Performance measurement is seen as a main pillar of public management reforms. However, prior research did not mitigate the level of ambiguity on PMS development and use and many questions left unanswered. Our findings provide knowledge on the design and use of PMS in the public sector answering to calls for a complementary approach among institutional and contingency factors. The authors find that the design of PMS and the use of performance information for external purposes are increased to legitimate the organizational' work and to obtain external support. Moreover, structural variables such as type of agency and management autonomy are also determinants on the organizational responses. Finally, the authors find some coupling between the design of PMS and outcomes. So, both contingent and institutional approaches play an important role on the empirical model.
TopIntroduction
Every effective organization measures and manages their performance to know how well they are performing and to identify opportunities for improvement. In this way, performance measurement (PM) arises as a way to ‘reinvent the government (Osborne & Gaebler, 1993) and improve efficiency, effectiveness and quality in public sector organizations (Hood, 1995). As Johnsen (2005) said, ‘by performance measurement in the public sector, the authors refer primarily to those performance indicators (PIs) of efficiency, effectiveness and equity that are intended to be used to improve rational decision-making in administrative and political processes’ (p.9). So, the main goal of any performance measurement system (PMS) is the improvement of decision-making (both at an external and internal level) and the promotion of accountability (Bovens, 2005; Cunningham & Harris, 2005; Johnsen, 2005; Modell, 2001; Walker, Damanpour & Devece, 2011). However, some unanswered questions are whether PMS are applicable in the public sector and performance information is effectively used to improve public sector performance.
The adoption of PMS has been studied as a response to the specific organizational environment (Cavaluzzo & Itther, 2004; Dooren, 2005) as well as from a neo-institutional approach (Laegreid, Roness & Rubecksen, 2007). As noted by Carter (1991), ‘the problems of performance assessment and the methods adopted for the design of performance indicator systems … are related to the specific features of organizations and their environment’ (p.85). Many scholars find that the complexity of the task (and the knowledge of the production process) and the measurability of outputs are considered important contingency factors (Oliver, 1991) with great influence on the effectiveness of PMS implemented in the public sector (see, for example, Bogt, 2004; Dooren, 2005 ; Laegreid, Roness & Rubecksen, 2006; Verbeeten, 2008).
Moreover, the implementation of sophisticated accounting and other management tools does not necessarily ensure their competent managerial use (Anderson & Young, 1999; Carruthers, 1995; Cavaluzzo & Ittner, 2004; Geiger & Ittner, 1996). This is the main issue that distinguishes institutionalists from contingency theorists. Many scholars find that if the implementation of new managerial practices is intended to satisfy government mandates, they will be used to obtain legitimacy and external support and not for control and decision-making purposes (Brignall & Modell, 2000; Carruthers, 1995; Geiger & Ittner, 1996; Johnsen, 2005; Modell, 2001, 2009; Pollitt, 2001, 2006). When the influence of isomorphism (coercive, mimetic and normative) is significant, a great deal of decoupling is expected (Modell, 2009). The spread between formal organizational structures and actual organizational practices will drive PMS to have little or no effect on internal corporate operations (Meyer & Rowan, 1991; Scapens, 1994; Carruthers, 1995; Geiger & Ittner, 1996; Burns & Scapens, 2000).