Unlike procedural justice, distributive justice concerns itself with the distribution of goods within society, not the process of how those goods are distributed. Like procedural justice, distributive justice has right-wing and left-wing forms.
On the right, distributive justice primarily takes the form of desert theory: society ought to be constructed so as to deliver to each person what they deserve. What constitutes deservedness is rather slippery of course. But on the right, it tends to be defined in terms of marginal productivity: a person deserves to keep what they produce. In a world of independent yeoman farmers — the world Thomas Jefferson yearned for — desert based on marginal productivity is not too hard to figure out. You keep what you grow or raise. In a modern capitalist economy, the connection of work to productivity becomes more tenuous. Mitt Romney for instance “made” over $20 million in 2010 without doing any work at all: the money was just returns on investments.
When you really dig into the specifics of desert-focused distributive justice, it becomes really hard to imagine that the system the right-wing prefers — a fairly unregulated form of capitalism — actually tracks desert. But, according to the free market economists, even though individuals are certainly not always compensated equal to their marginal productivity (if they were, no profit would exist), in a very efficient economy, people are approximately compensated according to their marginal productivity. The right-wing desert theorists map onto this as their reason for supporting the system.
Distributive justice on the right-wing primarily consists of this sort of marginal productivity desert theory with some occasional utilitarianism thrown in for good measure. Distributive justice on the left-wing however encompasses an enormous number of philosophies and theorists. The most famous and most relevant set of those theories were written from the late 20th century to the present. John Rawls — the most famous of the bunch — argued that the main task for economic justice should be deciding upon the distribution of primary goods, i.e. goods that everyone wants no matter what else they want. These goods include things like rights, liberties, income, wealth, and social respect.
For Rawls, the best way to go at this problem is to imagine what kind of distributive principle someone would adopt behind a “veil of ignorance.” Under this hypothetical veil, a person will not be permitted to know who they will be in society, i.e. they will not get to know their gender, class, race, and so on. Behind such a veil, Rawls argues that people would adopt a principle that we should distribute all primary goods equally except where distributing them unequally increases the amount of primary goods everyone has access to, especially the worse off. If introducing some marginal amount of income inequality so increases the incentive to work that it improves even the lives of the worse off, then that should be permitted. In cases where inequality directly trades off, it is not permitted.
After Rawls, an array of distributive justice theorists followed. G.A. Cohen, Ronald Dworkin, and Amartya Sen are the most famous. These left distributive justice proponents try to answer two basic questions: 1) what is the proper unit of distributive justice, and 2) how should those units be distributed. For Rawls, the answer to (1) is primary goods. For Dworkin, it is resources, and for Amartya Sen it is capabilities. The answer to (2) for each of these thinkers is some form of modified egalitarianism. But the basic approach of distributive justice for these thinkers — and for the right-wing desert theorists as well — is to figure out what justice requires by determining who should get what and why.
I find myself way more moved by the distributive justice approach than the procedural justice approach for a variety of reasons. First, I am not convinced that any process actually meets — either in theory or practice — the standards of procedural justice. The ownership of any plot of land, for instance, requires some initial act of aggression because it can only come to be owned in the first place through violently and non-consensually extinguishing the ability of others to access the plot. The right-libertarian procedural requirements seem to actually make right-libertarianism itself impossible.
Similarly, the consensus requirements of left-libertarianism are internally incoherent. What happens if you cannot get full consensus one way or another on some decision that is necessarily binary in nature? The whole system falls apart. The system also fails to declare from whom consent must be given. Presumably not the entire world, but on what non-hierarchical and consensual basis could you figure out what community of people needs to consent to something for it to happen? I suspect it would be impossible.
In addition to the problems with procedural justice, I think distributive justice tracks better what people are truly interested in. I am doubtful that people are actually interested in the maintenance of abstract rules of process. Does the starving woman feel any better about her hunger when she knows that at least libertarian exchange processes are being upheld? Probably not. Ultimately, I suspect that people are actually interested in well-being, welfare, capabilities, and so on. Distributive justice frameworks center those things as the focal points of justice, which makes me more partial to that approach. That does not mean that process is irrelevant of course. It just means that processes should be primarily judged according to the distributive consequences they generate, not as just or unjust in themselves.