The Influence of Asymmetry on Territorial Dynamics in “Hybrid” Multinational States: The Case of Spain

Adam Holesch

Department of Political and Social Sciences Universitat Pompeu Fabra (UPF) - Barcelona

[email protected]

Preliminary draft – please do not quote

Paper presented in the Department of Political Science of the Central European University in Budapest 13. November 2014 2

1. Introduction:

Territorial dynamics in multinational states develop on two principal axes, the centralization – decentralization axis and the symmetry - asymmetry axis, with asymmetry being an important part of the decentralisation process. Questions of decentralization are in vogue and have become a well-analyzed topic. Not the same can be said about the asymmetric solutions, which stay a barely inquired subject. This gap is surprising due to two facts. First, all multinational states have some asymmetrical aspects. Second, even if institutionally applied, asymmetrical solutions do not really match the preferences of the citizens. For example in 2010 in the Spanish “State od Autonomies” the majority of Spaniards (67,3%) was against institutional asymmetry.1 Nevertheless, this analysis will show, that even if not congrunet with the preferences of its citizens, asymmetry become an essential factor of the political decentralisation and federalisation process in Spain. The research question to be asked is the following. How does asymmetry influence the territorial dynamics in the Spanish “State of Autonomies”? I will use the case study of Spain for reasons of expositional simplicity. It presents the advantage of undergoing a constant institutional remodeling in form of reforms of the “Statutes of Autonomy” of the AC. The discussion how far the multinational Spain could be interpreted as federal is far from being closed.2 But because we deal in this

1 Surprising is that also over 54,2 % of the Catalans and 48,3% of the Basques were against the fact that the Autonomous Communities (AC) could provide different services in each AC. (CIS 2011) 2 The degree to which Spain is a federal state continues to be an object of debate and discussion. On the one hand, the high level of autonomy or self-rule of the seventeen Spanish AC supports a federal interpretation. For Riker a federal state is to be found where “(1) two levels of government rule the same land and people, (2) each level has at least one area of action in which it is autonomous, and (3) there is some [constitutional] guarantee . . . of the autonomy of each government in its own sphere” (Riker 1964:11). According to Riker’s definition, Spain could be classified as federal. Watts (2008) argues with similar arguments that Spain has structural attributes typical for a federal political system like two levels of government, constitutional distribution of legislative and executive authority and allocation of revenue resources, a Constitutional Court, which arbitrates disagreements, and institutions that can ease intergovernmental cooperation. But finally Watts calls Spain not a full federation, but a federation in praxis. Many scholars like Burgess (2006), Rodden (2006) or Moreno (2007) use a similar approach. Agranoff (1996) argues that Spain is a unitary system that appears to be in the process of becoming federal. On the other hand, there are numerous academics like Elazar (1987), Filippov et. (2004), Requejo (2005) or Nagel (2010) who doubt or refuse to define Spain as federal, because of the following institutional features: The AC are barely involved with the institutions of the central state. The Spanish upper house, the Senate, in particular, is no chamber of the ACs, as three quarters of the senators represent administrative sub-units of the central state, e.g. the 50 provincias which are also the electoral districts of the lower house. Besides that, the Senate has not many powers. In addition, the provincias depend heavily on the centre, while eluding the AC. Coming back to Riker, the AC (besides Basque Country and Navarre) apparently do not have any area of action, where they act autonomous, because the central control have been maintained and eventually re-established by using framework legislation. In the Catalan case some central state politicians threaten with the

3 article with quasi-federal dynamics its definition is of big importance. We claim, that the official term “State of Autonomies” expresses to some extent the “hybrid” aspect of Spain on the unitary-federal axis, which we will define as a “hybrid” multinational state. We will position this paper in the strand of historical institutionalism (HI), which see institutions as continuities and puts the emphasis on path-dependency. The HI- approach allows us to connect questions dealing with territorial dynamics with the literature on the constitutional origins of Spanish state. The HI approach will be mixed with the actor-centered institutionalism approach developed by Mayntz/Scharpf (1995) and Scharpf (1997). It claims that political decisions are not made in a political vacuum, but between different actors acting in different institutional settings. For that reason in order to analyze the political decisions we will distinguish the type of actors, the positions of the actors and the institutional settings they work in. With that above all Scharpf (1997) showed that what really matters in the analysis of political decisions are the institutions as actors and not the social or cultural factors or informal rules.3 The mixing of these two approaches will allow us for simple game-theoretical modeling. Even if in its essence this paper is of a qualitative nature, we would like to define our variables. The dependent variable is the dynamics of the development of the Spanish “State of Autonomies”, in particular its political decentralisation/federalisation process. As independent variable I will use four variations of asymmetry, which will be developed in this paper: de facto asymmetry, institutional de jure asymmetry, symbolic de jure asymmetry and ultimate de jure asymmetry.4 The HI-methodology binds us to a chronological approach. After explaining the concepts used and presenting a short literature review, we will analyze how the actors bargained aspects of asymmetry and decentralisation in the constitution. Then, we will concentrate on territorial dynamics identifying two phases: first, the

withdrawal of autonomy applying the Art.155 of the Spanish Constitution. If that would be the case, following our conceptual definitions, Spain would qualify as a unitary and not a federal state. Besides that, the AC do not have any stake in constitutional amendments nor do they have direct influence upon the election of the judges of the Spanish constitutional court. Moreover, terms like federalism or federation do not appear in the Spanish constitution. If both terms appear in the Spanish debate, then usually they mean a future reform suggestion. 3 However they can strongly influence the development of the institutions and the position of the actors. 4 We are aware that there are also other variables influencing decentralisation like questions of efficiency or democratization.

4

“constitutionally conform accommodations of asymmetry” and second, the “constitution-destabilizing development of asymmetry”. The latter development reduced the territorial dynamics to a game between the central state and Catalonia, which we will explain with a simple analytical game-theoretical model. When talking about institutional change, we have to define clearly the relevant institutions.5 In the Spanish “State of Autonomies” the territorial dynamics did not take place inside the Constitution6, but within the “Statutes of Autonomy”. Starting from this point this analysis claims the following: First, that the permanent reform of the “Statute of Autonomies” challenges the view that the Spanish constitution established a path-dependency from which the actors could difficulty deviate. Second, that the alleged growth of the asymmetrical demands by the Minority Units brought this reform process to an end, because no reform or replacement has been any more possible. After exhausting the possibilities at the “Statute”-level, the requests have to be solved at the highest constitutional level. Third, that the crucial actor of the territorial dynamics has been the central state on two levels. Once in form of the two statewide parties (SWP): Popular Party (PP) and the Socialist Party (PSOE), and second as the sum of the central state institutions and the regional Majority Units, which are clearly opposed to the two Minority Units. Without the approval of these two parties, the Autonomous Communities (AC) in general and the Minority Units in particular have no leverage for a possible constitutional reform.7 Seen from the game-theoretical perspective, it is because of the missing threat of the Minority Units that the “ultimate asymmetry” game can find only one equilibrium: the Status Quo.

5 We define institutions as architecture and as rules that determine opportunities and incentives for behaviour. 6 The constitution has been resistant to change with two small exceptions in 1992 and 2011. The first reform was to adapt the voting rights of foreigners from the European Union (EU) to the Maastricht Treaty. The second reform planed to establish strict budget rules. Both are strongly influenced by exogenous factors: the EU-Treaties or the political pressure of the EU. 7 There is however one exception, when the minority government of PP or PSOE needs support from the NSWP parties in the government.

5

2. Concepts

2.1. Majority Units (MaU) and Minority Units (MiU)

Since the 19th century, the widely ideal case of liberal thinkers was that every state should be a one-nation state. However, there have always been multinational8 states, which included more than one demos. Within these states one demos formed a majority, while the other(s) a minority. They often identified themselves as nations. Kymlicka defines minority or sub-state nations as “nations, which do not currently have a state in which they are a majority, but which may have had such a state in the past, or which may have sought such a state.” (Kymlicka 2001: 23) These “minority” nations ask for some forms of autonomy or self-government to guarantee their survival as a distinct society within the state ruled by the majority nation. Kymlicka connects these minority nations to territorial units, distinguishing between Nationality-Based Units (NBU) and Regional-Based Units (RBU). The NBUs are the Autonomic Communities or “Lands”, where the minority nations form a majority. The RBUs form its own sub-state unit but are members of the majority nation. I will take this taxonomy one step further and reduce it to two actors: to the Majority Units (MaU), which represent the majority nation be it in form of the central state or the Regional-Based Units and the Minority Units (MiU), which represent the Nationality- Based Units. In the Spanish “State of the Autonomies” Catalonia and the Basque Country are the two ACs, which we could identify as Minority Units. In 2013 in both AC around half of the population defined its “regional” identification as a “national” one, and the AC as a nation (CIS 2014). In both there have been a permanent pressure for different kind of asymmetrical recognition. Taking this “national” identification as a valid measurement, we discover that there has been a big distance to the next AC, Galicia. In the last decades in Galicia only an average of 15 % has been claiming that Galicia is a nation. For that reason Galicia and the others AC will be from now on identified as the regional-based units (RBUs) of the Majority Unit (MaU). This analysis will show, that within the territorial dynamics of political decentralisation/federalisation,

8 In this analysis, I will use the term multinational, even if many authors use the term “plurinational” which has just a slightly different meaning. Some authors prefer the term ‘plurinational’, because the prefix ‘pluri’ helpfully describes cases of ‘not one’; that is, it covers both ‘two’ and ‘more’ (McGarry 2007).

6 within possible state-destabilizing critical junctures, both the central state and the RBUs act as one player. At the same time, crucial for the understanding of the territorial dynamics will be the fact, that not only Minority Units, but also the Majority Units claim some historical rights as arguments in favor of more asymmetry or resymmetrisation.

2.2. Asymmetry

The first author to theorize asymmetry was Tarlton (1965), for whom asymmetry “expresses the extent to which component states do not share […] common features”. Tarlton and later Duchacek (1970) showed, that some degree of asymmetry is a feature of all compound units in federal political systems and agreements, even if by the wording of the constitution they should be equal. These degrees can be expressed by structural conditions, which include geographic, cultural, social, and economic factors, which are commonly termed, leaning on Elazar (1987), as “de facto” asymmetries. Duchacek (1970: 280) called these “de facto” asymmetries “power ingredients”.9 They have to be distinguished from de jure asymmetries, which deal with the authority which a constitution in a federal state (or a law in unitary state) assigns to a regional unit or that this regional unit enjoys by law or in practice in regard either to self-rule (in regional communities) or shared-rule (at the national level) or both (Agranoff 1999, Watts 1999).10 De jure asymmetry arises usually due to deeply rooted historical factors, being in place, when these federations were formed (Watts 1999). The emphasis of the historical difference leads us to the normative aspects of asymmetry emphasized by

9 A different typology offer Burgess/Gress (1999), who besides of the de facto factors identifies also some socio-economic and cultural-ideological pre-conditions of asymmetry: Political Cultures and traditions, social cleavages, territoriality, socio-economic factors, demographic patterns. Burgess/Gress (1999:51) claim that it is impossible to translate asymmetrical perceptions into real power relations correctly. With that they in some parts follow Duchacek, who claimed that the constituent state units can interpret “the objective facts of disparity of power either correctly or incorrectly” (Duchacek 1970:282) Agranoff (1999), identifies conditions and outcomes of asymmetry. The conditions of asymmetry relate to the de facto factors like wealth, ethnicity, why the outcomes are aspects of how good these conditions are reflected in the federal system. 10 Bauböck (2002:3) distinguishes between three types of de jure asymmetry: a horizontal relation between the powers of different constituent units; a vertical comparison of powers between institutions of federal and of constituent unit government; and the consensual arrangement between different institutions at the federal level, such as a symmetric distribution of powers between the two chambers of a bicameral parliament.

7 scholars like Keating (2001), who states that asymmetries reflect historic rights and traditions, which should be limited only to the communities which claim them. These historic differences can be accommodated through institutional and/or symbolic de jure recognition. The institutional recognition usually concentrates on societal aspects like a minority culture in general or minority language in particular. Also some kind of an historical fiscal deal is usually included. The request for symbolic de jure asymmetry concentrates on the constitutional recognition of a Minority Unit as a nation. Usually, the Minority Unit feels not fully accommodated, until both aspects are included into the constitution of a multinational state. If this recognition is refused by the central state, then the Minority Unit could ask for an “ultimate asymmetry”, meaning a constitutional recognition of a right for self-determination. Once introduced and constitutionally recognized, this kind of “ultimate asymmetry” can lead to the break of a state.11 Because the “ultimate asymmetry” has been seriously requested only by the Minority Units, it brings the process of political decentralization to its limits. Taking all these definitions of asymmetry into account, I would like to offer a new taxonomy of asymmetries in a multinational state. We should distinguish between: a) De facto asymmetry, which includes differences in territorial and economic size. This de facto asymmetry is closely related to Duchasek’s “power ingredients” that usually establish a power structure, which the national and subnational actors are aware of and within which these actors act. b) Institutional de jure asymmetry, which usually relies on historical differences and can contain the recognition of a minority language or a different financial system (institutional recognition) c) Symbolic de jure asymmetry, which relies on the “need of recognition”, usually of the status of a Minority Unit as a “nation” (symbolic recognition) d) Ultimate de jure asymmetry, which settles down the request of the Minority Unit to recognize constitutionally its right of self-determination. The area of asymmetry as an important factor of stability in multinational states has not attracted much academic attention after Tarlton’s (1965) pioneering work. There are however exceptions. Agranoff (1999), Palermo et al. (2007) and Watts (1999; 2005) analyzed asymmetry from the angle of federalism studies, Ghai (2001),

11 The break can also take place without the recognition of the right for self-determination. It can also take place as an expulsion (Singapore).

8

McGarry (2007), McGarry/O’Leary (2009), Weller/Nobbs (2010) from the viewpoint of ethnic conflict. Also political theorists like Bauböck (2002) or von Beyme (2005) published on this topic. Some scholars intended to analyze asymmetry in its dynamic form as a process. Zuber (2011) claims that asymmetric solutions are usually applied after the claims of NBUs for an upgrade of power. Other academics discover “spill- over” or “catching-up” responses to asymmetrical solutions (Moreno 2001, Giordano and Roller 2004, Watts 2008, Hombrado 2011), or destabilizing effects of the “resymmetrisation” attempts (Maiz et al. 2010; Requejo and Nagel 2013). The concept of resymmetrisation has been seen as an instrument of the central government to “water down the distinctive institutional position of ethno-national communities” (Agranoff 1999b: 107–112). Requejo and Nagel (2013) reverse the orthodox connection between decentralization and asymmetry, which claims that the resymmetrisation process is related to a process of centralization and show that resymmetrisation, can also take the form of an increased degree of decentralization.

3. The influence of asymmetry - Territorial statics

Following the approach of historical institutionalism we will distinguish in this chapter between the origins and the evolution of multinational states. For that reason we will differentiate between the static approach in form of a constitutional setting of the path dependence rules (part 3) and the dynamic development and changes within this path (part 4). The dynamic part itself will be divided into two parts, the “constitutionally conform accommodation of asymmetry” (part 4.2) and the “constitution destabilizing development of asymmetry” (part 4.3) demands.

3.1. The Spanish Constitution – development of path-dependence rules?

A Constitution usually sets the formal rules of the game, which regulate power relations across territories. They typically establish some kind of stability through impeding dramatic shifts in the distribution of authority among different levels of government. In the Spanish case, the preparation of the constitution was lead in the atmosphere of political optimism, but also of fear and terror. The political optimism was due to the possible democratization of the country, the fear due to the imaginable return of the civil war and the terror with the state-destabilizing activities of the Basque terror organization ETA. There were numerous cleavages, which had to be

9 solved within the constitution: The political cleavage between democratic and authoritarian options, the religious cleavage between the Church and the advocates of a laic state, the institutional cleavage between the supporters of monarchy and the supporters of republicanism or the socio-economic cleavage between the company friendly Right and the Union friendly Left. There was also the territorial cleavage between the periphery and the centre and the national cleavage between members of the Castillan majority and the Minority Units Basque Country and Catalonia. As we will see especially the territorial and national cleavage, both topic of this analysis, has been unsolved.

3.1.1. The Actors and the Satisfaction of its Preferences

Already before the constitution talks the first Suarez government (1977-1979) established two pre-autonomies (Catalonia and Basque Country) and then extended this solution to the rest of Spain. In this process most of the administrative frontiers (14 with the exception of Cantabria, La Rioja and Madrid) of the future “State of Autonomies” were drawn. The formation of the pre-autonomies had a big impact on the development of the state. Even if there has been no officially recognized influence of its administrative form on the creation of the Spanish constitution, it implied that decentralization would be applied to the whole territory (Aja 2003). In order to understand the impact of the symmetry/asymmetry axis on the centralisation/decentralisation axis, we have to understand how the preferences of the actors influenced the final form of the constitution.12 It is widely known, that the decisive organ to define central points of the Constitution was the Constitutional Council (Ponencia constitutional). Its seven members represented the principal parliamentary groups, voted in the 1977 elections. There were three members from the governing UCD (Unión de Centro Democrático/J.M. Pérez Llorca, M. Herrero de Miñon, G. Cisneros) one from PSOE (Partido Socialista Obrero Español/G. Peces- Barba), one from PCE (Partido Comunista Español/ J. Solé-Tura), one of AP (Alianza Popular/Manuel Fraga) and one from Pacte Democràtic per Catalunya (PDC/M. Roca

12 In this point we will rely on the analysis of Colomer (1995). We should emphasize that the relevant actors has been the political parties, and not an assembly of representents of Majority and Minority Units.

10

Junyent). With that just the last mentioned of the seven members represented a region (Catalonia).13 Following Colomer (1995) during the discussion about the territorial future of the Spain there were generally three issues to be solved. First, the maintenance or abolishment of the provinces, second some kind of administrative decentralisation or even federalisation following a general formula and third a particular autonomy, above all for Catalonia and the Basque Country. During the discussions the parties had the following preferences. The AP wanted to maintain the provinces and favored some kind of administrative decentralisation, but not for all regions. The PDC wanted decentralisation, but with special level of asymmetry for Catalonia and the Basque County and the abolishment of the provinces. For both parties (AP and PDC) the voting on the territorial issue was their first constitutional preference.14 Among the parties that gave only secondary importance to the territorial vote, the more confusing was the position of the UCD, not only because of their endogenous conflicts. The ruling party wanted a kind of a special formula for Catalonia, the Basque Country and Galicia, but because of the Andalusian pressure, the UCD changed its preferences and has chosen to promote some kind of a general administrative decentralisation instead. The PSOE and PCE were due to their pact within the Franco opposition with the parties of the Minority Units not opposed to some kind of a special autonomy. Formally their platforms even advocated the right of self-determination, even if not secession. Nevertheless some kind of a general decentralisation and federalisation formula was closer to its party identity based on solidarity. Due to the change from sincere preferences to sophisticated voting, not only elements of asymmetry in form of a particular autonomy for Catalonia and the Basque Country but also a general decentralisation formula have been adopted. Moreover, the provinces have been maintained. (Table 1) This quite paradoxical and contradictory result had a big impact on the development of the territorial dynamics since.

13 Even if also the PCE participant, J. Solé-Tura was Catalan. 14 For other parties the decentralisation issue have been the second or even the third preference. For UCD the maintaining of the monarchy and the framing of the principles of free enterprise was more important. The PSOE and the PCE developed more intense preferences regarding the constitutional recognition of civil rights and socio-economic issues.

11

Table 1: Voting about the territorial future of Spain

Sincere preferences Sophisticated voting Provinces/ Particular General Provinces/ Adm. Particular General Adm. Autonomy formula Decentral. Autonomy formula Decentral. AP-1 YES NO NO YES NO NO UCD-3 YES NO YES YES YES YES

PDC-1 NO YES NO YES YES YES PSOE-1 NO YES YES YES YES YES PCE-1 NO YES YES YES YES YES Result YES 4:3 NO 4:3 YES 5:2 YES 7:0 YES 6:1 YES 6:1 Source: Colomer (1995: 97)

When we consider the satisfaction of the actors involved in this bargaining, we can see that all actors besides the Catalan representative satisfied two of their three possible preferences. The PDC got what they wanted in questions of an asymmetric particular autonomy, but had to change its preferences on two other aspects, among others the contradictory general formula for decentralisation. Opposed to the Catalans, the non-present Basque Nationalists of the PNV15 did not accept any formula, which would introduce a perspective of symmetrical competences among all the future Autonomous Communities. The Basques insisted on asymmetry, which was part of the Basque self-understanding. Their major objective was the introduction of “the historical right of the old lands” into the Constitution.16

3.1.2. The Influence Of Historic Differences On Constitutional Asymmetry

The Spanish Constitution recognized the following “Historical Differences” (hechos diferenciales) in the constitution: in Article 3 the question of language, in Article 141.4 the subject of the islands “cabildos”, in Article 149.1.8 the civil law

15 The PNV officially did not qualify to the Ponencia due to the weak representation in the national elections, but was together with PDC in favour of asymmetric deals. 16 Many authors see the Basque opposition to the constitution as a tactic in order to get their first preference: a special fiscal treatment in form of the reestablishment of the fueros. Unlikely Navarre, which collaborated with Franco during the military uprising, only one of the three Basque Provinces took Franco’s side (Álava). The other two (Biscay and Gipuzkoa) defended the Republic and were punished by Franco with the suspension of their special fiscal regime (which in the Basque case weren’t exactly the fueros). Paradoxically, on the edge of democracy, not Franco collaborates, but Franco adversaries had to tremble if they get back their rights lost after the Civil War. This confirms that the Spanish way to democracy, at least at the beginning was more a slow transition than a “pacted rupture. “

12

(legislación civil), be it foral or special and in article 149.1.29 the issue of the autonomic police. These historical differences were “translated” in the following manner into the historic-geographical panorama of the Spanish “State of Autonomies”. For Basque Country: the historical territories, language, foral civil law, autonomic police and own fiscal rules (concierto fiscal). In Catalonia: the language, special civil law and autonomic police. In Galicia: Language and own foral civil law. In Navarra: own fiscal rules (concierto foral), foral civil law, autonomic police and Basque language, where Basque-speaking minority. In the Canary Islands: Cabildos and special fiscal regime. In the Balearic Islands: language, Consejos Insulares, special civil law. In the Valencian Community: language and civil law and finally in Aragon: civil law.17 The most important source for initial asymmetry was to establish that statutes could rule different issues. They gave the possibility to delegate powers by the central law with the exception of foral laws and islands. In the constitution all asymmetries have been open for all due to possible opt-ins for example when talking about the title of “nationality” or police forces. Also exceptions were included. Very important was the “Disposition Adicional Primera”, which settled down the historical rights of the foral territories (three Basque Provinces and Navarra) and the “Disposition Adicional Tercera”, which introduced fiscal and economic regulation for the Canary Islands. Only within the “Additional dispositions” specific AC were mentioned. This quite long catalog of asymmetric exceptions shows that the asymmetry in Spain was introduced due to the historical particularities of different regions18, and not due to the

Basque or Catalan ethnic nationalism.

3.1.2.1. Asymmetric way to autonomy

Considering the way to autonomy the constitution established two ways with different

17 There is also constitutionally established asymmetry for shared rule like senators of the Spanish Islands, or special treatment for the Autonomic Cities Ceuta and Melilla. 18 Andalucía, Aragon, Asturias, the Balearic Islands, Galicia, Navarre or Valencia had always a different identity than the Castilian one. Also the geographic distant Canary Islands always felt a kind of estrangement from the center. During the Habsburg reign many of these regions (Aragon, Basque Country, Catalonia, Navarre and Valencia) had the opportunity of self-rule or a special taxation system. What distinguish the two NBUs from the RBUs are two facts, the first being earlier economic differences. Catalonia and the Basque Country were since the 19th century economic centers and their had to act against a poorer agricultural core, consisting of Castile, Andalucía and Estremadura, which exercised the political control. Second, a further political difference was the existence of a politically organized national movement in the NBUs since the end of the 19th century.

13 requirements, which the different nationalities or regions could use. The “extraordinary” track (Art.151) was foreseen for two types of regions, first the historical nationalities which had realized or planed a referendum during the Second Republic (Catalonia, Basque Country or Galicia). Second, this track was also open to regions that show a strong will of “self-rule”, expressed through a majority of the municipalities and the ratification of the decision through a referendum. The “normal” way to autonomy was through Art.143, which had to be ratified by smaller majorities - for example it did not have to pass a referendum.19 Even if the constitution established the possible way, the development seemed initially open. The Constitution could have facilitated a very asymmetrical decentralization, during which the territories could have selected various paths towards autonomy (or none). This development sets a puzzle if the constitution really established a path-dependency. The decisions made at the Ponencia Constitutional level with a concurrent impact on the quite contradicting “general formula” and “particular autonomy” aspects, but also the two tracks to autonomy show that the established “path-dependency” consisted in allowing new decentralisation claims by nearly all actors. Both, the statewide parties SWP and the regional non-statewide parties (NSWP) should use that uncertainty in the decentralisation/asymmetry process since.

3.1.2.2. Fiscal institutional de jure asymmetry

The Spanish “State of Autonomies” distinguishes between a dual taxation system, divided into a “special regime” and a “common regime”. This fiscal asymmetry is based on exceptions made due to the historical fueros.20 The financial concessions to Navarre have been always in vigor, with that the Constitution in reality only confirmed to respect them. The Basque fueros were abolished in 1876, but already in 1877 the central state offered the three Basque Provinces a special financial deal, allowing them to collect their own taxes in form of a concierto economico. After the Civil War, the dictator Franco maintained two of these exceptions in the cases of Navarra and Alava, but suspended it for Biscay and Gipuzkoa, which he considered as rebel provinces.

19 There is also the way to Autonomy through the Art.144 for the Autonomous Cities Ceuta and Melilla. 20 Fueros were the charter of rights, which regulated some aspects of local governance.

14

The disposición adicional primera of the Constitution has become an essential part of the Basque Statute of Guernica, which also incorporated the concerts of Guipuzcoa and Vizcaya. The Spanish government proclaimed in 1981 that the Basque Country could maintain, establish and regulate its own fiscal system within the territory of the AC. The Economical Agreement (Acuerdo Economico) with Navarre was established in 1982. Both AC are with that part of the “special regime” and can collect all taxes except the ones on petroleum products, tobacco and the custom duties. They submit a so-called cupo, a formula-based amount that is negotiated with the central government to compensate the center for their AC’s share of national expenses. There are also some minor tax concessions for the Canaries (harbour and petroleum taxes are retained, and the Canaries are not subject to the EU value added tax). The other AC operate in some form under the “common regime” system. They get their funding from central taxes as well as through tax sharing and central state grants. In the discussion on asymmetry many authors claim that Catalonia received some kind of a greater regulatory and operational control over its financial institutions. Catalonia collects for example a fee for supervision, and allocates these funds to economic development. It also has unique shared legislative controls over banking operations. Customarily, Catalonia has experienced higher tax sharing to a greater extent than do other “common regime” regions. Nevertheless these fiscal exceptions are not specific for Catalonia. They are based on the statute and all statutes of the “common regime” may have the same content, if claimed by the AC or ruled so by the central parliament. Therefore, these asymmetries are not necessarily limited to Catalonia, while the Basque or Navarra fiscal system is limited to the Basque Country or Navarra. The Catalan representents did not request a special financial deal during the negotiations of the constitution, mainly because their saw the Catalan particularity in the matters of culture and above all language. Nevertheless, how we will see in the following analysis, they changed their preferences over time.

3.1.2.3. Linguistic institutional de jure asymmetry

The Article 3 of the Constitution states that Castilian is the official language in all parts of Spain. Besides that, there are four co-official languages, which shall be official in the respective AC in accordance with their Statues. These languages are Catalan, Euskara (Basque language), Galician and Valencian (the last usually

15 considered derivatives of Catalan). In the Statute of the Balearic Islands Catalan has been defined as co-official. There are AC that grant some official status/co-officiallity in parts of their territory: Navarra for Basque, Catalonia for Aranese and Aragon for Lapao and Lapapyp. There is an essential difference between the fiscal and the linguistic asymmetry. Probably most of the rich ACs would ask for a financial “special regime”, but only the ACs with a different language can ask for an asymmetrical solution.21 With that we can distinguish between ACs with their own languages and the Castilian-speaking AC. The linguistic asymmetry is especially important in questions of education. Contrary to the Majority Units with a different language (RBU with historical particularities), above all Catalonia, but also the Basque Country uses teaching in their own language as part of their parallel nation-building project.

3.1.2.4. Symbolic de jure asymmetry

The Article 2 of the Spanish constitution deals with questions of recognition and symbolic de jure asymmetry. Its development was highly problematic. Colomer (1995) claims that finally a veto of the government and not the Ponencia Constitutional decided the outcome. At the beginning, especially the Basque Country pushed for a symbolic recognition as a “nation”, which was refused by the francoist actors like Alianza Popular and the Army. As a kind of settlement, the term “nacionalidades” should have been adopted for Catalonia, Galicia and the Basque Country. But even this compromise was after a short period too “risky” for the actors of the central state. During the negotiations the government handed in a hand-written sheet adding the phrases of “common and indivisible Fatherland” and “indissoluble unity of the Spanish nation” to the previous draft. The term “nacionalidades” was not removed, but framed in this in constitutional law nearly unique resilient formulation. It should exclude all future discussions. Nowadays every AC can define itself by Statute as a “nacionalidad” and several have done so, including Aragon and Andalusia.

21 The intent to define Andalusian as a vernacular language has failed.

16

4. The influence of asymmetry - Territorial dynamics

4.1. The Actors: Parties and its Preferences on the Decentralisation issue

In Chapter 3.1.1. we have already analyzed, which preferences the political parties had during the constitutional bargaining. Before starting with the analysis of the evolution of the territorial dynamics, we will examine how their preferences developed during the more than three decades of democracy. Many academics argue that in the Spanish case above all the rightist parties AP/PP express their preferences for centralisation and the PSOE has stronger preferences on the decentralisation axis. In order to check this assumption, we will use the Party Manifesto Data, analyzing the discourse of the statewide (SWP) and non-state wide parties (NSWP) on decentralisation. Even if the PMD has many limitations, it allows us for a solid longitudinal analysis. 22

Table 2: SWP parties + NSWP parties: Party manifesto

Party&Discourse:&Decentralisa1on&2&Centralisa1on& 30%#

25%#

PCE/IU# 20%# PSOE#

15%# AP/PP# CIU#

10%# PNV# ERC# 5%#

0%# 1977# 1979# 1982# 1986# 1989# 1993# 1996# 2000# 2004# 2008# 2012# Source: The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR), Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB)

The Party Manifesto Data shows that both statewide parties PSOE and PP have a similar discourse on decentralisation in their party manifesto. The PSOE expressed the strongest support for decentralisation in 1977, but subsequently, their discourse diminished and stabilized between 1% and 3% in 1986 – 2004. After a small recuperation in 2008 (4%) it followed the general negative trend of the statewide parties and diminished again. The AP/PP had a very stable positive discourse on

22 There is no similar data on asymmetry available.

17 decentralisation, close to 2% between 1977 and 2004. That number went down to less then 0,4% since 2008.23 Among the statewide parties the PCE/IU has been the strongest supporter of decentralisation using for it 8% of their party manifesto in 2000. But also the former communists reduced this discourse since 2004 dramatically. When we confront the data from statewide and nationalist parties, the proximity of the PSOE and AP/PP programs becomes even clearer. All three major nationalist parties (CiU, ERC, and PNV) use in average between 10% and 25% of their manifestos on questions of decentralisation, which differs strongly from the 0,4% up to nearly 5% of the statewide parties. In summary, we can say that the presumed dominance of the PSOE on aspects of decentralisation cannot be confirmed. In average, the PSOE lays just insignificantly over the AP/PP. Both parties seem to lose interest in decentralisation since 2008. When connecting it to the bargaining of the constitution, we can state that the two statewide parties seem do have adjusted their preferences. During the constitutional bargain the positions of the AP and PSOE were much more different. On the other hand, the difference of preferences between the SWP and NSWP persisted, making the positions of the actors in the decentralisation clear, on the one side the SWP and on the other side the NSWP.

4.2. Constitutionally conform accommodations of asymmetry

The territorial dynamics in the Spanish “State of Autonomies” are not solved through reforms of the constitution, but through the reform or replacement of the “Statutes of Autonomy”. These Statutes determine the contents of powers and usually encompass high thresholds for revision in the Spanish and regional parliament.24 Statute reforms are initiated and approved by the respective regional parliaments with a qualified majority; the final text is written and passed in Madrid. Then the text is ratified either by referendum (Catalonia, Andalusia, Basque Country, Galicia and Valencia) or by the regional parliaments. Considering the development of the Statutes reforms we identify five waves, which can be summarized into three stages.

23 We have to take into account that the AP majority opposed the constitution in 1978 because of the title VIII, which coped with decentralisation. Under Aznar the PP made the constitution (inclusive title VIII) into the main symbol of the Spanish nation. Moreover the Basque PP accepts and defends the Basque concierto. 24 The ruling of the central state through Organic Laws could be considered another instrument.

18

Table 3: Three stages of Statute reform in the Spanish “State of Autonomies” Autonomous 1st wave + 2nd wave 3rd wave + 4th wave 5th wave Community (AC) First Stage Second Stage Third Stage Basque Country 1979 2005 (failed) Catalonia 1979 2006 Galicia 1981 Andalucía 1981 2007 Asturias 1981/1991 1994/1999 Cantabria 1981/1991 1994/1998/2002 La Rioja 1982 1994/1999 Murcia 1982/1991 1994/1998 Valencia 1982/1991 1994 2006 Aragon 1982 1994 2007 Castile-La Mancha 1982/1991 1994 2011 (failed) Canarias 1982 4th wave – 1996 Navarra 1982 4th wave – 2001 Extremadura 1983/1991 1994/1999 2011 Islas Baleares 1983 1994/1999 2007 Madrid 1983/1991 1994/1998 2010 Castile -Leon 1983 1994/1999 2007 Source: own elaboration

The first stage can be established in 1979-1983. It took place nearly parallel with the launch of the constitution. The second stage, which I will call the resymmetrisation process, holds on approx. between 1990 and 1999, usually in order to upgrade the ACs established by article 143. The third stage can be observed again between the years 2003 - 2011.

4.2.1. The first stage 1977-1983 – The ambitions of the Majority Units

The first stage between 1977-1983 was strongly influenced by an asymmetric dynamic. The two ways to reach autonomy expressed not only the diversity of the Spanish regions, but also took into account their different will to reach “self-rule”. Moreover, it established most of the future rules of the game. But though the constitution wrote down a possible evolution of the territorial dynamics, their development seemed initially open. The special treatment given to the Minority Units Basque Country and Catalonia was from the beginning challenged by two Majority Units, Galicia and Andalusia.

19

Considering Galicia, the UCD government wanted the autonomic powers not to be established by a statute, but delegated by a state law - a mechanism also called “clausula competencial”. After one year of discussions that plan was given up. Galicia was finally treated like Catalonia and the Basque Country and derived its competencies directly from a “Statute of Autonomy”. In the case of Andalucía the government wanted the AC to use the “normal” way to autonomy. Andalusia should also receive just the lower level of competencies. But Andalusia made use of the Art. 151 and expressed by mobilizing all its social forces its strong “self-will” and finally obtained the maximum level of competencies in 1981.25 Also other Majority Units had its upgrade ambitions, which lead to the Autonomic Pacts (APA/Pactos autonomicos) and the projected LOAPA (Ley Orgánica de Armonización del Proceso Autonómico).26 The APA were based on the consensus of the two leading parties in the year 1981, UCD and PSOE. They decided to define the autonomic map, choosing the regions, which finally would become ACs. The APA maintained the two levels of competencies and decided that all the pendent Statutes would be ratified by the “normal” way.27 In 1983 all 17 statutes had been approved, seven of them got a higher level of competences (Basque Country, Catalonia, Galicia, Andalusia, Navarre, Valencian Community and Canarias) and the other initially got a lower level of competencies, susceptible to upgrade after 5 years. The decisive actor were the two leading SWP, which allowed with the APA an important and general resymmetrisation.

4.2.2. The second stage: “Resymmetrisation process”

After five years some AC with a lower level of competencies like Aragon, Castile and Leon, Baleares and Asturias asked for the reform of their statutes by using the article

25 Andalusia failed the first referendum (Almeria has not delivered) and was given a second chance by the central state, which adapted the rules. 26 LOAPA has foreseen a harmonization of the autonomy process, but was partly declared unconstitutional by the Constitutional Court (CC); CAT and BC submitted the suits. The CC emphasized that any state law should not limit the Statutes of the AC. With that the CC confirmed that there couldn’t be any state law between the constitution and the Statutes, which with that sentence got nearly a constitutional character (Aja 2003). However, there is the exception in form of the Organic Law LOFCA about the financiancion of the Autonomus Comunities (Ley Orgánica de financiación de las Comunidades Autónomas). 27 An exception was made for the Canary Islands and Valencian Community, which started the “faster way” following article 151 before the ratification of APA. But, in order to follow the new rules, it was finally decided that both would take the “normal” way, even if getting the highest level of competences.

20

148.2. The governing PSOE was reluctant to give up further powers; also the Minority Units governed at that time by regional NSWP (PNV/Basque Country and CIU/Catalonia) were against a further resymmetrisation in form of more “café para todos”. Finally it was again the two big parties, the PSOE and PP, who pushed through a resymmetrisation adjustment and signed new Autonomic Pacts in 1992. Following the Art. 150.2 of the Constitution, the PSOE government unilaterally upgraded the ten AC that had established their statutes by article 143 CE, which got nearly the same powers as the higher level AC of Art.151 for example in the field of education. Furthermore, the PSOE, even if it did not need the support, decided to include the biggest opposition party PP. Further Statute upgrades followed.28 In this stage of the process, the “benign” central state, represented through both SWP parties and mixed with a persistent preference of the regional Majority Units for resymmetrisation has been the decisive factor for the reduction of asymmetry. This resymmetrisation processes was accompanied by an another instrument - the additional framework legislation of the central state in form of Organic Laws, for example in the field of education. (Holesch/Nagel 2012) Education has been one of the public policies, which in Spain has been assigned to the Autonomous Communities (AC). This transfer of powers could be considered as proof of the strong “self-rule” of the AC, but a closer analysis shows that the powers over education have been more divided. Due to the state’s exclusive power to regulate the basic conditions guaranteeing the equality of all Spanish citizens, the important and final decisions are taken at the center through the framework legislation. Both, the resymmetrisation through the upgrading of the Majority Units and the important weakening of the whole decentralisation process through the framework legislation led to a request by the Minority Units for stronger asymmetry. This asymmetry could be only offered, when the ruling party at the center needed the peripheral votes of the Minority Units. This happened when the parties of the

28 In 1996 new Statute reforms came on the table, following the initiative of Aragon, an AC, which had started the autonomy on the “normal” way and changed its statutes in 1994 incorporating the additional powers approved by the Pactos Autonomicos of 1992. In that case the AC proposed the reform by resolution of the regional parliament and there was also a basic consensus between PP and PSOE in the Spanish parliament. Aragon in 1996, Castile - La Mancha in 1997, Murcia, Cantabria and Madrid in 1998 and Asturias, La Rioja, Islas Baleares, Castile and León and Extremadura in 1999, reformed in that way their Statutes. Further reforms concentrated on improving the already transferred competencies and the working of regional parliaments (Aja 2003: 91). The central state led by minority governments, be it the PSOE (93-96) or the PP (96-2000), allowed that resymmetrisation.

21

Minority Units (CiU and PNV) gave support to the minority governments of PSOE (1993/1996) and PP (1996-2000) and won the necessary leverage for further asymmetry requests, which both SWP finally delivered. Considering the development of territorial dynamics between 1978 – 2004 the following statements can be made: - Institutional fiscal de jure asymmetry: the fiscal decentralisation was based on bilateral negotiations, but all Autonomous Communities of the “common regime” got the same treatment. The distinction between a “special” and a “common regime” did not change. The Catalan government did not demand an upgrade to a “special” regime. - The institutional linguistic de jure asymmetry became important when the powers over education were transferred to all ACs in the 1990s. It gave the Minority Units the possibility to start an own nation-building project. The central government challenged this through a framework legislation, which defended the unity and equality of Spanish citizens. This asymmetry introduced a permanent moment of tension between the Majority and Minority Unit. - Symbolic de jure asymmetry, the term nationality has been watered down because of the symbolic upgrade of some of the Majority Units like Aragon as “nacionalidades”. This could be one of the reasons why the Minority Units asked for a constitutional upgrade as nation.29

4.2.3. The development of the ”power ingredients”

4.2.3.1. Population and territory

Considering the territorial divide, the biggest AC has been Castile-Leon with 94.225 km2. On the other side of the continuum we can find the Balearic Islands with 4.992 km2. The most relevant AC for this analysis differ strongly in their size. Andalusia (87.599 km2) is nearly three times as big as Catalonia (32.112 km2). Under the smallest AC we can find Madrid (8.028 km2) and the Basque Country (7.234 km2).

29 We should keep in mind, that a part of the Minority Unit always used the term “nation” in order to define itself.

22

Table 4: Asymmetric power ingredients: Territory and population 1981-2011 (without Ceuta and Melilla)

Territory (km2) Population (1981) % in 1981 Population (2011) % in 2011 SPAIN 505.990 37.636.201 100 46.651.079 100 Castile and Leon 94.224 2.582.043 6,7 2.540.188 5,4 Andalusia 87.599 6.429.151 17,1 8.371.270 17,9 Castile -La Mancha 79.461 1.647.876 4,4 2.106.331 4,5 Aragon 47.720 1.196.430 3,2 1.344.509 2,9 Extremadura 41.634 1.064.289 2,8 1.104.499 2,3 Catalonia 32.113 5.948.177 15,8 7.519.843 16,1 Galicia 29.575 2.809.201 7,5 2.772.928 5,9 Valencia 23.255 3.642.816 9,7 5.009.931 10,7 Murcia 11.314 953.852 2,5 1.462.128 3,1 Asturias 10.604 1.128.986 3,0 1.075.183 2,3 Navarre 10.391 508.679 1,3 640.129 1,4 Madrid 8.028 4.679.696 12,4 6.421.874 13,7 Canary Islands 7.447 1.364.616 3,6 2.082.655 4,4 Basque Country 7.234 2.139.860 5,7 2.185.393 4,7 Cantabria 5.321 512.579 1,4 592.542 1,3 La Rioja 5.045 254.201 0,7 321.173 0,7 Balearic Islands 4.992 655.134 1,7 1.100.503 2,3 Source: Own elaboration based on Instituto Nacional de Estadística (INE) and Ministerio de Economia y Hacienda

Considering the population the most populated AC is Andalusia with 8.3 million residents. The two identified Minority Units; Catalonia and the Basque Country differ in this aspect strongly. Catalonia is with over 7.5 million residents the second biggest AC, even if it has far less territory than for example the slightly more populated Andalusia. The Basque Country with it nearly 2.2 million residents is in the middle of the Spanish population ranking. Even if the Spanish population grew strongly between 1981 and 2011, the differences between the different AC did not change heavily.30 Catalonia’s share of the population in the Spanish territory grew by 0,3% and that of Basque Country went down by 1%. Surprisingly, the economic motors of the Spanish economy, Catalonia, Madrid and the Basque Country did not have an important rise of their population.

30 The biggest changes have been absorbed during different emigration waves between the 1950 and 1970s.

23

4.2.3.2. Economic Powers: GDP of the Autonomous Communities

Table 5: Asymmetric power ingredients: Spanish GDP by Autonomous Communities (AC) (without Ceuta and Melilla)

General GDP/Capita (1980) GDP/Capita (2012) Share of PIB (2012) SPAIN 100 100 100 Basque Country 130,5 134,8 6,2 Catalonia 119,9 118,5 18,9 Madrid 122,9 129,7 17,9 Andalusia 77 75,1 13,4 Castile and Leon 83 98,7 5,2 Castile -La Mancha 83 79,4 3,4 Aragon 95,3 111,2 3,2 Extremadura 51 67,9 1,6 Galicia 81,9 91,2 5,4 Valencia 106 87,4 9,5 Murcia 99,4 80,9 2,6 Asturias 82 93,6 2,1 Navarre 122,5 127,8 1,7 Canary Islands 109,7 84,9 3,9 Cantabria 96,2 97,3 1,2 La Rioja 99 113,0 0,8 Balearic Islands 124,7 105,8 2,5 Source: Own elaboration based on Instituto Nacional de Estadística (INE) and Ministerio de Economia y Hacienda

In terms of the shares of the region in the national PIB, 31 Catalonia leads with 18,9% followed by Madrid with 17,9% and Andalusia with 13,2%. Due to its not so big population, the richest AC, the Basque Country, shares only 6,2% of the national PIB. Within Spain income disparities have diminished due to EU structural funds, redistribution and central government investments. In 1980 there have been 5 AC over 115% of the Spanish average and 6 AC under 85%. In 2012 the differences diminished, only 4 AC lied over 115% and 5 AC under 85%. Above all Extremadura made an important rise from 51% to 67,9%. Nevertheless, there are winners and losers of the economic development. We can see that the two Autonomous Communities with a fiscal “special regime”, the Basque Country (130,5% to 134,8%) and Navarre (122,5% to 127,8%) increased they wealth compared to the Spanish mean. However, there are strong disparities within the “common regime”. While Madrid got richer (122,9% to 129,7%), Catalonia’s (119,9% to 118,5%) PIB/capita

31 The author is aware of the imperfections of this method. However, he uses it due to the wide availability of the GDP/capita or PIB index data.

24 diminished in the Spanish average. With that Catalonia, one of the most populated AC with the biggest share of the Spanish PIB percepts itself as a relative loser of the economic development in the democratic period. The discrepancy between both MiU (the Basque Country and Catalonia) and between Catalonia and the AC Madrid has grown and Catalonia blames it on its type of fiscal regime. That perception of the power ingredients is important, when dealing with the requests of the Catalans in the following chapter. It could be the most important explanation, why the Catalans changed their preferences asking for a Basque-like “special” financial regime.

4.3. Constitution-destabilizing influence of “ultimate asymmetry”

4.3.1. The Ibarretxe Plan – The Basque Country

It was not until the 2000s that especially two political demands of the Minority Units questioned the Statutes of Autonomies to such extent, that if ratified, they would imply not only a reform, but also the introduction of elements of “ultimate asymmetry” into the Constitution. The starting point was surely the Ibarretxe Plan (IP), proposed by the Basque Lehendakari Juan Jose Ibarretxe (PNV). It was announced in 2001 and approved by the Basque parliament in 2004. It requested the “ultimate asymmetry” in form of the right of the Basque Country for self- determination, which could end up in a free state – association. Besides that, the IP asked for the recognition of the symbolic de jure asymmetry, the recognition of the Basque nation. It asked for a kind of a double nationality with Basques and Spanish passports for every citizen of the Basque Country. The Ibarretxe Plan, even if formulated as a Statute reform, was seen by the central government be it in form of PP or PSOE as a plan, which aimed to break with the Spanish constitution. This kind of “ultimate asymmetry” was finally widely refused by most of the political actors, the governing PSOE and the opposition PP included.32 In the case of the Ibarretxe Plan the Spanish government did not react with resymmetrisation tactics, but with a complete refusal, above all because the “ultimate

32 After the prohibition of the Basque referendum on self-determination, which was announced by the Basque government after the refusal by the central state, the Basques government called out an election in order to get a stronger mandate, but finally lost power to a coalition of the two state-wide parties PSOE and PP.

25 asymmetry” of the Ibarretxe Plan was clearly constitution-destabilizing and could not be counterweighted by any resymmetrisation processes anymore.

Table 6: Types of Asymmetry in the “State of Autonomies” stressing Catalonia and Basque Country Asymmetry in Spain Kind De facto Institutional Asymmetry Symbolic Asymmetry Ultimate Asymmetry (de of Asymmetry: (de jure) ( de jure) jure) Asymmetry Power ingredients ACs involved All ACs AC with “historical Basque Country (BC) Basque Country (BC) and differences” and Catalonia (CAT) Catalonia (CAT)

Levels of Population Fiscal decentralization: Recognition as nation Constitutional recognition Horizontal Different level Different level of the right of self- Asymmetry of BC (strong) vs. Same level/refused determination Minority Economic Catalonia (weak) Units: the power BC and CAT (weak), Same level/refused Basque Same level Cultural decentralisation nationality not Country and BC, CAT, Same level enough also because BC and CAT (weak), Catalonia MAD - BC (strong)/Cat (strong) taken by f.e. Aragon refused by central state strong actors or Andalusia which through government (both vs. MaU weak Administrative/Political are MaU state-wide parties) , CC actors decentralisation (Constitutional Court), Similar level parliament and the population of the Majority Units Source: own collaboration

4.3.2 The Catalan “Statute” reform

Nearly parallel to the Ibarretxe Plan another important process started: the Catalan Statute reform. Looking for the support of Catalan voters, the than opposition party PSOE made them an electoral promise - a possible reform of the Catalan Statute and a federalization of Spain.33 After the PSOE victory in 2004, the Catalans started to prepare their new Statute. But they asked for more, than the PSOE expected. First, they requested a fiscal regime similar to the Basque and Navarra “special regime”.34 Besides a fiscal upgrade, Catalonia also asked for the recognition of the symbolic de jure asymmetry, the recognition of the Catalan nation.

33 The contents of the Basque and the Catalan plan were different. Moreover, the Ibarretxe Plan had no support of any SWP, not even in the Basque Country while the Catalan one had the support of the PSOE. 34 It was similar and not identical, because they suspected that the Constitution does not allow transferring the recognition of “historical rights” of the disposición adicional primera to other territories than Navarra and the three Basque provinces.

26

The Catalan referendum on the Statute reform was successful, 73,2 % of the voters voted in favor. On the other hand there was also a high abstention rate, with only 48,8% of the Catalans participating. With that only about 1/3 of the Catalans voted in favor, what showed, that the territorial upgrade was not a salient issue for the Catalans at that period. The nationalist mobilization started some years later for several reasons. First, already before the referendum, the PSOE was watering down the possible constitution-destabilizing parts of the Statute before the Spanish Parliament. Second, some of the Majority Units AC have contested the Catalan Statute and also the opposition party PP wanted it curtailed. The objections were raised against the symbolic de jure asymmetry (nation/cultural heritage) and the lack of a "solidarity between regions" principle in questions of fiscal and cultural decentralisation (education). When the PP failed dismissing the text due to a lacking majority in the Parliament, the PP and some socialists brought the Statute before the Constitutional Court. Third, at the same time, the central government used again the mechanisms of resymmetrisation. Between 2003 and 2007 the Statutes of Autonomy of Valencia, Catalonia, Andalusia, Aragon, Balearic Islands and Castile-Leon were reformed.35 Since Valencia was the first region to revise its Statute, its president Camps was keen to ensure that if in the subsequent reforms the Minority Units would surpass the level of competences attributed to Valencia, the Valencian Community could upgrade up to the same level. This was assured through a mechanism in the revised Valencian statute known as the ‘Camps’ clause’ or “the most favored AC” clause. After identifying parts of the Catalan Statute as constitution destabilizing, the central government (PSOE) started to cooperate with the regional branch of the PP, which governed in Valencia. The central state sought to advance the Valencian reform as a “ideal type” for a consensual model of institutional change, which accommodated the more federal project of the PSOE as well as the more unitary one presented by the PP.36

35 Madrid reformed its Statute in 2010 and Extremadura in 2011. Galicia and Canarias abandoned the project before sending it to Madrid, Castile - La Mancha (2011) abandoned its short before ratification. Everywhere where both major SWP agreed, statues were reformed. 36 At the same time some parts of the media presented the Valencian statute as the “good statute” in contrast with the “bad” ones like the Basque or the Catalan.

27

Fourth, in 2010 the Constitutional Court finally pronounced its sentence on the Catalan Statute. It interpreted some of its articles as unconstitutional. Some were declared null; others received compulsory interpretations by the court. The article 5 of the Catalan Statute about the historical rights was said to be not equivalent to the disposición adicional primera of the Constitution. With that it did not give the Catalan government the right to justify the same institutions like the Basque Country. Important was also the clear refusal of the Article 8, which dealt with the symbolic de jure asymmetry. Following the interpretation of the CC the Catalan statute cannot speak of a Catalan nation, but only of nacionalidad, which should not question the "indissoluble unity of the Spanish Nation".37 Summarizing the four points, the response of the central government to a constitution destabilizing Statute reform can be shortened as following. The key player is the central government, usually one of the two SWP parties. The central government can give the Minority Units some asymmetrical powers well knowing about its available mechanisms to block it. It has two possibilities to influence the asymmetry dynamics, one on the identity and the other on the institutional level. First, there are the resymmetrisation mechanisms for the constitutionally conform accommodations of asymmetry. In this case the common national identity Spanish/Castillan helps to resymmetrizese the “Autonomic State” through the regional Majority Units. Second, the Constitutional Court helps to refuse the constitution-destabilizing aspects of asymmetry with its strict interpretation of the constitution.

37 Moreover, the sentence said that the Statute couldn’t bind following the Spanish legislation state attributes like for example the justice system.

28

Table 7: The asymmetric development in the Spanish “State of Autonomies” 1. Phase 1978-2003 Central state (UCD or PP or PSOE) vs. Basque Country and Catalonia (more Majority Units (request of catch-up) asymmetry and decentralisation) Inside Constitution Inside Constitution 2. Phase 2004 – 2009/2010 Central state (PSOE + CC) vs. Basque Country Catalonia (symbolic + Majority Units (used by (Sovereignty/Association) financial asymmetry) the CS units of Outside Constitution Inside Constitution resymmetrisation) Inside Constitution 3. Phase 2010 – 2014 Central state: (PP + CC) vs. Basque Country: OUT Catalonia (referendum - Majority Units: OUT Sovereignty/Association) Outside Constitution Source: own elaboration

5. Game-theoretical interpretation of the game between the CS and Catalonia

The Basque and Catalan requests for the ”ultimate” asymmetry brought the development of the constitutionally conform accommodations of asymmetry to its end (Table 7). Like already shown in point 4, with the Basque Country (temporary) and the Majority Units permanently out of the process, the asymmetric territorial dynamics could be reduced to two actors. We find at central state level a majority PP government and in Catalonia a minority CiU government, an NSWP-party. The reduction to only two actors, allows us for a game-theoretical modelling. After the end of the process of constitutionally conform accommodations of asymmetry and the failure of the accommodation of constitution destabilizing aspects of asymmetry, the future of the Spanish “State of Autonomies” will be decided in an “Ultimate Asymmetry” game, 38 which could be seen as a logical consequence of the asymmetrical development in Spain.

38 Greif/Laitin (2004) claim that game theory does not capture fully the complexity of the interrelationships between individuals and the institutions influencing their behavior, but it provides a useful analytical tool for studying self-enforcing beliefs and behavior of institutions in a given situation. This is exactly the situation, which we can find in Spain since 2010.

29

5.1. The “ultimate asymmetry” game

In order to model the preferences of the actors, we have to reconstruct their development first. After the verdict of the Constitutional Court they changed rapidly. After 2010 the Catalan CiU government was not concentrating on the reconsideration of the symbolic de-facto asymmetry in form of “nation”, but was pushing principally the financial upgrade, if possible in form of the Basque “special regime”.39 In case that these bilateral negotiations would fail, the Catalans had to create a credible a threat. At this very start they threatened with some kind of a not nearly formulated federal incompliance, be it through creating an own tax collecting agency or some kind of nationalist mobilization. During two following two years the Catalan reform proposals were permanently refused by the central government, which mainly argued that Catalonia has already received a better financial treatment within the Statute Reform of 2006. The repeated refusal led to radicalization of the preferences of the Catalan actor. In the advanced 2012 elections, the secessionist party ERC had heavy wins on costs of the more moderated CiU. Also the civil society organized itself within the structures of the ANC (Catalan National Assembly). The political and civic forces mobilized parts of Catalan society in mass demonstrations during the Catalan national holiday: “La Diada” in three following years 2012-2014. Within that period Catalonia introduced the option of an “ultimate asymmetry”, the referendum, into its possible preferences set and enlarged with it its strategy space. Instead of confining itself to a choice among available strategies, Catalonia redefined from its point of view the rules of the whole game. From their point of view, it was partly due to the refusal of the other alternatives like the Statute reform or the fiscal “concert” and the impossibility of a reform of the Spanish Constitution. In a nutshell: Already from 2005 on, the Catalan governments wanted an institutional change (IC), which was widely supported by a wide part of the Catalan voters. Since 2011/2012 the referendum option was introduced, which was requested above all by the supporters of the nationalist parties, among them the secessionist Esquerra Republicana (ERC). The central government, which was not interested in any change of the institutional setting and its asymmetry, had as its preferences the Status Quo and Centralisation.

30

The biggest part of the PP voters (and also PSOE voters) was against any asymmetric accommodation of the Catalan preferences. At the same time, since 2010 big parts of the Spaniards showed a growing disappointment with the Spanish “State of Autonomies” and the preferences for the centralisation have grown massively. With that, in 2014 we find the following preferences of the actors:

REF: Referendum IC: Institutional Reform SQ: Status Quo C: Centralisation

Following that discussion the preferences of the Catalans were Referendum, followed by an Institutional Change, the Status Quo and Centralisation. Even if it questionable if “Referendum” has been really the first preference, its introduction should help to achieve a possible sub-optimal outcome, the “Institutional Change”. It is clear that besides that the Catalans preferred the “Status Quo” to “Centralisation”. The central state’s first preference was the “Status Quo”, followed by “Centralisation”. It could be questioned if these preferences could be reversed. But, in 2014 in the Spanish “State of Autonomies” there have been still a consensus between both SWP that defends the autonomic version of the state. The following preferences of the central state would be “Institutional Reform” and “Referendum”.

CS: SQ>C>IC>REF

CAT: REF>IC>SQ>C

31

Table 8: The “ultimate” asymmetry game

Catalonia

Cooperate Defect

Cooperate IC (2/3) REF (1/4)

Central State Defect C (3/1) SQ (4/2)

Source: own collaboration

In this game we find equilibrium when both actors defeat. The dominant strategy for CAT is defeat with REF/SQ, and for the central state C/SQ. With that the central state gets its first-order preference and the Catalonia the third-best. But why do we get this result? Above all, because the central state does not take the referendum “threat” seriously and considers it as “cheap talk”.40 This strongly influences the outcome. When Catalonia could make a credible threat than the available preferences set could change. In this case the centralisation option could be removed, and a new preference “Failed Relations” introduced. The term “Failed Relations” could consist of many elements. We can speak of “Failed Relations” for example if the central government would be not sure, if international partners would support Catalonia’s unilateral declaration of independence.

REF: Referendum IC: Institutional Reform SC: Status Quo FR: Failed Relations

If including a credible threat, the preferences of both actors would be the following: The preferences of Catalonia are “Referendum”, followed by “Institutional Change”,

40 We concentrate on the ‘threat” and not for example on the transition costs of the secession, because in the Spanish case the important player is the central state. Considering a possible change of the constitution be it through inclusion of elements of the right for self-determination he has a reputation for toughness. For that reason we consider the threat and not future “transition costs” are relevant.

32 the “Failed Relations” and “Status Quo”. Also here is questionable, if “Failed Relations” would come before the “Institutional Change” or vice versa. We claim that the uncertainty and possible high transition costs would make “Failed Relations” only the third-best preference for Catalonia. The central state’s first preference would be like in the first game the “Status Quo”, followed by an “Institutional Reform”. However, than the central state would prefer the option “Failed Relations” to a “Referendum” above all because of the high transition costs and the uncertainty, which could frighten the Catalan public opinion. It would be not clear if with the option “Failed Relations” Catalonia could express a unilateral declaration of independence. However, a referendum win for the “independence” option would give Catalonia a much stronger leverage.

CS: SQ > IC > FR > REF

CAT: REF > IC > FR > SQ

Table 9: The “ultimate” asymmetry game with a credible threat

Catalonia

Cooperate Defect

Cooperate IC (3/3) REF (1/4) Central State

Defect SQ (4/1) FR (2/2)

Source: own collaboration

In the “ultimate asymmetry game with a credible threat”, the game would take the form of a prisoners dilemma. “Failed Relations” would be the Nash-equilibrium outcome, which would mean uncertainty for both actors. The costs for the “Failed Relations” would be for both sides so high that the central state would probably offer Catalonia a binding contract, which could solve the Prisoners dilemma with the Pareto optimal outcome of cooperate/cooperate – the “Institutional Change”.

33

In a nutshell: In the game played between the central state and Catalonia not only the strategy place, but especially the question of a credible threat has been central. The Minority Unit Catalonia has its problems with making that threat credible.41 There are no international partners supporting Catalonia’s ambitions for a referendum and the European Union even if not condemning Catalan ambitions, tends to have a negative opinion on that question. Within the Catalan population there is an overwhelming majority for the referendum, but not the same can be said about their support for independence, which varies between 35-55% depending on the Research Institute, the question asked and the choices offered. These numbers also strongly influence the strength of a suggested civic disobedience, which would be quite new in a modern democratic country when dealing with the nationalist cleavage. Moreover, the Catalans do not have an own military force. For that reason, there is no exit option.

6. Conclusions

In this paper we answer the research question: How does asymmetry influence the territorial dynamics in the Spanish “State of Autonomies” through the analysis of the fourth types of asymmetry in its horizontal and its vertical pattern. Considering the horizontal axis we can identify two asymmetries. First, the horizontal de facto asymmetry between the Majority and the Minority Units. Catalonia has the biggest share in the national GDP, but considers itself as a relative loser in the economic development since the transition. Catalonia detected the fiscal institutional de jure asymmetry in form of two different taxation systems for the Basque Country and Catalonia as a possible reason for that development, the other one being the transfers to the central state. This institutional de jure asymmetry was one of the reasons for the change of the preferences of Catalonia, which during the constitutional bargain did not ask for a special financial treatment yet. The other aspect of the institutional de jure asymmetry, the linguistic differences between different AC, do influence the territorial dynamics in other points. Where a different language meets a Minority Units, like in Catalonia and the Basque Country, the decentralisation of the educational powers leads to two opposite nation-building

41 That could be different in the case of the Basque Country, where a possible terror of the ETA could have imposed a possible threat scenario.

34 projects. This leads to a kind of estrangement between the Majority and Minority Units. The two other kinds of asymmetries take place on the vertical axis between the central state and the Minority Units. The symbolic de jure asymmetry, the request of the recognition of the Minority Units as a nation, has been permanently refused by the central state. This refusal together with the rejection of a fiscal deal and the ongoing nation-building projects of the Minority Units lead finally to the request for the second kind of vertical asymmetry, the “ultimate asymmetry”, the constitutionally guaranteed right of self-determination of the Minority Units. This analysis shows that even if during the negotiations of the Constitution mainly historical reasons were responsible for the introduction of aspects of asymmetry, during the evolution of the territorial dynamics the conflict between the central state and the Minority Units has been crucial. Asymmetry has been avoided, until the ad- hoc coalition partners from the Minority Units forced the minority central governments into asymmetry deals. But the central government, (always one of the two statewide parties PP or PSOE) had two possibilities to influence the asymmetry dynamics. First, the constitutionally conform asymmetry requests can be opposed through the resymmetrisation process due to the instrumentalization of the regional Majority Units. This strategy is based on a common national statewide identity. Second, the constitution destabilising aspects of asymmetry can be opposed on the institutional level. Here above all the Constitutional Court with the strict interpretation of the constitution has been the fundamental actor. Nevertheless, when the central state opposes the symbolic de jure asymmetry or the regional Majority Units ask for a resymmetrical deal, the Minority Units try or to protect their asymmetrical status or to extend it further to confirm its being different. Simultaneously, if the nation-building project of the Minority Units has been successful, it could have already influenced or even strengthened its different national identity. Both developments are parallel and have immense influence on each other and usually lead to requests of the “ultimate asymmetry”. This kind of asymmetry questions the stability of the state. The regional Majority Units are not more able to participate, and the conflict/game has been reduced to two actors, the central state and the Minority Unit. Because in the first “asymmetry game” in 2014 the Minority Unit Catalonia was not able to make a credible threat, the outcome of the game is not an “Institutional Reform” including new aspects of asymmetry, but the “Status Quo”.

35

That shows that the Minority Units have not the leverage for the recognition of its “ultimate asymmetry” or any kind of asymmetry yet. They need the approval of the central state. There is also no possible exit option. A possible way out of this deadlock would be a return to the only successful strategy of the Minority Units in the democratic past: a statewide party (SWP) minority government on the central level, which would need the support of some of the Catalan or Basque regional non statewide parties (NSWP).

36

7. Bibliography:

Agranoff, R. (1999). Power Shifts, Diversity and Asymmetry, in: R. Agranoff (Ed.), Accommodating Diversity: Asymmetry in Federal States, Baden-Baden: Nomos Verlagsgesellschaft, pp. 11-23.

Agranoff, R. (1999b). Intergovernmental Relations and the Management of Asymmetry in Federal Spain, in: R. Agranoff (Ed.), Accommodating Diversity: Asymmetry in Federal States, Baden-Baden: Nomos Verlagsgesellschaft, pp. 11-23.

Agranoff, R. (1996). Federal evolution in Spain. International Political Science Review, 17(4), 385- 401.

Aja, E. (2003). El Estado Autonómico. Madrid: Alianza.

Bauböck, Rainer (2002). United in Misunderstanding? Asymmetry in Multinational Federations, Vienna: Austrian Academy of Sciences, Research Unit for Institutional Change and European Integration working paper. Vienna.

Benz, A., & Broschek, J. (Eds.). (2013). Federal Dynamics: Continuity, Change, and the Varieties of Federalism. Oxford University Press.

Broschek, J. (2013). Between path dependence and gradual change: Historical institutionalism and the study of federal dynamics. Federal Dynamics: Continuity, Change, and the Varieties of Federalism, pp. 93 - 116.

Burgess, M. (2006). Comparative Federalism in Theory and Practice. London: Routledge.

Burgess, M. & Gress, F. (1999). Asymmetrical federalism in Canada, the United States and Germany: comparative perspectives. Accomodating diversity: asymmetry in federal states, Baden-Baden, pp. 169-188.

Colino, C. (2009). Constitutional Change without Constitutional Reform: Spanish Federalism and the Revision of Catalonia's Statute of Autonomy. Publius, 39(2), pp. 262-288.

Colomer, J. M. (1995). Game theory and the transition to democracy: the Spanish model. Edward Elgar.

Duchacek, I., (1970). rev. 1987. Comparative Federalism: The Territorial Dimensions of Politics. New York: Holt, Rinehart and Winston; Lanham, MD: University Press America.

Elazar, D.J. (1994). Federal Systems of the World: A Handbook of Federal, Confederal, and Autonomy Arrangements (New York: Stockton)

Elazar, D.J., ed., (1987). Exploring Federalism. Tuscaloosa: University of Alabama Press.

Elazar, D.J. (1976). Federalism vs. Decentralization: The Drift from Authenticity, in: Vol. 6, No. 4, Dialogues on Decentralization, pp. 9-19.

Filippov, M., Ordeshook, P. C., & Shvetsova, O. (2004). Designing Federalism. A Theory of Self- Sustainable Federal Institutions. Cambridge: Cambridge University Press.

Ghai, Y. (2001). Constitutional asymmetries: Communal representation, federalism and cultural autonomy. In: The architecture of democracy, ed. A. Reynolds. Oxford: Oxford University Press.

Giordano, B., & Roller, E. (2004). Té para todos'? A comparison of the processes of devolution in Spain and the UK. Environment and Planning A, 36(12), pp. 2163-2181.

Greif, A., & Laitin, D. D. (2004). A theory of endogenous institutional change. American Political Science Review, 98(04), pp. 633-652.

37

Hombrado, A. (2011). Learning to Catch the Wave? Regional Demands for Constitutional Change in Contexts of Asymmetrical Arrangements. Regional & Federal Studies. Special Issue: Federalism and Constitutional Change: Theoretical and Comparative Perspectives, 21(4-5), pp. 479-501. Keating, M. (2001). Plurinational Democracy. Stateless Nations in a Post-Sovereignty Era, Oxford University Press, Oxford.

Holesch A. / Nagel K.J. (2012), “Educational policy in Spain – a federal illusion?, UPF Political Theory Working Paper 2/2012.

Kymlicka, W. (2001). ‘Western Political Theory and Ethnic Relations in Eastern Europe, in: Will Kymlicka and Magda Opalski (eds.) Can Liberal Pluralism be Exported? Western Political Theory and Ethnic Relations in Eastern Europe, Oxford University Press, Oxford: 13-106.

Kymlicka, W. (1998). ‘Is federalism a viable alternative to secession?’, in Percy Lehning (ed.) Theories of Secession, Routledge, London: 111-150.

Kymlicka, W. (1995). Multicultural Citizenship. A Liberal Theory of Minority Rights, Oxford University Press, Oxford.

Maiz, R., Caamaño, F., & Azpitarte, M. (2010). The Hidden Counterpoint of Spanish Federalism: Recentralization and Resymmetrization in Spain (1978–2008). Regional and Federal Studies, 20(1), pp. 63-82.

Mayntz, R., & Scharpf, F. W. (Eds.). (1995). Gesellschaftliche Selbstregelung und politische Steuerung (Vol. 23). Campus Verlag.

McGarry, J. (2007). Asymmetry in federations, federacies and unitary states. Ethnopolitics, 6(1), pp. 105-116.

McGarry, J., & O'Leary, B. (2009). Must pluri-national federations fail? Ethnopolitics, 8(1), pp. 5-25.

Moreno, Luis (2007): Federalization in Multinational Spain, in: M. Burgess and J. Pinder (eds.), Multinational Federations. London: Routledge, pp.142–189.

Moreno, Luis (2001), The federalisation of Spain, London: Frank Class.

Nagel, Klaus-Jürgen (2010): Spanien – Auf dem Weg zur Föderation?, in: Jürgen Dieringer/Roland Sturm (Hrsg.): Regional Governance in EU-Staaten, Opladen/Farmington Hills, pp. 149-170.

Palermo, F., Hrbek, R., Zwilling, C., & Alber, E. (2007). Auf dem Weg zu asymmetrischem Föderalismus? [On the way toward asymmetrical federalism?]. Baden-Baden, Germany: Nomos.

Requejo, F. (2005). Multinational federalism and value pluralism: the Spanish case. Routledge.

Requejo, F., ed., (2001). Democracy and National Pluralism. London: Routledge

Requejo, F., & Nagel, K. J. (Eds.). (2013). Federalism Beyond Federations (Epub) Asymmetry and Processes of Resymmetrisation in Europe. Ashgate Publishing.

Riker, W.H., (1964). Federalism: Origin, Operation, Significance. Boston: Little, Brown.

Rodden, Jonathan A. (2006): Hamilton’s Paradox. The Promise and Peril of Fiscal Federalism. Cambridge: Cambridge University Press.

Scharpf, F. W. (1997). Games Real Actors Play: Actor-Centered Institutionalism in Policy Research. Boulder: Westview Press.

Streeck, W., & Thelen, K. (2005). Beyond continuity. Institutional Change in Advanced Political Economies.

38

Tarlton, C.D., (1965). “Symmetry and Asymmetry as Elements of Federalism: A Theoretical Speculation.” Journal of Politics 27(4): pp. 861-74.

Watts, R.L., (2008), Comparing Federal Systems. 3rd ed. Montreal & Kingston: McGill-Queen’s University Press.

Watts, R.L., (2005), A comparative perspective on asymmetry in federations. Asymmetric working paper 4. Kingston: IIGR School of Policy Studies, Queen’s University.

Watts, R. (1999). The Theoretical and Practical Implications of Asymmetrical Federalism. In R. Agranoff (Ed.), Accommodating Diversity: Asymmetry in Federal States Baden- Baden: Nomos Verlagsgesellschaft, pp. 24-42.

Weller, M., & Nobbs, K. (Eds.). (2010). Asymmetric autonomy and the settlement of ethnic conflicts. Univ of Pennsylvania Press.

Zuber, C. I. (2011). Understanding the multinational game: Toward a theory of asymmetrical federalism. Comparative Political Studies, 44(5), pp. 546-571.

Von Beyme, K. (2005). Asymmetric federalism between globalization and regionalization. Journal of European Public Policy, 12(3), pp. 432-447.