The primary argument in this book is that the member governments of the European Union delegate power and discretion to the European Commission and the Court of Justice in order to reduce the transactions costs of EU decision-making, and that they deliberately design and tailor a wide range of control mechanisms to limit agency discretion and maximize the benefits of delegation across issue areas and over time. Once created, however, supranational agents develop their own distinct preferences, generally for greater integration, and they pursue these preferences as 'engines of integration', albeit within the bounds of the discretion allocated to them in the original act of delegation.
In this chapter, I lay out in greater detail the theoretical argument of the book, as well as the research design and methods used in the subsequent, empirical chapters. The chapter is divided into five parts. In the first, I examine the decision by a group of principals to delegate authority to an agent, and I derive hypotheses about the functions likely to be delegated to agents and the conditions under which principals will allocate greater or lesser discretion to those agents. In the second part of the chapter, I turn to the supranational agents themselves, putting forward hypotheses about the preferences of supranational agents and the conditions under which they might be able to influence policy outcomes in ways that depart from the preferences of their principals. Third, I discuss both formal and informal agenda setting, and the very different conditions under which we would expect each type to occur. Fourth, I briefly summarize an alternative approach to the problems of delegation, agency, and agenda setting, drawn from sociological institutionalism, and derive some competing hypotheses from this approach. The fifth and final section discusses the research design, methods, and sources used in the book's substantive chapters.