Along with Meg, I'd like to add a response to Prof. Hashimori's assertion that
rational choice political scientists make more assumptions about aggregate
utility functions than economists do. This, in general, is not true. Aside from
IR theorists who assume the rational actor model at the nation-state level,
ratchers strive to provide microfoundational causal arguments, tracing the
causal path through decisions that individuals make. When we assume that
politicians want to get reelected or that MOF bureaucrats help protect the
public trough, we recognize that politicians and bureaucrats have many other
goals than these, but that the incentive structures they face channel their
behavior in these ways, as Bernie Silberman pointed out. [Note that we do NOT
have to assume that MOF bureaucrats are fiscal conservatives by selection,
self-selection, ministerial socialization or collective ideology or anything
else, but merely that their behavior is channelled in that way. Nor does it
preclude some of those things, but we want to be careful to avoid what Steve
Reed rightfully points out as the attribution fallacy--the assumption people are
behaving a certain way because they believe that is right, as opposed to acting
in response to incentives, learning, etc.]

To find out what those incentive structures are in the first place, we rely on
"soak and poke" like everyone else -- hard-nosed empirical work about career
paths, veto points and other decision making road blocks. (Perhaps this is the
"scientific punch" David Leheny wants to put into hermeneutical analysis.) When
the incentive structures are not binding, we may observe multiple electoral
strategies and bureaucratic slack or shirking, for example. These, too, are
objects of our study and are not assumed away.

Best,
Frances Rosenbluth