ABSTRACT
Over the last generation there has been a shift in theory and practice towards recognizing that collaboration across boundaries is important for tackling complex problems of government policy making and service delivery (Mandell 2001; Edwards 2001; Goldsmith and Eggers 2004). The traditional top-down models of public administration have been modified by more flexible models of partnering, contracting and incentives, often with a greater focus on ‘horizontal’ models of decentralized coordination. Much of the literature on collaboration of the 1980s and 1990s was concerned with marshalling arguments stating why collaborative approaches could be desirable and useful for reducing tensions and solving problems (Gray 1985, 1989; Gray and Wood 1991; Kanter 1994). The assumption that collaborative methods could resolve a wide range of previously intractable problems led to a normative bias in favour of collaboration, without the support of an empirical scholarship examining the costs and benefits under a wide range of specific conditions. However, recent research has made a more careful examination of the balance of costs and opportunities arising from working across boundaries (Gray et al. 2003; Goldsmith and Eggers 2004; Kamensky and Burlin 2004; O'Flynn and Wanna 2008).
