Peter McDonald explores the opportunities and challenges of a smaller government
Standing on an exposed common in rural Carmarthenshire, deep in blanket bog, debating with local farmers and agricultural specialists in the driving rain, was the moment I realised the true scale of my journey away from Whitehall to Welsh Government. Only one month earlier, I had come to the end of a 12-year stint in HM Treasury, where my practical experience was limited to searching for an available hot-desk each morning. Now, somehow, I was leading a 60-strong team for whom ‘out in the field’ meant just that.
I confess I had expected an easier transition. On the face of it, I was leaving behind the frenetic pace of a job preparing biannual budgets and advising on UK-wide policy levers while buffeted by the political winds that constantly blow down Parliament Street. But while the contrast between central and devolved government has proved stark, I have found the Welsh experience to be no less rich, challenging or rewarding – it has simply been ‘a different type of hard’.
As I have adapted to devolved life (since 2017), I have increasingly come to appreciate that the differences reflect far more than just geography. To understand devolution – and to be effective within the devolved system – I first had to learn its story.
A history primer
In 1999, three ‘new’ governments were formed in the UK: the Welsh Assembly, the Scottish Government and the Northern Ireland Executive. But this was reincarnation rather than creation: each body had previously existed under the guise of the Scottish, Welsh and Northern Ireland Offices, respectively. The advent of devolution meant those organisations became accountable to new devolved parliaments and led by new structures of Cabinet government. This was a huge shift for those sitting in Edinburgh, Cardiff and Belfast – in the space of months, officials went from serving a single Secretary of State to a full Cabinet of devolved Ministers, accountable to a wholly new parliamentary architecture. It is hard to overestimate the scale and speed of change.
As such, the devolved governments are both new and old institutions. Similar to any Whitehall department, their institutional memory stretches back decades, but in constitutional terms they are still young. Not only that, their remit has expanded since birth, with many new powers added to the original devolution settlements. Give or take, the areas which a devolved government is responsible for encompasses health, education, skills, transport, environment, housing, culture, regional development, agriculture and many others. But there are important differences between each nation, and the UK is now a patchwork quilt built in layers (all the more varied when England’s combined authorities are included). For that reason, this article can only reliably consider the Welsh experience; a future Heywood Quarterly article would benefit from accounts of the Scottish and Northern Irish perspectives.
Certainly in Wales, devolution took some time to find its feet. The 1997 referendum was won by the narrowest of margins (50.3% to 49.7%). This was a tentative mandate which mirrored the original design of the Welsh Assembly, its competence constrained to narrowly defined areas without primary legislative powers. These fetters have been removed through successive settlements and the process continues: in 2026, the membership of the Welsh Parliament will expand by over 50% in order to increase the legislature’s capacity to scrutinise the executive. To work in devolved government is thus to experience regular change – with changes to date always involving more devolution and more responsibility, not less.
The question of optimal size
Notwithstanding the expansion of responsibilities, the first noticeable feature of Welsh devolved government is its small size. At around 6,000 staff members, the entirety of it is around the size of a single modest Whitehall department (similar to the Department of Transport; smaller than the Cabinet Office). Consequently, officials can be spread thin and most have a remit far wider than their UK government counterparts; it is not uncommon to find a single official covering the same brief as a whole Whitehall team (the same is equally true of Ministers – one Welsh Cabinet Minister could easily have more than five opposite numbers).
This raises the question of the optimal size of an administration. The modern reality is that good government has a fixed cost (all the nations of the United Kingdom need a set of policies backed by consultation, analysis and legislation). That might suggest economies of scale will arise from larger units. At the same time, I have come to appreciate that being small has its virtues – sometimes significantly so.
To begin with, small government facilitates – perhaps even necessitates – a strong relationship between policy and delivery, often because the two are done in the same place. A single Senior Civil Servant-led team might typically cover the overall strategy, detailed policy and practical implementation for a specific area. Vertical integration can work well, but it requires multi-disciplinary teams with a wide range of skills, some of which can be hard to come by in a small government. Path dependency is also a strong force: the old Welsh Office focused on administrative delivery, and devolution meant its character changed to the point where Ministers expected more policy development and innovation. The Welsh Government has therefore been continually tested to build new capability to respond to successive devolution settlements (particularly the skills required to prepare primary legislation).
Further, just as a start-up can prove more nimble than a large corporate, a small government allows for low-friction cross-government working. While each individual team is small, organisational physics makes it easier to convene than in Whitehall. This applies to all levels of government: even before the rise of virtual meetings, it was perfectly feasible to assemble all local authorities, police forces or health boards in a room and have a relationship with each. This all helps when solving a problem requires intervention from multiple quarters: for example, a piece of transport advice could easily draw on insights from health, environmental, planning and economy colleagues before ever being seen by a Minister – indeed Welsh Ministers have come to demand this degree of partnership in the work they receive.
As such, the fences within government can be flexible and low, enhanced by the strong Welsh culture of using Cabinet for collective decision-making, strategy and coordination. One particular manifestation is that Special Advisers are appointed by the First Minister and are technically part of his or her office; they operate as a single team in support of Cabinet, a very different dynamic from Whitehall. As a result, there is less sense of each Minister having their own Praetorian Guard. With fewer reshuffles than in Westminster, these devolved Ministers also tend to stay in their roles for longer, which influences the time horizon of decision making.
Policy in the open
These ways of working shape the Welsh policy-making process, which takes place in its own political atmosphere, its own level of outside scrutiny and its own tradition of civil society. With far fewer backbenchers, no second chamber and a limited number of Wales-specific media outlets (most Welsh people consume UK news), there is a greater opportunity to do policy in the open. This can allow for the sort of debate that might struggle to overcome the political economy of Westminster.
None of this necessarily makes decision-making any easier, but the process has a distinct flavour. In particular, the expectation of consultation and working with partners is a major feature, balanced by less scrutiny from more traditional sources. The principle of ‘co-design’ is frequently invoked and, just as with cross-government working, often driven by Ministers themselves.
It may be that this culture has developed because the system of devolved government was inherently designed to be more pluralist than Westminster. In Northern Ireland, it is hard-wired into the Cabinet architecture; elsewhere, it is reflected in the voting system that makes political majorities the exception (Wales has never had a majority government since the start of devolution). The idea of working together is reflected both in the political and the physical: the Scottish and Welsh Parliaments are rounded horseshoes. In Cardiff Bay, parliamentary committee work is done with all attendees – Committee scrutinisers and invited witnesses alike – sitting around the same quadrilateral table.
The challenges of devolution
Before readers pack their bags and head west, they should not assume that devolution has created governmental utopia. Far from it: Wales is steered by its own political dynamics and has its fair share of constraints, just like Scotland and Northern Ireland, or indeed any region of England. These challenges occupy a broad spectrum. At a micro level, proximity to the electorate in a small country means there is an expectation that government should be able to account for its impact on individuals as well as groups (whereas there is an acceptance in Whitehall that government must sometimes paint in broad strokes). At the other end of the scale, the interaction with UK-level politics means that the Welsh Government is shaped by two electoral cycles rather than one.
This is especially relevant given that devolved budgets are the product of the UK Government’s plans. The mechanical Barnett Formula is the cornerstone of the so-called ‘fiscal framework’, essentially the set of rules for the regulation of devolved spending and some taxation. It is possible to perceive these arrangements as a constraint, particularly as future devolved budgets can only be finalised once the UK Government has set budgets for Whitehall departments.
As a former Treasury official who worked on devolution finance, I can entirely appreciate the trade-off between maintaining control of the aggregate public finances and providing flexibility for different levels of government within the system. However, in a devolved administration, it can be hard to reconcile the many risks and liabilities of devolved responsibilities within what is a tight budgetary framework. For example, as a rail operator, the Welsh Government manages all the risks and liabilities of running train services. On the face of it, this is no different from the Department for Transport’s challenge in England, but the ups and downs are much harder to manage within a balance sheet more than 95% smaller than the UK’s – akin to landing a jumbo jet on a postage stamp.
The potential of ‘natural experiments’
In all of this, it is easy to lose sight of the fundamental purpose of devolution – to bring decisions closer to the citizens affected by them. Ergo, we should not be surprised if there are different policies in different parts of the UK, especially if those places are led by governments of different political colours.
As well-intentioned public administrators, this does not always sit comfortably; for understandable reasons, our instincts often err towards uniform, streamlined business. But devolution is not simply about having four different administrative regimes for the same policy; it is about facilitating the right policy in the right place.
Through this diversity, devolution has influenced all parts of the UK, including England. Many UK-wide policies began as devolved schemes; the opt-out system for organ donation, the plastic bag charge, and the ban on smoking in enclosed public spaces are but three examples of policies pioneered by a devolved government and then replicated in each of the four nations. At the same time, devolution has, as you would expect, resulted in divergence. Three examples are the privatisation of buses (undertaken across the UK in the 1980s with the exceptions of Northern Ireland and London), tuition fees (each nation has a different level of support for students) and prescription charges (abolished in Wales, Scotland and Northern Ireland but not in England). Now that many powers have been returned to the four nations from the European Union, the post-Brexit world may allow for new areas of variance.
In one sense, these differences can be viewed as a package of ‘natural experiments’, allowing us to analyse outcomes from different policies among fundamentally similar countries. In 2016, the Institute for Government (IfG) eulogised that “devolution provides us with a wonderful opportunity to develop and share innovative and creative approaches to social policy.” But some have argued that the opportunity has not been fully grasped; the same report goes on to conclude that “the UK has failed to live up to this promise of becoming a ‘living laboratory’ for policy exchange and development.”
Clearly, the workings of democracy are a constraint: between the General Elections in 2010 and 2024, no party was in power in more than one of the four nations at any given point. As our duty is to serve the Ministers of our respective governments, it is perfectly valid for politics to set the scene, including in relation to the Union itself (each of four governments often has a different perspective on the constitution).
Opportunities for co-operation
However, this does not mean we lack opportunities for objective engagement, learning and cooperation. Many civil servants know this to be true: up and down the country there are countless examples of officials working together across the UK – entirely in the open, but rarely in the limelight.
More co-operation and learning is surely possible and desirable, but it is not my place to offer a prescription for the entirety of the Civil Service. What I can say is that I regret not doing more myself. While working in the UK Treasury, my focus was on advising Ministers on the macroeconomic levers within their control. This approach rarely involved strengthening relationships with devolved administrations. I can now see this was short-sighted: in the clear service of Treasury Ministers, I could have delivered more with a greater knowledge of what was going on outside my immediate fiefdom, more openness to dialogue and a better attitude to learning.
I might have made different choices were I in today’s Treasury, which now benefits from a second site in Darlington. This really is revolution rather than evolution: even a decade ago, the idea of ‘regionalising’ the department was largely unimaginable. While a multi-location Treasury is different from devolved government in that it still works to the same political centre, there is far greater socio-economic similarity between the North-East and Wales than between London and any other region.
From my vantage, introducing the Treasury to perspectives and people far from Westminster can only be a good thing. Indeed, I would not be the first to argue that Civil Servants aspiring to senior echelons should first take a tour of duty outside of Whitehall, be it in devolved government, an English region or local government – anything that changes their frame of reference. Most UK Government officials in London-based and London-centric departments would probably benefit from practising their trade in a smaller field, with wider responsibilities and working closer to the operational level. But it cuts both ways: many devolved officials could equally gain from experiencing policy making on a broader landscape at a greater distance from the citizen with the choices that entails.
If you ever get the opportunity, I would strongly recommend conducting your own natural experiment. I have learnt much from mine.
Peter McDonald, Director of Transport and Digital Connectivity in the Welsh Government
