The “Great Enrichment” is suffering from success. A significant problem for poor people in developed nations is obesity. And the problems of extreme poverty have largely been banished in every nation that connects with the worldwide market for trade. The result is that the main talking point for leftist politicians is an outrage called inequality, when an actual concern for the poor would more properly be focused on poverty.

But inequality is still the big concern, though it often manifests as the sin of envy, dressed up in fancy “social justice” clothing. One of the most intellectually coherent defenses of limitations on the degree of wealth inequality was John Rawls, especially in his 1971 book A Theory of Justice. The Rawlsian “veil of ignorance” is often used as a bludgeon in debates over inequality, as if it were self-evidently correct. Anyone who disagrees “needs to read Rawls.”

I have been teaching Rawls to political economy classes for the past 25 years, and I have tried to find a way to communicate the basic logic in a way students can understand and think about. Most of the people who cite Rawls apparently don’t understand the actual argument. And for those who do understand the argument, it is important to recognize that there is an empirical problem the maximin assumption — and an implementation problem the assumption that patterned, end-state systems of justice are achievable, inert, and permanent, but also consistent with liberty. I claim that these two unstated premises, together, render the classical Rawlsian approach irrelevant for considering intellectually serious problems in the real world of institutions, and wealth distribution.

Behind The Veil

As I have described in several places, I have an “exercise” I do with students when I teach Rawls. To illustrate “the veil of ignorance,” I buy 150 North Carolina “Education Lottery” scratch-off tickets, so that I have enough for all the students in the large “PPE Gateway” class that I teach.

The premise of the exercise is that we each have an “outcome” in the world of realized institutions after we all scratch off the ticket and see what we’ve got. But before we scratch off the ticket, we (I should say “WE” because since this is a philosophy class, we are assuming that there is a “WE” that decides these things, for some reason) have to decide what rule will be used for the distribution of the wealth that is arbitrarily allocated by the lottery.

The analogy, of course, is trying to capture the Rawlsian claim that all differences in wealth result from random — and therefore morally irrelevant — variations in inheritance. If you inherit wealth, you didn’t earn it; if you inherit character and a work ethic, that is because you were lucky and were born to parents who raised you that way; none of these have anything to do with you, but are just the product of chance.

I let the students discuss the possibilities, and impose (in the interest of time) a rule that if a majority votes for a rule, then that rule is chosen as being binding on the entire group. The two alternative decision rules are always some versions of the following:

A. Everyone keeps his own lottery ticket. It’s fair, after all, in the sense that it’s random, and everyone is equally likely to win or lose, because the probabilities are the same for each ticket.

B. Everyone puts her lottery ticket into the commonwealth, and the proceeds are then divided equally across all participants.

In every case I have run the simulation, alternative “A” (accept random inequality) wins by a large margin. There is something about holding the ticket in one’s hand that creates a sense of ownership: “this is my ticket.”

Now, the chances of winning much money, for any individual, are slight. (The expected value of a scratch-off ticket appears to be about $0.25, which of course makes the price tag of $1.00 seem pretty steep. But remember, it’s an education lottery, so that’s okay, I guess.) But in a group of 150-ish, the chances are that 15 or more people will win something, and one or two people might win a reasonable prize, according to the published odds list. 

Specifically, then, I handed out the cards to all the students present that day, and then said, “Before we scratch off the cards, we have to decide how to divide the proceeds. As it stands, each of you has the same expected payoff. But after we scratch off the covering, it will be revealed that some are rewarded and most are not. These differences may be quite large, and they are surely morally arbitrary, given the random distribution of cards.”

Proposal A is always the winner, and then I ask the students to scratch off their cards, and say that we were going to go around the room so that each student can share the news of his or her winnings. Now, it is possible — though unlikely, since the odds are worse than 1 in a million — that someone might win the top prize just by chance. But I cheated, and had enlisted the aid of a helper in advance. I always ask one of the students to wait about 10 seconds after scratching off the ticket, make some kind of squawking noise, and then jump up and wave the ticket while shouting, “I won $10,000! I won $10,000!” The actor then is scripted to run out of the room, yelling excitedly.

And then….this has never failed, folks….one of the students raises a hand, and asks, “Can we vote again?” I always enlist a second actor-student to wait 30 seconds and ask the question, just in case, but so far I have not had to use them, because it happens on its own.

I ask “Why? We already voted!”

The students, many of them by now, all give some version of the “Rawlsian” answer: well, differences in wealth are morally arbitrary, and fairness demands that morally arbitrary effects, differences that are not morally deserved, should be minimized. It’s just “much more fair” to collectivize wealth, and then to distribute it equally.

I then text the actor-student, who is waiting outside. He or she returns, and bows, and explains the subterfuge. And then I make two points, the empirical point and the implementation point, that I mentioned above.

Empirical Point:

The Rawlsian logic depends crucially on the empirical claim that people in the setting of the “original position” would overwhelmingly choose the set of institutions that would deliver the best outcomes for the worst-off. But decades of experiments (work much more serious than the classroom simulation I have described here!) have shown that almost no one chooses the “maximin.” Without this claim, Rawls’ support for the difference principle is simply an assertion of an ethical intuition, and is quite unpersuasive.

Implementation Point:

The point of the exercise is that the notion of the end-state pattern will always be subject to reevaluation and ex post recontracting, if the rules allow that. So, far from representing an outcome chosen behind the veil of ignorance, political pressures will always be driven by full knowledge of the benefits realized by self-interested actors, even if they mouth formulaic invocations of “social justice” while they do it.

The simulation exercise allows students to (re)discover the fundamental problem of fair rules, rather than focusing on fair outcomes. Ex ante agreement on rules is not the same as ex post acceptance of outcomes. Political philosophy should follow Buchanan and Tullock in focusing on the “demand” for rules, or the set of institutional arrangements that allow us to expect to capture gains from cooperation or exchange.

But then the problem is the highly constrained “supply” of stable, workable sets of rules; rules that are self-enforcing and universally accepted. As Jeremy Bentham pointed out in his Critique of the Doctrine of Inalienable Natural Rights:

In proportion to the want of happiness resulting from the want of rights, a reason exists for wishing that there were such things as rights. But reasons for wishing there were such things as rights, are not rights; — a reason for wishing that a certain right were established, is not that right — want is not supply — hunger is not bread.

In the “original position,” far and away most people choose the lottery, violating the prediction of maximin. And after the results are announced, people who did not win want to change the rules to the system that they rejected when it was fairly presented. It is only when they know their own interests that they seek to invoke “fairness,” ex post. But this is precisely what the veil of ignorance was supposed to prevent: a “right to equality” will be rejected by free citizens, because they want something better. The Rawlsian system is incoherent.