July, 23 2019   |   Clare FitzGerald & Franziska Rosenbach

Findings from Are We Rallying Together?, a report from the University of Oxford


As concern over Brexit fomented in Whitehall throughout 2018/19, politicians seemed to focus on everything but the challenges local governments face. After a decade of austerity, local councils are scrabbling for money to help the most vulnerable, often turning to their communities for help. As a research team we wondered: is this a brave new era in public service delivery, characterized by collaboration and power-sharing within communities, or does it represent the end of universal, state-funded public services?


From autumn 2018 to early 2019, we—the Government Outcomes Lab (GO Lab) at the Blavatnik School of Government, University of Oxford—attempted to address this question. As part of our ongoing research into improving policies for complex social problems, we were increasingly hearing talk of the approach. Buzzwords like ‘collaboration’, ‘place-based working’ and ‘community participation’ were used to represent substantive changes in practice across the UK. We wanted to know more. We familiarized ourselves with 10 locally based collaborations in an effort to begin to describe the landscape of collaborative practice in the UK. Here is what we found.[1]


If privatization got us in here, collaboration will get us out


Collaboration has broad appeal as a solution to complex social problems. This is not a new approach: the UK public sector has a long history of cooperation with the voluntary and private sectors. Still, there are several trends that have shaped contemporary forms of collaboration: the persistence of ‘wicked problems’, the ‘move to privatize’, and the ‘move to partner’.[2]


‘Wicked problems’ are those social, public and economic problems for which there are no clear answers, definitions, or links between cause and effect.[3] Examples include homelessness, chronic unemployment or educational underachievement. Causes and solutions to ‘wicked problems’ are often intensely debated in Parliament (and around the dinner table), but progress has been slow. These kinds of problems are not likely to just go away. If anything, they are likely to become more complicated, perhaps because of the ‘move to privatize’. 


In addition to actual privatization, the ‘move to privatize’ relates to a set of public service reforms beginning in the 1980s that centered around applying private sector logics to government operations. A central belief was that market forces could increase efficiency and quality in the delivery of public services. There were many indirect policy tools used to embed these market forces, including selling state-owned assets and enterprises; contracting out public services to reduce state-monopoly delivery; and encouraging private sector delivery of government services.[4] Given these ideals, governments increasingly turned to third parties to provide core public goods and services. As the number of provider organizations grew, so too did fragmentation in local systems as well as competition between members of those systems.[5]


Perhaps as a response to such complexity, we have seen a ‘move to partner’, with growing recognition from “professionals, foundations, researchers, government agencies, and groups of organizations and volunteers” of a “clear need for greater communication, collaboration, and coordination of organizational efforts to achieve desired outcomes in local communities.”[6] As some issues cannot be addressed in isolation, partnering allows stakeholders to “jointly address seemingly borderless problems”[7] towards achieving outcomes for communities. The 10 projects we spoke to gave four broad justifications for collaborating:


  • To share financial and service delivery responsibility across organizations and sectors.
  • To give members of communities a more significant role in shaping and delivering services which affect them.
  • To deliver better overall impact and value.
  • To make the public sector a better place to work.


Collaboration can mean very different things


Collaboration happens when individuals in multiple organizations coordinate and share resources in order to support one or more policies.[8] There are fairly well-established models for groups to collaborate which are characterized by different levels of decision-making power and administrative responsibilities across member organizations: collaborations can self-organize, appoint one of their members to take charge of organizing things, or create a new body to do the organizing. We observed a variety of models in the places we went. We termed them as follows: 


  • Collaborative councils are broad programs of change where entire councils are attempting to change the way they work. Examples are Oldham, Wigan, and Wirral.
  • Collaborative markets aim to change relationships between local voluntary sector delivery organizations from competitive to collaborative. Examples are the Plymouth Alliance Contract, and Young People’s Foundations.
  • Agents of change sit outside the public sector, and seek to change the system from the outside. Examples are Ignite, and Golden Key.
  • System connectors aim to better integrate the public and voluntary sectors by leveraging existing resources, without fundamentally uprooting existing relationships and structures. Examples include West London Zone, and Doing the Right Thing.


Engagement not exploitation


We saw big changes in ways of working to enable collaboration within and among local communities. The style of leadership shifted from being about decision making exerted through traditional hierarchies to the facilitation of relationships. Organizational culture changed, giving frontline staff more discretion to work with service users. Each collaboration had infrastructure needs, a common theme among them being investing in data capture systems. This included integrating new methods, like ethnographic research tools, into ways of working. Co-location was sometimes identified as a facilitator for integrated teamwork, but was not always essential, nor enough on its own. Many sites used new types of meetings to improve communication and relationships between teams, and some provided access to shared IT systems.


Communities were always involved, though in varied ways. Sometimes community members were innovators, coming up with ideas that broke the bureaucratic mold of local government contracting. To implement these ideas, some councils set up community funds for grassroots projects. In an atmosphere of tighter local finances, many places tried to change the conversation with citizens from “what can we do for you?” to “what can we help you do for yourselves?” In some areas, ownership of assets like libraries and swimming pools was transferred to the community. While the impetus for these decisions seemed to be an effort to keep non-statutory services open, such a shift in responsibilities comes with risks in terms of ensuring service quality and maintaining equitable access. Was this about empowering the disempowered? Or was it just pushing critical public responsibilities onto organizations and people who may not be equipped to deal with them nor compensated for their efforts?  

If everyone is responsible, who is accountable?


Collaboration rests on mutual trust between partners, underpinned by an empowered and enfranchised workforce. Who is in charge? Who measures success, sets targets, and is accountable if things go wrong? Unlike ‘targets and terror’ methods of performance measurement, we found data capture in these collaborations was bottom-up, place-based and included qualitative insights. Measures were used to learn, not punish.  


Whilst it is challenging to define and measure success across many stakeholders, there are also great challenges around accountability within collaborations. The jury is out on whether collaborative approaches enhance or diminish democratic accountability. Some scholars argue that collaborative arrangements are “less democratic” by traditional measures[9] because government cedes control of services. There is no longer a straightforward mechanism by which policymakers are held to account by the electorate. Yet accountability rarely functions so simply, even in a centralized system with well-defined contractual relationships. Other scholars argue that multi-centered governance has equal or greater legitimacy.[10] Such dispersed governance provides more checks and balances than centralized systems. It can offer more opportunities for citizens’ voices to be heard, and for local or innovative solutions to be developed. However, this type of governance presents accountability challenges. Elected politicians and the public may prefer simple lines of accountability, even if these are sometimes illusory. Thus, collaborations require a compelling and accurate narrative to explain their legitimacy to others. They need to ensure that inequity is not exacerbated through ‘capture’ by particular organizations or vocal community groups, or neglect of groups that lack resources to participate.


We considered how sites demonstrate ‘external’ accountability, such as to commissioners or funders, elected councilors in local authorities, regulatory bodies, service users and citizens. Some collaborations are required to report to external funders and commissioners. This was sometimes an onerous duty when multiple funders were involved, with several reporting lines with different formats and timescales. Collaborations featuring market- and outcomes-based financing usually had more defined reporting requirements than those with grant funding. Some collaborations are responsible to their local authority—specifically the elected council—for fulfilling statutory duties and the council’s public ‘best value duty’. This type of scrutiny often overlaps with internal accountability mechanisms, particularly for council-wide collaborations. Some collaborations are also subject to inspection by regulatory bodies—which can prove complex, as traditional inspection processes are more suited to dealing with single organizations than with partnerships.


Still, most collaborations we spoke to are ultimately accountable for, and thus base their legitimacy on, improving outcomes for service users. This underpins the importance of having a deep understanding of both client need and the factors that generate better outcomes, so that delivery systems can be based on this understanding. In turn, this is dependent on a high level of good quality information and feedback, and willingness for partners to work in the interests of service users even if that does not necessarily align with their organizational interests. 

Should we all start rallying together?


As with any research-inspired exercise, we found ourselves ending this phase of investigation with more questions than answers. In particular, we think the following are worth consideration. 


  • What regulatory and statutory constraints prevent collaborative approaches?
  • Does collaboration deliver better value and impact than the status quo? Do we need to show that it does before politicians and policy-makers will be convinced?
  • Is collaborative practice resilient in the face of setbacks? Can it withstand a scandal?
  • How can you maintain quality and access to services when decisions are made by frontline workers and community organizations? What role does middle management play in such a shift?
  • Is there a system of measurement that can deal with the complexity of the issues, allow learning, and deliver accountability?
  • What is the ‘right way’ to involve citizens in the delivery of their own public services?


We would welcome more conversation, collaboration, discussion and debate.


Clare FitzGerald is Research Fellow at The Government Outcomes Lab, Blavatnik School of Government, University of Oxford. Franziska Rosenbach is Research Assistant at The Government Outcomes Lab, Blavatnik School of Government, University of Oxford.



[1] The report on which this article is based, Are we Rallying Together?, can be found at: https://golab.bsg.ox.ac.uk/our-projects/about-future-state/

[2] Koliba, C., Meek, J., Zia, A., Mills, R. 2019. Governance Networks in Public Administration and Public Policy. Routledge, New York.

[3] Koliba, C., Meek, J., Zia, A., Mills, R. 2019, p. 22.

[4] Wise, C. 1994. The public service configuration problem: Designing public organisations in a pluralistic public service. In A. Farazmand (ed.), Modern Organizations. Praeger, Westport, CT., p. 84.

[5] Christens, B., Inzeo, P. 2015. Widening the view: situating collective impact among frameworks for community-led change. Community Development 4(6), 420–435.

[6] Christens, B., Inzeo, P. 2015, p. 423.

[7] Koliba, C., Meek, J., Zia, A., Mills, R. 2019, p. 29.

[8] In this report, the term ‘collaboration’ is used as a catch-all for the various names and forms of inter-organisational working including but not limited to: partnerships, collectives, collaboratives, task forces, collaborations, and collective impact initiatives.

[9] Cairney, P., Heikkila, T., Wood, M. 2019. Making Policy in a Complex World. Cambridge University Press, p. 44.

[10] Ostrom, E. 2010. Polycentric systems for coping with collective action and global environmental change. Global Environmental Change 20(4) 550–557; Hooge, L., Marks, G. 2003. Unravelling the central state, but how? Types of multi-level governance. American Political Science Review 97(2) 233-243; Cairney, P., Heikkila, T., Wood, M. 2019, p. 45.

Share this news

Comments (0)

It is mandatory to be registered to comment

Click here to access.

Click here to register and receive our newsletter.

Partners Program

Executive Master (EMPA)


Public 50