https://orcid.org/0000-0002-3780-6361
![Sample our Urban Studies journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110826/TOC-Subject-Sample-2021-Urban-Studies.jpg"})
ABSTRACT
Modern states are organized in multilevel governance structures with economic and political authorities dispersed across them. However, although there is relatively widespread consensus that this form of organization is preferable to a centralized authority, the same cannot be said about its jurisdictional design – that is, how to transfer authority from central states to both supra- and subnational levels. This lack of consensus also exists in contexts with explicit initiatives to strengthen political ties such as the European Union (EU), and even within EU member countries, a situation which is aggravated by the relative scarcity of contributions that measure the advantages and disadvantages of different territorial organizations. We explore these issues through a study of one EU country, Spain, whose provincial councils (diputaciones) are often the subject of debate and controversy due to their contribution to increasing public spending and their purported inefficiencies, corruption and lack of transparency. Specifically, we combine a variety of activity analysis techniques to evaluate how they impact on local government performance. Results suggest that, in general, the presence of a provincial council has a positive impact on local government performance, but when their activity levels are too high, the effect can become pernicious.
ACKNOWLEDGEMENTS
The authors thank Carles Gríful Miquela, Miguel Ángel Malo, Cristina Pita, Rosario Gandoy, Leticia Blázquez, Carmen Díaz Mora, Francisco Puerta Seguido, Consuelo Alonso García, José Baños as well as the participants at the XXVIth Encuentro de Economía Pública, Oviedo, Spain, January 2019. The authors are particularly grateful for the comments of three anonymous referees and the associate editor, Raquel Ortega-Argilés, which contributed to the overall improvement of the paper. The usual disclaimer applies.
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.
Notes
1. The year 1984 is the first in which all the regional governments had a budget for the entire period.
2. NUTS is Nomenclature des unités territoriales statistiques (in French); LAU is local administrative units and refers to the classification of territorial units for statistics used in the European Union. LAU-1 and LAU-2 levels correspond to former NUTS-4 and NUTS-5 levels, respectively.
3. This literature was surveyed recently by Narbón-Perpiñá and De Witte (Citation2018a, Citation2018b). We do not extend the literature review to the performance of other public sector bodies for reasons of space.
4. For instance, Seifert and Nieswand (Citation2014) examine the case of France, although with different aims, methods and objectives to our study.
5. Municipalities and provinces are the two basic forms of territorial organization at the local level. However, other local-level institutions exist such as the local government bodies in the islands (cabildos and consejos insulares), territories smaller than municipalities (aldeas), associations or consortiums of neighbouring municipalities (mancomunidades), districts (comarcas), and metropolitan areas.
6. Law 27/2013 on the rationalization and sustainability of the local administration substantially modified some articles of Law 7/1985 regulating the local system with the aim of clarifying municipal powers. The new law also had an important effect on the competencies and institutional role of the intermediate local governments, as will be shown below.
7. Specifically, municipalities are classified into four population tranches: < 5000 inhabitants; between 5000 and 20,000 inhabitants; between 20,000 and 50,000 inhabitants; and > 50,000 inhabitants.
8. Chartered provincial councils (Diputaciones de Régimen Foral) in the Basque Country are special provincial councils in which the provinces are constituted as ‘historical territories’. They differ from the common provincial councils (Diputaciones de Régimen Común) in terms of both powers and financial resources.
9. Island councils, known as Cabildos in the Canary Islands and Consejos Insulares in the Balearic Islands, are the governing bodies that assume the responsibilities of the common provincial councils (Diputaciones de Régimen Común) in the islands. In the case of the Canary Islands, the scope of the Cabildos does not coincide geographically with the provincial boundaries as each island has its own island council.
10. As stated in the introduction, there has been considerable public debate about the presence and/or reform of this intermediate level of government. Favourable positions argue that provincial councils provide administrative help and technical support, ensuring the effective provision of municipal services and facilities for those municipalities with severe problems due to their scale. On the other hand, arguments against the existence of provincial councils point out that the corporate governance mechanisms are unclear (since provincial councillors are not directly elected), and that in some cases some local institutions could be duplicating the work of others. However, it is important to note that the literature on decentralization is not conclusive, and this debate has been framed entirely in political and qualitative terms. Quantitative evidence is therefore necessary to complement the discussion.
11. We do not include a hypothesis to test the population size effect. The paper focuses on the role the provincial councils (diputaciones) play in municipalities’ performance (in which we also take into account population size), but not on the impact of municipal size on municipal efficiency as an end in itself (which has been widely studied in previous literature, as explained above).
12. As stated above, the latest local regulation reforms had a particular impact on the provincial councils in the local government framework. The reforms aimed to clarify the tasks performed by the provincial councils and strengthen their responsibilities in terms of their cooperation with and coordination of municipal services.
13. If provincial councils detect that the costs of the services provided by the municipalities are higher than those of the services coordinated or provided by the provincial councils, they should offer their collaboration through a more cost-efficient management that would reduce the costs.
14. Given that in public sector outputs are established externally (Balaguer-Coll & Prior, Citation2009) and goods and services are frequently unpriced due to their non-market nature (Kalb, Citation2014), we adopt a cost-minimization approach, a strategy widely applied in previous studies on municipal efficiency.
15. Non-parametric methods do not impose a particular functional form and allow for the simultaneous modelling of several inputs and outputs. See Fried et al. (Citation2008) for further details on efficiency measurement.
16. These two methods have been widely applied in the previous literature to estimate long-term efficiency scores (Drew & Dollery, Citation2015). They construct a frontier including all the DMUs during the period of study, which in the intertemporal analysis model is the entire period, whereas in the windows analysis it corresponds to shorter periods of time. However, in both methods, each DMU in each year is regarded as a different observation, without considering any panel structure of the data. The result is that each observation is compared not only with other observations but also with itself in different time periods.
17. At the municipal level, some previous studies have already used this technique to analyse the performance of different groups of municipalities operating in different contexts. For instance, Balaguer-Coll et al. (Citation2013) control for the environmental conditions that affect municipal efficiency by using meta-frontier techniques in combination with cluster analysis. Other examples include studies analysing organizational arrangements and alternative ways of managing municipal waste collection (Pérez-López et al., Citation2016), urban public transport (Campos-Alba et al., Citation2020), wastewater disposal services Blaeschke and Haug (Citation2018) or drinking water services (Suárez-Varela et al., Citation2017), among others.
18. Given that an increase in the ratio has a negative correlation with the gap between the local frontier and the meta-frontier, O’Donnell et al. (Citation2008) referred to this measure as a ‘metatechnology ratio’ in order to avoid confusion.
19. Listed in Article 26 of the Spanish local government law (Ley reguladora de Bases de Régimen Local).
20. Spain differs from other European countries in the areas of education, care for elderly and health services, which are not responsibilities at the municipal level.
21. The ‘curse of dimensionality’ implies that an increase in the number of variables, or a decrease in the sample under analysis, entails higher efficiency scores (Daraio & Simar, Citation2007).
22. Municipalities are mainly characterized by their territorial diversity and heterogeneity in terms of size.
23. Appendix F in the supplemental data online provides a robustness check that compares the overall cost-efficiency results for the traditional cross-section DEA and the panel data DEA.
24. The technology gap determines the importance of each effect, indicating the extent to which the different municipalities’ conditions cause municipal inefficiency. However, if in a specific group there are some particular units that present high inefficiency scores, the technology gap does not take them into account because the local frontiers consider the ‘best-practice’ units in the specific group (i.e., the efficient municipalities).
25. For other studies dealing with moderating effects, see, for instance, Andrews et al. (Citation2011).
26. Simar and Zelenyuk (Citation2006) adapted the Li (Citation1996) test for efficiency applications using DEA and free disposal hull (FDH) via bootstrapping techniques.