Zwolinski says that parents do not waste cash benefits for kids based on data showing that cash benefits reduce child poverty. But how is child poverty measured--based on the tangible things kids are receiving (food, clothes, etc.) or based on the reported income of their families? I don't know, but if it's the latter, then the data prove nothing about whether parents waste cash benefits for kids on things that don't benefit them.
I was going to ask the same question. As far as I can tell the US metric is absolute income based (with the threshold adjusted for family size and composition), making the argument tautological, and non-responsive to Caplan's argument.
Yeah, this guy is either dishonest or dim. He also cites America's allegedly "highest post-tax, post-transfer child poverty rate" and cites a source* that uses "relative poverty," meaning that the threshold of poverty is set lower in poorer countries, making it an apples to oranges comparison.
Yes, this second issue is even worse than the first. Getting the reported income part “wrong” imo is more likely an honest mistake (and do understand that there is no such data available on the point Caplan is asking about), and to me doesn’t necessarily make him dim.
But getting the poverty rate comparison wrong *is* imo either dishonest or dim.
Right?! Zwolinski left that way too ambiguous to use as an example to supports his thesis. With a hypothetical poverty line of $1200, a family with $1,000 income that spends $300 on their children is "in child poverty", but an identical family that receives $500 in UBI ($1500 total income) and still spends $300 on their children somehow isn't? Terrible metric...
Unless I'm misunderstanding something, it's willfully dishonest, and I wouldn't trust anything this person says about anything. I've lost my patience with academic frauds. They do serious damage to the country.
He included a snarky cartoon, making fun of anyone who would be concerned that cash benefits might be wasted. That didn't age well. He wrote an entire book on UBI, and some internet randos like me immediately spotted the problem in his argument.
To each his own, but I'm not inclined to a charitable interpretation. Purported conservatives and libertarians who say what progressives want to hear tend to get a pass on the rigor of their arguments, and it looks to me like he's been a beneficiary of this.
I don't really think it's really making fun of people. It's just making a point that has plausibility at least on its face in comic form. Maybe it offends some people, but people use thought experiments and memes that offend people all the time regardless of intention, and it's best not to put stock on that. There's nothing clearly insulting in the comic.
This is a great point! Less notable, but still worth noting, is that the author references:
> Fewer children living in poverty, along with all the short-term and long-term benefits that relief brings with it.
As Cremieux Recueil has repeatedly shown, though, almost all the commonly touted effects of poverty, are correlational, rather than the effects of actual causation. Lottery studies and within family studies bear this out. Particularly, whatever "long term" effects the writer has in mind likely don't exist.
The linked study claims to be measuring poverty based on pretax/transfer family income levels. In theory, that should have eliminated the direct effect of the CTC on the family's income. But we very much *do* want to measure the direct effect of the CTC on the poverty experienced by children; why would this study have seen any effect at all? I note that I am confused.
Right. Shifting goal posts too. The "no one is calling for" rhetoric is just embarrassing for someone presenting themselves as a serious intellectual doing serious analysis. Constructive discourse is an impossibility without setting precise details, definitions, and assumptions at the very start.
I don't think you're understanding his argument. The basic proposal is that you fund a UBI with a flat income tax. People with smaller incomes would be net-gainers, people with larger incomes would be net-losers. That's the sense in which it would both be a UBI and be means tested. Other possible sources of funding are possible, but all would be "means tested" in the sense that they would be net transfers from people with high wealth/income to people with low wealth/income.
Apart from the question of definition, I think you are suggesting that these two scenarios should be judged as equivalent:
1. Status quo
2. UBI, wherein the Govt takes much more of my income and then returns some of it.
Assuming NO DIFFERENCE in net dollars, I would still be strongly against #2, because it puts the Govt in the center of even more of my livelihood. After that, the obvious next step to ease welfare administration (the most important topic around which we should build our society) is for my company to send my income directly to the Govt, which will provide me an allowance to live on. UBI is another step on the slippery slope towards socialism, and those who feel that every step reduces human flourishing are right to resist it.
I agree that these two scenarios are not equivalent, but the status quo for most people is *already* that the government automatically takes income and then returns some of it. If you implement UBI as a negative income tax, for instance, the structure would be exactly the same as it currently is, but with a slightly different effective tax rate for most people.
If you don't like this, then your real beef is with the IRS rather than with a UBI (per se). I agree with that, actually: the IRS is deeply problematic.
If we take your argument literally, and the UBI replaces existing welfare spending (so no net tax increase in aggregate), then most of the rich would be “net winners” because they would be getting a handout from “the government” that they are not getting today, right?
And this is separate from the reality that a *switch* to a flat tax system would make most (not some of the crony capitalists) of the rich “net winners” as well.
But if your imagined “same structure” UBI is implemented as a *new* “negative income tax” layered on top of the existing taxes, then the cost of UBI is in fact massive.
You can’t have it both ways. Mega-details like this matter.
I’m no fan of the current IRS, but the structure/reality of the IRS has nothing to do with this discussion.
That's not the definition of "means-tested". If you make one up, then I agree it could comply with it. From GPT-4, given the neutral prompt "Define "means-tested" in the context of a discussion about government tax & benefits.":
> "Means-tested" refers to a method for determining eligibility for government tax relief or benefits based on an individual's or household's financial means, such as income and assets. The principle is to provide these benefits or tax relief primarily to those who need them most. In this context, the government sets thresholds for income and assets, below which individuals or families may qualify for assistance or tax benefits, and above which they would not. This approach targets resources towards those deemed to have insufficient means to support themselves without assistance.
There is no "determining eligibility" in the proposal or your defense of it, hence using the word means-tested is incorrect and deceptive.
You can quibble about definitions, but the *idea* behind "means-tested" is that the poor people who need the help should be the ones getting the benefit; it shouldn't be going to rich people who don't need it. Under any non-regressive funding program, a UBI would satisfy this requirement.
Also - one of the major touted benefits of UBI is the simplicity of administration / consumption and avoidance of benefit cliffs.
If means-testing is now part of the proposal, then it will have all of the drawbacks that it decries of the existing welfare state:
* bloated bureaucracy to administer
* low-income people who qualify but couldn't successfully navigate the process
* moral hazard / benefit cliffs / very high effective marginal tax rates
At that point, it's just adding yet another welfare state appendage, except one which provides the consumers with funds which can go to uses such as buying drugs, which is at least difficult under the current system of food & medical services provided.
If you implement a UBI as a negative income tax, then the administration would come through the IRS. IMHO, the IRS is terrible and could definitely be improved, but the administration wouldn't be much *worse* than they currently are (admittedly, it would mean more tax returns would be filed by people who don't work). Any other agencies which could be cut in the process (e.g. food stamps, Social Security) is where we could see the efficiency gains.
I agree with most all of this comment except the last point.
MONEY IS FUNGIBLE, so (medical services aside) it’s not at all difficult for current system welfare to effectively go to buying drugs. Let alone the obvious example of selling EBT cards.
Not that how a tiny number of drug addicts spend their welfare should be a deciding factor in how a proper welfare system is designed.
I think you're mistaken -- this isn't quibbling about the definition, it's the actual heart of the point he was trying to make. Here my paraphrasing of the argument:
Caplan: "UBI for everyone would be ruinously expensive. Taxes would have to rise hugely on everyone".
Zwolinksi: "No serious UBI proponent suggests giving money to everyone without means-testing. As a result your calculated expenditure is too high"
You seem to be agreeing with me that Zwolinksi is being deceptive or incorrect in his answering of Caplan's objection regarding the cost of UBI. Regardless of the definition of means-testing, if the point is that UBI goes to everyone, then I don't how Caplan's objection about the cost could be considered answered.
What's theoretical about the $3000 net cost in your example? You're talking about cost to the US Treasury here, which is affected just as much by a loss of $X of tax revenue as by an expenditure of the same amount.
“The basic proposal is that you fund a UBI with a flat income tax. People with smaller incomes would be net-gainers, people with larger incomes would be net-losers.”
Uh, sorry, but no. Or at least mostly no. With a *flat* income tax, everyone gets the same amount. It’s true that the amount matters less to the rich person, and it’s true that the rich person is *already* paying much more of the taxes to fund it. But unless your flat tax proposal to fund is on top of all existing taxes - back to Caplan’s issue about the cost - then it’s not really appropriate to claim that people with large incomes would be net-losers, even if it *might* be technically accurate. But of course, almost by the very definition of a redistributive state, the rich *already are* “net losers” today.
However, if you assume a *progressive* tax system - i.e. like the one we have now - then your comment about gainers versus losers is essentially accurate.
[Also, btw, in a truly flat tax system to get the desired net amount of income targeted to the poor, you need to gross the amount up for the taxes that will be collected on said UBI. This increases the headline “cost” of the program, and puts the poor back into the tax system (which for incentive reasons to me is a *good* thing), even if it doesn’t net cost more money. But who in the world thinks that the U.S. is going to switch to a pure flat tax system?!?]
It’s just strikes me as an absurd belief that a UBI would replace any meaningful part of the current welfare system. It just won’t, dude. All it will take is the tiniest visible subset of recipients using their cash transfer unwisely (for drugs or whatever), and we’ll have people insisting that we need a targeted program (ie the current welfare system) to help those people.
It will never, ever replace the current welfare system! This is as close to a self-evident truth as anything in social science.
I’ll consider a UBI if and only if the legislation that introduces it
1) Explicitly says that any additional government transfers outside the UBI are illegal, and
2) explicitly recognizes that some subset of UBI recipients will squander their money and will starve in the street barring private charity
And it’s worth noting that this failure mode doesn’t require a significant fraction of recipients to be addicts or otherwise pathologically irresponsible. Once everyone sees the government giving out food stamps and Medicaid to those who waste their UBI check, others will simply pocket the cash and go collect all the direct benefits anyway.
Unless we’re willing to let people who squander it die in the streets rather than give them traditional direct assistance, a UBI inevitably deteriorates into just yet another multi-trillion dollar welfare program on top of all the others.
It's ridiculous for someone to argue for something they are calling UBI which is neither U nor B!
Not U, because the literal definition requires "without a means test"!
Not B, because far too small on its own to "meet a person's basic needs".
Universal but small is "demogrant" / "citizen's dividend".
Means test and small and some incentive to work means "guaranteed minimum income", negative income tax, EITC, etc. Richard Nixon and Milton Friedman called saying that happened 55 years ago.
Hmmm... it is universal in the sense that EVERYONE would have the safety net of a guaranteed after-tax positive income should they become unable (or unwilling) to earn much money. But I think that’s actually, in many cases, a bad thing. People already suffer from the vice of laziness. And I don’t mean to moralize. I mean that many people would enjoy hard work if they got it, but are tempted to indolence and actually suffer for it. (Speaking from experience here, particularly my own haha.) Adding more temptation may do more harm than good.
You *are* moralizing. And moralizing is completely appropriate here. A big enough UBI to matter would indeed be a disincentive to work.
But in fairness... many of today's welfare programs have very high marginal tax rates on the poor. These are an even bigger disincentive to work. This is the nature of the so-called Welfare Trap. Many low-income people on welfare programs face *far* higher marginal tax rates than do the [rich / 1% / 0.1%].
Personally, I've generally lost interest in UBI discussions because the idea that they will in any actual implementation replace most/all welfare programs [even putting healthcare aside] is laughable, and if they do not do that replacement, they are too expensive in addition to their other demerits.
Unlike what at least one person argued above, "means testing" in fact usually means imposing very high marginal tax rates at the cutoff / phaseout points. These are a massive disincentive for work.
If somehow UBI *were* implemented to replace most other welfare programs, and done in or akin to the Negative Income Tax method, and it resulted in relatively *low* marginal tax rates for the poor, then - and only then - it might actually be worth considering.
I don't get it. Means tested UBI is self contradictory. He made it into a program to give $500 a month to poor people which isn't what UBI is typically about. We'd still have the benefit cliff, it's still blatantly a transfer, we still have the issue of the poor being wasteful (or worse) with cash payments, and on top of that $500 isn't enough to meaningfully cover living expenses, at least to the levels bleeding heart types would find tolerable - so it'll be layered on top of (or under) more programs.
The current system sucks, but cash payments to people who are likely to be substance abusers is guaranteed to make things so much worse.
Regarding the uncertainty of the costs of UBI it should be borne in mind that governmental programs have a ineluctable tendency to grow – both in extent and cost. Would the endowed bureaucracy of any UBI scheme not seek to remedy any residual 'deficiencies' on behalf of its beneficiaries?
1. The rationale for providing services instead of cash is strong -- if someone receives UBI but has no money and needs medical care, will society have the stomach to say no? That seems very implausible, in which case it's not an argument about which one, but one or both.
2. (a) Simplifying the currently-complex consumer experience of welfare and (b) avoiding benefit cliffs are the two things highlighted as to why welfare is insufficient. I don't see why both of those things can't be fixed for much less effort and change from status quo than introducing a UBI. Simplifying administration can be done with computer technology even without changes to laws, and avoiding benefit cliffs should have bipartisan support. I would naively think both of those things would have bipartisan support as improvements to quality-of-life of low-income folks and removing incentives to get back to work.
I find the above points strong arguments against pursuing a UBI.
Re: 1, I am *not* a proponent of UBI, but imo medical care [alone] can and should be kept in a separate bucket.
If in fact UBI replaced all other forms of government welfare, then it would solve 2 a) and 2 b) better than the in-situ changes you are proposing. Some of the cliffs are inherent in the separate nature of each distinct welfare program.
That said, I agree that the far stronger arguments against UBI are that in practice it will not be implemented as a replacement of all other [non-healthcare] welfare programs, and that as an add-on to existing welfare programs it is a completely crazy thing to do with far more downsides than any upsides which might exist.
Zwolinski's computation for costs of a UNIVERSAL basic income excluded some age groups, thereby resulting in the costs of basic income for some, but not all, U.S. citizens, which by definition means Zwolinski's basic income idea is NOT universal because it does not apply to all citizens.
The author (Zwolinski) said Bryan's 3.3 trillion dollar figure was absurdly high, and then went on to say that UBI was even more expensive. I found that annoying.
Bryan said: Simple cost estimates for a UBI are simply astronomical. $10,000 a person times 330M Americans is $3.3 trillion. That’s more than double what the U.S. will spend on Social Security in 2023.
Zwolinski says of course 3.3 trillion is an insane amount of money. With UBI you only cut checks to people whose income falls below a certain threshold. If the UBI was $1000 a month and included just people > 18 yrs and < 65 yrs it would cost $2.335 Tln. Using the costs for UBI including > 65 yr olds ($3.029 Tln) and < 18 yr olds ($3.326 Tln) I can compute the cost of $1000 UBI for old people (3.029 - 2.335 = $ 0.694 Tln cost of UBI for > 65 yr olds) and kids (3.326 - 2.335 = $ 0.991 Tln cost of UBI for < 18 yr olds). Computing the cost of a universal, i.e., for everyone, $1000 per month income requires summing the cost of a $1000 per month income for each of the age groups Zwolinski made:
$ cost of UBI using Zwolinski's numbers = $0.991 Tln (kids) + $0.694 Tln (old) + $2.335 Tln (else) = $4.02 Tln.
$4.02 Tln is more than the $3.3 Tln dollars Bryan mentions. Which makes sense because $10_000 per year is less than 12 * 1000 = $12_000 per year. It follows that the costs of a UBI than Bryan mentioned are not absure because they are lower than the costs Zwolinski mentions and still around twice the amount the U.S. spent on social security in 2023, which we can deduce is around $1.65 Tln = 3.3 / 2.
At a high level, one of the problems I have with UBI is that if it did result in a net transfer of money from those who are means-tested out of receiving any to those who get UBI, that is a reduction of savings and an increase in consumption for the country. As it is we save very little, so reducing savings even more will further reduce money available for the kind of business investment we need to keep productivity, and therefore living standards, rising. Even if UBI turned out to be effective at raising the income and consumption levels of the poorest today, it will hurt their (and everyone else's) incomes in the future. A less prosperous society will have more problems than a richer one.
That isn't to say that only UBI stands in the way of rising living standards. We waste massive amounts on so many things that we pretend are "investments" that have zero chance of bringing any return on investment. We have a national policy of diverting business investment and R&D away from things that will enhance productivity toward what will produce the same things as before more expensively, only they will give off less CO2. So UBI isn't some kind of unique culprit. There is a lot we should change first before adding this transfer from savings to consumption on top of everything else.
I think inflation is a risk too. I keep thinking about the COVID stimulus--back-of-the-envelope calculation: Suppose a family of four at about poverty line ($30,000) got stimulus check ($3,340), but the stimulus program induced inflation (+13% over 2 years). When all’s said and done, the family’s income went up 11% (but for only a year or two), but their income moving forward loses about 12% of its value, indefinitely. Money isn’t wealth, and inducing people to buy more and work less by just giving them other people’s money might literally improve their standard of living not at all, with bad luck.
"But we don’t need to speculate about what the overall results of increased cash transfers would be. We know. Fewer children living in poverty, along with all the short-term and long-term benefits that relief brings with it."
I'm confused by this. As far as I can tell, his argument is something along the lines of "people given more money will have more money to spend". Is that not obvious? Or does the phrase "poverty" explicitly mean something beyond having more money?
As I understand it the phrase "along with all the short-term and long-term benefits that relief brings with it" was left as implicit. Do we know what those benefits are? Is it not feasible that there are at least some negatives to paying people to just sit around and not be economically productive?
One major negative of per-child cash subsidies for low income parents is the dysgenic effect of incentivising child-bearing by the least competent, least productive adults. What's needed instead is incentives for child-bearing by the most competent, who have been reproducing at a rate well below replacement level in recent decades.
You have changed subjects, but this particular problem you cite is easily addressed by having child subsidies be a combination of a smaller tax *credit* and a larger tax *deduction*.
This wouldn’t work as well in a 100% pure flat-tax only system, but we ain’t ever gonna be there (and imo would be a great problem to have if somehow we were)…
That'd be an improvement over what we've got now but I doubt it would suffice to raise the mean lifetime childbearing of highly intelligent women up to, let alone above, replacement level, as the main disincentives for them are cultural rather than financial.
I didn't suggest it would suffice alone to solve the fertility problem. Though I don’t agree with you that the financial disincentives are unimportant, even if I do agree that the cultural disincentives are quite important.
If this is a problem that interests you, I suggest you read Robin Hanson's substack.
Although I think that the low "fertility" of highly intelligent women is mainly due to cultural disincentives that's not to say that financial disincentives are unimportant.
Dysgenic fertility does interest me; indeed, it wouldn't be stretching a point to say that it preys on my mind. So thanks for that tip.
Why do we have separate programs for food stamps and free school lunches? If we could trust parents to not waste cash benefits, we could increase payments by $60 per month and eliminate the federal and state bureaucracies that manage the free lunch programs and save money.
Incidentally, school lunches are wildly wasteful too (speaking as a former teacher in a very poor public school in Meridian, Mississippi). Roughly half the lunches in my school were literally thrown directly in the trash. They didn’t taste bad either. I would treat myself sometimes by buying the school lunch instead of packing a lunch. My school required (unofficially, I think) all students to take a lunch even if they didn’t want it so that they wouldn’t lose funding for the cafeteria staff.
I have an even better idea. Why not just make booze and fentanyl free, since the parents will then spend all their savings on the kids? Win win!
Seriously though, this is really deja vu on the old BHL blog. The authors would make bad arguments, the audience would tear it apart, and the authors would neither respond or revise their silly arguments. Honestly I read the old blog for the comments, which were often excellent.
The problem with a UBI that I think is insurmountable is to make anything meaningful work, you're going to have to get through a program that replaces much more generous SS and Medicare payments with a lower, cash UBI. I'd be fine with that, but until we have something approaching a societal collapse, I don't think that's going to be a political winner.
This sounds reasonable. However as an OG member of the Yang Gang, Andrew 100% absolutely did run on the proposition of giving $1,000 to EVERY American per month
I also have a problem the use of those poverty numbers which do not measure if a child lives in poverty at all.
I'd also like to say, I support a $260/week UBI given to all US citizens over 18 years old but it would replace TANF, SNAP, minimum wage laws, housing subsidies etc. and yes Social Security also. There would no longer be a reason to have those programs if we had a UBI. And I think giving more money to parents for their children would dangerous in that a very few but not 0 might have children that they neglect to get more money and that would look bad for the UBI and in fact be bad. Children are already subsidized through schools and free lunch programs and charities like to help children and the UBI would make the problem of child poverty small enough that charity could cover it. I would also start taxing the 1st dollar earned at 33% instead of 15.2%. The tax rate/bracket could lowered once income goes about enough to pay for the UBI ($40,969.70/year).
The other merits and demerits of UBI completely aside, if you were to replace SS with a $13,500 annual UBI payment, as you propose, that would be a MAJOR wealth redistribution away from middle and upper middle class retirees the value of whose SS contributions is anywhere from 1.5x to 4x that amount.
To say nothing of the “fairness” issue of yanking those larger SS payments (vs, say, trimming them, which I for one have no per se problem with) from retirees and near retirees who have organized their lives around same.
Well, I don’t agree with UBI for (many) other reasons, but that method of implementing the SS transition would be far less horrible, agreed. At least until inflation goes way up again due to government and Fed policy like we had in 2020-2022.
Of course, current widespread public support for the SS taxes needed to fund the Ponzi scheme that is SS (absent a strong economy and a continuous increase in payers into the system) would instantly dry up now that it is clear that SS is just an additional pure income tax on top of existing income taxes, as under your plan at least 70%, and perhaps as high as 85%, of payees will be net worse off over their lifetimes.
>Matt recently published a critique of my response to Chris Freiman on the UBI, and he’s has kindly agreed to let me cross-post his critique here. Enjoy!
The links in this sentence are wrong. The link to "my response" goes to an article from 6 years ago that isn't in response to Freiman. And the link to Freiman goes to a book, that isn't even by Freiman but by Zwolinski.
Zwolinski says that parents do not waste cash benefits for kids based on data showing that cash benefits reduce child poverty. But how is child poverty measured--based on the tangible things kids are receiving (food, clothes, etc.) or based on the reported income of their families? I don't know, but if it's the latter, then the data prove nothing about whether parents waste cash benefits for kids on things that don't benefit them.
At least according to this information, child poverty is measured based on family income: https://www.childtrends.org/publications/lessons-from-a-historic-decline-in-child-poverty-how-poverty-is-measured-in-the-united-states
If that's correct, then the writer is extraordinarily foolish or dishonest.
I was going to ask the same question. As far as I can tell the US metric is absolute income based (with the threshold adjusted for family size and composition), making the argument tautological, and non-responsive to Caplan's argument.
You are very charitable. I have much less charitable adjectives for an argument like that, and someone who would make it.
Yeah, this guy is either dishonest or dim. He also cites America's allegedly "highest post-tax, post-transfer child poverty rate" and cites a source* that uses "relative poverty," meaning that the threshold of poverty is set lower in poorer countries, making it an apples to oranges comparison.
*https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6087662/#:~:text=In%20the%20selected%20set%20of%20countries%20relative%20child%20poverty%20ranges,1).)
Yes, this second issue is even worse than the first. Getting the reported income part “wrong” imo is more likely an honest mistake (and do understand that there is no such data available on the point Caplan is asking about), and to me doesn’t necessarily make him dim.
But getting the poverty rate comparison wrong *is* imo either dishonest or dim.
Right?! Zwolinski left that way too ambiguous to use as an example to supports his thesis. With a hypothetical poverty line of $1200, a family with $1,000 income that spends $300 on their children is "in child poverty", but an identical family that receives $500 in UBI ($1500 total income) and still spends $300 on their children somehow isn't? Terrible metric...
Unless I'm misunderstanding something, it's willfully dishonest, and I wouldn't trust anything this person says about anything. I've lost my patience with academic frauds. They do serious damage to the country.
I've read several things by him, and Zwollinski is good. This is an overreaction.
He included a snarky cartoon, making fun of anyone who would be concerned that cash benefits might be wasted. That didn't age well. He wrote an entire book on UBI, and some internet randos like me immediately spotted the problem in his argument.
To each his own, but I'm not inclined to a charitable interpretation. Purported conservatives and libertarians who say what progressives want to hear tend to get a pass on the rigor of their arguments, and it looks to me like he's been a beneficiary of this.
I don't really think it's really making fun of people. It's just making a point that has plausibility at least on its face in comic form. Maybe it offends some people, but people use thought experiments and memes that offend people all the time regardless of intention, and it's best not to put stock on that. There's nothing clearly insulting in the comic.
Do you highly recommend Zwolinski? He’s been on my reading list for a while. Should I make the time to read him?
For sure. I've enjoyed the stuff on exploitation and sweatshops that he's written on.
Yeah, my jaw dropped when I read that foolish argument. Like, seriously?! Needs a retraction or clarification--badly.
It’s bad form to assume dishonesty when incompetence can also explain it.
And usually it is *very* difficult to discern the one from the other.
I'm sure it's an honest mistake
This is a great point! Less notable, but still worth noting, is that the author references:
> Fewer children living in poverty, along with all the short-term and long-term benefits that relief brings with it.
As Cremieux Recueil has repeatedly shown, though, almost all the commonly touted effects of poverty, are correlational, rather than the effects of actual causation. Lottery studies and within family studies bear this out. Particularly, whatever "long term" effects the writer has in mind likely don't exist.
The linked study claims to be measuring poverty based on pretax/transfer family income levels. In theory, that should have eliminated the direct effect of the CTC on the family's income. But we very much *do* want to measure the direct effect of the CTC on the poverty experienced by children; why would this study have seen any effect at all? I note that I am confused.
"No serious UBI proponent suggests giving money to everyone without means-testing"
Feels like motte-and-bailey. I haven't seen means-testing to be a common part of UBI proposals and it obviously violates the name itself.
Right. Shifting goal posts too. The "no one is calling for" rhetoric is just embarrassing for someone presenting themselves as a serious intellectual doing serious analysis. Constructive discourse is an impossibility without setting precise details, definitions, and assumptions at the very start.
I don't think you're understanding his argument. The basic proposal is that you fund a UBI with a flat income tax. People with smaller incomes would be net-gainers, people with larger incomes would be net-losers. That's the sense in which it would both be a UBI and be means tested. Other possible sources of funding are possible, but all would be "means tested" in the sense that they would be net transfers from people with high wealth/income to people with low wealth/income.
Apart from the question of definition, I think you are suggesting that these two scenarios should be judged as equivalent:
1. Status quo
2. UBI, wherein the Govt takes much more of my income and then returns some of it.
Assuming NO DIFFERENCE in net dollars, I would still be strongly against #2, because it puts the Govt in the center of even more of my livelihood. After that, the obvious next step to ease welfare administration (the most important topic around which we should build our society) is for my company to send my income directly to the Govt, which will provide me an allowance to live on. UBI is another step on the slippery slope towards socialism, and those who feel that every step reduces human flourishing are right to resist it.
I agree that these two scenarios are not equivalent, but the status quo for most people is *already* that the government automatically takes income and then returns some of it. If you implement UBI as a negative income tax, for instance, the structure would be exactly the same as it currently is, but with a slightly different effective tax rate for most people.
If you don't like this, then your real beef is with the IRS rather than with a UBI (per se). I agree with that, actually: the IRS is deeply problematic.
If we take your argument literally, and the UBI replaces existing welfare spending (so no net tax increase in aggregate), then most of the rich would be “net winners” because they would be getting a handout from “the government” that they are not getting today, right?
And this is separate from the reality that a *switch* to a flat tax system would make most (not some of the crony capitalists) of the rich “net winners” as well.
But if your imagined “same structure” UBI is implemented as a *new* “negative income tax” layered on top of the existing taxes, then the cost of UBI is in fact massive.
You can’t have it both ways. Mega-details like this matter.
I’m no fan of the current IRS, but the structure/reality of the IRS has nothing to do with this discussion.
That's not the definition of "means-tested". If you make one up, then I agree it could comply with it. From GPT-4, given the neutral prompt "Define "means-tested" in the context of a discussion about government tax & benefits.":
> "Means-tested" refers to a method for determining eligibility for government tax relief or benefits based on an individual's or household's financial means, such as income and assets. The principle is to provide these benefits or tax relief primarily to those who need them most. In this context, the government sets thresholds for income and assets, below which individuals or families may qualify for assistance or tax benefits, and above which they would not. This approach targets resources towards those deemed to have insufficient means to support themselves without assistance.
There is no "determining eligibility" in the proposal or your defense of it, hence using the word means-tested is incorrect and deceptive.
You can quibble about definitions, but the *idea* behind "means-tested" is that the poor people who need the help should be the ones getting the benefit; it shouldn't be going to rich people who don't need it. Under any non-regressive funding program, a UBI would satisfy this requirement.
Also - one of the major touted benefits of UBI is the simplicity of administration / consumption and avoidance of benefit cliffs.
If means-testing is now part of the proposal, then it will have all of the drawbacks that it decries of the existing welfare state:
* bloated bureaucracy to administer
* low-income people who qualify but couldn't successfully navigate the process
* moral hazard / benefit cliffs / very high effective marginal tax rates
At that point, it's just adding yet another welfare state appendage, except one which provides the consumers with funds which can go to uses such as buying drugs, which is at least difficult under the current system of food & medical services provided.
If you implement a UBI as a negative income tax, then the administration would come through the IRS. IMHO, the IRS is terrible and could definitely be improved, but the administration wouldn't be much *worse* than they currently are (admittedly, it would mean more tax returns would be filed by people who don't work). Any other agencies which could be cut in the process (e.g. food stamps, Social Security) is where we could see the efficiency gains.
I agree with most all of this comment except the last point.
MONEY IS FUNGIBLE, so (medical services aside) it’s not at all difficult for current system welfare to effectively go to buying drugs. Let alone the obvious example of selling EBT cards.
Not that how a tiny number of drug addicts spend their welfare should be a deciding factor in how a proper welfare system is designed.
I think you're mistaken -- this isn't quibbling about the definition, it's the actual heart of the point he was trying to make. Here my paraphrasing of the argument:
Caplan: "UBI for everyone would be ruinously expensive. Taxes would have to rise hugely on everyone".
Zwolinksi: "No serious UBI proponent suggests giving money to everyone without means-testing. As a result your calculated expenditure is too high"
You seem to be agreeing with me that Zwolinksi is being deceptive or incorrect in his answering of Caplan's objection regarding the cost of UBI. Regardless of the definition of means-testing, if the point is that UBI goes to everyone, then I don't how Caplan's objection about the cost could be considered answered.
Zwolinski is making the point that the theoretical cost of a UBI is not the same as it's actual cost (in terms of the size of the transfers).
Simple example with a negative income tax and $1000 UBI:
Person A owes $0 in taxes: receives $1000 on tax day.
Person B owes $1000 in taxes: pays $0 on tax day.
Person C owes $2000 in taxes: pays $1000 on tax day.
Theoretical cost of UBI: $3000, Actual size of transfers: $1000
This is also means-tested in the sense that the people with lower income are the one who receive the net benefit.
What's theoretical about the $3000 net cost in your example? You're talking about cost to the US Treasury here, which is affected just as much by a loss of $X of tax revenue as by an expenditure of the same amount.
“The basic proposal is that you fund a UBI with a flat income tax. People with smaller incomes would be net-gainers, people with larger incomes would be net-losers.”
Uh, sorry, but no. Or at least mostly no. With a *flat* income tax, everyone gets the same amount. It’s true that the amount matters less to the rich person, and it’s true that the rich person is *already* paying much more of the taxes to fund it. But unless your flat tax proposal to fund is on top of all existing taxes - back to Caplan’s issue about the cost - then it’s not really appropriate to claim that people with large incomes would be net-losers, even if it *might* be technically accurate. But of course, almost by the very definition of a redistributive state, the rich *already are* “net losers” today.
However, if you assume a *progressive* tax system - i.e. like the one we have now - then your comment about gainers versus losers is essentially accurate.
[Also, btw, in a truly flat tax system to get the desired net amount of income targeted to the poor, you need to gross the amount up for the taxes that will be collected on said UBI. This increases the headline “cost” of the program, and puts the poor back into the tax system (which for incentive reasons to me is a *good* thing), even if it doesn’t net cost more money. But who in the world thinks that the U.S. is going to switch to a pure flat tax system?!?]
It’s just strikes me as an absurd belief that a UBI would replace any meaningful part of the current welfare system. It just won’t, dude. All it will take is the tiniest visible subset of recipients using their cash transfer unwisely (for drugs or whatever), and we’ll have people insisting that we need a targeted program (ie the current welfare system) to help those people.
It will never, ever replace the current welfare system! This is as close to a self-evident truth as anything in social science.
I’ll consider a UBI if and only if the legislation that introduces it
1) Explicitly says that any additional government transfers outside the UBI are illegal, and
2) explicitly recognizes that some subset of UBI recipients will squander their money and will starve in the street barring private charity
And it’s worth noting that this failure mode doesn’t require a significant fraction of recipients to be addicts or otherwise pathologically irresponsible. Once everyone sees the government giving out food stamps and Medicaid to those who waste their UBI check, others will simply pocket the cash and go collect all the direct benefits anyway.
Unless we’re willing to let people who squander it die in the streets rather than give them traditional direct assistance, a UBI inevitably deteriorates into just yet another multi-trillion dollar welfare program on top of all the others.
Let's be honest. If it's means-tested, it's not universal.
It's ridiculous for someone to argue for something they are calling UBI which is neither U nor B!
Not U, because the literal definition requires "without a means test"!
Not B, because far too small on its own to "meet a person's basic needs".
Universal but small is "demogrant" / "citizen's dividend".
Means test and small and some incentive to work means "guaranteed minimum income", negative income tax, EITC, etc. Richard Nixon and Milton Friedman called saying that happened 55 years ago.
Hmmm... it is universal in the sense that EVERYONE would have the safety net of a guaranteed after-tax positive income should they become unable (or unwilling) to earn much money. But I think that’s actually, in many cases, a bad thing. People already suffer from the vice of laziness. And I don’t mean to moralize. I mean that many people would enjoy hard work if they got it, but are tempted to indolence and actually suffer for it. (Speaking from experience here, particularly my own haha.) Adding more temptation may do more harm than good.
You *are* moralizing. And moralizing is completely appropriate here. A big enough UBI to matter would indeed be a disincentive to work.
But in fairness... many of today's welfare programs have very high marginal tax rates on the poor. These are an even bigger disincentive to work. This is the nature of the so-called Welfare Trap. Many low-income people on welfare programs face *far* higher marginal tax rates than do the [rich / 1% / 0.1%].
Personally, I've generally lost interest in UBI discussions because the idea that they will in any actual implementation replace most/all welfare programs [even putting healthcare aside] is laughable, and if they do not do that replacement, they are too expensive in addition to their other demerits.
Unlike what at least one person argued above, "means testing" in fact usually means imposing very high marginal tax rates at the cutoff / phaseout points. These are a massive disincentive for work.
If somehow UBI *were* implemented to replace most other welfare programs, and done in or akin to the Negative Income Tax method, and it resulted in relatively *low* marginal tax rates for the poor, then - and only then - it might actually be worth considering.
I don't get it. Means tested UBI is self contradictory. He made it into a program to give $500 a month to poor people which isn't what UBI is typically about. We'd still have the benefit cliff, it's still blatantly a transfer, we still have the issue of the poor being wasteful (or worse) with cash payments, and on top of that $500 isn't enough to meaningfully cover living expenses, at least to the levels bleeding heart types would find tolerable - so it'll be layered on top of (or under) more programs.
The current system sucks, but cash payments to people who are likely to be substance abusers is guaranteed to make things so much worse.
Regarding the uncertainty of the costs of UBI it should be borne in mind that governmental programs have a ineluctable tendency to grow – both in extent and cost. Would the endowed bureaucracy of any UBI scheme not seek to remedy any residual 'deficiencies' on behalf of its beneficiaries?
UBI vs Welfare state - 2 points
1. The rationale for providing services instead of cash is strong -- if someone receives UBI but has no money and needs medical care, will society have the stomach to say no? That seems very implausible, in which case it's not an argument about which one, but one or both.
2. (a) Simplifying the currently-complex consumer experience of welfare and (b) avoiding benefit cliffs are the two things highlighted as to why welfare is insufficient. I don't see why both of those things can't be fixed for much less effort and change from status quo than introducing a UBI. Simplifying administration can be done with computer technology even without changes to laws, and avoiding benefit cliffs should have bipartisan support. I would naively think both of those things would have bipartisan support as improvements to quality-of-life of low-income folks and removing incentives to get back to work.
I find the above points strong arguments against pursuing a UBI.
Re: 1, I am *not* a proponent of UBI, but imo medical care [alone] can and should be kept in a separate bucket.
If in fact UBI replaced all other forms of government welfare, then it would solve 2 a) and 2 b) better than the in-situ changes you are proposing. Some of the cliffs are inherent in the separate nature of each distinct welfare program.
That said, I agree that the far stronger arguments against UBI are that in practice it will not be implemented as a replacement of all other [non-healthcare] welfare programs, and that as an add-on to existing welfare programs it is a completely crazy thing to do with far more downsides than any upsides which might exist.
Zwolinski's computation for costs of a UNIVERSAL basic income excluded some age groups, thereby resulting in the costs of basic income for some, but not all, U.S. citizens, which by definition means Zwolinski's basic income idea is NOT universal because it does not apply to all citizens.
The author (Zwolinski) said Bryan's 3.3 trillion dollar figure was absurdly high, and then went on to say that UBI was even more expensive. I found that annoying.
Bryan said: Simple cost estimates for a UBI are simply astronomical. $10,000 a person times 330M Americans is $3.3 trillion. That’s more than double what the U.S. will spend on Social Security in 2023.
Zwolinski says of course 3.3 trillion is an insane amount of money. With UBI you only cut checks to people whose income falls below a certain threshold. If the UBI was $1000 a month and included just people > 18 yrs and < 65 yrs it would cost $2.335 Tln. Using the costs for UBI including > 65 yr olds ($3.029 Tln) and < 18 yr olds ($3.326 Tln) I can compute the cost of $1000 UBI for old people (3.029 - 2.335 = $ 0.694 Tln cost of UBI for > 65 yr olds) and kids (3.326 - 2.335 = $ 0.991 Tln cost of UBI for < 18 yr olds). Computing the cost of a universal, i.e., for everyone, $1000 per month income requires summing the cost of a $1000 per month income for each of the age groups Zwolinski made:
$ cost of UBI using Zwolinski's numbers = $0.991 Tln (kids) + $0.694 Tln (old) + $2.335 Tln (else) = $4.02 Tln.
$4.02 Tln is more than the $3.3 Tln dollars Bryan mentions. Which makes sense because $10_000 per year is less than 12 * 1000 = $12_000 per year. It follows that the costs of a UBI than Bryan mentioned are not absure because they are lower than the costs Zwolinski mentions and still around twice the amount the U.S. spent on social security in 2023, which we can deduce is around $1.65 Tln = 3.3 / 2.
At a high level, one of the problems I have with UBI is that if it did result in a net transfer of money from those who are means-tested out of receiving any to those who get UBI, that is a reduction of savings and an increase in consumption for the country. As it is we save very little, so reducing savings even more will further reduce money available for the kind of business investment we need to keep productivity, and therefore living standards, rising. Even if UBI turned out to be effective at raising the income and consumption levels of the poorest today, it will hurt their (and everyone else's) incomes in the future. A less prosperous society will have more problems than a richer one.
That isn't to say that only UBI stands in the way of rising living standards. We waste massive amounts on so many things that we pretend are "investments" that have zero chance of bringing any return on investment. We have a national policy of diverting business investment and R&D away from things that will enhance productivity toward what will produce the same things as before more expensively, only they will give off less CO2. So UBI isn't some kind of unique culprit. There is a lot we should change first before adding this transfer from savings to consumption on top of everything else.
I think inflation is a risk too. I keep thinking about the COVID stimulus--back-of-the-envelope calculation: Suppose a family of four at about poverty line ($30,000) got stimulus check ($3,340), but the stimulus program induced inflation (+13% over 2 years). When all’s said and done, the family’s income went up 11% (but for only a year or two), but their income moving forward loses about 12% of its value, indefinitely. Money isn’t wealth, and inducing people to buy more and work less by just giving them other people’s money might literally improve their standard of living not at all, with bad luck.
Regarding the political feasibility of UBI, check out this case study of Nixon's "negative income tax" proposal, which was enacted by the House but died in the Senate: https://digitalcommons.chapman.edu/chapman-law-review/vol23/iss1/3/
"But we don’t need to speculate about what the overall results of increased cash transfers would be. We know. Fewer children living in poverty, along with all the short-term and long-term benefits that relief brings with it."
I'm confused by this. As far as I can tell, his argument is something along the lines of "people given more money will have more money to spend". Is that not obvious? Or does the phrase "poverty" explicitly mean something beyond having more money?
As I understand it the phrase "along with all the short-term and long-term benefits that relief brings with it" was left as implicit. Do we know what those benefits are? Is it not feasible that there are at least some negatives to paying people to just sit around and not be economically productive?
One major negative of per-child cash subsidies for low income parents is the dysgenic effect of incentivising child-bearing by the least competent, least productive adults. What's needed instead is incentives for child-bearing by the most competent, who have been reproducing at a rate well below replacement level in recent decades.
You have changed subjects, but this particular problem you cite is easily addressed by having child subsidies be a combination of a smaller tax *credit* and a larger tax *deduction*.
This wouldn’t work as well in a 100% pure flat-tax only system, but we ain’t ever gonna be there (and imo would be a great problem to have if somehow we were)…
That'd be an improvement over what we've got now but I doubt it would suffice to raise the mean lifetime childbearing of highly intelligent women up to, let alone above, replacement level, as the main disincentives for them are cultural rather than financial.
I didn't suggest it would suffice alone to solve the fertility problem. Though I don’t agree with you that the financial disincentives are unimportant, even if I do agree that the cultural disincentives are quite important.
If this is a problem that interests you, I suggest you read Robin Hanson's substack.
Although I think that the low "fertility" of highly intelligent women is mainly due to cultural disincentives that's not to say that financial disincentives are unimportant.
Dysgenic fertility does interest me; indeed, it wouldn't be stretching a point to say that it preys on my mind. So thanks for that tip.
Why do we have separate programs for food stamps and free school lunches? If we could trust parents to not waste cash benefits, we could increase payments by $60 per month and eliminate the federal and state bureaucracies that manage the free lunch programs and save money.
Incidentally, school lunches are wildly wasteful too (speaking as a former teacher in a very poor public school in Meridian, Mississippi). Roughly half the lunches in my school were literally thrown directly in the trash. They didn’t taste bad either. I would treat myself sometimes by buying the school lunch instead of packing a lunch. My school required (unofficially, I think) all students to take a lunch even if they didn’t want it so that they wouldn’t lose funding for the cafeteria staff.
I have an even better idea. Why not just make booze and fentanyl free, since the parents will then spend all their savings on the kids? Win win!
Seriously though, this is really deja vu on the old BHL blog. The authors would make bad arguments, the audience would tear it apart, and the authors would neither respond or revise their silly arguments. Honestly I read the old blog for the comments, which were often excellent.
The problem with a UBI that I think is insurmountable is to make anything meaningful work, you're going to have to get through a program that replaces much more generous SS and Medicare payments with a lower, cash UBI. I'd be fine with that, but until we have something approaching a societal collapse, I don't think that's going to be a political winner.
This sounds reasonable. However as an OG member of the Yang Gang, Andrew 100% absolutely did run on the proposition of giving $1,000 to EVERY American per month
2,500 years ago, Aristides made the proposition that a silver find should be distributed equally to EVERY Athenian. Ended up being spent on warships.
I also have a problem the use of those poverty numbers which do not measure if a child lives in poverty at all.
I'd also like to say, I support a $260/week UBI given to all US citizens over 18 years old but it would replace TANF, SNAP, minimum wage laws, housing subsidies etc. and yes Social Security also. There would no longer be a reason to have those programs if we had a UBI. And I think giving more money to parents for their children would dangerous in that a very few but not 0 might have children that they neglect to get more money and that would look bad for the UBI and in fact be bad. Children are already subsidized through schools and free lunch programs and charities like to help children and the UBI would make the problem of child poverty small enough that charity could cover it. I would also start taxing the 1st dollar earned at 33% instead of 15.2%. The tax rate/bracket could lowered once income goes about enough to pay for the UBI ($40,969.70/year).
The other merits and demerits of UBI completely aside, if you were to replace SS with a $13,500 annual UBI payment, as you propose, that would be a MAJOR wealth redistribution away from middle and upper middle class retirees the value of whose SS contributions is anywhere from 1.5x to 4x that amount.
To say nothing of the “fairness” issue of yanking those larger SS payments (vs, say, trimming them, which I for one have no per se problem with) from retirees and near retirees who have organized their lives around same.
For SS recipients, I'd phase it in slowly. IE not do the cost of living increases for the higher amounts until everyone gets just the UBI amount.
Well, I don’t agree with UBI for (many) other reasons, but that method of implementing the SS transition would be far less horrible, agreed. At least until inflation goes way up again due to government and Fed policy like we had in 2020-2022.
Of course, current widespread public support for the SS taxes needed to fund the Ponzi scheme that is SS (absent a strong economy and a continuous increase in payers into the system) would instantly dry up now that it is clear that SS is just an additional pure income tax on top of existing income taxes, as under your plan at least 70%, and perhaps as high as 85%, of payees will be net worse off over their lifetimes.
>Matt recently published a critique of my response to Chris Freiman on the UBI, and he’s has kindly agreed to let me cross-post his critique here. Enjoy!
The links in this sentence are wrong. The link to "my response" goes to an article from 6 years ago that isn't in response to Freiman. And the link to Freiman goes to a book, that isn't even by Freiman but by Zwolinski.