I love this format, and I love both the rebuttal and re-rebuttal here.
I don't love Bruenig already being an outright dick to his colleague in the first week of this publication's launch. His writing is fantastic and his depth of knowledge impressive, and then I tune out every time I see him start slinging needless barbs at people who are acting entirely in good faith.
My father once told a young version of John From FL that "just because you are smart doesn't mean everyone else is dumb." Matt would benefit from similar advice.
I think the least-technical yet most compelling part of KP's original column was that the quotes from the study authors who were surprised these efforts weren't as effective as expected. From my view, these are people who likely would have been very happy to see more promising results - the fact that they were disappointed speaks volumes to me.
I thought Matt Darling's responses on twitter was the most succinct and appropriate response - even though unconditional cash transfers don't remediate some of the first and second order issues we commonly expect to be strongly associated with poverty, preventing the 0th (or most first) order effect of preventing poverty in the immediate is very important.
Still, it isn't promising that people's financial situations (income and assets) had negative effects after receiving the cash transfers.
It's important to have metrics for success, though. Like the point of reducing poverty isn't that having a bigger number in your checking account is intrinsically good; it's that there are certain things that keep people alive and make them happy, and a lack of money prevents them from obtaining some of them. Questioning the purpose is like responding to a study showing that a medicine has no effect on someone's health outcomes by yelling that anyone citing it hates healthcare.
So we should be able to measure the results of cash transfers in some way or other. There are some interesting findings here; perhaps there are some positive outcomes that other measures of well being would capture that these surveys are missing. More interestingly, perhaps program design matters. We have plenty of evidence that, for instance, people gain a greater sense of self worth from getting paid to do a job compared to receiving cash for nothing. That matters. We similarly have evidence that an incremental bump in take home pay is much more effective than a lump sum payment at stimulating economic activity, but that lump sum payments are much more politically popular (because they're noticeable). These kinds of data points matter. And they're the kinds of things we should be thinking about, not taking the Bruenig approach of mocking while saying nothing.
Which is why I liked the opening of Bruenig's piece quoting the 1969 Report of the President's Commission. But then he just didn't follow up with any clarity as to how cash transfers have improved intangible expenditures like "going out on dates" and "having clothes for church" in the US. I'm not anywhere near as boned up on the white papers as he certainly is, so I imagine there's *some* literature out there that finds a way to quantify this, but he just left us guessing.
Well, that's just the issue-- he begged the question. One advantage to the original piece is that it asked the logical question: what did people do with the funds? And it found that they mostly worked a bit less, paid down some debt, and bought kids some more stuff. And that's well and good. But, again, what's the value of that if it doesn't show up elsewhere?
Like we see this debate in discussion of forgiving student loan debt. And it's important to think about it in those terms. Like I graduated law school with lots of student loan debt-- it could have been a mortgage worth in most of the country. But it didn't meaningfully impact my life. I saved a bit less than I would've (though I still saved). I maybe bought a smaller house than I would've if I hadn't spent my twenties paying off student loan debt. But that's about it. My quality of life didn't meaningfully change. And forgiving that student loan debt for people situated like me would've cost a lot of money. Now, there's probably good evidence that student loan debt meaningfully impacts, say, someone who borrows 10% of what I did to get a toilet paper degree from the University of Phoenix. And that debt probably should be forgiven. But my strong hunch is we'd probably see meaningful changes on metrics like stress and quality of life from doing that. And that's how we should be thinking about that.
" But, again, what's the value of that if it doesn't show up elsewhere?"
This seems like you're holding welfare spending to a higher standard than we apply to nearly any other type of program. You never hear Republicans talk about whether tax cuts actually improve well-being; having more money in people's pockets is simply presumed to be a benefit in and of itself. Why shouldn't we apply the same viewpoint to welfare spending, especially since (unlike tax cuts) it's typically going to those who need more money the most?
This was my father's reaction to this piece! He texted me 'Anyway, if the marginal utility of cash transfers is undetectable, doesn't it imply that the marginal disutility of higher marginal tax rates is probably very low??' It's a great take imo.
There certainly should be! That might be a gotcha that works on Republicans, but not on anyone else. And there's been academic work done based on those assumptions. Like the pretty well-known Diamond-Saez paper from 2011 suggested a 73% top marginal income tax rate based on the assumption that the marginal utility of additional income is quite low for high earners.
But yes, that should be the type of analysis we do. We can presume there's some marginal value to producing more, but we should weigh the disincentive supply side effects of higher taxes against the welfare effects of redistribution. And within redistribution, we should study the actual impact of various welfare state policies to determine what should be prioritized. That's the smart way to design taxation and spending policy. Conservatives like to pretend that there is no level at which the welfare effects of taxation can offset the ostensibly horrific moral costs of paying more in marginal taxes. But lots of leftists also like to pretend that all social programs are self-evidently good, and that we can fund them all with more taxes on billionaires. Both claims are self-evidently wrong.
I think what you’re asking for is a direct measurement of utility (or “well-being”). The short answer is: No, there’s no clear way to quantify this.
Think about it: How can we convert dinner dates into church wardrobes? You could use the monetary value a person attached to these things. This is the method economists use for cost-benefit analysis. But all the economists will be quick to tell you that’s it’s not an actual measurement of well-being. For example, it weights benefits to a rich person more highly than to a poor person.
You could also look at how a person rates how happy something made them on, say, a 1-10 scale. This is similar to how researchers measure “subjective well-being.” But it’s fuzzy for obvious reasons.
At a certain point you just have to recognize that there’s some tension between wanting a policy that can address diverse needs and wanting a it’s success to be clear via single metric.
(And fwiw, the experiment in Finland did show an improvement in subjective well-being among those who got cash, but not the experiments in the US)
Agreed on all points. Thankful the work was done (and hope more occurs), but these results should dampen expectations and be digested in an impactful way. I don't think hand-waving the "bad results" and then going "we know functionally that poverty can be solved because three countries did it" is really compelling.
"But other countries did it" is the world's laziest argument. You can't copy and paste something that may work in one place into another and imagine that that ends the argument. It's especially common in the healthcare space.
I don't find 'the study authors were disappointed by how small the effects were' to be an adequate substitute for the numbers.
How much did it help? The answer that I want to that question is a number (a point estimate) with a confidence interval, plus enough context to be able to interpret what that number means (e.g. how big the effect size would be if it matched correlational studies on the relationship between log income and that well-being measure). I want to be able to decide for myself how disappointed to be about the effect sizes, rather than only getting information that's filtered through other people's opinions of the study.
Part of the problem with this is that there were literally hundreds of metrics collected, across a bunch of different studies. I tried dumping them all into a massive spreadsheet earlier this spring, and it was basically impossible.
Matt's argument here reads to me as hostile and condescending. I think arguments are more productive between people who, if they don't like each other, can fake it for the duration of a discussion.
I wonder if this would work better if the authors wrote in second person, or as a dialogue composed of shorter messages.
I think this was good. I’m not as left wing as Bruenig but I like how he forces Vox center left types to actually admit they are not as left wing on policy issues rather than masquerade as “technocratic problem solvers”.
I was first introduced to Bruneig debating Ezra Klein on Medicare for All (https://podcasts.apple.com/us/podcast/the-gray-area-with-sean-illing/id1081584611?i=1000446691804) It was quite revealing how Bruenig (who comes across super rude no doubt) ultimately forced Ezra to admit that it wasn’t just a matter of political prudence or electoralism, he disagreed ideologically. It’s good to lay out differences in values clearly.
From what I see, the animating value behind Bruenig's argument is maximizing how left-wing a policy is, whereas Piper's piece and rebuttal is focused on maximizing benefits in a society that places limits on government power and will.
I honestly don't see an argument in Bruenig's piece other than, "it doesn't fit with my preconceived notion of how the world works."
The difference in values is what the purpose of a social safety net is and how much of that purpose is investment for future/economic development versus simple poverty alleviation.
Easiest way is to answer the question what do you think is the purpose of primary public education? Bruenig, I assume, would argue the greatest benefit of American public education has been less about optimizing human capital and more about providing a safe, reliable place for children to go each day. Schools function as custodial and social infrastructure, offering supervision, meals, and climate-controlled spaces that allow parents to work and communities to operate.
While Piper would say, I assume, purpose is preparing students for the economy. I’m not as left as Bruenig but I do tend to agree the most most consistent and indispensable role of public schools has been logistical—ensuring all children, especially children from abusive homes, are housed, cared for, and out of the labor market during part of the day at least.
We actually have a different word for that. It's "daycare." What separates "school" from "daycare" is actually the instruction and preparation. Learning to read and write and arithmetic or whatever. Or, more aspirationally, for participating in the economy or society or civic life in general.
Discarding those functions of a school is... alarming. I don't think anyone serious (including Bruenig) would actually hold that view of what school is good for.
In a similar vein, discarding the "ulterior" motivations for the social safety net is likely counter-productive. For example, I think the social safety net should encourage a worker feeling safe enough to leave a bad employer to find a better one or to try a hand at entrepreneurship. These actions grow the economy, and a vibrant economy is better for everybody. Including people relying on the social safety net.
And, crucially, they can benefit everyone who has to vote to install such a program, and to vote against the people who threaten to remove the program.
I think the argument Bruenig was making is that at least half the recipients of the social welfare state actually don’t work or have no plausible means/will to work in the future so such incentives in terms of work output seem partially irrelevant. Most of the recipients are elderly, disabled, parents of young children (mothers) or children themselves. We should take care of them even if they might not be participating in economic growth, at least directly.
For one, the studies that Piper cites did not measure work output and focused on stuff like mental health, child development, and housing, so... I'm not sure what you're going on.
And second, nobody is saying that we shouldn't take care of them (except for the strawmen that Bruenig sets up).
The main idea of Piper's piece is that large rigorous studies with direct cash transfers do not show the same secondary benefits (ex. reduction in psychological stress or benefits in child development) that the smaller splashy pilot studies did. And thus to improve general mental health or child development or housing status, cash transfers will probably not be the best option.
I am constantly baffled to see Klein described as “center-left.” His first choice in the 2020 primaries was Warren, followed by Bernie. He supported passing M4A via reconciliation. He supports reparations, not just for slavery, but for colonialism, globally.
But then I see clarifying things like this:
“I like how he forces Vox center left types to actually admit they are not as left wing on policy issues rather than masquerade as “technocratic problem solvers”.
It’s tempting to assume that people who disagree with you, either in ideology or approach, are simply lying about how awful they are. But it’s much more likely that they, like you, want a better world and have sincere differences in belief, temperament, and tactics.
That doesn’t mean you have to agree with them, or even like them. I think Matt Walsh sincerely holds terrible ideas, and I dislike him very much. But you should be able to extend some charity to someone who shares 80% of your values.
The convo lays out the idealistic case for Ezra’s healthcare plan (assuming no political constraints) and it preserves choice in a way Bruenig would disallow it in order to keep the system more egalitarian since he believes choice would only be afforded to rich people in a way that would make the public plans used by poors less robust. Bruenig believes we should force the rich and youngs to participate in public system so we can subsidize poors and old, which means we can’t have choice that allows rich to consume parallel or duplicative plans that are modernized in a way public plans aren’t. Ezra believes it’s important to preserve choice, optionality, and dynamism in health insurance, even if he believes we should have universal coverage. Do you not see that as a difference in value priorities?
It sounds again like you’re talking about a Warren versus Bernie approach. Yes, they had differences, but they also both supported M4A. And I remember how nasty Bruenig was about Warren in the 2020 primary.
Klein is not center-left. He is extremely progressive. In some regards, he is more progressive than Bruenig. But I see tons of leftists dismiss him because he’s wonkish and speaks in a mild tone of voice.
I think Ezra is very socially liberal (progressive) but he is not a socialist, which explains why he doesn’t feel comfortable putting a ceiling on rich people’s options. He is, however, very enthusiastic about spending on poor people. Do you think I am wrong? Did you listen to the pod?
I haven’t listened to that specific one, but I’ve followed Klein for at least a decade. I was a devout Weeds listener, and I follow his writing/podcast at the NYT. I don’t know that we are disagreeing particularly. Klein would agree he’s not a socialist. I don’t think he’d need to be strong-armed into acknowledging that, so I don’t know how Bruenig adds any value by having him say it.
I would recommend it I enjoyed the middle/later parts especially. I agree Ezra is smart and that’s why it was interesting watching him being challenged a bit by someone who’s (no doubt very rude) also very policy aware and wonky
I agree with this but I was talking about how at the end of the M4A convo, it was clear that Ezra thinks preserving choice was slightly more important and Bruenig thought that if choice was only afforded to rich people, we shouldn’t allow it. Ezra clearly favors putting a floor for the poors but is uncomfy putting a ceiling on how much rich can consume. Bruenig clearly believes keeping supply constant, we should ration healthcare based on need not ability to pay, which means rich people shouldn’t be allowed to get ahead in line.
“Ezra clearly favors putting a floor for the poors”
If he clearly favors that, then did Bruenig actually force him to admit anything? Also, I’m not sure if you’re being tongue in cheek or not, but I read your phrasing as indicating that you think Klein looks down on poor people, which again, is both uncharitable and factually incorrect. Do I misunderstand you?
God no, I meant Ezra is a lib and he favors creating a social safety net/floor for poor people in society so they don’t ever go below a certain floor in terms of standard of living. But he is uncomfortable in putting a ceiling in how much rich people can consume. Maybe you can say that’s because he thinks putting a ceiling reduces dynamism, innovation and fosters a zero-sum worldview and is slightly authoritarian even. Thats a perfectly fine liberal position to hold, but that’s what distinguishes him from people to his left. Bruenig is much more comfortable putting a government mandated ceiling and reducing choice by banning private insurance/healthcare altogether so everyone is forced to participate in the same public system.
Okay, thank you for clarifying. I am reacting to this: “ I like how he forces Vox center left types to actually admit they are not as left wing on policy issues rather than masquerade as “technocratic problem solvers”.”
Most people at Vox are not center-left. *I* am center-left. Most people at Vox are to my left. They are progressive. I haven’t followed Bruenig all that much, but my understanding is that he’s a leftist. While Vox writers and I aren’t perfectly aligned politically, we are both technocratic, and I’d also say I’m a pragmatist.
What I really object to is people claiming that being technocratic, mild and practical mean that I’m secretly much further to the right than I claim to be, and that if I’m interrogated aggressively enough, that veil will be torn away.
I don’t support M4A, and I don’t think it’s particularly ideological. I sincerely think we should model our healthcare system on what works best globally, and I think the most successful models use a blend of private and public insurance.
I am generally pro-market, and so sure, that’s ideological and going to affect my assessment. But my ideology is not a secret, and it’s not the “real” reason I oppose M4A. My concerns are what I say they are: 99% practical.
I’m also not as left as Bruenig on every issue but I’m very left on healthcare and do support M4A. I do think there’s a tendency among center left types to not articulate their ideological/value based differences with leftists/socialists and ONLY highlight how they are more pragmatic or more aware of political constraints. I think leftists find it frustrating and it would be helpful to highlight the role of morality and ideology in the construction of policy design and not assume it’s all just a strategic/technocratic difference. Bruenig is wonky himself and therefore him arguing with center left types clarifies their ideological differences is all I was trying to say.
'One quirk of the Finnish study is that the government ran it and was therefore able to literally force the randomly selected individuals to participate in it. The studies Piper cites rely on volunteers, which technically makes them nonrandom, despite the fact that they are called “randomized control trials.”'
Congratulations, Matt Bruenig, for throwing out nearly all pharmaceutical research!
The "random" in RCT is, of course, that treatments given to any participant in the study are randomized, not that the population being studied is randomly selected from the actual total population of anyone who has the relevant disease. Medical RCTs have to recruit subjects just like these UBI studies did. Yes, RCTs can be questioned by having populations that are skewed in systematic ways (women's general non-inclusion in pharmaceutical trials for decades because what if those monthly hormones cause the medicine to not work, or what about the always potential baby that could appear!?), but Bruenig doesn't even gesture toward that kind of criticism, much less state it outright. Surely Bruenig has to know all this about RCTs, so why is he being so obstinate to attack these studies in such a transparently dumb manner.
i thought bruenig was writing in good faith up until i read that specific sentence... if he's going to do throwaway barbs at his debate opponent, fine, not everyone is cut out for SSC discussion norms. but that one... that was just a barb for the sake of hostility and nothing else. it made it clear that if bruenig saw an opportunity to demean his ideological opponents, he was going to take it, whether or not it made sense.
i was already kinda shocked at the implication that kelsey piper of all people might not have ever *considered* any of the ideas he laid out... but i thought it was possible, just barely possible, that bruenig wasn't very familiar with piper and was actually making the criticism in good faith
but after seeing that sentence you pointed out, i downgraded my assessment. he is performing marketing, not truthseeking.
I have a couple thoughts. I am very glad to see Matt Bruenig bring up the fact that our K-12 schools, in terms of "outcomes," do very well. I think what a lot of people have a problem with in some public schools are widespread disobedience and disorder that, while bad, don't actually make much of a difference in terms of learning. Kelsey Piper's argument that school reforms haven't caused improvements in learning doesn't actually contradict Bruenig's point. Schools can be good and also not improving.
Also, I really enjoyed Piper's original piece, but I feel like it and possibly this piece aren't really addressing the actual UBI question. To me, the question that many people have (myself included) isn't "does giving people some extra money make their life better?" It's this: "When many people will literally be unable to find work because AI and automation have taken everyone's job, will we be able to simply pay everyone what they need to live and still have a functioning society?" Of course, this is much harder to test out with experiments in the present.
Also, I see a lot of comments taking issue with Bruenig's tone, but... this feature is literally called "MAD libs." I think he's just following the assignment. (And I've followed him as a writer for a long time and I know he is passionate about this issue.) And to be honest, this publication needs to get off the ground, and if all they offer is dispassionate policy discussions based around fact-checking, no one is going to read it.
It’s crazy to me that nobody’s acknowledging that if the predicted big and dramatic benefits from giving people money are real, we wouldn’t have seen them yet. This is stuff like the earnings of cash recipients’ kids when they grow up, the criminality of cash recipients’ kids when they grow up, and the health of cash recipients’ kids when they grow up.
I’m definitely more skeptical that these benefits are going to materialize given the recent experiments failed to produce the predicted short-term benefits. But we still don’t really know!
Also: Matt is being way too hostile, and Kelsey is right to be disappointed…but Matt is right that simple poverty reduction is a good result.
Edit: Some shameless self-promo - I wrote a much more measured take on the recent experiments here: https://jg12402863.substack.com/p/giving-people-money-down-but-not (why should you care what I think? Honestly, no really good reason, but I did work pretty hard at it, and I was an RA for baby’s first years a few years ago)
In general I have a strong skepticism at this point of claims that long-term effects will manifest when short term effects do not. If a program like this is to have long term benefits, it has to be via immediate effects, and time is going to introduce noise and other variables. If we can't measure anything now I am very very skeptical we'd be able to measure anything later.
I also think that the main cases where you see lifetime effects on eg income or criminality, it's via literal brain damage - lead poisoning, in the developing world severe malnutrition, etc. Preventing that should of course be high among our priorities as a society, but targeted programs probably work best there.
I mostly agree, but want to flag that there are *some* intermediate effects we’ve measured - like increased healthcare consumption (as well as consumption of pretty much everything else).
Re criminality, reductions due to cash are predicted by quasi-experimental work on the Eastern Cherokee cash payments in North Carolina. Social scientists have previously used this to predict some pretty large effects of cash on crime. But I do have some skepticism about exogeneity in that work, especially after these recent null results.
Is everyone now in agreement that leaded gasoline caused the huge spike in crime in the 80s? If so, I’m curious, did it also correlate to a drop in national IQ?
In Piper's framing, the question isn't whether "simple poverty reduction is a good result," the question is whether that's the *best* result possible with the money going towards direct transfers. And to be fair, a large section of the concluding paragraphs of the original article is about exactly that.
Yeah, I agree that this is the question we should be asking! But as far as I’m aware, we don’t have many similarly strong RCTs showing better effects from in-kind transfers. (Though we should try some and find out!)
I definitely think these studies cast doubt on the most optimistic visions of cash benefits. Still, they have a lot of limitations, so I want to emphasize the uncertainty as you do. They’re not that much money in the grand scheme of things, especially for families with kids, and they were time-limited. Also, as Kelsey acknowledges, there might be community-wide benefits if everyone got a cash benefit.
Some (though certainty not most) economists predicted this way back before the results were released - Hoynes & Rothstein said the experiments would be good for getting better estimates of labor supply elasticity, but wouldn’t tell much about long-term/universal cash
As is typical, Bruenig’s rebuttal is ideologically driven (cash transfers must work in America because he WANTS them to work in America) and seemingly full of personal animus. She’s your colleague, man.
This piece was quite typical of Bruenig-- lots of big claims, but nothing resembling any evidence of thought or willingness (or ability) to grapple with the argument. His collection of words (I hesitate to call it a piece) amounts to an extended tantrum-- first, that it's self-evident that giving people cash makes their lives better (in the 9th grade, where Bruenig's intellectual development appears to have ended, we call this "assuming your conclusion"); second, that other studies find different results (let's name those studies and why they're methodologically better than the studies that Piper cited); and third, that we should just "be Finland." Bruenig writing embarrassing drivel would be fine if he weren't so incredibly smug and dismissive. I don't mind arrogance, per se; if you're very very smart, not suffering silliness is fine. Bruenig is not very very smart, or even a little smart. He's very much the opposite.
So unfortunately, the only takeaway I get from reading this is that Bruenig has nothing worthwhile to say. But that doesn't mean that there aren't good reasons to think that the studies might be missing or failing to measure something relevant, or whether something about the studied programs' design makes them ineffective. But the messenger for that will have to be someone with far far far more intellectual horsepower than Bruenig.
I am not sure to want extent I agree with Bruenig but I disagree with your point here; he's a good debater and his piece here is useful in pushing the cash assistance-skeptical side to be more specific about what their goals are, what standards they hold policies to, and what their view is as of which policies are and are not politically feasible. I assume that this was basically why he was hired (though would also be curious as to why he was interested in the job) and I think if the less lefty writers here learn to think with his critiques in mind they will be better off for it. I did find some the antagonistic tone a little unpleasant to read but hey, part of the point of this publication is to help liberals shed our reputation as pathologically agreeable.
That’s the thing— he’s not a good debater. He’s a very bad debater. The piece is very specific as to the standards to which it’s holding the policies— it’s a survey of studies that found that, broadly, unconditional cash transfers didn’t improve outcomes over a number of measures of well being relative to a control group. A smart critique would have suggested alternate measures that the studies might have missed. Or it would have suggested alternate program designs that might better improve outcomes on those measures.
He put together… a string of verbal diarrhea that amounted to “of course cash assistance is good, look at Finland.” To call this a bad argument would be too generous. It’s not an argument at a; it’s restating a conclusion.
Now, I don’t per se care that he was also an asshole. Lots of smart people are dismissive. It’s not that he’s an asshole and a leftist; it’s that he’s dismissive and an idiot. That’s a bad combination. Now, should we consider leftists’ critiques? Sure, if they’re thoughtful. But what he wrote wasn’t thoughtful. It was painfully stupid. And he was a smarmy asshole on top of that.
We should be open to all forms of thoughtful critique, from anywhere on the political spectrum. If something isn’t thoughtful but is courteous, pushing back courteously is good. If something isn’t thoughtful and is smarmy, well, it’s good to let that person know that if they can’t be smart, they should try not being a dick.
"Many people who would support the safety net if it were to succeed, oppose it precisely because they perceive it as failing"
This is such an odd assumption to make. By far the biggest opponents of the safety net are Republicans, and their disinterest in whether or not it actually works is palpable--you can tell because they make no effort whatsoever to confine their cuts to the things that (they believe) don't work, nor do they make any effort to bolster spending on the things they think do.
The real story in American anti-poverty policy is that a large share of conservative elites simply opposes spending money on poverty alleviation, on moral grounds. They believe that cutting taxes for the rich is morally better than spending money on helping the poor. This belief, not a principled dispute over program efficacy, is what drives opposition to the safety net. You're not going to convince these people to change their bedrock moral views simply by doing an RCT that finally convinces them that this one intervention really will work.
While I agree with Matt’s main argument in favor of basic income (the effect on class) and consider myself a big fan of such a program even with Kelsey’s piece in mind- I think he completely misread the original piece and simultaneously criticized it for a failing that his own critique suffers from more prominently. In arguing that the question of class stratification is solved by the Nordic model he ignores the obvious fact that there are other very consequential forms of social division that cash payments may not rectify! Kelsey rightly points out that in the United States, maybe housing policy has the potential to frustrate simple cash transfers, I agree. I can think of a few more reasons why one country or population might be different from another!
Matt beats whatever point he has so throughly against a straw man he’s left with useless nub.
It's a little unfortunate, in some ways, that the argument here ended up getting followed up immediately one twitter, but it did helpfully reveal that one key point of disagreement was how to define the marginal dollar: for Bruenig, you save it by reducing administrative costs via M4A, for Piper that's not possible so you look to incrementally increase funding to programs with greater records of success than cash benefits. I'd love to see a follow-up along those lines; I know Bruenig does not take a very high view of "what could we get passed" analysis, but it would be interesting to see each writer explain what their ideal welfare state would look like and, separately, what they think is best given current political realities.
It’s a good exchange, I think he lays out his case and where his values diverge from Piper’s pretty clearly. I happen to agree with Bruneig- and Matt Darling’s take that cash transfers are simply a better way to conduct transfers to the extent we do them (fragmented in kind benefits). It’s not magic, just an anti poverty program. Overall I think Bruneig is simply more correct about the purpose, outcomes, and goal of the welfare state.
The problem is America doesn’t have the level of welfare state as the more developed welfare states (eg no child allowance) is piper really so oblivious to make all these claims about how poverty is more complicated when the CTC halved child poverty instantly for 100B a year? And it doubled when it expired? I challenge her to find another proposal that would halve child poverty for a similar amount or less. that I don’t think she appreciates how much poverty is a structural issue caused by non workers (children, elderly, disabled, etc) half of parents wouldn’t be in poverty had they not had kids. The Welfare State is the only solution that can solve this problem. She mentions the growth of welfare state spending as a % of GDP but doesn’t compare it to our countries like the Nordics which, especially on children, is lacking.
Per the OECD: the US had a 21% child poverty rate in 2019. The Nordics had an average of half that or less throughout. The CTC brought it down to 14%(!) in 2021- which considering this is a relative measure and the US is at a higher baseline state of inequality is very impressive. No doubt if it increased family spending as a percent of GDP relative to the Nordic countries we’d see a further reduction.
You write: "Piper is simply wrong that there has been no progress since the war on poverty."
Meanwhile, the real Piper writes: "You can see the Great Society on there — it did genuinely help! In 1969, 24% of people lived below the poverty line under this measure before transfers and 22.5% after them. In 2023, 22% lived below the poverty line before transfers and 11% after. Those are real gains."
You misrepresented her writing in such an egregious manner as to cast the rest of your comment in serious doubt.
Sorry I’ll edit it to say not enough progress, and I think she implies a stagnation in progress (later on in the quote you mention) relative to the growth in spending and my links I reference confirm that isn’t true. These programs are effective and they work and they’ve gotten more effective as the welfare state has expanded in fits and starts.
Poverty has been reduced as a result of transfers much more now than in the 70s (47% v 9%) this is in the CBPP source I linked. I also think her pivot to pre transfer income dodges the whole point of the discussion which is how can we best reduce poverty via government spending. Because if we want to do that we’d expand the circle to minimum wages and reviving organized labor and that is beyond the scope of the discussion.
I’d again encourage her to find a better program that costs less and reduces poverty more.
I’d say that the halving of post transfer poverty is a BFD. We are nowhere near the ceiling of what can be accomplished via transfers as the CTC expansion reminds us. Piper even acknowledge that the research supports the expansion and we didn’t see any deleterious effects like a reduction in labor supply
And everything I’ve said is backed up with direct empirical evidence, if I didn’t cite a figure I can find it for you.
I wasn’t intentionally trying to mischaracterize anyone- I was typing quickly on my phone and have gone back and edited it with more sources multiple times
Sort of a side point, but given GOP and Trumpian authoritarianism and social conservatism…is M4A actually a good thing in the US context? I work in healthcare so I see all the inefficiencies, and quite frankly the only way around them is for government to take responsibility. Part of that is hospital/insurer rents, and a big part of that is we are very overregulated compared to any existing socialized system (government run systems in general acknowledge trade offs and realize that rules have compliance costs and inefficiencies which sometimes aren’t worth it).
My concerns in the US contexts are (1) here the government is absolutely willing to overregulate itself, so I’m not sure the cost savings materialize to the extent we like (especially with doctors and nurses being part of the Democratic base and having coalitional power to keep extracting rents, especially if/when we unionize) and
(2) how would we prevent the GOP from just banning abortion, transgender care, vaccines etc when they are in power? Decentralization is inefficient but it might be protective given what we know about the actually existing US
Really enjoying this format, I definitely want to see more of it. Mostly convinced by Piper here - I think my cruxes are (a) you can't assume that policies that work in Finland work here, indeed that's the surprising part of these studies; and (b) there are real tradeoffs here that Piper is discussing and Bruenig seems to be ignoring. Bruenig's best part was the opening, but I think he drifts away from that core point and doesn't acknowledge that the studies in question are indeed mainly looking at non-work outcomes.
I would love to read a follow-up on schooling in the US and how we should think about the metrics on both sides (Bruenig - we do solidly on international assessments vs Piper - we're not close to the results we saw in Mississippi).
> Transfer payments to individuals have risen, in 2019 dollars, from $417 billion (in 2024 dollars) in 1969, or 4.8% of GDP, to $3.2 trillion in 2024 — or about 12% of GDP.
Is this in 2019 dollars or 2024 dollars? Not sure if it's a typo or just different sources, but it should be reconciled to the same standard. I think the point the same either way, it's just confusingly phrased.
Right, that's one we tested and it worked well (and it would be great if we reinstated it). Less targeted cash transfers were tested here and (surprisingly) did not work well. Some Finnish policies succeed here, others don't seem to, so we can't assume that they all will.
I mean is it really targeted if the payout was practically to all parents, working or not? Seems more like a universal basic child allowance than a means tested targeted benefit like TANF or SNAP.
Yeah, it’s definitely less targeted than most similar programs. I think that led to a lot of the predictable gains in reducing overhead, both on the govt side and on the side of applying for benefits. That’s another big part of why the unconditional transfers not working so well was surprising to me too!
But if the UBI programs had positive effects on the scale that ECTC did, shouldn’t we see that in these studies? I’m not saying I’m totally confident in why we don’t, but it seems like that amount of targeting really does make a difference.
Maybe, but I’m not too sure. The full amount of $3,600 was given at 0$ and only begins to phase out at 112k, and then only phases out again after that at 200k to $0 at 400k a year
So is targeting to parents the solution this combination of facts suggests?
It’s also plausible to me that one or more of these is true:
-the ECTC didn’t actually have significant effects beyond lowering child poverty - I haven’t done much lit review on it, it’s possible that the same thing happened as in these studies, i.e. lowering of ‘child poverty’ as a measure didn’t lead to broad knock-on improvements in other measures.
-both the ECTC and unconditional cash transfers do have broad positive effects beyond ‘poverty’ but they are diffuse and hard to measure
-some other part of the ECTC’s approach/conditions/targeting/timing made a bigger difference
Of course that’s not to say that ‘poverty’ is a bad thing to reduce, just that when I talk about poverty in general I mean a whole system of poor outcomes that are not just a lower amount of money. If the other outcomes (like health and housing and hunger) don’t change, I’d prioritize a program that affects those over one that affects ‘poverty’.
I love this format, and I love both the rebuttal and re-rebuttal here.
I don't love Bruenig already being an outright dick to his colleague in the first week of this publication's launch. His writing is fantastic and his depth of knowledge impressive, and then I tune out every time I see him start slinging needless barbs at people who are acting entirely in good faith.
My father once told a young version of John From FL that "just because you are smart doesn't mean everyone else is dumb." Matt would benefit from similar advice.
I think the least-technical yet most compelling part of KP's original column was that the quotes from the study authors who were surprised these efforts weren't as effective as expected. From my view, these are people who likely would have been very happy to see more promising results - the fact that they were disappointed speaks volumes to me.
I thought Matt Darling's responses on twitter was the most succinct and appropriate response - even though unconditional cash transfers don't remediate some of the first and second order issues we commonly expect to be strongly associated with poverty, preventing the 0th (or most first) order effect of preventing poverty in the immediate is very important.
Still, it isn't promising that people's financial situations (income and assets) had negative effects after receiving the cash transfers.
> I thought Matt Darling's responses on twitter was the most succinct and appropriate response
Me too.
If you keep on being succinct and appropriate, I'll have no choice but to keep citing you.
It's important to have metrics for success, though. Like the point of reducing poverty isn't that having a bigger number in your checking account is intrinsically good; it's that there are certain things that keep people alive and make them happy, and a lack of money prevents them from obtaining some of them. Questioning the purpose is like responding to a study showing that a medicine has no effect on someone's health outcomes by yelling that anyone citing it hates healthcare.
So we should be able to measure the results of cash transfers in some way or other. There are some interesting findings here; perhaps there are some positive outcomes that other measures of well being would capture that these surveys are missing. More interestingly, perhaps program design matters. We have plenty of evidence that, for instance, people gain a greater sense of self worth from getting paid to do a job compared to receiving cash for nothing. That matters. We similarly have evidence that an incremental bump in take home pay is much more effective than a lump sum payment at stimulating economic activity, but that lump sum payments are much more politically popular (because they're noticeable). These kinds of data points matter. And they're the kinds of things we should be thinking about, not taking the Bruenig approach of mocking while saying nothing.
Which is why I liked the opening of Bruenig's piece quoting the 1969 Report of the President's Commission. But then he just didn't follow up with any clarity as to how cash transfers have improved intangible expenditures like "going out on dates" and "having clothes for church" in the US. I'm not anywhere near as boned up on the white papers as he certainly is, so I imagine there's *some* literature out there that finds a way to quantify this, but he just left us guessing.
Well, that's just the issue-- he begged the question. One advantage to the original piece is that it asked the logical question: what did people do with the funds? And it found that they mostly worked a bit less, paid down some debt, and bought kids some more stuff. And that's well and good. But, again, what's the value of that if it doesn't show up elsewhere?
Like we see this debate in discussion of forgiving student loan debt. And it's important to think about it in those terms. Like I graduated law school with lots of student loan debt-- it could have been a mortgage worth in most of the country. But it didn't meaningfully impact my life. I saved a bit less than I would've (though I still saved). I maybe bought a smaller house than I would've if I hadn't spent my twenties paying off student loan debt. But that's about it. My quality of life didn't meaningfully change. And forgiving that student loan debt for people situated like me would've cost a lot of money. Now, there's probably good evidence that student loan debt meaningfully impacts, say, someone who borrows 10% of what I did to get a toilet paper degree from the University of Phoenix. And that debt probably should be forgiven. But my strong hunch is we'd probably see meaningful changes on metrics like stress and quality of life from doing that. And that's how we should be thinking about that.
" But, again, what's the value of that if it doesn't show up elsewhere?"
This seems like you're holding welfare spending to a higher standard than we apply to nearly any other type of program. You never hear Republicans talk about whether tax cuts actually improve well-being; having more money in people's pockets is simply presumed to be a benefit in and of itself. Why shouldn't we apply the same viewpoint to welfare spending, especially since (unlike tax cuts) it's typically going to those who need more money the most?
This was my father's reaction to this piece! He texted me 'Anyway, if the marginal utility of cash transfers is undetectable, doesn't it imply that the marginal disutility of higher marginal tax rates is probably very low??' It's a great take imo.
There certainly should be! That might be a gotcha that works on Republicans, but not on anyone else. And there's been academic work done based on those assumptions. Like the pretty well-known Diamond-Saez paper from 2011 suggested a 73% top marginal income tax rate based on the assumption that the marginal utility of additional income is quite low for high earners.
https://www.aeaweb.org/articles?id=10.1257/jep.25.4.165
But yes, that should be the type of analysis we do. We can presume there's some marginal value to producing more, but we should weigh the disincentive supply side effects of higher taxes against the welfare effects of redistribution. And within redistribution, we should study the actual impact of various welfare state policies to determine what should be prioritized. That's the smart way to design taxation and spending policy. Conservatives like to pretend that there is no level at which the welfare effects of taxation can offset the ostensibly horrific moral costs of paying more in marginal taxes. But lots of leftists also like to pretend that all social programs are self-evidently good, and that we can fund them all with more taxes on billionaires. Both claims are self-evidently wrong.
Because Republicans are wrong to do so. Why would we copy a bad approach?
Because unilaterally holding only one side to a higher standard is worse than holding neither side to that standard. Consistency matters here.
I think what you’re asking for is a direct measurement of utility (or “well-being”). The short answer is: No, there’s no clear way to quantify this.
Think about it: How can we convert dinner dates into church wardrobes? You could use the monetary value a person attached to these things. This is the method economists use for cost-benefit analysis. But all the economists will be quick to tell you that’s it’s not an actual measurement of well-being. For example, it weights benefits to a rich person more highly than to a poor person.
You could also look at how a person rates how happy something made them on, say, a 1-10 scale. This is similar to how researchers measure “subjective well-being.” But it’s fuzzy for obvious reasons.
At a certain point you just have to recognize that there’s some tension between wanting a policy that can address diverse needs and wanting a it’s success to be clear via single metric.
(And fwiw, the experiment in Finland did show an improvement in subjective well-being among those who got cash, but not the experiments in the US)
Agreed on all points. Thankful the work was done (and hope more occurs), but these results should dampen expectations and be digested in an impactful way. I don't think hand-waving the "bad results" and then going "we know functionally that poverty can be solved because three countries did it" is really compelling.
"But other countries did it" is the world's laziest argument. You can't copy and paste something that may work in one place into another and imagine that that ends the argument. It's especially common in the healthcare space.
I don't find 'the study authors were disappointed by how small the effects were' to be an adequate substitute for the numbers.
How much did it help? The answer that I want to that question is a number (a point estimate) with a confidence interval, plus enough context to be able to interpret what that number means (e.g. how big the effect size would be if it matched correlational studies on the relationship between log income and that well-being measure). I want to be able to decide for myself how disappointed to be about the effect sizes, rather than only getting information that's filtered through other people's opinions of the study.
I think KP has the links to the study in her original piece, and at the very least in a comment chain I had w/ her.
Part of the problem with this is that there were literally hundreds of metrics collected, across a bunch of different studies. I tried dumping them all into a massive spreadsheet earlier this spring, and it was basically impossible.
Matt's argument here reads to me as hostile and condescending. I think arguments are more productive between people who, if they don't like each other, can fake it for the duration of a discussion.
I wonder if this would work better if the authors wrote in second person, or as a dialogue composed of shorter messages.
I agree that this would feel more natural in a second-person format.
I think this was good. I’m not as left wing as Bruenig but I like how he forces Vox center left types to actually admit they are not as left wing on policy issues rather than masquerade as “technocratic problem solvers”.
I was first introduced to Bruneig debating Ezra Klein on Medicare for All (https://podcasts.apple.com/us/podcast/the-gray-area-with-sean-illing/id1081584611?i=1000446691804) It was quite revealing how Bruenig (who comes across super rude no doubt) ultimately forced Ezra to admit that it wasn’t just a matter of political prudence or electoralism, he disagreed ideologically. It’s good to lay out differences in values clearly.
What difference in values do you see here?
From what I see, the animating value behind Bruenig's argument is maximizing how left-wing a policy is, whereas Piper's piece and rebuttal is focused on maximizing benefits in a society that places limits on government power and will.
I honestly don't see an argument in Bruenig's piece other than, "it doesn't fit with my preconceived notion of how the world works."
The difference in values is what the purpose of a social safety net is and how much of that purpose is investment for future/economic development versus simple poverty alleviation.
Easiest way is to answer the question what do you think is the purpose of primary public education? Bruenig, I assume, would argue the greatest benefit of American public education has been less about optimizing human capital and more about providing a safe, reliable place for children to go each day. Schools function as custodial and social infrastructure, offering supervision, meals, and climate-controlled spaces that allow parents to work and communities to operate.
While Piper would say, I assume, purpose is preparing students for the economy. I’m not as left as Bruenig but I do tend to agree the most most consistent and indispensable role of public schools has been logistical—ensuring all children, especially children from abusive homes, are housed, cared for, and out of the labor market during part of the day at least.
We actually have a different word for that. It's "daycare." What separates "school" from "daycare" is actually the instruction and preparation. Learning to read and write and arithmetic or whatever. Or, more aspirationally, for participating in the economy or society or civic life in general.
Discarding those functions of a school is... alarming. I don't think anyone serious (including Bruenig) would actually hold that view of what school is good for.
In a similar vein, discarding the "ulterior" motivations for the social safety net is likely counter-productive. For example, I think the social safety net should encourage a worker feeling safe enough to leave a bad employer to find a better one or to try a hand at entrepreneurship. These actions grow the economy, and a vibrant economy is better for everybody. Including people relying on the social safety net.
And, crucially, they can benefit everyone who has to vote to install such a program, and to vote against the people who threaten to remove the program.
I think the argument Bruenig was making is that at least half the recipients of the social welfare state actually don’t work or have no plausible means/will to work in the future so such incentives in terms of work output seem partially irrelevant. Most of the recipients are elderly, disabled, parents of young children (mothers) or children themselves. We should take care of them even if they might not be participating in economic growth, at least directly.
For one, the studies that Piper cites did not measure work output and focused on stuff like mental health, child development, and housing, so... I'm not sure what you're going on.
And second, nobody is saying that we shouldn't take care of them (except for the strawmen that Bruenig sets up).
The main idea of Piper's piece is that large rigorous studies with direct cash transfers do not show the same secondary benefits (ex. reduction in psychological stress or benefits in child development) that the smaller splashy pilot studies did. And thus to improve general mental health or child development or housing status, cash transfers will probably not be the best option.
I am constantly baffled to see Klein described as “center-left.” His first choice in the 2020 primaries was Warren, followed by Bernie. He supported passing M4A via reconciliation. He supports reparations, not just for slavery, but for colonialism, globally.
But then I see clarifying things like this:
“I like how he forces Vox center left types to actually admit they are not as left wing on policy issues rather than masquerade as “technocratic problem solvers”.
It’s tempting to assume that people who disagree with you, either in ideology or approach, are simply lying about how awful they are. But it’s much more likely that they, like you, want a better world and have sincere differences in belief, temperament, and tactics.
That doesn’t mean you have to agree with them, or even like them. I think Matt Walsh sincerely holds terrible ideas, and I dislike him very much. But you should be able to extend some charity to someone who shares 80% of your values.
The convo lays out the idealistic case for Ezra’s healthcare plan (assuming no political constraints) and it preserves choice in a way Bruenig would disallow it in order to keep the system more egalitarian since he believes choice would only be afforded to rich people in a way that would make the public plans used by poors less robust. Bruenig believes we should force the rich and youngs to participate in public system so we can subsidize poors and old, which means we can’t have choice that allows rich to consume parallel or duplicative plans that are modernized in a way public plans aren’t. Ezra believes it’s important to preserve choice, optionality, and dynamism in health insurance, even if he believes we should have universal coverage. Do you not see that as a difference in value priorities?
It sounds again like you’re talking about a Warren versus Bernie approach. Yes, they had differences, but they also both supported M4A. And I remember how nasty Bruenig was about Warren in the 2020 primary.
Klein is not center-left. He is extremely progressive. In some regards, he is more progressive than Bruenig. But I see tons of leftists dismiss him because he’s wonkish and speaks in a mild tone of voice.
I think Ezra is very socially liberal (progressive) but he is not a socialist, which explains why he doesn’t feel comfortable putting a ceiling on rich people’s options. He is, however, very enthusiastic about spending on poor people. Do you think I am wrong? Did you listen to the pod?
I haven’t listened to that specific one, but I’ve followed Klein for at least a decade. I was a devout Weeds listener, and I follow his writing/podcast at the NYT. I don’t know that we are disagreeing particularly. Klein would agree he’s not a socialist. I don’t think he’d need to be strong-armed into acknowledging that, so I don’t know how Bruenig adds any value by having him say it.
I would recommend it I enjoyed the middle/later parts especially. I agree Ezra is smart and that’s why it was interesting watching him being challenged a bit by someone who’s (no doubt very rude) also very policy aware and wonky
I agree with this but I was talking about how at the end of the M4A convo, it was clear that Ezra thinks preserving choice was slightly more important and Bruenig thought that if choice was only afforded to rich people, we shouldn’t allow it. Ezra clearly favors putting a floor for the poors but is uncomfy putting a ceiling on how much rich can consume. Bruenig clearly believes keeping supply constant, we should ration healthcare based on need not ability to pay, which means rich people shouldn’t be allowed to get ahead in line.
The values are egalitarianism in healthcare versus choice, that’s a difference in values. Listen to the whole conversation: https://podcasts.apple.com/us/podcast/the-gray-area-with-sean-illing/id1081584611?i=1000446691804
“Ezra clearly favors putting a floor for the poors”
If he clearly favors that, then did Bruenig actually force him to admit anything? Also, I’m not sure if you’re being tongue in cheek or not, but I read your phrasing as indicating that you think Klein looks down on poor people, which again, is both uncharitable and factually incorrect. Do I misunderstand you?
God no, I meant Ezra is a lib and he favors creating a social safety net/floor for poor people in society so they don’t ever go below a certain floor in terms of standard of living. But he is uncomfortable in putting a ceiling in how much rich people can consume. Maybe you can say that’s because he thinks putting a ceiling reduces dynamism, innovation and fosters a zero-sum worldview and is slightly authoritarian even. Thats a perfectly fine liberal position to hold, but that’s what distinguishes him from people to his left. Bruenig is much more comfortable putting a government mandated ceiling and reducing choice by banning private insurance/healthcare altogether so everyone is forced to participate in the same public system.
Okay, thank you for clarifying. I am reacting to this: “ I like how he forces Vox center left types to actually admit they are not as left wing on policy issues rather than masquerade as “technocratic problem solvers”.”
Most people at Vox are not center-left. *I* am center-left. Most people at Vox are to my left. They are progressive. I haven’t followed Bruenig all that much, but my understanding is that he’s a leftist. While Vox writers and I aren’t perfectly aligned politically, we are both technocratic, and I’d also say I’m a pragmatist.
What I really object to is people claiming that being technocratic, mild and practical mean that I’m secretly much further to the right than I claim to be, and that if I’m interrogated aggressively enough, that veil will be torn away.
I don’t support M4A, and I don’t think it’s particularly ideological. I sincerely think we should model our healthcare system on what works best globally, and I think the most successful models use a blend of private and public insurance.
I am generally pro-market, and so sure, that’s ideological and going to affect my assessment. But my ideology is not a secret, and it’s not the “real” reason I oppose M4A. My concerns are what I say they are: 99% practical.
I’m also not as left as Bruenig on every issue but I’m very left on healthcare and do support M4A. I do think there’s a tendency among center left types to not articulate their ideological/value based differences with leftists/socialists and ONLY highlight how they are more pragmatic or more aware of political constraints. I think leftists find it frustrating and it would be helpful to highlight the role of morality and ideology in the construction of policy design and not assume it’s all just a strategic/technocratic difference. Bruenig is wonky himself and therefore him arguing with center left types clarifies their ideological differences is all I was trying to say.
'One quirk of the Finnish study is that the government ran it and was therefore able to literally force the randomly selected individuals to participate in it. The studies Piper cites rely on volunteers, which technically makes them nonrandom, despite the fact that they are called “randomized control trials.”'
Congratulations, Matt Bruenig, for throwing out nearly all pharmaceutical research!
The "random" in RCT is, of course, that treatments given to any participant in the study are randomized, not that the population being studied is randomly selected from the actual total population of anyone who has the relevant disease. Medical RCTs have to recruit subjects just like these UBI studies did. Yes, RCTs can be questioned by having populations that are skewed in systematic ways (women's general non-inclusion in pharmaceutical trials for decades because what if those monthly hormones cause the medicine to not work, or what about the always potential baby that could appear!?), but Bruenig doesn't even gesture toward that kind of criticism, much less state it outright. Surely Bruenig has to know all this about RCTs, so why is he being so obstinate to attack these studies in such a transparently dumb manner.
this one also bothered me
i thought bruenig was writing in good faith up until i read that specific sentence... if he's going to do throwaway barbs at his debate opponent, fine, not everyone is cut out for SSC discussion norms. but that one... that was just a barb for the sake of hostility and nothing else. it made it clear that if bruenig saw an opportunity to demean his ideological opponents, he was going to take it, whether or not it made sense.
i was already kinda shocked at the implication that kelsey piper of all people might not have ever *considered* any of the ideas he laid out... but i thought it was possible, just barely possible, that bruenig wasn't very familiar with piper and was actually making the criticism in good faith
but after seeing that sentence you pointed out, i downgraded my assessment. he is performing marketing, not truthseeking.
I have a couple thoughts. I am very glad to see Matt Bruenig bring up the fact that our K-12 schools, in terms of "outcomes," do very well. I think what a lot of people have a problem with in some public schools are widespread disobedience and disorder that, while bad, don't actually make much of a difference in terms of learning. Kelsey Piper's argument that school reforms haven't caused improvements in learning doesn't actually contradict Bruenig's point. Schools can be good and also not improving.
Also, I really enjoyed Piper's original piece, but I feel like it and possibly this piece aren't really addressing the actual UBI question. To me, the question that many people have (myself included) isn't "does giving people some extra money make their life better?" It's this: "When many people will literally be unable to find work because AI and automation have taken everyone's job, will we be able to simply pay everyone what they need to live and still have a functioning society?" Of course, this is much harder to test out with experiments in the present.
Also, I see a lot of comments taking issue with Bruenig's tone, but... this feature is literally called "MAD libs." I think he's just following the assignment. (And I've followed him as a writer for a long time and I know he is passionate about this issue.) And to be honest, this publication needs to get off the ground, and if all they offer is dispassionate policy discussions based around fact-checking, no one is going to read it.
Many hugely successful substacks take this approach! SSC, Slow Boring and Silver Bulletin, off the top of my head.
Damn, this is what's going behind the paywall? I feel like this format is the central selling point of the argument. Sorry for being that guy lol
Yes, that is the general idea of selling points.
It’s crazy to me that nobody’s acknowledging that if the predicted big and dramatic benefits from giving people money are real, we wouldn’t have seen them yet. This is stuff like the earnings of cash recipients’ kids when they grow up, the criminality of cash recipients’ kids when they grow up, and the health of cash recipients’ kids when they grow up.
I’m definitely more skeptical that these benefits are going to materialize given the recent experiments failed to produce the predicted short-term benefits. But we still don’t really know!
Also: Matt is being way too hostile, and Kelsey is right to be disappointed…but Matt is right that simple poverty reduction is a good result.
Edit: Some shameless self-promo - I wrote a much more measured take on the recent experiments here: https://jg12402863.substack.com/p/giving-people-money-down-but-not (why should you care what I think? Honestly, no really good reason, but I did work pretty hard at it, and I was an RA for baby’s first years a few years ago)
In general I have a strong skepticism at this point of claims that long-term effects will manifest when short term effects do not. If a program like this is to have long term benefits, it has to be via immediate effects, and time is going to introduce noise and other variables. If we can't measure anything now I am very very skeptical we'd be able to measure anything later.
I also think that the main cases where you see lifetime effects on eg income or criminality, it's via literal brain damage - lead poisoning, in the developing world severe malnutrition, etc. Preventing that should of course be high among our priorities as a society, but targeted programs probably work best there.
I mostly agree, but want to flag that there are *some* intermediate effects we’ve measured - like increased healthcare consumption (as well as consumption of pretty much everything else).
Re criminality, reductions due to cash are predicted by quasi-experimental work on the Eastern Cherokee cash payments in North Carolina. Social scientists have previously used this to predict some pretty large effects of cash on crime. But I do have some skepticism about exogeneity in that work, especially after these recent null results.
Is everyone now in agreement that leaded gasoline caused the huge spike in crime in the 80s? If so, I’m curious, did it also correlate to a drop in national IQ?
In Piper's framing, the question isn't whether "simple poverty reduction is a good result," the question is whether that's the *best* result possible with the money going towards direct transfers. And to be fair, a large section of the concluding paragraphs of the original article is about exactly that.
Yeah, I agree that this is the question we should be asking! But as far as I’m aware, we don’t have many similarly strong RCTs showing better effects from in-kind transfers. (Though we should try some and find out!)
I definitely think these studies cast doubt on the most optimistic visions of cash benefits. Still, they have a lot of limitations, so I want to emphasize the uncertainty as you do. They’re not that much money in the grand scheme of things, especially for families with kids, and they were time-limited. Also, as Kelsey acknowledges, there might be community-wide benefits if everyone got a cash benefit.
Some (though certainty not most) economists predicted this way back before the results were released - Hoynes & Rothstein said the experiments would be good for getting better estimates of labor supply elasticity, but wouldn’t tell much about long-term/universal cash
As is typical, Bruenig’s rebuttal is ideologically driven (cash transfers must work in America because he WANTS them to work in America) and seemingly full of personal animus. She’s your colleague, man.
Different publication, same guy.
This piece was quite typical of Bruenig-- lots of big claims, but nothing resembling any evidence of thought or willingness (or ability) to grapple with the argument. His collection of words (I hesitate to call it a piece) amounts to an extended tantrum-- first, that it's self-evident that giving people cash makes their lives better (in the 9th grade, where Bruenig's intellectual development appears to have ended, we call this "assuming your conclusion"); second, that other studies find different results (let's name those studies and why they're methodologically better than the studies that Piper cited); and third, that we should just "be Finland." Bruenig writing embarrassing drivel would be fine if he weren't so incredibly smug and dismissive. I don't mind arrogance, per se; if you're very very smart, not suffering silliness is fine. Bruenig is not very very smart, or even a little smart. He's very much the opposite.
So unfortunately, the only takeaway I get from reading this is that Bruenig has nothing worthwhile to say. But that doesn't mean that there aren't good reasons to think that the studies might be missing or failing to measure something relevant, or whether something about the studied programs' design makes them ineffective. But the messenger for that will have to be someone with far far far more intellectual horsepower than Bruenig.
I am not sure to want extent I agree with Bruenig but I disagree with your point here; he's a good debater and his piece here is useful in pushing the cash assistance-skeptical side to be more specific about what their goals are, what standards they hold policies to, and what their view is as of which policies are and are not politically feasible. I assume that this was basically why he was hired (though would also be curious as to why he was interested in the job) and I think if the less lefty writers here learn to think with his critiques in mind they will be better off for it. I did find some the antagonistic tone a little unpleasant to read but hey, part of the point of this publication is to help liberals shed our reputation as pathologically agreeable.
That’s the thing— he’s not a good debater. He’s a very bad debater. The piece is very specific as to the standards to which it’s holding the policies— it’s a survey of studies that found that, broadly, unconditional cash transfers didn’t improve outcomes over a number of measures of well being relative to a control group. A smart critique would have suggested alternate measures that the studies might have missed. Or it would have suggested alternate program designs that might better improve outcomes on those measures.
He put together… a string of verbal diarrhea that amounted to “of course cash assistance is good, look at Finland.” To call this a bad argument would be too generous. It’s not an argument at a; it’s restating a conclusion.
Now, I don’t per se care that he was also an asshole. Lots of smart people are dismissive. It’s not that he’s an asshole and a leftist; it’s that he’s dismissive and an idiot. That’s a bad combination. Now, should we consider leftists’ critiques? Sure, if they’re thoughtful. But what he wrote wasn’t thoughtful. It was painfully stupid. And he was a smarmy asshole on top of that.
We should be open to all forms of thoughtful critique, from anywhere on the political spectrum. If something isn’t thoughtful but is courteous, pushing back courteously is good. If something isn’t thoughtful and is smarmy, well, it’s good to let that person know that if they can’t be smart, they should try not being a dick.
"Many people who would support the safety net if it were to succeed, oppose it precisely because they perceive it as failing"
This is such an odd assumption to make. By far the biggest opponents of the safety net are Republicans, and their disinterest in whether or not it actually works is palpable--you can tell because they make no effort whatsoever to confine their cuts to the things that (they believe) don't work, nor do they make any effort to bolster spending on the things they think do.
The real story in American anti-poverty policy is that a large share of conservative elites simply opposes spending money on poverty alleviation, on moral grounds. They believe that cutting taxes for the rich is morally better than spending money on helping the poor. This belief, not a principled dispute over program efficacy, is what drives opposition to the safety net. You're not going to convince these people to change their bedrock moral views simply by doing an RCT that finally convinces them that this one intervention really will work.
While I agree with Matt’s main argument in favor of basic income (the effect on class) and consider myself a big fan of such a program even with Kelsey’s piece in mind- I think he completely misread the original piece and simultaneously criticized it for a failing that his own critique suffers from more prominently. In arguing that the question of class stratification is solved by the Nordic model he ignores the obvious fact that there are other very consequential forms of social division that cash payments may not rectify! Kelsey rightly points out that in the United States, maybe housing policy has the potential to frustrate simple cash transfers, I agree. I can think of a few more reasons why one country or population might be different from another!
Matt beats whatever point he has so throughly against a straw man he’s left with useless nub.
It's a little unfortunate, in some ways, that the argument here ended up getting followed up immediately one twitter, but it did helpfully reveal that one key point of disagreement was how to define the marginal dollar: for Bruenig, you save it by reducing administrative costs via M4A, for Piper that's not possible so you look to incrementally increase funding to programs with greater records of success than cash benefits. I'd love to see a follow-up along those lines; I know Bruenig does not take a very high view of "what could we get passed" analysis, but it would be interesting to see each writer explain what their ideal welfare state would look like and, separately, what they think is best given current political realities.
It’s a good exchange, I think he lays out his case and where his values diverge from Piper’s pretty clearly. I happen to agree with Bruneig- and Matt Darling’s take that cash transfers are simply a better way to conduct transfers to the extent we do them (fragmented in kind benefits). It’s not magic, just an anti poverty program. Overall I think Bruneig is simply more correct about the purpose, outcomes, and goal of the welfare state.
Also piper mentions the progress since the war on poverty but I think she leaves out some good context; She mentions the SPM metric without mentioning that it is a quasi relative poverty metric. It remains the same (unanchored) because the share of people below 50% of the median income remains the same, even if the [bottom](https://www.vox.com/the-highlight/23383703/robert-greenstein-center-budget-policy-priorities) has increased. [transfers are more effective than they ever were at reducing poverty](https://www.cbpp.org/research/poverty-and-inequality/the-nation-has-made-progress-against-poverty-but-policy-advances). The effectiveness has grown as we have slowly expanded the welfare state over the decades, the poverty reduction matches the increase in spending.
The problem is America doesn’t have the level of welfare state as the more developed welfare states (eg no child allowance) is piper really so oblivious to make all these claims about how poverty is more complicated when the CTC halved child poverty instantly for 100B a year? And it doubled when it expired? I challenge her to find another proposal that would halve child poverty for a similar amount or less. that I don’t think she appreciates how much poverty is a structural issue caused by non workers (children, elderly, disabled, etc) half of parents wouldn’t be in poverty had they not had kids. The Welfare State is the only solution that can solve this problem. She mentions the growth of welfare state spending as a % of GDP but doesn’t compare it to our countries like the Nordics which, especially on children, is lacking.
Per the OECD: the US had a 21% child poverty rate in 2019. The Nordics had an average of half that or less throughout. The CTC brought it down to 14%(!) in 2021- which considering this is a relative measure and the US is at a higher baseline state of inequality is very impressive. No doubt if it increased family spending as a percent of GDP relative to the Nordic countries we’d see a further reduction.
https://www.oecd.org/en/data/indicators/poverty-rate.html?oecdcontrol-8027380c62-var3=2021
You write: "Piper is simply wrong that there has been no progress since the war on poverty."
Meanwhile, the real Piper writes: "You can see the Great Society on there — it did genuinely help! In 1969, 24% of people lived below the poverty line under this measure before transfers and 22.5% after them. In 2023, 22% lived below the poverty line before transfers and 11% after. Those are real gains."
You misrepresented her writing in such an egregious manner as to cast the rest of your comment in serious doubt.
Sorry I’ll edit it to say not enough progress, and I think she implies a stagnation in progress (later on in the quote you mention) relative to the growth in spending and my links I reference confirm that isn’t true. These programs are effective and they work and they’ve gotten more effective as the welfare state has expanded in fits and starts.
Poverty has been reduced as a result of transfers much more now than in the 70s (47% v 9%) this is in the CBPP source I linked. I also think her pivot to pre transfer income dodges the whole point of the discussion which is how can we best reduce poverty via government spending. Because if we want to do that we’d expand the circle to minimum wages and reviving organized labor and that is beyond the scope of the discussion.
I’d again encourage her to find a better program that costs less and reduces poverty more.
I’d say that the halving of post transfer poverty is a BFD. We are nowhere near the ceiling of what can be accomplished via transfers as the CTC expansion reminds us. Piper even acknowledge that the research supports the expansion and we didn’t see any deleterious effects like a reduction in labor supply
And everything I’ve said is backed up with direct empirical evidence, if I didn’t cite a figure I can find it for you.
I wasn’t intentionally trying to mischaracterize anyone- I was typing quickly on my phone and have gone back and edited it with more sources multiple times
Here’s a few:
US v OECD family spending
https://www.oecd.org/content/dam/oecd/en/data/datasets/family-database/pf1_1_public_spending_on_family_benefits.pdf
CTC studies on how the 2021 expansion affected the labor supply
1 https://www.nber.org/papers/w31110
2 https://www.nber.org/system/files/working_papers/w29823/w29823.pdf)
3 https://bfi.uchicago.edu/insights/child-tax-benefits-and-labor-supply-evidence-from-california/
Sort of a side point, but given GOP and Trumpian authoritarianism and social conservatism…is M4A actually a good thing in the US context? I work in healthcare so I see all the inefficiencies, and quite frankly the only way around them is for government to take responsibility. Part of that is hospital/insurer rents, and a big part of that is we are very overregulated compared to any existing socialized system (government run systems in general acknowledge trade offs and realize that rules have compliance costs and inefficiencies which sometimes aren’t worth it).
My concerns in the US contexts are (1) here the government is absolutely willing to overregulate itself, so I’m not sure the cost savings materialize to the extent we like (especially with doctors and nurses being part of the Democratic base and having coalitional power to keep extracting rents, especially if/when we unionize) and
(2) how would we prevent the GOP from just banning abortion, transgender care, vaccines etc when they are in power? Decentralization is inefficient but it might be protective given what we know about the actually existing US
Really enjoying this format, I definitely want to see more of it. Mostly convinced by Piper here - I think my cruxes are (a) you can't assume that policies that work in Finland work here, indeed that's the surprising part of these studies; and (b) there are real tradeoffs here that Piper is discussing and Bruenig seems to be ignoring. Bruenig's best part was the opening, but I think he drifts away from that core point and doesn't acknowledge that the studies in question are indeed mainly looking at non-work outcomes.
I would love to read a follow-up on schooling in the US and how we should think about the metrics on both sides (Bruenig - we do solidly on international assessments vs Piper - we're not close to the results we saw in Mississippi).
> Transfer payments to individuals have risen, in 2019 dollars, from $417 billion (in 2024 dollars) in 1969, or 4.8% of GDP, to $3.2 trillion in 2024 — or about 12% of GDP.
Is this in 2019 dollars or 2024 dollars? Not sure if it's a typo or just different sources, but it should be reconciled to the same standard. I think the point the same either way, it's just confusingly phrased.
“you can't assume that policies that work in Finland work here”
If only someone ran an experiment to see what would happen to child poverty of the US ran a Nordic style child allowance for a year
Oh wait they did and the studies show it halved child poverty and parents didn’t work less at all.
Right, that's one we tested and it worked well (and it would be great if we reinstated it). Less targeted cash transfers were tested here and (surprisingly) did not work well. Some Finnish policies succeed here, others don't seem to, so we can't assume that they all will.
I mean is it really targeted if the payout was practically to all parents, working or not? Seems more like a universal basic child allowance than a means tested targeted benefit like TANF or SNAP.
Yeah, it’s definitely less targeted than most similar programs. I think that led to a lot of the predictable gains in reducing overhead, both on the govt side and on the side of applying for benefits. That’s another big part of why the unconditional transfers not working so well was surprising to me too!
But if the UBI programs had positive effects on the scale that ECTC did, shouldn’t we see that in these studies? I’m not saying I’m totally confident in why we don’t, but it seems like that amount of targeting really does make a difference.
Maybe, but I’m not too sure. The full amount of $3,600 was given at 0$ and only begins to phase out at 112k, and then only phases out again after that at 200k to $0 at 400k a year
https://www.pgpf.org/article/what-to-know-about-the-new-monthly-child-tax-credit-payments/
Seems like the large majority of parents received the full or near full credit
So is targeting to parents the solution this combination of facts suggests?
It’s also plausible to me that one or more of these is true:
-the ECTC didn’t actually have significant effects beyond lowering child poverty - I haven’t done much lit review on it, it’s possible that the same thing happened as in these studies, i.e. lowering of ‘child poverty’ as a measure didn’t lead to broad knock-on improvements in other measures.
-both the ECTC and unconditional cash transfers do have broad positive effects beyond ‘poverty’ but they are diffuse and hard to measure
-some other part of the ECTC’s approach/conditions/targeting/timing made a bigger difference
Of course that’s not to say that ‘poverty’ is a bad thing to reduce, just that when I talk about poverty in general I mean a whole system of poor outcomes that are not just a lower amount of money. If the other outcomes (like health and housing and hunger) don’t change, I’d prioritize a program that affects those over one that affects ‘poverty’.