The mechanics of devolving power



In December 2020 the BBC reported that a group of ‘mainly Labour’ politicians had written to the Prime Minister, stating that the future of the Union was uncertain and demanding the devolution of significant additional power to localities across the UK. The BBC reported that the signatories say they have “seen the huge potential of devolution first hand, but also its weaknesses… Especially in England, it is constrained by a lack of adequate powers and funding. Overall, we are still more centralised, and more regionally unequal, than any comparable country…. The current relationship between our central government and the constituent parts of the UK is incoherent, often unjust, and ultimately unsustainable”.

These assertions mirror claims made in a large number of recent think-tank reports,[1] which critique the UK’s system of centralised administration as inherently inefficient. They also reflect an academic consensus: that decentralisation of power benefits states economically, administratively and democratically, and that UK governments’ policy towards ‘devolution’ in England has not been sufficiently muscular to reap these benefits.[2]

Since 2014, English devolution policy has developed in fits and starts. Throughout it has been marked by limited local discretion and uncertainty about the long-term role and purpose of the new ‘metro-mayors’. The limited powers available to them was a feature of the public disagreements between the UK Government and some Northern mayors in late 2020 over Covid-19 financial support. That episode was significant as much for its optics, and its effects on public perceptions, as for its substance. Northern leaders were able to muster instinctive local sympathy and cross-party support for their position; but their impact on outcomes was muted.

That episode also highlighted a consistent feature of the UK government’s response to Covid-19 during 2020: an instinct to bypass local institutions in favour of ad hoc central structures. This has not been a success: substantial evidence has built up of a lack of competence in those systems. This has been accompanied by demonstrations of what Adam Lent has called ‘weknowbestism’; a disregard for local knowledge or locally-sourced statistics. Unflattering comparisons have been drawn with the decentralised, localised response to Covid-19 in other states. The Institute for Government went as far as to claim that the coronavirus pandemic had seen a “fundamental breakdown of the working relationship between central government in Westminster and local government across England”.

The benefits of devolution

Exhortations of this kind have had negligible impact on the politics of English devolution. Indeed, the frequency with which such arguments have been made in recent years appears to stand in inverse proportion to their influence within Whitehall.[3] This may change with the anticipated Devolution and Recovery white paper, expected in summer 2021, but experience to date points toward constraint on any further transfer of power to local areas. In the meantime, metro-mayors have emphasised their ‘convening role’.[4] This parallels the academic concept of ‘place-based leadership’, denoting a broad network of governance relationships between elected officials, public sector bodies, and local leaders within the private and voluntary sectors. Mayors use inspiration and vision, and small quantities of cash and capacity, to help these networks act as more than the sum of their parts.

Focusing on the quality of relationships, and personal vision, is a natural stance for a new and somewhat insecure group of institutions. However, the metro-mayors have also periodically sought additional formal powers. Beyond brief demands, there has been a dearth of analysis of how greater legal powers would be transferred to them. Intuitively, this may sound like a non-issue. Ordinarily, research and commentary treats decisions to ‘devolve power’ as one-dimensional and unproblematic. Powers exist and governments devolve them (or not). But in practice, there are many ways to implement devolution, demonstrated by the huge variations globally in central-local government relationships.

The remainder of this paper explores how such decisions, and their impact on outcomes, can be understood. The paper draws on the recent practice of English devolution, still a live policy area, but it could equally be applied to other polities.

Dimensions of autonomy

The OECD’s 2019 report Making Decentralisation Work presents a ‘Local Autonomy Index’, based on the work of Ladner and Keuffer. It comprises seven dimensions of local autonomy:

  1. Legal autonomy measures the extent to which the existence of municipalities is constitutionally guaranteed and whether or not municipalities can, for example, be amalgamated against their will.
  2. Policy scope describes the range of services for which municipalities are responsible.
  3. Political discretion additionally asks whether municipalities also have some decisional power while fulfilling these tasks.
  4. Financial autonomy means that municipalities have their own financial resources, can collect tax and decide on their base and their rate and are able to borrow money.
  5. Organisational autonomy describes the possibility to organise and staff their administration and to decide on features of their political system.
  6. Non-interference is related to the vertical relations with the higher levels of state and consists in the way supervision is organised and whether financial transfers are unconditionally granted.
  7. Access captures whether municipalities can influence higher-level decisions.

Any polity where power is devolved beneath the central government level must take a position on each of these seven dimensions of autonomy, whether that position has developed organically or has resulted from an explicit decision. It follows that any decision to transfer additional power to sub-national governments must also, by definition, comprise a series of decisions regarding each of these dimensions of autonomy. For instance, devolving power does not consist solely of transferring control over particular services (dimension 2). It also comprises decisions about the degree of discretion and financial autonomy available to the recipient authorities (dimensions 3 and 4). Those decisions may constitute explicit change to previous practices; a retention or reflection of previous practices; or they may be disregarded or not made explicit at all. Either way, they constitute a critical influence on the operation of any new institutions or newly-devolved powers.

The following discussion explores how decisions in each of these dimensions might influence the capacity, effectiveness and legitimacy of a devolved authority.

  1. Legal autonomy

The constitutional position of local authorities in the UK is not guaranteed, in accordance with the UK’s unwritten constitution and the principle of parliamentary sovereignty. This is largely uncontentious in the UK context. There have been calls in recent years for constitutional protection for local authorities, and for the UK to take symbolic actions such as signing the European Charter of Local Self-Government, but these have been peripheral to the wider debate on devolution of power. Metro-mayors and local authorities have not sought legal autonomy as a foundation stone of their demands for additional powers.

  1. Policy scope

The majority of commentaries on the English devolution process advocate increasing the functions of combined authority mayors, local authorities, or both. However, identifying functions to devolve to recipient authorities may be complex. For instance, the Greater London Authority’s 2019 Skills and Employment Call for Action ets out a clearly-reasoned proposal for a “devolved holistic skills and employment system” for London. But the proposed functions are located in a number of agencies and bodies, each with distinct funding streams; and potentially with distinct legal foundations. Granting a recipient authority legal powers to act in this area and devolving an existing function to them are two separate matters, and the former does not automatically imply the latter.

A second challenge is the existence of concurrent functions, where multiple tiers of government hold identical functions, with none having legal priority over the other. Concurrent functions are the norm in European systems of multi-level government (OECD 2019). Where governments hold equivalent functions, their practical influence in the policy areas concerned will turn on other dimensions of the autonomy index, such as access to funding, local capacity, and the absence of central interference. This point is critical to understand English metro-mayoralties, which lack the funding or capacity to exercise many of the powers legally devolved to them.

The upshot is likely to be that ‘competing’ public bodies – normally central agencies – continue to dominate policy outcomes in those areas. In turn, this threatens to undermine the capacity and legitimacy of devolved institutions, which will be perceived as unable to deliver on their strategic aspirations.

  1. Political discretion

Recipient authorities may exercise legal powers but face practical restrictions in doing so. This might derive from a central-local relationship that comprises ‘delegation’ rather than devolution, or that amount to contractual relations. English devolution has featured many process-based demands from central government, that hem in the capacity for recipient authorities to take innovative actions in line with local strategic priorities. Their political discretion is arguably considerably less than it appears.

Curbs on political discretion may be low-profile, emerging in discussions between officials after high-level agreement on devolved tasks. But they are a critical element of the functioning of devolved powers in practice. They may derive from long-held assumptions about the nature of central-local relations or the requirements of effective governance. As such, the rationales behind them, or their potential effects, are rarely articulated. Understanding how the scope for discretion might influence a new institution’s priorities, and legitimacy, is a crucial element of institutional design.

Alternatively, devolved authorities may frequently be expected to take on responsibility for ‘orphan policy’ areas, for which no tier of government is currently clearly responsible. This can be a political opportunity. A policy issue might rise up the political agenda, but addressing it may not be a statutory duty for any one institution. Funding and strategic thinking affecting the issue can be piecemeal, or spread across several institutions. This scenario provides more oportunity for political discretion, because action by a devolved authority threatens fewer pre-existing institutional or political interests. Recent examples include air quality, mental health, and homelessness. In England, metro-mayors have featured all of these in their priorities since being elected.

  1. Financial autonomy

As noted above, access to funding is essential for any public body to be able to exercise its powers. Devolution of legal powers to a recipient authority will have little effect if it cannot access funding with which to exercise those powers. Both the quantity of funding available, and its sources, influence recipient authorities’ consideration of strategic priorities. The nature of a recipient authority’s sources of income, the amounts available, and the degree of flexibility open to them will therefore influence its decision-making substantially.

Local governments’ financial resources typically derive from a mixture of transfer grants from ‘higher’ tiers of government and local revenue sources (taxes, fees, and commercial activities). Local revenue sources can afford recipient authorities a degree of financial flexibility that increases their capacity for decision-making. Local revenue sources also offer an alternative to relying on transfer grants, and thus the contingencies of central decision-making, for financial health. This avoids the threat of ‘unfunded mandates’, where local governments receive insufficient funding to fulfil legal duties.

Studies that aim to measure local authorities’ financial autonomy can focus on the percentage of local authority revenue that they raise themselves. This aligns with the frequency of the assertion that effective devolution is impossible without substantial fiscal devolution. This forms part of a narrative that sees direct financial accountability to taxpayers as a foundation stone of good governance.

Financial autonomy is potentially a more extensive concept than this, however. For instance, where transfer grants are generous and central governments practise non-interference in spending decisions, a local government’s ‘decision space’ may be broad – even if they raise little of their own funding. Where local revenue forms a large proportion of a local government’s funding, but the total available is insufficient to deliver effective services, the opposite may be true.

Metro-mayors, and local authorities, in England have some taxation and borrowing powers, but in practice these are significantly restricted by a legal framework. They have small amounts of money; are highly dependent on short-term central grant funding; and have little capacity to raise funds locally. This works against their ability to build capacity and institutional memory in the long term. It is that threat that underlies the recent support in some quarters of the sector for ‘fiscal autonomy’ – broadly, a desire no longer to be financial supplicants. Dependence on local income is untypical in local government systems in other states, and it would have sharp implications for equity between local areas.

  1. Organisational autonomy

Local authorities in the UK have long had freedom to organise their staff administration as they see fit. The UK Government holds a small number of reserve powers, and local authorities must fill a small number of statutory positions. Local authorities cannot change their voting system and are required by law either to adopt a directly-elected mayor, leader and cabinet, or committee-based decision-making system. Organisational autonomy is uncontentious within the current debate in England.

  1. Non-interference

The concept of ‘non-interference’ amounts to a demand for permissive relationships between central and local governments. The degree to which a central authority operates veto powers, reserve powers, accountability requirements and financial management strictures will influence the decision-making and strategic priorities of a recipient authority.

The OECD’s report proposes that the role of central government within a devolved polity should be permissive and facilitative, “putting in place the strategies, framework conditions and regulations, incentives, rather than … the implementation of policy… enabling, advising and assisting, ensuring consistency, facilitating the work of local governments and sometimes helping share good practices” (OECD 2019:94).[5]

At present, central-local relationships in England are dominated by financial accountability requirements. This is not purely a manifestation of ‘weknowbestism’ or hyper-centralism, nor does it arise purely from the idea that central government ‘lacks confidence’ in local leadership (TBI 2020: 6): it is a long-standing feature of British governance. Where central government funds pay for devolved activities, the paradigm of financial accountability requires central assurance requirements to follow the funding.

Metro-mayors have seen some movement away from this principle. A number of their funding streams are amalgamated into a ‘single pot’, subject to a single assurance process. Devolution deals include commitments to joint working on future projects, and metro-mayors have scope to borrow against future funding streams for projects chosen locally. Despite the substantial requirements for financial accountability that it operates, it is conceivable that the UK Government sees this as a facilitative, light-touch, permissive approach to central-local relations. In other words, the current financial assurance regime is a minimum requirement of a functioning system of governance.

The OECD’s report recommends that financial assurance of devolved authorities should take place after the event, preferably with the involvement of an external and independent institution. It notes the existence of region-based ‘audit chambers’ in a number of states. The Centre for Governance and Scrutiny, an independent UK consultancy, has proposed a network of ‘Local Public Accounts Committees’ in England. These would audit both devolved and deconcentrated services within their area, providing a place-based approach to accountability.

  1. Access

Many federal or regionalised states include formal mechanisms through which municipal or regional governments can influence decisions by a higher tier of government. This is not the case in the UK: there is no formal representative assembly of local authorities, or legal provisions permitting local authorities to veto, delay or opt out of central government decision-making. UK local authorities are dependent for their influence on individual and collective lobbying efforts, which are inevitably coloured by personal relationships, party politics and policy priorities. Current debates in England on sub-national governance have not addressed this issue.

These seven dimensions of autonomy provide a useful framework to assess proposals for devolution of additional powers. But most importantly, the aspirations, operations, and priorities of recipient authorities are framed by the interaction between these seven dimensions. For instance, a sub-national body’s policy scope (dimension 2) may be limited to one or two areas, but it may have broad political discretion (dimension 3) in those areas. Its policy scope (dimension 2) could be broad in legal terms, but limited financial autonomy (dimension 4) might constrain its ability to act. Its discretion to act (dimension 3) might be broad, but potential or actual levels of interference by government (dimension 6) might be high.



Academic debates on place-based leadership (Durose and Lowndes 2021; Beer, Sotarauta and Ayres 2020) remind us that not all political change is driven by legislation, finance and institutional reorganisation. The ‘convening power’ of political leaders can extend their reach beyond their formal powers. Nevertheless, the institutional architecture within which political leaders work strongly determines the decision space available to them and the incentives that they face. As Gaskell et al argue (2020: 2), structures “frame the way that decision makers work and operate, what they can see and hear, what tools they think they have available to them, what resources they view as at their disposal”. Even in the context of “overloaded” governance (Skelcher 2000), where large numbers of actors exert overlapping influence on policy and delivery, institutions and the location of power matter. Even the strongest of networks can struggle to escape from what Jonathan Davies has called “the shadow of hierarchy” (Davies 2011).

Views will differ on how far English devolution to date, or the proposals in the upcoming White Paper, provide for an effective approach to the transfer of power. Durose and Lowndes note that designs for urban governance are often ‘incomplete’. Institutions may not function as their creators intend; they may be intended to grow into their role; or apparent shortcomings may be intentional design features. Nevertheless, any existing multi-level governance system embodies a position on each of the seven dimensions set out here. Further changes to the English metro-mayoral system will also embody Government positions, whether explicitly or implicitly. This is not to mention the challenges of English sub-national geography, a location where most commentators fear to tread.[6] This paper offers a means by which to assess how new or existing proposals for devolution of power will function in practice – and thus to what degree they will be able to meet the aspirations and expectations of those who support them.


[1]                For instance, the Centre for Cities’ Levelling up local government in England (2020); Achieving Levelling-Up by the LIPSIT project at the University of Surrey (2020); New Local’s No Strings Attached (2020); Fixing the Plumbing (2020) from the Tony Blair Institute for Global Change; Will Jennings, Adam Lent and Gerry Stoker, Place-based policymaking after Brexit: in search of the missing link (2018); LGIU, Power Down to Level Up, 2020; the UK2070 Commission’s publications Make No Little Plans and Go Big Go Local (2020)

[2]                For instance, see Tomaney (2016); Kenealey 2018); Pike & O’Brien; Roberts; Riachrds & Smith; Warner et al 2021

[3]                As Heather Jameson said in a Municipal Journal editorial, “local government lives in an echo chamber of decentralisation. If only, the orthodoxy goes, central Government would trust local government, let go the levers and allow local democracy to forge its own path, public services would thrive. But while it echoes round and round local government, rarely has the sentiment escaped outside the confines of the local arena.” [Heather Jameson, MJ, 22/10/20]

[4]                See IFG, In conversation with Andy Burnham, 8 Mar 2021, 7’40’’

[5]                See Public Administration and Constitutional Affairs Committee, The Evolution of Devolution: English Devolution, HC-825, Q54, 20 Oct 2020; Power Down to Level Up

[6]                Notable exceptions are the Centre for Cities’ 2020 report Levelling up local government in England, and Localis’s The making of an industrial strategy (2017).