understand the [45] changing intergovernmental system in Germany, and in the case of Scharpf’s work, its similarities to the EU. However, these concepts are less useful for broader comparative analysis. In contrast, the concept of MLG is valuable because it allows us to analyse features shared in common by both federal and decentralised unitary systems, as well as the EU.
As reflected in the terms “unitary federalism” and “hidden unitary state,” a major concern shared by academics and politicians at the federal and Land levels is that German federalism is too centralised. There have been a number of reforms to the German federal system since its creation. However, most of these reforms did little to address concerns over the centralised nature of the federation, and even exacerbated some issues related to the “joint-decision trap” (Auel 2008). In the early years of the twenty-first century, there was a strong push for further reforms (Moore et al 2008: 396). The government set up the Federal Reform Commission in 2003 to explore ways of reducing the degree of joint-decision making in the German federal system. However, the commission’s reform process failed, and instead, the Grand Coalition government, led by Angela Merkel, introduced constitutional changes that came into effect in 2006 (Moore et al 2008: 396). Overall, these reforms decentralised powers to the Länder, giving them, for example, exclusive competences in the areas of education, including post-secondary-education (Benz and Zimmer 2011: 167; Moore et al. 2008: 398).
A number of political scientists have criticised the reforms for not going far enough5. Auel points out that, on paper, the reforms of 2006 are the most broad-ranging changes to the Basic Law since its creation in 1949, at least in terms of the actual number of amendments (Auel 2008: 427). However, Auel suggests that the reforms may be counter-productive and do not offer any “escape” from Scharpf’s “joint-decision trap” (Auel 2008: 425). Benz writes that as a result of the reforms, German governance is “neither more effective nor more democratic” and that the “federal system is in danger of losing necessary flexibility” (Benz 2008: 440). He suggests that the legislative system is still too centralised to be effective (Benz 2008: 442). Jeffery notes that one reason put forth for the fact that the reforms did not go as far as necessary is because Germany has a “unitary political culture.” The country does not have the territorial cleavages that are often present in other, more decentralised, federations. Without these cleavages, the German system leans towards uniform policies across the Länder (Jeffery 2008: 589). Scharpf agrees with the general consensus that the reforms did not go far enough (Scharpf 2008: 509). In terms of the outcome of the reforms, Scharpf does not find a move away from the jointdecision trap, in the sense that “the need for compromises between the government majority and the opposition, and thus the possibility of party political blockades, remains pretty much unchanged” (Scharpf 2008: 514).
[46] There is little discussion of Type II MLG in Germany in the federalism literature. This mirrors our findings in our earlier conference papers with respect to the literature on federalism in the United States and Latin America, and to a lesser extent, Canada. For example, in his discussion of multilevel governance in Germany and Switzerland, Braun acknowledges the existence of Type II multilevel governance, but chooses not to include an examination of this type of MLG (Braun 2011: 181, endnote 1). We suggest that political scientists studying intergovernmental relations in federal and decentralised unitary systems should increasingly take note of interactions that can be best understood through the conceptual lens of Type II MLG. We find more discussion of Type II MLG in the public policy, public administration, and local governance/urban studies literature. From a comparative and analytical perspective, an interesting form of Type II MLG that is emerging in Germany is inter-municipal cooperation (Wollmann 2010: 265). Harfst and Wurst (2011) provide another empirical example of Type II MLG in their study of environmental rehabilitation of mining regions in the former East German Länder.
In the German system, political parties, and specifically, party politics, play an important role in intergovernmental relations (Däubler and Debus 2009: 74). “[P]arties at the regional level are integrated into a coherent national party system” (Benz 2007: 432). Party politics at the Land level are intertwined with political competition and the legislative process at the federal level. This is a two-way interaction process that is “top-down from the federal to the state level and bottom-up from the state to the federal level” (Däubler and Debus 2009: 74). Political parties and interest groups have created “linkage structures” that span levels of government (Benz and Zimmer 2011: 159). These structures constitute “vertically integrated multilevel systems” in which regional associations also operate at the Land level (Benz and Zimmer 2011: 159).
Conclusions
The MLG framework fits better and accounts more accurately for changing forms of governance, political conditions and intergovernmental relations in both of these countries. There are several aspects in which an MLG approach is able to describe these two states better than a traditional federalism approach that places federal and unitary systems in separate “boxes.” Our argument also appears to follow an emerging trend in the literature, as highlighted by the work of Toonen (2010), among others. German scholars in particular are quite open to studying their own federal system, as well as the EU, through the lens of MLG. As our research suggests, MLG is also an approach that is gaining [47] ground in the study of UK devolution. Overall, we suggest that what might otherwise be seen as important institutional and constitutional differences between these two polities are less significant when they are viewed through the lens of MLG.
A typology of multilevel governance systems can be viewed to encompass different forms of federal, “quasi-federal,” and unitary systems. Multilevel governance is an overarching concept, under which variation along two spectra can be subsumed. The two axes are higher and lower levels of territorial diversity, and a range from a high level of centralisation of power in the hands of the national/federal government to a high degree of decentralisation to substate units. We concur with Gagnon’s (2011) view that a common flaw in MLG theorising is that it overlooks political cultural and socio-economic cleavages inherent in multinational federations. We therefore call for further empirical and theoretical work on how multi-national federalism and autonomism can add to our understanding of multilevel governance.
As a result of devolution, most observers now consider the UK to be a “quasi-federal” system (Bache and Flinders 2004), while federal Germany prior to 2005 was described as a “hidden unitary state” (Abromeit 1992, as cited in Moore et al 2008: 396; 406, endnote 12) that was too centralised. The MLG framework is the best way to understand the changes that have taken place as well as the calls for further changes in both of these states. In this sense, MLG is much more flexible, and thus better able to subsume the study of these changing forms of governance than are traditional comparative theoretical approaches. The MLG approach, in contrast to the Westminster Model and the various conceptions of federalism applied to Germany, such as “interlocking federalism” (Benz and Zimmer 2011: 149) and “unitary federalism” (Hesse, as cited by Moore et al 2008: 396; 406, endnote 9), is broader and more encompassing. Although the concept of MLG has been criticised for being too “overstretched” (Piattoni 2009), its elasticity does have benefits. While it could be argued that applying MLG to decentralised unitary states stretches the concept even further, we might also point out that the terms federal and unitary systems are themselves often considered to be problematic in meaning (Toonen 2010: 36).
An interesting point of comparison is the discussion of UK as a “lopsided state” (Jeffery and Wincott 2006:4) versus the constitutionally uniform powers of the Länder in Germany. There has been an acceptance of asymmetry in the UK versus an acceptance of a more “unitary” federal state in Germany, although the Länder with larger economies may be more influential in some regards (Scharpf 2008: 512). As a concept, multilevel governance does not specify a uniform or asymmetrical allocation of power to subnational authorities, giving it greater flexibility in its application. As illustrated in the typology, the concept of multilevel governance can encompass both symmetrical and asymmetrical federal and unitary systems.
[48] Another area for comparison is the devolutionary reforms in the UK and federal system reforms in Germany. Both reforms brought about changes leading to more decentralised systems, and both generated questions regarding the overall outcomes of the reforms and possibilities for