Freedom, Markets, and Well-being Fall 2018

Dworkin’s Auction


Dworkin’s project is to spell out what it would be for a society to treat its members as equals. He asserts that this is possible only if a society devotes an equal share of resources to the lives of each member of the society.1 This paper is devoted to stating what it means to devote an equal share of resources to the lives of every member of society.

One surprising feature of Dworkin’s article is that it makes the case for economic markets on the grounds of equality. Markets are usually described as the source of inequality. But if Dworkin is right, there is no way of saying what having an equal share of resources involves outside of a system of market exchanges. That is a novel way of thinking about the relationship between markets and equality.

Dworkin’s auction

The idea is that the members of a society have equal shares of resources when what Dworkin calls the “envy test” is satisfied. This does not mean that the society is free from the emotion of envy. What it means is that no one in the society would trade their bundle of goods for anyone else’s bundle. He maintains that the way to satisfy the envy test for the initial distribution of resources in a society is to auction the available resources.

Dworkin believes that shares of resources should be sensitive to two aims:

  1. The tastes and ambitions of the people who get the resources.
  2. The tastes and ambitions of the people who do not get the resources.

A simple kind of equality would give everyone exactly the amount of exactly the same thing. That’s equal! But, given the diversity of tastes and ambitions people have, it is unlikely to be a satisfactory way of devoting an equal share of resources to each person’s life. As Crystal noticed, it falls short of the first aim.

We could try to deal with this problem by giving people whatever they need to do whatever they want; we could try to make everyone equally satisfied with their shares, in other words. If we did that, we would give people with expensive tastes and ambitions a lot more than people with more modest aims get. Dworkin thinks that would run afoul of the second aim since the share you get depends on what you want without taking anyone else’s interests into account.

The auction, by contrast, is supposed to achieve both aims.

  1. People bid on the things that they want.
  2. The price that they pay to get what they want depends on how much others want them; in the economist’s terms, it reflects the opportunity cost of each person’s getting the resources that they get.

If the things I want are in high demand, I will have to pay a high price for them to outbid others. By contrast, if what you want is in low demand, you will get it at a low price.

So you could wind up with a lot of what you want and I could wind up with a little of what I want and we would have equal shares of the society’s resources. That is because the value of our bundles is determined by how much others bid for them. Thus my tiny apartment in Manhattan could be equal in value to your large ranch in Wyoming.

Is the Auction Determinate?

Prof. Brown demonstrated that the auction is indeterminate. Markets could clear at different prices. In her example, Bob and John have equal shares when Bob has 4 baskets of eggs and 1 bottle of wine and John has 2 baskets of eggs and 5 bottles of wine or when Bob has 4 baskets of eggs and 2 bottles of wine and John has 2 baskets of eggs and 4 bottles of wine.

Is that a problem? It’s hard to say.

On the one hand, Dworkin can consistently say that either set of prices could establish equality of resources. After all, the envy test is satisfied either way as neither Bob nor John would trade his bundle of resources for the other guy’s bundle in either case. The fact that there could be two ways of meeting the envy test shows that there are at least two ways of achieving an equal distribution of resources.

On the other hand, the different price schedules make a difference in how satisfied Bob and John are by the outcome of the auction. The price schedules are not subject to the envy test and so Dworkin’s theory cannot say that one is better than the other. Since the choice of price schedules does make a big difference to Bob and John in our example, that appears to be a significant gap in the theory.

Luck and Insurance

Suppose we complete the auction. What happens when we turn people loose in the real world?

Remy noticed one thing that will happen: people’s tastes will change. By itself that doesn’t raise a problem for the envy test. The envy test fails only if you are willing to trade your entire bundle for someone else’s bundle. The fact that you no longer like the potatoes you bought in the auction doesn’t necessarily mean that there is another total bundle that you would prefer to the spud-laden bundle that you have. Leaving that point aside, I think Dworkin’s idea is that many of these cases can be accommodated through a functioning market economy. If you decide you don’t like bicycling so much and would rather play music and I decide that I would rather exercise than stay inside practicing, we might have a mutually beneficial trade available that would enable us to be more wholly satisfied with our bundles.

Tommy hit the sort of thing that Dworkin himself worries about: luck. The crops could fail, my instrument could break, bankers might have crashed the economy by making crazy bets on the mortgage market that cost me my job, … and so on. If Abel gets hit by a string of bad luck, he might want to trade his bundle for someone else’s. And if Baker has a string of good luck, other people will want to trade their bundles for his. In either case, the society won’t pass the envy test.

What’s the solution? Crystal for the win: insurance! When you have the option of buying insurance against suffering a loss, you can convert your brute luck into option luck. Suppose Professor Brown and I go gambling, we play the same game, she wins, and I lose. She has more chips than me, but that doesn’t raise a problem for the envy test. If we compare the resources we had going into the casino plus the gambles we took, we’re still equal: we took the same gamble, just with different outcomes. Because the outcomes reflect our choices, they don’t fail the envy test. Insurance is a kind of gambling, so we use it to convert bad brute luck into option luck. If you chose to take a trip rather than buying homeowner’s insurance, the fact that your house burns down doesn’t by itself make the difference between meeting the envy test and not meeting the envy test. If you preferred your bundle of resources with a trip but no insurance to my bundle with insurance but no trip, then the fact that you lost your insurance gamble doesn’t mean that we fail the envy test, even though you would now rather have my house than your pile of rubble.

Does insurance solve all of the problems we might worry about? Peter N. says no. There are conditions you cannot buy insurance against, especially ones that have already happened. So people who are born with congenital handicaps can’t buy insurance against being born with handicaps. No one would sell it to them because there is no risk to insure against. The cost of the policy would have to exceed the benefit to be paid. But if I’m born with a condition that means I can’t hold a paying job, I will quickly come to envy the bundles of resources held by other people and there is no plausible way of saying that my bundle is so small because of my choices and tastes.

Dworkin’s solution to this problem is hypothetical insurance. We are supposed to imagine what insurance policies people would have bought if they did not know whether they were born with a debilitating condition or not. Then we take the average policy, charge everyone the premium needed to cover its cost through the tax system, and pay the benefit to everyone who has the relevant condition. Once the benefit is paid, according to Dworkin, we will have equal resources again.

I can see two kinds of objection to this. Some people might object to the idea of hypothetical insurance. “If I don’t choose to buy hypothetical insurance, how does it convert bad brute luck into bad option luck? I never exercised any options!” Others might object to the possibility that the average policy would not pay enough to cover a disabled person’s needs. “The point of this was to treat everyone’s lives as equally important. If we can’t guarantee that this class of people have a roof over their heads, food to eat, and clothes to wear, then we aren’t really treating them as if their lives are as important as everyone else’s lives are.”

One thing Peter N. said at the very end of class is worth flagging here. There are lots of other ways of taking care of the handicapped: families, churches, charities, and so on. More generally, I think it’s legitimate to say that an equal share of resources may not be what is morally required. There is more to morality than equality, after all. So you might say that society owes more than an equal share to the handicapped out of compassion, humanity, benevolence, or charity (pick your favorite term). The fact that Dworkin’s theory of equality doesn’t cover everything you think a society should do just means that equality is only one value among others.

Jeremy had a very interesting point about how wealth can enable some people to predict congenital health problems that raises a host of complicated issues. At this point, I have gone on way too long. I will close by saying that we’re going to get back into the hypothetical insurance game on Thursday when we talk about insurance and the labor market. This is where we will get into the welfare state.


Dworkin, Ronald. 1981. “What Is Equality? Part 2: Equality of Resources.” Philosophy & Public Affairs 10 (4): 283–345.

  1. Pay attention to that initial step that moves from treating people as equals to devoting equal shares of resources to their lives. This is where Anderson is going to say Dworkin’s project goes wrong.↩︎