Despite its popularity as a conceptual framework, institutionalism is characterized by a deep methodological uncertainty. Doctoral students struggle to pin down the actual institutions that they wish to study, and, in the work of many established scholars, the operationalization of institutionalist concepts is frustratingly vague or surprisingly flexible. It has been 30 years since “new institutionalists” March and Olsen (1984) argued that “the organisation of political life makes a difference.” They were reacting to “undersocialised” perspectives like behavioralism and early rational choice theory and asserting that informal conventions were as significant as formal structures and procedures. This broad conception of institutions has been both the strength and the weakness of the wave of research that followed. It has enabled new institutionalists to build a more fine-grained and realistic picture of what really constrains political behavior and decision making. But an expanded definition of “institution” runs the risk of “conceptual stretching” (Peters 1999, 216). March and Olsen (1989, 17) themselves make no clear distinction between institutions and social norms in general, while Douglas North (1990, 83) goes as far as to include tradition, custom, culture, and habit in his definition. Researchers also operate at radically different temporal and spatial scales—from microlevel studies of decision making to analyses of whole government systems (like legislatures, for instance) or historical accounts of policy change. Bo Rothstein (1996, 145) cautions that if the concept of institution “means everything, then it means nothing.” Guy Peters (1999, 145) argues that institutionalists need “more rigour in conceptualisation and then measurement of the phenomena that are assumed to make up institutions.”