House of Commons - Justice Committee

Devolution: A Decade On - Justice Committee Contents

Memorandum submitted by Alan Trench, Hon Senior Research Fellow, The Constitution Unit, University College London

1. I am an academic researcher at The Constitution Unit, part of the School of Public Policy at University College London. In this role I have been responsible for undertaking research on various aspects of intergovernmental relations in the devolved United Kingdom since 2001, mostly funded by either the Leverhulme Trust or the Economic and Social Research Council. I have edited several volumes in the Constitution Unit's series of devolution yearbooks The State of the Nations (and am in the process of editing the 2007 volume, expected to be published in November 2007), as well as a volume entitled Devolution and Power in the United Kingdom, which is due to appear from Manchester University Press in July 2007 and which is concerned with how intergovernmental relations have worked since 1999. I was also specialist adviser to the House of Lords Select Committee on the Constitution for its inquiry into Devolution: Inter-Institutional Relations in the United Kingdom in 2002-03.[28]

2. In this submission, I will draw upon my own academic work and that of colleagues, to address a number of the issues that the Committee have raised. Underlying this is a large number of interviews with those involved in devolution (mainly officials, and also politicians), carried out since 2001 and still ongoing. I also draw upon experience of interviewing those involved in intergovernmental relations in a number of other decentralised and federal systems (notably Germany, Spain, Australia and Canada). In particular, I will address questions 3-7 as set out in the Call for Evidence.

3. There are two general points to make about devolution in the UK, the importance of which is not always fully appreciated. First, the working of devolution to date is that it has been underpinned by the dominance of the Labour party in all three British governments. (Even in Northern Ireland, the fact that the British parties do not contest elections means that there has been no competition from the parties there.) Labour's dominance of the three governments has had important and far-reaching implications, and not all aspects of that are always appreciated. One is that there has been substantial consensus on political objectives and values, which have limited the scope for governments to disagree with each other. However, perhaps more important than this is the shared electoral interest of all parts of the Labour party, and how this affects intergovernmental relations. Labour has a strong interest in making devolution appear to be a success, and it has treated the absence of overt disputes between governments as an indicator of that success. There has therefore been the opportunity and incentive, as well as the means, to resolve any differences that do arise between governments quietly and behind the scenes, rather than to do so openly in public. If (or when) the Labour party loses an election, and different parties hold office in London and Edinburgh or Cardiff, the goodwill that has underpinned relations since 1999 will be significantly weakened, and this shared interest in avoiding open disputes is likely to vanish.[29]

4. One might add that this sort of behaviour by the Labour party is understandable in UK terms, but is not normal in other regionalised systems let alone federal ones. In such systems parties are conscious of the need to put the interests of territory ahead of party, and be seen by the electorate as a strong voice for Victoria, Baden-Württemberg (or wherever) in dealing with the federal or central government, and if necessary standing up to the central government.

5. What devolution involves in the UK is highly unusual in another comparative respect. The UK has a developed, industrial (or post-industrial) economy and an advanced welfare state, both established in the context of the UK as a single state. Devolution has involved transferring a number of functions related to the welfare state and to a lesser degree the economy to the devolved administrations in Scotland and Wales (these were always separately administered in Northern Ireland, because of devolution to Stormont between 1922 and 1972). This has meant a sort of "spinning out" of functions, and a disentanglement of functions previously exercised by one government. This is a complicated process that has left many overlaps of competence, even for Scotland, where the settlement is clearest in both constitutional and administrative terms. The effect of this is that it is almost impossible for any government (whether the UK Government or a devolved administration) to make any policy that does not have some sort of spill-over effect for other governments. The high level of formal legal autonomy that Scotland has (and that Wales may develop under the Government of Wales Act 2006) is not mirrored by the practicalities of governing. Many policy actions create some sort of implication for other matters, which have to be identified and resolved if the policy is to be successfully brought into effect.

6. Most other decentralised or federal systems have built their economy and welfare state around those institutions (even Spain, which may have only established its asymmetric decentralised structure in 1981 but which had very limited welfare provision and a generally backward economy until the fall of Franco). The only other developed country state to attempt a similar sort of decentralisation is Belgium, which resembles the UK in having similarly complex intergovernmental relations, and a peculiar if not asymmetric constitutional structure.

DEVOLUTION AND WHITEHALL

7. Whitehall's response to devolution has been a limited and low-key one. It has two aspects—what happens in ordinary or line departments, with policy functions, and what happens at the centre of government. The latter is discussed in more detail below. As regards line departments, around 1998-99 arrangements were put in place in most departments to manage devolution matters, including setting up devolution or constitution desks or teams. These sought to co-ordinate the department's actions regarding devolution, help prepare bilateral or departmental concordats, and generally to provide internal advice. These were generally dismantled after 2001 (they survive in the Foreign and Commonwealth Office and Ministry of Defence, however). A network of departmental "devolution contacts" still exists, but is a far cry from the sort of detailed attention to devolution matters that there used to be.

8. In principle, devolution is considered to have "mainstreamed" across departments. It is one of the many areas in which officials are expected to have expertise, as one of the basic elements of civil service training. In practice, there is good reason to question how effective mainstreaming has been. Awareness of devolution and its practical implications has improved since 2001, but remains somewhat patchy and inconsistent. Some officials and teams are sensitive to devolution issues, spot them early and resolve them effectively. Others are not and do not. (Effective resolution very often means getting in contact with the territorial offices, which are the UK Government's main source of technical expertise about the devolution arrangements, as well as having a broader view of overall UK Government policy.) Some departments (and their ministers) are inclined to be helpful to the devolved administrations, and others are not. This is made more acute by the fact that practically all Whitehall departments are responsible for a mixture of England-only and UK- or Great Britain-wide functions (and none have a organisational map explaining the territorial impact of their responsibilities).[30] Both looked at from the inside and judging by outcomes, the overall pattern remains highly variable and inconsistent.[31] Two particular cases have drawn the attention and ire of the Welsh Affairs Committee: the Children's Commissioner for England (who has functions in Wales as well, for non-devolved matters relating to criminal justice), and policing and police re-organisation.[32] Similar problems for Scotland have not attracted public or Parliamentary attention to the same degree, but nonetheless appear to have arisen (the transfer of responsibility for railways and the Scotrail franchise is an example).

9. I should emphasise that the problem here is not that the Government's procedures are not the right ones. By and large they are. It is that those procedures are not always followed in a systematic way, and that there are no authorities at the centre of government to enforce them. This is partly a consequence of bureaucratic organisation, and partly of the working of the present Government. In particular, the tendency for 10 Downing Street, on behalf of the Prime Minister, to take control of a particular policy issue, can result in the devolution rules being overlooked.

10. Regarding concordats, these must be among the least-used documents in government. The Memorandum of Understanding[33] mostly sets out high-level principles for intergovernmental relations, largely designed to preserve ways of working that had grown up before devolution. Bilateral concordats between particular Whitehall departments and devolved administrations largely repeat those principles. Neither is regularly consulted, partly because they do not address most tricky situations that have arisen and partly because there are no meaningful consequences for breach. (Most notably, the Memorandum of Understanding contains a commitment for the Joint Ministerial Committee in its plenary form to meet every 12 months—but it has not done so since November 2002.) Preparing bilateral concordats, in particular, was clearly a useful exercise in 1998-2000, for making Whitehall departments think about how devolution would work and what it would mean for them. Since then, however, they have largely gathered dust. It is hard to resist the conclusion that the Memorandum of Understanding is no longer fit for purpose, and that neither it nor the existing bilateral concordats will be effective instruments when real differences between governments emerge.

THE ROLES OF THE TERRITORIAL OFFICES AND THE TERRITORIAL SECRETARIES OF STATE

11. My remarks in this section are addressed to the situation concerning Scotland and Wales, not in general toward Northern Ireland. The special circumstances of Northern Ireland mean that it will still be some time before any incorporation of arrangements for Northern Ireland into those for Great Britain can be considered.

12. The territorial offices seem, at present, to work well. They serve as general sources of expertise and advice on the devolution arrangements for Scotland and Wales, as well as playing important parts in identifying and resolving differences between administrations that in other circumstances might turn into open disputes, and dealing with particular matters concerning each country (the Government of Wales Act 2006 is an obvious example). A problem of disparity of size and role that existed in 2002-03 (when the Scottish Office had 85 staff but relatively little business, while the Wales Office had only 41 but a great deal of work to do) no longer exists, as the Wales Office has grown in size while the Scottish Office has shrunk.[34] The two offices appear to work well together, and to spot issues affecting one set of devolution arrangements that have implications for the other (a so-called "read across"). In fact, the offices prove to be a vital part of the glue of devolution at the UK end, as they identify and resolve many of the large number of difficult technical issues that the entangled nature of the devolution arrangements (discussed in para 5 above) throw up.

13. The position of the territorial offices within the Department of Constitutional Affairs has created some lurid press speculation, but appears to work well. The principal problem that arises is that there is a missed opportunity for ensuring better overall co-ordination within the UK Government. Such co-ordination has not existed since 2001 or thereabouts, when the Cabinet Office lost its role in relation to devolution and its Constitution Secretariat was formally wound up. Some of the officials involved in the Constitution Secretariat continued to provide overall co-ordination (and acted as secretariat for the Joint Ministerial Committee), first within the Office of the Deputy Prime Minister and later within the Department for Constitutional Affairs, but that role has shrunk so it now takes relatively little time and gets relatively little official or ministerial attention. (In May 2006, the Ministerial Committee on Devolution Policy was wound up, and its functions subsumed within the broader Constitutional Affairs Committee.)

14. The roles of the territorial Secretaries of State are more complex. Again, the fact that the present holders of the posts of Secretary of State for Scotland and Wales only spend part of their time on those territories has attracted a good deal of press comment. In practice, it appears that each Secretary of State spends about 10-20% of his time on Scottish or Welsh matters, the amount depending on the flow of business (and the demands of their other ministerial post). This has included keeping in regular contact with the devolved administration generally, and the devolved First Minister in particular. (Peter Hain and Rhodri Morgan used to meet every week, but that appears to have been reduced in the last few months. Personal contact between the Scottish Secretary and First Minister was never so frequent.) This seems to be pretty much sufficient; it has meant that there is a senior minister able to intervene when necessary, without having to spend time on activities that are not otherwise necessary in order to justify their office or salary.

15. In effect, there is a "hollow centre" at the heart of government when it comes to devolution. At ministerial level, responsibility is shared between five ministers (Secretaries of State for Scotland and Wales, the Lord Chancellor, the Secretary of State for Communities and Local Government, who leads on English regions, and the Deputy Prime Minister). With the exception of the Deputy Prime Minister, each has other, more time-consuming matters in their portfolio as well. This is likely to become slightly worse with the establishment of the Ministry of Justice and transfer of offender-management functions to DCA from the Home Office.

16. However, this situation creates problems. First, it assumes that there is a generally benign climate of relations between the UK Government and the devolved administration involved, and that the Secretary of State and his counterparts in the devolved administration have substantially similar views. They are also predicated on the Secretary of State being able to speak for their party in that territory—on an understanding of its political needs and conditions, and the interests of the UK Government there. If different parties hold office in London and Cardiff or Edinburgh, that is unlikely to be the case, as UK Secretaries of State may be concerned with the position of their own party in Scotland or Wales, rather than developing or maintaining good relations between governments. If the party in office in London has little or no electoral support in Scotland or Wales and little prospect of improving it, the standing of the Secretary of State in the UK Government is likely to suffer and consequently they will be a less effective advocate for the territory.

17. Second, it limits the portfolios a minister can hold. The issues that affect Scotland and Wales mainly come out of the Government's domestic policy agenda. To be an effective (and time-efficient) Secretary of State, a minister needs to be involved in that business as a result of their other portfolio. Combining the post of Scottish or Welsh Secretary with a post like Defence or International Development would dramatically increase the number of issues of which the minister has to be aware, the papers they have to read and the committees they have to attend. Yet this domestic portfolio must be one that includes a substantial amount of UK-wide business; there would be conflicts of interest and other difficulties if Wales or Scotland were combined with a portfolio like Health, or Education and Skills which is mainly concerned with England. The number of posts which can be combined with being Secretary of State for Scotland or Wales is rather low, and that has an implication for the freedom of the Prime Minister in forming a cabinet if the post of Secretary of State for Scotland or Wales has to be combined with a domestic portfolio.

18. A further consequence of the present way of doing things, at both ministerial level and in terms of bureaucratic organisation, is that no-one is able to take a view of the territorial make-up of the UK as a whole and of the constitutional and policy issues that affect it. Instead, territorial issues are approached in a fragmented way that focuses on bilateral concerns and issues (Scotland-UK, Wales-UK, Northern Ireland-UK), not UK-wide matters. This fits with a more general approach to devolution as a set of ad hoc pragmatic responses, lacking any broader focus on territorial management. As matters stand, this system is only just able to cope with the demands placed on it. it will come under very serious pressure if or when these conditions change. For this reason, the Lords Constitution Committee recommended a "department of the nations and regions" to embrace relations with Scotland and Wales, the English regional agenda, and perhaps eventually Northern Ireland matters too.[35] That recommendation still appears to confer significant advantages, and will provide the capacity for greater co-ordination if territorial relations become more difficult constitutionally and politically.

DISPUTE RESOLUTION AND THE COURTS

19. As is well known, the courts have played a very minor part in the practice of devolution. A small number of cases raising "devolution issues" have been considered by the courts, but almost all have related to issues of Scottish criminal law or practice and human rights; there has been no intergovernmental litigation, and no cases before the Judicial Committee of the Privy Council where third parties have challenged the validity of devolved legislation or executive action. (Ironically, cases about Scottish criminal law and human rights have made their way to the Judicial Committee as devolution issues, but could not have reached the House of Lords as direct appeals before the Human Rights Act 1998.) This has been contrary to the expectations of many before 1999, and has been due to three factors which are not always fully acknowledged.

20. The first is the general climate of seeking to avoid disputes, discussed in paragraphs 3-4 above.

21. The second is the flexibility built into all three sets of devolution legislation, to enable issues that in other systems might be referred to the courts to be resolved without litigation if the governments involved can agree about the solution. There is a wide-range of order making powers under the Scotland Act 1998 and for orders to be made or the Secretary of State's consent to be given under the Northern Ireland Act 1998, to enable legislation that touches on reserved or non-devolved matters to be passed by the devolved institutions. Similarly, the Government of Wales Act 1998 enables further matters to be devolved by order. These powers also relate to a number of European Union functions. (This is quite separate from issues such as Westminster legislation on devolved matters made under the Sewel convention, or legislative competence orders under the Government of Wales Act 2006.) These mechanisms are used surprisingly often to adjust the boundaries of devolved competence, and resolve some of the technical difficulties that devolution creates. Similar issues have often led to litigation in other decentralised countries, because with a written constitution it is hard to avoid taking such a formal approach to resolving them. Measures for Scotland seem to crop up at Westminster three or four times a year, usually producing several instruments made under different provisions of the Scotland Act 1998. The effectiveness of this depends heavily on there being an underlying agreement about what should be done, and if it should become harder to reach that agreement the mechanism will become less useful or usable. In that circumstance, it is likely that devolution litigation, including inter-governmental litigation, will become more common.

22. There remain a number of legal issues that cannot be resolved by making appropriate orders. The nature of things means that these relate principally to Scotland or Northern Ireland (where there is legislative and executive devolution), not Wales (where devolution has hitherto been executive only). In such cases, government has needed a significant degree of legal certainty—but has not wished to bring the matter before the courts. The solution adopted has been to seek the opinion of the Law Officers in each administration. In the early days of devolution this went so far as to produce joint opinions of devolved and UK law officers; more recently, these have been law officers of only one administration, and they are not normally shared between governments. But the impact of an opinion of the Attorney General or Advocate General for Scotland (the UK Government's law officer specialising in Scots law) within the UK Government is very powerful, being binding in the absence of a contrary court judgment. This is a further case of a pragmatic response to a problem which may store up problems for the future. One problem is that the advice of the Law Officers is becoming central to the practice of government, but remains essentially untested. If the courts should at some future date take a contrary view to that taken by the Law Officers, large areas of government activity will need to be changed. The second problem is one of transparency. Law Officers' opinions (and indeed matters that are the subject of such opinions) are confidential. It is impossible for those outside government to know what has been decided by these opinions. The influence of the opinions is nonetheless far-reaching. The result is that we are governed by what is, in effect, a form of private public law. This seems at odds with the principles of government closer to the people that are supposed to underlie devolution.

INTERGOVERNMENTAL RELATIONS, AND THE RELATIONSHIP BETWEEN DEVOLVED AND RESERVED/NON-DEVOLVED MATTERS

23. The formal mechanisms of intergovernmental relations have played only a modest part to date. The British-Irish Council (BIC) has continued to meet regularly, despite suspension of devolution in Northern Ireland, in both its "plenary" and specialist formats. However, it appears to be little more than a talking shop, albeit a useful one. Ministerial attendance (from the Republic of Ireland and UK as well as devolved administrations) tends to be at a relatively low level, and unless the BIC had some actual rather than symbolic role to play it is hard to see how it could work otherwise. For the present, it belongs more to the ornamental than the functional part of the constitutional arrangements of the British Isles.

24. More interesting, and potentially more serious, is what has happened to the Joint Ministerial Committee (JMC). As noted above, this has not met in its plenary form since October 2002 despite a formal obligation that it meet annually. Its "functional" formats have largely ceased to meet too, except for its Europe format which met frequently while the EU constitution was under consideration by the Giscard Convention and Intergovernmental Conference, and which continues to meet regularly but less frequently. Meanwhile, two other sets of ministerial meetings—of agricultural ministers, meeting more or less monthly, and of finance ministers, meeting twice a year, take place outside the JMC framework. The official attitude (manifested, for example, in responses by the Lord Chancellor when questioned by the Lord Constitution Committee in November 2004) is that there is no need for the JMC to meet in the absence of any disputes to resolve.[36] Certainly the meetings that took place in 1999-2002 served little practical purpose, with the agenda padded out with "information-sharing" items of the sort that scarcely need ministerial attention, and certainly do not merit that of the Prime Minister or First Ministers. But this approach disregards the valuable symbolic role the plenary JMC plays (not one which the BIC can fulfil), the signals such a meeting sends to other parts of the UK and devolved governments and to the wider world, or its representative function for dealing with reserved/non-devolved matters (discussed further below).

25. Otherwise, intergovernmental relations proceed in a low-key, informal, bilateral way. They are largely ad hoc, driven by items of particular business on one or other government's agenda at the time rather than by any attempt to manage relations in a more systematic way. Most links are at working level. There are relatively few bilateral meetings of UK and devolved ministers, and few ones of senior officials. These only take place to resolve specific questions, not to co-ordinate policy generally or maintain contact over a wider agenda.

26. What is important is less the format of relations than that they can serve the function of managing the differences between governments that arise. So far, that has been the case—although this would appear to be due more to a combination of particularly favourable circumstances rather than because a robust and effective structure is already in place.

27. This becomes especially important when it comes to issues where the devolved territories, and devolved administrations, have particular interests in reserved or non-devolved matters. Examples include immigration and Scotland (both in relation to the treatment of asylum-seekers in Scotland, and Scotland's attempts to attract more highly-skilled immigrants to address its demographic problems), and broadcasting in both territories (in Wales, there are obvious links to language and cultural matters for which the National Assembly is responsible). The plenary JMC was supposed to be a setting for raising and dealing with such issues, but its failure to meet means that such a forum is lacking. The sort of bilateral arrangements that exist are not suitable for dealing with such matters, so they go un-aired. An attitude can be found among many in Whitehall that devolved responsibilities and interests are limited to devolved matters, and that it is inappropriate or improper for the devolved administrations to interest themselves in such matters. This is clearly contrary to the way devolution is in fact structured, and what the Memorandum of Understanding provides.

28. These issues take their sharpest form when it comes to relations with the European Union. The devolved administrations are closely involved in many aspects of the UK's relations with the European Union, through the Council of Ministers, the UK Permanent Representation to the EU, and a variety of bureaucratic methods. Until recently, the general view (expressed by both devolved and UK Government sources) was that this arrangement worked well overall. The recent leaking to the press of a report prepared for Michael Aron, EU Director in the Scottish Executive, which suggested that the sort of problems identified above in relation to domestic UK matters also occur in relation to EU-related ones. The EU presents few distinctive issues, but plenty of opportunity for problems also apparent elsewhere in the system to manifest themselves.

29. This supports the view that the problem with the present system in the UK is that the system permits an undesirable degree of variation and inconsistency in the way it deals with devolution issues. Underlying this is the fact that the UK is still largely able to do what it wishes, while the devolved institutions often require the co-operation of the UK, tacitly (by not obstructing them) if not actively. For this reason, I have dubbed the extensive freedom of the devolved administrations in practice a form of "conditional" or "permissive autonomy", as it would be possible for the UK Government to undermine it easily, even inadvertently, with very serious constitutional and political consequences.[37]

30. A further consequence of the non-meeting of the plenary JMC is that it sends a signal across UK Government that devolution is not a high priority for the Government, and that a failure to deal satisfactorily with devolution matters will not result in serious consequences. This is in keeping with the more general Whitehall view that devolution is an event, not a process, and not something that requires ongoing attention. This approach is also dangerous, as the lightly-institutionalised framework for dealing with devolution matters means that such ongoing attention is precisely what is needed.

ASYMMETRIC DEVOLUTION, AND RECOMMENDATIONS FOR THE FUTURE

31. Asymmetric devolution as such would appear to be a workable and sustainable model for maintaining a United Kingdom, and is likely to be the only way of doing so. However, it will only work if that is acknowledged to be its objective, and action is taken at all levels to ensure that this framework is maintained in good order. Allowing the present state of affairs to continue is unlikely to achieve that.

32. The first step that needs taking is to ensure that political management of the system is in hand, and not left to bureaucratic interactions or ad hoc responses to particular crises. That means that the UK Government must be fully engaged, at all levels, and that leadership must come from the very top. It is not for the UK to dictate, or try to dictate, to the devolved institutions. That would be contrary to the spirit and letter of the present arrangements, and present grave constitutional and political hazards. But it can seek to lead the United Kingdom as a whole, actively rather than inadvertently.

33. This in turn leads to a need for:

(a) More use of formal mechanisms in intergovernmental relations—for example and most notably, of the plenary Joint Ministerial Committee, which should certainly resume the practice of annual meetings.

(b) Greater transparency in relation to devolution matters—for example, more reporting to the UK Parliament about what happens in intergovernmental meetings, and more publication of relevant documents.

Memorandum of Understanding

(d) Possibly a restructuring of the UK Government, to give greater coherence to the management of devolution matters and the UK's territorial constitution. That implies a "ministry for devolution", along the lines suggested in the Lords Constitution Committee report in 2003.

34. A further factor that needs to be borne in mind is the financial structure that underpins asymmetric devolution. This is not an issue addressed in the Call for Evidence and perhaps is beyond the scope of the present inquiry, but it remains a key question. The present arrangements, relying heavily on the Barnett formula, give a high level of spending autonomy to the devolved administrations, but tie the overall package of funds available to them to that allocated by the UK Government for spending on "comparable functions" in England. Devolved finance is therefore tied into the UK system, creating a form of dependence for the devolved administrations on the UK Government and its actions. This is important in itself, but also has significant implications for other matters notably the so-called West Lothian Question and participation by MPs from Scotland (and perhaps in due course Wales and Northern Ireland) in Parliamentary business apparently relating only to England. The financial arrangements have the further impact of making the devolved administrations responsible for spending decisions, but not revenue-raising ones. They have little incentive to reduce their spending, and this also limits their accountability to their electorates. If the present system delivered a form of equity in the allocation of public spending that might be more understandable, but we do not know whether it does (if the data to show this exist, they are not in the public domain), and many academic estimates are that it does not. At some point, possibly soon, these financial issues will need to be addressed; and they cannot be looked at in isolation, as they have implications for other aspects of devolution as well. Any successor to the present arrangements will need to ensure that it delivers adequate financial resources to the devolved administrations, is financially sustainable and is also politically legitimate (which will mean that it is perceived as addressing devolved spending needs), as well as ensuring that the devolved administrations retain their political and policy-making autonomy. This is a tall order.

April 2007