Moral arbitrage

by John Q on January 7, 2009

I’ve been planning for a while on a post motivated by the discussion of trolley problems a while back, but recent discussions have raised some more serious examples (the Iraq war, Gaza and so on).

Looking at the discussion, it seems as if nearly everyone is concerned about the (foreseeable) consequences of their actions, but there are a lot of claims that some consequences should be treated differently from others (intended vs unintended, direct vs intermediated by the predictable reactions of others, and so on).

To an economist, what this naturally suggests is the possibility of moral arbitrage.

Opportunities for arbitrage arise when the same good (or financial asset, or moral consequence) is priced differently in different markets. Someone who can buy in a market where the good is cheap and sell where it is dear has the opportunity for arbitrage profits.

So, if you want to raise the moral value of a particular action, what you need to do is make sure that the positive aspects of the action are valued in markets where the price is high, and the negative aspects where the market is low. For example, an advocate of the Iraq war can be a virtue ethicist as regards their own heroic standard against Ba’athist dictatorship, a deontologist regarding obligations to punish the criminal behavior of their enemies, regardless of the unintended effects on the millions of people living in the general vicinity, and a consequentialist regarding the necessity to excuse the criminal behavior of their leaders for fear of subsequent bad effects on the polity.

As this example shows, with arbitrage opportunities, all sorts of things can be made possible. A consistent virtue ethicist (for example, a Jeffersonian) might reasonably conclude that the criminal behavior inevitable in a long occupation of a largely hostile country is unacceptable to someone who wants to maintain a virtuous disposition. A deontologist would object to violations of well-established principles of just war theory. A consequentialist would certainly conclude that the foreseeable costs of a war exceed the benefits. But a moral arbitrageur can mix and match these principles to reach a conclusion none of them would individually support.

The trolley-crash toy examples seem perfectly designed to encourage moral arbitrage of various kinds. By shifting consequences between direct and indirect, intended and unintended, close and remote, it seems as if moral virtue can be claimed for any course of action you like.



dsquared 01.07.09 at 12:30 pm

I think that the problem is not so much the active traders or arbitrageurs, but rather the number of people who have decided that they can’t beat the averages and so decide to simply “buy and hold” an index of ethics, provided by such low-cost suppliers as the State of Israel, the Socialist Workers’ Party, etc.


engels 01.07.09 at 12:47 pm

I think Alasdair MacIntyre noticed this phenomenon first, and he didn’t think it was confined to Israeli partisans or Trotkyists…


Barry 01.07.09 at 2:22 pm

dsquared, it’s pretty much the same – another metaphor is holding a hand of cards, and being able to pull the one which is of high value at the moment (trump or ace high or king high,…)


Anderson 01.07.09 at 2:52 pm

Dunno on arbitrage, but the bit about foreseeability reminded me of a Valéry quote:

“With effects so rapidly becoming independent of their causes, and even antagonistic to their causes, perhaps it will now be considered puerile, dangerous and insane to seek out events — a habit essentially due to history and sustained by it. It is not that, in the meanwhile, there will no longer be events and monumental moments; there will be prodigious ones! But those whose function it is to await them, or prepare them, or to ward them off, will of necessity learn more and more to beware of their results. It will no longer suffice to combine the will with the ability in order to undertake some enterprise. Nothing has been more destroyed by the last war than the pretension to foresight.”

No big Valéry readers in Hamas or the Israeli gov’t, I’m thinking.


Picador 01.07.09 at 3:30 pm

This applies to political economy as well as moral philosophy, of course. Gore Vidal’s adage about free enterprise for the poor and socialism for the rich is a pretty succinct example: one shifts one’s economic frame depending on the group under consideration.


Alex 01.07.09 at 4:16 pm

For example, an advocate of the Iraq war can be a virtue ethicist as regards their own heroic standard against Ba’athist dictatorship, a deontologist regarding obligations to punish the criminal behavior of their enemies, regardless of the unintended effects on the millions of people living in the general vicinity, and a consequentialist regarding the necessity to excuse the criminal behavior of their leaders for fear of subsequent bad effects on the polity.

Quite often, within the same comments thread.


bianca steele 01.07.09 at 4:22 pm

“Positive and negative aspects” can be anything, though. For example, if I believe that my reputation in an Internet discussion group has fallen so low that younger posters are demonstrating their prowess by taking apart my arguments in other groups I don’t read, then presenting third parties’ arguments against “their” arguments as their arguments against mine, a cost-benefit analysis might lead me to pretend to opinions that I believe so ludicrous no intelligent person would even take them seriously. (I don’t do this — not anymore :| .)

But zounds, I’m on a different page than the other commenters here.


John Emerson 01.07.09 at 5:32 pm

I’m convinced by now that consequentialism is not ethics but a corrective or override to ethics. Where there’s no conflict (where the “right” thing to do by some given standard coincides with the one that will bring the best outcome, or where the “wrong” thing to do brings a bad outcome) you don’t need consequentialism. Consequentialism only is meaningful where there’s a conflict. And what happens is that you’re faced with a choice — how much ethical cost to pay for a good outcome, and how much consequential cost you’re willing to pay for ethics.

Consequentialism strikes me as most valid in a political or administrative context, where you have to deal with the differences between the acts of leaders of large groups and the acts of individuals facing other individuals in communities. But even then, omelette-making argument is often horribly overused.

Consequentialism is also often used as a kind of sophistry by bright, argumentative gameplayers to put inarticulate moral actors at a disadvantage, often for some self-serving reason. The accusation of irrationality is used aggressively. Lifeboat ethics is the worst case of this, and self-serving people eat it right up.

In the worst cases, the rational calculation is a purely hypothetical one, and the ability to predict outcomes is imaginary. The sole use of these arguments is to soften people up by getting them to admit that doing bad things for good purposes is not only permissible, but a duty.

The crime of Leninism was just in setting up a competing state, of course. All states are omelette-makers. A loyal citizen of any modern state must grant that under some circumstances bombing day care centers is permissible; in the worst case, to insist otherwise might amount to demanding that your nation be left helpless in the face of its enemies.


Paul Gowder 01.07.09 at 7:34 pm

I’m not sure I get the point of moral arbitrage. My comments became too long, so I sprung them out into a post of my own, but the basic worry is that there are two readings of the claim “moral arbitrage exists.” On the first, it’s a valid form of reasoning, and that can’t be true. On the second, it’s a human psychological blunder, in which case, of course it is, it’s old-fashioned wishful thinking.


Barry 01.09.09 at 2:38 am

Wow. An Israeli-Palestinian thread with only 9 comments. CT isn’t measuring up – or down, perhaps.


J Anderson 01.09.09 at 5:06 am

I don’t think I’m grasping how one can “be” a virtue ethicist and “be” a deontologist and “be” a consequentialist. If one understands each of these to be a moral outlook based on some underlying principle(s), and believes that to “be” one is to adopt those underlying principles and use them to arrive at moral judgments, then one might not be able to “be” all of them at once. For example, if to be a consequentialist is to accept the notion that acts are to be judged solely by their outcomes, while to be a deontologist is to accept that acts are to be judged not by outcomes but by some inherent property of the act itself, then you can’t be both. On this understanding of what it is to “be” something, arbitrage is just inconsistency.

But this objection (which Paul Gowder raises in a somewhat different way in his post) seems so obvious that I wonder if Quiggin has something else in mind: a sort of moral pluralist who has managed to cobble together elements of virtue ethics, deontology, and consequentialism into a coherent outlook. For example, one could be a consequentialist to the extent that one believes consequences matter but not to the extent that they are all that matters, thus leaving room for other considerations such as character and duty. Was that it, John?


John Quiggin 01.09.09 at 7:33 pm

I had in mind, for example, the Doctrine of Double Effect, which seems to require you to be a deontologist for some purposes (roughly, intended consequences) and a consequentialist for others (unintended consequences). Actions can then be permitted or not, depending on how consequences are classified, and this seems to create all sorts of arbitrage opportunities as the thread on that topic showed.


Chris Bertram 01.09.09 at 7:57 pm

Well, of course, John I see your point. But it does rather rely on the proposition that people can reclassify at will, just as it suits them to. However any moral outlook can be manipulated for PR purposes by those who are keen to do so. You don’t even need to switch between outlooks. Take consequentialism: if you big-up the long-term or short-term consequences, factor in tiny risks of total catastrophe etc, you can get things coming out to suit you. Bad faith? Sure. But different moral frameworks are only one of the resources available.


Chris Bertram 01.09.09 at 8:00 pm

Oh and whilst DDE may, as you put it “require you to be a deontologist for some purposes and a consequentialist for others” so, in a sense, does any sane moral outlook. After all, consequences matter, but they aren’t all that matters.


John Quiggin 01.09.09 at 10:51 pm

I haven’t fully worked this out, as I meant to say in starting the post. But it does seem to me that the arbitrage idea points up the difficulties in combining moral outlooks and particularly the problems with the (intuitively appealing) idea that consequentialism should be combined with other considerations.

It certainly seems to me that, in practice, reasoning based on DDE is often casuistical (in the pejorative sense of the term) and that the easy availability of moral arbitrage is a large part of the reason.


Chris Bertram 01.09.09 at 11:05 pm

The basic point that you’re making is that in any situation where a range of different principles, criteria and values are in play there will be opportunities for those of bad faith to emphasise what suits their interests, downplay what doesn’t, engage in trade-offs, goalpost shifts etc. That’s a regrettable feature of human affairs, but it doesn’t, as such, give us any reason to believe that there isn’t a plurality of values.


Tom Hurka 01.09.09 at 11:59 pm

It’s also wrong to say the DDE is consequentialist about unintended consequences. (1) Its proportionality rule may require the benefits resulting from even unintended harms to be not just a little but a lot greater than the harms. (2) In weighing benefits against harms it may not count all resulting benefits as relevant, e.g. it may not count as relevant to assessing the proportionality of a military tactic that collaterally harms civilians that it will give pleasure to the bombers or enable their superiors to test military equipment. And (3) it may discount some resulting harms for other people’s wrongful contributing agency, e.g. (very relevantly to the current Gaza situation) if the civilians were only there to be harmed because the enemy wrongfully placed them there to be used as shields. Consequentialism does none of these things, so the DDE’s treatment of even unintended consequences can be far from consequentialist.


John Quiggin 01.10.09 at 1:04 am

I mentioned (3) in the opening sentence of the post, as an example of the kind of thing I find problematic. To push it a bit further, it seems as if, provided we can convince ourselves that particular claims are just (not too hard, when we are judges in our own case) any bad consequences of actions we take to enforce them can be blamed on the wrongful resistance of others.


John Quiggin 01.10.09 at 6:35 am

The discussion has helped to clarify my thoughts a bit, or maybe replace old confusions with new ones. Here’s a provisional restatement. Suppose that, in deciding how to act, we take a range of considerations into account, including consequences. It seems to me that we can do this consistently if either
(i) Some considerations, such as duties or prohibitions, trump consequences; or
(ii) For some kinds of acts, consequences aren’t relevant
My problem arises when a principle (like DDE) allows you, in effect, to evaluate an action by its consequences, but to price different consequences differently. Granted that there is always some possibility of doing this, DDE and similar approaches seem to allow (and even require) it on a much larger scale. That’s an objection from a consequentialist view; I’d think a deontologist would be concerned that these approaches turn duties into high-priced consequences.

Comments on this entry are closed.