|
|||||||||||||||
 |
|||||||||||||||
  |
|
||||||||||||||
|
Devolution and Public Policy in
the United Kingdom. Seminar on 'Devolution in Practice', Institute for Public Policy Research, London, 29 October 2001 In Page Navigation Devolution and policy divergence
That devolution should generate substantive differences in public policy across the component parts of the United Kingdom may seem obvious, since that is precisely what it is meant to do. Yet in the protracted debates over more than thirty years, relatively little systematic attention has been given to this or to its implications, while since 1999 the media have got into the habit of describing every distinct measure taken for Scotland or Wales as an ‘anomaly’. Comparative experience shows that decentralization and federalism do provide scope for policy differentiation, but that there are also powerful pressures to convergence. This paper looks at the scope for policy divergence under the UK devolution settlement. It is organized in six sections. This introduction sets the scene. Then I examine structural factors, notably the formal devolution settlement and intergovernmental relations; political factors including the party balance and the emergence of territorial policy communities; the context of the single market and welfare state; and European influences. Finally I look at the dynamics of policy making in complex systems and the prospects for change. The first point to make in this discussion is that policy making in modern government is typically an incremental process in which the weight of existing commitments limits the scope for innovation. The difficulties of the present British government in actually spending the money it has recently committed to enhance public services is a recent testament to this. Devolution in the United Kingdom, in contrast for example to the Spanish experience, builds on an existing and well established system of administrative devolution in which each of the component territories had a distinct way of making or adapting policy and delivering services. Against some of my colleagues, I never believed that the Scottish Office and its associated agencies had a substantial policy autonomy, but there were matters on which it was allowed to go its own way within the limits of overall UK policy. The Welsh Office, on the other hand, was more closely tied into Whitehall networks, sponsored very little legislation and took fewer initiatives. Government in Northern Ireland was an inheritance from the Stormont regime, which had extensive devolved powers although tending to shadow British welfare state provision. How much policy divergence occurs under devolution must therefore be measured not against some abstract model of uniformity, but against the pattern of convergence and divergence existing in the past. Structural FactorsAn obvious point to start the analysis is the formal division of powers under the respective devolution Acts. Both the Scottish Parliament and the Northern Ireland Assembly have a general competence over all matters not specifically kept at Westminster, although these are expressed rather differently. Scotland faces just a single list of reserved matters, while in Northern Ireland there is a distinction between reserved matters, which may be devolved in the future, and excepted matters, which are kept permanently for Westminster. Both bodies have primary legislative competence over non-reserved matters, but also have executive responsibility in matters where the primary responsibility remains with the centre. An examination of the two Acts shows some common elements but also the traces of past provisions such as the Stormont system, or the Scots law tradition. There is also a tendency for the centre to be concerned in Scotland with keeping matters related to the common UK market, while in Northern Ireland the prime concern is with security. The National Assembly for Wales, on the other hand, has only secondary legislative and executive responsibility for a list of powers devolved from Westminster and must get parliamentary approval for any changes in primary laws. Reserved and devolved powers are listed in Appendix 1. Modern government does not work on the basis of watertight divisions of competences and there is a high degree of functional dependence between devolved and reserved questions. This is particularly noticeable in welfare state matters where, generally speaking, Westminster has reserved cash payments while personal social services, training and housing are devolved. This has already caused some problems in the treatment of housing benefit, and in the Scottish proposals for financing long-term care for the elderly when the Scottish Executive has tried to claim back from the Department of Work and Pensions funds that would be saved by its new scheme. There are large interdependencies between the unemployment benefits scheme (reserved) and labour market policy (partly devolved) which have been highlighted by the New Deal programme. These might produce policy distortions or perverse incentives – a cynic might argue that there is no incentive for the devolved governments to spend money getting people into work since that would serve to reduce total welfare payments into their territories.[1] There are also functional overlaps in energy matters, with Whitehall responsible for policy but the Scottish Executive retaining power to approve the construction of new power stations on planning and environmental grounds. Given the present UK government’s apparent intention to resume the construction of nuclear power stations, there will be pressures within the Scottish Parliament, if not the Executive, to resist this. Overlaps have also arisen in the treatment of asylum seekers. While the division of powers intended by the Acts is reasonably clear, there is also scope for devolved administrations to expand the application of their competences by invoking the aims of policy rather than the means. Already the Scottish Executive has launched a strategy on broad band telecommunications although telecommunications is clearly a reserved matter. Before devolution, the territorial offices were tied into the Whitehall policy network through a dense network of ministerial and official committees and regular working contacts among ministers and civil servants. Papers were exchanged, each level knew what the other was doing and there was care not to get too far out of line. Yet while the Whitehall departments generally took the lead in joint policy making there was no formal hierarchy and indeed no UK department at all in large areas of policy including education, housing, local government and social services. The devolution settlement builds on this administrative heritage with the effect that, over large areas of public policy, there is now no ‘centre’ at all. This contrasts with Germany, where the division of powers (legislation to the Bund, execution to the Länder) means that both levels have parallel ministries, or Spain, where the central ministries have shown little inclination to retreat in face of the autonomous administrations. There was an attempt in Italy to solve the problem by abolishing ministries but these managed to pop up again in new forms.[2] Even Canada and the United States have federal departments for health and the US (but not Canada) has one for education. This, combined with the end of the old interministerial committees and contacts,[3] could have powerful centrifugal effects. It is partly to combat this tendency that the Joint Ministerial Committees have been set up but, while these appear to be a means of exchanging ideas and innovations, they are a far cry from the interdepartmental committees that prepared policies in the pre-devolution age and there would obviously be great political sensitivity about their being seen as a top-down measure to impose policy from the centre. Another mechanism for co-ordination are the Concordats, non-statutory frameworks for agreement on areas of functional interdependence (listed in Appendix 2). These provide for policy variation to be limited and for a negotiation of differences, but it is clear that London will have the last word in the event of disagreement. It is by no means clear how they would fare were different political parties in control at the two levels. There are also legal mechanisms for pulling devolved administrations and assemblies into line should they exceed their powers but these, too, are untested given the consonance of political alignments between London and Edinburgh and Cardiff. This makes a big contrast with Spain, where political antagonisms produced a spate of court references in the 1980s, which served to set the bounds for devolved government. A more serious constraint in practice may be the limited policy capacity of the devolved administrations. This reflects their historic inheritance so that Scotland has quite a large capacity for research and development in education and housing. It also has a disproportionate share of UK agricultural research, previously seen as contributing to a common effort and which may now come into question. On health policy, on the other hand, the devolved administrations may have to select areas for attention and then look to experience elsewhere. While this could involve a certain amount of horizontal learning among the devolved bodies, or from Europe, Whitehall is likely to be the main reference here. There have been some changes in policy capacity since devolution, with the appointment of many more ministers with their own ideas and the recruitment of specialist advisors. Innovative groupings of functions, such as the combination of industry with further and higher education in Scotland, or the replacement of ministries of agriculture with departments of rural affairs, are another force for innovation. Devolution was intended not only to transfer powers downwards territorially but to foster a new type of politics, with more participation on the part of backbenchers and civil society generally. Experience of the Scottish Parliament shows that the committees, while they could never function like their counterparts in the US Congress,[4] are more important than those at Westminster. The Parliament is certainly more important than most of its counterparts in other federal or devolved systems, no doubt reflecting the salience (if not the power) of Parliament in British public life. It sustains a debate about public policy that was scarcely possible before devolution, and draws in many more actors to the policy process. In areas outside party contention, like the future of the enterprise support network, committee input has been mentioned as a significant contribution to policy. This reinforces a distinct pattern of political life in Scotland and encourages new policy communities. Politics and Policy CommunitiesParty politics is obviously a factor in determining the degree of convergence or divergence of policy but so far we have limited evidence from which to judge the practical importance of this. Scotland and Wales are controlled by Labour-Liberal Democratic coalitions not far removed politically from the government in Whitehall, although coalition politics has already produced two highly salient shifts in Scotland, on student fees and long-term care for the elderly. In the event of differences in control at the two levels of government, there would obviously be more pressure for divergence, which might put strain on the system of accommodation put in place at a time of consistent political alignments. I believe that the system is perfectly capable of accommodating this, but new rules of the game would have to be learnt – the courts would certainly enter into the policy process. A radical difference in ideological stance between London and the territories on basis welfare state and common market issues, however, would be difficult to sustain, for reasons discussed below (Context). Even in the absence of partisan divisions, however, there could be differences, as we see from the hints that the Scottish Executive might not be as keen on private provision of public services as the government in London. The two main British parties have traditionally been highly centralist and have found it very difficult to adapt to the political realities of devolution, as Labour’s troubles in London and Wales have shown. Scottish Conservatives are torn in a battle between those who want to rebuild themselves as a more authentically Scottish expression of centre-right opinion and those who want to cleave to the anti-European and right-wing views prevailing down south. These intra-party differences could be important in the case of diverging pressures coming from territorial policy communities, that is the constellation of interest groups forming around the devolved institutions. Scotland has always had its own interest groups, which fall into three types. There are purely Scottish groups, Scottish groups affiliated to British or UK federations, and Scottish branches of British groups. The balance among these owes a lot to historic factors, for example the refusal of the British Trades Union Congress to accept affiliations from trades councils, which led to the establishment of the Scottish Trades Union Congress.[5] Scottish groups in the fields of law and education have never been part of wider UK bodies, since these areas have been treated separately since the union of 1707. In other cases, the growth of Scottish groups is a response to administrative devolution and the need for an interlocutor with the Scottish Office. Wales has many fewer distinctive interest groups, although there has been some growth since the 1970s. Northern Ireland is more complex again, with some groups operating on a Northern Ireland basis, some on a UK basis and some on an all-Ireland basis. In Scotland, devolution has led a more formal division of responsibilities between Scottish groups and their UK counterparts. The Confederation of British Industry, for example, has given autonomy to its Scottish branch in devolved matters and increased its research capacity. Some bodies that operated mainly at the local level have established an all-Scotland presence. Many groups have sought to strengthen their policy capacity by hiring researchers, policy consultants and parliamentary liaison officers. They have also had to reorient their networks and lobbying to take account of the new three-level politics. There are emerging patterns, which we can analyse by reference to three sets of institutional factors. First is the location of the group and its members, whether within Scotland, at the UK level or international. In the case of business firms and sectors, this would refer to the ownership of the firms and the location of their production. In the case of trades unions or social organizations, this would refer to the location of their members. Second is the field of operation of the group or, in the case of firms, the market. Third is the level at which the activity is regulated. So the main teachers’ union is based in Scotland, organizes only Scottish teachers and the sector is regulated entirely at the Scottish level. The main focus for policy attention is therefore the Scottish Executive and Parliament. Big firms doing business in Scotland are mostly externally owned, are regulated mainly at UK and European levels and trade in UK and global markets. We find, accordingly, that the main focus of attention for their representative groups are UK government departments. They do, however, have to pay attention to the Scottish level more than in the past, because some matters affecting them, such as transport or planning policy, are devolved, and they do have to play in the Scottish arena to sustain legitimacy in the face of increased political exposure. Small firms tend to be locally owned, trade locally and are more dependent on the public goods produced by devolved governments, including enterprise support services and infrastructure. They are also less well integrated into the UK-wide networks of influence and so are more strongly oriented to the Scottish level. Other groups are subject to conflicting pressures. The Scotch Whisky industry operates locally, but is externally owned, depends heavily on export markets and is regulated at UK and, to some extent, European levels. It therefore tends to look still to London, although there is a Scottish dimension. Agriculture, an industry owned and operating locally, is almost entirely devolved but at the same time highly Europeanized. This has led to the maintenance of strong links to the UK level at the nexus of European and Scottish networks. Important financial and banking interests are headquartered in Scotland, in some cases Scottish owned, but operate in global markets and are regulated at UK and European levels. This leads them, too, to tie into UK policy networks, but without neglecting the Scottish dimension. Business groups are fairly unanimous in opposing anything that could be seen as erecting economic barriers within the United Kingdom, although they have become more supportive of local and regional business development initiatives. This makes them a force for policy convergence. Individual rade unions are, with a few exceptions, organized on a UK basis. Labour regulation is a largely reserved matter and there is a tendency now to leave this to the TUC and UK-level of the unions. Training policy and responsibility for local development and responses to industrial closures, however, is devolved, so the unions are drawn into the Scottish networks. Trade unions, while generally supportive of devolution up to the 1920s and since the 1970s, oppose any divergence in labour regulation across the United Kingdom, although they are very supportive of local development initiatives. They are also concerned with a range of social issues that draw them into Scottish networks. Public sector trade unions and associations are sometimes organized on an UK basis, like Unison or the British Medical Association, while others, like the Educational Institute of Scotland, are purely Scottish. One might expect greater pressures for uniformity in public sector pay and conditions where the negotiators are UK-based, as with doctors, since members would be unlikely to accept radically different conditions on either side of the border, and there is some evidence already of this happening. Of course, even where negotiators and negotiations are quite separate, as in teaching, people will make comparisons. Social lobby groups and voluntary service organizations tend to be less well resourced than the others and operate usually at the Scottish level, respecting the division between reserved and devolved powers. This means that they can enter into policy networks in matters like housing, which are almost entirely devolved. They tend to be strong supporters of devolution since it has increased the points of access into the political system and given them more sympathetic interlocutors in the Parliament and its committees, and they might well prove an important agent of policy innovation. While Scotland has long had its own interest groups and policy communities, there has been an important change in style and operation since devolution. The main role for the community before 1999 was to lobby, whether for more resources or for some bending of UK programmes to take account of Scottish interests. This allowed Scots to sustain the idea of being in a common cause, bound by shared values and a broad social consensus. Indeed this experience probably underlay some of the more naïve ideas floating around before 1999 that the Scottish Parliament could operate on the basis of consensus, banishing partisanship, conflict and lobbying. Now the various groups must confront each other in an open political system, where they are vying for attention and competing for limited resources. They are gradually adapting to the new balance of consensus, conflict and compromise, although the failure to appreciate that disagreement is an essential part of a live democracy may be partly responsible for the tendency of sections of the media to describe every political conflict as a crisis of the system. Many groups have found the transition from lobbyist to participant in a policy process difficult. Several of them commented to us that while in the past they would just girn, secure in the knowledge that nobody would respond, now they are asked what they would do instead. The mass of consultation conducted by the Scottish Executive and Parliament is welcome in itself but has put massive pressure on groups, many of which lack the resources to respond. There has been some growth of think tanks, almost all based in Edinburgh, but there is still a shortage of policy making capacity in Scottish civil society. The devolution settlement provides for block grants for the devolved assemblies, calculated on a share of increases or decreases in corresponding English programmes (the Barnett formula). While these sums are freely disposable in Scotland, Wales and Northern Ireland, groups are likely to look at the corresponding totals to see if they have got their ‘share’. UK-wide interest groups may be most aware of the implications of English spending decisions and able to make comparisons, providing another pressure for convergence. This may depend on the amount of competition for resources and on the transparency of the public accounts. In the first phase of devolution there was plenty of money to go around following the public expenditure review, the main difficulty being in spending it all; and the government presented the accounts in such a way as to make comparisons of increases in individual functions very difficult. In future years, there will be more competition for resources and for greater clarity; the 2001 public expenditure statement for Scotland already makes for easier reading. ContextThe context for UK devolution is a modern, integrated welfare state and there are questions about how much divergence is possible in practice in these conditions. Three important sets of constraints arise from the existence of a UK common security area, a common market and the welfare state. While criminal justice is largely devolved, the lack of barriers among the parts of the UK and the need for common standards leads to a lot of co-operation and concern not to create conditions that could be exploited by wrongdoers. Hence there is a great deal of co-ordination in matters of criminal law. Concern to maintain the common market is responsible for many of the limitations on devolved bodies, especially in Scotland and Wales. There are no powers to tax business, except through the Unified Business Rate and development grants are kept in line through the concordats to prevent market distortions or bidding wars. Competition among the territories for mobile investment will also be a pressure for uniformity, since all will want to maintain a friendly business environment. Of course, this does not imply uniform policies, since there are many different models of local and regional development, but it will impose limits on the range of policies adopted and may produce convergence in regulatory, planning and environmental policies. The common labour market similarly imposes limits on the ability of devolved administrations to establish their own model of industrial relations, even within the limits of their competences. While there is a lot of support for devolving active labour market policies, neither side of industry has ever shown support for devolving labour law or regulation. The common welfare state also poses practical and political limitations. Fears that differences in welfare provision will trigger migration have found only limited support in the experiences of other countries, since other factors such as employment and family networks have proved more powerful, but there may be some distortions. More important perhaps are the shared assumptions of the post-war welfare settlement for broadly equivalent basic services free at the point of use. ‘Social citizenship’ in the United Kingdom has been linked to a British identity and indeed many have credited the welfare state is forging a sense of British, if not UK, unity during the twentieth century. Scotland’s decisions on student fees (‘Cubie’) and long-term care for the elderly (‘Sutherland’) have tested these limits, provoking politicians and the media into discovering all manner of ‘anomalies’. It is here that the financial settlement and the Barnett formula come into play. The block grants to the devolved administrations are freely disposable, that much is clear. The devolved powers in the area of social and welfare policies are also reasonably clear, so that there is no doubt about Scotland’s right to proceed with Cubie and Sutherland. Yet if finance raised on a common tax base is used to fund substantively different entitlements in the various parts of the United Kingdom, this is going to cause major political opposition. This is not to say that convergence will always be to the English line; it may that pressure within England will lead to the adoption of Scottish or Welsh innovations there. Even more problems could arise were innovations pioneered in England, since resources are immediately adjusted to take account of this, with the knock-on effects for the devolved territories merely a consequence. So if a future government were extensively to privatize the welfare state, promoting private schools and medicine and charging for services, the devolved administrations would automatically lose their corresponding public funding and have to respond. Alternatively, they would seek to reopen the settlement and gain their own powers of taxation. So marginal differences in welfare state provision are likely under devolution, as we have already seen but radical changes would destabilize the whole settlement. EuropeMany devolved functions are subject to European Union law, notably in agriculture, fisheries, environment and industrial policy. This has become a major issue of contention in Germany and Belgium, where regions complain that it entails a loss of power not only to Europe but to central government, since it is the state as a whole that determines European policy. The response in Germany and Belgium has been to use the clause in the Treaty on European Union allowing the Länder, regions and communities to participate in the Council of Ministers and even to represent the member state where regional matters are involved. Yet while this may enhance regional influence, it does not permit policy divergence, since they must represent the state as a whole and agree a common line. There is provision in the United Kingdom for devolved governments to participate in the Council of Ministers but again they must support a common line and the relevant concordats make it clear that Whitehall will have the last word on the line to be followed. The UK government is also responsible for the implementation of EU policies and has over-ride powers in case devolved administrations do not fulfil their obligations. The Structural Funds, often regarded as a means of regional emancipation, if fact have the opposite effect. Since the UK does not recognize additionality at the territorial level, the effect of structural fund designation is to earmark a part of the block grant deemed to represent the European contribution and oblige the devolved administrations to allocate another tranche as ‘matching funds’. These moneys are then ring fenced and unavailable for allocation to other priorities.[6] Relations with Europe may change as a result of the debate on a European constitution, the White Paper on Governance, and the initiatives of the ‘constitutional regions’. It has been suggested that regions with broad competences could contract more directly with the Commission for the implementation of EU policy, allowing for a more differentiated application and combinations of instruments. There has also been pressure to allow regions to appeal to the European Court of Justice invoking the principle of subsidiarity where they feel that their competences have been invaded, but all this is for the future. The devolution settlement in the United Kingdom is at an early stage and the actors are still learning the new rules of the game. Since the administrations in Scotland and Wales are broadly in line politically with that in Whitehall and the power-sharing executive in Northern Ireland has not really begun to function, there have been no causes for major policy divergence. Nor, in contrast to countries like Spain or Canada, has the division of powers been tested in the courts to produce a body of case law. In fact there has not to date been a single challenge by central government to the competences of the devolved bodies. We are therefore limited in our ability to speculate about the future, but we can frame some general typologies of policy, on the basis of the history of administrative devolution before 1999, of developments since then, and of experience overseas. Five types of policy might be distinguished. First, there are non-comparable policies, where an issue exists only in certain parts of the United Kingdom. So one of the earliest initiatives of the Scottish Parliament was to address the historic issue of land reform, which has no counterpart in the other three countries. Welsh language policy and many issues in community relations are similarly unique to Wales and Northern Ireland respectively. Second, there is policy autonomy, where the Whitehall and the devolved bodies make their own policy according to local needs and preferences. This is likely to occur in wide areas of education and social services, and in organizational matters including the structure of local government in Scotland. Third, there are concurrent policies, where the territorial administrations pursue the same broad policies because of pressure from UK groups, similarity of conditions, European regulation or whatever. This may well be the case across much of health policy, where the demands are similar, interest groups operate across borders and make comparisons, and technology is an important factor in policy innovation. It is also likely to be case in higher education, where universities have historically been suspicious of devolution as something that might narrow their horizons; hence the organization of the Research Assessment Exercise and the Quality Assurance Agency on a UK basis, sponsored jointly by the four countries. Third there is policy uniformity, where practical considerations or external pressures make for a single line. Legislation on establishing the International Criminal Court, for example, has been brought forward separately in Scotland, although the effect is identical to that in England. Sometimes such matters have been handled by ‘Sewell resolutions’ in which the Scottish Parliament votes to agree to Westminster legislating in devolved matters. Indeed, up to spring 2000, there had been more Sewell resolutions than separate items of Scottish legislation. This, however, is politically sensitive and it is likely that, parliamentary time permitting, the Scottish Parliament will choose to make its own laws. Fifth, there is policy competition, a form of policy autonomy but which is worth highlighting separately. In this case, the four countries make policy for themselves in a way to demonstrate their credentials in innovation and imagination. Such initiatives make the news and are taken up by interest groups in the other jurisdictions and the effect is a diffusion of innovation and perhaps a reconvergence around the new practice. This type of convergence might be considered one of greatest contributions of devolution, in encouraging experimentation and learning from best practice but it could also produce a focus on what the Canadians have called ‘boutique policies’, highly visible initiatives to gain political impact in competition with other levels of government, but not necessarily making the best use of resources. Before 1999 the territories were distinctive more in how they did things than in what they did. After devolution, they will continue to work within the parameters of the same welfare settlement but have considerably greater scope in choosing how to adapt to it. Devolution is a learning process and, as they learn, we are likely to see ever more policy innovation. AcknowledgementThis paper draws on research under way in collaboration with Sean Loughlin under two programmes, the Leverhulme-funded programme on Region and Nation in the United Kingdom directed at UCL by Robert Hazell; and the ESRC programme on devolution. The emphasis in the paper on Scotland reflects my own work there, with the assistance of Linda Stevenson. Appendix 1.
|
|
||||
|
||||