Bryan Caplan  

Libertarianism Against the Welfare State: A Refresher

PRINT
Krugman's confusing argument f...
I'm a hard-core libertarian who defines libertarianism broadly.  If you think voluntarism is seriously underrated and government is seriously overrated, you're a libertarian in my book.  I also strive to treat others with common decency regardless of their political views.  That includes libertarian apostates.  People sometimes cease to be libertarians even on my broad definition - and when that happens, the proper reaction is not anger and ostracism, but friendliness and curiosity.

In recent years, I've heard many libertarians expressing new-found appreciation for the welfare state.  This is most pronounced at the Niskanen Center, but that's only part of a broader trend.  If the revisionist position were a clear-cut, "Sure, most of the welfare state is terrible, but the rest of okay.  We should cut social spending by 80%, not 100%," their libertarian credentials would not be at issue. 

When libertarians start describing Danish "flexicurity" with deep admiration, however, I don't just doubt their libertarian commitment.  More importantly, I wonder why they changed their minds.  And to be honest, the more I listen to them, the more I wonder.  The most enlightening path, I think, is to restate what I see as the standard libertarian case against the welfare state, and find out exactly where they demur.  Here goes.

Soft-Core Case

1. Universal social programs that "help everyone" are folly.  Regardless of your political philosophy, taxing everyone to help everyone makes no sense. 

2. In the U.S. (along with virtually every other country), most government social spending is devoted to these indefensible universal programs - Social Security, Medicare, and K-12 public education, for starters.

3. Social programs - universal or means-tested - give people perverse incentives, discouraging work, planning, and self-insurance.  The programs give recipients very bad incentives; the taxes required to fund the programs give everyone moderately bad incentives.  The more "generous" the programs, the worse the collateral damage.  As a result, even programs carefully targeted to help the truly poor often fail a cost-benefit test.  And while libertarians need not favor every government act that passes the cost-benefit test, they should at least oppose every government act that fails it.

4. "Helping people" sounds good; complaining about "perverse incentives" sounds bad.  Since humans focus on how policies sound, rather than what they actually achieve, governments have a built-in tendency to adopt and preserve social programs that fail a cost-benefit test.  Upshot: We should view even seemingly promising social programs with a skeptical eye.

Medium-Core Case

5. There is a plausible moral case for social programs that help people who are absolutely poor through no fault of their own.  Otherwise, the case falters. 

6. "Absolutely poor." When Jean Valjean steals a loaf of bread to save his sister's son, he has a credible excuse.  By extension, so does a government program to tax strangers to feed Valjean's nephew.  If Valjean steals a smartphone to amuse his sister's son, though, his excuse falls flat - and so does a government program designed to do the same. 

7. "No fault of their own." Why you're poor matters.  Starving because you're born blind is morally problematic.  Starving because you drink yourself into a stupor every day is far less so.  Indeed, you might call it just desserts.

8. Existing means-tested programs generally run afoul of one or both conditions.  Even if the welfare state did not exist, few people in First World countries would be absolutely poor.  And most poor people engage in a lot of irresponsible behavior.  Check out any ethnography of poverty.

9. First World welfare states provide a popular rationale for restricting immigration from countries where absolute poverty is rampant: "There just coming to sponge off of us."  Given the rarity of absolute poverty in the First World and the massive labor market benefits of migration from the Third World to the First, it is therefore likely that existing welfare states make global absolute poverty worse.

Hard-Core Case

10. Ambiguity about what constitutes "absolute poverty" and "irresponsible behavior" should be resolved in favor of taxpayers, not recipients.  Coercion is not acceptable when justification is debatable.

11. If private charity can provide for people in absolute poverty through no fault of their own, there is no good reason for government to use tax dollars to do so.  The best way to measure the adequacy of private charity is to put it to the test by abolishing existing social programs.

12. Consider the best-case scenario for forced charity.  Someone is absolutely poor through no fault of his own, and there are no disincentive effects of transfers or taxes.  Even here, the moral case for forced charity is much less plausible than it looks.  Think of the Good Samaritan.  Did he do a noble deed - or merely fulfill his minimal obligation?  Patriotic brainwashing notwithstanding, our "fellow citizens" are strangers - and the moral intuition that helping strangers is supererogatory is hard to escape.  And even if you think the opposite, can you honestly deny that it's debatable?  If so, how can you in good conscience coerce dissenters?


Personally, I embrace all twelve theses.  But even the Soft-Core Case implies radical opposition to the welfare state as it currently exists.  My questions for lapsed critics of the welfare state: Precisely which theses do you reject - and what's the largest welfare state consistent with the theses you accept?


Comments and Sharing






COMMENTS (16 to date)
BC writes:

On K-12 education, isn't the rationale for public funding of K-12 based on positive externalities of education, that people other than the child benefit from that child's education? In fact, I think Caplan has argued that people should have more children due to positive externalities: more people to trade with and more potential future spouses for one's own children. Presumably, one would want one's children to choose from an educated spousal pool.

Other than that, his soft-core case against universal welfare seems spot on. Why tax someone just to return their own money back to them?

Thomas B writes:

Gosh, I consider myself pretty libertarian, but I think even your soft-core case falls on its face as an influential argument, which suggests to me that there might be a more useful way for you to think about this.

1. Universal programs - "taxing everyone to help everyone makes no sense".
I don't think most people would agree with you.
Essentially, if you are a starving old person, we're going to wind up feeding and housing you. And that's going to be true EVEN if you once had the means to save for your old age, but partied instead. So, we're going to force everyone who can save for retirement to do so, at least to a minimum extent. Whether Social Security is the best design for that, is debated, but the basic concept of "taxing everyone to help everyone" is not, per se, altogether invalid. This is why the government increasingly looks like a heavily armed insurance company.

2. "Absolutely poor through no fault of their own".
I don't think this one gets universal - or even majority - agreement either. I think a LOT of people think that in a society as rich as ours, there is a certain (moving) threshold of basic needs and, yes, wants that can and should be met - EVEN if the person is not "worthy". I understand that you don't, but enough people do that you will lose the discussion IMMEDIATELY if you emphasize this point. They will find your position - and that of people who "have succeeded in our social system but don't want to help the less fortunate" - morally repugnant.

Finally, with your view that "ambiguity should be resolved in favor of the taxpayer", I think you're going to get blank looks from most people. At present, in our society, people are not going to know where you got that idea - and many of them will quickly see that, since there is always SOME ambiguity, your principle would eliminate all social programs - which they will reject without further consideration. They would not see your "principle" as any more valid than "ambiguity should be resolved in favor of the person with the greater need" or "ambiguity should be resolved through the democratic process". You're stating a subjective value and not, as you may think, the obvious.

On the other hand, perverse incentives are much easier to discuss and you'll get much less push-back. I also think it's possible to frame "responsible behavior" in ways that look less like blaming/shaming (increasingly considered morally reprehensible themselves) and more like helping/enabling. Same behaviors, though.

Liam writes:

I must say I don't see why it is wrong in principle to steal to help people who are only relatively poor. If I knew I could get away with stealing a multi-billionaire's tenth Rolls-Royce and use it to fund cataract treatments for poor old ladies or an acne treatment for a teenager who was seriously depressed about their appearance, it would seem to me the right thing to do. If you vividly imagine the increase in well-being for these people, as compared to the slight decrease in utility for the billionaire, it seems to me the right thing to do. I think some libertarians are too caught up imagining the horrors of a mugging situation and they think, 'No! Nothing could justify this!'. If you exercise a bit more moral imagination, I think the intuition against stealing will lose some of its force, especially when you imagine the stealing taking place within a more stable and predictable system of government welfare so that you avoid the terror of having guns pointed in people's faces in the night.

Lauren writes:

Hi, Liam.

So, um, where do you live? What's your address? I'd like to steal from you and give the proceeds to people I care about who are not only just relatively poor but even more destitutely poor than you or me.

I know this sounds facetious. But the important issue you raise is this: In principle it may sound fine and dandy to steal from someone else to give to those who are poor. But if you are the one being stolen from, wouldn't you have a problem with that?

You, Liam, are richer than many others--for sure, because you have access to a computer and are able to read and able to type up your comment and post it on EconLog. So, where would you draw the line? When is it okay to steal from someone else, but hey, you yourself are not okay to steal from?

The concept of Robin Hood in the days of yore was about a desperate situation that is not apt today.

I am trying to reach out to you because, frankly, I don't want anyone to steal from me in my home. But that's how your comment comes off when you say things like it's okay to steal so long as it's to help people who are only relatively poor. So, it's okay to steal so long as your motives are--what? to give to those poorer than who? You? the person you stole from? So, me, if you steal from me? I guess I'm just baffled by your saying you don't see why it is wrong in principle to steal so long as it's to help people who are relatively poor. Relatively poor compared to whom? You? Me from whom you say you see no problem in principle from stealing? Someone abstract you'd like to imagine who is some kind of King John of yore who first stole from his citizens and was deserving of punishment, and who certainly did not, like you, or like me or my parents and grandparents here in America, earn their keep and save and struggle to put aside a nest egg so as to take care of themselves or their grandchildren? Steal from me in my home? My parents' or grandparents' homes? My child's home, who finally has a good job and is socking away money to save for a rainy day after the long, hard financial crisis of 2008? All okay to steal from "in principle" so long as they are not "relatively poor"--so, say, poor relative to whom? you? the person from whom you steal, however much or little that person may have? You draw the line where? If you are richer than me--and you make it sound like you are--may I steal from you without concern or moral repercussions so long as I'm going to give it to others poorer than me? I give a whole lot to people poorer than me. Let's do it together! Why wait for the government?! What's your address?!

ZC writes:

@Thomas B

You (somewhat unintentionally) highlight the fundamental fallacy of the welfare state -- " we're going to wind up feeding and housing you". Talk about perverse incentives...

I have no moral or ethical obligation to care for someone I don't know who has made decades of poor decisions. That you sincerely believe that the hard work of individuals should be forcibly taken and given to those who 'partied instead' (your words) is troubling to say the least.

Denver writes:

I don't even think the moral case, as you've presented, holds any water. Taxation isn't equivalent to stealing, it's more akin to mugging. If you evade paying taxes, the state doesn't say "ah, you got around us, better luck next year". No, they use coercion to take your money.

And while there might be a moral obligation to help the truly destitute. It would be difficult to argue that there is a moral obligation to force others into helping the destitute. If you pass an undeservingly poor man on the street, do you have an obligation to give him money? Arguably. If you pass an undeservingly poor man on the street, do you have an obligation to coerce the rich guy next to you into helping him? I doubt many would say yes, and even less would actually do it if the situation arises.

The only real moral justification here is the one libertarian weasel: consequentialism. Does having a redistribution program, as opposed to not having one, provide significant benefits to society as a whole?

Liam writes:

Hi Lauren,

Thanks for your reply.

I think it is OK to steal if it will increase overall utility.

Does this mean people should, in principle, steal from me? Yes, it does. I think people should steal pretty much everything I have and give it to starving children. Do I want this to happen? No, because I am far too selfish. I don't care enough about being moral or creating a moral society to actually want it in reality, unfortunately.

Currently, I allow loads of people to die horrible deaths or go blind from poverty because I am too selfish - like most people. I use the money to buy things like cinema tickets and ice creams. This is not morally acceptable, in my view. I honestly believe the state ought to compel me to save more lives and prevent more people from going blind. I think the libertarian view on all this is severely tainted by self-interested bias. Libertarians don't want to give away almost all their income, so they want it to be true that it's not an enforceable duty to do so.

If I know I can prevent a teenager from being really miserable by giving them the funds to buy an expensive acne treatment, and I fail to do so, I think I act immorally. So the state should take it from me.

It might seem to be a somewhat demanding view of morality, but it's the one that makes the most sense to me.

Philo writes:

Liam,
Why wait for the State to save you from your irresistable compulsion to wrongfully selfish behavior? Give Lauren your address; she'll take care of it!

pyroseed13 writes:

I think the problem that Will Wilkinson and Jerry Taylor have is that they find the Rawlsian argument for the welfare state far more compelling than your typical libertarian. And I agree with them in part. While it is true that irresponsible behavior partially explain why poor people continue to remain poor, it just seems unfair to punish the children of those families who had no control over which family they were born into. Private charity certainly can play a role here, but I wonder if would be adequate. In general, I'm just not comfortable telling people that society has absolutely zero obligation help its poorest citizens.

rowbigred26 writes:

Liam,

If Lauren were to forcible steal from you, would you call the police? If so, why? If not, why not?

JLV writes:

Libertarianism is one big moral hazard problem: non-poor libertarians have no skin in the game if their policies fail, and reap financial rewards regardless (in the form of lower taxes).

Michael Savage writes:

I have a particular problem with 5 and 7: how do you define 'fault'? I think you segue from solid pragmatic arguments to questionable moral ones. Conscientiousness itself is at least partly heritable. So is low IQ. There are lots of reasons for poverty that can't be accounted for purely as acts of will. Of course there are good reasons for society to value and reward industriousness, because those traits increase overall welfare. But you imply judgment of moral worth that raises even my hackles. It starts to look like a statement of self-interest on behalf of people with high conscientiousness and low neuroticism.

As stated, I think 10 is in bad faith, because there is always ambiguity. It's just a way of saying tax is never justified; there must be scope for debating probabilistic costs and benefits. Your phrasing is in tension with 4, which I do agree with.

Steve J writes:

The people you are complaining about are most likely libertarians that are influenced by utilitarian thinking. In this piece you mention cost-benefit analysis - why would cost-benefit analysis matter in this argument? Don't allow yourself to think too critically about considering outcomes or you may soon be joining your enemies.

Jeff writes:

Overall, quite a good summary, but I think 7 and 8 are a little muddier issues than you make them out to be. For example, someone who is poor because they were born with a severe learning disability would be easy to classify as deserving poor, but if you consider that some people are poor because they have a severe lack of conscientiousness which prevents them from holding regular employment, which is likely also a product of some accident of birth, then how do classify that second group? My instinct is to categorize these people as undeserving poor, but how can I reasonably justify that? It's difficult to distinguish between situations where someone is unavoidably held back by physical or mental limitations they have no control over vs those people who could overcome personal defects if only they tried harder or what have you. I guess you can kind of cover that with #10: when in doubt, choose in favor of less coercive redistribution, but nonetheless I think grouping people into deserving and undeserving poor is a more difficult task than it might at first seem, and someone might easily retort that given, as you state, that first world poverty is pretty rare, we should actually err on the side of classifying people as deserving, because if they are that big an outlier, it's probably due to factors outside their control.

I guess if I were a bleeding heart libertarian type, I might say something about the welfare state stopping people from free-riding, too. If you have a moral obligation to help the poor, then in a strictly voluntary charity model, it's too easy for people to free-ride off the charity of others. I guess you could also make an argument where, even if I never have to use social safety net programs, I benefit psychologically from their existence (ie, it gives me peace of mind), so maybe it's not completely unreasonable to ask me to help pay for them, assuming you admit it's legitimate to have them in the first place.

People on the left also like to make the argument that public education is justified because we all benefit from living in a more educated society. Someone so inclined could make a similar case that we all benefit from having fewer poor people running around stealing bread in the first place or drinking themselves retarded and passing out in public places, etc.

kyle r writes:

Liam

As Michael Munger says, the state is not a unicorn. There is no reason to think the government (made up of fallible, mortal humans) is any more virtuous than those outside the government. In fact there are many arguments to the exact opposite (e.g. the principal-agent problem).

I wonder how much Americans would donate to charity if they weren't taxed so heavily by the state. Based on the median household income, number of households, and total amount donated to charity by individuals, the average American currently donates 4% of her income. No doubt this would increase if people paid less taxes and could not assume the welfare state would take care of all the problems (not that it is doing a very good job now). This spending would probably be much more localized and be better suited to improve the community. The issue would be in areas prone to cycles of abject poverty. These areas already see a lot of charitable work, but it could be possible that more would be needed.

Floccina writes:

Well as long as we will have a welfare state and we will, we should try to make it efficient.

Replace almost all of it (SS, TANF, SNAP, housing and energy aid) with a BIG.

Make Medicare and Medicaid pay only for the treatments with evidence of large positive margins. This would cut Medicare and Medicaid to less than 1/2 of their current costs.

Also I do not mind programs for the mentally retarded, the Schizophrenic and severely disabled.

POST A COMMENT




Return to top