public administration and development Public Admin. Dev. 22, 19–30 (2002) Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/pad.203 GOVERNMENT–NONPROFIT PARTNERSHIP: A DEFINING FRAMEWORK JENNIFER M. BRINKERHOFF* George Washington University, Washington, DC, USA SUMMARY Partnership has emerged as an increasingly popular approach to privatization and government–nonprofit relations generally. While in principle it offers many advantages, there is no consensus on what it means and its practice varies. Following a review of partnership literature, the article refines the partnership concept, developing two definitional dimensions: mutuality and orga- nization identity. Based on these dimensions, partnership is defined on a relative scale and is distinguished from other relation- ship types: contracting, extension, and co-optation or gradual absorption. Examples of each are provided. The model enables actors to assess their relative tolerance for partnership approaches, and provides a common language among potential partners. Linking its defining dimensions to partnership’s value-added assists partners to advocate for partnership approaches from an instrumental as well as normative perspective. The model and inter-organizational relationship matrix can inform continuing theory building and practical experimentation both to refine defining dimensions and indicators of partnership practice, and to enhance responsiveness to partners’ expectations of partnership. Copyright # 2002 John Wiley & Sons, Ltd. INTRODUCTION The last few decades mark a notable trend towards the privatization of public service delivery. This began as a response to what some perceived as the excessive growth of the welfare state, necessitating a third-party govern- ment solution (see Salamon, 1989). It was then extended to a recognized need to increase responsiveness and results orientation alongside efficiency, the purported benefits of the New Public Management (NPM) (see Ferlie et al., 1996). Regardless of the precise rationale, there is a growing popular consensus that the private sector— commercial and nonprofit—is generally more efficient and effective than government, and that government should steer, or at least facilitate, and the private sector should row (see Osborne and Gaebler, 1992). The trend is now global, as evidenced by: re-engineering in the United States (Gore, 1993); the concepts of pluralism and partnership in the United Kingdom (Young, 2000); the prominence of the New Zealand model (Boston et al., 1996); the importation of similar models into donor policy and programmes for developing coun- tries (see, for example, Wallis and Dollery, 2001); and the international donor community’s ‘New Policy Agenda’, where donors look to non-governmental organizations (NGOs) as implementers of donor-driven development pol- icy, bypassing government and co-opting NGOs (Hulme and Edwards, 1997). 1 In addition to their emphasis on efficiency and results, these models are also consistent with the principle of subsidiarity. That is, responsibility for meeting individuals’ needs should always be vested in the ‘units of social life’ closest to the individual (e.g., the family, parish, community and nonprofit organizations), and larger/higher units should only intervene when meeting the need exceeds the capacity of those lower levels (Salamon and Anheier, 1998). Copyright # 2002 John Wiley & Sons, Ltd. *Correspondence to: Jennifer M. Brinkerhoff, Department of Public Administration, School of Business and Public Management, George Washington University, 805 21st Street NW, Suite 601, Washington DC 20052, USA. 1 Traditionally, ‘NGO’ is used to connote nonprofit organizations in the international development arena, while ‘nonprofit’ typically refers to not-for-profit organizations and/or associations in industrialized countries. In this article, the terms nonprofit and NGO are used interchangeably.