What About a Social Minimum?

We discussed Waldron’s criticism of Rawls’s difference principle. Rawls argues that the parties in the original position would prefer his principles of justice over utilitarianism. Waldron ask asks what the parties would do if they had to make a different choice. What if they had to choose between Rawls’s principles and utilitarianism with a floor of a guaranteed social minimum?

Distributive Vs. Needs Based Standards

Waldron’s social minimum is a needs-based (his term) or absolute (mine) standard that is based on what people need. Rawls’s difference principle is a distributive or relative standard that is based on where the members of different classes stand relative to one another. The difference is that a society committed to Rawls’s principles has to redistribute from rich to poor as the rich get richer. A society committed to Waldron’s social minimum only has to ensure that the needs of the poor are met; it can allow the gap between the rich and poor to grow.

In order to illustrate the point, I returned to our old friend: Figure 6 (Rawls 1999, 66). This charts the share of primary social goods (which I will just call “wealth”) held by two groups. The chart is meant to illustrate the effects of incentives: the people in X1 will produce more if they are allowed to earn more than a strictly equal share and this will make the people in X2 better off even if it means they will have less than the people in X1.

Figure 6 from (Rawls 1999, 66); red lines added by me.
Figure 6 from (Rawls 1999, 66); red lines added by me.

The difference principle is a relative standard because it treats inequality as undesirable in itself. It permits inequality, but only so long as it is necessary to raise the wealth of the worst off class. In Figure 6, inequality is permitted only as long as the worse off group (X2 in this graph) is made better off. Once the society hits point a, no additional inequality is allowed because it would not improve the position of worst off group.

In Rawls’s original graph, going to the right of point a would mean gains for the people in X1 but losses for the people in X2. But the difference principle would not allow further gains for the people in X1 even if they did not make the people in X2 worse off. Thus the difference principle forbids moving to the right of point a even if the line were flat (as it is with the red dashes). Inequality either benefits the worst off class or it is not permitted at all.

Work and the Difference Principle

Kenny made a great point about the difference principle. People could belong to X2 in Figure 6 for a variety of reasons. They could be in the worse off group because the skills they have are not in high demand. Or they could be in this group because they do not want to work. (There are other possible reasons, but these two will do.)

Rawls’s difference principle is insensitive to the difference. It does not draw any distinctions between those in X2 who want to work but cannot find a market for their skills and those in X2 who do not try at all. In fact, it is conceivable, though unlikely, that everyone in X2 would be unwilling to work. The difference principle would still require that they be made as well off as possible.

Rawls did not directly confront this point. But one of his remarks about the inadequacy of what he calls “Liberal Equality” (roughly, free markets plus an educational system geared towards equalizing opportunity) seems relevant.

The extent to which natural capacities develop and reach fruition is affected by all kinds of social conditions and class attitudes. Even the willingness to make an effort, to try, and so to be deserving in the ordinary sense is itself dependent upon happy family and social circumstances. (Rawls 1999, 64)

Why do some people have a drive to work while others do not? The difference is due to either nature or nurture: natural or social causes. But these factors are beyond anyone’s control. So just as no one can fully deserve the talents, drive, and good fortune that make them rich, no one can fully deserve the lack of talent, drive, and good fortune that make them poor.

A couple of things are worth bearing in mind here. First, the productive will come out ahead and the unproductive will fall behind even in a society governed by Rawls’s principles. It’s just that the productive can gain only if the unproductive do as well. The unproductive, in turn, are not on their own but rather rise along with the productive. Second, there are a variety of reasons why someone might not work for wages. For instance, some people stay out of the workforce in order to raise children. They are still working, just not for pay. A society may well be concerned about their welfare even if it means making welfare payments to undeserving bums.

With all that said, I think Kenny is right to raise an eyebrow at this. How far can we go with the observation that everyone’s behavior is determined by factors outside their control? Is no one responsible for anything? Could we function as a society if we believed that sort of thing about one another and ourselves? Maybe we could!

Wilt Chamberlain and the Difference Principle

Chris asked whether the difference principle would mean that Rawls runs afoul of Nozick’s Wilt Chamberlain argument. Roughly, it seems that once a society has hit point a, it has to forbid anyone from paying Wilt Chamberlain extra to watch him play basketball since that would move them off of point a. If that is right, then it appears that sticking with the difference principle would have to prevent people from paying Wilt Chamberlain to play basketball and that just looks wrong.

I can’t remember exactly when Chris said this, but I remember that I did not answer it very crisply. I’m going to take another crack at it now.

To begin with, the difference principle governs the distribution of resources across classes. So it does not really apply to one person’s windfall. Wilt Chamberlain’s extra $250,000 would raise the amount held by his class, but not significantly.

But let’s suppose that a large number of people are big winners in the labor market. If they get paid at the rate they can command, that would shift the distribution of resources to the right of point a.

How a society governed by the difference principle would respond depends on what those workers are doing. If their work benefits the rest of society, then giving them the incentive to do it would, in effect, raise the OP curve higher on the y-axis than it would otherwise be. That’s good from the perspective of the difference principle. So the government would seek to identify the combination of taxes and regulations that allows wages that meet their reserve price (what they demand in order to do the job) while trying to capture everything above that price for the rest of society. (It is, of course, notoriously difficult to do this with much precision.)

But if their work does not benefit the rest of society, then the government might well tax away their jobs. For instance, financial firms pay people a lot of money to do things that, from the point of view of society, are completely useless such as paying computer programmers to make trades post microseconds faster. That gives firms an edge on their competitors but it does nothing to improve the efficiency of the stock market. We would get the same information about the value of companies if the trades were made whole seconds or even minutes later than they could be. High rates of taxation or burdensome regulations on socially useless behavior like that would be OK under the difference principle. The relevant “capitalist acts among consenting adults” would not necessarily be prohibited, but they would be rendered unprofitable (Nozick 1974, 163). The programmers will have to make a living writing software for some other purpose and the financiers will have to make do with slower trading.

What about entertainers like Chamberlain? Do their jobs benefit the rest of society? It’s an interesting question. Entertainers certainly produce what economists and utilitarians call utility: people enjoy entertainment! But they do not directly produce what Rawls is concerned with: primary social goods, that is, things anyone would need for a rational plan of life including wealth, opportunities, and a sense of self-worth. However, as Taylor pointed out, there is a lot of economic activity surrounding entertainment: people work in the arena where the games are played, for instance, and people work at their jobs to earn money to spend on entertainment. So Wilt Chamberlain’s work might well benefit the rest of society, even when benefits are measured in terms of primary social goods.

Does that apply to the programmers? Maybe! But bear in mind that the relevant question would be comparative: do programmers working on high speed trading support more activity than they would if they were working on something else? While these matters are complicated and anything is possible, I find it hard to believe that the answer is yes.

Waldron’s Argument

Waldron thinks the best arguments for the difference principle are the ones about finality and stability. According to these arguments, the parties would reject utilitarianism because they know they could be big losers in a utilitarian society. The rules of the original position tell them they must choose principles that they could accept as final, irrevocable commitments (the finality condition) and that people in the society would grow to accept and live by (the stability condition). Since big losers could not keep their commitments to the rules or accept and live by them and the parties know they could be big losers under utilitarian rules, the parties cannot choose utilitarianism. Consequently, they would reject utilitarianism in favor of Rawls’s two principles of justice.

Waldron’s point was that a society that met the social minimum could accomplish the same goal. Such a society would ensure that people were comfortable enough that they would not find it difficult to comply with its rules. But it would not be committed to doing more than that.

Our Discussion

The case for being concerned with absolute standards of welfare is clear enough: poverty is bad and needs are important. The case for caring about relative standards is trickier. We spent most of our discussion time on reasons for caring about relative standards.

For example, Jennifer noted that inequality can reduce the purchasing power of the poor. Roughly, as the rich bid up the price for food and housing, the poor need more money in order to afford to meet their basic needs. So even someone committed to an absolute standard will have to pay attention to some relative measurements.

I mentioned Thomas Piketty’s book Capital in the Twenty-First Century. Piketty has documented rising inequality. He maintains that wealthy countries experienced an unusual period of relative equality in the middle of the twentieth century that he thinks is due to the two world wars they fought. With the passing of the wartime economy and the institutions of social mobilization, we are reverting to historical norms of greater inequality. He believes that the natural outcome will be that capitalist societies will revert to their historical forms as well, with wealthy families controlling almost all the wealth. He worries that this would mean moving away democratic and meritocratic norms to rule by the wealthy and advancement through marriage into great families. We discussed Piketty in this year’s PPE senior seminar; you can read the notes from that session for more information about his theory if you are interested.

Natasha said that one disadvantage of inequality is that it makes mobility more difficult. As the gap between the classes grows wider, it is harder to move from one class to another. As it happens, this is a theme of another author we discussed in the PPE seminar, Miles Corak.

There is a related thought about inequality and mobility that I suspect a lot of people have. Namely, that the rich skew the system in their favor, preventing the poor from getting ahead; the only way to keep this in check is to limit the inequality of wealth that leads to unequal influence over the political and social system. Here the objection might not just be that inequality reduces mobility but that it makes the competition unfair.

In presenting these points, I am not saying that I think they are decisive or even that they are correct. I just want to show that there are reasons for caring about where people stand relative to one another that are different from the ones Rawls gives.

Key Concepts

  1. How a social minimum is different than the difference principle.
  2. Why the parties in the original position might prefer a social minimum to the difference principle.
  3. Different reasons for caring about relative measurements of the gap between classes in society.


Nozick, Robert. 1974. Anarchy, State, and Utopia. New York: Basic Books.

Rawls, John. 1999. A Theory of Justice. Revised edition. Cambridge: Harvard University Press.

Waldron, Jeremy. 1986. “John Rawls and the Social Minimum.” Journal of Applied Philosophy 3 (1): 21–33.