Performance Management in Practice: A Comparative Study of Executive Agencies

This article reports a study of performance management practices in four functions across four European Union member states (Finland, the Netherlands, Sweden, and the United Kingdom). The focus is on how and to what extent performance indicators influenced the top management of the agencies concerned and the degree to which performance data were used by ministries as steering instruments. The research uses a historical institutionalist perspective combined with a model that identifies primary task characteristics as a source of significant variation. Thus the design explores both the influence of task characteristics (through contrasts among the four different functions) and embedded national system characteristics (through contrasts among the four countries). I show that both primary task characteristics and national system characteristics had some of the theoretically predicted effects on the management regimes. Equally, however, certain general tendencies embraced all countries and most functions. These include, first, the incremental growth of more sophisticated performance indicator systems and, second, the feebleness of ministries in developing performance-based strategic steering. Over the past two decades the introduction of performance management and the creation of executive agencies have been two of the most widespread international trends in public management (De Bruijn 2002; Holzer and Yang 2004; Ingraham, Joyce, and Donahue 2003; Joustie 2001; Mol and Kruijf 2004; Office of Public Services Reform 2002; Organization for Economic Cooperation and Development [OECD] 2002; Pollitt and Bouckaert 2004; Pollitt and Talbot 2004; Pollitt, Talbot, Caulfield, and Smullen 2004; Van Thiel 2004; Yesilkagit 2004). While these two (linked) types of development have been ‘‘translated,’’ in different ways in different countries and sectors, there can be little doubt of the pervasiveness of the reform rhetoric featuring one, or often both, of these ‘‘solutions.’’ As yet, however, we are rather short of comparative empirical research into the actual practice of ‘‘agencification’’ and its relationship to performance management. This article is a single-authored derivative of the EUROPAIR project, which was a team effort involving Janice Caulfield, Amanda Smullen, Colin Talbot, and myself. Adrian O’Reilly and Karen Bathgate also assisted in the early stages. EUROPAIR was supported by a grant from the U.K. Economic and Social Research Council and by the Centre for Public Management, Erasmus University Rotterdam. Address correspondence to the corresponding author at