Abstract

How effective are global regulations in preventing financial crimes such as money laundering, terrorism financing, and corruption? We conducted several large-scale “mystery shopping” experiments in which we posed as would-be criminals and approached tens of thousands of financial institutions worldwide. We embedded randomized experimental conditions into our approaches and examined the extent to which financial institutions were willing to offer companies and bank accounts that could enable financial crimes. Our research reveals some striking patterns. Descriptively, tax havens frequently demonstrate greater compliance with transparency standards than OECD countries, such as the United States. Experimentally, while certain high-risk profiles, such as those suggesting terrorism financing, modestly reduce access to financial services, most financial institutions remain largely insensitive to risks and rewards. These findings challenge dominant behavioralist models focused solely on individual rationality and point to the need for experimental social science to engage more deeply with institutionalist theory—especially where shared norms and embedded scripts shape organizational behavior. We use this empirical puzzle to advance and deepen the dialogue between behavioralist and institutionalist theories, a synthesis we term behavioral institutionalism, which I explore further in this talk.

Reference

Michael Findley, “Banking bad: Behavioral institutionalism in the global financial system”, IAST General Seminar, Toulouse: IAST, May 27, 2025, 11:30–12:30, room Auditorium 4 (First floor - TSE Building).