The Canadian philosopher David Gauthier died at the age of 91 some days ago. Gauthier was a widely recognized expert of Hobbes’s philosophy and is mostly known for his contractarian theory of morality essentially developed in his 1986 book Morals by Agreement.[1] The very name of this substack is related to this book, as Gauthier devotes a full chapter to discussing the concepts of Archimedean point and Archimedean choices.
Morals by Agreement is a fascinating book in several respects, even though it is almost now unanimously agreed – including by Gauthier himself[2] – that the contractarian account of morality it set forth ultimately fails. Nonetheless, it continues to inspire moral and political philosophers who very often feel obliged to return to Gauthier’s account, if only to explain why they don’t follow it. Because it uses game theory, Gauthier’s contractarianism has also captured the attention of several philosophically-minded economists, though they generally have harshly dismissed it.[3] Among the many issues discussed in Gauthier’s book that are interesting and worth discussing, one in particular echoes contemporary debates about the justification of high incomes. In what follows, I briefly Gauthier’s account of this issue and explain why I think it is unconvincing.
The question at stake is whether high incomes, for instance those received by professional sportsmen, can be justified. Gauthier’s target is Robert Nozick’s well-known defense of high incomes based on his procedural theory of justice.[4] Taking the example of the basketball superstar of the 1960s and 1970s Wilt Chamberlain – the first basketball player to have earned more than 100000$ a year – Nozick argues that high incomes are fair as long as they are the result of fully voluntary transactions. Chamberlain’s income comes from the voluntary choices of tens of thousands of persons who are willing to pay a significant price to see Chamberlain play basketball. People are willing to pay because (i) Chamberlain is gifted at playing basketball and (ii) people give value to the fact of watching gifted basketball players. Because Chamberlain’s income results from those people’s willingness to transfer their property rights on money in exchange for the opportunity to watch him playing absent any coercion, the resulting allocation of resources is fair, whatever its pattern. Justice is fully procedural.
In Morals by Agreement, Gauthier discusses and rejects Nozick’s procedural account. As a Canadian who is writing in the 1980s, he takes the alternative example of the ice hockey player Wayne Gretzky to make his point. Gauthier’s discussion takes place in his broader contractarian account of morality based on the controversial idea of “constrained maximization.”[5] The basic idea is as follows. In the context of a market society, morality and justice make their appearance as soon as there are market failures or rents (according to Gauthier, perfectly competitive markets are “morally-free zones”). In cases of market failures, the optimizing behavior of rational agents leads to suboptimal outcomes. A moral agent is one who behaves as a “constrained maximizer,” that is, who chooses an action that permits one to reach the optimal outcome. In a bargaining situation, this corresponds to the more specific principle of the minimization of the maximum relative concession made by one of the parties.[6]
What are the implications in the case of Gretzky’s high income? According to Gauthier, Gretzky’s income corresponds to rent. This is so because, beyond some level, Gretzky’s offer of his abilities is totally inelastic to the price: Gretzky’s offer will not increase even if the price increases. This is a relatively common situation, not only on the market for talented sportsmen. Such a rent situation is characterized by the fact that any increase in demand is fully converted into a higher income received by the supplier. At this point, Gauthier departs from Nozick: Gretzky is not entitled to the excess income corresponding to the rent. Gretzky is entitled to this part of the revenue that is needed to make his participation in the economic activity (e.g., playing professional ice hockey) rational for him, i.e., what economists call the participation constraint. Beyond this, everything Gretzky receives should be redistributed to the society. Gauthier’s claim is based on his idea of constrained maximization. Moral agents behaving as constrained maximizers view any decision problem as a “cooperative venture.” Society itself, when inhabited by constrained maximizers, is such a cooperative venture. In such a venture, moral agents must acknowledge that no one is responsible for the initial situation, in particular each agent’s natural and material endowments. Neither Chamberlain nor Gretzky is responsible for the fact of living in a society where people are willing to pay a lot of money to watch gifted sportsmen in action. The fact that their talent is a scarce resource is not an intrinsic property but is fully dependent on the state of supply and demand. In this context, taxing the rent is fully justified and would not affect the functioning of the market.
Gauthier’s view with respect to the status of high incomes, at least when they are related to a rent situation, is thus the opposite of Nozick’s. Both are maximalist in their own rights. Nozick contends that no redistribution is permissible. On the contrary, Gauthier seems to consider that the totality of the rent should be redistributed. There are however several difficulties with Gauthier’s view. Some of them are just related to the inadequacies of his broader contractarian account and its notion of constrained maximization. There is also a problem with the way Gauthier uses the concept of rent. In particular, when measuring the rent enjoyed by Gretzky, Gauthier only considers alternatives outside professional ice hockey. This is obviously a mistake because Gretzky’s salary must cover his opportunity costs, and those include the fact that, instead of playing say for the Edmonton Oilers, Gretzky could be playing for the New York Rangers. If this part of the income is taxed, this will definitely affect the functioning and, possibly, the efficiency of the market.
Nonetheless, the repeated controversies about the high incomes of sportsmen but also CEOs for instance indicate that Gauthier’s view singles out something about the reason why many of us view them as unfair. When high incomes are due to rent situations as described by Gauthier, we tend to consider that those who benefit are not entitled to their situations. This is more generally related to the fact that market outcomes are largely the results of stochastic factors in an uncertain context. This feature of market outcomes has long been recognized by proponents of free markets, such as Friedrich Hayek.[7] But contrary to Gauthier and, for that matter, Rawls, Hayek refutes the idea that the stochastic nature of market processes implies that the resulting outcomes are unfair. Randomness is neither fair nor unfair, it is just a constitutive part of the market order, and more generally of life. In itself, it doesn’t justify any form of coercive interference to redistribute resources.
There would obviously be much to say about this view. Where Hayek has a point, I think, is that attributing unfairness properties to randomness is opening Pandora’s box. Because virtually any outcome of the “game of life” has a stochastic component, there is in principle no limit to the justification of coercive interferences for correcting outcomes – this has of course been a never-ending source of inspiration for dystopic stories. Now, is there a third way between Nozick’s and Gauthier’s maximalist accounts? Let me sketch here such a third way based on the idea of public reason and using Lebron James as a contemporary illustration.
As for Chamberlain and Gretzky, James’s (very) high income is the result of a mix of factors. An incredible innate talent that definitely has genetic roots, a hard-working mentality, and a sense of devotion,[8] and the fact of living in a society where people are willing to pay an insane amount of money to watch basketball games, even at the last row of the Madison Square Garden. Strictly speaking, James is responsible for none of those factors, including his mentality that he has never chosen to have. Now, the idea of morality is necessarily tied to the ascription of agency to oneself and others, as I have argued here. We have therefore strong and decisive public reasons to see others as agents. In the contrary case, there would just be no point in looking for public justification of principles. Without the ascription of agency, the very idea of public reason is self-defeating. As an agent, Lebron James must be viewed as partially responsible for the outcomes that affect him, and the resulting benefits and burdens he receives. Where responsibility stops and pure randomness comes is not well defined, but we know that such a frontier has to exist.
As I said above, randomness is not, in itself, sufficient to justify coercive interferences aiming at redistributing resources. At least, some persons may have reasons strong enough to reject such interferences, those making their public justification impossible. What has more chances to pass the requirement of public justification however is the old claim that money has diminishing marginal utility. The idea has been controversial but is still well alive, including in economics. It can be argued against based on the claim, for instance, that some persons have expensive tastes and that there is no reason that we make it impossible for those persons to satisfy them. But this is a weak argument. First, because we cannot appeal to the unfairness of randomness, that people have not chosen their expensive tastes is irrelevant here. Second, while the identification of expensive tastes obviously depends on a value judgment, this cannot be an objection. Public justification is concerned with the assessment of normative reasons, and normative reasons are obviously related to value judgments. Third, expensive tastes have an opportunity social cost, and this opportunity cost, like all opportunity costs, is relevant to assessing allocations of resources.
There is the obvious difficulty that we cannot expect everyone to agree on what expensive tastes are and how we should estimate the marginal utility of money. Fair enough, but this is again a weak argument. Consider the analogy with expertise and competence. There are controversies about what is required to be considered as an expert on some issue. People routinely disagree on who the experts are. But this disagreement is local, contingent, and most often settled by norms and formal rules. There is no problem with the general idea that there are experts and that we need them, and its implementation is relatively effective. I think the same applies to the marginal utility of money. There is no reasonable objection to the general idea of the diminishing marginal utility of money. People may disagree on what this implies in specific contexts and for specific issues of redistribution. But this disagreement can be settled by an impartial umpire taking the form of, for instance, democratic institutions. Different societies will not have the same level of tolerance for expensive tastes, but in general, all will consider that the redistribution from high incomes is publicly justified.
Interestingly, on this public reason account, James’s high income is not unjustified, contrary to what Gauthier was contending in Gretzky’s case. There is nothing about the market order that is unfair. Nonetheless, coercive interference to redistribute part of this income is publicly justified because there are public reasons to consider that this income is more than enough to satisfy all of James’s tastes and preferences that are not (too) expensive.
[1] David Gauthier, Morals by Agreement (Oxford University Press, 1986).
[2] David Gauthier, “Twenty-Five On,” Ethics 123, no. 4 (July 1, 2013): 601–24.
[3] See in particular Kenneth Binmore, Game Theory and the Social Contract: Playing Fair (MIT Press, 1994).
[4] Robert Nozick, Anarchy, State, and Utopia (Basic Books, 1974).
[5] This should not be conflated with the economist’s standard notion of maximization under constraint. Gauthier’s constrained maximization is indeed largely antithetical to the latter.
[6] As Gauthier recognized himself, in 2-player bargaining games, this corresponds to the so-called “Kalai-Smorodinsky” bargaining solution. One of the reasons why economists have generally dismissed Gauthier’s account is its incompatibility with the Nash bargaining solution.
[7] Hayek, Law Legislation and Liberty: The Mirage of Social Justice (Chicago: University of Chicago Press, 1977).
[8] Whether you like James or not, this must be recognized in light of his still impressive individual performances at almost 39. This has never been seen in the history of the game.
Very interesting discussion. For Hayek, I seem to remember that he said in LLL that the randomness of economic outcomes is in a way best kept unknown, so that people try hard to be successful. I could not find that back when looking at it. Did I dream it?
My response to Nozick from Crooked Timber https://crookedtimber.org/2016/10/16/locke-nozick-locke/
I’ll respond to Nozick’s Wilt Chamberlain example, which seems to have impressed quite a few readers. The basic point is that, if lots of people would be willing to pay a small amount to watch Wilt play, he could get very rich, while the watchers would be better off. At the core of this example is simply the observation that, given any initial allocation of property rights, there are potential gains from trade and that voluntary trade between two parties will make them both better off (though it may make others worse off, a point Nozick does not address). Furthermore, there is nothing inherent in this process that prevents some people from getting very rich.
But the choice of sport as an example is self-refuting. As the history of sport demonstrates, marginal differences in the structure of property rights could produce very different outcomes. If today’s intellectual property laws had been in place in 1891, James Naismith could have patented the idea and copyrighted the rules (the copyright would have expired in 2009, too late to help Chamberlain). If his invention were viewed as a work for hire, it might be the property of the YMCA. Or, if basketball was organized like college sports, Chamberlain wouldn’t be able to cash in at all, while his coaches could earn a fortune.
The actual structure of property rights under which Wilt Chamberlain acquired a net worth estimated at $10 million is one that includes an obligation to pay taxes at rates determined by legislation. Nozick gives no coherent reason why this particular feature of the property rights system is unjust, while the rest of the system is fine.