Bryan Caplan  

Why Libertarians Should Oppose the Universal Basic Income

Beckworth interviews Eggertsso... Bryan Caplan's Best Line and M...
Here's my opening statement for my Students for Liberty debate with Will Wilkinson.  Enjoy.

Libertarians have a standard set of fundamental criticisms of the welfare state. 

1. Forced charity is unjust.  Individuals have a moral right to decide if and when they want to help others.

2. Forced charity is unnecessary.  In a free market, voluntary donations are enough to provide for the truly poor.

3. Forced charity gives recipients bad incentives.  If the government takes care of you, you're less likely to take care of yourself by work and saving.

4. The cost of forced charity is high and growing rapidly, leading to a future of exhorbitant taxes or financial crisis.


Taken together, I think these criticisms justify the radical libertarian view that the welfare state should be abolished.   But this is an extremely unpopular view, so it's natural for libertarians to consider more moderate reforms like the Universal Basic Income.  And when you're considering moderate reforms, the right question to ask isn't: "Is it ideal?" but "Is it better than the status quo?" 

My claim: the Universal Basic Income is indeed worse than the status quo.  In fact, all the fundamental criticisms of the welfare state apply with even greater force.

1. Some forced charity is more unjust than other forced charity.  Forcing people to help others who can't help themselves - like kids from poor families or the severely disabled - is at least defensible.  Forcing people to help everyone is not.  And for all its faults, at least the status quo makes some effort to target people who can't help themselves.  The whole idea of the Universal Basic Income, in contrast, is to give money to everyone whether they need it or not.  Of course, the UBI formula normally reduces the net payment as income rises; but if a perfectly able-bodied person chooses never to work, the UBI gravy train never stops.

2. The UBI is an extremely wasteful form of forced charity.  Helping the small minority of people who can't help themselves doesn't cost much.  Giving an unconditional grant to every citizen wastes an enormous amount of money.  If you were running a private charity, it would never even occur to you to "help everyone," because it's such a frivolous use of scarce charitable resources.  Instead, you'd target spending to do the most good.  And unlike the UBI, the status quo makes some effort to so target its resources.

3. Overall, the UBI probably gives even worse incentives than the status quo.  Defenders of the UBI correctly point out that it might improve incentives for people who are already on welfare.  Under the status quo, earning another $1 of legal income can easily reduce your welfare by a $1, implying a marginal tax rate of 100%.  But under the status quo, vast populations are ineligible for most programs.  Such as?  You guys!  If you're an able-bodied adult, aged 18-64, who doesn't have custody of any minor children, the current system doesn't give you much.  Switching to a UBI would expand the familiar perverse effects of the welfare state to the entire population - including you.  And if taxes rise to pay for the UBI, the population-wide disincentives are even worse.

4. A politically acceptable UBI would be insanely expensive.  Libertarian economist and UBI advocate Ed Dolan has a detailed, fiscally viable plan to provide a UBI of $4452 per person per year.  But every non-libertarian I've queried thinks it should be at least $10,000 per person per year.  Even with a one-third flat tax, that implies that a family of four would have to make $120,000 a year before it paid $1 of taxes.  This is pie in the sky.

But doesn't the UBI give people their freedom?  In some socialist sense, sure.  But libertarianism isn't about the freedom to be coercively supported by strangers.  It's about the freedom to be left alone by strangers.

If abolition of the welfare state is extremely unlikely and the UBI is worse than the status quo, does this mean libertarians should accept the welfare state as it is?  Not at all.  There's a straightforward moderate path to a freer world: AUSTERITY.  Cut benefits.  Restrict eligibility.  Remind the world of the great Forgotten Man: the taxpayer.  We probably can't convince the majority to end the welfare state.  But "Welfare should be limited to genuinely poor people who can't help themselves" has broad appeal - and unlike the UBI, it's a clear step in the libertarian direction.

Comments and Sharing

COMMENTS (20 to date)
Thomas writes:

I agree with your argument. But #2 is its weakest link; it needs empirical support. Lacking such support, it's easy for supporters of UBI (and other redistribution schemes) to maintain that a lot of people will starve for lack of voluntary charity.

Also, you're being kind to call the UBI a type of forced charity. Like other such schemes, it's theft.

Arthur B. writes:

A weak point of the argument is that any income-tested welfare scheme based is equivalent to a UBI + the adequate tax but with more overhead. Does your argument apply to income tested welfare schemes in general, or is it something specific about the UBI?

UserGoogol writes:

I don't agree with any of these but I'm not a libertarian so I'll focus on how your response to (3) seems to miss the point. One of the popular benefits of a UBI is that because it's universal, it doesn't incentivize any particular course of action: you get the money no matter what you do, as opposed to conditional benfits which reward meeting the qualifying conditions.

Of course, it does impact behavior, but it does so by the income effect: people who have more stuff have different goals for what to do. But there's no particular reason why the income effect should lead to irrational behavior the way misincentives do. Arguing that people are made worse off by getting more money is a rather bold claim which has a lot of implications beyond the welfare state.

Of course, a UBI would (presumably) be payed for with taxes, which has incentive distorting effects. But in so far as the argument is merely that a UBI is an improvement, replacing the sharp cutoff of conditional benefits with the more gradual distortion of taxes seems like a net improvement on its face.

Hazel Meade writes:

As a devils advocate, it seems like you're arguing against welfare in general, not the UBI.
A major argument for the UBI is "We're going to have forced charity anyway, so it might as well be universal."

Also #2 and #3 are pretty weak. Voluntary charity cannot guarentee that everyone who needs help will get it. Imagine say, an autistic orphan atheist - he might not have ANY family or friends or church, and yet be a completely deserving person on the merits.
Plus, incentives don't matter if you're disabled and unable to work.

#4 is really the strongest argument. It's hard to imagine how a UBI wouldn't become financially unsustainable once people realize then can vote themselves money from the public treasury.

#1 is strong from a moral perspective, but will be unconvincing to those who believe in a positive duty to rescue.

Jon Murphy writes:


One of the popular benefits of a UBI is that because it's universal, it doesn't incentivize any particular course of action...

Be careful here, because this statement is incorrect. The fact that everyone gets UBI doesn't mean it doesn't change incentives. Let's take a simple hypothetical as a counterexample:

You have Jack. Jack lives under a system where there is no welfare and no charity whatsoever. If you want money, you must work for it. Jack would rather not work, wanting to play video games, but seeing as he wants to eat, he must. In short, he is incentivized to work. Now, let's say that a UBI is imposed, say $20k. That new UBI is more than enough for Jack to eat, so he quits his job, stays home, plays video games, and lives off the UBI. In short, the UBI incentivized him not to work. This is true whether everyone gets the UBI or not.

Or, if you don't like my hypothetical, let's look at real life: universal health care. Universal health care incentivizes people to seek more health care than they ordinarily would because it reduces the out-of-pocket cost of health care (just like how Jack was incentivized not to work by the UBI reducing the out-of-pocket costs of not working). Universal health care applies to everyone, too.

Incentives don't appear because some people are affected and others not. They appear because choices exist. When you have something like UBI, it just means everyone is affected.

Jon Murphy writes:

UBI is something I, as a libertarian, go back and forth on. I don't disagree with anything Prof. Caplan wrote here (although I might quibble with Welfare Criticism #2), but I'm not entirely opposed to UBI either (one might say I tentatively support the idea). I've written on a conditional UBI (I know, contradiction in terms) in the past (see here #shamelessselfpromotion).

UBI is an intriguing option, I think. Even if libertarians ultimately opt against it, I'm glad we're at least discussing it.

MikeP writes:

But every non-libertarian I've queried thinks it should be at least $10,000 per person per year.

This is the entirety of the matter, isn't it? Libertarians should be able to support a UBI if the 'B' stands for 'Basic', not 'Bacchanalian'.

If your entire income is a properly sized UBI, you won't starve and you won't go without a roof over your head. But you won't have a nice house. And you certainly won't be among those who spend only 30% of their income on rent. You will be able to live. And you will be incentivized to get another source of income.

The UBI should very explicitly generate a lower tier of income. It should not get you into the middle class. That's the point of the grand bargain: in return for removing the shackles from the labor market, those who are left behind -- along with those who choose to be left behind -- will be able to eat and have a place to live.

Craig writes:

Like Jon I don't necessarily disagree with anything Prof. Caplan writes. I do, however, find the silence regarding efficiency pretty loud. My understanding of the benefits of UBI is that it much more efficiently gets resources into the hands of people. Overhead, when compared to the status quo, is next to nil. There should be data to show how exactly what is being wasted on the status quo's administrative overhead and duplicate programming. The saving of those costs is the only reason I find UBI attractive but they are not mentioned here.

David R. Henderson writes:

Read this.

Christophe Biocca writes:

#3's universal availability is probably the most interesting topic to discuss given it's also why UBI is being brought up again after spending 30 years mothballed, and is popular with young people.

This is not necessarily for the super-cynical reason that it's welfare they otherwise wouldn't get and they just don't want to work.

A lot of them see the appeal of being able to take a few years of travel before starting to work, and having less student loans + a (small) revenue stream can help make that happen.

Some of the more entrepreneurial-minded think that this would help them start a company (by removing the saving-money-or-getting-seed-funding step).

Arguably both of these are examples of misallocation of resources enabled by a UBI.

The first one is the most obvious: people who just earned degrees and are in their prime learning and professional development years end up not using that degree at all for some time. If the cost was fully borne by them they'd almost never take a bank loan against their future earnings to finance a trip right at the start of their career.

The second one is trickier, but we do have some evidence in Canada, where SR&ED refundable tax credits can allow a pre-revenue company to stretch their funding to an absurd degree. Sometimes this just lets founders keep more equity they'd otherwise have to give up in exchange for funding (so it's just a wealth transfer from taxpayers to high-earning tech industry founders). Other times it allows zombie companies (entities that have no chance of ever becoming viable) stick around for longer, keeping people working on dead-end projects when they could be employed elsewhere.

Pat writes:

You have convinced me. I was in favor of a UBI because it was clear to me that the sharp phaseouts of benefits creates an implicit marginal tax rate that is too high on too many people. The UBI would reduce marginal tax rates relative to the average tax rate but it would make the average tax rate much too high.

Shailesh writes:

The real problem with UBI, or any other 'less evil' proposal is that, in practice, it is more likely to add to (rather than replace) the status quo. This is despite what its proponents say.

That's why even consequentialist libertarians should always oppose all 'less evil' proposals. That's how you increase the chances of replacing (vs. adding).

Emma Casey writes:

So you seem rather confused here as to what the actual aim is. Our aim should surely be to promote economic efficiency, non-governmental support structures, and markets. This proposal seems to do none of these things.

The present welfare system is not mostly harmful because of its cost. Suppose the EU decided as its latest act of charity that it would fund the US welfare system *as presently implemented*. This would perpetuate the dependency, the joblessness, the inefficiency in family size and house value. It would also ensure there's never economic growth in failing areas by throwing more disincentive to work at the places the such policy will harm most. Which in turn means a continuation of "the poor you will always have with you".

By insisting that the money we give to the poor be spent in ways that promote dependency we make your proposal to lower the amount we spend both less humane, and (more significantly for this conversation) less politically realistic. At present the lower rungs of society are dependent on the state because they have a strong incentive not to work. Austerity, in this context, necessarily produces visible human suffering, which harms the political project you are engaged in.

It also risks rolling back our commitment to markets. Suppose we insist on tight government regulation of welfare, with all the attendant costs. Will this not be used as an example of a case where "markets do not work", and hence as a general promotion of government interference in other areas? While a paperwork-minimal automatic system without state management can be used to argue for eg replacing environmental regs with Pigovian taxes, or with replacing public schools with a voucher system. These seem sensible incrementalist steps in the direction of less state control. Why does UBI not?

By taking such a deontological approach you seem to be forgetting to think like an economist. For instance:

"And unlike the UBI, the status quo makes some effort to so target its resources." Rather assumes it succeeds in so doing. Yes, rich people (mostly) don't get government help (except homeowners, college students, people who want health insurance ...), but that's like using overhead as a charity metric. It assumes that spending is spending, no matter what it's on. Which obviously isn't true. Programs that cut the incentive to work are more harmful that those which don't. Subsidies for childcare and for renting expensive homes distort the market vastly more than the equivalent in helicopter money.

"Under the status quo, earning another $1 of legal income can easily reduce your welfare by a $1, implying a marginal tax rate of 100%. But under the status quo, vast populations are ineligible for most programs." is a non-sequitur. For the population concerned this is a perfect dis-incentive to work. The fact that others are not in this population is irrelevant.

"Switching to a UBI would expand the familiar perverse effects of the welfare state to the entire population - including you." Seems flatly false. UBI is a slight disincentive to work via a wealth effect. But this is not at all similar to 'the familiar perverse effects of the welfare state'. UBI ensures that the marginal tax rate is not more than the listed tax rate, whereas welfare trap effects are the bulk of the welfare sate's failing.

Yes, a UBI of $10k p.a. is impractical for the reasons you outline. But to conclude from this is a UBI of $4k p.a. is worse than the present welfare system seems fundamentally to not take the idea seriously.

Matt Skene writes:

I would think a better option than a UBI would be something like a "starting out in life fund." A one-time tax-free payment of, say $50,000 would meet most of the concerns of UBI advocates at a lower cost and offset almost all causes of need later in life that at least arguably justify taxation to meet. Not sure why this wasn't the first idea rather than a UBI.

Jay writes:

@Matt Skene

While I don't disagree in principle, practically and politically, you have to answer the question "what do you do with people who blow through their $50,000?". Let them die in the street? At least the UBI (even a low one) would answer this question.

JdL writes:

I don't understand how anyone who calls himself a libertarian can at the same time support a UBI. Libertarianism is founded on the Non-Aggression Principle, and forcibly stealing money from one group of people and giving it to another is the exact opposite.

Craig writes:

@ David

Thank you for the link to your article. That absolutely put the cost in a much better perspective for me. I formally withdraw my concerns :)

Paul Marks writes:

One trick the Universal Basic Income people play is phony mathematics.

They pretend that it will "pay for itself" by the government getting rid of the Welfare State.

But then it turns out that they do NOT intend to get rid of such things as government (taxpayers) paying for education and healthcare.

In reality the UBI is ON TOP OF most of the rest of the Welfare State.

Robert Hotchkis 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.]

Joey Donuts writes:

Almost all of the arguments against the BIG because it could provide government payments to those not qualifying for existing programs assume that the BIG would appear as soon as an individual stops working.

We can easily limit the number of individuals who currently do not qualify for existing programs by simply using last year's income as the basis for the BIG. Someone already working and earning sufficient income to not require the BIG would have to stop working and earn nothing and receive nothing from the government for a year before qualifying. Of course those with enough saving to do so could do this but they would have to give up something to do it. Of course we could adjust how far back we look to make it even more difficult e.g. average of two year's previous income.
Once one has a tool like this arguments 1,2, and 3 fall by the wayside.
How big should we make the BIG presents more diffiuclties. However, we can estimate the size of the total value of all grants the "poor" currently receive.

Comments for this entry have been closed
Return to top