Games Real Actors Play is a book about conceptual tools that social scientists could and should use in comparative case studies of complex policy interactions. Given that standard methods of theory testing will not work well for complex and potentially unique cases, the framework of "actor-centered institutionalism" shifts the emphasis from testing to the development and evaluation of theoretically disciplined and empirically grounded hypotheses. Since valid explanations must reconstruct the interactions among intendedly rational (individual, collective, and corporate) actors, Scharpf argues throughout that empirical researchers should, and could, rely on the logic of game-theoretic explanations. But to be empirically viable, such explanations must systematically exploit the information about actor orientations, capabilities, and constraints that is contained in the institutional settings within which policy interactions take place.
In the context of policy research, Scharpf argues that substantive policy problems need to be mapped onto the constellations of policy actors involved, and he shows how these constellations can be represented by relatively simple and transparent game-theoretic models. Next he argues that for any given actor constellation, policy outcomes are directly affected by institutional structures and the modes of interaction that these allow. Hence the explication of the policy implications of various modes of interaction -- such as noncooperative games, negotiations, majoritarian and hierarchical decisions, and the frequently encountered combinations of these -- constitutes the major part of the book. Throughout, the theoretical and methodological discussion is enriched by frequent references to, and some extensively presented, examples from comparative policy research in Europe and in the United States.
Beyond that, Games Real Actors Play addresses the need to integrate the positive and normative aspects of policy studies without violating the integrity of either empirical research or normative discourse. Scharpf shows how the criteria of "good" policy, which must inevitably be used in assessing the empirical effects of institutional arrangements, can be formulated in such a way that idiosyncratic value judgments can be ruled out within the context of interaction-oriented policy research.
Fritz W. Scharpf is codirector of the Max Planck Institute for the Study of Societies in Cologne, Germany, and a former director of the International Institute of Management and Administration, Wissenschaftszentrum, Berlin. He has taught at the Yale Law School, the University of Chicago Law School, and the University of Konstanz. He has published widely on constitutional law, democratic theory, policy formation and policy implementation, political economy, negotiation theory, and game theory.