Roundtable Write-up: Institutions

Setting and purpose

On 25th March 2026, a roundtable was held as part of the Centre for the Edge initiative, co-hosted by Kinship Works and Joseph Rowntree Foundation. The roundtable asked: is the public sector stuck in an institutional monoculture, dominated by institutions that look similar, and behave in similar ways? If so, what is locking us into this monoculture? And what could be done to induce more institutional diversity?

The roundtable brought together practitioners, policymakers, senior leaders and innovators from across the public sector and civil society to explore these issues. Two provocations were shared in advance:

• Escaping the institutional monoculture, by James Plunkett, Kinship Works

• Hidden wiring — and why we need an army of boring innovators, by Sophia Parker, JRF

The discussion was held under Chatham House rules; in the write-up below, comments are unattributed.

Summary of discussion

The conversation opened with two participants sharing their experiences. One has built new institutions to support and drive community-led work at the edges of our struggling old systems. The other has worked in various centres, in local and national government, supporting innovation. There was then a wide-ranging discussion in which nine broad themes emerged.

1. The confidence gap: why the centre feels stuck

Several participants felt that the lack of confidence in central government was a first-order problem — anxiety makes it harder to escape from unsuitable or outmoded institutions. Despite holding considerable formal power, it is striking how senior figures at the centre feel a profound lack of agency. In the words of one participant, if you talk to Permanent Secretaries about their level of agency, “it’s like talking to someone entirely removed from the centres of power.”

Why has the centre lost confidence? Participants suggested the reason was partly historical, partly psychological. The previous dominant theory of state capacity — broadly, New Public Management — was limited, but at least it gave leaders a coherent framework for action. This worked, for a while, to some extent. Now, however, NPM is widely seen to be inadequate, but senior leaders don’t yet feel that anything has replaced it. There is a feeling that the old textbook has been lost, but a new one has not yet been found. This feeds anxiety, which makes people want to cling on tightly to old methods as a form of control. There is no malice to this — it is a reflexive response, like the tightening of muscles — but it squeezes out diversity of thought and hinders institutional experimentation.

What is more striking is that local parts of the system often seem more confident than the centre, although not always. Strategic Authorities are one example — at their best, they are confidently making space for civil society to step into, allowing innovation to thrive. This kind of self-assurance is often lacking in the centre. One participant described attending a gathering of local and civic leaders and finding “a room of a hundred confident people with the bit between their teeth”. Another spoke of community-led work in Grimsby that feels “joyful”. Work at the edges is not easy, but it often feels full of possibility.

This prompted a paradoxical thought: if the centre was more confident and self-assured, would it be less overbearing? If so, how do we help senior leaders in the centre to re-find their confidence?

2. The need for new, more adaptive institutions

It was suggested that bold work at the edges often requires new institutions — and those institutions tend to be organic in their form and adaptive in their development. This was illustrated by a pioneering example of community-led development. Starting with £80,000 against a surveyor’s estimate of £1.9 million, the project discarded the professional survey and developed a derelict nine-storey office block, working floor by floor as money became available. What began as financial necessity was later recognised as a replicable methodology: phased organic development. The community has been involved from day one and the form of the new space has emerged in response to need.

New institutions were developed to support this work, and they drew on the logic of the commons, with a three-part setup: a common resource, in this case buildings and land; commoners with rights and responsibilities; and “commoning” as an active, ongoing practice — the thing they do together. This is not one organisation but an ecosystem of separate legal entities, each created as a tool to serve the vision, with a shared brand, board and team but distinct legal forms.

At a larger scale, a national network of community asset developers is now building mutual trust and solidarity, aiming to aggregate members’ assets into a mutual that can pool risk and attract investment. The goal is to build the capabilities of a mycelium: being able to sense the whole, and adapt to changes or threats in the environment, without having a centralised brain or organ of control.

These initiatives were described as prefigurative: ways of working today that enact a better future, before it has arrived. This work is showing, with living examples, that alternatives are possible, and exposing current norms as inadequate by contrast.

3. Structural drivers: funding, accountability, compliance

Why don’t institutional innovations spread? Participants pointed to a number of structural mechanisms that keep current institutions locked in place. One example is the shift, over the 2010s, from grant funding to contracts in civil society. This tightened the hold of commissioning, leaving less space for experimentation. Charities became delivery arms of government, and began to look more like their government funders.

Accountability also functions in a way that inhibits experimentation. Organisations seeking public funding are asked for particular types of evidence, and have to adopt certain forms of governance. Together, this means conforming ever more closely to the bureaucratic norm. This leaves innovators facing a constant choice — do they adapt to these requirements, and become ever more like mainstream institutions, or do they stay true to their innovative forms and forgo access to funding? A common story is a gradual drift — as innovative organisations get bigger, they begin to resemble the old institutions they had hoped to replace.

This dynamic often holds even when the centre tries to devolve budgets or decisions. In these cases, it often feels like central government is searching for whatever part of the local system looks the most like them. The hope is that they can then hand over budgets, or power, and be confident that a local body will make the same decisions they would have made at the centre. The result, as one person put it, is “the same outcomes, just organised differently”.

4. Crisis and the allure of smooth transitions

What role does crisis play in breaking a monoculture? One participant challenged the assumption that the transition could be smooth or carefully managed. Perhaps the whole idea of incrementalism — gradually building our confidence; slowly cultivating a better relationship between the centre and the edges — doesn’t fit the moment we’re in. It seems at odds with the scale of the climate crisis, for example, or the crisis of biodiversity, or the rise of populism.

There was a debate about the upsides and downsides of crisis-driven innovation.

On the one hand, moments of genuine institutional innovation have often come from crises, and not from carefully managed processes. The UK vaccines taskforce was cited as an example: it was an extraordinary feat of coordination, pace and institutional creativity, operating outside the normal logic of government procurement and delivery. It was made possible by a combination of exceptional leadership, political mandate and the forcing function of an emergency. It is hard to think of equivalents during normal peacetime operations. Perhaps genuine institutional innovation needs a crisis. If so, maybe the key is to be ready, when the moment comes.

On the other hand, one participant asked how sustainable it is to birth innovation from crisis. Many Covid-era innovations collapsed after the pandemic ended, in part from exhaustion — much as the human body collapses after a period of intense exertion. We cannot stay in crisis mode indefinitely. And we should also be mindful that it is often the victims of a crisis who take on the brunt of the response.

There was also a suggestion that we should look at this another way: maybe what we need are new institutions that can cope with a permanent state of crisis. One participant suggested that we are like a boat that has passed from a smooth river to rapids: the vessel and the navigation skills that worked in calm water are not the ones we need now, in more turbulent times. The military was referenced as a source of learning: they have sophisticated procedures for entering and exiting high-intensity activity, for rotating leadership through periods of sustained pressure, and for managing the psychological strain. There might be more to learn from these methods than from the smoother and more managed methods that tend to dominate debates about public service reform.

5. The big prize: changing the middle ground

Should we be more ambitious? Much of the conversation about institutional diversity, it was suggested, operates on an implicit assumption: that the task is just to pull the centre back a little, and create some space for innovation at the margins. This is important. But is it enough? Should we focus instead on transforming the middle of the delivery landscape? If our goal is to break the institutional monoculture, then surely the real work is to open up the middle, not the edges. We need to change the forms through which public services are delivered at scale — changing the large-scale institutions that shape our daily lives.

The challenge of changing the middle is more demanding. We need to get beyond tolerating diversity at the periphery to actively transform mainstream provision. Participants were honest about how difficult this will be. There has been a long history of efforts to introduce provider diversity, with some incremental gains, but few examples of genuine transformation.

6. Visibility and attunement

How well do the centre and the edges see each other? One participant suggested that the centre and the edges of the public sector often seem mutually invisible. So perhaps this is less an issue of hostility and more an issue of perception. The people doing the most generative work at the edges are often not well-understood by people in the centre. Meanwhile, leaders in the centre — their anxieties, and the constraints they have to navigate every day — are not truly seen by people at the edges.

The word “attunement” was offered as a way to understand this. Maybe what we need at the centre is less the capacity to drive change, but rather a fine-grained sensitivity to what is happening at the edges, and what is needed. The example of Buurtzorg was given, a care network in the Netherlands. It grew from a single small team to 15,000 people, working in around 1,000 local teams. This growth was not driven from the centre. Instead, it was enabled by a centre that was highly attuned to what the model needed, and that was self-confident enough to provide these enabling conditions.

The suggestion was that we need more space for the centre and the edges to talk to one another. One local project, for example, brought together officials from a local authority with a group of local social innovators — it was the first encounter between the two groups, and it seemed to unlock something. Could we create more of these encounters, at scale? How can we create more opportunities for the centre to become better attuned to what is needed at the edges?

On a similar theme, one participant offered the concept of “interfaces”. Just as different software systems can exchange information via the interface of APIs, while remaining very different, maybe public institutions need better interfaces between each other — ways of funding, evaluating, commissioning and holding to account that can bridge different practices and mentalities, without requiring all institutions to become more like each other.

7. Trust — in both directions

When we worry about trust in public life, we tend to mean that citizens do not trust institutions. But trust runs both ways, and it’s also clear that public institutions often don’t trust citizens.

This lack of trust from state to citizen is part of what keeps us locked in today’s top-down, control-oriented institutions. We see this in the design of top-down services that centre professionals, and assume users need to be managed. We see it in healthcare services that assume patients have little to offer when it comes to managing a chronic health condition. We see it in intensive monitoring of public servants, and in compliance-oriented accountability systems. And we see it in the reluctance of central government to devolve genuine decision-making to local communities.

One participant offered a story from Bologna as an illustration: a local man painted a dilapidated public bench and was given a fixed penalty notice for criminal damage. Luckily, the Mayor stepped in — they rescinded the notice and set up a “ministry of civic imagination”, telling public officials to trust the public and see their role as helping citizen innovators. Perhaps we need to emulate this shift across the UK public sector.

8. Non-anxious leadership

The theme of leadership came up repeatedly. There was a consensus that we need something very different to the mode of public sector leadership we currently develop and reward.

The capabilities mentioned most often were not the traditional capabilities of public sector leaders: strategic planning, performance management, communication, and so on. What participants returned to were capabilities related to holding complexity, rather than resolving it — the capacity to tolerate ambiguity, to work through conflict, to maintain direction without imposing solutions, to know when to step in and when to get out of the way. One participant described this as “non-anxious leadership” and noted that this was in short supply in a public sector that often seems gripped by anxiety.

How do we change leadership culture? Partly we need to change incentives. Pay and promotion are tied to managing existing systems, not to imagining new ones. Leaders who invest time and energy elsewhere — supporting emergent alternatives, building relationships across boundaries, closing down old organisations — do so largely outside the reward system, often at significant personal cost.

The question of “leading for endings” was given particular attention. Ending an organisation is one of the most courageous and necessary acts of leadership in the current moment — recognising when a particular institutional form has come to the end of its useful life, even if the wider mission goes on. Leading an institution through an ending like this requires a multi-year perspective, significant personal and organisational confidence, and a willingness to voice the unvoiceable — that the current institution is unfixable. There should be more support for the work of closure and transition.

9. How granular do we need to be?

Is it useful to talk about institutional innovation in such general terms? One participant felt that the ecosystems people work in are so different — social care, vs education, vs scientific innovation — that there is limited value speaking about institutional innovation at a general level. In each of these sectors, there are hundreds of expert professionals who have spent their whole careers thinking about issues of diffusion, innovation and institutional dynamism. The risk when we speak in general terms is that we all nod along, but really all of the useful diagnoses and solutions sit within sectors.

Participants agreed that each ecosystem is different. But there was also value in exploring these issues across sectors. One participant spoke of the importance of field-building. By speaking across disciplines and ecosystems we can develop a field — a shared language about institutional innovation, shared frameworks, and shared stories about the predicament we’re in, and the work to be done. Second, we will spot surprising connections, or analogies across sectors: someone working in social care, or education, might be inspired, for example, by work being done in metascience to make the scientific discovery process more dynamic.

One suggestion — in service of field-building — was to go beyond conceptual and narrative work on the institutional monoculture, and undertake some more granular work: mapping, visualising and quantifying the institutional landscape to make it more visible and legible. The question of what exists, where, in what form, and with what track record, is still only partially answered. Without a clearer picture of our institutional landscape, it is difficult to make the case for deliberate intervention to cultivate institutional diversity.