of the concept of the ‘differentiated polity”2. They also borrow the term “multilevel polity” from Gamble to describe the post-devolution UK state. Gamble defines this concept in rather broad terms as “a form of MLG that stresses the variety of institutions and processes through which societies are governed” (Gamble 2000: 290). Bache and [42] Flinders offer a view of MLG that they characterise as an “organising perspective” or “framework for analysis” (Bache and Flinders 2004: 33). They contrast this perspective with what is generally called the “Westminster Model” (WM), which is the most frequently applied approach to the study of British governance. Bache and Flinders (2004) acknowledge that the WM has provided a valuable framework for academic research in the past, a useful behavioural guide for politicians, and a significant influence on public perceptions. It has also served as a meaningful normative standard. But in recent years it has shown a tendency to exaggerate the importance of the political elite, to adopt narrow and simplistic assumptions about politics, and to be too insular in its focus. Moreover, it reflects a resistance to adopting broader, foreign (non-British) approaches and methodologies and epistemological/ ontological philosophical views on political questions. MLG, on the other hand, offers an alternative approach both to the WM and the state-centric model of the intergovernmentalists.
Hogwood et al. (2000) provide a detailed nuanced view of “asymmetrical devolution,” and its impact on UK EU policy-making. They emphasise more than Bache and Flinders the uneven manner in which constitutional and institutional change deriving from devolution occurs. Unlike Bache and Flinders, Hogwood et al. believe that it is important to place their analysis of UK devolution within a wider picture of reciprocal relationships between the constitutional context and the actual patterns of participation of territorial actors in multilevel governance. Therefore, they argue that the constitutional and institutional changes stemming from devolution do not take place in an even manner throughout the polity. Hogwood et al. (2000) nevertheless conclude optimistically, like Bache and Flinders, that trends to increasing manifestations of MLG will continue and ultimately dominate in UK politics.
Bulmer at al. (2006) provide an elaboration of an intermediate position between the polar extremes of the intergovernmentalists and more dogmatic MLG theorists. They point to the need to place the relationship between these UK “devolveds” into a broader context involving interactions between European subnational authorities and EU decision-makers, a “significant field of study … [which is] generally termed multilevel governance” (Bulmer et al. 2006: 76). Within this field of study, they acknowledge the strong disagreements as to whether power is concentrated within the EU on the member state (or national) governments (as alleged by intergovernmentalist theorists), or whether it is shared by these national governments with supranational and subnational governmental authorities (as claimed by proponents of MLG theory). Bulmer et al. (2006), cite the work of Jeffery (2000) in declaring their preference for an intermediate position between these two points of view.
Gamble (2006) takes a historical and constitutional-institutional approach to devolution. He views the British state since devolution very differently from strong MLG exponents such as Bache and Finders (2004). In fact, he makes [43] no mention of this concept, and does not cite its literature, either in the article or in his bibliography. With respect to the historical context, he notes that, “Britain has never been a pure type of unitary system, in which all power is concentrated and centralised” (Gamble 2006: 21). In this way, he separates himself from the staunch exponents of the UK state as a pure manifestation of the ideal type Westminster Model. A major reason for this is the practical constraints imposed on the central government by the “informal territorial constitution” (Gamble 2006: 22). He considers these territorial arrangements to be “at best quasi-federal”, but prefers to label the UK type of quasi-federalism a “federacy”. He defines it as “a large political unity to which smaller units are federated, even though the larger unit is not itself a federation.” However, he concedes that, “Britain has been a multinational state rather than a single-nation state.” Therefore, the UK has sometimes been described as a “union state”, or “state of unions”, rather than as a unitary state. For Gamble, the political and institutional context of devolution is best understood in terms of the pattern of electoral and political competition between the major political parties.
Hopkin (2003; 2009) provides a perspective on UK devolution in terms of the party system. In the earlier paper, Hopkin (2003) had argued that there is a need for party specialists to focus more on the “territorial dimension” of electoral politics, or what he calls “spatial and geographical aspects of party competition” (Hopkin 2003: 227). In his more recent work, Hopkin (2009), analyses the relationship between political devolution/decentralisation and the organisation of political parties, using the Labour Party in the UK and the Socialist Party in Spain as his comparative case examples. “party” is a significant causal factor within the matrix and multiplicity of conditioning variables that shape the emergence of MLG in mature and advanced polities.
The British state is becoming increasingly decentralised, regionalised, fragmented and polycentric, and is no longer amenable to systematic study using more traditional frameworks of analysis. In particular, the highly centralised and executive-centred Westminster Model, despite its continued widespread application to that system by both British and non-British scholars, now seems to be largely outdated and inapplicable to changing political realities. Notwithstanding its continued status as a relatively centralised unitary state in many comparative textbooks, devolution has brought some major changes to overall patterns of UK governance. At the same time, it has spawned several important efforts among scholars to fashion novel or recast earlier analytical frameworks that are better able to highlight the rapidly changing realities of the current economically globalised and politically interdependent world. The multilevel governance approach is only one of several new analytical frameworks that are currently being devised by scholars, but it is clear from our summary above that it is among those that are most frequently applied to UK politics. In fact, somewhat paradoxically, this framework may prove to be more adaptable to [44] the British state than to many theoretically more decentralised political systems, including parliamentary and presidential federations.
Germany: Federal System Reforms (2005 to the Present)
There is an important link between the origins of the concept of multilevel governance and the German federal system. Scharpf’s (1988) work on the “joint-decision trap” in the German federal system was cited as a major theoretical reference point for the concept of multilevel governance (Jachtenfuchs 2006, cited in Stein and Turkewitsch 2008: 7). In its relatively short existence, the concept of multilevel governance has therefore had important ties to German federalism. Research and theorising about German federalism has brought us important insights about multilevel governance and intergovernmental relations in the EU. The two-way interaction process between the concepts of federalism and multilevel governance has continued. Developments in the MLG literature can now be applied in a new light to the German system. We suggest that when applied to the German federal system, multilevel governance is increasingly an important conceptual tool through which scholars can analyse recent political and institutional developments.
The German system includes the local level, counties, inter-municipal bodies, the Länder, the federal level; the EU level; as well as Type II multilevel governance structures that exist at different levels, and often span these levels. However, the emergence of new forms of governance, incorporating the supranational (EU) level and expanding to include NGOs and other actors, has been controversial from the perspective of the Länder, whose governments resent the loss of policy making autonomy and perceive it as a threat to democratic legitimacy (Jeffery 2007: 24).
A number of terms, including “cooperative federalism” and “inter-locking federalism” have been used to describe the linking of federal and Länder level policy-making (Benz and Zimmer 2011: 149). Others have used the term “unitary federalism”3 to describe the system (Moore et al 2008: 396). Scharpf suggested that the interlocking nature of the system leads to a lack of transparency in decision-making and makes it hard to tell which level of government holds the ultimate responsibility (Moore et al 2008: 396). This reasoning led Abromeit to describe Germany as a “hidden unitary state”4 (Moore et al 2008: 396). All of these terms refer to the