Libertarianism, Safety Nets, and Ideal Theory

The purpose  of this post is to emphasize a problem in our thinking about social safety nets that I believe is often ignored. Virtually all non-libertarians, and even most minimal state libertarian philosophers, will endorse  the following: “If a state social welfare program, funded by coercive taxation, is both a necessary and sufficient means of preventing grave harm to the welfare of innocent persons, its implementation is morally justified” (“Proposition #1”).  

But note that this conclusion inhabits the realm of ideal theory, i.e. an abstract moral proposition that need not come to grips with any potential costs or problems of implementation.  By its nature, Proposition #1 does not have to take into account issues of institutional design/competence, political pressures that might push well-intentioned policies off course, or other problems identified by public choice theorists.  Thus, support for Proposition #1 does not commit us to endorsing Proposition #2: “Whenever, in the absence of state intervention,  grave harm will occur to the welfare of innocent persons in the United States (or other liberal democracies), the government should intervene to ameliorate this harm.”

There are at least two reasons why, in the real world, we might endorse Proposition  #1, while rejecting #2.  First, it is entirely possible that the contemplated program will not work as planned, and may not only waste scarce resources, but might actually worsen the position of the intended beneficiaries or their offspring. I don’t believe that this is at all a fanciful idea. As President Clinton said upon signing the Welfare Reform Act of 1996: “A long time ago I concluded that the current welfare system undermines the basic value of work, responsibility, and family, trapping generation after generation in dependency and hurting the very people it was designed to help.” I regard it as an open question whether  the new and improved welfare system does much better, but I will leave this discussion for another day.

Putting this issue aside, there is an even more intractable difficulty encountered in moving from Proposition  #1 to #2, which might be described as the problem of inescapable “mission creep.” In other words, it may be the case that any state that establishes an effective welfare program (“W”) for a particular needy group will, as a result of the factors cited in my second paragraph, also undertake programs  X, Y and Z, which in the aggregate will (in the long run) harm not just the people served by W, but society at large. I believe this is a quite realistic scenario.

Until recently, Greece was a modern, first world society, with the (usual) wide range of welfare and entitlement programs.  This governmental largess was funded largely by foreign borrowing, and as soon as Greece maxed out its credit card, terrible consequences ensued. Incredibly, the unemployment rate is now 27% overall, with youth unemployment standing at an appalling 61.7%. Since the start of the financial crisis in 2008, the Greek economy, as measured by GDP, has shrunk a depressing 20%, and is expected to contract by an additional 4% this year. See

There is no end in sight to this economic meltdown. Many thousands of young Greeks with marketable skills have fled the country, and thousands more will follow, taking with them the very talents needed to reverse their country’s decline. Those left behind are not getting paid, strikes and lawlessness are endemic, and basic services are breaking down. It is no exaggeration to say that the very fabric of civilization is unraveling before our eyes:

If a large portion of formerly middle class Greeks are facing dire poverty, what will become of the most vulnerable members of society, i.e. the disabled, the old and infirm, those with no marketable skills, etc? Will the welfare state really benefit them in the long run, or would they have done better under unfettered free markets?  Is Greece simply a “one-off” or is it just further down the road to serfdom that all states with this social model are destined to travel?

Well, I have my suspicions, but no conclusive arguments for them. But, equally, those inclined to endorse Proposition #2 cannot simply assume these questions away, and that given their complexity, and the history that is rapidly unfolding before us, convincing answer are hard to come by.  Those who support social safety nets, funded by nonconsensual taxation, should be prepared to make their case. In the absence of persuasive evidence, those of us who highly disvalue governmental coercion, will resist.








This entry was posted in Blog. Bookmark the permalink.

9 Responses to Libertarianism, Safety Nets, and Ideal Theory

  1. Far am I from being a philosopher, so please bear with me, but without any coercive taxation, wouldn’t that necessarily lead to anarchy? If you believe some taxation is morally justified to support a minimal state, say a defensive military and court system, what differentiates taxation to that end from a social safety net?

    I’m not terribly disposed towards safety nets myself, but I’m having trouble in my own mind distinguishing a significant difference between the two. It seems that once we accept some form of taxation as moral, we’re all statists; we’re now just arguing over the price.

    • Mark Friedman says:

      Hi Alex:
      Thanks for reading my blog and stopping by with this comment. Your question is quite reasonable, and I do think that an essential feature of the state is coercive taxation. In its absence, you would have anarchy, but I do think that such taxation is morally justified for military/courts and (as a theoretical matter at least) for a social safety net for the innocent needy. So, I am a (minimal) statist. But my reasoning here is pretty complex, and hard to summarize in a few sentences. Very briefly, I think that taxes for military defense can be justified because they are required to preserve our moral agency, and without coercion the free rider problem would likely cause a serious undersupply of this good.

      With respect to law enforcement and a social safety net, there is an argument within Nozick’s political architecture for a relatively minimal redistribution of wealth, sufficient to provide police protection and food/housing/basic medical care, for those who would otherwise be unable to purchase it for themselves. I try to spell this out in a little more detail here: // and in more detail in my book.

      • Mark,

        I appreciate your taking the time to respond. I read the link provided, as well as the interesting comments below the blog post.

        Quite clearly, deterring or preempting foreign attacks and international terrorism promotes rational agency in a way the basic scientific research does not…Therefore, the coercion of rational agents to support national defense is an exception to Nozickian side constraints because it can be justified in terms of the very value, rational agency, which generates those constraints. NLP, 95.

        This in particular was clarifying for me. While I think the line you draw could still be considered arbitrary, to some, it’s one I am willing to accept.

        Thanks for hosting such an interesting blog; I look forward to reading your posts.

        • Mark Friedman says:

          Hi Alex:
          Thanks, and I salute you for the interesting blog you are running over at EC. As you can probably infer from my quotation of the Talmud in the banner, I am religious in my own way. All the best.

  2. Jake says:

    I wonder what you would think of the concept that similar to education tax credits, the charitable deduction should be expanded to a tax credit to promote competition among private and public welfare/charity programs.

    • Mark Friedman says:

      Hi Jake:
      If there was a way for taxpayers to designate some of the taxes they now pay to fund the welfare state to go instead to private organizations for the support of the needy, I would be all for it. Ideally, I would outsource the entire anti-poverty effort to private organizations, who would compete for tax dollars based on their meeting objective performance criteria.

      • Jake says:

        The reason I ask that is that many libertarians want to abolish the charitable deduction in the name of tax reform (which I support) and argue that a flat tax without deductions would be sufficient to increase income and therefore charitable giving, but advocate education tax credits in the same breath, and that sounded like a contradiction to me.

        Also, what you seem to be describing sounds like faith-based initiatives, do you think there’s a way to get around that without violating Separation of Church and State?

        • Mark Friedman says:

          Hi Jake:
          Thanks for the clarification. I do favor a flat tax with no deductions, credits, etc. And, I also favor, if necessary, taxpayer funded vouchers for the children of the needy who can’t afford to educate them. However, I do not see this as a contradiction, because all moral principles have limits. The disvalue I attach to coercion is quite high, but the right to justly earned income is not stringent enough to withstand the suffering of innocent children.

          In my anti-poverty scheme, the bidding would be open to all groups, i.e. profit, non-profit, religious, secular, etc. So, just as vouchers have so far withstood constitutional, scrutiny, I would hope this plan would too. But, I am making a moral argument here, and we could always amend the constitution.

          • Jake says:

            The tax credits I’m referring to are non-refundable, so they’re not really vouchers. I think they are better than vouchers because more people support them than vouchers, and they usually have less regulations attached to them.

            And the bidding your referring to isn’t the same as vouchers, because at least recipients can choose where their voucher goes, this is direct government funding of charities (including religious ones) and this can be problematic when dealing with spiritual services provided by eligible charities.

Leave a Reply

Your email address will not be published. Required fields are marked *