Introduction

A growing number of studies characterise multi-level public–public and public–private inter-organisational partnerships as ‘governance networks’ (Klijn,1996; Jones et al, 1997; Kickert et al, 1997; Lowndes and Skelcher, 1998; Torfing, 2005; Klijn and Skelcher, 2007; Provan and Kenis, 2007; Koliba et al, 2010). While multi-disciplinary enthusiasm about the characterisation and analysis of governance networks has grown considerably in recent literature, much more theoretical and empirical work remains to be done to understand how performance management systems are institutionalised in multi-scale and polycentric governance networks (Kettl, 1996; Bardach and Lesser, 1996; Milward, 1996; Frederickson, 1997; Milward and Provan, 1998; Agranoff and McGuire, 2001; Papadopoulos, 2003; Benner et al, 2004; Scholte, 2004; Papadopoulos, 2007; Harlow and Rawlings, 2007; Koliba et al, 2011a). Page (2004) notes that while accountability and performance management are closely linked, the latter focuses on narrower, more measureable targets that are linked to specific goals. Performance management systems are often part of larger accountability regimes; both face challenges in the context of collaborative governance networks (Page, 2004; Ansell and Gash, 2008). Papadopoulos (2007, 2010) argues that multi-level governance may improve accountability to stakeholders while simultaneously reducing accountability to the general citizenry. The issues of scale, both the jurisdictional scale of inter-organisational network as well as the scale of vertical and horizontal collaboration in which a given inter-organisational network engages, are critical for assessing performance of polycentric governance networks.

The application of fair and effective performance measurement in ‘traditional’ vertical public administration and policy studies is no easy task. Beryl Radin, for instance, warns that, ‘despite the attractive quality of the rhetoric of the performance movement, one should not be surprised that its clarity and siren call mask a much more complex reality’ (2006, 235). Performance management is a complicated matter within individual organisations, let alone inter-organisational networks. Herbert Simon (1957) and Charles Lindblom (1959) were some of the first to discuss the limits of rationality within social organisations. We argue that the same factors that lead to ‘bounded rationality’ and incrementalism in the course of day-to-day management and policy making cloud performance measurement in polycentric governance networks.

Theodore Poister (2003) describes performance measurement as a continuous cycle of inquiry that encompasses the collection and processing of data, the analysis of this data, and the utilisation of this analysis to adjust actions and behaviours.