Agency Fever? Analysis of an International Policy Fashion

In the last 15 years, the governments of many OECD countries have transferred a wide range of functions to new, agency-type organizations. Allowing for the fact that, for comparative purposes, it is difficult precisely to define agencies, and further acknowledging that in many countries agencies are far from being new, it nevertheless remains the case that there seems to have been a strong fashion for this particular organizational solution.This article investigates the apparent international convergence towards agencification. It seeks to identify the reasons for, and depth of, the trend. It asks to what extent practice has followed rhetoric. The emerging picture is a complex one. On the one hand, there seems to be a widespread belief, derived from a variety of theoretical traditions, that agencification can unleash performance improvements. On the other hand, systematic evidence for some of the hypothetical benefits is very patchy. Furthermore, the diversity of actual practice in different countries has been so great that there must sometimes be considerable doubt as to whether the basic requirements for successful performance management are being met.     

Performance Regimes—The Institutional Context of Performance Policies

Performance measurement, targeting, reporting and managing of public services has spread across jurisdictions in recent years. The most usual stance adopted by governments in developing performance policies has been shaped by principal-agent theory and a hierarchy of principal-agent relationships from core executive to service delivery. Such notions have been challenged from several directions, both in theory and empirically. Writers on accountability and those analyzing the “audit explosion” and the growth of “regulation inside government” have pointed to the way in which multiple actors and accountabilities have grown. Drawing on these and other sources this article develops a “performance regimes” perspective that offers a heuristic analytical framework of the main groups of institutional actors who can (but do not always) attempt to shape or steer the performance of service delivery agencies. The aim is to create a framework that can be applied comparatively to study changes in total performance regimes over time and between jurisdictions and sectors.