In a podcast with Erik Torenberg, Bryan Caplan trashes the universal basic income. He gives the following arithmetic example:
UBI for a family of four of $48,000.
Tax rate of 25 percent.
That means that the “breakeven” family income (where you net zero from the government) is $192,000. That is ridiculously high. Ergo, he says, we need targeted welfare policies.
But. . .
What is the target? If your target is to ensure that no family falls below $48,000 a year, then you give a family earning $40,000 a year $8000 and you give a family earning $48,000 a year nothing. Essentially, you have a 100 percent marginal tax rate on any earnings until the family hits $48,000.
Our current welfare system provides nearly this level of disincentive, with tax rates for low income people around 80 percent.
I assume Bryan would agree that these high marginal tax rates are unwise. But he does not articulate his preferred approach.
If you go to my essay on the UBI, you will see that my preferred approach is a smaller UBI, say $10,000 for a family of four, with local governments and charities providing additional targeted support based on family circumstances other than willingness to work. A family with a disabled child would get more local support, for example. So would a family where the parents are unable to work. The low overall UBI would make the breakeven point lower. The targeted support based on family circumstances would keep marginal tax rates low for people who are able to work.
If it will help, you can think of the UBI as a negative income tax. A UBI of $10,000 with a 25 percent tax rate is arithmetically equal to a negative income tax rate of 25 percent for a family with an income of $40,000 or less.
A simpler rule.
If you earn less than $80,000, then government has no interest in your paycheck, one way or the other. You are strictly unregulated cash labor (except possibly safety rules), on your own.
The idea is that if you choose to be middle class you can also choose to dump government entirely, no tax forms, no special hedges to worry about, no weird lawsuits). A lifetime contract, you promise not to get filthy rich and government leaves you the frig alone.
Government tries to maximize revenue take. The middle class have enough for it to be worth taking, but not enough to fight the government and win. Hence middle class taxes will always be pretty high.
It’s the same with local property/income taxes on the middle class when one sees little possibility of escaping. The property tax in Baltimore is famously high, done mainly because it’s much harder to move a house than to move the legal status of your income. But the only people who pay that tax are middle class residents. The poor get exceptions to the tax rate (to the extent they even own real estate). Wealthy businesses negotiate special tax exemptions. Only middle class residents and small businesses have to pay that rate.
I actually meant federal government should leave us alone. Agt some point the middle class pays, but lower down on the ladder, please. Make them pay at the local level where budgets can be better negotiated.
Part of the problem is that the middle class already faces somewhat high marginal tax rates. Between the 12% federal income rate, a 5-10% state rate for most people, and a 15% payroll tax rate (including employer contribution) and you’re looking at 30% or so already on the middle class. Then you tack on another 25% for the UBI and they’re suddenly keeping less than half of a marginal dollar.
And then let’s say you have a single person with $20k in market income plus $10k in UBI. They pay about $1k in income taxes, $3k in payroll taxes, and say $1k in state income taxes (all of these are assuming the UBI is untaxed). They also pay $5k of the UBI back. Their net income after all this is… $20k. So the UBI basically just paid back the taxes they took. Plus now they face a 50%+ marginal rate.
There’s something wrong here. If we’re saying that a $10k UBI makes a big difference in terms of poverty, but all it’s doing is compensating people for money the government took, then were they really poor to begin with?
https://en.wikipedia.org/wiki/Negative_income_tax
I think it could make sense to have a UBI at very low levels, maybe even below what you are suggesting for a family of four, in order to provide a floor on living standards.
For most people, I am leaning towards a wage subsidy as probably the best way to help poor people in an economically efficient manner. The US EITC is like a wage subsidy, but also combines with it a child tax credit. That could be split up into two programs with the wage subsidy expanded as appropriate. The advantage of the wage subsidy over the UBI is that the wage subsidy is conducive to people staying in or coming into the labor force, while the UBI can encourage people exiting.
What to do about poor people, who choose not to work when they have the freedom to choose not to work?
Arnold says – give them free money, so they can eat.
I say, offer them a job in National Service. To everybody. But then they have to put in an honest effort to do the work assigned. A LOT of that national service will be: health care assistants for old people (cleaning up, being available, etc.); child care offerings; early ed assistants.
We need to learn how to have the gov’t hire people who “need jobs”, at a lower net cost — thru having those jobs provide benefits to other taxpayers so the net utility price (job value – money cost of program) is less negative. Accept that the National Service program will cost more, in money, than the jobs are worth. Still try to minimize that loss.
Poor people need jobs. For self respect. Only their own accomplishments can give them self-respect. Not Gov’t handouts, not UBI.
Track the employment rate of those 18-68 (? normal retirement).
Hire the disabled, the stupid, the lazy — and then give them doable jobs.
+1
I wouldn’t even try to make it seem like a discount job program (that may be difficult to achieve), for several reasons.
We don’t want people being pigeonholed in the program.
The self-esteem people get from working a job would probably be severely attenuated if it was obviously make work.
Ideally, we’d want to be able to pay people well enough so that they can make a good living, and also so that turnover isn’t so high that the jobs are completely useless.
This is how’d I’d structure it.
Look at the number of jobs required to reduce the unemployment rate to 2.5%, and provide sufficient funds (federal level) to create that many openings. The majority of the funds would go to individual cities, so that a poor urban city with 8.0% unemployment gets many openings whereas a wealthy suburb with 2.8% unemployment gets relatively few, though states should be able to create useful jobs at that level as well.
Provide wage/pension funding at 75% of GDP per capita per job, plus 50% of that per job for any overhead expenses. Cities can set wage levels for any particular job (e.g. lower levels for younger adults with simple tasks, higher levels for experienced older people who may have a lot of bills and are well trained, but make it a real pay scale).
Think about useful things that people can do – and as Tom says, they don’t need to be MB = MC. There’s a lot of social benefit from the mere existence of a worthwhile job that isn’t captured monetarily.
I hate waiting at the DMV. Why not create enough new locations such that the typical wait time during a busy time falls from 45 minutes or an hour to 15 minutes? We pay for people to have a decent living, and in turn we receive at least that much of a tangible benefit.
There are lots of dangerous areas. Also, there are many people worried about school shootings. Why not train quasi-police (perhaps we can call them auxiliary security officers) to patrol dangerous neighborhoods in groups, or to have a group of security officers posted at each school?
There’s a lot of urban blight, we could create teams of people to beautify parts of cities and make them more attractive and livable areas.
I would also keep disability benefits, as there are probably a number of people disabled enough that it wouldn’t be reasonable to ask them to work a job (I’m sure those who aren’t that disabled might leap at the chance to take a low physical demands job for $40,000/yr over their, say, $22,000 disability paycheck).
Tom G.
You stated …
Oops – meant to read …
Tom G:
You stated:
“Poor people need jobs. For self respect. Only their own accomplishments can give them self-respect.”
You couldn’t be more wrong. But I know you are not the first to be so, nor are you alone in that regard.
May I suggest a poem by Edwin Arlington Robinson, circa 1890’s, titled: “Richard Cory”.
High EMTRs are unwise … if … one only cares about encouraging people to maximize their potential for (formal, reported) income.
But we care about other things, and so maybe high EMTRs aren’t that important.
What someone like me would prefer is extremely high behavioral tax rates. If you follow the bourgeois script – married, 40 hours a week at the steady job, etc. – them the welfare floodgates will open to you. Deviate from the script, then watch the government benefits evaporate.
If we are obligated to be non-judgmental and so willfully blind in our neutrality that we are only allowed to consider the AGI block on some tax form, then the whole question of proper incentives seems absurdly narrowed, and, indeed, it is not actually neutral, since every choice of payouts tends to incentivize something.
Or, whatever the amount of the dole, I think it’d be an improvement to at least make welfare more narrow for those who ‘deviate from the script:’ food stamps for food, vouchers for healthcare and housing, rather than cash transfers. If you’re temporarily out of work because your company closed, you get a monetary unemployment benefit; but where poverty is behavior-related, the benefits ought to be paternalistic.
It’s not really a matter of paternalism per se, and anyway, vouchers are just barely paternalistic. We don’t need means, testing so much as we need ways-testing, i.e., “In what way did you get into trouble? Bad luck or bad decisions?”
The way to solve the EMTR problem of benefits decreasing with income is to think outside the financial box and to trade off with a heavy tax on overall utility which also decreases with income. That acts like compensation for having to use your extra income to pay for everything yourself.
Yeah, maybe the consumption profile legible by statistics agencies hasn’t improved very much, but consumption isn’t everything, and people care a lot about other things too. Things we can take away. There’s a lot more than taxes and benefits that go into those indifference curves.
That means making receiving assistance must be as annoying, frustrating, and humiliating as possible for the bad-decision makers who could be making better decisions and earning more.
For example, imagine the difference between paying for a normal health-insurance plan, and getting “Medicaid Parole”, in which you have to show up at Sick Call clinical inspection every other weekend.
Or think about a young man with the chance to mooch and live at home, but under mom and dad’s supervision and house rules, or who could go out and live on his own, albeit at a much reduced lifestyle, at least in terms of that “measurable financial consumption profile.” It’s very easy to see that a typical person in that situation might be indifferent between circumstances at very different levels of consumption, in exchange for status, esteem, pride, freedom, independence, privacy, etc.
And we can leverage that. We should! Because it solves the problem.
So, the overall welfare system should be a combination of material and psychological benefits. At the low end, you just get MatBens, but lots of negative PsychBens. As your income goes up, sure, the MatBens go away, but the PsychBens go way up.
People sell their food stamps right outside grocery stores. And they tend to pay for junk food anyway. Vendors I work with are fine with EBT card holders. It’s a reliable flow of revenue purchasing high margin items.
Housing vouchers are a whole can of worms, Section 8 destroys middle class communities.
There really isn’t a lot of cash welfare in the world today. It’s all “the government pays for some service”.
According to Wikipedia (I’m too lazy to do better reseach): Outlays for Social Security, Medicare, Medicaid, “Other” mandatory outlays (which are transfer programs), and “non-defense discretionary spending” (more transfer programs), were $ 2.5 trillion in 2018.
At a slightly smaller UBI of $10,000 per adult and $5,000 per child (the child UBI should be small enough that it doesn’t create incentive effects), the cost would be roughly $2.9 trillion.
The $400 billion deficit (and then some) would easily be made up by the incentive effects of a 25% flat tax.
Charities would have to pick up some slack, especially in health care costs. They probably would — remember that most hospitals were founded as charities. The change to high-cost profit making establishments was coincident with government involvement and subsidies.
The problem here is that while it might be desirable in a theoretical sense to replace all welfare transfer programs with a UBI, it will prove politically impossible. Many people will find their benefits reduced, such as higher-end social security recipients, and they will scream bloody murder and vote accordingly. And those accustomed to gaming the various welfare systems will do likewise (talk to any ER nurse about this). And when it’s discovered that people of color are in some marginal way coming out worse than some non-POCs, game over.
Not even higher end Social Security recipients – consider a married couple on Medicare and Social Security in the bottom 30% of the income distribution for the elderly. How eager do you think they would be to take $20,000/yr instead of Social Security and Medicare?
I also don’t think you’d get any positive economic incentive effects to cover the expense with new tax revenue, I think you’d end up with economic losses as many second earners would drop out of the work force, particularly those who have to pay for child care. You’d have to have a barebones UBI like Arnold proposes for the incentive effects to be minimal.
Milton Friedman proposed the Earned Income Tax Credit to Reagan to solve this same incentive trap, and it has worked well. The way to make EITC work better would be to start merging some or all of the hundreds of other welfare programs into it, so that they go away while EITC’s income brackets expand and its payment amounts increase. (And so that most of the bureaucrats who administer the present mess can be laid off — so even if they remain net tax recipients, at least they will no longer be making the system worse.)
The broader problem is the progressive income tax itself. Any tax produces revenue in proportion to the average tax rate, while disincentivizing work in proportion to the marginal rates on each earner. A progressive income tax combines low average rates with very high marginal rates on the most productive earners, thus pessimizing both results.
“Tax rate of 25 percent.”
The effective spending rate across all government levels is 40% of income – $7.1 trillion.
The reason the math isn’t working is because your tax rate is too low. It’s below even the current tax burden of 28%.
Also UBI is part of taxable income.
Approximate non-old people, non education transfers across all levels is 3.5 trillion, or about $10k per American. Assuming our 40% flat tax, this gives us about $16,000 tax credit per person. Call it the unearned income tax credit if you prefer.
The real issue is trying to put a ubi on top of existing transfers without increasing the tax rates. On top of that the US government at all levels is currently under taxing the population by about $2 trillion a year in favor of paying higher taxes at some point in the future. UBI doesn’t add up in this context, but neither does the status quo. You can make the math work with if ubi is allowed to run deficits too.
It’s frustrating that Caplan refuses to do similar basic math of why we can’t have open borders given our current social spending.