Scott Sumner  

Finland's Universal Basic Income experiment

The Noise-to-Signal Ratio as a... Dave Barry Highlights...

Finland has begun an experiment with a universal basic income (UBI) program:

Finland has started a radical experiment: It's giving 2,000 citizens a guaranteed income, with funds that keep flowing whether participants work or not.

The program, which kicks off this month, is one of the first efforts to test a "universal basic income." Participants will receive €560 ($587) a month -- money that is guaranteed regardless of income, wealth or employment status.

The idea is that a universal income offers workers greater security, especially as technological advances reduce the need for human labor. It will also allow unemployed people to pick up odd jobs without losing their benefits.

The initial program will run for a period of two years. Participants were randomly selected, but had to be receiving unemployment benefits or an income subsidy. The money they are paid through the program will not be taxed.

If the program is successful, it could be expanded to include all adult Finns.
The Finnish government thinks the initiative could save money in the long run. The country's welfare system is complex and expensive to run, and simplifying it could reduce costly bureaucracy.

The change could also encourage more jobless people to look for work, because they won't have to worry about losing unemployment benefits. Some unemployed workers currently avoid part time jobs because even a small income boost could result in their unemployment benefits being canceled.

Many free market economists favor this sort of program, partly for reasons outlined above. And it's very possible that the program would constitute an improvement over Finland's current welfare system. But I'd also like to offer a few words of caution.

Although I am all for evidence-based policymaking, including policy experiments, this evidence needs to be viewed with caution. For instance, consider a UBI that equaled 80% of the monthly earnings of low skilled labor. A policy experiment lasting two years might yield very different results from a permanent shift to a UBI, for two reasons. First, workers may be reluctant to give up low skilled jobs to live off the UBI, because they fear the program would end after two years, and they might not be able to get their old jobs back. Under a permanent UBI, they might quit the job and make up the 20% earnings shortfall with a bit of part time work. In that case, the UBI would increase leisure time and reduce GDP. Second, a permanent UBI would gradually lead to cultural change, which would lessen the stigma of living on "welfare". That's less true of a two-year experiment.

So what's wrong with a bit of extra leisure time? If freely chosen, I have no objection. But a UBI program tends to reduce the incentive to work. There is a risk that providing an income to people who chose not to work will lead to this sort of outcome (from a Reason article describing coal country in West Virginia):

I ask FACES' Whitt why so many young unmarried women in the county become pregnant. She sighs and notes that birth control is freely available at school. Most of the girls and women are "on medical cards" (that is, enrolled in Medicaid) that would pay for contraception as well. It doesn't matter. "There are no consequences to pregnancy--they get immediate access to a medical card, food stamps, a check, WIC, and home visits," she explains. "They have all the welfare benefits as long as their kids are not adopted, plus there's no babysitting, since the grandparents will look after the kids."

FACES organizes a Second Time Around support group for folks who are raising their grandkids or great-grandkids. It meets once per month. Slagle notes that her friend and her friend's husband are raising two of their grandkids despite health problems. "Neither one of them is able to do it," Slagle says. "You know, if I weren't rooted here, I would take the kids and go."

Cold Turkey

So why don't people just leave? That question is actually surprisingly easy to answer: They did. After all, 80 percent of McDowell's population, including my grandparents, cleared out of the county to seek opportunities elsewhere during the last half-century.

But as the mines mechanized and closed down, why didn't the rest go, too? Reed, Whitt, and Slagle all more or less agree that many folks in McDowell are being bribed by government handouts to stay put and to stay poor. Drug use is the result of the demoralization that follows.

A well-constructed UBI can provide more incentive to work than our current welfare programs, which sometimes leads to implicit marginal tax rates of above 100%. But I would argue that a well-constructed wage subsidy program is even better.

In the end, it's likely the case that a very small UBI, combined with a wage subsidy, is better than either program in isolation. That's because a very small UBI would not stop very many people from working, but would be easier to administer than an equal size wage subsidy. So you might want to provide a minimal UBI, topped off with a more generous wage subsidy program.

I don't know much about the Finnish economy, but $587/month sounds like it's not high enough to cause very many people to stop working. Hence I expect the experiment will be a "success". On the other hand, if you really wanted the UBI to replace all welfare programs, the monthly stipend would have to be much larger. And the long run impact of that sort of UBI is much less clear.

A wage subsidy tends to encourage people to work more hours. It also tends to discourage the acquisition of human capital. However, that disincentive is largely offset by massive subsidies to education at all levels of government.

Comments and Sharing

CATEGORIES: Income Distribution

COMMENTS (18 to date)
Peter Gerdes writes:

The example you give, far from being a problem with UBI, is actually a substantial argument for it (at least if done correctly).

The situation in the example is one in which undesirable behavior (early pregnancies and continued residence in a depressed area) is incentivized by government benefits for that behavior. Under a UBI scheme that replaced all such situational benefits with a fixed payment to adults we would avoid the harmful behaviors associated with government benefit seeking.

Indeed, generally under a non-UBI solution the government offers substantial extra benefits to those individuals who demonstrate an inability to work (mental/physical disability) or exceptional need (many mouths to feed). This incentivizes individuals to get disability diagnoses and to have extra children.


Also your argument is fallacious when it assumes that because UBI changes the incentive to work it somehow isn't 'freely chosen' or that the overall societal utility isn't greater if people choose not to work because they have UBI.

Thaomas writes:

Would it not be wise to start with removing the wage taxes before instituting a wage subsidy? Reporting of incomes could still be done for benefits calculations. If we wished to go farther, the EITC could be increased. With a generous EITC work requirements should not be necessary as part of UI, SNAP, etc. making them less expensive to administer.

Roger Sweeny writes:

A wage subsidy tends to ... discourage the acquisition of human capital. However, that disincentive is largely offset by massive subsidies to education at all levels of government.

That assumes that going to school automatically leads to an acquisition of human capital. That is simply not true. At the present time in the United States, I doubt that sending people to school for longer periods would have much effect on human capital acquisition. In fact, I suspect it would have a negative effect, teaching young people that the only jobs worth taking are white collar, and that once you have a degree, you are owed a good living and a pleasant job.

AntiSchiff writes:

Dr. Sumner,

It will be interesting to see whether technology will eventually reduce the number of jobs in an economy. Historically, it has destroyed jobs in some areas while new jobs were created in others. Do you have an opinion on this? Also, do you think we're already seeing a reduction in the value of human labor overall, without having yet reduced the number of jobs, with falling compensation as a consequence?

Scott Sumner writes:

Peter, Obviously I don't agree. Most UBI schemes that I've seen also provide benefits for children.

As far as being "freely chosen", suppose the implicit MTR in the system were 90%, and suppose that this tax rate sharply reduced work effort. Would you view that reduction as voluntary?

In any case, what we call it is less important than the fact that when you distort incentives in that way, you tend to get a suboptimal amount of labor supplied.

Thaomas, I think the best solution is a progressive wage tax, and the EITC can be a part of that system.

Roger, Fair point. I'd still argue that schooling adds something to human capital, but it's clear that the two concepts are far from identical.

Antischiff, I don't see much evidence that the value of labor is falling, but perhaps it is increasing at a slower rate than usual.

AntiSchiff writes:

Dr. Sumner,

Approaching full employment should be interesting then, as compensation growth could pick up, assuming productivity doesn't get even weaker.

Lots of interesting mysteries in most major economies in the world right now. My hunch is that those claiming the economic slowdown in the US, for example, is at least partially responsible for the apparent productivity slowdown are correct.

MamaLiberty writes:

Has Finland discovered a free money tree? Where would the money for this UBI come from? How in the world can any "libertarian" consider this viable, much less moral? All tax is theft. Theft is supposedly against their principles.

Diamond Dog writes:


Firstly I think a discussion of the benefits and disadvantages of any proposed policy are useful, especially one such as this which is gaining traction in many circles.

Clearly it is unjust by libertarian standards (although there seems to be a softening of these at foot...), but there are strong proponents of the group which think that the lessening of any taxation/subsidy/other illegitimate imposition by government is good. The strategy of many libertarians is to support policies which reduce government activity.

In the case of UBI, I personally do not agree with the libertarian supporters of it. It purports to lessen the spending on bureaucracy surrounding welfare. I fear that once government institutions come into being, they rarely die easily. These huge systems may be re purposed for another use? I cannot predict what that would look like, but I very much doubt UBI will reduce the governments role in peoples life.

I concede, however, that there are certain conditions that would convince me. I am very sceptical.

Roger Sweeny writes:

Scott, I completely agree that "schooling adds something to human capital" in many situations. The ability to read directions, to write a simple note, to do basic math are almost always useful. However, as Bryan points out, much that is "learned" in school is simply forgotten.

One way school builds human capital is causing students to develop certain habits and attitudes. Many high grade students act in accordance with, "I will work at this and do well at it, even though I'm not really interested and, to be honest, don't see the point." That's quite a useful habit in a potential employee.

Unfortunately, many unsuccessful students develop negative habits, "This is useless and uninteresting so I'm going do as little as I can get away with." Fortunately, real jobs are different enough from school that many turn it around, "This job IS important and it matters if I do well, so I'll put in the effort to get paid and be kept on and maybe get raises and promotions."

Thaomas writes:


In what way would a progressive wage tax be superior to a progressive consumption tax (with a negative rate range)? It seems vulnerable to gaming the system in the same way that "carried interest" is counted as "capital" income, not wage income.

Tim writes:

The problem with UBI is in the cost of administering a program to every working age adult. Even a modest sum of $7k a year becomes gigantic when you multiply it across the entire working age population of 205 million people. At $1.4 trillion it would eat up half the Federal Budget. Without means testing or addressing how much of the budget this could offset the program is politically DOA in Congress.

Cloud writes:

I don't think this one is UBI, as it is only applied to "selected" people rather than "Universal", and the selection is based on income level, in this case, it is only for poor people.

Therefore, I don't think this particular Finland experiment can be a good benchmark for the usefulness of UBI

arqiduka writes:

When discussing the budgetary impact of an UBI scheme it just won't do to simply look at the expenses side, as we do for any other welfare program.

An UBI is still a redistributive scheme, although one that works the redistribution through giving everyone the same stipend and taking back from the producers in the form of taxes (not income taxes, hopefully). So, any UBI will also increase the tax revenues by a decent amount (it is not "free money" for all).

By how much? Here a pilot program would help but not of the Finish kind: you need a truly random sample. If you pick folks on welfare to begin with, your sample just won't work.

Floccina writes:

I think to be worthwhile a UBI must be combined with the elimination of the minimum wage. That way it might even increase employment.

HokieCoug writes:

MamaLiberty wrote:

I fear that once government institutions come into being, they rarely die easily. These huge systems may be re purposed for another use? I cannot predict what that would look like, but I very much doubt UBI will reduce the governments role in peoples life.

I agree. Suppose all welfare programs are eliminated in favor of a UBI. How long would it be before well meaning politicians say "but look how these poor folks whose only income is UBI are suffering, we need a program to help them with medicine, childcare, food (insert any other aspect of current welfare).

Won't we eventually (I'd predict within 10 years) end up with everyone receiving UBI AND basically the same government welfare programs in place?

Justin writes:

--"The Finnish government thinks the initiative could save money in the long run. The country's welfare system is complex and expensive to run, and simplifying it could reduce costly bureaucracy."--

I've never understood this argument. Unless the UBI is very small, this will undoubtedly increases costs both immediately and in the long run. The primary dollar cost of a welfare program is the benefits provided, not the cost of the bureaucracy implementing it. As Tim writes, providing a UBI of a mere $7k/yr to the US working age population (presumably 18-64) would be over $1.4 trillion, and that sum would probably require some basic administration (fraud prevention especially). The amount spent in total on benefits and administration combined for existing welfare programs to that population is far less.

Scott Sumner writes:

Thaomas, The carried interest loophole is about to be closed. And there can be big loopholes in consumption taxes too. Is a mansion consumption or investment?

ma writes:

[Comment removed pending confirmation of email address. Email the to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Comments for this entry have been closed
Return to top