemental Nutrition Assistance Program, is Robert Greenstein, the visiting fellow with the Hamilton Project here at Brookings, and also the founder of the Center on Budget and Policy Priorities. Bob, thanks so much for talking to us today.

GREENSTEIN: Thanks for having me.

 PITA: Can you start us off with what are some of these cuts that are being considered and why are they coming up as a particular target?

GREENSTEIN: Well, the president and Republican leaders in Congress want to pass a big reconciliation bill. Reconciliation bills operate under procedures in which they bypass the filibuster in the Senate and only need a simple majority of votes in the Senate, 51 or 50 senators and the vice president. Rather than the 60 votes in the Senate, it takes to pass many other kinds of legislation. So this is an all-Republican bill. It is unlikely there will be a single Democrat in the House or the Senate who votes for it. But what they’re trying to do is to put the biggest parts of their domestic agenda in many ways, all together into one massive bill with enormous tax cuts and having those tax cuts partially – but only partially – offset by very deep spending cuts, far bigger spending cuts than really we’ve ever seen in the last number of decades of the country’s history. The way reconciliation works, the cuts have to be in what we call mandatory or entitlement programs. Cuts in annually appropriated programs operate through a different process. But having said that, if you want to get the overall picture of what does this mean, for example, for low-income families, you would want to look not only at the big budget cuts in Medicaid and SNAP, maybe Pell Grants, a variety of other programs that will go into the reconciliation bill, but then also note the extremely large cuts in a number of social programs that are appropriated programs that President Trump proposed just last Friday, May 2nd, when he released his first budget of his new administration to Capitol Hill.

PITA: Maybe can I ask you to put that into a little bit of context for us? There are a wide variety of programs, as you said, in addition to Medicaid and SNAP, things like the earned income and child tax credits, children’s health insurance, and many more. You recently put out a paper, in fact, that looks at the relatively recent history of the safety net since 1970. What have we seen in terms of changes to, and particularly about the effectiveness of the safety net in recent decades?

GREENSTEIN: Yes, and I think this has direct relevance to the reconciliation debate that’s about to start. So, I looked at the evolution of the U.S. safety net over the last 50 years, especially at what’s happened to social programs that are targeted by income on people with low incomes, modest incomes, as distinguished from universal programs like Social Security and Medicare. And what we find is it’s a little bit of a complex story. Some social programs, some of the targeted social programs, expanded and some others like cash welfare assistance were cut. But overall, the picture was one of substantial expansion and the results are really pretty stunning from a couple of perspectives.

First, if you look at 1970, at that point about one in every four Americans was below the poverty line, and we had a very limited safety net. So that the safety net, social programs that provide income-like supports, the safety that lifted out of poverty only about 5% of people who were in poverty otherwise. If we fast forward then to 2023, the latest year for which we have poverty data, we find very interestingly that the poverty rate before counting government benefits, benefits from social programs and tax credits and the like, that the poverty rate before you count those is pretty similar to what it was back in 1970. Economic growth itself has not changed the poverty much, this tells us. But when you look at the poverty rate after counting government benefits, instead of one in every four Americans being poor, now about one in eight Americans are being poor. We’ve cut poverty virtually in half. Why is that? As I noted, in 1970, social programs, what we call taxes and transfers, lifted from poverty only 5% of those who would otherwise be poor. Today, they lift out of poverty nearly half, 45% of those who would otherwise be poor. The expansions of safety net programs have been a big success in reducing poverty and, particularly important, we now have a growing body of quite impressive research showing that expansions in safety nets, particularly for young children, not only reduce their poverty in the short run, but increase their human capital, their health, their lifetime trajectory, in essence, when they grow up to be adults.

And the other thing, of course, is that a key part of the expansions of the safety net have been in health insurance. 40 years ago, one of every four children in America, about 25%, had no health insurance, today it’s 5%. Among the population as a whole, the share of the population that’s uninsured has been cut in half primarily because of the Affordable Care Act enacted back in 2010. And I think this is particularly relevant because if we cut deeply into the key programs that underlie this progress like Medicaid and SNAP – which we used to call food stamps – then I think it’s inevitable that we will see a reversal of a substantial part of the progress. We’ll see poverty go back up, and we will clearly see, if you cut Medicaid substantially, you’re going to have millions more people who lack health insurance. So that’s part of the issues and the concern with the reconciliation process upon us.

And I should add, though it’s part of the appropriations process, not the reconciliation bill, the budget the president issued on Friday, May 2nd, calls for cuts of something like 40% in low-income housing programs. It would severely cut back rental assistance, particularly for families with children that are low income. And one would then have to think, if that were to become law, about the combined effect of the Medicaid cuts, the SNAP cuts, and the rental assistance cuts, and the conclusion that I come to is that if all of those things are actually enacted, that we will probably see the biggest increase in poverty and in the ranks of the uninsured in the country’s modern history. Now, I’m not saying the poverty level would go all the way back to where it was in 1970; I’m talking about increasing poverty and increasing the ranks of the uninsured from their level today. If these things pass, this will be the biggest increase in poverty and in the ranks of the unsured in the modern history of the United States.

PITA: One thing I thought was curious in this reconciliation process is that the question of these cuts to Medicaid, and maybe to a slightly lesser extent the cuts to SNAP that are being considered, that there is actually a fair amount of debate within Republican ranks about those. Why is that? What do cuts to Medicaid mean for people, and why is that a sticking point with some Republicans?

GREENSTEIN: Let me do a little bit of a political overlay here. You sometimes hear debates in which people, particularly some people on the left, say universal programs are the only ones that do well. If they’re targeted by income, they do poorly. In fact, the programs targeted by income have actually grown more than the universal programs in percentage terms over the last 50 years. And a key part of that is that programs like Medicaid and SNAP are no longer limited just to people below the poverty line. They serve millions of hard-pressed working families, the people, to use the overworked phrase, that work paycheck-to-paycheck. They’re near-poor, they’re modest income. Maybe they’re at 25, 30, 35,000 a year, not at five or 10,000 a year. But with today’s costs, that’s still pretty tough. And so, in my view, it was good policy that, over time, programs like Medicaid and SNAP have been broadened to bring in those families. But it also means that the programs now have a stronger base in terms of who the beneficiaries are. And particularly, given that the Republican Party as remade by Donald Trump tries very hard to appeal to, and did in the 2024 election appeal to, millions of lower income modest income working families. Much of that part of the Trump base vote are people who indeed receive Medicaid and SNAP. I read recently that House Speaker Johnson’s district in Louisiana has over 100,000 Medicaid beneficiaries, just in his district alone. So I think we’re particularly seeing now a number of Republican members of the House who are from swing districts as opposed to deep red districts, becoming increasingly concerned about the political consequences for their reelection if they go deeply into Medicaid and SNAP cuts.

And that has prompted the leadership in the White House to spin these cuts as though nobody will lose benefits. It’s really just waste, fraud, and abuse, et cetera, et cetera. But when you look at the cuts that are emerging, that are being talked about, big chunks of them, hundreds of billions of dollars of what they’re talking about in Medicaid and SNAP cuts involved transferring tens of billions of dollars of costs from the federal government to the states while giving the states broad leeway to cut the programs if they didn’t want to pay that much money. And states are, almost every state is required to balance its budget each year. It’s no politically easier to raise taxes at the state level than at the federal level. This is really a prescription for doing deep cuts in the programs while claiming right now, in order to get the votes to pass them, we’re not cutting benefits, find me a place where I’ve changed the benefit structure. No, you make the states be the bad guys. And I’m expecting that to the degree they do these things, they will also phase them in, so there is not much effect before the 2026 midterm election, and not full effect before 2028 presidential election, but in the years after that the full extent of the cost shift to the states comes in.

To give listeners a sense of what we’re talking about, in President Trump’s first term, he did have a proposal like this at one point, which Congress didn’t pass, to require states to pay a share of SNAP benefits. And when my old organization, the Center on Budget and Policy Priorities, investigated, they found, to take the state of Pennsylvania as an example, my home state, that the share of food stamp costs, SNAP costs, Pennsylvania would have to pay, exceeded everything the state finances, exceeds 100% of the funding the state puts up for all community colleges statewide. So we’re talking about big money here and we’re taking about a federal budget cut battle that may then spawn state budget cut battles for a number of years to come.

PITA: One of the other alternatives that’s being considered is instead of doing cuts, there’s been talk about imposing work requirements as a condition for Medicaid eligibility and other eligibility for programs. These are often touted as a way of targeting waste and fraud. Have there been studies about the effectiveness of work requirements and what they do for programs like this?

GREENSTEIN: Yes, but before getting into the studies, I want to make a point that I hear all too infrequently. People think work requirements are work requirements. There are differing forms of work requirements. When I came to Washington in 1972, there were debates about work requirements in SNAP. We called it food stamps at that point. Then, what was meant by work requirements in the 70s, in the 80s, and the early 90s was If you don’t have a job and you’re able-bodied, you’re not incapacitated, you’re not elderly, and you are not caring for young children, that you were expected to look for a job if you didn’t have one. If you looked and you couldn’t find one, your benefits were not cut off. That is no longer any Republicans on Capitol Hill or the Trump administration mean when they say work requirements. In 1996 Newt Gingrich came up with a new kind of work requirement in SNAP that passed as part of the ’96 welfare law, under which to meet the requirement you had to actually be working at least half time. If you searched for a job but couldn’t find one, it didn’t matter, you’re thrown over the side anyway after three months. In terms of the current SNAP requirement, all of the work requirements that are being looked at for inclusion in the reconciliation bill, in Medicaid and in SNAP, to the best of my knowledge, are the Gingrich-type requirement where you have to prove that you’re working at least half-time or whatever the specified number of hours per week is, and if either you look for work and can’t find it, you have hours here and there and there that you piece together, but you have difficulty documenting it in time, you get thrown over the side anyway.

So, there have been such work requirements experimented with in Medicaid in the last few years in Arkansas and in Georgia. And in both cases, what the research shows is virtually no increase in employment, the stated goal. Huge decreases in participation, large numbers of people thrown over the side, and in most cases the research shows the people who were thrown over side were actually working and meeting the requirement or eligible for an exemption, disability, incapacity, whatever it may be, but could not get through the red tape of documenting in time and proving their compliance, and hence were removed from the program. There’s a very important new paper on Medicaid by Brookings Matt Fiedler, one of the top Medicaid experts in the country, just out in the past few days, and Matt estimates that under the Medicaid work requirements being considered on Capitol Hill, being considered for people who are part of the ACA’s Medicaid expansion – that’s the prime population on which the new Medicaid work requirements that are being discussed, the prime group to which it would be applied – Matt estimates that one of every three of them will lose coverage as a result of the work requirements, and in most cases not because they’re refusing to work.

So, you mentioned in addition to cuts, they’re also considering work requirements. Work requirements are actually, in my view, one of the major cuts they’re considering. If the goal was really promoting work and the well-being of families and children, and not just saving money to finance tax cuts, then you would design work requirements that provided a job search assistance to people that recognize looking for a job, even if you can’t find one. Yet another issue is, will they write exemptions, temporary exemptions into the work requirements in recessions? We’ll see. One idea that’s floating is to say something like, people are exempt from the work requirement in months when the unemployment rate exceeds, say, 10%. 10%, most of our really big recessions have unemployment rates of 7.5% or 8%. And if you have a recession and the unemployment rate for the country, the labor force as a whole is 8 percent, then for low-income unskilled workers, it’s way up in double digits. But under some of the options for how to do this being discussed on the Hill, they would act like effectively it would mean that it didn’t matter if you were in a recession, it would just mean more people got cut off then, which also would undermine the countercyclical nature of some of these programs.

PITA: I want to end on that sort of budget question. The targeted safety net programs are often talked about as if they are significant drivers of budget deficits, of the federal debt, that sort thing. Is that an accurate picture? And if it’s not, what sort of alternatives would you recommend as a way of addressing these fiscal concerns?

GREENSTEIN: Well, I think there are a couple of questions there, so let me unpack them. There was an op-ed by former Senator Phil Gramm and the chairman of the House Budget Committee, Jodey Arrington, who’s one of the key designers of these cuts. There was an op-ed they had several months ago in the Wall Street Journal that claimed that the means-tested or targeted social programs are the main cause and the driving force the increase in deficits and debt of recent decades. If I were Glenn Kessler doing a fact check for the Washington Post, you would give it four whatever the emoji is –

PITA: Pinocchios.

GREENSTEIN: — absolutely totally false. So from 1998 to 2001, the budget was balanced every year. Deficits have come back since then and an increase in debt. You can look at the data on what are the driving factors that caused that and the single biggest factor unquestionably is a big drop in tax revenues as a share of GDP as a result of the 2001 and 2003 tax cuts under George W. Bush and the 2017 Trump tax cuts, all of which either got made permanent or, in the case of the Trump tax cut, extended and now about to be extended again in the reconciliation bill. Yes, there were increases in expenditures as a share of GDP on targeted safety net programs. The overwhelming bulk of those were in programs like Medicaid and CHIP, because A, we insured tens of millions of people who formerly had no health insurance, and B, because health care costs have risen system-wide throughout the U.S. health care system, not just in Medicaid or CHIP.

And by the way, if you’re really interested in saving money and addressing health care costs, there are a number of targets, as other scholars at Brookings have shown, a number targets in Medicare, where without reducing benefits, you could, in Medicare Advantage and things like this, you could save money, but President Trump has ruled out any change in Social Security or Medicare, including, it appears, inefficiencies and what you could really classify as waste in Medicare. And they’re focusing solely on programs like Medicaid in the health area.

There are multiple other routes to moving in a more fiscally responsible direction. I think they start with increases in revenues. And here, while I assign the biggest blame to Republicans for repeated tax cuts in the last quarter century, Democrats have played a role here, too. I find irresponsible the position, although I understand it politically, from a policy standpoint, the notion that you can’t raise a dime in tax on anyone making under $400,000 a year is a very serious problem. And were we able to institute, ideally, a climate tax or even a value-added tax alongside big reforms of the income tax focused primarily on people with higher incomes and closing various corporate loopholes and things of that sort, you could raise a lot of money there. We need to restore long-term solvency to Social Security and Medicare, and there are changes that can be made there on both sides, revenue and benefit structure, as a recent paper, also another Brookings paper by Wendell Primus and others here, and some people at the Urban Institute, have shown.

And I think you also look for things that are less effective and lower priority throughout the budget. But to do a combination of massive, poorly designed, highly regressive tax cuts that explode the deficit, and then to accompany them with budget cuts aimed at the people with the least means to get by, particularly with the potential for a recession to come as a result of the ill-advised tariff policy, this is really a prescription for, in my view, taking us back decades in the kind of country we are and really making it a much more difficult country to live in for tens of millions of our fellow Americans who don’t do as well as we do.

PITA: Well, thank you, Bob, very much for explaining all this to us today, and we’ll have links to your paper, as well as the papers of our colleagues, Matt Fiedler and Wendell Primus and others that you mentioned for our listeners.

GREENSTEIN: Thank you and thanks for inviting me.