There are plenty of people in Washington, DC, advocating on behalf of corporate industries, or well-endowed universities, or the idiosyncratic interests of wealthy donors. A much smaller number spent their days monomaniacally focused on addressing the needs of the poor.
But among the latter is Robert Greenstein, president and founder of the Center on Budget and Policy Priorities. Since 1981, Greenstein and his team have been at the center of countless debates over food stamps, welfare reform, housing assistance, the earned income tax credit, Medicaid, and more. Most recently, they helped convince Congress to make permanent expansions of refundable tax credits for the poor last year.
So it was somewhat jarring when Greenstein came out against the idea of a universal basic income — an idea that many commentators, myself included, have suggested could end poverty altogether.
In an essay posted on the CBPP website, Greenstein argued that adding a basic income on top of the existing safety net would require tax increases that are simply not politically viable. He also argued that using basic income to replace the safety net would actually increase poverty, because the spending wouldn’t be as well-targeted. It’s a thoughtful, important argument that UBI fans ought to take seriously.
I recently called Greenstein to hear about his concerns with UBI, why means-tested programs don’t necessarily do worse than universal ones, and whether Americans will ever take action to help poor people who can't find work. A transcript, lightly edited for clarity and length, follows.
Dylan Matthews: CBPP usually works on more incremental, in-the-trenches political debates, like fights over expanding the EITC or preventing food stamp cuts. What got you interested in basic income, which is still a pretty pie-in-the-sky proposal?
Robert Greenstein: I've been interested in it for quite some time, from a couple of perspectives. Sometime in the '90s, one of the godfathers of the UBI movement internationally, Sen. Eduardo Suplicy from Brazil, reached out to me, and he would see me when he was in Washington. He actually brought me down to a UBI conference in 1998 in Brasilia.
We’ve had this long dialogue, where I would say that I very much shared the ideas of UBI but in the United States, I didn't think it was feasible or practical. There were, however, ways to move in that direction, such as big expansions of the earned income tax credit and the like. It was an ongoing conversation with Eduardo for a number of years.
On a different front, I for a long time have been in this debate with primarily liberal academics who maybe think a little more philosophically and aren't as close as I am to the policy process, over whether universal programs are always politically stronger and better than means-tested programs.
I wrote a piece on this back around 1991 where I think I demonstrated that, looking at what's actually occurred in recent decades, that's much too simplistic an understanding. It’s not really an accurate characterization of how US social programs and policies work.
I began to get particularly concerned when I began to see, in recent years, the emergence of this notion of going to UBI through a left-right coalition — of which a key building block, for the right, is eliminating most or all means-tested programs. It was at that point that I got much more heavily engaged again. I found that a lot of my fellow progressives did not fully appreciate or understand that that could well move us backward on poverty and inequality, rather than forward.
DM: Talk me through your argument that universal programs aren’t as rock-solid as people believe — and that means-tested programs like food stamps aren’t as politically vulnerable as often assumed. I think that conclusion surprises a lot of people, especially when they see Republicans keep putting out budgets that would gut food stamps and EITC but leave Medicare basically unscathed.
RG: No one has written more about the cuts in Paul Ryan’s budget to programs for the poor than we have, but let’s put that to the side, as none of that’s been enacted. Let's look at what's actually occurred.
Issue No. 1 is that universal social insurance programs in the United States — Social Security, Medicare, unemployment insurance — every one of them is tied to work. We have no universal programs for people who aren't working or don't have an extensive work record. I think that's a crucial point that sometimes gets lost. If, for example, you compare universal social insurance programs that are tied to work to means-tested programs that are tied to work, like the earned income tax credit, you’d have to come to the conclusion that in recent decades, the means-tested programs have done much better politically than the universal programs.
The evidence is clear. Social Security has suffered a 14 percent across-the-board benefit cut. The 1983 Social Security rescue legislation raised the retirement age from 65 to 67. People don’t fully understand that raising the retirement age is an across-the-board cut. No matter what age you retire at, it’s a cut. It's a 7 percent cut for each year you raise the retirement age.
One of the programs that’s been cut the most in recent decades is unemployment insurance. It never fully recovered from the cuts under Reagan in the early 1980s. The percentage of the unemployed getting unemployment benefits, the last time I looked, is below 30 percent, significantly lower than it was before the period of the Reagan budget cuts, which made substantial changes to the underlying structure of unemployment insurance.
If we go to the means-tested side, it's a big mistake in doing this kind of analysis to put all means-tested programs into the same bucket. Means-tested programs that are for the very poor, primarily people who don't work, and do not encompass the working poor and lower-middle-income working families and the like, and on top of that pay their benefits in cash, do very badly politically. Exhibit A: cash welfare assistance.
By contrast, means-tested programs that put together the working poor, near-poor working families, maybe up into the lower middle class, where working families are often the majority of the beneficiaries, and on top of that which provide benefits more in kind, even if it's close to cash (e.g., vouchers to purchase food, Medicaid, etc.) or, if they do provide benefits in cash, do it through the tax system, like the earned income tax credit — those are the programs that have expanded the most.
DM: And it’s not just that universal programs like Social Security have been cut — means-tested programs have, in some cases, been significantly strengthened over time.
RG: Yes, there have been cuts in SNAP [Supplemental Nutrition Assistance Program, or "food stamps"] in 1981 and 1982, again in 1996. But over time, it's kind of been like two steps forward, one step back, three steps forward, one step back. Within some number of years of each group of cuts in the SNAP program, the program overall has been more robust and stronger and more expansive than it was before those cuts. That’s why today we have 44 million people in the SNAP program, whereas not that long ago it was in the 15 million to 25 million range.
I feel that often when people say, "Universal programs do so much better than means-tested programs," they're really comparing something like Social Security, that's tied to work, to cash welfare assistance, that's primarily for people who don't work. They're conflating universal versus means-tested dimension with the work dimension.
I am not trying to make an argument that means-tested programs are always stronger politically than universal, but the often-heard claim is much too simplistic. The reason in recent years that means-tested programs have done better isn't entirely mysterious. You can make substantial gains in means-tested programs — think of things like EITC— for less cost than expansions in universal programs, often.
As has been the case for some time in Washington, you have a better time winning and attaching gains to legislation when the costs aren't massive, especially if you have to offset the costs.
It's a complex picture, and you have to look at each proposal on its own merits. Blanket claims that any universal program is stronger politically than any means-tested program really is not supported by the historical evidence.
DM: One thing your analysis suggests is that there’s just zero political appetite for anything that gives cash — rather than food or housing — to poor people who don’t or can’t work.
There’s a lot of evidence from researchers like Kathryn Edin and Luke Shaefer, which CBPP has highlighted, that millions of Americans are living more or less without cash income, just off food stamps, and are in a very desperate situation as a result. There just aren’t jobs for them, and there aren’t the supports like child care and transportation assistance necessary to make work possible.
Is there anything we can do for those people that’s politically viable? Or is that just a group that the American public is never going to support doing anything to assist?
RG: It's a great question. I think this is a very serious issue, but for multiple reasons I don't think UBI is the answer. For one thing, even UBI’s staunchest supporters say we can get there in 15 to 20 years. I am totally not comfortable with any policy prescription that says we wait 15 to 20 years to deal with very deep poverty.
Second, in US political culture, the hardest area we have encountered, in trying to support people in need, is getting cash to people who aren't working and have no work record. No universal program now does that. The things that we had that did it before were general assistance [aid to poor childless adults] and Aid to Families With Dependent Children [AFDC, or pre-reform welfare], highly imperfect as they were.
Now general assistance is gone in much of the country, and Temporary Assistance to Needy Families [TANF, or post-reform welfare] is a shadow in terms of cash assistance of what AFDC used to be.
Among my fears of UBI is that given US political culture, if we ever got to a form of UBI in the first place (which I think is extremely unlikely), I think the odds are high that it would exclude people who have no earnings and have no work record. Unless you're going to dramatically increase poverty by eliminating most or all low-income programs as part of the financing (which is what right-wing support of UBI is premised on), you're going to need massive tax increases.
There's nothing in US political culture to suggest that there's openness to doing big tax increases, that’d extend well beyond people just at the top, in order to finance cash payments for people who have no earnings and little or no work record. I personally am in favor of doing that! But I don't see support for that. I think they'd likely be excluded.
So if we want to help these people, how do we really deliver the goods? What's the alternative? I think there are a series of steps that could significantly ease the deep poverty at the bottom. They're not politically easy but [are] I think in a number of cases doable, at least to a significant degree.
One would be a substantial subsidized jobs program. We had a variant of that in 2009-2010 through the Recovery Act. It created, within a year, 250,000 job positions for low-income parents and youth. It ended at the end of 2010, but there's some significant conservative support for it now. That's not all one would need to do; that'd just be one step.
I would love to see a European-style universal children's allowance. I think that's highly unlikely for the foreseeable future for the same reason; the children are part of families, and if the families have no earnings, I certainly don't see Republicans agreeing to that.
But we could move in that direction by, for example, beginning to phase in the child tax credit with the first dollar of earnings, and to phase it in at a much faster rate than 15 cents on the dollar. Most of the families in the $2 a Day book don't have zero earnings for a calendar year as a whole. They're in and out of the under-$2-a-day range; sometimes they have a job that falls apart. They may have a few thousand dollars of earnings over the course of the year.
By and large they're eligible for little or no child tax credit now, but under the changes I'm talking about they could get more. There's a provision to this effect in a recent bill that Sen. Michael Bennet (D-CO) introduced. A lot of extremely poor families would get a full $1,000 per child, or much closer to it.
DM: So that’s the children piece of this. What about other areas?
RG: We recently co-hosted a forum on Matthew Desmond's book Evicted. I think one of the most important things to do is to significantly increase efforts to address extreme housing unaffordability at the bottom. That would mean more Section 8 rental vouchers, perhaps expanding the low-income housing tax credit.
There are also a number of really good proposals, that would not solve the problem but make significant progress, in the Obama budget that came out in February. There's an $11 billion proposal to try to end family homelessness by 2020. It's a combination of more rental vouchers targeted at the very poor plus some other steps in the housing area. There's another proposal there for moving toward saturation job creation for low-income youth.
There's yet another proposal there that starts small as kind of a demonstration, but would create a competitive grant program to find the best ways to intervene when poor families like the ones in $2 a Day spiral downward, with the goal of finding the most effective things and then expanding them into a much more substantial program.
Then, of course, there are families that don't have more earnings in part because they don't have stable child care and can't afford to work, and we clearly, in my view, should do something like the proposals that have been in recent Obama budgets to have universal child care availability for people up to, say, twice the poverty line. There's a proposal finally in the Obama budget to start adjusting the TANF block grant upward and requiring a larger share of the money go to cash assistance, child care, and job training programs.
I don't think there's any silver bullet here. We need to push on all of these areas. I don't know whether this would be the case or not, but it would be my hope that if, for example, [Hillary] Clinton is elected president, this becomes a high priority. This isn't necessarily a pro- or anti-UBI argument i'm making, but this absolutely is not something where we can say, "Take 15 to 20 years and see if the politics changes." We have to be moving on all those things right now.
DM: Let me press you a bit on the issue of cost. I kind of think any plan that’s going to really cut the poverty rate dramatically will be about this expensive. If you look at negative income tax proposals — basic incomes that phase out for high earners, basically — one large enough to eliminate poverty, phasing out at 50 cents for every dollar in earnings, would cost $219 billion a year. That’s a lot of money, for sure.
By contrast, there was a plan that the Community Advocates Public Policy Institute in Wisconsin put out that packaged a bunch of more conventional proposals together: increasing the minimum wage, subsidized jobs for the unemployed and underemployed, EITC expansion, topping up Social Security benefits for poor seniors, etc. It would cut poverty by more than half, but it also cost $300 billion to $400 billion a year.
So isn’t any policy package capable of delivering a massive decline in the poverty rate going to cost a fair bit of money? Is this really a problem specific to basic income?
RG: I like the concept of UBI, and I like the proposals that were in that Wisconsin plan, but I had somewhat of the same reaction to the Wisconsin plan. To me, it doesn't help get us to where we have to go. It's so big and expensive that I don't see it being taken seriously within the political system. The problem with all those plans is how to pay for it and where the money would come from.
At the end of the day, if people like it or not, except in situations like the Great Depression social progress in the US basically does occur incrementally. The Affordable Care Act is at the edge of what you can do incrementally, and that's in a lot of political turmoil and not firmly embedded yet, though I hope it will be after the election.
So the concern that someone like myself has is that when you have the potential to really make progress, particularly on poverty, if you have big, very expensive, and therefore highly politically unrealistic proposals, then I worry that people will look at them and say, "Okay, we can do one or two pieces," and too often the pieces that get selected out are pieces where a lot of the money goes to the middle or upper middle class. I want to prioritize things that go to people at the bottom, or maybe the lower middle class.
DM: One reason I write a lot about basic income is not that I think it’s going to pass soon, but because I think giving cash aid to poor people, including the nonworking people, is a very good thing, and I view it as part of my job as a writer with a platform to try in some small way to change public opinion on that.
And the way that basic income has caught on as an idea makes me think it’s a vehicle through which people, particularly the kind of high-income people with jobs that give them enough freedom to read articles about basic income all day, can come to think, "Maybe poor people aren’t lazy; maybe we should trust them to spend money as they see fit."
Do you think there’s any value in basic income as a persuasive tool that can translate to more sympathy for comparatively modest expansions of the safety net?
RG: I certainly agree with that goal, and this is an ongoing effort — one CBPP and I have been involved in for almost 40 years now; I wish we’d been more successful — to try to alter people's assumptions on this.
I very much agree with the guaranteed income goal. The question is how do you get there, and, given the math and US political culture and budget politics, make sure that one is making progress toward that rather than going in the wrong direction? I would view UBI proposals like Charles Murray’s, or even other proposals that don’t eliminate Social Security or Medicare but do eliminate all or most means-tested programs, as clearly steps backward when you do the math.
Having said that, I do think that while it's hard for me, because of the attitudes you've mentioned, to see universal children's allowances for the foreseeable future, I do think there's a clear path toward making however much progress you can make toward that. We have the child tax credit. It is not universal in two senses, one of which doesn't bother me at all and the other one of which does. The one that doesn’t bother me is that for a married couple with two kids it begins to phase down at $110,000 a year, and phases out at $150,000.
This is another point in the universal versus means-tested debate. The child tax credit is effectively means-tested in that it's not universal. It goes to $0 at about $150,000 a year, and the fact that it does has not weakened it politically at all. If it went up to millionaires, it wouldn't make very much political difference, because it covers the vast bulk of the middle class already.
The fact that it phases out at $150,000 doesn't bother me, but it would be great if up to that level it could be universal, and people who didn’t have earnings got the $1,000 per kid as well. Legislation, which I'm glad we helped lead the fight for, was enacted in December, that made permanent the improvements in the child tax credit and EITC originally enacted in the Recovery Act [stimulus package] in 2009. It means that the child tax credit starts phasing in now when you have $3,000 of earnings rather than when you have $13,000 to $14,000, as it would’ve been if the legislation hadn’t passed.
The next step is to start phasing it in at $0. Once you get there, the question is, 10 or 15 years down the road, is it possible to make it fully refundable, to enable families with kids that don’t have earnings to get the full credit as well?
That's an area where I share the goal with some UBI people. I’d love to get there. But I worry a little about the UBI interest being a little bit of a distraction from the immediate steps and fights that actually move toward that. I’ve had this discussion with a couple of UBI people, about starting with the child credit and moving to phase in at $0 and so on, and it's sort of like we're talking past each other. It’s smaller, it’s incremental. But to me, that's how you get toward the goal.
I would argue that the SNAP program is a step towards the goal. It’s the one near-universal income-like floor we have for poor people today. I view the work I've done for 40 years to expand the SNAP program as moving toward the same goal.
DM: Are there other paths, outside of the child tax credit and SNAP?
RG: The other option you and I haven’t discussed is that for other reasons, environmental and global warming reasons, I, like many people, think we need a robust carbon tax. If we could ever get one, I do think there may be a potential to do a modest-size universal payment with a portion of the revenue that’d grow over time.
To me, that's a different route. The biggest obstacle there isn't UBI; it's getting the support to actually impose the tax. But if global warming continues to become more and more of a problem, one certainly hopes that at some point our political system accepts that you’ve got to do something about that. I do think that’s a potential platform.
The 2009 cap-and-trade bill that the House passed, one of the arguments that opponents made against it was is that it would screw the poor and low-income communities, because it would raise fossil fuel energy prices; the poor spend a greater percentage of their income on fossil fuel energy and energy in general than higher-income people do, so it’d have a regressive effect on them.
So for Waxman-Markey we designed a measure that was in the bill that passed that took about 15 percent of the proceeds from auctioning off the permits and used it to, on average, make the bottom fifth of the population whole. The limitation was you only had 15 percent, because in order to pass the bill you had to do things for coal-mining regions. There were things politically you just needed to do with the money. And there were other things that are important to fund, like adaptation, alternative energy research, and the like.
But what Waxman-Markey did was that for the first 10 years the bulk of uses weren't income-related (it was just that 15 percent), but their conclusion, which made a lot of sense to me, was to view the other things like money for coal-mining country as transitional. The way they designed the bill, somewhere maybe 15 years down the line the vast bulk of the revenue would transform into a universal dividend type of payment.
I do think there’s a potential model there. If you get a carbon tax, you’d have to use some of the revenue for other stuff just to pass it, like, as much as I wish this weren’t the case, a corporate rate cut. I wouldn’t otherwise do it, but if that were the price, given how serious the global warming threat is, I'd do that. You'd have to do some transitional stuff.
But hopefully you could design it so that over time a growing share of proceeds were available to make consumers whole. You could move from making sure people at the bottom weren’t made worse off to gradually turning it into a modest universal payment.
