CONSTRAINTS ON POLITICAL AUTONOMY IN DECENTRALIZED SPAIN? ANALYSING THE IMPACT OF TERRITORIAL FINANCING AND GOVERNMENTS’ IDEOLOGY ON REGIONAL POLICY-MAKING

I. introduction[up]

The effects of decentralization in States is a major issue in comparative literature from various disciplines, mainly due to the diversity of social, economic and political impacts that this process may have across the territory. Hence, most of these studies focus on the potential inequalities that may occur between regions ( ‍McEwen, N. and Moreno, L. (2005). The territorial politics of welfare. Oxon: Routledge/ECPR.McEwen and Moreno, 2005;  ‍Gallego, R. (dir.). (2016). Descentralización y desigualdad en el estado autonómico. Valencia: Tirant lo Blanch.Gallego, 2016;  ‍Gallego, R. (dir.). (2014). Descentralització i autonomia política: L’impacte de la ideologia i el finançament territorial en els models sanitaris de Catalunya i Andalusia. Barcelona: Institut d’Estudis Autonòmics.Gallego, 2014;  ‍Guillén, A. M. and León, M. (2011). The Spanish welfare state in European context. Surrey: Ashgate.Guillén and León, 2011;  ‍Rodríguez-Pose, A. and Gill, N. (2004). Is there a global link between regional disparities and devolution? Environment and Planning, A (36), 2097-‍2117. Available at: https://doi.org/10.1068/a362.Rodríguez-Pose, 2004). However, in this article, we start from the basis that the States are, in fact, geographically diverse in their political, economic and social conditions and that this holds true whether they are unitary or decentralized. Therefore, we understand that decentralization will have different impacts on different territories because these are, to begin with, different. That is, even supposing that the powers and decentralized resources are equivalent in the different territories of a State, their governments will deploy actions in different political, social and economic contexts, so it is expected that the content and outcome of their performances will also be different.

In this regard, our research interest is the impact of decentralization on one aspect of territorial diversity, specifically on the self-government policy discretion of territories, i.e. on the political autonomy of sub-central governments to make their own policy decisions. But what are the factors that condition the margin for policy discretion of these governments as an expression of their political autonomy? The literature has explored several factors that condition policy discretion, such as productive structure, social capital, political culture, government ideology, social capital, political context, budgetary resources, previous policy paths… However, a dichotomy has arisen from the political debate in decentralized States, focusing on two such factors: to what extent are sub-central governments limited in their autonomy by the models of territorial financing, or to what extent do the factors of political will and ideological orientation of those governments primarily determine the direction and scope of their decisions and action? Also, in the academic debate, these same main determinants of the discretion to govern have been identified and explored: the budgetary resources available, on the one hand, and the ideology of the political forces in the government in question, on the other. The influence of resource allocation between levels of government has been studied by the normative principles of fiscal federalism, and the impact of ideology in government decisions has been one of the centres of interest in the literature on public policy.

This research uses these approaches to focus on the experience of decentralization in Spain, where this debate is particularly intense both in political and academic venues, and which may shed light for further comparisons with other decentralized or federal States. Numerous studies have found that, although located in a common legal and institutional framework, the Spanish Autonomous Communities (AC) have expressed their capacity of self-government by taking different public policy options. Since their creation in the early 80s, the AC have played an essential role as legislators and managers of complex systems of welfare services for citizens. Thus, within their margin of autonomy, the regional governments have developed their own policies of income and spending that have enabled them to develop public management policies that have produced differing models of regulation and service provision ( ‍Gallego, R. , Gomà, R. and Subirats, J. (2005). Spain, from State Welfare to Regional Welfare. En N. McEwen y L. Moreno (eds.). The Territorial Politics of Welfare (pp.103-126). London: Routledge.Gallego et al., 2005;  ‍Gallego, R. and Subirats, J. (2011). Regional welfare regimes and multilevel governance. En A. M. Guillén y M. León (eds.). The Spanish welfare state in the European context (pp. 97-‍118). Oxon: Routledge.Gallego and Subirats, 2011, 2012;  ‍Moreno, F. J. (2009). Del sistema sanitario de la Seguridad Social al Sistema Nacional de Salud descentralizado. En L. Moreno (ed.). Reformas de las políticas del bienestar en España. (pp. 101-‍136). Madrid: Siglo xxi.Moreno, 2009). Therefore, we ask: Why have regional governments taken different public policy options? In what way, and to what extent, were their policy decisions and actions conditioned by the financing model and by their different ideologies?

To explore this question, we have selected two AC —Catalonia and Andalusia— which are contrasted, among other things, by the options that they have developed for their healthcare models. Healthcare is the budgetary sector with greatest weight in the hands of regional governments (they spend more than a third of their budgets on it), being an essential element of their respective welfare systems. While these AC show similarity in relation to their political and normative discourses on the right to healthcare and in relation to the levels achieved regarding the extent and intensity of health coverage, numerous studies show that the contrasts are profound if we look at the models for the governance and management in the provision of health services: Andalusia reinforced and deployed a system of direct public provision based on an integrated hierarchical model, being built on the model of the structures transferred by the central government; Catalonia set up and consolidated a model of indirect public provision added to the structures of the integrated model transferred by the central government, and which has been based on a contractual network in which the diversity of the legal nature of providers prevails, as well as the interrelationship between public and private healthcare activity. These options for different models of healthcare governance were initially developed under governments of political continuity for more than two decades in both cases: the Spanish Workers’ Socialist Party (PSOE) in Andalusia since the first regional elections of 1983 and the centre-right, nationalist Convergència i Unió (CiU) in Catalonia from 1980 to 2003. Moreover, these developments have been accompanied by controversy over the territorial financing model and its impact on the political autonomy of the ACs, which has always been present in the general political debate.

Given this scenario, our study focuses on the first decade of the 2000s, during which we observe decisions by the Andalusian and Catalan regional governments involving changes of direction in their healthcare governance policies. In Catalonia, after more than two decades of consensus on the impossibility of reforming the Catalan Institute of Health (that is, the set of structures for integrated, direct service provision, transferred by the central government), a law was negotiated and approved that transformed it into a public company. In Andalusia the option of transforming the Andalusian Health Service into a public company was rejected and, on the other hand, a network of High Resolution Hospital Centres was developed, with different management formulas and which was external to the prevalent integrated model. Looking at the financing-ideology dichotomy, in both cases, these decisions were made in a context that was expansive in various ways. In late 2001, the transfers of powers to all regions, including healthcare, were completed, although Andalusia and Catalonia had already had this since the early 80s; and the territorial financing model was modified so that the ACs would increase their resources and earmarking would be eliminated. In addition, the economic environment was upward; therefore, decisions during that decade were not a reaction to the changing economic cycle. In Andalusia such decisions began to be taken prior to 2001 and under continuity of the PSOE in the regional government. In Catalonia such decisions began after the elections of 2003, after an ideological shift in government —namely, by a coalition government of pro-independence forces and left-wing catalanist forces representing the first ideological shift in the Catalan government since 1980. All this gives us a path of broad contrasts and allows us to ask: How can we explain these decisions, again disparate, in the policies of public healthcare management of Andalusia and Catalonia? Are the decisions being made highly conditioned by the territorial financing system? Or are they the result of ideological options arising from the exercise of self-government?

This paper addresses these questions by contributing new evidence to the intense debate around the impact of these two variables on the political autonomy of regions in Spain. To do so, it provides a detailed analysis of updated, empirical data that include, on the one hand, the official data on the public resources available to the two ACs being analysed (income coming not only from the territorial financing model, but also from outside it –namely, income decisions by the regional governments themselves), as well as data on their spending decisions. On the other hand, the data analysed include official documents and semi-structured interviews to 20 key officials who either participated in the decisional processes analysed or have expert knowledge about them. The position and number of the interviewees were: senior politicians of the Catalan and Andalusian governments (9), public management and healthcare sector professionals (5), trade union representatives (2), members of non-profit organizations (2) and academic researchers (2). The guidelines of the interviews include the following general topics: characteristics of the health policy sector in both AC, availability and constraints in budgetary resources, intergovernmental relationships and key decisions in health public management (particularly about the Catalan Institute of Health and the network of High Resolution Hospital Centres in Andalusia).

In this paper we present, first, the theories of reference for this approach –fiscal federalism and policy analysis. Second, we analyse the availability of budgetary resources of Andalusia and Catalonia, distinguishing their volume and origin (internal or external to funding model, and tax collection or grants), together with its implications. Third, we explore the influence of the ideology of the political forces in power in each region in order, fourthly, to make a joint interpretation of both cases. We rely on the analytic dialogue between theory and empirical data to then generate an interpretation of the two selected cases. We integrate the interpretation of the influence of both factors —financing and ideology— in these cases, to finally present the findings. Understanding the elements that condition the decision-making processes at these levels of government and, therefore, the nature of their political autonomy, can contribute to comparative knowledge and debate about the operation and functionality of decentralization.

II. FISCAL FEDERALISM AND POLICY ANALYSIS: THE THEORY OF REFERENCE[up]

How has the need to decentralize revenue-raising capacity and expenditure of the central government to sub-central governments been argued? The “Oates decentralization theorem” ( ‍Oates, W. E. (1972). Fiscal Federalism. Nueva York: Harcourt Brace Jovanovich.1972) may be considered the starting point of the theory of fiscal federalism, showing that the diversified supply of public goods by sub-central governments provides more welfare (more efficiency) than the uniform supply by a single government, to the extent that a better adjustment of supply to the preferences and needs of each territory occurs. Therefore, the heterogeneity of preferences is one of the main arguments that justify the existence of sub-central governments. The same argument is applied to healthcare, a decentralized function in many States. Therefore, different decisions about the provision of this service would have positive effects on the efficiency, according to the principles of Oates, but could have negative effects on the degree of inter-regional equity, which should be considered ( ‍Ayala, L., Martínez, R. and Ruíz-Huerta, J. (2013). Desigualdad y redistribución en los países de la OCDE. En Informe sobre la desigualdad en España (pp. 25-‍73). Madrid: Fundación Alternativas,.Ayala et al., 2013).

From this point, the question arises: On what criteria should design models for territorial funding be based? According to the main principles of fiscal federalism, the financing model for sub-central government must reconcile the exercise of financial autonomy with a “certain” principle of equity. Thus, revenue sharing must meet the following principles: a) Financial autonomy: all governments should be able to take decisions both on spending (how much and on what they spend) as on income (how much and from where they get their resources). This is only possible if funding is largely not earmarked and comes from own taxes; b) Vertical balance: the different levels of government should have the same degree of coverage of their spending needs; c) Horizontal equity: governments that are part of the same level must have an equal or similar volume of resources to meet their spending responsibilities, asking the same tax effort of their citizens; d) Other principles are transparency, coordination and institutional loyalty.

If the governments of Andalusia and Catalonia have made different decisions in relation to the provision of healthcare, this may be due to at least two reasons. One, the regional governments have provided these services expressing heterogeneous preferences for health output between AC (which supposedly represent the citizens who voted for them) towards a greater collective welfare. Two, the different behaviour of regional governments could be due to the characteristics of the financing model and its results, mainly the available per capita resources, their origin and the decisions that have been taken on them.

In the Spanish case, these options are raised within an institutional framework that is consistent for all regions in certain aspects, such as: the scope of the basic legislation, invasive (or not) effect of the doctrine of the Constitutional Court, the control period of the central administrative unit on the regional territory, the administrative scope or extent of earmarked financing condition the degree of the sub-central government policy autonomy. In this context, key areas of the Welfare State are decentralized to the AC, but framed in equal rights and market unity. That is, the relevance of the various preferences and territorial needs is recognized, but the ultimate responsibility for the provision is preserved by the higher level of government.

However, from the examination of public policies, other factors are included in the analysis that also influence the development of policies and explain that, even within similar institutional frameworks, sub-central governments make decisions on different public policies. Thus, the ideology of the political forces in power not only results in the construction of a discourse supporting government decisions, but also a priori steers the direction of public policy. The ideology implies a certain identification and definition of the issues to be addressed, bringing a preference for a certain intensity, extent and type of intervention in such issues, and supports the proposal of a particular model of relationship between the external stakeholders and government institutions.

Another factor influencing government action is the nature of the public policy sector in question. This includes, inter alia: the relative weight of public and private provision, management tools used, the characteristics of the sector actors that influence policy making, as well as the type of interaction these have with the political actors. Although these features have a certain structural base, they are also factors that typically become an implicit or explicit object of intervention and modification by government action. Finally, all this unfolds in the context of dynamic contextual aspects of economic, political and social types - both specific for the territory and relevant in a relationship of multilevel governance. Thus, the analysis of capacity constraints and government action necessarily integrates all these factors (financing, ideology and structural and relational characteristics) in the interpretation of the political process.

III. THE REGIONAL FINANCING MODEL: SAME BUDGET CONSTRAINTS FOR ANDALUSIA AND CATALONIA?[up]

In the 2000s, the governments of Andalusia and Catalonia took different decisions regarding healthcare. We look at whether these differences can be explained by the characteristics of the financing model. We assume that the model and its evolution have a significant explanatory weight in the political decision-making capacity of regional governments ( ‍Vilalta, M. (2005). La reforma del sistema de financiación de las comunidades autónomas. En VV. AA. Estado autonómico: integración, solidaridad, diversidad (pp. 635-‍652). Madrid: Instituto Nacional de Administración Pública.Vilalta, 2005,  ‍Vilalta, M. (2007). Los problemas actuales de la financiación autonómica. En S. Lago (dir.). La financiación del Estado de las autonomías: perspectivas de futuro (pp. 11-‍40). Madrid: Instituto de Estudios Fiscales.2007;  ‍Bosch, N. and Durán, J. M. (eds.). (2008). Fiscal Federalism and Political Decentralization. Lessons from Spain, Germany and Canada. Chentelham: Edward Elgar. Available at: https://doi.org/10.4337/9781848442719.Bosch and Duran, 2008;  ‍Bosch, N. and Vilalta, M. (2008). Informe sobre el finançament de les comunitats autònomes. Any 2006. Barcelona: Generalitat de Catalunya, Departament d’Economia i Finances.Bosch and Vilalta, 2008;  ‍Ruiz-Huerta, J., Benyakhlef, V. and Vizán, C. (2010). Las comunidades autónomas ante la crisis económica: impacto territorial de la recesión, políticas autonómicas de reactivación y tensiones en las cuentas públicas. En M. Tornos. Informe comunidades autónomas 2009 (pp. 89-112). Barcelona: Instituto de Derecho Público.Ruiz-Huerta et al., 2010) but we want to investigate how and to what extent this influence occurs. So we ask: How does the financing model distribute resources between these territories? Have Andalusia and Catalonia had the same per capita budget? What part of the budgets of the Andalusian and Catalan governments comes from the regional financing model and how much comes from other sources (debt, etc.)?

In accordance with the stated principles of fiscal federalism, we can affirm that the agreed models described below are characterized by: a lack of financial autonomy; a lack of global resources to address spending responsibilities; erratic and unjustifiable results according to the principle of horizontal equity.

IV. HEALTH POLICY DECISIONS IN ANDALUSIA AND CATALONIA: THE INFLUENCE OF IDEOLOGY?[up]

Regarding the influence of ideology as a conditioning factor of government capacity we ask, for the decade under investigation: What was the per capita health budget in Andalusia and Catalonia? How have these autonomous communities structured their health spending? What model of service delivery and management has been developed in each?

V. FINANCING MODEL AND IDEOLOGY: HOW DO THEY INFLUENCE REGIONAL POLITICAL AUTONOMY?[up]

After more than two decades in which Andalusia and Catalonia consolidated different models of public healthcare provision, in the early 2000s, their governments undertook reforms in this area which again were different. Andalusia boosted the CHAREs bringing public governance closer to private management instruments, but consolidated and strengthened the model of direct public provision of SAS. Catalonia maintained the contracted network, but transformed the ICS direct public provision into a public company. These decisions again different; do they respond to an ideological option or have they been conditioned or constrained by the territorial funding model?

The period analysed in this study is part of an upward and expansive economic cycle for both regions. Catalonia and Andalusia had the same level of competence in healthcare provision since the early 1980s. In the analysed period, the reforms of the territorial funding model of 2001 and 2009 expanded their powers of income and expenditure and increased the volume of their available budgetary resources - both for those coming from the territorial funding model and for those from outside the model. This allowed the per capita healthcare expenditure to increase exponentially until 2008 in Andalusia and until 2010 in Catalonia, then a descent began, linked to the crisis. However, until 2008, Catalonia received annually fewer resources per capita than Andalusia from the funding model, accentuating the difference until the reform of the model in 2009 reversed this situation. According to officials interviewed for this study, increasing available resources always facilitated the proposal for reforms, and, as they indicate, in both cases the proposed reforms required increased resources. Throughout the decade under study, the relative weight of resources from the funding model to total revenues of the regional government has been greater in Andalusia than in Catalonia, and in both cases this weight has decreased since 2009. In the budgets, for 2008, allocated income taxes represented 88 % of revenues in Catalonia and 55 % in Andalusia, making the latter more dependent on transfers or grants from other levels of government. In any case, resources per capita in both AC were finally similar, because Catalonia resorted to financing from outside the model.

The greater weight of taxes and debt implies greater responsibility in accountability, and it is relevant to note that, throughout the period, healthcare spending per capita has been higher in Catalonia than in Andalusia. Also, the structure of health spending reveals a policy option for different delivery models. In Andalusia, the remuneration of staff and intermediate consumption has increased from 68 % to 69 % of total spending, indicating that the weight of the integrated model (direct public provision) has not changed despite the analysed reform. In Catalonia, the weight of the integrated model (remuneration of personnel and intermediate consumption) has increased from 42 % to 54 % of total health expenditure, while the weight of the model of indirect provision (private partnerships) has decreased from 33 % to 24 %. This indicates a strengthening of the direct public provision model following the reform and throughout the period. Therefore, budget data indicate that in both cases the reform reinforced the model of direct public provision, which fits with the discourse of its promoters in both cases, but not with the diversification of management involving both the CHAREs as well as the conversion of ICS to a public company.

The influence of the ideology factor also raises ambiguities, particularly regarding the link between propositional principles traditionally assigned to ideologies on the left-right axis, on the one hand, and the programmatic proposals and government actions of the parties, on the other. During the last two decades of the xx^th century, Andalusia was ruled by a leftist party (PSOE) and Catalonia by a nationalist coalition of the centre-right (CiU). This would strengthen the traditional bond of leftist ideology with a model of direct public provision in an integrated structure (Andalusia) and a right-wing ideology with a model of indirect public provision (Catalonia). However, in early 2000, in Andalusia there is continuity of the PSOE in government but they undertake reforms that involve the diversification of forms of management and distancing from the integrated model. In Catalonia, there is an ideological shift to the left in government (PSC-ERC-IC), but the new government does not question the contractual delivery network and transforms the single integrated structure of the healthcare system —the ICS— into a public company, away from the integrated model and opening the door to disaggregation and indirect public provision in the ICS space.

However, if we understand ideology as a complex link between principles and interests expressed through a programmatic and justifying discourse of government action, the analysis of these cases shows that ideology has a great influence on government decisions and if the scope of government action is not well determined —being limited by available resources— it does determine its direction. Therefore, as seen in these two cases, reforms of healthcare management that choose management tools that depart from direct public provision or integrated model do not necessarily imply an ideological shift to the withdrawal of the public sector in the provision of services, but rather that budget indicators can reveal the reinforcement thereof. The officials interviewed for this study insist in the political will for this, whether or not it be the case. Moreover, they explicitly insist that not only are they moving in an ideological axis of left and right, but also in a territorial centre-periphery axis that shows in each case projects other than identity affirmation.

VI. CONCLUSIONS[up]

The interest of this research is the impact of decentralization of States on the political autonomy of sub-central governments. In both academic and political debates on the factors that condition the policy discretion of those governments, there are voices that highlight the model of regional financing and voices that insist on the political will and the ideological orientation of parties in power, as alternative explanations. To explore this question, this article presents a study in decentralised Spain, on: Why have regional governments taken different public policy options? In what way, and to what extent, were their policy decisions and actions conditioned by the financing model and by their different ideologies? Theories on fiscal federalism and theories of public policy analysis have guided the analysis.

We selected the regions of Andalusia and Catalonia where, after more than two decades in which they consolidated different models of public healthcare provision, in the early 2000s, their governments undertook reforms in this area which again were different. Andalusia boosted the CHAREs bringing public governance closer to private management instruments, but consolidated and strengthened the model of direct public provision of SAS. Catalonia maintained the contracted network, but transformed the ICS direct public provision into a public company. Do these decisions respond to an ideological option or have they been conditioned or constrained by the territorial funding model?

In a context of upward economic cycle, in 2001 the territorial financing model was modified so that the CA increased their resources and the economic transfers they received from the central government for health were no longer earmarked. The analysis of the evolution of resources from the territorial funding model shows that the model approved in 2001 always provided more resources per capita to Andalusia than to Catalonia and, instead, starting from the agreement of 2009, this situation was reversed. Thus, while per capita resources from the model have been, during the years 2001-‍2008, higher in Andalusia than in Catalonia, in the end both regions end up having a similar volume of resources per inhabitant. Catalonia has therefore sought a greater share of its income from outside the model (higher debt and greater use of its regulatory powers over taxation) and, from 2005, health spending per capita in Catalonia has tended to be higher than in Andalusia. Key officials interviewed for this study agree in interpreting that the widespread increase in resources facilitated the reforms undertaken in both AC, that is, the scope of government action. In Catalonia, a reform that had been considered by actors of all ideological lines as unapproachable for more than two decades was undertaken, and in Andalusia, a reform that had hardly been considered in previous years was deployed.

Faced with the different orientation of these reforms, these sources also agree that, as in previous decades, the content and direction of them respond to the political preferences of the political forces in government. However, the influence of the ideological factor in these decisions does not provide conclusive evidence: in Andalusia, under the continuity of the PSOE, there are reforms that involve the diversification of forms of management and their distancing from the integrated model —which does not fit the traditional preference of the left for the model of direct public provision that had been defended until then—; in Catalonia, there is an ideological shift to the left in the government, but the new government does not question the contractual provision network and transforms the single integrated structure of the healthcare system into a public company, thus moving away from the integrated model and opening the door to disaggregation and indirect public provision in that space.

These results confirm that the nature and volume of budgetary resources of sub-central governments determine the extent of political autonomy, but not its direction, which would be marked by ideological choices. However, these findings raise new questions. According to the theories of fiscal federalism, greater weight of own resources in the budget as opposed to transfers/grants from other governments brings a greater responsibility for accountability towards citizens. But to what extent are citizens and public decision makers themselves aware of the degree of dependence on their resources? Finally, according to the political theories, right and left tend to identify with certain options of management models; but when the evidence does not show it to be so, does the management model stop being ideologically relevant?