Consequentialism, Libertarianism

What Sort of (Libertarian) Consequentialist are You?

Last post, I claimed that lots of libertarians are consequentialists. In this post, I want to try and state the kind of consequentialism I think most libertarians would sign on to if asked. This is the conception of consequentialism that is “implicit” in the writings of most (non-philosopher) libertarian consequentialists.

Let me be circumspect: many libertarians are consequentialists about politics if not morality generally. They think that coercive laws and policies are justified when they produce good consequences for people. Such libertarians are skeptical of additional moral criteria, like natural rights. Their mantra is “Maximize efficiency” but the mantra obscures a lot, as consequentialism can be qualified in many ways. The purpose of this post is not merely to articulate the version of consequentialism most common among libertarians but to introduce you to the genus consequentialism if you are interested.

My bet is that most libertarian consequentialists are what I shall call AMORE-3 utilitarians (“Three Love” utilitarians, if you like!). They are average, maximizing, ordinalist, rule, expected, egalitarian, efficiency utilitarians. Let’s get clear on what this means.

Nearly all of these definitions come from this encyclopedia entry, though they are modified to concern when political coercion is permitted.

Before I begin, note: everyone values good consequences; arguing for a law or policy based on its consequences does not make you a consequentialist. And so:

(1)   Consequentialism: whether coercion is permitted depends only on consequences (as opposed to the circumstances or the intrinsic nature of the act or anything that happens before the act).

The definition does not by itself require a maximizing response to consequences.

(2)   Maximizing Consequentialism: whether coercion is permitted depends only on which consequences are best (as opposed to merely satisfactory or an improvement over the status quo).

Maximizing is often contrasted with satisficing (doing well but less than the best). I think most libertarian consequentialists are maximizers partly due to their economics background which may lead them to think that satisficing is generally the most efficient way to maximize.

(3)   Expected Consequentialism: whether coercion is permitted depends only on expected not actual consequences.

Libertarian consequentialists want consequentialism to be action-guiding. Accordingly, we should select among expected consequences not objective/mind-independent consequences.

(4)   Egalitarian Consequentialism: in determining whether coercion is permitted, benefits to one person matter just as much as similar benefits to any other person.

As Bentham said, each person is to count for one and not more than one.

(5)   Indirect (ultimately, rule) Consequentialism: whether coercion is permitted depends not merely on the value of the consequences of the act of coercion in question.

Indirect consequentialism opposes direct consequentialism, which holds that whether coercion is permitted depends only on the consequences of the act in question. Indirect consequentialism is generally contrasted with act-consequentialism. Libertarian consequentialists tend to focus on the evaluation of institutional rules and laws and probably want to avoid the drawbacks of act-consequentialism. And so:

(6)   Rule Consequentialism: whether coercion is permitted depends on the general consequences of a rule.

Libertarian consequentialists understand rules differently. They might mean a specific law or something as broad as an entire system of coercive law. They also reject other forms of indirect consequentialism such as virtue consequentialism which evaluates things based on the consequences of possessing some set of virtues.

(7)   Utilitarian Consequentialism: whether coercion is permitted depends only on the utility of the coercion in question.

I would say 99% of libertarian consequentialists are utilitarians. The thing to be maximized is utility not something more objective like The Good. What utility consists in is a matter of dispute in philosophy, but most libertarians think that utility consists in the satisfaction of preferences. Libertarian consequentalists are willing to put just about anything people prefer into a utility function, including stuff like having a nicer car than your neighbor and consuming the bitter tears of despair of your enemies. Libertarian consequentialists are nigh universally subjectivists about value – whether a state of affairs has value depends entirely on whether it is valued by at least one agent. So utility is understood in terms of subjective valuing.

(8)   Ordinalist Utilitarianism: whether coercion is permitted depends only on ordinal utilities.

Ordinal utilities are comparative and non-numerical. Libertarian consequentialists tend to be skeptical that individuals have cardinal utilities (how would we number the scale?). They’re also skeptical of interpersonal comparisons of utility and so reject any cardinal scale of utility that applies across persons. In general, libertarian consequentialists think we should select those laws that will move people up their utility functions in terms of their rankings.

(9)   Average Utilitarianism: whether coercion is permitted depends on the average utility among the relevant group’s members.

Average utilitarianism is contrasted with total or aggregate utilitarianism which holds that we should maximize utility for the entire set of persons. But Derek Parfit developed the famous “mere addition” paradox for total utilitarianism which holds that you can maximize utility simply by adding persons to the universe with lives whose utility is just barely positive. This leads to the “repugnant conclusion.” One response to the repugnant conclusion is to change over to average forms consequentialism. It doesn’t really avoid the repugnant conclusion, but some people think it helps. It avoids mere addition but still runs into its own problems (like preferring a 1 person/100 utils society to a million people with 99 utils society).

I don’t really know how libertarian consequentialists want to go on this one, but I figure they want to avoid the repugnant conclusion insofar as they can. Perhaps they want to throw in some kind of minimally decent life threshold standard as well. Tyler Cowen has some thoughtful reflections on these matters. Robin Hanson is pretty friendly to the repugnant conclusion, so he’s one exception. So if you’re worried about average utility and friendly to the repugnant conclusion, then perhaps “A” should stand for “aggregate” where what is to be maximized is aggregate utility rather than average utility.

(10)  Efficiency Utilitarianism: coercion is permitted when it maximizes efficiency.

Libertarian consequentialists often talk about maximizing something called “efficiency,” but efficiency is usually just a stand-in term for utility. When libertarian consequentialists analyze the value of state coercion, they employ efficiency measures that somewhat vary. Some efficiency theorists understand efficiency in terms of the (weak) Pareto criterion, where a distribution is made more efficient when at least one person is better off and no one worse off. But the Paretian standard is pretty demanding because its bars policies that make anyone worse off at all. So for that reason libertarian consequentialists usually endorse the Kaldor-Hicks efficiency criterion. On this view, an outcome is more efficient if those made better off could in theory compensate those made worse off, leaving them better off than before the law or policy was implemented.

Bringing it all together:

(11)  AMORE-3 Utilitarianism: coercion is permitted when the rule on which it is based maximizes average ordinal expected (Kaldor-Hicks?) efficiency.

The egalitarian descriptor is built into the average descriptor and efficiency is understood in terms of utility. And remember, AMORE-3 Utilitarianism concerns only institutional justice, i.e., the proper use of coercion, not right action generally.

I think AMORE-3 Utilitarianism is probably the plurality view among libertarian consequentialists (of course, I don’t know the percentages!). But if I’m right about this characterization, I’ll raise some problems for views in the general vicinity of AMORE-3 utilitarianism in future posts. So again I’d like your feedback.

This is the time for you to get all nit-picky and difficult about the sort of consequentialist you are or that you would be if you were a consequentialist or whether I’m right about the conception of consequentialism most libertarians affirm. Have fun.

  • All right, let me begin with the obvious, maybe at the risk of spoiling one of Kevin’s future posts: the “average” and “ordinal” criteria are irreconcilable.

    • Kevin Vallier

      Well, three things: (a) you might think that with an additional ethical premise you can get cardinal numbers out of ordinal numbers as Harsanyi once contended (, (b) if money is roughly monotonic with utility that you can do the averages in terms of dollars. (b) is a pragmatic but not an in principle deviation from ordinalism. And (c) the Kaldor-Hicks counterfactuals may provide a way to get averages, as people will be willing to pay some amount in order to secure certain gains. But this is the area of consequentialist ethics where economists quickly become towering experts, so I should be careful not to overstep my expertise.

      But you are on to one of my general concerns, namely that a powerful objection to consequentialist moral theory is that its most plausible pieces conflict with one another in really debilitating ways that I think non-consequentialist views can avoid.

  • Anonymous

    Am I wrong to think that the repugnant conclusion is no big deal in the real world because the government won’t seek to continually add minimally preference-satisfied individuals to the population? 

    In other words, does this philosphy classroom example matter to someone who wants to judge realistic public policies?

  • A technical point–economic efficiency may serve as a proxy for maximizing utility, but it isn’t the same thing. The Hicks-Kaldor criterion, or Marshall’s earlier version (which I prefer), measures costs and benefits by willingness to pay. That corresponds to utility only if we assume that everyone’s marginal utility of income is the same–that a dollar represents the same amount of utility to a beggar and to Bill Gates, which is implausible.

  • I’m trying to get a handle on this article because I think there are some important points being made, but the language reminds of the title of an article Matt posted a few days ago entitled:  “Consequences of Erudite Vernacular Utilized Irrespective of Necessity: Problems with Using Long Words Needlessly.”

  • Kevin Dick

    I’m a non-philosopher libertarian with an econ background.  I’ll just throw in my vote that these 9 properties are all desirable.  However, as Jacob Levy and David Friedman have pointed out,  it may be the case that we’ve got an Arrow’s Theorem problem and there is no formal ethical system that can meet all 9 criteria or that such a formal system would have to violate some other obvious criterion. 

  • Glen Whitman

    “Maximizing is often contrasted with satisficing (doing well but less
    than the best). I think most libertarian consequentialists are
    maximizers partly due to their economics background which may lead them
    to think that satisficing is generally the most efficient way to

    I think this requires clarification.  Satisficing is not always maximizing.  But it does turn out to be maximizing in the context of a dynamic search.  For instance, if you’re searching for a job and the process of searching is costly, then your optimal search rule will dictate accepting any wage offer equal to or greater than your reservation wage, where the value of the reservation wage is determined by the parameters of the model.  At that point, your job search would stop.  (Of course, this result depends on some assumptions that I’m not stating here — such as not caring about non-wage characteristics of the job.)

    In the context of choosing the rules of a just society, it’s not obvious to me how search is involved.  I suppose your philosophy could be to some degree agnostic about the right way to organize society, and changes in the rules over time could be characterized as a search process.  Come to think of it, that would be an intriguing approach.  But it doesn’t sound much like any position I’ve heard in discussions like this. 

  • I suspect that, in addition to the general kind of embarrassing-counterexample-avoidance reasons people have for moving toward rule utilitarianism, utilitarian libertarians will have a reason of an additional sort. Namely, it’s hard to see how one could defend a strict minimal-state form of libertarianism if each act had to be evaluated on the basis of its particular consequences. Making that argument would seem to require showing that the consequences of exceeding the tight confines of the minimal state could never be (expected to be) superior to the consequences of remaining within it, not even on rare occasions. That’s an empirical claim, and I guess there’s nothing logically incoherent about it. But it does strain credulity.

    Still, though, there are other ways of building a concern with rules rather than acts into one’s consequentialism than going full-bore rule-utilitarian, right? I haven’t kept up on the literature on this topic well, and Tyler Cowen had what looked to be an interesting piece defending RU in a recent Social Philosophy and Policy. But given that the view is thought to have insurmountable difficulties, utilitarian libertarians might find a kind of rule-sensitive act-consequentialism a more attractive option. I have in mind something along the lines of Mill’s position, assuming (as I do) that it is incorrect to read him as a rule utilitarian strictly speaking.

    • Anonymous

      For those of us a little rusty on our Mill, could you please briefly describe what rule-sensitive act consequentialism looks like. Thanks.

      • Well, briefly, a rule utilitarian claims that rules define right action. The right action is the action specified by the rule which, if generally observed, would maximize utility. Rule-sensitive act utilitarians will think that we should often follow rules when deciding how to act, but that in cases where the rule leads us to do something non-optimific, we will have acted wrongly. Blamelessly, perhaps, but wrongly.

        David Brink’s essay on Mill at the SEP gives a nice overview: 

        • Anonymous

          OK, thanks.

    • Kevin Vallier

      There are lots of rule-sensitive alternatives to rule-utilitarianism, just invent an alternative maximand for indirect consequentialism and you’ll probably find one (virtue-consequentialism would be sensitive to rules and so might a very coarse-grained institutional consequentialism), but my impression is the act-consequentialism is more obviously problematic than rule-consequentialism, not the other way around, especially given how utterly poorly it sits with our considered judgments.

  • I’m the type of libertarian consequentialist who believes that libertarians like Ron Paul should have to face the consequences of all that crap he put out under his name in the ’80s and ’90s. ;o)

  • I’m the type of libertarian consequentialist who believes that libertarians like Ron Paul should have to face the consequences of all that crap he put out under his name in the ’80s and ’90s. ;o) 

  • Andrew Prock

    I think you’ve put the cart before the horse.  Since all laws, polices and social mores are coercive, it’s not clear what you mean by “permit coercion”.  If you want to have any kind of society, you’re going to have to “permit coercion”, and the degree to which it is permitted is a function of many things.

    I think you need to take a step back and discuss coercion with a little more precision, and not assume that everybody understands it the same way.

    • I define coercion in this case as aggression under color of official warrant, as in, “Comply or we’ll fine you. Don’t pay the fine, we’ll imprison you. Decline to submit to imprisonment, we’ll kill you.”

  • Anonymous

    I am a little confused here, because #1 is almost exactly the opposite of what I think of libertarianism. Let’s take drug laws, for example.

    A consequentialist approach to drugs might be to say ‘Drugs are harmful to society and hence we need to ban them’, or it could be flipped around (as I would argue) and say ‘The laws against drugs are so impossibly harmful at multiple levels of society that at this point it’s inconceivable that legal drugs could do more harm’.

    However, that’s not the argument I tend to hear from libertarians, which is ‘The government has no right to tell us what to put in our body’. Now, I actually have sympathy for this position, (I think the government’s regulation of people should basically stop at our skin, or at least at our brain chemistry.) and agree as long as it’s a general policy and we all agree it can be restricted based on time and place locations. For example, no drugs while operating motor vehicles, and if you want to do PCP, you have to do it in private.

    So that’s all well and good, but the problem is, ‘The government has no right to tell us what to put in our body’ is not a consequentialist approach to the law. It, like almost every libertarian argument I’ve ever heard, is rights-based approach, i.e., it is based on the ‘intrinsic nature of the act’, to quote above. It’s a rights-based approach that even I (And I’m a progressive who has very little patience with the idea that the government is ‘not allowed’ to do things that the recipients think are beneficial.) agree with, but it’s still a rights-based approach.

    So I know I’m a little late to this discussion, but if someone could give some sort of example here of a libertarian consequentialist approach.

  • Pingback: Against Utilitarianism and Self-Ownership Defenses of Libertarianism | Ellen Wolchek()