Devolution: the importance of scale and coterminosity
JACK SHAW
Devolution has emerged over the last decade in an incremental fashion and, in having done so, significant attention has been paid to the process of deal-making rather than a long-term strategic vision for governing England. The Government has stated its intent to address this deficit in vision. At the same time the Government sees devolution and strong local institutions ‘with skin in the game’ as a vital part of delivering its growth mission “with good jobs and productivity growth in every part of the country”.[1]
In its early months, the Government has seen significant progress. It has signalled an important shift of tone, namely the “partnership” approach it wants to pursue with combined and local authorities – realised in Labour’s decision to invite England’s dozen mayors into Downing Street in its first week of government. It has set about designing the institutional architecture necessary to facilitate ‘knowledge flows’ between Whitehall and England’s sub-regions, which is necessary to boost growth and productivity.[2] This is evident in the establishment of a Council of Mayors – chaired by the Deputy Prime Minister, Angela Rayner – as well as a Council of the Nations and Regions, though it remains unclear whether they will be formally involved in policy development or consultative fora. And it has been integrating combined authorities with its five missions by asking mayors to take responsibility for Local Growth Plans, co-designing industrial strategy and developing regional spatial strategies. At the Budget in October, the Chancellor similarly announced ‘integrated settlements’ for Greater Manchester and the West Midlands, giving them more funding flexibility, and took steps to codify devolution by setting out the criteria by which other authorities can follow in their footsteps.[3] This greater codification will enable mayors to move away from being ‘snake-charmers’ responsible for attracting the gaze of central government in order to secure funding.
Though further information on the specific investment and responsibilities that combined authorities will be entitled is not expected until the revised Devolution Framework in the forthcoming Devolution White Paper (and English Devolution Bill) authorities have already submitted ‘Expressions of Interest’ (EoI) following a request from Rayner. These EoIs have set out authorities’ preferred geographies and governance arrangements.[4]
The EoIs provide the first real insight into how the Government is approaching the task of ensuring that combined authorities are established England-wide – and how authorities are responding in turn. They are instructive of the geographies that achieve a degree of political consensus, though some are no more than compromises between the ‘coalition of the willing’ and intended as ‘conversation starters’, rather than specifically aligned with local identities or, more substantively, sensible economic geographies. Given the Government still has significant progress to make, with 46 per cent of England’s economic output, 52 per cent of its population and 74 per cent of its land mass not covered by devolution, such alignment is essential.[5]
The institutional design of combined authorities – their scale, alignment, capacity and governance arrangements – are central to ensuring that devolution can promote spatially-balanced economic growth.[6] While some tolerance for the messiness of devolution is required, too much misalignment between identities, economies and politics, or combined authorities that are too under-powered or under-resourced, undermines the prosperity that combined authorities are intent on delivering. A critical analysis of the geographies that have emerged is therefore necessary.
Figure 1: England’s proposed combined authorities, according to submissions from local authorities
Source: Analysis of authorities’ Expressions of Interest[7]
Following the closure of the Ministry of Housing, Communities and Local Government’s request for submissions, 17 new combined authorities have been proposed. If approved, there would be 36 combined authorities as well as the Greater London Authority.
These new settlements would cover another 20.8 million people in England, with an average size of 1.2 million and significant variation between them. Bournemouth, Christchurch and Poole would be the smallest by population at 400,000 and Hampshire and the Solent the largest with two million.
There are 11 authorities – accounting for over 1.8 million people – that are at risk of being “orphaned” or stuck in “devolution deserts”.[8] They include Shropshire and Herefordshire, which have decided to work with their neighbouring Welsh authorities Monmouthshire and Powys rather than explore a devolution settlement. As a result, Telford and Wrekin has no clear option available to it – though the prospect of joining the West Midlands Combined Authority (WMCA) has been mooted. Telford does not have a land border with the WMCA – separated by a three-mile stretch of either Shropshire or Staffordshire, depending on where the lines are drawn – but neither does the City of York with West Yorkshire Combined Authority (WYCA), which once held ‘non-constituent’ status.
Elsewhere, no agreement has been reached for Swindon (Labour) to join the Wessex settlement, made up of Wiltshire (Conservative), Somerset (Liberal Democrat) and Dorset (Liberal Democrat). Liberal Democrat-run Bournemouth has decided to seek a devolution settlement on its own, which is unlikely to be successful given it falls below the population threshold. Worcestershire has decided not to submit an EoI for reasons that remain unclear. And Plymouth has very few options. It proposed the idea of a peninsula-wide settlement led by a mayor and incorporating Cornwall and Devon and Torbay – both of which have already agreed a settlement.[9] That was swiftly rejected by Cornwall. With the exception of Worcestershire – which is of a sufficient size to pursue a county-wide settlement alone under current arrangements – these authorities are likely to find themselves in the same position as North Somerset, which is exploring the prospect of joining the West of England Combined Authority. Given North Somerset decided not to join the West of England in 2016 at its inception, and given the electoral consequences of a Liberal Democrat authority joining a Labour-run mayoralty, it’s not clear whether the Government will approve North Somerset’s request.
While some local authorities are intent on understanding the details of a ‘mayoral premium’ before considering a mayoralty, the spatial dimensions of EoIs are revealing. The North of England is already represented by seven elected mayors – and authorities not yet in receipt of a devolution settlement in northern England are similarly supportive of mayoralties. Similarly much of the South East, particularly sub-regions proximate to London, are also supportive of mayoralties. Given northern England and London have the most established mayoral-led combined authorities, proximity to those may be one dimension of their pro-mayoralty stances. At the other end of the spectrum, an ‘anti-mayoral belt’ is concentrated across the country from the South West to East of England – with many of those authorities held by Conservative or Liberal Democrat administrations and for the most part not in proximity to established combined authorities.
Does the size of combined authorities matter?
One of the values of combined authorities is their strategic, spatial function. Under the Devolution Framework set out by the Levelling Up White Paper, the population threshold for a combined authority is 500,000.[10] The criteria determining that threshold is unclear and it should not come as a surprise if the Government increases it. At present, 10 potential combined authorities (and another four that already exist) have populations fewer than one million – with recently approved Buckinghamshire, Cornwall, Warwickshire and Hull and East Yorkshire marginally over the threshold (Figure 2).[11]
Though the Government has suggested that approving these settlements is a “downpayment in good faith” – with the assumption that they will evolve over time – it poses a question mark over the legitimacy and longevity of some of those authorities if the threshold increases, as well as whether they may be expected to merge in the future.[12] Similarly, with the spectre of unitarisation, it is not implausible that an increasing number of local authorities will be more populated than some combined authorities – with the Chair of the County Councils Network recently calling for minimum thresholds for unitaries to be set at 500,000, despite that making it the same as the current threshold for combined authorities.[13]
Figure 2: Combined authorities by population size (under one million)
| Combined Authorities | Population | Constituent Members | Status | Mayoralty |
| Bournemouth, Christchurch and Poole | 404,050 | 1 | EoI submitted | Against |
| Cumbria | 505,063 | 2 | EoI submitted | Open |
| East Sussex | 555,484 | 1 | EoI submitted | Open |
| Buckinghamshire | 566,694 | 1 | 2025- | Against |
| Cornwall | 578,324 | 1 | 2025- | Against |
| Warwickshire | 617,823 | 1 | 2025- | Against |
| Hull and East Yorkshire | 622,061 | 2 | 2025- | Expected (May 2025) |
| Gloucestershire | 659,276 | 1 | EoI submitted | Against |
| Tees Valley | 698,931 | 5 | 2016- | Exists |
| Oxfordshire | 750,230 | 1 | EoI submitted | Open |
| York and North Yorkshire | 834,409 | 2 | 2024- | Exists |
| Cambridgeshire and Peterborough | 919,082 | 7 | 2017- | Exists |
| Devon and Torbay | 972,893 | 2 | 2025- | Against |
| Cheshire and Warrington | 989,908 | 3 | EoI submitted | Open |
Source: Author’s analysis of Expressions of Interest. Existing combined authorities are in black. Combined authorities agreed but not established are in grey. Combined authority proposals – which have not yet been agreed by the Government – are in blue.
If combined authorities are established on the geographies set out by their EoI, they will, on average, be significantly smaller than international counterparts. International evidence of ‘meso-level’ governance – states, provinces, länder, cantons, autonomous communities, prefectures, or regional levels – suggests that they typically represent populations of three million or more.[14] England’s combined authorities do not compare favourably.
Figure 3: combined authorities by population size (over one million)
| Combined Authorities | Population | Constituent Members | Status | Mayoralty |
| Greater Lincolnshire | 1,111,230 | 3 | 2025- | Expected (May 2025) |
| Leicestershire, Leicester and Rutland | 1,154,438 | 3 | EoI submitted | Against |
| Staffordshire and Stoke-on-Trent | 1,161,678 | 2 | EoI submitted | Against |
| Brighton and Hove and West Sussex | 1,180,499 | 2 | EoI submitted | Open |
| West of England | 1,203,401 | 4 | 2017- | Exists |
| Hertfordshire | 1,215,387 | 1 | EoI submitted | Open |
| Surrey | 1,228,671 | 1 | 2025- | Against |
| Berkshire | 1,268,006 | 6 | EoI submitted | Against |
| South Yorkshire | 1,407,072 | 4 | 2014- | Exists |
| Wessex | 1,483,933 | 3 | EoI submitted | Against |
| Lancashire | 1,570,373 | 3 | 2025- | Against |
| Liverpool City Region | 1,586,149 | 6 | 2014- | Exists |
| “Orphan” | 1,832,403 | N/A | N/A | N/A |
| East Anglia | 1,708,385 | 2 | EoI submitted | Against |
| Greater Essex | 1,896,590 | 3 | EoI submitted | Open |
| Kent and Medway | 1,897,051 | 2 | EoI submitted | Open |
| South Midlands | 1,958,182 | 6 | EoI submitted | Against |
| North East | 2,012,449 | 7 | 2024- | Exists |
| Hampshire and the Solent | 2,035,872 | 4 | EoI submitted | Open |
| East Midlands | 2t,251,679 | 4 | 2024- | Exists |
| West Yorkshire | 2,402,161 | 5 | 2014- | Exists |
| Greater Manchester | 2,948,633 | 10 | 2011- | Exists |
| West Midlands | 2,980,936 | 7 | 2016- | Exists |
| Greater London Authority | 8,931,847 | N/A | 2000- | Exists |
Source: Author’s analysis of Expressions of Interest. The keys in Figure 2 apply to Figure 3.
The literature is clear that devolved forms of governance at larger scales can facilitate stronger and more equal economic growth than highly centralised governance arrangements that exist in England specifically, and the United Kingdom more widely.[15] Similarly, smaller combined authorities – especially those without mayors – make limited progress on the international stage, making attracting Foreign Direct Investment through international collaboration difficult. The threshold for joining C40 Cities – a global network of mayors from nearly 100 cities collaborating to tackle climate change – is a population of three million by 2030; only the Greater London Authority is a member.
The smaller scale of combined authorities is also likely to create institutions with insufficient capacity and capability, as well as ‘weaker voices’ that are more easily ignored by government. As Diane Coyle has set out, the absence of adequate sub-national governance prevents “knowledge inputs” of locally specific information being heard in government. Instead the concerns and priorities of London (and to a lesser degree, Greater Manchester) will remain dominant.[16] This undermines policy development and encourages short-termism. And as Philip McCann has outlined, this is not a “market failure” but instead a “problem associated with poor institutional design”.[17] For these reasons, smaller combined authorities undermine the productivity-enhancing potential of devolution. Scale – alongside governance arrangements – will also have a bearing on the powers and funding streams available to combined authorities, either directly or indirectly.
The population size of authorities poses other implications for governance which are also problematic. Hull and East Yorkshire and York and North Yorkshire are among over half a dozen combined authorities that will have only two local authority ‘constituent’ members with voting rights. One emerging interpretation is that the arrangement gives those combined authorities a competitive advantage, with agreement among fewer authorities enabling them to move at pace. But an equally plausible scenario is that it places them at greater risk of paralysis, especially given many combined authorities with two constituent members are from different political parties. The logic of consensus-driven combined authorities is, at times, in tension with the pace at which mayors are expected to make progress.
Smaller combined authorities also pose questions of equity and of the role of such authorities in the Council of Nations and Regions. How will the Government defend its decision to exclude Scotland’s City Region Deals – some of which have larger economies and populations than English combined authorities? And how will the Government navigate a fora with the potential for 12 English mayors for each representative from the devolved nations?
Boundaries
The EoIs also raise questions about the vexed issue of boundaries. They reveal a number of poorly designed geographies, which lack coterminosity with either NHS Integrated Care Board, Fire or Police boundaries. And the failure to align them at their inception risks embedding complexity and preventing combined authorities from adopting a ‘total place’ lens that might enable them to drive public service productivity or reform.
Figure 4: Integrated Care Boundaries overlaid over Expressions of Interest
Source: Analysis of authorities’ Expressions of Interest
Figure 5: Police constabularies overlaid over Expressions of Interest
Source: Analysis of authorities’ Expressions of Interest
Figure 6: Fire authorities overlaid over Expressions of Interest
Source: Analysis of authorities’ Expressions of Interest
Sussex is a case in point of misaligned boundaries. Brighton and Hove and six West Sussex district authorities have submitted an Expression of Interest. West Sussex County Council has submitted its own Expression of Interest, without the support of its districts. And East Sussex County Council has submitted an Expression of Interest, though with the support of its districts. Combining these interests into a Sussex-wide settlement would provide the necessary scale and alignment, with 1.7 million people and an economy of £48.7 billion across the county footprint, as well as strong connectivity with Gatwick airport, the second busiest airport in the UK.[18] And it aligns well with NHS, Police and Fire authorities (see Figures 3, 4 and 5) as well as historic boundaries. West Sussex acknowledges that the Government is “likely to want to seek to devolve to an area larger than West Sussex”.[19]
A second example is Oxfordshire and the six authorities in Berkshire, which are seeking separate devolution settlements despite their long history of collaboration, not least through Thames Valley Police. Establishing a combined authority of that scale – alongside Buckinghamshire at a later date, since the Government has already agreed a devolution settlement there – would bring together the innovation and research assets of Oxford, industries in Reading and Slough and the connectivity of the M3, M4 and M40 corridors. It is also plausible that, if Swindon does not join a proposed Wessex devolution settlement, it will have to look to Thames Valley too.
Where coterminosity does exist, it’s not homogeneous. Fewer than half of England’s combined authorities are run by mayors that are also Police and Crime Commissioners – Greater Manchester, London, South Yorkshire, West Yorkshire and York and North Yorkshire. And in cases where mayors are beneficiaries of health devolution, those powers are relatively underdeveloped. The Mayor of South Yorkshire, Oliver Coppard, is co-Chair of the Integrated Care Partnership (ICP). Andy Burnham is similarly co-Chair of Greater Manchester’s ICP alongside Sir Richard Leese.[20] Neither are Chair of their Integrated Care Board (ICB), which has significantly more responsibility, including for planning NHS services. These are only possible with coterminous boundaries.
A ’maximalist’ approach to coterminosity would enable mayors (or an appointee of their choice) to become Chair of their ICB, with responsibility for setting their strategic direction, aligning resources and – as is the case with policing – recruiting and dismissing the Chief Executive. Doing so would provide more democratic oversight sub-regionally over NHS activity. A similar question could be levied at the Department for Education and its Regional Directors responsible for overseeing school performance, though as London is the only authority with a coterminous boundary it should be the first to pilot such devolution. Multi-agency partnerships such as Local Resilience Forums should also be re-aligned. Given public sector productivity – which increased by 0.7 per cent annually between 2010 and 2019 – was largely driven by productivity in education and healthcare, if combined authorities are not aligned with and have oversight over those sectors, their ability to improve prosperity and increase productivity will be limited.[21]
As Sam Freedman recently set out, coterminosity has already begun to bear some fruit. In Greater Manchester a long-term programme for women with multiple needs has reducing re-offending rates by taking a systems-approach. Issuing women with a Court Order rather than a custodial sentence has enabled them to stay within their communities and families, which has a significant and long-term impact. The programme is co-commissioned with Greater Manchester Probation Service following the devolution of the national ‘Dynamic Commissioning Framework’ led by His Majesty’s Prison and Probation Service, but probation geographies do not align with other combined authorities either, so Greater Manchester is the only beneficiary to date.[22]
Leaving aside that approaches to coterminosity are uncodified, it is not clear whether it is even a desired outcome for the Government. This should be corrected. And while there is an opportunity cost to wholescale re-organisation of local authorities – something that Labour will be aware of now that it has proposed in the Budget working with local authorities on this issue – there are instances where central government has progressed boundary changes without significant friction; notably in 2018 when Local Enterprise Partnerships were asked to review and change their boundaries if they overlapped. In some cases, pan-regional partnerships would also merit alignment, though the Government is currently consulting on terminating its investment in pan-regional partnerships so their ongoing existence is far from certain.
Why does the design of combined authorities matter?
The preoccupation with the design of combined authorities is not inimical to outcomes and is instead central to them. Communities across England are better served with devolution settlements that include suitable governance arrangements. Enhancing their growth potential through appropriately scaled combined authorities and aligning them with other services is likely to generate allocative efficiencies and improve the quality of public services, which is particularly important given the public sector accounts for 20 per cent of GDP in the United Kingdom.[23] Moreover, coterminosity is a gateway to further fiscal devolution and there is significant evidence of the relationship between further devolution and economic growth.[24] And the Chancellor has already been explicit that her plans to boost public service productivity are contingent on “the integration of services at a national and local level”.[25]
While EoIs submitted by local authorities represent the starting point for ‘negotiated autonomy’, they expose the broad contours of sectoral appetite. Their revealed preference is for combined authorities that are smaller in scale – and without coterminosity in public services, despite the potential this has for public sector reform and productivity gains across a broader range of services and funding streams.
The Government has, thus far, adopted a pragmatic approach – had it not done so, fewer places would have come forward with devolution proposals – it must now determine the suitability of proposals, which are likely to be robustly contested, including among some Labour authorities.
- Given the proposed scale of some combined authorities – which would on average be less populous than existing combined authorities and not in keeping with meso-level governance seen elsewhere – the Government should set out a minimum threshold that all new authorities observe. Bournemouth, for example, does not have the necessary scale. There will, however, need to be exceptions to the rule; a straitjacket will consign some local authorities to ill-suited geographies. Doing so would, in turn, align the expectations of authorities and Ministers.
- Second, the functions of combined authorities need to be better integrated with services operating at similar geographies – principally the NHS, policing and fire services. There is no codified framework to advance this agenda, which is why the Government should give consideration to a ‘coterminosity duty’. This could be set in motion in the White Paper. It would give the Government a better chance of delivering on its Growth Mission and give combined authorities the opportunity to realise long-term change in their communities including addressing poor productivity performance in England’s second cities.
- Third, governance arrangements are not afforded the attention they deserve. While this note explores questions related to the number of constituent members – namely whether fewer of them facilitate agility or paralysis – there are a wider set of governance arrangements, including the executive functions vested in elected mayors, that need to be codified.
- Fourth, in setting the rules of the game, the Government must also be willing to enforce them. In circumstances where authorities are unwilling to countenance a devolution settlement, the Government must impose one. On the surface this is contrary to the spirit of devolution, but the alternative – abandoning a small number of authorities – is not an option. To encourage consensus-building, the Government should give authorities sufficient warning by setting a date by which it expects devolution proposals to be agreed. And in instances where there is no agreement, it should be clear about the consequences.
Ensuring that the whole of England is represented by combined authorities is of national significance. Establishing combined authorities England-wide will better equip the Government to support place-based growth strategies and public service reform. And it will prevent the postcode lotteries that uneven administrative arrangements risk creating. Not only must they be established England-wide, but they must be of sufficient scale, capacity and capability to make meaningful change, be entrusted with suitable governance arrangements and be integrated with a broader set of public services.
Jack Shaw is a Policy Fellow at The Productivity Institute’s Productivity Policy Unit.
[1] The Labour Party, Mission driven government: growing the economy, 2023.
[2] Coyle, D. and Muhtar, A. UK’s Industrial Policy: Learning from the Past? Productivity Insights Paper No. 002, The Productivity Institute, 2021.
[3] Ministry for Housing, Communities and Local Government, Integrated settlements for Mayoral Combined Authorities, 2024.
[4] MHCLG, The next steps to devolution, 2024.
[5] The Institute for Government, Explainer: English devolution, 2024.
[6] Coyle and Muhtar, UK’s Industrial Policy: Learning from the Past?, 2021.
[7] Thanks to William Sarsfield and Raquel Ortega-Argiles for their help visualising the analysis. The visualisations are available to view on Figshare.
[8] Marinko, P. Minister: England’s devolution map must avoid ‘orphan areas’, The MJ, 23 October 2024. MHCLG, Deputy Prime Minister kickstarts new devolution revolution to boost local power, 2024.
[9] Plymouth Herald, Devolution deal between Plymouth and Cornwall could be on the table, 5 September 2024. In 2023 a motion presented at Cornwall Council’s Full Council proposing a “National Assembly of Cornwall or a Cornish Parliament” was only defeated by two votes: it was highly unlikely that Cornwall would countenance a cross-peninsula settlement, given its nationalism.
[10] MHCLG, Levelling Up White Paper, 2022, p.137
[11] In Figure 3, West of England’s population data has been updated to include North Somerset, but as it stands the authority represents a population of 982,255.
[12] MHCLG, Four devolution deals signed off and others progressing, 2024.
[13] Jameson, H. Write off the debts or reorganisation will fail, The MJ, 13 November 2024.
[14] McCann, P. The fiscal implications of ‘levelling up’ and UK governance devolution, Productivity Insights Paper No. 008, The Productivity Institute, 2021.
[15] Carrascal-Incera et al, UK interregional inequality in a historical and international comparative context, Cambridge University Press, 2020; McCann, P. The fiscal implications of ‘levelling up’, 2021.
[16] Coyle and Muhtar, UK’s industrial policy: learning from the past?, 2021.
[17] McCann, P. The fiscal implications of ‘levelling up’, 2021.
[18] Greater Brighton Economic Board, Devolution, Item 11, 16 October 2024.
[19] West Sussex County Council, Expression of Interest, 2024.
[20] Greater Manchester Integrated Care Partnership, Mayor of Greater Manchester is the new co-chair the Greater Manchester Integrated Care Partnership, 27 September 2024.
[21] Van Ark, B. Making Public Sector Productivity Practical, 2022.
[22] Freedman, S. Public Service Reform and Devolution, 2024. Greater Manchester Combined Authority submission to Justice and Home Affairs Select Committee, 2023.
[23] Van Ark, Public Sector Productivity, 2022.
[24] Blöchliger, H. and Égert, B. Decentralisation and Economic Growth – Part 2: The Impact on Economic Activity, Productivity and Investment, OECD Working Papers on Fiscal Federalism, 2014.
[25] Chancellor of the Exchequer, Public Spending: Inheritance, 29 July 2024.
