Wilkinson on libertarian principles and welfare policy

We really could have been linking to everything that’s appeared on the new Niskanen Center’s blog in general lately, and in particular to the contributions from former BHL guest-blogger and original liberaltarian Will Wilkinson, and from Jerry Taylor.

But we really should link to Will’s latest, in which he foreshadows the Center’s forthcoming work on welfare policy with explicit connections to the work we’ve done at this blog, and the work many of us have done in various ways in our scholarly careers. He also discusses what it means to see the broad family of views characterized by BHL and by Niskanen as still libertarian views. (Our libertarianer-than-thou critics like to emphasize that Will has sometimes disavowed the word, but I think this piece should be treated as a fuller statement of his views than those comments.) In the process he adopts, via John Tomasi and Jason, the only phrase I find even uglier than “bleeding-heart libertarianism” for those views, but c’est la vie.

Disclaimer, I guess: like Matt and Kevin, I have signed on to the Niskanen Center in a formal capacity, though as yet it’s a matter of form not function, i.e. everyone is trying to figure out quite what our participation will mean. Anyone who knows how these things work will be unsurprised to hear that no money has changed hands, nor is there any expectation that any will. I don’t think there’s good reason to keep putting this disclaimer in for future BHL links to writings over there, but I thought I’d note it once. BHL is its own thing, the Niskanen Center is its own thing, but there are obvious affinities and some of us will be involved with both.

Published on:
Author: Jacob T. Levy
  • Aeon Skoble

    It’s not necessarily a “libertarianer-than-thou” purity-test mania that makes some people skeptical of coercive redistributivism. You will anticipate that one objection is that _because_ the redistributivism is coercive is enough to make it a bad idea, and I actually think that’s true, but leave that aside for now. We have sufficient reason to be skeptical because we have seen a zillion times that state power multiplies and grows itself based on the logic of bureaucracy, and that it necessarily lacks the knowledge to make the kinds of plans it needs to make. Never mind Rand and Rothbard, how about Hayek and Buchanan and Beito?

    • jtlevy

      “It’s not necessarily a “libertarianer-than-thou” purity-test mania that makes some people skeptical of coercive redistributivism.”

      I didn’t say it was, and don’t think it is! We have plenty of critics. My comment was about a subset of them– a subset I do think exists, and I don’t think includes you.

      • Elsewhere, I have seen other Niskanen affiliates make similar claims about similar, nameless libertarian critics. How many times must this point be made before it starts to take on the appearance of a straw man?

        The Niskanen Center seems dedicated to the promotion of things that stand athwart of traditional libertarianism and Wilkinson & Taylor have admitted as much. If their task is to solicit buy-in from the more reasonable subset of libertarians, then maybe it’s time to stop addressing the unreasonable ones and causing this same point of contention that keeps coming up again and again. Don’t you think?


      You don’t get it. This particular coercion is different than all the others because it’s endorsed by philosophically sophisticated, really nice, well intentioned, neoclassical liberals. Got it now?

      • Reasonable Extremist

        I’m still unclear as to what the nature of the entire “neoclassical enterprise” is. I remember reading some posts on here in the early days attacking libertarians who would refuse to countenance redistribution even if something like 15 percent of the population was starving/ homeless through no fault of their own. I have no doubt such people exist but they seem to be few and far between and without any sort of serious representation in philosophy departments. In any event, I’m happy to stipulate the morality (or at least lesser immorality) of redistribution in such a hypothetical instance. But then you get into articles praising Denmark and Wilkinson endorsing Sanders and it’s like “Wait what?” I don’t think one has to be a “hard libertarian” to view this with skepticism.


          Well, if it makes you feel any better, I am in the same boat. It seems to my naive, unschooled brain that any flavor of “libertarianism” must embrace a strong presumption against coercion, even in the service of noble causes. Of course, we can argue about how strong this should be, but libertarian philosophers should at least be able to articulate a plausible ethical foundation for this presumption. Not sure how strong the distaste for coercion is with the neoclassical liberals, or what it is based on.

          • Reasonable Extremist

            Right. I mean I can appreciate the idea that any system of property rights that lead to innocent suffering is incompatible with the Lockean proviso. But the Nordic model goes so far beyond that in redistribution wealth. Indeed much of the re-distribution goes to middle income people in the form of highly subsidize maternity leave and the like. This is particularly objectionable as it involves the state providing funding for certain life choices but not others.

          • MARK_D_FRIEDMAN

            Agreed. I am convinced that any plausible version of libertarianism cannot countenance the state forcibly subsidizing some lifestyle choices at the expense of others.

          • Swami Cat


            One such principle is “consequences matter”. I am not arguing for coercion, just that we are choosing between real social institutions with real consequences. Imaginary ideals based upon philosophical principals is a fine starting place to test reforms, resist changes, or even to try to spin off new start up colonies. Until tested and proven though, it is just speculation.

            My distaste for coercion is that it tends to lead to worse outcomes. If it routinely and unanimously led to better outcomes for all, I would be all for it. Wouldn’t you?

            As for the way to have both redistribution and non-coercion, the most promising angles to investigate seems to me to increasingly make social participation voluntary among competing entities with easier exit and entry. I would choose society with strong social safety nets which do not encourage free riding. But I could be wrong. In the end the answers will be emergent, not constructed by top down philosopher kings…. Not even libertarian philosopher kings.

          • MARK_D_FRIEDMAN

            I’m a natural rights libertarian, so my answer to the question posed in your 2nd para is, “yes,” but only if you assume that everyone affected preferred this outcome. Autonomous agents are, in my view, entitled to decide for themselves what consequences they prefer, subject to the rights of others. And, if they willingly elected these consequences, it is no longer “coercion.” So, to take a real world example, I am not in favor of forcing people not to smoke, even if in some sense this produced a “better outcome.”

            Your third para sounds like what Nozick had in mind in Part III of ASU, where he describes his “framework for utopia,” and I endorse this idea. But it doesn’t eliminate the need for philosophy. There is still the need to determine what constitutes “aggression” by one community against another, the rights of children, etc. Also, of course, we don’t live in Nozick’s utopia, so we still need a yardstick to measure the justice of our state’s policies and laws.

          • Swami Cat

            I broadly concur. I think a productive area of inquiry is that rather than start from the more narrow category of “coercion,” to establish a foundation starting with the much broader category of “harm.” I can come up with twenty distinct forms of harm.

            The problem is that there is no possible way to eliminate all harms in social settings. Indeed the reason we are a cooperative species is that we can minimize the natural harm that the universe throws at us by cooperating. But in doing so, we create new types of harms (such as harms due to refusal to cooperate, or negative externalities, or reductions in scarce resources, or relative status harms, or harms established proactively to discourage negligence or coercion). The best we can do is optimize net benefits/gains/ solutions and minimize the costs/harms/problems.

            A successful society manages the tradeoffs well and thus creates greater net benefits for citizenry than those who manage the tradeoffs poorly. The trouble of course is that we have different goals, weights, contextual conditions, skills, and so on. Thus we are unlikely to ever agree on what the best social institutions and mindsets are, even assuming we could be confident that our opinions are right even about what WE desire longer term and how WE should attempt to achieve it. In other words, we don’t know what the best path is either. It needs to be discovered over time. It is emergent.

            The only reasonable answer I see is create institutions which allow choice. It sounds at first like it would lead to infinite societies, but this ignores the fact that much of the value of society is that lots of others are operating by the same rules and expectations. Thus, there is a strong attractor toward, larger, more successful and more impartial institutions of cooperation. People will get feedback on how well their cooperative systems work, and will have the opportunity and incentive to both revise their system or abandon it for more cooperative and less harmful ones.

            Call it evolutionary Rawls without the default minimax assumption.

          • MARK_D_FRIEDMAN

            Yes, a friction-free right of exit combined with the right to establish new communities would make the world a much better place. What we have in place now here is almost exactly the opposite.

          • Reasonable Extremist

            Hi Mark,

            As somebody who got into libertarian thought from the left, the issues that initially brought me in were things like the drug war so I’m very sympathetic to what you say but I wonder how you respond to the idea that certain substances ought to be banned because they are fundamentally antithetical to the maintenance of the critical faculties that make autonomy possible. I’m thinking here of the people who claim that certain drugs are so addictive that the autonomy argument simply cannot be made. I’m quite skeptical of these claims but on empirical grounds (work by scholars like Carl Hart have shown that few, if any drugs, really have this sort of hold on people). I do wonder, though, if there were drugs that really did have these effects on users if we could argue against bans on deontological autonomy/ agency grounds rather than consequentialist ones.

          • MARK_D_FRIEDMAN

            Interesting question. I think it depends on the effect the drug has. If, for example, it makes you the robotic slave of the first person who talks to you, turning the user into a useful hit man or terrorist, I believe society could be justified in prohibiting use (assuming of course that this ban would generally be effective). Here, your right to take the drug is overridden by the dangers to innocent others.

            On the other hand, if the drug just plugs you into your own fantasy land (see Nozick’s “experience machine” thought experiment), then I can’t see a good argument for forcing you not to surrender your autonomy. Society might be justified in imposing conditions to assure that you are really making an autonomous choice, but otherwise this case seems indistinguishable from the right to commit suicide.

          • Reasonable Extremist

            Interesting answer. In your view, are there any circumstances in which paternalism is acceptable apart from those involving minors? This paper raises an interesting challenge in my view http://ocw.mit.edu/courses/linguistics-and-philosophy/24-235j-philosophy-of-law-spring-2012/assignments/MIT24_235JS12_Paternalism.pdf

          • MARK_D_FRIEDMAN

            I read the linked paper quickly, and I am in agreement with his or her argument (I couldn’t find the author’s name). The conclusion is this:

            This argument does not imply absolute anti-paternalism. An action with either of the following properties may be paternalistically regulated given that it is known that the action would not be made in a rational state of mind:

            • The intent of the action is to achieve death or severe irreversible brain damage

            • The action, regardless of intent, entails death or severe irreversible brain damage.

            His/her point is that making bad decisions is part of our growth process, and should not be inhibited. That seems right to me. Note the exceptions are for decisions NOT made in a rational state of mind, i.e. not autonomous.

          • Reasonable Extremist

            Right. I found it very interesting. I suppose the challenge is what constitutes a rational state of mind. One might think that people who mutilate themselves are irrational for example. As with many other things, there’s a line drawing problem.

          • MARK_D_FRIEDMAN

            Philosophers need a SHARP pencil.

    • Swami Cat


      You are comparing the follies of real systems to the perfection of an ideal state which has never actually existed or been empirically proven to work at large scale. I point to the US or Sweden and you are pointing to an imaginary ideal or a book by Mark.

      As of today, all successful modern societies which I am familiar with have some degree of coercive redistribution. And the degree of success, compared to historical average, is absolutely awesome. The state of human prosperity is assuredly better than ever before in our history, and better than any other species we have ever learned of. To throw this away based upon unproven speculation is absurd.

      Don’t get me wrong, I am all for incremental steps or for enterprising libertarian anarchists to start up new colonies based upon their principles and thus prove them. I strongly hope they do. But I do not assume the ideal concept of society is actually workable until it has met the test of reality.

      • Aeon Skoble

        No, that’s exactly what I _wasn’t_ doing. My point was in the real world, the one we actually live in, there are zero examples of state power not growing, or bureaucracies not expanding, of political actors pursuing agendas other than their nominal charge. I reread my post, and I can’t see how you could possibly infer from it what you’re accusing me of. I was noting that, in the real world, things don’t work out as neatly as idealists think they will. So that makes _me_ idealistic?

        • Swami

          A thousand apologies. Sincerely. My bad.

          My point, and mine alone, is that it is true that all successful societies have an element of coercion and coercive redistribution. It is also true that coercion, bureaucracy, rules, rent seeking and redistribution tend to grow over time, usually (possible even always) eventually to the point of weakening society. It therefore implies we should actively favor and experiment with institutions which try to slow or counter this process.* One such idealistic extreme is to build a society bottoms up without any coercive redistribution at all.

          I seriously doubt that this ideal will work, but I am certainly open minded and hopeful. Society is too complex to be master planned off self consistent axioms.

          * there is another implication too — if we are selfish we should capitalize on the trend by seeking to join government or become a rent seeker or privileged position. I reject this course on moral grounds, but I think the majority of people would not.

  • Jameson Graber

    This is pretty exciting stuff; but the vocabulary really is going to be a problem. Words used as banners are incredibly helpful to spreading ideas, and when you fail to find a good banner (or lose control of one), you really face an uphill battle. Let’s face it, the word “liberal” has been lost (by true liberals) for generations. Adding “neoclassical” to it does not help.

    • Sean II

      The ultimate indignity: the progressives who took the word liberal from liberalism don’t much want it anymore.

      Which means minarchism is like a kid who had his bike stolen and thrown away.

      • jtlevy

        You made me laugh twice in one day!

        • Sean II

          I thought of an even better follow-up:

          “Who knew…that renouncing the use of force would lead to so much being pushed around?”

    • jtlevy

      Yes. “Niskanenism” would be better than “neoclassical liberalism.” Hell, “Zwolinskiism” would be better, and that’s unpronounceable.

      • Besides, “neoclassical liberalism” is too close to “neoliberalism.”

        • Swami Cat

          My thought is that it is too similar to the new catch all derogatory label that the progressives are currently using “neoconservative.” I prefer classical liberal, like Hayek.

      • Zwolinskiism it is!

      • Phil Magness

        I question whether it’s appropriate to name a philosophy after a deceased person who was often at tension with several specific tenets of that philosophy in real life. Though I do agree that it could benefit from having another distinctive name.

  • Reasonable Extremist

    I guess my question for Wilkinson and everybody else is this: to what extent does your libertarianism (or “neoclassical liberalism” or whatever you wish to call it) depend on its ability to advance the interests of Rawl’s “least well off.” Would you favour, say, a tax rate of 70 percent on incomes over 500,000 if it could be shown that this would be somewhat better for the least well off than a tax rate of 35 percent? In other words, could a Rawlsian say to you “Look you have to favour this new program or higher tax because empirically it is better for the least well off than a more limited state?” Because that seems like an absurdly demanding standard no?

    • jtlevy

      I think that of everyone involved in the overall project, John Tomasi is the only one who’s concerned to track Rawls so closely that he even might might follow Rawls that far.

      • Reasonable Extremist

        Okay fair enough and thanks for the response. I just wonder what principle or set of principles are at play in justifying the welfare state. Because one can imagine a purely pragmatic case holding that the masses will not accept the shifts and swings inherent within any dynamic free market economy (not that we exactly have such an economy right now) without a safety net. Creative destructive is great but it imposes costs and the welfare state helps temper those costs and maintains public support for markets. That seems pretty sensible. But to go further and justify the welfare state on first principles grounds… I’m not saying it can’t be done but I’d be curious to get a clearer and firmer idea of what the normative foundations really are (sufficientarianism, prioritarianism, an appeal to the Lockean provisio?) other than a rejection of Rand/ Rothbard style “property rights no matter what” position.

  • Pingback: Educated Liberty()