The legitimacy trap

In many ways, the public want more government: to shape a fairer, more dynamic economy and protect ordinary people from the consequences of a turbulent world. But they also see, all around them in their everyday lives, the evidence of a state that doesn’t work: roads full of unfixed potholes, GP appointments unavailable for weeks, and rivers full of sewage.

After over 15 years of stagnant productivity, weak business investment, and low GDP growth, the case for a more activist economic state has entered the mainstream. Yet many of those shaping this agenda are also deeply critical of the state as it actually operates. It appears bloated and unresponsive; staffed by civil servants who cannot deliver at the pace required. More diffusely, the public also feels that the British state is too expensive and wasteful – taxes are too high, civil servants too numerous – and yet still unable to meet everyday needs. 

Changing that will require a willingness to disrupt. But, in the search for disruption, too often policymakers latch onto simple solutions (such as arbitrary cuts), which replicate the mistaken thinking that brought us here. Progressives need to put the same work into rethinking how the state works as they have done to the economy. The state cannot act as an economic motor without public legitimacy; and the legitimacy of the state is in crisis. 

Addressing this crisis is critical to any serious progressive economic and social agenda. It requires that progressives build institutions that demonstrate the capacity of politics to both solve problems and bring people together around shared identities, with renewed agency. 

A legitimacy crisis 

The current crisis of the state is not simply a crisis of delivery, although services have deteriorated. It is not simply a crisis of trust, although trust has fallen. It is not simply a crisis of economic governance, although the economy has underperformed for 15 years. It is a legitimation crisis: a situation in which the structural failures of the state are systematically undermining the conditions under which democratic government can be renewed. 

Evidence of this crisis is visible in everyday life: in asylum seekers housed in hotels while local services collapse; children with special educational needs taxied to private equity-owned schools; and a growing number of university graduates burdened with debt for degrees whose labour market value is uncertain. These are not isolated policy failures. They are symptoms of a state whose multiple functions are becoming dysfunctional at the same time, in mutually reinforcing ways.

Understanding this crisis requires thinking clearly about the multiple overlapping roles of modern states. The state is a deliverer, which provides services and fulfils social commitments. But it is also a society builder, which provides a framework for holding together diverse populations through shared commitments, rules, norms, and a sense of common fate. And it is motor, which strategically shapes economic life, directs investment, and positions the nation in a changing world.

These roles can reinforce each other. Effective delivery builds trust; shared institutions sustain solidarity; and strategic action generates resources and security for the people. Britain today is experiencing the reverse: a doom loop in which failures in each domain compound the others, eroding the legitimacy of the democratic state as a whole.

As Sam Freedman argues, British state institutions are poorly matched to the needs and lived realities of their citizens: an overly centralised bureaucracy, operating at a distance from everyday experience, produces policies and institutions ill-suited to local conditions. Critical services are governed at arm’s length, through contracts with private bodies whose accountability to citizens is limited or absent. Professional judgement and agency, which might foster civic trust, become mired in rules and audit culture.

Local authorities sit at the sharpest end of this. English councils have faced cumulative real-terms cuts over the last decade, while their statutory obligations have often grown. A widening gap exists between the state’s obligations, and its capacity to fund and deliver them. The consequences are hollowed institutions, depleted professional workforces, and weakened infrastructure.

This yields three linked problems of legitimacy: people cannot see what the state does, delivery function becomes weaker and obscured, and delivery failures risk corroding trust further.

The crisis of accountability: delivery without agency

When people do not see the state solving problems, faith that politics can improve their lives fades. This erosion is experienced first at the level of everyday life: in those potholes, waiting lists, and polluted rivers. Small-scale local problems can accumulate into national discontent. 

The damage is compounded when there appear to be no actors with the authority or ability to fix them. Market-conforming, low-visibility interventions – from planning reform to “nudge” policies – have been both too modest for the scale of the problems, and they have failed to produce a clear demonstration of state effectiveness. Furthermore, serious high-profile failures – such as the infected blood scandal – do not seem to result in adequate learning on the part of system designers. Without visible improvements in public services, the state will struggle to rebuild trust; yet repeated failures to improve them have simultaneously eroded support for the very measures (most notably higher taxation) required to do so. 

The crisis of attribution: weakening rather than building solidarity

Failures of delivery create a second crisis, one that undermines rather than builds cohesion. This is best understood as a fiscal crisis that becomes experienced as a moral crisis: a diffuse sense that the state is not being honest about how it is allocating resources and who it is really serving. In pronounced cases, this generates a public mood of anger: ordinary people are being ripped off, resources are going to undeserving others, the whole system is a racket. 

When states appear impotent, service failures both demobilises and radicalises voters. In 2024, just over a tenth of voters in the British social attitudes survey “almost always” or “mostly” trust politicians to put the nation above the party (figure 1), a decline from even a decade ago. Research across Europe links austerity measures and service cuts to increased support for radical right parties.

Why do cuts push voters to the right rather than the left? Research on “zero-sum thinking” offers one answer. Where voters experience prolonged resource constraints alongside weak state performance, they are more likely to perceive the economic pie as fixed. Hence politics becomes a contest over exclusion – who should receive limited benefits and services – rather than over growth and investment. Populist parties exploit this logic systematically, framing social provision as something to be defended by restricting access. 

The result is a doom loop: faced with fiscal constraints, governments cut spending on public services. Underinvestment fuels dissatisfaction and populist backlash, which in turn undermines growth, political stability, and future investment - reinforcing populist pressures and limiting solidarity.

The crisis of governance: replicating rather than rebuilding

These symptoms are not accidental, nor, despite the harm done by 2010s austerity, primarily the result of bad decisions by individual politicians. They are the product of a specific theory of the state – applied systematically over four decades – that has produced structural consequences now proving very difficult to reverse. 

As the political economist Abby Innes has argued, neoliberal restructuring was not simply an economic programme but a model of governance: one that deliberately transferred operational knowledge and capacity from public institutions to private contractors. When services are outsourced, the knowledge of how to deliver them migrates too. The state retains a commissioning function but loses the capacity to evaluate what it is buying, to renegotiate when contracts fail, or to bring provision back in-house. Public agency is lost.

As delivery has faltered and trust has fallen, successive governments have fallen back on this approach to reform, yet again reorganising Whitehall or the NHS, with yet another efficiency agenda. These actions feed into the weakening trust and capacity that spurred them in the first place: alienating staff, confusing users, and sharpening lines of on-paper accountability without providing actors with the resources or responsibilities to solve problems themselves. 

Three false exits

It is precisely because the legitimacy trap is hard that shortcuts look appealing. There is no shortage of proposals. But many of these proposals advance singular solutions to the above problem, which cannot address the triple, interlinked crises. Without rethinking what makes public organisations both functional and legitimate, these solutions risk replicating the problems they wish to fix.

Streamline and automate. A first response sees technology as the solution. If the current state cannot process, decide, and deliver at the speed required, perhaps AI can handle benefit assessments, planning applications, visa processing, and clinical triage, yielding a leaner, faster state. Announcements of arbitrary cuts to civil service headcount accompany these promethean visions. 

There is a role for technology in public administration. Used well, it can give civil servants greater scope for creative problem-solving. But treating AI as a substitute for institutional capacity, rather than a complement to it, is at best risky.

Introducing AI systems at scale will likely create further dependence on (foreign) contractors, weakening the state’s domestic capacity. When things go wrong, this can be devastating. One example comes from the Netherlands, where an algorithmic system wrongly accused tens of thousands of families of fraud, with catastrophic consequences and no effective redress.

More fundamentally, the experience of being processed by a system that cannot explain its reasoning and cannot be argued with is not the experience of living in a society that takes you seriously. Legitimacy cannot be automated into existence.

Fire the consultants. The mirror-image of the above response is not to outsource to technology but “insource”, bringing capacity back in-house.

The UK government spends billions annually on management consultants and contractors, and this outsourcing has produced genuine inefficiencies.  But the problem is not simply that consultants are expensive. It is that decades of contracting out have left the state without the institutional knowledge, trained personnel, or organisational culture to do many things itself. Dismissing the consultants in a slash-and-burn approach without rebuilding what they replaced does not restore state capacity, it creates a vacuum.

More fundamentally, many of the activities which the state coordinates, like the care sector, benefit from rich interconnections with society and local businesses. A more capable state  needs to foster these interconnections, not look to direct ever more from an expanded Whitehall.

Spend more. Or spend less. The third and fourth responses are the most traditional. If the state has been starved, the answer is to feed it. If the state is bloated, drain it. Pulling these fiscal levers often goes hand-in-hand with centralisation, intended to eradicate ‘waste’ and guarantee outcomes – but frequently the result is further dysfunction. 

There is no question that many services and infrastructure projects have been starved of resources. Equally, there are untapped opportunities for public sector efficiencies. But viewing state legitimacy through an exclusively fiscal lens can obscure other ways in which the state matters to people in their everyday lives, and how it appears to them. Many of the problems afflicting the state-citizen relationship are about trust and agency. They will not be adequately understood if they are only ever viewed in terms of money and quantifiable outcomes.  

Addressing the triple crisis of legitimacy: an agenda

All these responses contain a grain of truth. The state does need to be faster, to rebuild in-house capability, and spend more in some areas and less in others. But each substitutes a component for the whole, and so sidesteps the question: what would it take for the state to act, visibly and effectively, in ways that citizens recognise and trust?

This requires serious thought across three areas. 

First, a theory of power — what can and should the state do in the economy and society? Today, progressives both want the state to do more and distrust it. Resolving that tension requires being clear-eyed about where state action is the right instrument and where it is not.

The case for an active state is strongest where power is concentrated and individuals cannot push back alone: confronting market concentration, preventing the spillover of economic power into political power, protecting workers from exploitation, pooling risks too large for households or firms to bear, and making the long-term investments that markets will not. This is the terrain on which the new economic thinking has rightly focused.

But a theory of power must also include recognition of what lies beyond the state. Much of what holds a society together happens in places the state cannot see and should not try to direct: in families and care relationships, in local businesses serving their communities, in the civic organisations that hold neighbourhoods together. Here the state's role is to convene, to fund, to protect the conditions under which others can act. 

Second, a practice of governance. Breaking the loop means demonstrating agency. Not performing activity – more announcements, more initiatives, more reviews – but showing that the state can identify a concrete problem that affects people's lives and actually solve it. Yet under conditions of genuine uncertainty (AI, climate, demographics), the old planning-and-delivery model cannot work: no central planner knows the answer.

Demonstrating that the state can improve lives involves visible agency: giving specific parts of the state, often at local level, the authority, resources, and licence to try things, iterate, and be judged on outcomes. Many have called for more “experimentalism” within the state, but how to produce it is not always clear. The British state already practices this in some areas: for instance, it is currently running experimental programs around AI use, and  local authorities have enacted highly effective local models of experimentation in service delivery.

Thinking more systematically about how to rebuild agency is likely to yield richer and more sustainable ways forward than focusing on headcounts. That means rethinking how we evaluate the public sector, and learn from local success and failure. The state needs some internal slack – even some “inefficiencies” – that allow civil servants to take risks. It also requires systems of evaluation that move beyond easily measurable singular metrics – ROI on physical infrastructure investment – to support a social and cultural infrastructure on the ground. 

Third, a politics of legitimacy. The 20th century state built its legitimacy partly through genuine organisational innovation: new welfare institutions, new forms of public provision, new partnerships between state and civil society. Faced with a radical Right attempting to colonise the language of social cohesion, progressives should not abandon the idea that policy can build a more cohesive and legitimate state through social and economic policy. But the approach and vision need renewing.

The state needs to demonstrate that it can improve lives, making the case to the public that collective institutions, rather than competitive ones, are the best way to do so. That requires moving beyond narrow delivery, to thinking about how public sector reform can embed democratic principles and create visible, shared experiences of the state. One way forward would be thinking about new social problems – regulating safety in AI, or addressing the housing shortage – not only in technocratic terms, but in symbolic ones, and in terms of enabling democratic action: a GB housing corporation, a democratic AI safety council.