Bryan Caplan recently gave Charles Karelis an opportunity to offer some thoughts on poverty. Karelis had this to say about a Universal Basic Income program:
UBI has negatives too. It would be way, way more expensive, pushing US tax rates into the middle of the EU pack. There would be no positive substitution effect, since it’s not work-contingent. The pros are that it would eliminate the class antagonism inherent in EITC, since everybody would get it. What’s more it would be easy to administer and receive. Above all, it would eliminate poverty, which is a miserable condition and an embarrassment to a rich country like ours. And as a cash program, it would get government out of the business of telling people what their spending priorities should be. People could spend their money on health care, or education, or leisure, or whatever they themselves thought was most valuable.
I agree with much of that, but not the part about eliminating poverty. If I thought a UBI would eliminate poverty, then I’d support the program. Instead, I favor low wage subsidies.
There are at least three ways of thinking about poverty, and in each case a UBI program would fall short.
1. Perceptions of consumption: Poor people might be defined as those who seem to have very low levels of consumption. Even with a UBI, there would be plenty of people who make poor (or unusual) choices and thus end up with a consumption bundle that looks inadequate. Some would end up homeless, for instance.
2. Relative poverty: Some argue that the poverty line is a moving target, and as we become richer we redefine poverty to include real incomes that were formerly viewed as adequate. A UBI will not eliminate relative poverty because the concept will simply be redefined in the public’s mind to reflect UBI incomes. Think about how the term “minimum wage job” implies a lousy job, regardless of where the minimum wage is set.
3. Absolute poverty: A UBI will require European-style tax rates, as Karelis correctly points out. I’d expect these to eventually lead to European levels of GDP/person. (Not right away.) Thus America’s GDP/person will fall by roughly 25%. When that occurs, the UBI payments will have to be scaled back to levels that are lower than anticipated, and absolute poverty will reappear.
This doesn’t mean the UBI is a bad idea, but we should not expect it to eliminate poverty.
READER COMMENTS
Benjamin Cole
Aug 31 2019 at 9:30am
Unfortunately, I think everything Scott Sumner said in this post is true.
I go even further than Scott Sumner, in that I do not think wage subsidies are a good idea, and the vast bulk of welfare programs are counter-productive.
That said, all policies should be geared to keeping domestic labor markets very tight (the Fed saluting a 5% unemployment rate is criminal and asinine) and taxes (including payroll taxes) should be radically lowered people in the lower wage brackets. Eliminating property zoning and laws that criminalize push-cart and truck-vending would help a lot.
I do not know how to fix healthcare, except to say that Great Britain spends about half as much per capita and gets about the same results. Sure, free markets blah-blah-blah, but turning away someone who needs medical help seems unbearable, even if they “deserve” to be turned away.
James Roberts
Aug 31 2019 at 2:30pm
The goal of UBI is not to eliminate poverty, it’s to keep those at risk from falling into poverty. You’re not accounting for the fact that 75 million jobs are going to be lost to automation by 2022. The golden idea of automation was to allow everyone to enjoy a better quality of life, however that’s been corrupted by individuals using it to eliminate their workforce and maximize their profits. Automation hits those at the bottom of the ladder the hardest, cause in this society people can’t rationalize paying people the same amount of money to work less.
Example: it currently takes 2 people 8 hours to carry a thing. New machine comes out allowing 1 person to carry a thing in the same amount of time. Now you could A) keep the 2 people and have them work 1 at a time for 4 hours each to move the thing, or B) Keep 1 person working 8 hours to move the thing. Historically employers have done the latter, because then they get to pocket off of the loss of one person.
The question is, why is that okay? A person is not a commodity. When you take someone as an employee you are then obligated for their well being, just as they are obligated to do a job that effects your well being. Employers have no legal obligation currently, and lobbyists will see that they never will.
Jonathan S
Sep 1 2019 at 9:54am
75 million jobs are going to be lost to automation by 2022
Is this sarcastic?
john hare
Sep 1 2019 at 12:26pm
I certainly hope he is being sarcastic. That would account for totally misstating the relationship between employers and employees. It’s a business transaction of supplier and consumer, not an adoption.
Mark Brophy
Aug 31 2019 at 9:19pm
Britain, France, Italy, and Germany all have different health care systems but California, Texas, New York, and Florida are controlled centrally from Washington. If the EC in Brussels controlled all Euro health care, it would be as bad as our system.
John Paine
Aug 31 2019 at 3:05pm
First, “Perception of Consumption” is a nonsense conception of poverty. The entire thrust of the “elimination of bureaucracy” argument for UBI is that no one is fit to decide in detail what an “adequate consumption bundle” is for anyone else. If someone wants to forgo housing in order to eat like a king, who are you (or I) to tell them that their choice is wrong?
Second, “Relative Poverty” will always exist as long as there is any inequality in our system of distribution. Therefore, this is another meaningless way to interpret the promise to “eliminate poverty”. No one is promising that UBI will prevent you from making less money than other people. They are promising that you will always have the money to feed and clothe yourself.
And finally, “Absolute Poverty” is the only reasonable way to interpret the promise to eliminate poverty, especially considering that it is the default common definition people use when discussing poverty and it’s elimination. People talking about “Relative Poverty” generally call it “Income Inequality”, and people generally don’t talk about “Perception of Consumption” at all because it is unquantifiable and subjective.
Anyway… The author “expects” that raising taxes in order to fund UBI will lower GDP by approximately 25% because… No particular reason. Well, color me unconvinced.
In short, the article offers two bogus interpretations of “poverty” and fails to support, in any way, the argument that UBI can’t eliminate poverty defined using the only contextually useful definition.
Edward Zimmer
Aug 31 2019 at 3:12pm
Another case of using a term without bothering to define it. A UBI providing people with a basic income (which may be the writer’s definition) is much different than a UBI guaranteeing people a basic income (which still leaves open the definitions of “basic” and “people”, needed to draw any meaningful conclusions).
Ellen Belluomini
Aug 31 2019 at 7:31pm
This is how we get UBI to work for those populations
https://socialworksdigitaldivide.blogspot.com/2019/08/ubi-for-21st-century-from-social-work.html
Thaomas
Aug 31 2019 at 8:14pm
I, too, favor an EITC that mimics a low wage subsidy, but why would an UBI (even if not financed with a progressive consumption tax but something like a VAT) lead to a fall in per capita income?
[What could Kerelis possibly mean by the class antagonisms of an EITC/Child Tax Credit?]
MarkW
Sep 1 2019 at 8:32am
…but why would an UBI (even if not financed with a progressive consumption tax but something like a VAT) lead to a fall in per capita income
I’d say because of the disincentives for both work and consumption. If a UBI is to eliminate poverty, it must be high enough to live on and a person living on UBI alone must be eligible for Medicaid. I believe a life of UBI+health care would prove immensely attractive to many low-wage workers (all your time for yourself, no boss or customers to kowtow to, no drudgery, no commuting, no physically-demanding, dirty labor). It would also make early retirement much more feasible. One way to think of it is ‘No questions asked SSDI’ or ‘No age-limit Social Security’. Does anybody really think that if anybody could sign up for social security retirement or disability benefits at any time without any eligibility rules, that the numbers would not go WAY up and retirement ages would not decline?
And on the other side, the VAT imposed to pay for UBI (so not replacing other taxes but in addition) would discourage consumption. Add 20% to the cost of all goods and services and people would cook more and eat out less often. More would mow their own lawns, clean their own houses, fix their own cars, and decide that they don’t really need that kitchen remodel. Much of our current consumption is optional and surely would be reduced in the face of a European-style 20% VAT.
nobody.really
Sep 2 2019 at 4:15am
I share the suspicion that a UBI would reduce per capital GDP, but I don’t know by how much.
First, a UBI makes the most sense when confronting a world of automation-induced unemployment. When automated vehicles eliminate taxis, Uber/Lyft, over-the-road trucking, and potentially even pilots and sailors, we may face some problems. In such a world, it is unclear that excluding more people from the labor force will harm productivity much.
Second, a tax–yes, even a consumption tax–will drive up prices, which may shift the supply curve, leading to “dead weight loss.” But a UBI will increase people’s income, shifting the demand curve, leading to “dead weight gain.”
Even with a consumption tax, I expect the net effect would be progressive, meaning that poorer people would end up with greater spending power than today, and richer people less, all else being equal. Poorer people tend to engage in less savings than richer people, so this policy might be expected to simulate demand among inferior goods and depress investment and demand for superior goods, all else being equal.
MarkW
Sep 2 2019 at 8:09am
First, a UBI makes the most sense when confronting a world of automation-induced unemployment. When automated vehicles eliminate taxis, Uber/Lyft, over-the-road trucking, and potentially even pilots and sailors, we may face some problems.
You can relax. This is not going to happen. Autonomous vehicles that can handle all roads and driving conditions (rain, snow, fog, construction zones, etc) are not close to market-ready. But even if it did happen, the effect would be minuscule compared to the mechanization of agriculture, housework (modern electric appliances), and transportation. At the beginning of the 20th century, an enormous number of Americans worked as agricultural laborers, domestic servants, and in the ‘horse industry’. Virtually all of these jobs disappeared due to automation. To get an idea of the differences in scale, in 1900 about 40% of Americans lived and worked on farms whereas currently 3% of Americans work as drivers.
nobody.really
Sep 3 2019 at 11:59am
You make a fair point. To be clear, in arguing for automation-induced decline in demand for labor, I’m making the classic “but this time is different!” argument–an argument with a notoriously bad track record.
Those disclaimers made, I suspect that this time is different. Automation is growing increasingly sophisticated, while the demand (and compensation) for unskilled labor has plateaued or declined. That is NOT like the past.
And the scenario I describe is not exactly unknown in the literature. Yes, automation decreased the number of Americans working in agriculture. Yes, former agricultural workers were able to find employment elsewhere–except for those that couldn’t. In particular, in 1915 the US employed 20 million … horses. Technological innovation rendered their jobs obsolete. And by 1959 their number had fallen by more than 75%. Hey, that’s supply and demand.
It is perhaps not a coincidence that as nations industrialize, family size declines. We may be gradually following the path of the horse. Supply and demand.
Edward Zimmer
Sep 2 2019 at 5:00pm
And why would that be bad!? Businesses who need workers will increase wages enough to get or retain them. Do you believe no one would take those jobs if they had sufficient income to survive (or even live in reasonable comfort)? Do you believe that it’s morally bad for one to live in leisure – including the rich?
Mike Sandifer
Sep 1 2019 at 2:51am
Scott,
In the past, you’ve said you also don’t support Milton Friedman’s negative income tax idea. Do you think it would be better than a simple UBI?
Scott Sumner
Sep 2 2019 at 12:25am
In my view, the two policies are quite similar.
Mike Sandifer
Sep 2 2019 at 9:23pm
My impression is that the negative income tax combines the UBI with wage subsidies. You dirty support wage subsidies, so I might assume you’d support a negative income tax more than a simple, flat UBI.
DF
Sep 1 2019 at 7:32am
Scott, any insight into the inflation effects of UBI?
Scott Sumner
Sep 2 2019 at 12:24am
I don’t see much inflation effect.
Mork
Sep 1 2019 at 11:01pm
Wage subsidies don’t help stay at home moms, volunteers or artists like the UBI does.
Duncan E
Sep 2 2019 at 1:38am
I think you are right here. A UBI might help with the basics like food and clothes, but a UBI on its own wont solve Housing or Healthcare costs.
nobody. really
Sep 2 2019 at 3:43am
I expect a UBI might help a lot with housing. If a UBI actually permitted people to leave the paid workforce and retire to low-income subsistence living, people would be free to leave places where work is prevalent and instead move to places where the cost of living is lower. In small-town America, jobs are scare but houses are plentiful.
Michael Pettengill
Sep 2 2019 at 2:46pm
But that would prevent the creative destruction of small town businesses, and even keep schools open – Yang’s UBI is per person, rewarding having children – which would produce more educated young people, the from which coastal elites will draw the elites to increase the pool of high income consumers liberal elite capitalists give jobs too to promote high consumption to drive up living costs, ie money paid to the capitalists selling the costly consumption.
By creatively destroying small town America, everyone will be forced into high cost cities controlled by liberals.
Just read how all the population growth in Texas is in cities run by liberals while the conservative run rural areas are losing population. Texas will join Virginia in becoming New York with the big cities with most of the population in control of liberals, ignoring the rural conservatives, who have pursued policies depopulating the smaller cities and towns once filled with consumers with as much spending money as those in the big cities.
nobody.really
Sep 3 2019 at 5:48pm
Cute. But you raise an interesting question: What effect would a Universal Basic Income have on reproduction?
Above, I argue that a UBI is most appropriate in a world with declining demand for labor–and the fact that we already observe declining birthrates in industrialized nations suggests that we’re already in that world. But a UBI might alter these dynamics somewhat, both regarding supply and demand.
If parents received an extra $1000/mo for each kid born, I expect that the policy might encourage more procreation (relative to the status quo). If the payment kicked in only after the kid reached the age of majority, I expect the incentive effect would be more muted.
That said, the UBI might permit more people to substitute leisure for work. Work has many aspects: It fills people’s day and gives them a sense of identity and purpose. People who found themselves with an abundance of leisure, but lacking a social role, might be more prone to pick the role of … PARENT! That is, I suspect that many people would opt to have kids, or have MORE kids, if they only had the time and money to do so. A UBI would provide (some) more of both.
And even if people didn’t PLAN to have more kids, there’s an adage that says, “Idle hands are the devil’s playground.” When I and my significant other have more time, we have more sex. And more sex can lead to more people.
Todd Kreider
Sep 2 2019 at 1:06pm
If Scott’s estimate is about right, the 25% figure doesn’t accurately enough reflect the difference in the standards of living, which is what is being discussed with respect to a UBI.
First, inequality is notably higher in the U.S. than in Europe so skews a direct GDP per capita downward for Europe. Second, workers in most European countries work fewer hours than American workers and that isn’t considered:
Germany 25% fewer hours
France 17% fewer hours
UK 6% fewer hours
Italy 3% fewer hours
Spain 5% fewer hours
Netherlands 20% fewer hours
Denmark 20% fewer hours
Norway 20% fewer hours
Austria 10% fewer hours
Sweden 9% fewer hours
—————————-
Poland 8% more hours
Portugal 3% more hours
Michael Pettengill
Sep 2 2019 at 2:30pm
It seems to me Scott needs to argue that Yang is wrong.
Argue that by working hard to increase profits by cutting the incomes of most customers by lowering minimum wages, benefits, etc, and that by eliminating welfare for those customers businesses should never pay to work, businesses will sell more at higher profits to a smaller pool of customers with a lower aggregate income to spending buying from businesses.
Yang is advocating a UBI because he is seeking to have a larger pool of customers with higher incomes to buy stuff from the businesses in the conservative regions of the US.
He notes the liberal coastal elites suck out of conservative areas all the best customers, the customers businesses are happy to pay high incomes so they can buy lots of stuff from businesses run by coastal liberal elites.
With the coastal liberal elites using the best from all of the US to produce the most desirable and high profit products, conservative parts of the US can’t afford to consume much of the highly desirable products because conservatives argue they should consume less, not more, so conservative consumers feel left out of the US economy, getting to consume less and less.
While $1000 in consumption is poverty among the coastal elites, in the left behind coal country, depopulating plains farming communities, $1000 a month is solidly middle class consumer. Businesses with sell more and perhaps thrive instead of struggling against, and finally ending up bankrupt.
There are many towns in Kansas, Iowa, Ohio that have suffered from Scott, et al’s view that customers desire to spend less by not being paid to work or being paid less.
Bob
Sep 5 2019 at 10:22am
Scott,
I mostly agree with you. My understanding is that EITC (combined with welfare reform) has been very successful. I believe the concept of “job, better job, better job, career” has been proven true. And I think it is not irrational for someone to opt not to go to work when the costs of working eclipse the returns, as they often do for low marginal utility workers. I have no idea why we would want to limit the benefits (to individuals and to society) of EITC to people with children. To me, UBI addresses the biggest flaw in EITC- EITC assumes that poor people are good at filling our forms and it puts the federal government in the business of assessing worthiness. UBI only asks the Fed Gov to collect and distribute money which it does quite will. My thought has always been that the net effect on GDP of UBI or wage subsidies should be similar since the difference in taxes between the two would be offset by distributing UBI to all participants. And UBI would be administratively cheaper. Wage subsidies do have the benefit of rewarding work. Can you explain why UBI would cause more excess burden than a wage subsidy with similar net costs?
Comments are closed.