Introduction: Policy-making, Learning and Devolution

Abstract

This collection explores the idea that devolution in European states can provide ‘laboratories of democracy’ as states experiment in different ways to address social and economic problems and engage their citizens. While there is a substantial literature on policy diffusion and learning among US states, and on transfer and learning between countries, there is not much on learning among European federated and devolved governments. This collection fills that gap with a series of studies based primarily on experiences in Germany, Spain and the UK (including a particular focus on Northern Ireland). The introduction sets up the analytical framework, discussing the concepts of learning and transfer, the direction of transfer (for example, from the centre to the devolved territory), the mechanisms involved (from voluntary to coercive) and the degree of transfer.

Keywords:

• Policy learning
• policy transfer
• devolution

Policy-making, Learning, Policy Diffusion and Devolution

Federalism has famously been described as providing laboratories of democracy as states experiment in different ways to address social and economic problems and engage their citizens. Devolution in European states has similarly been hailed as an opportunity to try out new solutions, adapting them to local circumstances and, if they are successful, applying them elsewhere. Yet actually benefiting from the experiences of others and adopting successful innovations is an extremely difficult matter. It implies a capacity to make policy at the devolved level, and mechanisms to transfer or diffuse the innovation, whether this is by more or less compulsory conformity, by policy competition or by policy transfer and lesson drawing from experiences in different places. While there is a substantial literature on policy diffusion and learning among states, there is not much on learning among federated and devolved governments, whether within individual states or on a wider international or European basis. This collection addresses that question through a series of studies based in different systems.

Policy transfer and learning in federal and devolved systems can be particularly difficult to identify since there are many reasons why one government might adopt policies similar to those being carried out elsewhere, not all of which are to do with improving policy performance. Convergence, like divergence, might not even be the result of conscious decisions by policy-makers. In order to locate the role of policy transfer and learning, therefore, we first need to explore patterns of convergence and divergence and their drivers within multi-level systems.

Divergence and Convergence in Multi-level Systems

Federalism and devolution are mechanisms both for dividing and for sharing power. They are equally ways of dividing and sharing policy-making, creating new dynamics and posing challenges for our understanding of the policy process. Multi-level policy-making might lead to divergence within states, to convergence around negotiated solutions, or to complex patterns of differentiation, co-operation and learning.

Classical theories of federalism portrayed two levels of government each as sovereign within its own sphere, pursuing policies concurrently in ways that might or might not diverge. In so far as governments were responding to different conditions and varying electoral demands, we might expect a significant degree of divergence. These theories—never more than ideal types—have long given way to theories emphasizing the interdependence of levels, the prevalence of co-operation and the importance of intergovernmental relations to the understanding of policy processes and outcomes. The rise of the Keynesian interventionist and welfare state cut across many of the traditional divisions of competences and put a premium on joined-up government, while assumptions about national solidarity reduced the scope for local and regional variations in citizen entitlements. Theories of co-operative federalism then placed less emphasis on autonomy and more on joint policy-making, implying that there would be less scope for divergence in policy, although there might be room for experimentation and variation in the detailed application. There is a large literature on intergovernmental policy-making in federal systems, which often highlights the uniformity of policy around common definitions of problems and solutions. The existence, within states, of a common market, a common security area and a single welfare state created spill-overs and externalities such that what one government does has an impact on its neighbours, calling for yet more co-ordination and uniformity.

Since the 1990s, however, there has been another shift under two influences. On the one hand, states have sought to simplify and disentangle intergovernmental relations and clarify responsibilities, although understanding that complete separation of competences is neither possible nor desirable. On the other hand, co-operation among regional and federated governments has to some degree given way to competition in two senses. Regions compete for economic advantage in national, European and global markets, notably to attract inward investment. They also compete in policy innovation, seeking to impress their own electorates and to attract highly qualified and dynamic residents. The theme of competitive regionalism, however much it might be contested intellectually, has been taken up by states, the European Commission, international organizations and many academic observers.

Competitive regionalism could lead to policy divergence as governments seek their own solutions and adapt policy to local circumstances. Competition also puts a strain on joint policy-making and national welfare programmes as wealthy regions complain about contributing to fiscal equalization for their poorer compatriots. To the extent that they succeed, we might see more differentiation.

Yet, in the longer term, competition could also lead to convergence. In so far as governments face the same economic constraints and opportunities, we might expect them to be forced by circumstances into similar policies. One version of this is the widespread idea of a ‘race to the bottom’ as jurisdictions seeking to attract investors and wealthy residents cut taxes (where they have the power), relax standards and seek to contain social overheads. This would imply convergence of a sort and that governments keep a close watch on what their competitors are doing.

In practice, there is limited evidence in the comparative literature for a race to the bottom. It seems that just as frequent is a race to the top as governments seek to impress their own electorates and potential incomers with the quality of public services and the efficiency of their operations. If the race to the bottom implies that all governments do similar things, the race to the top may involve rather different routes. Governments observe and learn from each other although they do not necessarily imitate each others' policies completely. Ideas about best practice are diffused but taken up in different forms in different places.

Dente (1997) has traced the evolution of federal models and practices as a response to the increased complexities of government, arguing that co-ordinated federalism was appropriate in a simpler world where both problems and solutions were known and understood. Co-operative federalism was more appropriate for a world in which problems were known but not solutions, so that flexibility and the combination of diverse powers and resources were appropriate. In the modern era, when neither problems nor solutions are well understood or shared, he argues, competitive federalism may be more appropriate. Solutions may be tailored to the needs of particular areas and different solutions can be tried out and eventually be compared to see what works and how.

The extent to which devolved governments converge or diverge in policy is an interesting one, as are the reasons for this but we should remember that divergence itself is rarely a policy objective, as opposed to an outcome. Governments are, rather, interested in dealing with policy problems and managing their political environments. They may be interested in policy innovation, in so far as this provides them with the means to achieve their aims. The focus on this collection is on this specific aspect of innovation—that is policy diffusion and transfer and policy learning among governments. This implies a degree of policy autonomy, so that governments can innovate and produce some divergence, at least initially, so that there is something to learn. Divergence may then be followed by convergence as governments face external constraints or learn from each other. It also implies that policy divergences and innovations can be identified, which is not always easy, given the slippery nature of policy and the differences between intentions and actual practice and outcomes. Comparison of policies across jurisdictions is also very difficult, given differences in institutions and context.

Policy Transfer and Policy Learning

Policy transfer may take a number of forms and be more superficial or profound (see Cairney, 2012 for a review of the literature). One form is simply copying policies that seem to work somewhere else. There is a genre of literature in management—including public management—that consists of purportedly successful cases, taken out of context and presented as exemplary. We also find many examples of ‘photocopying’ in which legislation or institutions are taken from a leader, who has the time and resources to develop policy, and applied in smaller jurisdictions. This is easier within countries where there is a compatibility of legal norms and institutions than across countries although, even within states, there may be important variations, such as the Scottish legal system. The problem is that the policy might not work in the different institutional, cultural and political circumstances of the receiving jurisdiction. In other cases, there is policy adaptation, as policies are reshaped to fit local circumstances and existing policies in related fields. More profound concepts are those of policy learning, and of lesson drawing, which take into account the circumstances of different places and the relevance of the example. So Rose (1991: 4; 1993) posits the question: “Under what circumstances and to what extent would a programme now in effect elsewhere also work here?” This requires that all relevant circumstances, including similarities and differences, be taken into account. Learning also implies that someone in the receiving jurisdiction is willing and able to undertake the necessary research and adaptation. These are challenging conditions, unlikely to be found in the real world. In practice, policy-makers will at best engage in an exercise constrained by “bounded rationality” (Simon, 1976), taking into account some of the relevant contextual factors but not all of them.

The literature distinguishes between coercive transfer, in which one jurisdiction is obliged to emulate the example of others, and voluntary transfer, where governments learn freely from each other's experience (Dolowitz and Marsh, 2000: 13). While these are ideal types and clear examples of coercion are rare, they may have more resonance in multi-level systems when the centre can lay down the law. Coercive transfer may occur in systems where competences are shared and the higher level can pass framework laws, which may be more or less detailed according to the case. Systems of co-operative federalism may be more or less constraining. This has been a controversial matter in Spain, where some ‘autonomous communities’ complain that the centre has abused framework laws to regulate detail. Central governments may also be able to apply conditions to financial transfers or shared-cost programmes, also a contentious issue. Voluntary transfer implies a conscious decision by governments to find out what is happening elsewhere and to learn from it. In practice, transfer may be more or less voluntary. Financial incentives are somewhere between the coercive and the voluntary. They can be refused but refusal entails opportunity costs as revenues are lost.

Pressures arising from competitive federalism may produce transfers that are formally voluntary but are, in fact, constrained. Powerful economic reasons may push governments to match policies in neighbouring jurisdictions. Where devolved governments have fiscal autonomy, tax levels tend to harmonize because increasing taxes too much risks competitiveness, while cutting them reduces revenue. Spill-overs from one region to another may limit the scope for divergence as citizens seek public services where conditions are most favourable, putting pressure on governments unless strict residency conditions are imposed. University fees in the United Kingdom and Germany are a case in point. Environmental policies and transport infrastructure have obvious spill-over effects.

Another form of spill-over is political, as citizens and groups demand parity with neighbouring jurisdictions for individual services, rather than engaging in trade-offs among priorities within their own jurisdiction. The media play an important role here in highlighting anomalies. In the United Kingdom the cliché ‘post-code lottery’ is used to imply that citizens are being treated unfairly simply for living in different places. Behind this rhetoric, there does lie a more serious issue of the extent to which basic welfare services should be provided equally on a state-wide basis and the tensions that this may pose with decentralization. The political outcome of these debates may be a recentralization, as the state imposes national standards, or a race to the top, as governments converge on the most generous forms of provision.

Voluntary transfer implies an ambition to learn on the part of the receiving jurisdiction. This requires detailed attention to issues and conditions and a shared understanding of the problems and solutions. It also requires mechanisms to encourage and sustain learning.

Directions of Transfer

In federal and devolved systems, policies can be transferred vertically, from centre to periphery or vice versa, or horizontally among regional jurisdictions. Vertical transfer is most likely in systems of co-operative federalism, where there are regular mechanisms for exchange and working together, rather than in systems where the tasks of the two levels are distinct. Even in the absence of co-operative federalism, however, higher-level governments often have schemes to encourage innovation, or provide incentives to take up ideas pioneered at the centre. This form of learning is common in the European Union. For example, the European Commission has used the Structural Funds as a way of diffusing new ideas about regional development and gradually incorporated notions of equal opportunities and environmental sustainability into its conditions. Transfer from the periphery to the centre may take place where there is feedback on the application of policies and a capacity at the centre to use this information to reshape its own policies. It might also occur where the centre has the ability to observe good practice and to learn from it.

Horizontal transfer has been widely observed in federal systems, although this may be more or less mediated by the centre. The Canadian Social Union programme relies on voluntary co-operation under federal guidance, while the EU's Open Method of Co-ordination gives the Commission a role in helping diffuse best practice and monitoring progress. In other cases, there may be associations of governments or sectoral bodies that allow exchange without central intervention. Other forms of learning and transfer are bilateral. International learning among states is a staple of the literature (Bennett, 1991; Dolowitz and Marsh, 1996; 2000; Holzinger and Knill, 2005) and there is a growing focus on international agreements (for example, Busch and Jörgens, 2005; Alter and Meunier, 2009; Cairney et al, 2012). It can also occur among their constituent units or between these and independent states, where the relevant competences are in one case at the devolved level and in another at the centre. Independence-seeking governments (or at least those pushing for radical forms of autonomy) in the United Kingdom, Spain or Belgium may have political motivations to compare themselves with sovereign jurisdictions. Geography plays a part here, as governments often learn from their neighbours. Ideological affinity may also count, although not always in obvious ways. In recent years, the Scandinavian countries have attracted the interest both of social democrats, because of their shared traditions, and of neo-liberals who point to things like the Swedish free schools experiment to show that new forms of public management are not just an alien import from the United States.

Learning can take place at all levels of the political system. Politicians may be impressed with ideas elsewhere, especially where they seem to provide a way out of difficult dilemmas, such as reconciling choice in public services with allocative efficiency, or containing the cost of government. Their attention spans tend, however, to be limited and they do not have the incentive to look at the difficulties in transfer. Officials may have more time to study, although not much more; but learning implies the existence of policy-conscious staff with the skills to engage in analysis. There may also be learning at the operational level, about programming and delivery of services, or the use of technology.

Mechanisms

These are all ideal types and in any given system we will find areas in which governments go their own way, areas in which they co-operate and areas of competition. This implies a complex set of relationships, depending on the circumstances and the institutional configuration of individual systems. Some systems divide functions rather clearly between levels of government, allowing a large degree of autonomy in policy-making. In others, functions are shared, with the central level making the broad policy and the lower level implementing it. This is the basis of traditional German co-operative federalism. In other cases, the centre can pass framework laws, limiting the scope for discretion at the devolved level, as in Spain and Italy. The United Kingdom has an unusual asymmetrical constitution, in which Scotland and Northern Ireland have complete legislative and administrative competence in a range of matters, while England comes directly under the central government. Wales is subject to legislation made at the centre but can adapt it through secondary legislation and, under a rather complex procedure, can gain primary legislative competence in various fields.

In practice, there are elements of hierarchy and co-ordination in most systems, limiting the discretion of devolved and federated governments. There are mechanisms of intergovernmental co-operation around common problems; and there are intricate patterns of negotiation and learning at all levels.

In some countries, there are elaborate mechanisms for intergovernmental policy-making. The German joint tasks framework, partially dismantled following the federal reform, took large areas of policy-making into the negotiated intergovernmental arena. Joint Ministerial Committees exist in the United Kingdom, in Spain there are sectoral conferences and Italy has the State-Regions conference. These have, with some exceptions, not been important instruments for policy transfer and learning since they lack strong infrastructures or the capacity for sustained policy work and comparison. Nor are they well linked to the world of policy-making, academia or think tanks. Horizontal co-operation has also been encouraged between local governments and ad hoc agencies in order to develop new perspectives on policy issues within shared territories.

An increasingly popular form of voluntary learning is benchmarking or yardstick competition, under which governments agree to measure themselves on common standards, an issue covered in Arthur Benz's chapter. This also features in the EU's Open Method of Coordination (OMC) in the European Union. The Bologna process for the harmonization of educational systems follows the same principles, although it was launched before OMC was formally created. British governments have used benchmarking in a more coercive way to diffuse innovations among local governments. Its use as a learning device is limited, however, when differences in institutions, definitions and data availability make comparison of performance impossible, as has happened widely across the nations of the United Kingdom. Data can then easily be misinterpreted in the transmission (Wink, 2010). Benchmarking and yardsticks are a sensitive matter politically, since if there is to be comparison of performance with a view to learning, then it has to be carried out by an agency independent of, or at least trusted by, the governments concerned. Such comparisons are put into question when the evaluators have a commercial interest in the results or at least the research is sponsored by bodies with an interest in the results.

There has in recent years been an explosion of inter-regional co-operation and of paradiplomacy in Europe and North America (Keating, 2010). Cross-border schemes of co-operation exist at every border within the EU, while regions have widely engaged in forms of paradiplomacy and formal co-operation. These, in principle, should allow sustained communication and learning but the results have been mixed. Co-operation in infrastructure development and environmental policy has often been effective, but there has been less learning on policy matters. The attention of policy-makers is not easily engaged or sustained and there is little infrastructure devoted to learning. Barriers include language, terminology and different conceptions of the issues, as well as the danger of free-riding as some governments merely look for solutions without contributing to the common pool of knowledge (Wink, 2010). Federated and devolved governments participate in numerous European-wide and international associations. Where these have a clear sectoral focus, there may be policy learning but it is often difficult to align this with the policy agenda in the participating regions, so allowing learning to take place. The EU's Committee of the Regions could potentially be a mechanism for learning and for serious policy dialogue between regions and the European Commission but it has tended to see itself as more of a lobby and its working practices do not encourage long-term or in-depth policy work.

Political parties are often important mechanisms for policy integration, as they seek to avoid presenting different faces in different places. They may also diffuse new ideas, although the role of parties in developing policies has tended to be eclipsed in recent years.

Another channel for learning is through interest groups and wider policy communities. These may be state-wide, presenting common demands everywhere and so contributing to convergence. At the other extreme, they may be self-contained and non-communicating. In practice, we find a mixture of state-wide and regional interest groups, with sectoral policy communities more or less decentralized and articulated with the centre (Keating et al., 2008; Keating and Wilson, 2010). This permits a circulation of ideas and a filtration from one level to another and horizontally among governments. Professional associations are an important mechanism for learning as successive ideas of good practice diffuse and enter the mainstream and exchanges are facilitated by conferences and project meetings (Wink, 2010). The same is true of academic institutions and think tanks (Stone, 2000). Consultants have become an increasingly important conduit, perhaps more for policy copying than for learning. Management gurus are often able to relabel banal ideas (such as the recurrent reinvention of the idea of industrial districts) and sell them on, including sometimes to the very places where they got the ideas in the first place. We have seen examples of policy recommendations and plans copied almost word for word, with only the name of the payee changed. In some fields, epistemic communities (Haas, 1992) have emerged committed to a particular way of looking at problems and their solutions. These may evolve gradually over time or change rapidly as one paradigm replaces another. In some cases, however, professional learning may remain at the level of the specialists and not filter down to the organization as a whole; or it may encounter resistance from those whose existing perspectives would be upset.

It is not surprising that it is the larger and better-resourced governments that have been more successful in transferring ideas. This may be because they have more policy capacity and better ideas, but it may equally be because they have a higher political profile and more prestige.

Incentives to transfer are also important on both sides. It may be vital for a government engaged in innovation to export its ideas to gain external legitimation and positive feedback as these ideas are taken up elsewhere. The selling of New Public Management ideas by British administrators since the 1990s is remarkable. On the importing side, while policy-makers may like the idea of getting solutions from elsewhere, they do not always welcome having the political agenda broadened by influences from outside, when dealing with domestic pressures is difficult enough. There may then be a closing in or dismissal of external ideas as not relevant. In the United Kingdom, the centre, which doubles up as the domestic government of England, has been very resistant to importing ideas from the devolved territories, although it has not always been able to resist.

What is Transferred?

The substance of policy transfer can be elusive. In some cases, it may be mere rhetoric, as governments claim to innovate and to follow best practice, while not changing the substance. Fashionable terms and concepts may be borrowed to cover continuity or legitimize old practices. So clientelism may be relabelled as ‘partnership’ or ‘civil society’. Muddling through may be dignified as ‘governance’ or ‘negotiated order’. Rather vacuous concepts like ‘public value’ can spread across the world precisely because governments can interpret them in the way they want while appearing to be innovating. The appropriation of academic concepts by governments and international bodies, including the European Union, provides strong incentives for these semantic games. Ideas may also be borrowed and then rebadged to give them a local provenance, so that Spanish policy may become a Catalan model, serving to legitimize the devolved institutions themselves.

There may be partial transfer and adaptation as recipient governments seek to incorporate only the best of examples elsewhere, or strive to adapt imports to local conditions and political circumstances. This is often difficult, since the examples in question may be inherently embedded in wider systems, without which they cannot flourish. So, for example, free choice in selection of schools, and the wider ‘choice’ agenda, will work differently in an egalitarian society from one that is highly stratified already. Economic development policies may depend on the existence of locally-based economic actors and networks.

Policies may be transferred formally but not take root in a different environment. The Labour-led Scottish Executive imported most of the English legislation on Anti-Social Behaviour Orders (ASBOs) but local actors did not use the provisions, which remained largely a dead letter. The same thing had happened under the Conservatives in the days of the Scottish Office, when provisions for fund-holding by General Practitioners and for schools to opt out of local authority control were hardly taken up north of the border.

Less common is a reappraisal or redefinition of the nature of problems themselves. Policy learning often takes place through professional networks based on existing frames, rather than challenging these. Only rarely is there a capacity to see policy as a whole or to read complex systems and understand their logic. There are differences in the stage of the policy cycle at which learning takes place. Ideally, it should be at the beginning, as problems are appraised and alternatives considered. In practice, we know that the policy cycle is abbreviated, with policy-makers going directly from problems to ready-made solutions, so that policy learning is similarly truncated and unsuitable options chosen. The formal mechanisms for learning are not always aligned to the policy cycle, so that it is difficult for ideas learnt elsewhere to be brought into play at the right time.

An important area of learning has involved public service reform and delivery mechanisms. This is often presented as concerning merely the implementation end of the policy cycle, not affecting the broad policy options, although we know that the choice of delivery mechanism can affect the pattern of winners and losers. One of the most important variations among sub-state governments concerns the extent of adoption of the range of items under the heading New Public Management (a term still widely used although it is hardly new anymore). This is a matter on which there is wide scope for divergence, together with strong incentives to appear modern and, so governments hope, make their resources go further. Another widely diffused idea is that of the learning region, an economic development model based on flexible specialization, application of research and development, and linkages among local actors. This also has the advantage of presenting an attractive model, in which economic competitiveness and social cohesion are combined. The danger arises when it is presented as the path to a Utopia in which difficult policy choices can be avoided and the trade-offs that are the core of politics ignored. Another problem is that the model tends to be drawn from the experience of particular regions, which have carved out a position in the spatial division of labour that cannot be replicated, if only because the first regions are doing it already. Policy-makers in learner regions are then in danger of imitating a model designed (or more likely just emerging) in another place and at another time.

The Articles

The papers in this collection address these questions in diverse contexts. Arthur Benz examines yardstick competition, distinguishing it from the market competition identified in much of the public choice literature. Focusing on the case of Germany, he details the problems it faces, given the disinterest of politicians and the way that the agenda is set by parties and the media. Yardstick competition works best in areas that are not highly partisan or dominated by strong interest groups. It needs to be sustained so that learning can proceed by incremental steps. On the other hand, it cannot be treated as a purely technical process but must be seen as essentially political.

Joan Subirats and Raquel Gallego focus on the dilemma between state-wide equity and regional diversity in the case of social services in Spain. They note that in Spain, in contrast to other countries, decentralization and the welfare state were constructed simultaneously. The result is that some social policy innovations have started at the centre and others in the regions, before being diffused more widely. Competition has not provoked a race to the bottom, as often feared but rather has increased equity in standards for core services in education and health. Differentiation has occurred, on the other hand, in ancillary or complementary services or modes of service delivery. These sometimes reflect pre-devolution traditions, sometimes demographic factors, and sometimes policy choice. In social services, where there was little existing provision and few national directives, there has been more variation, together with considerable learning among regions and from European networks. Although the general public appreciates the improved services in health and education, there are fears that decentralization might lead to divergences in social rights. This reflects the perpetual dilemma between equity and territorial diversity.

Michael Keating, Paul Cairney and Eve Hepburn examine policy learning in the United Kingdom after devolution. One peculiarity of the system is its highly asymmetrical nature, with autonomous institutions in Scotland, Wales and Northern Ireland, while the UK government and parliament double up as those of England. This makes it difficult clearly to distinguish vertical transfer (from the centre to the devolved level) from horizontal transfers (between England and the other territories). There has also been an absence of formal intergovernmental institutions, although since 2007 the Joint Ministerial Committees have been revived. In practice, there has been more policy innovation in England, driven by the UK government, especially in public service delivery, with the devolved territories opting to keep and build on their own existing models. The nature of the settlement provides few opportunities for coercive transfer although funding mechanisms can be used to put pressure on the devolved territories where expenditure is reduced or charges introduced in England. There are spill-overs, particularly between England and Wales and important political spill-overs as the media highlight differences in service provision. Governments, especially the UK government, have sought to contain these with limited success. Political parties, the civil service and policy communities serve to diffuse ideas, especially from London to the periphery. Policy learning, however, has been sporadic and unsystematic.

Derek Birrell considers public sector reform in Northern Ireland, where government has been dominated by the separate Northern Ireland Civil Service and executive agencies established to replace the local government system across wide areas of service provision because of the discrimination which thrived under the old Stormont System. In recent years, there has been some learning from Whitehall, reflecting the traditional Northern Ireland pattern of copying and the importance of the civil service during the years of direct rule. There has not, however, been much learning from the, arguably more relevant, experience of Scotland and Wales or from the Republic of Ireland, in spite of the extensive reviews conducted on quangos and on local government. Birrell attributes this to the interests of the Northern Ireland political parties and to the weak policy capacity of the devolved administration. The civil service itself has proved rather conservative.

Elizabeth Meehan examines the area of human rights in Northern Ireland. Here political rather than economic considerations have been predominant, although the economic crisis is now introducing financial constraints on the policy. The institutional landscape here is complex, with learning between Northern Ireland and the United Kingdom as well as Scotland, Wales and the Republic of Ireland. The salience of human rights in Northern Ireland has meant that it is an incubator for policies and institutions, which have subsequently influenced policy at the centre, albeit often in a diluted form due to the tradition of treating Northern Ireland as an exceptional case. The proliferation of human rights legislation at the UK, devolved, all-Ireland and European levels has raised the question of the compatibility of the various regimes and institutions and the need for consistency.

In a concluding article, David Dolowitz links the case studies to theories and concepts in the wider literature on policy transfer and learning. Drawing on the insights produced by this collection of articles, Dolowitz considers our ability to compare the European multi-level experience with that of the US, which is often described rather loosely as a ‘laboratory of democracy’.

Acknowledgements

We acknowledge the support of the Economic and Social Research Council, for a Follow-On grant to Michael Keating for the conference on which this collection is based.