Introduction: Diversities in the Spanish State of Autonomies

Spain is a diverse country from many different perspectives, for reasons rooted in history and geography(Aja, 2014; Arzoz, 2019). One traditional feature of the so-called “State of Autonomies” (Estado de las autonomías) is their asymmetric nature, which is both de facto—geography, climate, language, population, income—and de jure, rooted in the Constitution (Arzoz, 2019). Both types of asymmetries partly explain the role played by political agreements (Ruiz Almendral, 2023, Ruiz Almendral & Vaillancourt, 2013).

The process of decentralization in Spain is linked to democracy. As it has often been pointed out, there was a strong correlation between becoming a decentralized country and acquiring a democracy status as a country (Colino, 2018). From the outset, the very model of decentralization potentially allowed for a substantial level of asymmetry. Thus, the procedural framework established by the Spanish Constitution in articles 143 et seq. provided two special procedures for the formation of the autonomous communities, which differed in the speed of gaining authority (fast and slow lanes). Eventually, slow-laners were able to increase their authority and gain access to the maximum level, provided that they follow the process established in article 148.2 of the Constitution.

The original idea was that some provinces and regions, with past experience of self-government, should be given the opportunity to become fast-laners from the very beginning, while the rest would have to start by being slow-laners. This is according to the second transitional provision of the Constitution, which establishes fast access to autonomy for those regions which had approved self-government statutes in the past (beginning of the Twentieth Century). Initially, these were supposed to be Catalonia, the Basque Country, and Galicia. In the end, however, seven Communities became fast-laners, as Andalusia, Navarra, Valencia, and the Canary Islands had access to the higher level of autonomy from the outset. The other regions remained with a lower level of autonomy until the first decade of the current century, when they “caught up” with the higher level of authority. A second, more controversial, explanation for asymmetry lies in the loose recognition of the “historic rights” of some regions, enshrined in the first additional provision of the Spanish Constitution. This eventually resulted in the Basque Country and Navarra having a much greater level of tax authority.

From a strictly legal perspective, Spain is not a federal state, but it is common in the specialized literature to refer to “fiscal federalism” to assess the distribution of powers in matters of expenditure and revenue in Spain, adopting the Anglo-Saxon terminology of those who have mainly dealt with these issues (Castells, 1988; Oates, 1999, 2005).

The Devolution Process and Its Asymmetries

Spain’s process to become the current State of Autonomies is unique in that it underwent a complicated process of fiscal decentralization in a relatively short time span. From a fully centralized country in 1978, it had already been divided into seventeen Autonomous Communities by 1982. At the same time, an arduous process of regime change (from Franco’s 1939–1975 dictatorship to the 1978 democratic Constitution and the entry in the European Union in 1986), named the Transición brought about substantial legislative reforms and paved the way toward a full democracy (Ruiz Almendral, 2023, Colino, 2013, 2018; Valdesalici, 2018).

At the same time, this also meant a substantial overhauling of the tax system. It is not an exaggeration to say that until the late 1970s, Spain did not have a tax system as such, at least not one that was generally implemented, or which followed the general structure of the tax systems of other OECD countries. The first modern personal income tax was established in the late seventies, with the first corporation income tax. The tax reform undertaken between 1978 and 1985 entailed a substantial increase of tax pressure. Tax revenues in fact quadrupled between 1975 and 1980 (OECD, 2021).

From a fiscal federalism perspective, the high vertical fiscal imbalance with which the Autonomous Communities started off in 1982 was partially resolved by transfers as well as mechanisms of tax sharing (known as “ceded taxes”) (Castells, 1988; Herrero et al., 2010; Rodríguez Bereijo, 2015). It is commonplace to state that the vertical fiscal imbalance arises when one tier of government—usually the central government—has a greater power to raise revenues than it actually needs for the exercise of its assigned level of authority, while another one (a subnational unit) is in the opposite situation (Bird & Vaillancourt, 2006; Oates, 1977; Ruiz Almendral, 2023; Ruiz Almendral & Vaillancourt, 2013). Specifically, since 1997, the vertical fiscal imbalance has been addressed partly by sharing the personal income tax revenue, what in practice means that the Autonomous Communities perceive a percentage (currently 50%) of the revenue accrued in their territory. On top of that, the Communities may also establish the tax rates to be applicable in their territory, introduce new tax credits, or increase/decrease those established by the central government.

The Evolution of the Constitutional Framework in Terms of Tax and Finance Power Decentralization

The Spanish Constitution (sections 133 and 157) bestows taxation powers upon the 15 communities which conform to the so-called “common regime”. The Constitution also recognizes two types of specific tax regimes: the so-called “foral regimes” (Basque Country and Navarra) and the Canary Islands tax and fiscal regime (Ruiz Almendral, 2003, 2023).

The Basque and Navarra foral regimes are rooted in history, as is the case of many other institutions in Spain (i.e., civil laws and specific fueros). But the legal origin of these exceptions is not to be traced back to historical rights, but to the Spanish Constitution. As the Constitutional Court has repeatedly stated, all subnational financing regimes are evolving systems which must fit in the Spanish Constitution. In fact, the Constitution does not simply recognize all previous historic rights, but rather “what the Constitution guarantees is the very existence of a foral regime, but not each and every one of the rights that have historically made it up” (Constitutional Court ruling (STC) no. 76/1988, FJ 4). The same reasoning has been reiterated, among others, in STC 86/1988, FJ 5; and STC 214/1989, FJ 26. The Constitution does not exactly recognize the historical rights as they once were, but enables the general updating of these foral regimes and provides for them to be applied and developed “within the framework of the Constitution and the Statutes of Autonomy” (STC 208/2012, FJ 2).

In accordance with the recognition of autonomy, the Spanish Constitution (SC) also recognizes communities’ “financial autonomy for the development and execution of their authority” (SC, 1978, art. 156). Apart from stating this principle of financial autonomy, the Constitution also enumerates the resources that could constitute the communities’ revenue base. The list includes almost all types of possible existing revenue sources: i.e., ceded taxes; surtaxes on existing central government taxes; their own taxes; public debt; and transfers (SC, 1978, art. 157.1).

However, article 157.3 of the Constitution (1978) also allows the central government to approve a special “organic” law (ley orgánica) regulating both how the resources listed in section 157.1 will be distributed among the communities, and the limits on the exercise of their financial power on the resources (i.e., whether and to what extent they may create new taxes, etc.).

In practice, this has resulted in the central government having significant power to both limit and control the financial and tax autonomy of the Autonomous Communities. Currently, this organic law is the special law for the financing of the Autonomous Communities, Law 8/1980, Ley Orgánica de Financiación de las Comunidades Autónomas (LOFCA), which imposes strict limits on communities’ capacity to create new taxes.

The most relevant limitation for tax powers is the prohibition of double taxation (articles 6.2 and 3), which prevents communities from establishing taxes similar to existing taxes created by the central government and the municipalities. Article 9 of the LOFCA adds specific limits related to the construction of the Spanish internal market, which are reminiscent of the European Union (EU) fundamental freedoms, as interpreted by the European Court of Justice (ECJ).Footnote 1 Specifically, article 9 c) of the LOFCA (1980) stipulates that taxes established by the Autonomous Communities

may not constitute an obstacle to the free movement of persons, goods and capital services, nor effectively affect the establishment of residence of persons or the location of companies and capital within Spanish territory, […] nor entail burdens that may be transferred to other Communities. This situation forces a complex dialogue between courts in the interpretation to be given to limits that, at least in their wording, appear to be similar.

The original limitation of Autonomous Communities’ common regime tax powers has an obvious explanation. When the Constitution (1978) and the LOFCA (1980) were approved, both the municipal and the central governments had already established taxes on most of the sources of revenues, which has left little tax room for communities.

Autonomous communities under the common regime have made use of the taxation power assumed in the Statutes of Autonomy and recognized by the LOFCA, but the LOFCA imposed prohibition of taxing taxable events previously taxed either by the State and or by the local entities has extensively limited, in practice, their available tax space. This also led to an increase in conflicts, which were traditionally resolved before the Constitutional Court, rather than by agreement.

At the same time, the exercise of regulatory powers through the creation of their own taxes has been uneven, with some innovative taxes (such as taxes on large commercial establishments, or the tax on sugary drinks), and others that have given rise to various problems (such as taxes on environmental installations, or on deposits in credit institutions). In all, tax revenues corresponding to the common regime autonomous communities barely account for 2% of total subnational government revenues, but do constitute the vast majority of tax conflicts between the central government and the Autonomous Communities.

The lack of available “tax space” inspired the reforms undertaken in 1996, which entered into force in 1997, when there was a fundamental change in the financing system of the communities. Simply put, in 1997, some taxes traditionally under the central authority, and including the personal income tax, were transformed into shared taxes (ceded taxes or impuestos cedidos), substantially increasing the taxing powers of the communities. Subsequent reforms in 2002 and 2009 have further strengthened communities’ powers over these taxes.Footnote 2

In principle, the goal of these reforms was to make communities more involved in the establishment of taxes and thus more directly accountable to their taxpayers for their expenditure. The reforms undertaken in 1997, 2001, and 2009 by allocating and increasing regulatory capacities over ceded taxes have indeed led to greater fiscal co-responsibility, at least in theory. Ceded taxes have substantially increased the Autonomous Communities’ tax autonomy. In a comparative perspective, few countries in the world have greater tax autonomy at the subnational level of government (Cuenca, 2022; Lago-Peñas, 2021; Herrero Alcalde & Tránchez Martín, 2011; Ruiz Almendral, 2023; Ruiz Almendral & Vaillancourt, 2013; Ruiz Almendral et al., 2018). However, the high decentralization of spending means that there is still a wide imbalance between the two sides of the budget and that common regime Autonomous Communities continue to depend, to a high degree, on transfers or territorialized shares in VAT and excise duties, over which they lack tax autonomy, although they are formally included in the tax revenue sharing mechanism.

The system of ceded taxes is currently in need of a substantial reform, as several issues have arisen over time. First, the original central government’s regulation of some ceded taxes, such as the inheritance and gift tax, net wealth tax and transfer tax, and stamp duty have become obsolete after decades of insufficient attention, in addition to deficiencies in control and management. Furthermore, some Autonomous Communities have substantially lowered effective taxation on the first two, so that they are on the verge of disappearing (Cuenca, 2015).

Second, in the case of personal income tax, the very structure of the regional financing system and its shared nature implies a mismatch between the measures adopted by the Autonomous Communities and their perception by the public (López-Laborda et al., 2020), so that “irrespective of their individual believes on the topic, citizens living in more pro-devolution regions tend to attribute powers to the central government to a larger extent. Put it differently, they tend to sub-estimate the level of current decentralization” (Herrero Alcalde et al., 2018, p. 38).

Third, according to some commentators, the decentralization of property taxation may have generated simulated mobility of tax basesFootnote 3 and other negative fiscal externalities, while also posing significant problems from an EU law perspective.

Finally, Communities have also established different deductions and tax benefits in ceded taxes which may be in breach of European Union law, while (mostly) they do not have a significant revenue impact (Ruiz Almendral, 2022a, 2022b, 2023).

The ceded or shared-taxes system has been designed in a way that bestows the central government the power to coordinate and therefore, if necessary, curb or limit the exercise of taxation powers by the Autonomous Communities. From a constitutional perspective, the tool employed is a so-called framework law or ley marco (SC, 1978, art. 150.1). In practice, tax sharing (or ceded taxes) follows, as far as regulatory powers are concerned, a system of delegation of regulatory powers (SC, 1978, art. 150.1), which entails that the central authority retains ownership of all ceded taxes (see an explanation of this system in the following rulings: SSTC 161/2012, FJ 3; 19/2012, FJ 11; 25/2016, FJ 2; and 33/2016). This delegation method has different consequences, the main one being that the cession or sharing can be clawed back by the central government (STC 16/2003, FJ 11), and “the [central] authority may modify the scope and conditions of the cession of a tax, or even abolish it [the tax]” (STC 35/2012, FJ 9), albeit on the condition that an agreement is previously sought with the affected communities. Furthermore, failure to comply with the terms of the transfer will imply the unconstitutionality of the regional regulations, as this automatically leads to an invasion of the central government competence over these taxes, as the Court has repeatedly pointed out (see STC 21/2022, STC 186/2021, STC 161/2012, STC 197/2012, and STC 35/2012).

This model of tax sharing through a “ceded-taxes” system has allowed a great development of the Autonomous Communities’ financial autonomy. However, at the same time, it poses challenges of coordination, bearing in mind that the traditional limits of the national tax system have evolved with the advancements of the integration into the European Union.

A very different system applies to the so-called foral regimes. For them, the functioning of the financial system is radically different to that of the common-system Autonomous Communities and is the result of bilateral negotiation resulting in a convention (Navarra) or an agreement (Basque Country). In a nutshell, under the cupo (Basque Country) and quota (Navarra) systems, the foral regions run all the risk themselves, and no revenue guarantee is provided by the central government.

Presumably, this status is still attractive for these communities, as they are richer than the Spanish average. Another feasible explanation is that as some claim the actual payment of the quantities (the cupo and quota) was never been properly calculated, the result being that these foral communities may actually be paying less for the same services than the rest of (common-system) communities (De La Fuente, 2022; Monasterio, 2009; Zabalza, 2012). At the same time, the lack of transparency makes it harder to fully analyze it. In a recent work, De La Fuente (2022) points out that

there is not really, or at least it is not publicly known, a detailed methodology for calculating the quota and the cupo, but only a series of general principles and a one-page annex in which the final amounts of both contributions are fixed without much explanation. Consequently, the critical analysis of the foral model cannot be approached in general terms, as if there were a well-known methodology with technical and/or equity criteria. It necessarily becomes a more complex and risky exercise in which there is no choice but to start by proposing a methodology, which then has to be applied to real data to arrive at reference figures to be compared with the observed results of the model. Since there is certainly no single reasonable way of concretizing and applying the principles contained in the convention, the agreement and the Constitution, the numerical results of the present study must be considered as merely indicative. The most important thing is that this exercise forces to concretize, and therefore to put explicitly on the table, a whole series of issues that have to be addressed in order to arrive at concrete figures, and as such shows that many of these issues are not well covered or well solved either in the agreement and the convention, or in the rules and agreements that develop and concretize them. (p. 83)

Among other problematic issues described in De La Fuente’s work, attention is drawn to the calculation of the resources needed to finance the competences not assumed by the comunidades forales. This would not be correct, and would lead to higher financing, among other factors. This was in fact one of the conclusions of the report for the reform of the autonomous financing system published in 2017 (Comisión de Expertos para la Revisión del Modelo de Financiación Autonómica, 2017).

Spanish Fiscal Federalism in the EU Context

It has become commonplace to state that the interdependence of states has accelerated in recent decades.Footnote 4 Spain is a globalized economy, as well as one of the most decentralized and richest countries in the world (Cuenca, 2022; Lago-Peñas et al., 2017; OECD, 2018). Since January 1986, it has also been a member of the European Union. These three dimensions—globalization, decentralization, and Europeanization—fundamentally shape the scope of taxation powers in Spain and must be borne in mind for all diagnoses on the Spanish tax system, and in discussions of reform, as was repeatedly pointed out in the recent report on tax reform (Ruiz-Huerta, 2022).

In practice, the plurality of sources and the variable distribution of authority between the European Union and the Member States, and between the latter and their own subnational entities, such as Autonomous Communities in Spain, gave rise to different assumptions and degree of interference of the limits envisaged by European Union law.

EU law does not simply provide a system of limits, but a new legal framework within which all tax figures, whether harmonized or not, will have to develop. In this context, it is necessary to determine the current legal contours within which the tax systems of member states can develop. The EU law driven impact varies substantially depending on the type of tax and tax measure of concern. The EU legal framework itself has evolved to find its own system of rules, which largely depart from the accepted logic of international tax law, based on rules in force in the member states through double taxation treaties, whose bilateral nature is the most visible element of a different logical structure, with interpretative principles that are also different from those that inspire the model of construction of the European Union. This web of principles and rules translates into a polyhedric set of limits that impact the extent and the exercise of taxing power by member states (Ruiz Almendral, 2022a, 2022b, 2023).

However, the assertion that the taxing power of the Autonomous Communities is severely limited by EU law is premature. It is not primarily the EU legal framework, but above all the current context of (further) globalization, accelerated by technological change, that has a decisive impact on the national tax systems, designed for a world that in many respects no longer exists. This has been highlighted in different ways, by the recent efforts of the European Commission, and, above all, with the international tax avoidance maneuvers supported by changes in the way business is done. Problems, such as determining where value is created or generated in the case of new technology services or business models, or even what value should be assigned to data collection, have largely transformed the traditional foundations of taxation, as the OECD and G20’s current work shows (Ruiz-Huerta, 2022; Schön, 2018).

From a purely internal perspective, the vertical distribution of tax and financial authority in terms of revenue and expenditure is a key element in any Constitution, and indeed a central part of the Spanish Constitution. The financial and economic constitution in a decentralized country incorporates rules on the distribution of “tax space” and on the coordination between the legislative actions of the different authorities. There is however no single model of decentralization, and if the differences between the U.S. and the Canadian financial constitutions on the distribution of taxation power are substantial, the same occurs in the federal or partially decentralized models within the European Union itself.

An economic union may of course seem intuitively contradictory, by definition, with the development of fiscal federalism (Ruiz Almendral, 2022a, 2022b, 2023; Traversa, 2011). However, practice shows the opposite tendency of coexistence between agreements of greater integration and greater decentralization, without prejudice to the extensive literature that has shown the reasons, in terms of efficiency, that support the decentralization of certain tax figures, and in general a better coherence and correlation between the distribution of public spending and the responsibility for revenues, precisely to avoid problems of agency (“moral hazard”) that would imply a more inefficient use of public resources (Oates, 2005).

The adjustments will necessitate a more advanced model of fiscal federalism, which considers the processes of integration (in the EU) and of decentralization (within the country), present in all constitutions of complex states and therefore not uniquely a European problem.

One of the most obvious areas of influence and limitation to member states powers is tax harmonization. The founding treaties gave primary relevance to indirect taxation as a possible obstacle to building the internal market. In the mid-twentieth century, in an environment of much less mobility, direct taxation was not thought to pose significant constraints to the internal market. Thus, only indirect taxes were subject to harmonization. Article 113 of the Treaty on the Functioning of the European Union (TFEU) contains a mandate for harmonization of “turnover taxes, excise duties and other forms of indirect taxation to the extent that such harmonization is necessary to ensure the establishment and the functioning of the internal market and to avoid distortion of competition” (2012, p. 48). In fulfillment of this mandate, value added tax, excise duties, energy taxation, and taxes on the raising of capital were established.

The most tangible consequence of tax harmonization is the requirement for member states to establish harmonized taxes in their territory, as well as the impossibility of establishing other taxes equivalent to them. This second element has specific relevance for subnational taxation. It would not be possible to bestow any type of regulatory powers to the Autonomous Communities in the case of VAT or excise duties, except in a very limited way. The same applies to the transfer of regulatory powers in the case of corporate transactions of the tax on capital transfers and documented legal acts.

Thus, the shared-taxation system implemented in Spain as of 1997 (ceded taxes) did not provide for the attribution of regulatory powers over these taxes. Subsequently, part of the collection of these taxes would be attributed to the Autonomous Communities, following a model already in force in other federal states, where the decentralization of consumption taxation has been progressively ruled out.

Specifically, in the case of VAT, Article 33 of the Sixth Council Directive 77/388/EEC of May 17, 1977 on the harmonization of the laws of the member states relating to turnover taxes prevents the introduction of any other “turnover” tax, with limited exceptions such as those provided for in the so-called “outermost” regions (such as the Azores or the Canary Islands). In the case of excise duties, Article 3 of the Council Directive 92/12/EEC of February 25, 1992 on the general arrangements for products subject to excise duty and on the holding, movement, and monitoring of such products establishes a threefold limitation. First, the same products covered by the directive (mineral oils, alcohol and alcoholic beverages, and manufactured tobacco) may be subject to other indirect taxes for specific purposes, provided that such taxes comply with the tax rules applicable to excise duties or VAT for the determination of the taxable amount, assessment, chargeability, and monitoring of the tax (Sixth Council Directive, 1977, Art. 3, Sec. 2). Second, member states may impose taxes on different products, provided that this does not affect trade between member states because this would give rise to border-crossing formalities (Sixth Council Directive, 1977, Art. 3, Sec. 3). Finally, member states may tax the supply of services in so far as they are not in the nature of turnover taxes, including those relating to products subject to excise duties (Sixth Council Directive, 1977, Art. 33).

The prohibition of the establishment of VAT-like measures has been the subject of more than thirty cases before the ECJ, which has mostly rejected the equivalence of the measure at the origin of the dispute. To be equivalent to VAT, a tax must be applied across the board to all transactions in goods and services. The four characteristic features of VAT must therefore meet the following:

VAT applies generally to transactions relating to goods or services; it is proportional to the price charged by the taxable person in return for the goods and services which he has supplied; the tax is charged at each stage of the production and distribution process, including that of retail sale, irrespective of the number of transactions which have previously taken place; the amounts paid during the preceding stages of the production and distribution process are deducted from the VAT payable by a taxable person, with the result that that tax applies, at any given stage, only to the value added at that stage and the final burden of that tax rests ultimately on the consumer. (Joined Cases C-283/06 and C-312/06: KÖGÁZ rt and Others v. Zala Megyei Közigazgatási Hivatal Vezetője; OTP Garancia Biztosító rt v. Vas Megyei Közigazgatási Hivatal, 2007, p. 2)

As such, Article 33 of the Sixth Council Directive does not preclude the maintenance or introduction of a tax which does not have one of the essential characteristics of VAT (inter alia, Banca Popolare di Cremona, para 27, Judgment of 3 October 2006, Banca Popolare di Cremona, C-475/03, EU:C:2006:629). As recalled in KÖGÁZ rt and Others v. Zala Megyei Közigazgatási Hivatal Vezetője, the relevant element is to determine whether the measure established has the effect of jeopardizing the functioning of the common system of VAT by being levied on the movement of goods and services and on commercial transactions in a way comparable to VAT (citing Pelzl, Banca Popolare di Cremona; and EKW and Wein & Co - Judgment of 8 June 1999, Erna Pelzl, Joined Cases C-338/97, C-344/97 and C-390/97, EU:C:1999:285; Judgment of 9 March 2000, EKW and Wien & Co, C-437/97, EU:C:2000:110). This will be the case for those measures which have the essential characteristics of VAT, even if they are not identical to it in all respects.

A further example of the impact of harmonization on ceded taxes is the case of the tax on retail sales of certain hydrocarbons (Impuesto sobre las Ventas Minoristas de Determinados Hidrocarburos, “the IVMDH”). Originally a ceded tax, it was designed as a kind of surcharge that could be “activated” (or not) by each Autonomous Community, and whose collection was to contribute to the financing of health care competences, the decentralization of which had just been completed (hence the name “céntimo sanitario”, or “health cent”). Later, in the Transportes Jordi Besora case (Judgment of February 27th, 2014, Transportes Jordi Besora, C-82/12, EU:C:2014:108), the tax was declared incompatible with Directive 92/12/EEC.Footnote 5 The result of this judgment was not surprising. This undoubtedly influenced the refusal to limit the effects of the judgment, with the ECJ stating, among other considerations, that the incompatibility of the tax had already been established in the EKW and Wein & Co. judgment (Evangelischer Krankenhausverein Wien v. Abgabenberufungskommission Wien et Wein & Co. HandelsgesmbH v. Oberösterreichische Landesregierung, 2000), where it had already been recalled that the tax was incompatible with Directive 92/12/EEC. The reason for their incompatibility is that Article 3(2) of the Directive only allows mineral oils to be subject to indirect taxes other than the excise duty established by the Directive if they “pursue one or more specific purposes” and if, in addition, they comply with the tax rules applicable in relation to excise duties or VAT for the determination of the taxable amount, chargeability, and monitoring of the duty. Still,

member states shall retain the right to introduce or maintain taxes which are levied on products other than those listed in paragraph 1 provided, however, that those taxes do not give rise to border-crossing formalities in trade between member states. Subject to the same proviso, member states shall also retain the right to levy taxes on the supply of services which cannot be characterized as turnover taxes, including those relating to products subject to excise duty. (European Union Law, 2009, art. 3.3)

Intergovernmental Tax Relations and the (Limited) Role of Agreements

Overall, asymmetry in devolution of powers helps explain the role of bilateral and multilateral agreements in the assignment and actual implementation of authority.

The existing tension between, on the one hand, the desire to rationalize the distribution of competences by equalizing them among the different Autonomous Communities and, on the other, the persistence of the aforementioned underlying asymmetry, which has its origin in the “choice principle”, is manifested, as far as agreements are concerned, in the coexistence of multilateral and bilateral logics. This is clearly reflected in the matter of regional financing. Thus, the elevation to legal status of a series of matters that until now were exclusively covered by agreements does not eliminate the role of the latter, but rather brings them back to an appropriate logic, since it seems indisputable that the agreements between the state and the autonomous communities play an important role in the articulation of the transfer of taxes. These agreements take place in two bodies, the General Council for Fiscal and Financial Policy and the Joint or Mixed Commissions, which obey, respectively, a multilateral and a bilateral logic. This is expressly recognized, among others, in STC 13/2007, FJ 8; 31/2010, of June 28th, FJ 130; and 204/2011, FJ 7, which address the role the Fiscal and Financial Policy Council currently plays as a body for coordinating financial powers between the state and the autonomous communities.

Of course, such an institutional solution has often been criticized for its lack of transparency, as those agreements take place behind closed doors and the results are only partially made public. This is the so-called “executive federalism” that may give rise to a deficit of democracy (Cameron & Simeon, 2001; León-Alfonso, 2007). It has been argued that most of this political discussion should take place in the Senate, which, at least in theory if not in practice, is the representative chamber of the autonomous communities.

As is the case in all federations, political agreements about the allocation of resources have also played a relevant role in shaping the financing system of the Autonomous Communities. Agreements are an essential part of cooperative federalism and cannot always be substituted by debate in Parliament.

That said, a reform of the Spanish Senate is probably necessary, as this would be the best way to reinforce these agreements. Such a reform should serve to give the Senate sufficient authority to fully discuss legislation affecting autonomous communities, in a similar fashion to the German Federal Council.

At the same time, in practice, the so-called “choice principle” (principio dispositivo) has a very limited role, if at all, in determining the financing authority of Autonomous Communities.Footnote 6 In fact, the latter are not entitled to choose exactly how that authority is financed, at least not in the sense of determining the financing model.

This interpretation of the choice principle, as the right to decide the financing system has been expressly rejected by the Constitutional Court in STC 204/2011, (FJ 7). The Court recalls that the choice principle, contained in SC article 147.2 d., cannot be interpreted in the sense that each Autonomous Community can decide, on its own, which part of the financing system should be applicable to itself, or even if it should be applicable to itself at all, as it should be remembered that the central government, within the margins granted to it by the Constitution and respecting the principles and the autonomous financial competences stipulated therein (particularly in art. 157 EC), is constitutionally empowered to establish one or another system of autonomous financing […]. It is therefore a regulatory model whose apex (the LOFCA) is part of the block of constitutionality and which can vary according to the political decisions of the central legislator (organic and ordinary), with the participation of the Autonomous Communities, a model on whose goodness or functionality, as pointed out in STC 68/1996, of 4 April, it is not for this Court to pronounce (SSTC 192/2000, FJ 10 and 68/1996, FFJJ 3 and 9).

Consequently, to confer binding force on the will of the Autonomous Communities would not only annul the exclusive power of the State to configure the financing system of the Autonomous Communities that it considers most suitable, but would also deprive it both of exercising its powers of coordination (art. 156.1 EC), and of guaranteeing the effective implementation of the principle of solidarity enshrined in art. 2 of the Constitution (SSTC 13/2007, FJ 9 and 31/2010, FJ 135).

Therefore, according to the Court, the choice principle cannot be interpreted in the sense that the Autonomous Communities must be the ones empowered to choose the system under which they must be financed (STC 204/2011, FJ 7).

The Central Role of the Spanish Constitutional Court as a Conflict Resolution Mechanism

To date, the most relevant conflict resolution mechanism in Spain is the Constitutional Court. There was always agreement, during the 1978 Constitution discussions, for the need to create a constitutional Court (Pérez Tremps, 1985). Despite the name, it is not exactly a court, as it is not part of the judicial branch (see titles VI and IX of the Constitution, which separately deal with judicial powers and the Court) (Arzoz, 2021).

Taking into account that the vast majority of the matters listed in the Constitution are shared between the central and subnational governments, it is not hard to imagine that this has been a source of permanent conflict between these two tiers of government. As the only body competent to resolve such conflicts, the Court has undertaken a very important task in the evolution of the “State of Autonomies” (López Guerra, 1998).

This role has been reinforced by the open-ended or unfinished nature of the different constitutional provisions regarding subnational autonomy, and by a certain pedagogic tendency of the Court to fully explain and thus serve to clarify the rules governing the “State of Autonomies”.

It is often wrongly assumed that the Court has served to mitigate diversity or even limit it. In fact, the opposite is true, as has often been pointed out (Aja, 2014; Borrajo Iniesta, 2004; López Guerra, 1998). An analysis of the Court’s case law shows its upholding of Autonomous Communities’ authority in different areas, as López-Laborda, Rodrigo, and Sanz-Arcega show. There are many examples that illustrate the contribution of the Court to the devolution process (Arzoz, 2021).

For example, the Court has also been key in defending and upholding the official status of those regional languages recognized in the statutes of autonomy. Already in the eighties, the Court insisted that the 1978 Constitution recognizes the plurilingual reality of the Spanish Nation and that this reality is a cultural value not only acceptable, but also worthy of being promoted. From this reality, a series of legal consequences derive in terms of the possible attribution of official status to the different Spanish languages, the effective protection of all of them, and the configuration of individual rights and duties in linguistic matters (for example, SSTC 165/2013, FJ 4; 32/1986, FJ 1). This of course comes at a price, as Xavier Arzoz points out:

it is the Court who has given rationality and viability to the whole system. The other side of the coin is that the Court’s contribution has probably been excessive, as it has not merely applied but defined the rules, almost irreversibly for political actors. (Arzoz, 2021, p. 428)

The Court, of course, has also been crucial in upholding democracy and maintaining the rule of law, which was breached by an independentist minority in the Community of Catalonia in 2017 with the organization of an illegal referendum, after enacting a number of laws contrary to the Statute of Autonomy of Catalonia and the Spanish Constitution, in what has been named a process tantamount to an incomplete or attempted coup d'état. During September and October of 2017, the Government of Catalonia disobeyed every order issued by the Spanish courts, which had stressed that organizing a referendum for secession was in defiance of the Spanish Constitution and the Statute of Autonomy of Catalonia.Footnote 7 As in any parliamentary democracy, disobeying direct Court orders and ignoring the fundamental laws that govern them—such as the Spanish Constitution and the Statute of Autonomy of Catalonia—invariably carries harsh penalties. As the Spanish Constitutional Court has repeatedly insisted in more than twenty rulings addressing the Catalan independence issue, the people of Spain (including all Catalans, not only those citizens who aim at seceding) are the constituent power who have the right to decide their future. As the Court has also pointed out, independence is of course a legitimate political aspiration, but as is the case with any other legal ideology, it must follow the rule of law, which implies, among other things, obtaining the consent of the Spanish demos, and not just a part of it (among many others, STC 90/2017).

Another key element of the contribution of the Spanish Constitutional Court to the formation of the State of Autonomy is the case law on equality. This is particularly relevant since Spain is a country with such a long tradition of centralized rule. In this context, the Court stated early on that the principle of equality in a multilevel state such as the State of Autonomy should not be interpreted as precluding a different legal position of citizens. As early as 1981, the Court also recalls that although the authority of the Autonomous Communities is limited to the territory, this territorial limitation of the effectiveness of the rules and acts cannot mean, in any way, that these bodies, in the use of their own competences, are prevented from adopting decisions which may produce de facto consequences in other parts of the national territory. The political, legal, economic, and social unity of Spain prevents its division into watertight compartments and, consequently, depriving the Autonomous Communities of the possibility of acting, when their acts could have consequences beyond their territorial limits, would necessarily be tantamount to depriving them, purely and simply, of all capacity to act (STC 37/1981, FJ 1).

Also early, the Court denied the central government to use its harmonizing powers in a pre-emptive way before the Autonomous Communities had exercised their powers. In STC 76/1983, the Court upheld the appeal of unconstitutionality brought by the Basque Government, the Basque Parliament, the Executive Council of the Generalitat of Catalonia, the Parliament of Catalonia, and fifty members of the Cortes Generales against the Draft Organic Law Harmonizing the Autonomous Process (LOAPA). This is one of the fundamental rulings in the construction of the “State of Autonomies”. The Court recalls that there is no constitutional basis for defending equal rights for the Autonomous Communities. The Constitution enshrines the equality of individuals and social groups, but not that of the autonomous communities. Although equality does exist in the subordination of the autonomous communities to the constitutional order, the latter can be unequal in terms of the procedure for access to autonomy and the determination of the content of their statute and their powers; it is precisely the autonomous regime that is characterized by a balance between the homogeneity and diversity of the public legal status of the territorial entities that comprise it. Without the former, there would be no unity or integration in the state as a whole; without the latter, there would be no true plurality or capacity for self-government, which characterizes the state of the autonomous regions (STC 76/1983, FJ 2 a.)

In finance and taxation matters, the Court has reinforced the Autonomous Communities’ spending power (the leading case is still STC 13/1992, and earlier STC 37/1987) and the right of Autonomous Communities to create taxes, within the limits of the Constitution.

Therefore, even if the Court has substantially contributed to shaping the “state of autonomies”, since the nineties, different voices have proposed the Court to play a more limited role in this regard, in favor of a stronger role for the Senate (López Laborda et al., 2019). As López Guerra pointed out more than twenty years ago, it has become routine in Spain “to discuss any law of certain importance in two forums, a first debate takes place in the Parliament, a second and decisive one in the constitutional Court” (1998, p. 263).

A possible solution is to reinforce the role of the Senate, something traditionally rejected by some Autonomous Communities (the foral communities, but also Catalonia) that would rather continue having a bilateral relation, and not share the same forum with the other Communities. Autonomous Communities are represented at the Senate, which operates as a second revision legislative chamber. However, the Senate is only in theory a representative chamber of the Autonomous Communities. Two reasons explain this: first, most senators are elected by universal suffrage from provincial voting districts, while only a minority is appointed by the Parliaments of the Autonomous Communities. Second, the Senate has very limited powers vis-ávis the central government’s law-making process (i.e., no veto rights) (López Guerra, 1993).

The Future(s) of Spanish Fiscal Federalism

As I revise these lines (April 2023), there is a current debate on the constitutionality of the new impuesto temporal de solidaridad de las grandes fortunas (temporary solidarity tax on large fortunes, ITSGF), approved by the central government as a “complementary” tax to the impuesto sobre el patrimonio (IP), may raise doubts about its compatibility with the financial autonomy of the Autonomous Communities and, more specifically, with the very framework of tax devolution in force since 1997, which includes the delegation of certain regulatory powers to them. In fact, at the time of writing, the Constitutional Court has agreed to hear the appeals of unconstitutionality filed by the autonomous communities of Madrid and Andalusia against the new “temporary solidarity tax on large fortunes”.

As defended elsewhere, the tax may not be contrary to the Constitution (as argued by Madrid and Andalusia in their filed complaints before the constitutional Court), but it certainly is contrary to the spirit of ceded taxation (Ruiz Almendral, 2023). Legally speaking, it is true that ceded taxes belong to the central government, as established in their regulation. But the question arises as to whether this was the most suitable mechanism, as well as whether action should not have been taken by using control mechanisms, so that the questionable use of regulatory powers in ceded taxes is not resolved only via claims before the Constitutional Court.

On the other hand, there is a percentage of the central government using its powers to harmonize regional taxation. In the highly controversial case (there are four dissenting votes) resolved by STC 26/2015, of February 19th, the Constitutional Court declared the tax on bank deposits established by the central state, with a zero-rate tax, to be compatible with the Constitution, with the sole purpose of “harmonizing”, i.e., eliminating, taxes of the same type that had already been established by several autonomous communities. The Court affirms that although it is a zero-rate tax, it is a real tax, since the legislator has a wide margin to establish a tax with other additional purposes, not strictly speaking for tax collection but for organizational purposes. More importantly, in this case, the Court reiterates that the state has the power to coordinate its own tax system with those of the Autonomous Communities (SC, 1978, art. 149.1.14), as well as the pre-eminence for occupying taxable events pursuant to article 6.2 of the LOFCA. Thus, it can be concluded that the state will also be competent to create a tax whose central purpose is the coordination or harmonization of the taxation of credit institutions, in accordance with the title of coordination contained in articles 149.1.14, 133, and 157.3 of the Spanish Constitution. Furthermore, the very mechanism to cede taxes is, by itself, a coordination mechanism, which is why the ley marco (SC, 1978, art. 150.1) is used in it.

But of course, the long-term question is how fiscal federalism in Spain could and should be shaped. As I have argued elsewhere (Ruiz Almendral, 2023), the examination of Court cases brought about in the past 30 years, in particular on ceded or shared taxes, illustrates how the decentralization of taxation power in Spain has been carried out without sufficiently considering European integration. Often, measures established by the Autonomous Communities hardly comply with EU law. Of course, it must also be considered that the case law of the European Court of Justice also shows the challenge for national tax systems derived from EU law and the current interpretation of the principle of non-discrimination. But, even beyond the strict scope of the European Union, the difference in treatment between residents and non-residents, traditionally key in international taxation (or fiscal federalism), has since long been eroded. Of course, as has often been stressed, the problem with the ECJ's doctrine is that it often reveals an impossibility: in reality, to strictly comply with the non-discrimination mandates, perfect harmonization would have to be achieved in some taxes.

Ultimately any attempt to establish a system by means of case law is, by definition, doomed to fail. Even if the Constitutional Court has played a crucial role in the very existence of fiscal federalism in Spain, Courts answer the question put to it and decide based on the cases that are admitted and examined. They do not, by design, examine the system as a whole, as that would be beyond their power and role. It is for legislators to adapt the different tax laws and regulations and integrate them, so that the result is a coherent and systematic, albeit multilevel, tax system.

Tax decentralization is not inherently incompatible with economic integration, nor is it more inefficient. On the contrary, in a context of erosion of some tax bases, it is possible to think of local taxes that would better resist this phenomenon, to the extent that less-mobile bases are taxed. This requires a better process to undertake tax reforms, which will need to consider both the integration in the European Union and the highly decentralized nature of the Spanish State of Autonomies. This process will need to take into account the analysis of the expenditure side. All reforms to further increase tax decentralization had as one of the objectives the increase of Autonomous Communities’ fiscal responsibility. This can only be fully achieved if the exercise of tax powers is also linked to the degree of tax expenditure and the principle of fiscal and budgetary stability.

In any event, the concern for greater coordination in the exercise of regulatory powers and in the tax assignment mechanism is far from new. But it has become even more serious in recent years. In 2014, the so-called “Lagares Report” proposed eliminating the wealth tax (proposal 54), harmonizing the inheritance and gift tax (proposals 55 et seq.), and partially abolishing the transfer tax and stamp duty (proposals 62 et seq.) (Ministerio de Hacienda y Función Pública, 2014). In 2017, the Commission on the Autonomous Region Financing System (para. 42 and 44, p. 18) also proposed a sort of regulatory harmonization.

Finally, and much more recently, the (2022) White Book (Libro Blanco) for tax reform has also stressed the need to better coordinate the exercise of tax powers in a multilevel setting (Ruiz-Huerta). In particular, the case of wealth taxation (including the inheritance and gift tax), the race to the bottom has become obvious, and the problem is no longer that of clashing with EU law, but rather the disappearance of such taxes, which no longer would be a piece of the system to ensure better equality overall. The White Book states:

The exercise of regulatory powers over these taxes has involved an evolution that calls for reflection on the need to find an optimal framework of balance between fiscal co-responsibility, as an expression of financial autonomy, and the requirement to guarantee the application of wealth taxation with criteria of efficiency and equity, in accordance with the principles of justice in the tax system proclaimed in article 31.1 of the Constitution. (Ruiz-Huerta, 2022, p. 619)

In this context, the legal system has tools that allow, on a regular basis, better coordination of the exercise of taxation powers by the autonomous communities and local entities. All this considering that the Spanish Constitution does not design a single model of regional or local financing, but rather a legal structure with different options, in which the central government has ample powers to coordinate the system and ensure that the different pieces fit better together.