In The Construction of Social Reality (1995), John Searle develops a theory of institutional facts and objects, of which money, borders and property are presented as prime examples. These objects are the result of us collectively intending certain natural objects to have a certain status, i.e. to ‘count as’ being certain social objects. This view renders such objects irreducible to natural objects. In this paper we propose a radically different approach that is more compatible with standard economic theory. We claim that such institutional objects can be fully understood in terms of actions and incentives, and hence the Searlean apparatus solves a non-existent problem.

This paper discusses a common criticism of economic models that depart from the standard rational-choice paradigm - namely, that the phenomena addressed by such models can be ‘rationalized’ by some standard model. I criticize this criterion for evaluating bounded-rationality models. Using a market model with boundedly rational consumers due to Spiegler (2006a) as a test case, I show that even when it initially appears that a bounded-rationality model can be rationalized by a standard model, rationalizing models tend to come with unwarranted ‘extra baggage’. I conclude that we should impose a greater burden of proof on rationalizations that are offered in refutation of bounded-rationality models.

In Better Never to Have Been, David Benatar argues that existence is always a harm (Benatar 2006: 18–59). His argument, in brief, is that this follows from a theory of personal good which we ought to accept because it best explains several ‘asymmetries’. I shall argue here (a) that Benatar's theory suffers from a defect which was already widely known to afflict similar theories, and (b) that the main asymmetry he discusses is better explained in a way which allows that existence is often not a harm.