Só vim a entender um pouco de como Roemer pensa ao ler essa passagem da introdução:
I do not deal with the theory of exploitation because [...] after studying it for some time, I came to believe that it is not in itself a fundamental theory of (in)justice. I do not mean that workers are justly treated under capitalism, but rather that the view that what's unjust about their treatment is exploitation needs further articulation. For, according to Marx, or at least to my interpretation of his view, the exploitation of the worker is entailed by his receiving wage goods which embody less labor than the labor he expended for that pay. Now the unequal exchange of "embodied labor" in goods for "direct labor" in production is by no means obviously unjust. Indeed, if the capitalist is the rightful owner of the factory, then why cannot we view the "surplus labor"[...] as a rent the worker pays for access to that factory, for access, that is, to what he needs to render his labor fruitful? Thus the existence of surplus value, or unequal labor exchange in the above sense, is not sufficient to ground the claim that the worker is unjustly treated. I think that some egalitarian theory, of the Rawls-Sen-Dworkin-Arneson-Cohen variety, is needed to justify the Marxian accusation that workers are unjustly treated under capitalism. It may be unjust, for example, for any small group to own a factory, if, in a market economy, that makes the equalization of opportunities impossible. Or the method by which the capitalist came to acquire the factory may have been unjust. In either case, we need a deeper theory. (Roemer, 1996, p. 9).
Que coisa, não? Só falta o Marx pra esse marxista pelo jeito. Um marxista que gosta de teorias liberais de justiça.