Predominant models of financial regulation based on representative agents—in both the public interest and public choice traditions—assume that competitive pressures in financial markets undermine prudential behavior by firms in the absence of regulation. One empirical expectation of these models is behavioral: firms should adjust their risk-taking behaviors in response to the regulatory environment they face but should not over-comply with regulations. That is, the central tendency of bank behaviors should hew closely to regulatory minima and the variance should be small. I first demonstrate that this expectation is not borne out by the empirical record and then advance a theoretical argument that does not rely on a representative agent model. I argue that firms face a range of incentives from markets and governments that condition their risk-taking behaviors, and firms choose a “preferred habitat” within a market structure. Some of these incentives are towards greater risk-taking, while others are in the direction of greater prudence. This framework provides opportunities for examining financial market actors in a realistic context, and offers ways to unify micro-level and structural analyses of the political economy of global finance.