MPIfG Working Paper 05/8, September 2005


No Exit from the Joint Decision Trap? Can German Federalism Reform Itself?


Fritz W. Scharpf , Max Planck Institute for the Study of Societies






The unique institutions that make up Germany's "unitary federal state," long considered part of the country's post-war success story, are now generally perceived as a "joint-decision trap" impeding effective policy responses to new economic and demographic challenges at both levels of government. Nevertheless, a high-powered bicameral Commission set up in the fall of 2003 failed to reach agreement on constitutional reforms. The paper analyzes the misguided procedural and substantive choices that led to this failure, and it discusses the possibility of asymmetric constitutional solutions that might enhance the capacity for autonomous action at both levels.





Die im internationalen Vergleich einmaligen Institutionen des "unitarischen Bundesstaats" erschienen lange as Element der deutschen Erfolgsgeschichte in der Nachkriegszeit. Heute erscheinen sie jedoch als "Politikverflechtungsfalle," welche die Politik in Bund und Ländern an der Bewältigung der neuen ökonomischen und demographischen Herausforderungen hindert. Trotzdem gelang es einer im Herbst 2003 eingesetzten und politisch potent besetzten gemeinsamen Kommission von Bundestag und Bundesrat nicht, sich auf eine Verfassungsreform zu einigen. Der Text analysiert die prozeduralen und sachlichen Fehlentscheidungen, die das Scheitern erklären, und er erörtert die Möglichkeit asymmetrischer Lösungen, die den Spielraum für autonomes politisches Handeln auf beiden staatlichen Ebenen erweitern könnten.





1 Why the need for reform?
  1.1   The evolution of the joint-decision system
  1.2   The joint-decision trap


1.3   A window of opportunity?


The Commission for the Modernization of the Federal Constitution
3 What was agreed and why did it fail?


3.1   The Bundesrat veto
  3.2   The Constitutional Court changed the agenda
  3.3   Conflicts among the Länder
  3.4   The impasse over education: All or nothing at all?


Future options


Toward an asymmetric German federalism?






In the afternoon of December 17, 2003, most of the assembled members of the high-powered "Commission on the Modernization of the Federal Order" in Germany were surprised and disappointed to learn that all their efforts during the past year had come to naught. Even though there had been agreement on about 80 percent of the issues discussed, so they were told by the Commission's joint chairs, conflict over the remaining points, and in particular over the allocation of competencies in the field of education, had caused the prime ministers of the Länder to refuse adoption of the compromises that had in fact been reached. Predictably, public reactions were more strident, with comments ranging from ridicule to outrage over the inability of the German "political class" to get its act together and the country out of a quagmire of political deadlock and economic stagnation. For most of these commentators, it appeared self-evident that Germany's recent ills were rooted in its federal structures and practices - whose successful modernization was to be the "mother" of all the economic and social reforms the country needed but had failed to enact.


While such hyperbole would neither account for the earlier success of the "German Model" nor for the fact that economic and social reforms are not glaringly more successful in unitary and highly centralized France, the critics do have a point: While German federalism contributed to the political stability and economic success of West Germany during the post-war decades, it has in fact impeded effective political responses to the cumulative challenges of German unification, European enlargement, economic globalization and demographic change. In order to show why and how this should be the case, I will begin by describing the evolution of problematic features in German federalism, then go on to characterize the mandate, composition and modus operandi of the Commission, followed by an analysis of its almost-achieved solutions and the reasons for its ultimate failure. I will conclude with an assessment of the political and theoretical implications of this failure.



1   Why the need for reform?


The political institutions of post war (West) Germany have been characterized as a "unitary federal state" (Hesse 1962) - a paradoxical description which reflects the fundamental dilemma faced by the architects of the 1949 Basic Law: having to create federal structures that had to operate and find acceptance in a unitary political culture.


Bismarck Germany, it is true, started out as a monarchical federal state where limited powers at the national level were exercised jointly by the Chancellor (who was at the same time Prime Minister of Prussia) and a Bundesrat composed of the executive heads of the kingdoms, dukedoms, principalities and free cities that constituted the empire. But as the need for, and the political salience of, national legislation increased, ever more virulent demands for the democratization of German political institutions naturally focused on the elected Reichstag, where liberal, Christian and socialist parties came to outnumber the conservative forces protected by the Bismarck architecture. Thus, when monarchy was replaced by parliamentary democracy after the revolution of 1919, the change was also associated with a massive centralization of political power: Legislation was concentrated at the national level; the constituent Länder were reduced to the status of self-governing administrative provinces; and the functions of the Bundesrat - still composed of the executives of the Länder - in national legislation were reduced to a suspensive veto. Even these remnants of federal institutions were then abolished by the Nazi regime which, in terms of its formal institutions, created a fully centralized unitary state, supported by pervasive (and, so it seems, highly popular) appeals to national unity and Volksolidarity.


Moreover, after the end of the war, with cities and industrial regions in ruins and millions of refugees on the move, pre-war regional identities were scrambled while demands for solidaristic burden sharing increased dramatically. Hence, when the allies began their bottom-up reconstruction of German statehood, the new Länder - whose boundaries, with few exceptions, were redrawn to the convenience of the four occupying powers and with little regard for historical legacies - were perceived more as fiduciaries of the defunct German state than as autonomously legitimate political units. Instead, Land governments and all political parties, except for the Communists, were united in their commitment to national solidarity and in their protest against the division of Germany along the boundaries of occupation zones. Thus uniform national laws, if not revoked by the Allies, remained in force and were rarely replaced by Land legislation; and where new initiatives were required, burden-sharing and coordination among Länder executives were taken for granted. In short, the demands and expectations of national unity were probably strengthened, rather than weakened, by the collapse of Hitler's unitary and centralized state (Scharpf 1989).


However, Land governments were in place three years before the Federal Republic, and they were institutionally consolidated when the Western Allies finally gave orders to create a constitution for the three Western zones of occupation. Moreover, the Allied mandate not only insisted on a federal solution that would prevent the re-emergence of a strong centralized German state, but also made sure that the Länder would dominate the design and adoption of the constitution. Realizing, however, that they could not hope to profit from legislative autonomy in a unitary political climate, the Länder accepted a preponderance of federal legislative powers, but fought hard to preserve their de-facto monopoly in the implementation of federal legislation. Furthermore, and even more importantly, they succeeded in recreating Bismarck's Bundesrat model - i.e., a second legislative chamber with significant veto powers that was to be constituted by Länder governments, rather than by directly elected senators or representatives of Land parliaments.



1.1   The evolution of the joint-decision system


Once this institutional structure was in place, its path-dependent evolution seemed preordained (Lehmbruch 2002a; 2002b): Political demands were generally addressed to the national level, and if their satisfaction depended on expanded legislative competencies, Länder governments were generally willing to accommodate constitutional amendments - provided their administrative monopoly was respected and their veto in the Bundesrat extended.


In the early post-war period, the Adenauer government could generally rely on its superior resource base and its party-political preponderance to have its way with "coalitions of the needy and the willing" at the Länder level. But when the Constitutional Court finally intervened against the most glaringly biased national project,[1] opposition against discretionary and selective federal programs also gained ground in the political sphere. Furthermore, "Keynesian" and "technocratic" ideas of macro-economic management and comprehensive policy planning were on the rise during the 1960s, and when the Social Democrats entered the Grand Coalition in 1966, the time was ripe for a basic revision of the fiscal constitution that was also meant to regularize the proliferation of discretionary federal grants in a systematic structure of "cooperative federalism" (Troeger Kommission 1966; Schönhoven 2004).


At the same time, given the considerable diversity of the West German Länder in size, problem loads, economic prosperity and fiscal and administrative capacity, political demands and a constitutional mandate to create "equal living conditions" were accommodated not only by a fiscal regime of shared income and turnover taxes and ever more perfectionist vertical and horizontal "fiscal equalization" transfers, but also by the institutionalization of an elaborate system of jointly financed and jointly planned programs (Gemeinschaftsaufgaben and Finanzhilfen) in politically salient policy areas, such as regional economic policy, higher education, basic and applied research, social housing, urban renewal, and so on. Federal grants to states or provinces are of course quite common in comparative federalism. What is unusual is that, by the 1970s, the Länder obtained constitutional guarantees ensuring that such grants could no longer rest on bilateral agreements between the federal government and individual Länder, but had to be based on general legislation requiring agreement of the Bundesrat or on contractual arrangements with the consent of all Land governments (Scharpf et al. 1976).


While the "planning euphoria" did not survive the shocks of the first oil-price crisis, the constitutionalization of the joint-decision system has become a permanent feature of German federalism. What has also survived was the firm resolve of Länder governments to never again allow the federal government to pick and choose its allies and to play off some of them against the others, as Adenauer had done. By implication, this commitment to maintain a common front at almost any cost also required that the Länder must resolve conflicts among themselves before they are able to confront the federal government. Given the considerable diversity of interests between very big or very small Länder, with prospering or declining economies, and governments of different party-political orientations, that was never easy. When it was not a question of maximizing common demands against the federal government, the success of the cartel strategy depended on conflict-avoiding practices which amounted to a combination of egalitarian distribution rules and a disproportional influence of status-quo interests (Scharpf 1988).


In combination, the legislative veto of the Bundesrat, the regime of tax sharing and fiscal equalization, and the proliferation of jointly financed programs constitute the system of "joint decision-making" (Politikverflechtung) that has become the characteristic of German federalism. It implies that practically all politically salient policy initiatives (except foreign and defense policy) need to be based on broad consensus or even on unanimous agreement between the governing majority at the national level and the governments of originally eleven and now sixteen Länder.


Clearly, however, post-war Germany could not have been the success story that it was if political and administrative actors had not been able to develop institutional arrangements facilitating agreement and compromise, rather than persistent deadlock (Benz 1985; 1994). In practice, the conservative bias of the joint-decision system is sometimes overcome by federal side payments to the most reluctant opponents of its initiatives. Of greater importance is the fact that Länder governments are in charge of the implementation of practically all state functions and thus have an institutional self interest in shaping, rather than blocking, federal legislation. As a result, the machinery of dozens or even hundreds of specialized coordination committees at the bureaucratic level, regular conferences among ministers, and the diplomatic rituals of top-level meetings of the heads of governments have generally been able to resolve policy conflicts among the Länder and between the federal government and the Länder through compromises which, even if sub-optimal when viewed from a problem-solving perspective (Scharpf et al. 1976; Scharpf 1988), would still ensure the "governability" of the country.



1.2   The joint-decision trap


If that were all, German federalism could be characterized as a "compulsory negotiating system" (Scharpf 1997, ch. 6), or as a multiple-veto system (Tsebelis 2002) in which policy choices are shaped by the institutional self-interests and policy preferences of governments at both levels. In that regard, German institutions are similar to those of the European Union or of Switzerland. In contrast to the European Union, however, where the expanding role of the Parliament is not yet connected to electoral competition among cohesive political parties, party politics plays a significant role in Germany and Switzerland.[2] But whereas partisan conflict in Switzerland is moderated by the practice of super-majoritarian coalition governments, it is quite misleading to consider Germany a "consensus democracy" (Lijphart 1999, ch. 14).


In spite of the high consensus requirements of its institutional structures, the style of its parliamentary politics at the national level places Germany squarely in the class of "competitive democracies" (Bräuninger/Ganghof 2005; Lehmbruch 1998). With the exception of three years of a grand (CDU/CSU and SPD) coalition between 1966 and 1969, the "minimum-winning" coalition governments (Christian-Liberal, Social-Liberal or Red-Green) were always challenged by a strong opposition with its own policy program and with realistic hopes for displacing the government either through winning at the next national election or through a switch of the smaller coalition party. As a consequence, the "style" of interactions between governing and opposition parties tends to be competitive or even confrontational, rather than cooperative (Mayntz 1992; Scharpf 1997, 84-89). This has important repercussions for the functioning of federal institutions as well.


Given the unitary characteristics of German political culture and public opinion, the German party system is also unified.[3] Thus bottom-up and top-down processes of programmatic discussion, positional competition and legitimation occur within nationwide parties and their regional and local units. Within these parties, political leaders at the Land level play prominent roles, and prime ministers of Länder governments also have high visibility in national public opinion and considerable influence within their national parties. Thus, even if they should have no aspirations for national office,[4] it would be hard for prime ministers to dissociate themselves from the national party line - which must also affect their votes in the Bundesrat on issues with high party-political salience. These pressures increase when the Bundesrat vote may become politically decisive - i.e., when Länder governments controlled by the national opposition parties have sufficient votes to prevent a pro-government majority.[5]


Under these conditions of "divided government" (Laver/Shepsle 1991), opposition prime ministers may have three distinct motives for opposing federal initiatives:


- genuine Land interests,

- substantively differing ("policy-seeking") party-political preferences, and

- strategic ("position-seeking") interests in defeating initiatives that could strengthen the electoral position of the federal majority.[6]


The relative weight of these motives will vary from one policy issue to another and from one time to another. Generally, hard-core (e.g., financial) Land interests are unlikely to be overridden by considerations of party loyalty, whereas the strategic motive is most likely to dominate during election campaigns. When that is the case, the German joint decision system turns into a trap in which national policy initiatives are at the mercy of an opposition whose primary strategic interest is in unseating the government of the day.


Moreover, the likelihood of divided government is itself a function of national politics. Since Land elections determine the composition of the Bundesrat, they are generally conducted as "second order national elections" - which provide voters with the opportunity to express their dissatisfaction with the performance of the national government even if they should not (yet) want to throw it out of office. As a consequence, opposition majorities in the Bundesrat are most likely in periods of perceived crisis, when status-quo expectations are upset and unpopular remedies may be required (Manow/Burkhardt 2004). This was so during the oil-price crisis of the 1970s, when the Social-Liberal coalition faced an opposition-dominated Bundesrat; it was again true in the post-unification doldrums of the mid-1990s when the Social Democrats were able to frustrate the reform efforts of the Kohl government. Conversely, when the Red-Green government was elected in the fall of 1998, it was able to enjoy majorities in both houses only for half a year, and has been laboring under the constraints of an opposition veto ever since. As a consequence, since the mid-1990s, the joint-decision trap has prevented effective policy responses to increasing economic, fiscal and social challenges - with which some other European countries have been able to cope quite successfully (Scharpf/Schmidt 2000; Manow/Seils 2000).



1.3   A window of opportunity?


A permanent escape from the joint-decision trap would require changes of the federal constitution, and such changes must be adopted by a two-thirds majority in both houses of the legislature. In order to have any chance at all, therefore, reform efforts must assume that it will be possible to achieve agreement across three frontiers of potential conflict:


- between the governing coalition and the opposition in the Bundestag;

- between the federal government and the Länder as a whole; and

- among the diverse groups of Länder whose combined votes amount to a blocking minority (24 out of 69 votes) in the Bundesrat.


In the past, such efforts were not undertaken because it was assumed as a matter of course that the national opposition would not relinquish its most powerful instrument to frustrate the policy initiatives of the government of the day - and neither would Land prime ministers agree to eliminate the Bundesrat veto. In the last few years, however, these perceptions began to change.


While the opposition, playing the "game within rules," continued to exploit its Bundesrat veto to great effect, the deepening economic and fiscal crisis in Germany increased political frustration in all parts of the electorate and reduced the political payoff of "bloody-minded" blocking strategies. Moreover, as the government was launching a most painful reform package in 2003, allowing the opposition to soar in the polls, it was easy to predict that if it should form a new government after the next federal elections, it would also have to adopt highly unpopular economic and social policy reforms which could quickly re-create divided-government conditions. From the perspective of national party leaders in both political camps, therefore, there was reason to think about a new "game about rules" (Ostrom et al. 1994) which might allow German politics to avoid the joint-decision trap.


It was assumed, realistically, that this game could not be played successfully under conditions of party-political confrontation, so that constitutional reform only had a chance if it was completed well in advance of the next federal elections (scheduled to take place in the fall of 2006). Propitiously, moreover, the political calendar also foresaw no important Land elections between the fall of 2003 and the end of 2004. Taking advantage of this political window of opportunity, leaders of all parliamentary parties in the Bundestag launched a joint initiative for federalism reform in the summer of 2003.


But even if their national parties wanted them to do so, why should Länder prime ministers agree to relinquish their veto on federal legislation? In the abstract, there could be two motives: Some Land leaders anticipating future roles in the national government might wish to avoid the frustrations they would then encounter. At a less speculative level, it has become clear that the joint-decision trap is crippling not only national policy-making, but also the governing capacity of the Länder - which, in carrying out their implementation functions, are most directly confronted with the manifestations of politically unresolved economic and social problems.


More specifically, there has been a change of direction, or at least of emphasis, in the constitutional discourse over the last decade or so, from the affirmation of "cooperative federalism" to an assertion of the virtues of "competitive federalism."[7] In part, this can be seen as a response to the challenges of European economic integration which had the effect of opening national markets to competitors from all member states. Assuming that economic regulations that need to be uniform throughout the internal market are now defined at the European level, it is no longer clear that uniform national legislation would be the best response in areas where member states are still able to set the rules. In a country with highly diversified regional economies, the competitive advantages of specialization might be more effectively attained by a decentralization of legislative powers which would allow a more precise targeting of economic regulations to the needs of regionally concentrated industries (Scharpf 1999).


Instead of international economic competition among regions, political competition among Länder governments was emphasized by a second line of argument. Decentralized legislative competencies would empower Land governments and legislators to search for better solutions to common problems. In the context of nationwide media and an attentive public opinion, it was further assumed that political competition would then ensure the diffusion of "best-practice" solutions throughout the country. Unfortunately for the chances of reform, however, competition among the Länder also played a central role in a third line of arguments which in fact came to dominate the German debate over "competitive federalism."


It was derived from normative theories in institutional economics which emphasize the value of "systems competition" as a means for readjusting the balance between the state and the market (Sinn 1992;Vanberg/Kerber 1994): With decentralized competencies, so it was claimed, regional and local governments would have to compete for mobile taxpayers and factors of production - which would then force them to deregulate markets and to reduce the overall tax burden and the generosity of welfare-state transfers and services. Not surprisingly, proponents were also critical of the high degree of fiscal equalization among the Länder which, in their view, subsidized inefficient governments and eliminated incentives for pro-business policies at the regional level. Since the latter critique was taken up not only by business associations, the business press and the Liberal party (Friedrich-Naumann-Stiftung 1998), but also by the big and economically prosperous West German Länder, the promotion of "competitive federalism" came to be understood in the East as a renunciation of national solidarity, and in the West as an attack on the separate existence of very small and financially precarious Länder (such as Saarland, Bremen and Berlin).


In the German discourse about constitutional reform, "competitive federalism" thus became a derogatory label for the neo-liberal, or perhaps merely self-serving, rejection of the solidaristic commitments engrained in the post-war political culture and institutionalized in the constitution of West Germany - which the East Germans had been asked to accept unchanged. It could be challenged intellectually by reference to economic theorems of "ruinous competition" (Sinn 1995), and it was easy to mobilize the fears of potential losers against it. As a consequence, when the idea of federalism reform was seriously considered in the summer of 2003, a coalition of East German and small West German Länder succeeded in ensuring that three issues should not be allowed on the agenda of potential reforms:


-  No discussion about a territorial reorganization of German federalism (which would have created a smaller number of Länder, less unequal in size and in economic capacity);

-  no discussion about the existing system of fiscal equalization (which was to continue unchanged until the year 2019); and

-  no discussion of Länder legislative competencies in the field of taxation (which were associated with the dangers of ruinous tax competition).


With the most divisive issues thus out of the way, the Länder were then able to agree to the establishment of a reform commission in the fall of 2003 whose main mandate was to find ways to reduce the domain of Bundesrat vetoes in exchange for a major expansion of Land legislative competencies and a review of the jointly planned and financed programs.



2   The Commission for the Modernization of the Federal Constitution


The establishment of the Commission has its own history. Ever since the conclusion of the last round of fiscal-equalization negotiations in 2001, the heads of state chancelleries of a few Länder had been quietly talking with a few state secretaries at the federal level about potential reforms of the federal constitution. In April of 2003, the federal government issued a position paper on the modernization of German federalism that was long on demands addressed to the Länder and short on potential concessions (Bund 2003). Nevertheless, the talks continued, and on June 16, 2003 the chief civil servants of Bavaria, Bremen and the federal Ministry of Justice produced a draft suggesting ways to reduce the Bundesrat veto and to expand the legislative competencies of the Länder. Allegedly, the Chancellor's Office was less than happy about the drift of these talks. At any rate, the draft was leaked to the Bundestag, where it was interpreted as another attempt by governments to bypass parliaments in designing constitutional reforms behind closed doors. In response, all parliamentary parties insisted on a more inclusive and more open process of public deliberations about the future shape of German federalism.


The Liberal party, several public-interest organizations and some of the press would have preferred the "European model" of a constitutional convention which - with a representative membership, a free mandate and a single, strong chair - would deliberate in relative detachment from day-to-day federal-Länder business. However, among governments at both levels and in the big parties (SPD and CDU/CSU), this idea never had a chance. Since in the end a two-thirds majority would be required in both legislative chambers, so it was argued, deliberations ought to be structured in a way that ensured the ultimate adoption of proposals. Thus, the more pertinent model was thought to be the parliamentary conciliation committee (Vermittlungsausschuss) for the resolution of conflicts between Bundesrat and Bundestag votes. Composed of an equal number of members from both chambers, and bargaining or deliberating (Spörndli 2004) behind closed doors, its compromises are in fact rarely rejected by either chamber.


Hence the modernization of the federal order was entrusted to a Kommission von Bundestag und Bundesrat to be chaired jointly by Franz Müntefering, the leader of the SPD parliamentary party in the Bundestag and by Edmund Stoiber, the CSU prime minister of Bavaria.[8] Its voting members included the prime ministers or lord mayors of all sixteen Länder and an equal number of Bundestag MPs, divided proportionately among the parliamentary parties. Note, however, that the federal government was not among the voting members. Instead, three of its ministers (Justice, Finance, and Agriculture/Consumers) and the state secretary of the Chancellor's Office were granted a "guest status" that allowed them to participate in the debates. To emphasize the openness of the process, guest status was also assigned to some representatives of Land parliaments and of the associations of local governments as well as to twelve "experts" nominated by federal and Länder members and appointed by consensus in the first plenary session.


The commitment to openness was also reflected by the fact that monthly plenary meetings were in principle held in public, with verbatim records and all documents addressed to the full Commission available immediately on the Internet.[9] In contrast, two "working groups" - one dealing with the allocation of legislative competencies, the other one with financial relations - were always meeting in closed sessions (open to all voting and nonvoting members of the Commission). The same was of course true of the unofficial "preparatory meetings" preceding each of the official sessions, in which an attempt (largely unsuccessful) was made to coordinate the positions of federal and Länder members along party-political lines (Red-Green vs. Christian-Liberal).


With this structure in place, the Commission began its work in November 2003 with plenary debates on the general direction of reforms, complemented by "hearings" for which the experts were asked to prepare written analyses.[10] In the ensuing months, the overall agenda was then broken down into more narrowly framed issues to be resolved by the two working groups. The expectation was that the outlines of consensual constitutional reforms could be formulated before the summer recess, and that a draft of proposals addressed to the legislative bodies would be available before Christmas. Given the time constraints of top-level politicians, this was an extremely tight schedule - and by the end of spring, it was clear that it could not be realized within the procedural framework of one plenary session and two near-plenary working groups meeting once a month.


By early May, moreover, the Länder prime ministers, who had not shown much taste for plenary debates anyway, had been able to agree on a position paper which rejected nearly all proposals coming from the federal level, excluded all issues where conflict among the Länder could not be resolved, and presented a long and detailed list of their own demands for the transfer of legislative functions and compensating financial transfers (Positionspapier 2004). Given the seemingly firm commitment of all prime ministers to this list, general debates served no useful purpose any more. By mid-June, therefore, work was shifted to six (later seven) "project groups" composed of a small number of voting members, selected according to three-dimensional political criteria (federal/Land, left/right and East/West). Beyond that, participation was only by special invitation, and records and documents were treated as confidential even vis-à-vis other members of the Commission. Meeting at their own discretion during the summer, these project groups were expected to resolve specific subsets of the overall agenda and to report back to plenary meetings in the fall.


Paradoxically, however, these seemingly more efficient procedures did in fact impede the search for consensual solutions - and they did so for two theoretically interesting reasons. The first might well have been anticipated: Since each project group was only dealing with a very narrow range of issues, it was typically faced with asymmetrical constellations of interests, where one side was addressing demands to the other for which there could be no compensation within the same issue area. As a consequence, opportunities for negotiated compromises were rare. Therefore, even under the best of circumstances, the project groups were likely to produce a large agenda of unresolved issues for a final stage of negotiations in which cross-sectional bargaining with package deals and side payments would be possible.


The second reason is more complex: In the Commission, all parties had repeatedly rejected the insinuation that they might engage in Kuhhandel (horse trading), rather than in deliberations searching for the best solution. Unfortunately, however, this very commitment to "arguing," rather than "bargaining," (Elster 1986; 1998) proved counterproductive when it turned out that participants were not arguing about the same thing: When the prime ministers had agreed on their "position paper" in May, they had relied on the preparatory work of their state chancelleries which shared a "generalist" perspective focusing on the institutional architecture and pursuing a reallocation of broadly defined competencies that would create sufficient space for politically significant policy choices at the Land level. If their "specialist" ministries were consulted at all, they certainly had no major influence on the positions taken. In the project groups, however, the Länder generalists were confronted by specialists from the federal ministries, presenting their substantive-policy arguments in support of the status quo. As one frustrated Land official put it: "We were there to discuss competencies, but we had to listen to hours of lectures about child care" (interview, January 31, 2005).


Inevitably, these discussions became ever more fine grained, focusing at times on a single paragraph within a federal statute - so that concessions were also very narrow in scope. The same asymmetry also worked in the opposite direction. Where the federal government had asked for Länder concessions in such fields as tax administration, internal security, liability for violations of EU law and participation in EU decision processes, the Länder generalists were impervious to substantive-policy arguments and stood firm on their status-quo institutional positions. Thus, when the project groups reported back to plenary sessions of the Commission in October and early November, the list of agreed-upon solutions was disappointingly short and politically much less salient than the range of continuing controversies. In other words, after a year of deliberations, the Commission had failed to come up with consensual solutions to the problems on its agenda.


Thus bargaining across issue areas had neither been possible in plenary meetings nor in the specialized project groups. There was still a chance, however, that package deals involving federal concessions in some policy areas in exchange for Länder concessions in others might be worked out after the issues had been clarified in the project groups. But time was running out since the last meeting of the Commission was scheduled for December 17 - beyond which date the political "window of opportunity" was generally assumed to close rapidly in anticipation of two crucial Land elections. Hence the Commission agreed to entrust the management of the bargaining phase entirely to Stoiber and Müntefering, its dual chairs, who had less than six weeks to work out a consensual overall solution by consulting with whomever they thought it useful. Both of them, supported by their own staffs,[11] undertook to formulate tentative compromises and package solutions which they exposed to the critique of three rounds of a small Obleuterunde - i.e., a select group of Commission members supposed to represent the politically salient divisions mentioned above. Even after the last of these rounds on December 3 had ended in open conflict over education competencies, the joint chairs, nevertheless, continued their search for compromises practically down to the last hour before the announcement of failure in the afternoon of December 17.



3   What was agreed and why did it fail?


In these last six weeks, an astonishing range of issues were in fact resolved, so that the ultimate failure remained a puzzle even to members of the Commission who had been involved in the bargaining phase. In order to explore this puzzle, it is necessary to look at what was and what was not achieved from the perspective of those groups of actors that were veto players. According to original expectations, it should be remembered, reforms ought to result in significantly reduced veto powers of the Bundesrat exchanged for significantly enlarged legislative competencies of the Länder. In a nutshell, reform failed because both of these expectations were left largely unfulfilled.



3.1   The Bundesrat veto


In public perception, the most important motive for initiating the reform of German federalism was to reduce the number of occasions giving rise to a Bundesrat veto. Since the Commission's mandate did not include reforms of the tightly integrated fiscal relationships between the levels of government, however, it was soon realized that there was no way to abolish the veto in a wide range of highly salient and controversial issues. Hence the Commission focused exclusively on Art. 84 of the Basic Law which requires Bundesrat agreement for federal legislation that must be implemented by the Länder if the statute includes regulations of administrative procedure or organization. The solution proposed is straightforward and almost ridiculously simple: If the federal government is forbidden to adopt procedural or organizational regulations, the veto goes out as well.[12] Remarkably, the Länder succeeded in presenting this proposal as a major concession on their part for which a special compensation was required: Since in the past their veto on procedural grounds could also be used to defend other interests, they now insisted on a new Bundesrat veto over all federal legislation that imposed significant financial burdens on them. As a consequence, it was at best unclear if this reform would have led to significant, or any, reductions of veto opportunities.


In the Commission it was never even mentioned that the proposed solution could be achieved unilaterally by the federal government: If it would voluntarily abstain from procedural and organizational regulations, the veto would also go away - without the need for additional fiscal compensations. Even more remarkably, the Commission was not reminded by its legal experts of a decision of the Constitutional Court which had, two years earlier, legitimated the formal separation of a substantive statute and a statute regulating administrative procedures[13] - with the consequence that the veto would only apply to the latter.[14] Since the federal government must have been aware of these conditions, and since the Länder seemed unwilling to accept any of its other demands,[15] it seems plausible that the federal government was not very eager to offer additional concessions of its own in project-group deliberations. In fact, not only the Länder but also MPs of the governing coalition had the impression that federalism reform did not have a high priority at the center.[16] If that constellation of interests had not changed, there would have been little reason to expect much progress in the bargaining phase of the last six weeks.



3.2   The Constitutional Court changed the agenda


Things changed, however, with the judgment of the federal Constitutional Court of July 27, 2004,[17] which struck down an amendment to the federal framework law for higher education that had introduced the "junior professorship" as a regular step in academic careers.[18] The law had been opposed not only by conservative academics but also by some Land governments who challenged it as exceeding federal competencies. Given the excitement over the substantive consequences of the decision for the jobs and subsequent careers of junior professors who had already been appointed under the law, its wider institutional implications were not immediately realized: They derived from the fact that the court had not been content to find that the amendment exceeded the allowable limits of "framework" legislation. It had also based its decision on an interpretation of the "necessity" clause of Art. 72, para. 2 of the Basic Law which applied not only to framework laws but also to the long list of "concurrent" competencies that constitute the bulk of federal legislation.


This clause, which allowed federal legislation only if it was necessary ("erforderlich") for the maintenance of legal and economic unity or for the achievement of equivalent ("gleichwertige") living conditions in the country, had long been treated as a "political question" by the Constitutional Court - with the consequence that federal legislation had come to occupy the field to the practical exclusion of Land legislation. In response, a constitutional amendment adopted in 1994 had tightened the language of the clause and explicitly established the authority of the court to interpret and apply it. As it turned out, however, the court's interpretation was now extremely restrictive, limiting federal legislative power to constellations where it could be shown that divergent Länder legislation would have near-catastrophic consequences. If the same rule were now to be applied across the board, practically the whole body of existing national legislation - from civil and criminal law to labor law and traffic regulations - could be challenged not only by Land governments and parliaments but by private parties as well. In other words, the functioning of much of the legal system of the Federal Republic was now potentially in question.


From the perspective of federalism reform, the court had drastically changed the constitutional status quo to the disadvantage of the federal government - and thus the default outcome on the basis of which actors at both levels had defined their bargaining strategies. Although the decision came too late to be reflected in the plenary discussions of the Commission or in the project groups, the urgent need for damage control was evident to the federal Ministry of Justice and the Chancellor's office. Now it was no longer enough to allow individual ministries to argue against demands for the transfer of legislative powers. Instead, the cooperation of the Länder was needed to protect the integrity of core areas of federal legislation - preferably by moving competencies out of the concurrent category into that of exclusive federal competencies. Given their interest in orderly administration, the Länder were not entirely averse to such solutions,[19] but they would of course use the new leverage to ask for substantial compensation. In anticipation, the Chancellor's Office now called the ministries into a "confessions box," requiring them to review the portfolio of their legislative powers with a view to reducing them to their core competences. As a result, in November, the federal government was able to enter the final negotiations with the offer to transfer altogether 26 items of its legislative competencies to the Land level (Bund 2004) - far more, that is, than what had seemed possible at the end of the project-group phase.[20]



3.3   Conflicts among the Länder


So why did the reforms nevertheless fail in the end? To understand this, it is now necessary to look at the cracks in the united front the prime ministers presented in their position paper of May 2004. It was the big and economically prosperous West German Länder that had an overriding interest in enlarging their political action space. In the interest of unity, they had reluctantly agreed to exclude territorial reorganization and fiscal equalization from the agenda of the Commission, and they refrained, even more grudgingly, from voicing their interest in autonomous taxing powers.[21] In exchange, the group of Eastern and small and/or economically weak Western Länder had gone along with demands for the abolition of jointly-financed programs and for the transfer of autonomous legislative powers in the fields of regional economic, labor market, social and environmental policy and for the total elimination of federal competencies in the field of education - "from child care all the way to the Habilitation."


Given the dependence of the weak Länder on federal support and their fears of falling even further behind in interstate competition, these were no easy concessions - and they may not have been made entirely in good faith. At any rate, their representatives were often quite ready to concede to federal ministries arguing against such concessions in the project group phase; and of the 26 legislative competencies finally offered by the federal government in November, several (including, again, autonomous taxing powers) did not survive the negotiation phase.[22] In the end, the strong Länder thought they had obtained very little. Apart from the (largely undisputed) right to adopt their own civil-service regulations, the consensual package included mostly competencies that were narrowly circumscribed; instead of regional economic policy, the right to regulate shop closing hours and local fairs; instead of social policy the right to regulate old-age homes, or instead of environmental policy the right to regulate the noise of leisure activities. Given the latent conflict of interest among the Länder, it seems that none of these issues was seriously promoted in the final rounds. Instead, prime ministers insisted on full control over education as the one major field where all Länder seemed to agree - but also the field where the coalition parties and the federal government were unwilling to surrender the last remnants of federal competencies left by the Constitutional Court.



3.4   The impasse over education: All or nothing at all?


Education is indeed one of the areas where the national government never had undisputed competencies. As part of the wider field of "culture" (which also includes support for the arts and for research, and the regulation of radio and TV), it belongs to the domain which, in 1949, the Länder had reserved to themselves. Given the unitary character of the national culture, however, and the forced mobility of large parts of the population, they had seen it necessary to set up, even before the establishment of the Federal Republic, a "permanent conference of ministers of culture" which, over the decades, has created a highly constraining and extremely cumbersome system of negotiated coordination.


In the reform-minded 1960s and under the influence of a grand coalition at the federal level, it is true, the Länder had been persuaded to accept the financial support and guiding influence of the federal government. Institutions for the joint planning and joint financing of research support and university construction were set up, federal framework legislation for higher education was introduced, and an elaborate machinery for joint educational planning was established. Soon after the end of the grand coalition, however, ideological conflict over comprehensive schools put an end to joint education planning, and framework legislation for higher education became the battlefield, or punching ball, of university reforms as understood by Social-Liberal, Christian-Liberal and Red-Green federal governments (Mäding 1989). With the exception of jointly-financed research support, the prime ministers of the strong Länder wanted to get rid of all of this, and in this regard, the rest of their colleagues acquiesced.


Even before the Commission was established, the federal government had agreed to end the joint financing of university construction (engendering, predictably, conflict over the reallocation of the funds so released). Beyond that, however, it was politically committed not only to do something about the deficiencies, revealed by the PISA evaluation, of German pre-school and primary education, but also about the alleged lack of "excellence" and international competitiveness of even the best German universities. At the same time, however, its constitutional base for providing federal support to pre-school and elementary education and to universities was extremely fragile, to say the least, and the domain of its framework legislation for higher education was eroding under court decisions denying federal authority to introduce the junior professorship or, soon to follow, to prohibit the collection of student fees. In other words: If the government hoped to have any chance of realizing its political ambitions in the field of education, it was facing a steep uphill struggle in trying to build a broad coalition in the federalism Commission.


It is puzzling, therefore, that during the first six months, only the Länder position was presented in the Commission, and that it remained practically uncontested in plenary discussions. It was only at the beginning of the project group phase in early June that two SPD MPs introduced a position paper on the reform of jointly financed programs which, among other proposals, also included a clause that would have allowed the federal government and the Länder to "cooperate in the development of the education system" (Kröning/Runde 2004). The response was outrage among prime ministers of the CDU/CSU and puzzlement among the others. In order to avoid a blockade of discussions on all other issues, the authors had to place their proposal "in square brackets" until the final round of negotiations - where it was re-introduced and again squarely rejected by the opposition prime ministers.


In the hectic final weeks of bargaining in small circles and behind closed doors, there was of course no chance to invent, explain and build support for new constructive solutions that might have satisfied both sides. Instead, the only question was how far the federal side would reduce its status-quo competencies in order to reach a compromise. In view of the government's inevitable role in EU harmonization processes, the prime ministers seemed at least willing to accept a minimum of federal framework competencies for the regulation of university degrees and of access to universities - but even that potential agreement evaporated when the federal minister of research and education submitted a position paper interpreting the regulation of access so as to also include the regulation of student fees.


In the end, no compromise on education could be reached. Given the extreme demands of the other side, federal negotiators finally preferred their eroding status-quo competencies to any compromise on education that might be acceptable to the opposition prime ministers. At the same time, however, they expected the Länder to adopt the agreements that had been reached, or were within sight, on all other issues. As that expectation turned out to be wrong, federalism reform failed altogether.


Assuming that the agreements that had been reached were preferred to the status quo by all parties concerned, and considering that the status quo in education would continue in either case, the choice of the Länder seems to lack game-theoretic rationality. Allegedly, some of the prime ministers thought so as well, and would have been willing to accept the negotiated solutions even in the absence of agreement over education. So the failure must be explained from the perspective of the blocking minority of big, prosperous and politically ambitious West German Länder, especially those governed by the opposition parties. Even for them, it appears from interviews, the decision to let the whole reform effort fail was not entirely based on rational calculations, but was emotionally conditioned by disappointment and frustration: Having failed to gain the political action space they wanted in economic, labor market, social and tax policy, and now on education as well, they did not consider the list of narrowly circumscribed competencies they would have received[23] as worth their effort. "This is not what we came here for" (interview).[24]



4   Future options


Could the failure of reforms have been prevented? The answer cannot turn narrowly on the issues in education - which became the crucial issue vis-à-vis the federal government only because it was one on which all Länder could be made to agree, and on which no responsible federal government could or should accept total surrender. What really stood in the way of successful reform were the underlying conflicts of interest between the strong and the weak Länder. They had always existed, but after German unification their political salience greatly increased.


For the strong Länder, it matters greatly that now governments dependent on federal support and fiscal-equalization payments have a clear majority in the Bundesrat. Moreover, much of federal legislation has come to be determined by EU directives - whose content they are confident to influence more effectively through the lobbying activities of their large, or even grandiose, embassies in Brussels, rather than through the collective role of the Bundesrat in European affairs (Finanzreport NRW). In short, the post-war logic of trading Land legislative competencies in return for an increasing national role and a controlling influence on federal legislation has lost its persuasiveness for the prime ministers of the strong Länder. While they may still enjoy the national limelight, the desire for greater political autonomy at home has clearly gained in importance among their institutional preferences.


On the other hand, there are Länder for which greater autonomy appears as a threat, rather than a promise. German unification has added regions to the Federal Republic for which the constitutional promise of "equivalent living conditions" has come to depend entirely on massive West-East transfer payments - in the form of horizontal fiscal equalization among the Länder, of preferential treatment in the vertical allocation of tax revenues, and of additional federal grants under a variety of headings.[25] Once it was realized that, in spite of all efforts, the economic gap between East and West Germany would take much longer to close than had been imagined in 1990, the continuation of West-East transfers became the highest priority for East German leaders. Furthermore, having recently achieved guarantees for the period 2005 to 2019, they would not allow issues of fiscal equalization to be reopened in the Commission.[26]


This was regrettable as the existing regime of fiscal equalization is widely judged to be inefficient and in part counterproductive.[27] Even more damaging for the prospects of federalism reform, however, is the implicit association of uniform legislation with equality and solidarity. In fact, immediately after unification, East German prime ministers had protested against being subjected to the indiscriminate application of West German laws that did not fit the specific problems they had to deal with.[28] In the meantime, however, demands for legislative autonomy are muted by fears, never fully articulated and explicitly discussed in the Commission, that the East-West gap would also widen if greater legislative autonomy were to allow West German regions to further increase their relative attractiveness for firms, investors, and tax-paying citizens. These fears not only explain the rigid refusal to even discuss the possibility that the Länder might be allowed to set the rates on their own taxes, but also played a role in the quick collapse of the Länder front when the federal ministries argued against demands (raised in the seemingly consensual position paper of the prime ministers) for the transfer of legislative competencies in the fields of regional economic, labor-market and social policies,[29] or for the dismantling of jointly-financed programs of regional industrial policy. In effect, the prospect of allowing divergent Land legislation in these fields was equated with repudiating the constitutional promise of equivalent living conditions. More cynically, one might also think that governments in the East and in economically weak West German Länder were particularly reluctant to accept responsibilities in policy areas which might expose them to political blame for the poor economic performance of their regions.


In any case, at the end of the six-week negotiation phase, the list of legislative competencies that were to be transferred to the Länder was considerably shorter than the list the federal government had proposed at the beginning of November, and apart from the regulation of public-service employment and, perhaps, shop closing hours, it contained no matters of major economic, social or political significance. Hence, when it became clear that the federal government was not willing to move out of the education field altogether, the economically strong Länder felt that the game was not worth the candle and used their blocking minority to stop the whole project of federalism reform. They may have hoped, so one may surmise, that total failure would maintain the pressure for more far-reaching reforms under more auspicious future conditions.


Could there be such conditions? During the spring of 2005, Müntefering and Stoiber, the twin chairs of the defunct Commission, did in fact resume their informal talks trying to tie up the loose ends of their draft proposal which had been rejected on December 17. There were reports suggesting that they had in fact reached agreement among themselves, and that they might introduce a joint proposal into the parliamentary arenas after the elections in North Rhine-Westphalia at the end of May. But since these elections brought a resounding defeat of the Red-Green Land government and the prospect of national elections in the fall of 2005 (instead of 2006), Müntefering ruled out any action on federalism reform in the remaining months of the present Bundestag. At the time of this writing (mid-July 2005), these elections are still uncertain,[30] but political parties are preparing for the campaign and expectations are that the outcome would be a Christian-Liberal coalition at the national level. If that should happen, there would be at least a period of congruent party-political majorities in both houses.


But it is not clear that this new political constellation would do much to facilitate the reform of federalism. It seems unlikely that the Red-Green parliamentary parties would not constitute at least a blocking minority in the next parliament, and a Christian-Liberal government would soon enough see matters from the federal perspective as well. Moreover, party politics had played almost no role in the failure of the Commission.[31] What mattered were differences of institutional self-interest between the federal government and the Länder and, above all, the asymmetry of conditions shaping the institutional preferences of the strong and the weak Länder. These differences will not disappear after the election. There is no reason to think that an allocation of competencies that would satisfy the legitimate demands of the strong would be acceptable to the weak - and since both sides have a veto in constitutional reform, any uniform solution is likely to remain unsatisfactory or fail altogether. The question is, therefore, whether asymmetric solutions that would allow more autonomy to some Länder without activating the fears of others are conceivable.



5   Toward an asymmetric German federalism?


Such solutions were in fact discussed in the Commission. The most promising one, which would have allowed Länder parliaments to adopt laws deviating from existing federal legislation, has been on the agenda of constitutional reform since 1977.[32] It surfaced again in exploratory discussions among state secretaries in the summer of 2003, and then in the position paper of prime ministers in May of 2004, where opt-out rights (Zugriffsrechte or Abweichungsrechte) were presented as a fall-back solution if demands for the complete transfer of legislative powers should fail. At that time, however, the federal government was unwilling to consider opt-outs even as a second-best solution: Where there was a need for national legislation, it ought to apply throughout the country; and where diversity was acceptable, a clear-cut delegation of legislative powers to the Land level would be appropriate. To allow federal and Land legislation to exist side by side in the same field, and in combinations differing from one Land to another, could only amount to confusion. In this regard, the government was supported by most of the legal experts on the Commission.


Moreover, the project group that had been asked to examine the feasibility of opt-out solutions came up with a variety of models that differed so much in their features and potential implications that the subsequent plenary debate could neither clarify the issues nor persuade the skeptics.[33] As a result, the Commission reasserted its preference, in principle, for a clear-cut separation of competencies, but was willing to reconsider the possibility of opt-outs if agreement on the outright delegation of legislative powers could not be reached. But when, at the very end of the bargaining phase, it was clear that the range and the political salience of consensual transfers of legislative competencies was insufficient from the perspective of the strong Länder, there was neither the time nor the willingness of negotiators to return to the unfinished and seemingly most controversial business of exploring opt-out solutions.


Ironically, however, a feasible model had already been construed in the discussion about ways to reduce the incidence of Bundesrat vetoes. In principle, as I said above, the federal government could do so by simply avoiding the regulation of administrative procedures and organization in its substantive statutes. But ministries argued that procedure and substance were often hard to distinguish, and that in some cases (e.g., in the field of immigration) the regulation of procedures may constitute the core concern of the legislation. Accepting the argument, the Länder were willing to allow procedural regulations to be adopted without Bundesrat agreement, as long as they were allowed opt-out rights. If, however, the federal government should insist on imposing uniform administrative rules without the possibility of opt-out, the statute could only be passed with the agreement of the Bundesrat. In fact, this step-wise solution amounts to a conditional opt-out with constraints to be defined politically by the combined majorities of the federal legislature.


Even though it was pointed out in the Commission that the solution adopted for the regulation of administrative procedures could be taken as a model for substantive legislation as well, the suggestion was not taken up while discussions might still have had a chance. At that time, the Länder still hoped that they could achieve more attractive outcomes by way of a strict separation of competencies. They were bound to be disappointed, however, because the separation concept, derived from idealized images of the U.S. constitution, ignored both the multi-level characteristics of many policy areas and the asymmetry of conditions and interests among the Länder.


In most of the fields where full competencies had been claimed by the Länder - economic, labor market, environmental and social policy, as well as education - some aspects would clearly require regulation at the European level; for others, uniform national rules seem to be called for; and for still other aspects, regional or even local solutions would be preferable. However, given the sensitivity of these "functional" criteria to changes over time and differences from one issue to another, and the unitary character of the German political culture, the near-automatic response to demands for a clear-cut separation of governing responsibilities would favor the status quo and allow, at best, the transfer of narrowly defined subsets of federal competencies to the regional level. Even more importantly, the separation principle could not accommodate the diversity of interests and conditions among the German Länder. A constitution that would fit the needs and opportunities of Bavaria or North Rhine-Westphalia would be very different from a constitution reflecting the functional relationships between the national government and Länder like Bremen, the Saarland, or Mecklenburg-West Pomerania. If this underlying asymmetry was not respected (and if all parties were aware of the implications of alternative proposals), the outcome could only be deadlock or, at best, minimal reforms that could not stray far from the constitutional status quo - i.e., just the type of reforms that were rejected on December 17.


How could (conditional and controlled) opt-outs help to deal with both the multi-level problem and the asymmetry of Länder interests? On the one hand, they would allow federal legislation to deal with a given policy area systematically and comprehensively - without being constrained by the restrictive interpretation of the "necessity clause" adopted by the Constitutional Court.[34] In contrast to the present "framework legislation," there would also be no need to refrain from specific requirements (and no temptation to go too far in the regulation of details). In other words, the federal legislature could regulate all aspects that it considered necessary for a proper treatment of the matter. Länder that are satisfied with the content of national regulation could leave it at that, without the unattractive need to mobilize their own administrative and political resources to deal with the residual issues which federal law was presently not allowed to touch under either the "necessity clause" or the constraints of framework legislation.


On the other hand, however, every Land parliament would also be free to scrutinize the goodness-of-fit between federal legislation and regional conditions or preferences and to enact corrections where it thought it necessary to do so. The critical (and in the Commission unresolved) question was what the effect of such parliamentary decisions should be. The initial position of the Länder was that the opt-out, once exercised, should once and for all displace federal legislative competencies, whereas some federal MPs considered a lex-posterior rule by which a later federal law would override the opt-out, but could itself be modified by later opt-outs (Stünker-Röttgen 2005). This solution, criticized as "ping-pong legislation," found little support in other quarters. The federal government had no sympathy for constitution-based opt-outs in any form.[35] What was not fully explored was the potential of conditional opt-outs that would be either politically or judicially controlled at the national level.


In case of a judicial solution, the constitution itself might simply include a "reverse necessity clause," allowing Land opt-outs where there is no need for nationally uniform rules. This would leave the Constitutional Court in full control - which could simply continue to develop and apply the definitions which it presently uses to limit federal legislative competencies under Art. 72 para. 2 Basic Law. A political solution, in contrast, would give the Bundestag, acting with the agreement of the Bundesrat, the option to veto a Land opt-out. Unfortunately, however, the Länder, apparently unconvinced that sauce for the goose was sauce for the gander, did not like the reverse necessity clause, and they seemed even more averse to political controls. But some such controls would be unavoidable if opt-outs were to be allowed in the full range of concurrent legislative competencies. Otherwise, the solution would again activate all fears of ruinous competition and "beggar-my-neighbor" policies among the Länder, and of purely self-interested renunciations of national cohesion among federal actors.


But how much autonomy would be gained if national controls, judicial or political, would still be in place? The answer is, quite a lot. Compared to discussions in the Commission about the transfer of legislative powers, such controls would not be anticipatory but ex post. Thus federal actors would not be asked to speculate about the potential range and dangers of legislative action that future Land legislators might consider, but would be confronted with a single, specific initiative adopted by a democratically accountable Land parliament in full knowledge of the federal law and after full debate. In contrast to decisions based on the abstract anticipation of potential dangers, the ex-post examination of a specific opt-out, whether conducted in the judicial or the political arena, would be much more likely to favor the exercise of autonomous legislative powers. Moreover, the right of initiative would lie with the individual Land legislature, rather than with the federal legislature which at present could (but has never yet done so) provide an opening for Land legislation (Art. 72 para. 3 Basic Law). Nor would the Land first have to bring legal action against the federal government, and wait for a favorable decision of the Constitutional Court - which would then nullify the federal rule also for those Länder that would have been content to apply it.


As a result, federal majorities could legislate unencumbered by either the veto of the Bundesrat or the Constitutional Court's interpretation of a necessity clause, while Land legislature could modify federal rules where these did not fit local conditions or preferences - and if the federal side wanted to prevent such departures, it would need the agreement of either the Bundesrat or the Constitutional Court. In short, the introduction of conditional opt-outs would increase the autonomy of federal and Land legislation at the same time. The price to be paid (which weighed heavily for the federal government and for legal experts in the Commission) would be the anomaly that Land law could sometimes override federal law, and that federal law would not always apply throughout the federal territory. From a problem-oriented perspective, however, this anomaly is precisely what is needed to accommodate the otherwise incompatible requirements of greater political autonomy at both levels of government under the conditions of multilevel-interdependence and asymmetric interests and conditions.


In my view, therefore, conditional and controlled opt-outs are probably the only way in which the demand for significantly greater legislative autonomy of the strong Länder could have been realized within the constellation of veto players involved in attempts at constitutional reforms in Germany. As a complement to protecting the interests of the weak Länder, it would probably be necessary to establish a national authority to require comparative reports on the performance of governing functions in the Länder - combined with the continuation, or even extension, of selective federal grants alleviating deficiencies in these functions. In short, Germany must move to a more asymmetric constitutional form (Agranoff 1999) if its federalism should have a chance to regain the flexibility and resilience which facilitated the success of the post-war German model. For the time being, however, the failure of constitutional reform has deepened the pervasive pessimism that is clouding the politics of this country.







Agranoff, Robert (ed.), 1999: Accommodating Diversity: Asymmetry in Federal States. Baden-Baden: Nomos.

Benz, Arthur, 1985: Föderalismus als dynamisches System. Zentralisierung und Dezentralisierung im föderativen Staat. Opladen: Westdeutscher Verlag.

–––, 1994: Kooperative Verwaltung. Funktionen, Voraussetzungen und Folgen. Baden-Baden: Nomos.

Bräuninger, Thomas/Steffen Ganghof, 2005: Parteienwettbewerb im Zweikammersystem. In: Steffen Ganghof/Philip Manow (eds.), Mechanismen der Politik. Strategische Interaktion im deutschen Regierungssystem. Frankfurt/M.: Campus, 149-181.

Bund, 2003: Modernisierung der bundesstaatlichen Ordnung. Position der Bundesregierung vom 9. April 2003.

–––, 2004: Modernisierung der bundesstaatlichen Ordnung. Unveröffentlichtes Manuskript. Berlin, 7 November 2004.

Bundestag/Bundesrat (eds.), 2005: Dokumentation der Kommission von Bundestag und Bundesrat zur Modernisierung der bundesstaatlichen Ordnung. Zur Sache 1/2005. Berlin: Deutscher Bundestag.

Elster, Jon, 1986: The Market and the Forum: Three Varieties of Political Theory. In: Jon Elster/Aanund Hylland (eds.), Foundations of Social Choice Theory. Cambridge: Cambridge University Press, 103 132.

Elster, Jon (ed.), 1998: Deliberative Democracy. Cambridge: Cambridge University Press.

FinanzReport NRW, 2005: Interview mit Dr. Manfred Degen, Leiter der Landesvertretung NRW in Brüssel. In: FinanzReport NRW Online, March 2005.

Friedrich-Naumann-Stiftung, 1998: Für eine Neuordnung der Finanzverantwortung von Bund und Ländern. Manifest vom 21.8.1998. Bonn.

Ganghof, Steffen, 2003: Promises and Pitfalls of Veto Player Analysis, In: Swiss Political Science Review 9, 1-25.

Heinsen, 1977: Sondervotum zu Abschnitt 4.1 "Neuverteilung der Gesetzgebungszuständigkeiten" von Senator a.D. Dr. Heinsen. In: Schlußbericht der Enquete-Kommission Verfassungsreform des Deutschen Bundestags. Teil II: Bund und Länder. Zur Sache 2/77. Bonn: Deutscher Bundestag, 76-77.

Hesse, Konrad, 1962: Der unitarische Bundesstaat. Karlsruhe: C.F. Müller.

Homburg, Stefan, 2004: Stellungnahme zur öffentlichen Anhörung Finanzbeziehungen am 11. März 2004. Kommission von Bundestag und Bundesrat zur Modernisierung der bundesstaatlichen Ordnung. Kommissionsdrucksache 0024. Berlin.

Kröning, Volker/Ortwin Runde 2004: Vorschlag zur Neuregelung der Mischfinanzierungstatbestände der Art. 91a und 91b sowie Art. 104a Abs. 4 GG mit Erläuterungen, 2. Juli 2004. Kommission von Bundestag und Bundesrat zur Modernisierung der bundesstaatlichen Ordnung. Kommissionsdrucksache 0057-neu. Berlin.

Laver, Michael/Kenneth A. Shepsle, 1991: Divided Government: America Is Not "Exceptional." In: Governance: An International Journal of Policy and Administration 4, 250-269.

Lehmbruch, Gerhard, 1998: Parteienwettbewerb im Bundesstaat. Regelsysteme und Spannungslagen im Institutionengefüge der Bundesrepublik Deutschland. 2nd edition. Opladen: Westdeutscher Verlag.

Lehmbruch, Gerhard, 2002a: Der unitarische Bundesstaat in Deutschland: Pfadabhängigkeit und Wandel. In: Arthur Benz/Gerhard Lehmbruch (eds.), Föderalismus. Analysen in entwicklungsgeschichtlicher und vergleichender Perspektive. PVS-Sonderheft 32. Wiesbaden: Westdeutscher Verlag, 53-110.

Lehmbruch, Gerhard, 2002b: Restriktionen und Spielräume einer Reform des Bundesstaates. In: Wirtschaftsdienst 82, 197-201.

Lijphart, Arend, 1999: Patterns of Democracy. Government Forms and Performance in Thirty-Six Countries. New Haven: Yale University Press.

Lübecker Erklärung, 2003: Bekenntnis zum Föderalismus und zur Subsidiarität - Landesparlamente stärken! Lübecker Erklärung der deutschen Landesparlamente angenommen auf dem Föderalismuskonvent der deutschen Landesparlamente am 31. März 2003 in der Hansestadt Lübeck.

Mäding, Heinrich, 1989: Federalism and Education Planning in the Federal Republic of Germany. In: Publius: The Journal of Federalism 19, 115 131.

Manow, Philip/Eric Seils, 2000: Adjusting Badly: The German Welfare State, Structural Change, and the Open Economy. In: Fritz W. Scharpf/Vivien A. Schmidt (eds.), Work and Welfare in the Open Economy. Vol. 2: Diverse Responses to Common Challenges. Oxford: Oxford University Press, 264-307.

Manow, Philip/Simone Burkhart, 2004: Legislative Autolimitation under Divided Government: Evidence from the German Case 1976-2002. MPIfG Discussion Paper 04/11. Cologne: Max Planck Institute for the Study of Societies.

Mayntz, Renate, 1992: Social Norms in the Institutional Culture of the German Federal Parliament. In: Richard Münch/Neil Smelser (eds.), Theory of Culture. Berkeley: University of California Press, 219-240.

Ostrom, Elinor/Roy Garner/James Walker, 1994: Rules, Games, and Common-Pool Resources. Ann Arbor: University of Michigan Press.

Positionspapier, 2004: Positionspapier der Ministerpräsidenten: Föderalismusreform, 6. Mai 2004. Kommission von Bundestag und Bundesrat zur Modernisierung der bundesstaatlichen Ordnung. Kommissionsdrucksache 0045. Berlin.

Scharpf, Fritz W., 1988: The Joint-decision Trap: Lessons from German Federalism and European Integration. In: Public Administration 66, 239-278.

–––, 1989: Der Bundesrat und die Kooperation auf der "dritten Ebene". In: Bundesrat (eds.), Vierzig Jahre Bundesrat. Baden-Baden: Nomos, 121-162.

–––, 1997: Games Real Actors Play. Actor-Centered Institutionalism in Policy Research. Boulder: Westview.

–––, 1999: Föderale Politikverflechtung: Was muß man ertragen? Was kann man ändern? In: Konrad Morath (ed.), Reform des Föderalismus. Bad Homburg: Frankfurter Institut, 23-36.

–––, 2001: Notes Toward a Theory of Multilevel Governing in Europe. In: Scandinavian Political Studies 24, 1-26.

Scharpf, Fritz W./Vivien A. Schmidt (eds.), 2000: Work and Welfare in the Open Economy. Vol. 1: From Vulnerability to Competitiveness. Oxford: Oxford University Press.

Scharpf, Fritz W./Bernd Reissert/Fritz Schnabel, 1976: Politikverflechtung: Theorie und Empirie des kooperativen Föderalismus in der Bundesrepublik. Kronberg/Ts.: Scriptor.

Schönhoven, Klaus, 2004: Wendejahre. Die Sozialdemokratie in der Zeit der Großen Koalition 1966-1969. Bonn: Dietz.

Sinn, Hans-Werner, 1995: The Subsidiarity Principle and Market Failure in Systems Competition. CES Working Paper 103. Munich: Centre for Economic Studies.

Sinn, Stefan, 1992: The Taming of Leviathan: Competition Among Governments. In: Constitutional Political Economy 41, 85-107.

Spörndli, Markus, 2004: Diskurs und Entscheidung. Eine empirische Analyse kommunikativen Handelns im deutschen Vermittlungsausschuss. Wiesbaden: VS Verlag.

Stünker, Joachim/Norbert Röttgen 2005: Eckpunkte zu einer Neufassung des Art. 72 GG, 27. September 2004. Projektgruppenarbeitsunterlage 1/17. In: Bundestag/Bundesrat (eds.), Dokumentation der Kommission von Bundestag und Bundesrat zur Modernisierung der bundesstaatlichen Ordnung. Zur Sache 1/2005. Berlin: Deutscher Bundestag.

Troeger-Kommission, 1966: Kommission für die Finanzreform. Gutachten über die Finanzreform in der Bundesrepublik Deutschland. Stuttgart: Kohlhammer.

Tsebelis, George, 2002: Veto Players. How Political Institutions Work. Princeton: Princeton University Press.

Vanberg, Viktor/Wolfgang Kerber, 1994: Institutional Competition Among Jurisdictions: An Evolutionary Approach. In: Constitutional Political Economy 5, 193 220.







Plans for a public television network based on an agreement between the federal government ant its political allies at the Land level.



For the EU, I have argued (Scharpf 2001) that in important policy areas the intergovernmental and joint-decision modes are displaced by a supranational-hierarchical mode in which the Commission, the Court of Justice and the Central Bank are able to impose policy choices unilaterally. In Germany, that is also true of the role of the Constitutional Court whose strongly interventionist judgments have not only constrained but actively shaped policies in important areas. One such example will be discussed below.



Exceptions are the CSU (which can be described as a national party whose territorial base is limited to Bavaria) and the PDS (whose electoral support was in the past limited to East Germany, but which is now trying to compete nationally as the "Party of the Left").



Chancellors Kiesinger, Brandt, Kohl and Schröder had been Land prime ministers before, and ministers of the federal government had often played prominent roles at the Land level.



Legislation requiring the agreement of the Bundesrat must be passed with an absolute majority of Länder votes, thus abstentions count as "no" votes, and initiatives fail in a tie.



On the distinction between policy seeking (substantive) and position-seeking (strategic) preferences, see Ganghof (2003) and Bräuninger/Ganghof (2005).



In addition, the decentralization of legislative functions was generally supported by arguments emphasizing the "democratic deficit" of cooperative federalism - where the national majority is prevented from pursuing policies legitimated through its electoral mandate, and where parliaments at the Land level have lost almost all legislative functions, while chief executives cannot be held accountable for their role at the national level. These arguments found much support in the press and, not surprisingly, among the members of Land parliaments who, however, had no vote in federalism reform (Lübecker Erklärung 2003).



Stoiber was the opposition candidate for the office of the Chancellor in the 2002 national elections, and Müntefering became chairman of the Social Democratic party in early 2004. In combination, they reflected the highest possible level of political commitment to federalism reform.



These documents are available at: < 20Dokumente/6.2_20Kommissions-Drucksachen/index,templateId=renderUnterseiteKomplett.html>


In the meantime, all verbatim or summary records of all plenary, working group and project group meetings and all documents submitted to the Commission have been made available in a book publication that also includes a CD with search functions (Bundestag/Bundesrat 2005). All references to documents in the present article relate to this publication.



By discipline, the group of experts included eight professors of constitutional law, two of economics, and two of political science.


In post-mortem interviews, it was suggested by some of the Länder civil servants that the involvement of experts had been accepted primarily for the benefit of federal MPs who, unlike governments at both levels, could not draw on their own first hand experience in decades of federal-Länder negotiations. In any case, the explicit request for expert opinions was considerably reduced after the first few months of general discussion.



Stoiber could rely on the seasoned professionals of his Bavarian state chancellery; Müntefering had assembled a small staff headed by a former Land state secretary and including junior civil servants on secondment from federal ministries.



To be fair, the proposed solution is a bit more complicated. It does allow regulations of administrative procedures, provided the Länder are allowed to deviate from them, and it re-installs the veto if such changes are to be excluded.



BVerfG 1 BvF 1/01, July 17, 2002.



Admittedly, separating substance and procedure may be difficult in some cases - which may explain the preference of federal ministries for combining substantive and administrative regulations.



The government had tried to achieve more federal control over the implementation of tax laws, greater independence from Länder control in the conduct of EU-level negotiations, and the acceptance of burden sharing when Länder were responsible for violations of EU directives or of the Stability Pact.



Coalition MPs were dismayed when, during the summer, the Chancellor's Office invited the ministries to explore their options in case federalism reform should fail (interview, January 21, 2005).



BVerfG 2 BvF 2/02.



The intention was to accelerate the entry of young academics into autonomous teaching positions by displacing the requirement of a second (Habilitation) thesis.



In the end, they agreed to a few conversions into exclusive federal competencies. In a larger number of cases, however, they were only willing to exempt concurrent competencies from the operation of the necessity clause which, unlike the transfer into exclusive competencies, did not have the effect of excluding Land legislation even in areas not occupied by federal law. In those areas, however, where their demand for the transfer of legislative powers was ultimately refused by the federal government, they insisted on maintaining the status quo of the necessity clause.



In the same position paper, the federal government asked for the transfer of sixteen items from the status of concurrent into that of exclusive federal competence. Of these, only five survived the final negotiations.



In the Commission, autonomous taxing powers for the Länder were nevertheless (but unsuccessfully) promoted by some experts, federal MPs and representatives of (Western) Land parliaments. In its position paper for the final round of negotiations, the federal government also proposed to transfer legislative powers for those taxes that accrued entirely to the Länder. With a single exception regarding the setting of tax rates for real-estate sales, this offer was rejected.



Of the 26 items offered by the federal government in November, only 18 were still included in the draft proposal of the joint chairs of December 13.



Quisquilien ("peanuts"), a high ranking Land official called them in an interview.



In addition, there was also the more rational consideration that, given the drift of constitutional-court decisions, the status quo would be less comfortable for the federal government than for the Länder - so that time would work in their favor if federalism reform came on the agenda again.



In addition, pensions and unemployment insurance benefits received in East Germany are mostly financed by contributions collected in West Germany. Altogether, annual West-East transfers are estimated to have amounted to about four per cent of GDP over the last decade and a half.



However, the Eastern Länder overplayed their hand when, in the last week of negotiations, they demanded that the agreed-on transfers should also be enshrined in the constitution.



Its most serious congenital defect is that the criterion of redistribution is relative revenue intake, rather than relative wealth. Hence any changes in the tax system that affect Land revenues must entail difficult adjustments in the equalization system. In contrast, transfers based on GDP per capita could continue unchanged even if Länder had the authority to set their own tax rates (Homburg 2004).



These protests were articulated most forcefully by the Prime Minister of Saxony, Kurt Biedenkopf, who in his former West German career had been professor of competition law as well as general secretary of the CDU.



To the dismay of some prime ministers, and in spite of their professed preferences for "competitive federalism," Germany's centralized industrial associations also opposed the decentralization of economic and labor market regulations.



The German constitution does not allow discretionary dissolutions of the Bundestag, either by the government or by parliamentary majorities. Even after a vote of non-confidence, the decision lies with the Federal President and, ultimately, with the Constitutional Court.



In the end, it is true, the fact that current federal interventions in education had been shaped by social-democratic preferences for all-day schools and free university education, did contribute to the ire of CDU/CSU prime ministers.



It was first introduced as a dissenting vote the report of a parliamentary commission of inquiry on constitutional reform (Heinsen 1977). The proposal would have applied to the full range of federal concurrent legislation; and it would have allowed the Bundestag to veto deviating Land legislation before it would become effective.



Essentially, there were three interdependent issues to be resolved: (1) Should opt-outs be allowed in the full range of concurrent federal legislation, or should they be confined to a narrowly defined subset? (2) Should opt-outs be construed as an unconditional right of Land parliaments (whose exercise could not be affected by subsequent national legislation), or should they be conditioned and constrained by considerations of national interest? (3) In the latter case, finally, should these constraints be applied by the Constitutional Court on the basis of pre-defined legal criteria, or should they be left to the political judgment of the national legislature (the Bundestag, acting with the agreement of the Bundesrat)?



It was always considered self-evident, by all sides in the Commission, that legislation allowing Länder opt-outs would be exempted from the necessity clause of Art. 72 para. 2, Basic Law.



In place of constitutional options, the government suggested that individual federal statutes could allow for opt-outs - provided that such statutes would then be exempted from the "necessity clause" of Art. 72 para. 2 Basic Law.


In interviews after the failure of the Commission, it also appeared that, if pressed, the federal government might have accepted constitutional opt-outs in those narrowly defined areas where its offer of a complete transfer of legislative powers had not been taken up by a sufficient majority of the Länder.



Copyright © 2005 Fritz W. Scharpf

No part of this publication may be reproduced or transmitted without permission in writing from the author.

Jegliche Vervielfältigung und Verbreitung, auch auszugsweise, bedarf der Zustimmung des Autors.

MPI für Gesellschaftsforschung, Paulstr. 3, 50676 Köln, Germany





MPIfG:  MPIfG Working Paper 05/8

[Zuletzt geändert am 24.08.2007 11:15]