This country profile is part of a collective effort by the network members to map matching practices across Europe. If you find it useful and want to refer to it in your own work, please refer to it as “Calsamiglia, Caterina (2014), Matching Practices for elementary and secondary Schools – Spain, MiP Country Profile 17.”
Relevant country background
A seat in a public school is guaranteed to every child starting at age 3. That is when families enroll their children to school, since most schools include both preschool and primary school. School becomes compulsory in primary school, which starts at age 6, but more than 95% of the children attend school earlier. Most public schools include either preschool and primary school, or secondary school. This implies that at the end of primary school children need to be reallocated to a secondary school.
In Spain, there are very few fully private schools. Most of them are public or semi-public (publicly subsidized but privately run), the proportions being 50-50 for cities like Madrid or Barcelona. In principle semi-public schools may not charge tuition fees, but at the end of the day they end up charging for other services that apply to the vast majority of parents. Except for the private schools, all other seats in a school are allocated through a unique centralized mechanism. This profile describes the allocation procedure used in Spain for all public and semi-public elementary and secondary schools.
Spain is divided into 17 autonomous communities. In education, as in other services, the central government has delegated some regulatory power to the autonomous communities. In particular, there are organic laws (ley orgánica) that establish some rules, but autonomous communities can develop these rules further, through what are called decretos (decrees), as long as they do not directly contradict the organic law. More decentralized units correspond to provinces and then municipalities. A large fraction of educational policies are run at the autonomous community and municipality levels.
This is also true for school choice procedures: there is the organic law of education (Ley Orgánica de Educación ) that establishes some criteria that autonomous communities can use to deal with excess demand in schools. Then, each community creates its own specificities through Decretos de Educación. 
|ORGANIZATION OF HIGHER EDUCATION||Mostly public, quite decentralized to autonomous communities|
|STATED OBJECTIVES OF ADMISSIONS POLICY||n/a|
|WHO’S IN CHARGE OF ADMISSIONS?||The municipalities|
|ADMISSIONS SYSTEM IN PLACE SINCE||Since the 1980s|
|AVAILABLE CAPACITY||All children should have a seat guaranteed in some school|
|TIMING OF ENROLMENT||March choice, June allocation and enrollment|
|INFORMATION AVAILABLE TO STUDENTS PRIOR TO ENROLMENT PERIOD||Parents know the rules, and their coarse priorities for each school, but not the tie-breaker lottery|
RESTRICTIONS ON PREFERENCE EXPRESSION
|Parents can only list a limited number of school (usually in the range from 6 to 10)|
|MATCHING PROCEDURE||Boston Mechanism|
|PRIORITIES AND QUOTAS||Priority number based on neighborhood, sibling status and/or socioeconomics|
|TIE-BREAKING||Fair unique lottery|
|OTHER SPECIAL FEATURE|
Description of current practices
The Spanish government defines some minimal rules that all autonomous communities need to satisfy. In turn, the autonomous community defines some further basic principles in addition to those imposed by the state. Municipalities adapt their mechanism to their local contexts (proximity areas according to population densities, etc.), within the guidelines set by the government and the autonomous community. Art. 84 of the organic law (LOE) states that “whenever there is overdemand, admission should be based on siblings status, the presence of parents or tutors in the school, proximity of residence or work place, family income, coming from a large family (3 or more children), or disability of the child or a member of the family” and that “none of them are exclusive”. From there, autonomous communities have defined the weights given to each criterion and included additional criteria.  The main additional criterion is that schools can define their own priority criteria. This option has been implemented in Madrid since 2012 (point e) in Capítulo III of the Decreto de Libre Elección de Centro Escolar de Madrid. 
The size of classes is set to be at most 25. In the recent years it was increased to 28. Schools may increase class sizes, but need to ask for permission to the corresponding autonomous community’s authority; the size can never exceed 28.
School choice is implemented as follows: Families submit a list of schools (the size depends on the municipality, but is never larger than 10) to either the school they list as first choice, or to the central authorities. Schools and central authorities introduce the applications to a centralized computer system that applies an algorithm that corresponds to the Boston mechanism (also known as first preferences first) with priorities based on the criteria defined by the municipality.  The only specificity is that the priority points that are used for tie-breaking at all stages are computed on the basis of your first choice school.  For example, if the applicant lives in Barcelona and applies for a school in his residential priority area (30 points) and he belongs to a large family (15 points), then he will get 45 priority points at all potential ties at any round of the Boston mechanism. Both public and semi-public schools get students assigned through the exact same mechanism.
The assignment mechanisms for elementary and secondary schools are identical. The only difference for secondary schools is that grades and the school where you went to primary school provide the largest number of priority points.
Throughout the country, newspapers report that every year more than 90% of pupils get their first choice, suggesting that the mechanism works well. As Calsamiglia and Guell (2013) show most families apply for the school where they have highest priority, not necessarily the one they like most. Their study exploits a change in the priority structure that happened in Barcelona.
Recent policy changes
Many municipalities change the priorities structure over time: mostly redefining catchment areas and including or excluding individual health issues as a priority criterion. Madrid is one such example where residential priority is being removed.  In 2007, the municipality of Barcelona changed its neighborhood criteria, redefining the areas that constitute neighborhoods. 
Because living in catchment areas is the main channel through which you can get a seat in the best schools there has been great concern about parents cheating and claiming they live at a different address than they really do. Since about 2006, some additional conditions have been imposed for parents to prove where they live. If parents are caught cheating they are expelled from the choice system and are assigned some school that has free seats at the very end of the process.
The removal of residence priority in Madrid has left siblings and priority given by the school itself as the only tiebreakers. The fact that the reasons why schools give priority to families are not clearly specified reduces the transparency and the equality of opportunities of the system.
There are no publicly available data on school choice, but Calsamiglia and Güell have data about submitted preferences for Barcelona.
On performance though, the autonomous community of Madrid, since the 2004/2005 academic, has done external evaluations on children in 6th grade and 9th grade on math and languages and has made them publicly available at: http://goo.gl/wW5Pqs
Article 84 of the Organic Law of 2006 (National guidelines for admissions) http://noticias.juridicas.com/base_datos/Admin/lo2-2006.t2.html#a84
Decreto 29/2013 de la Comunidad de Madrid
Resources and References
 Calsamiglia, Caterina (2011), “School Choice is Spain: Theory and Evidence” Els Opuscles del CREI. http://pareto.uab.es/~caterina/OpuscleCalsamiglia.pdf
 Calsamiglia, Caterina and Maia Güell (2013), “The illusion of Choice: Empirical Evidence from Barcelona”, mimeo UAB https://www.minneapolisfed.org/research/wp/wp712.pdf
 Federation of Teachers (FETE) (2009), “Estudio Comparativo de las Normas de Admisión de Alumnos en los Centros Sostenidos con Fondos públicos en las Comunidades Autónomas” (Comparison of admission criteria in schools across autonomous communities in Spain), Gabinete técnico de la UGT
 See http://www.idescat.cat/cat/idescat/publicacions/cataleg/pdfdocs/dossier13.pdf for an overview of the financial sustainability of semi-private schools in Catalunya.
 See for example the one for the Madrid autonomous community: http://doe.gobex.es/pdfs/doe/2013/530o/13040028.pdf
 The following document describes the summary of the specifics for each autonomous community as of 2009: http://www.feteugt.es/data/images/2009/GabienteTecnico/DOCINFAdmisionalumnos.pdf
 Again, the full Decreto 29/2013 can be found at http://doe.gobex.es/pdfs/doe/2013/530o/13040028.pdf
 There is no particular reason why the Boston mechanism is used throughout Spain. The only aspect that the central authorities regulate is the criteria that municipalities can use to give priorities, but not the mechanism itself. The only reason is that it was easy to implement without a centralized computing system: schools collected first choice applicants and then handed it the rejected applicants to the central authority, which would assign the remaining manually (few of them, since in the Boston mechanism, most families get their stated first choice). Alternative mechanisms require computational power that authorities have not had until now.
 In the standard Boston mechanism, in every round, in order to resolve excess demand in a particular school, applicants are ordered according to the priorities that they have for that school. But in Spain the priority points that applicants have for the first school in the list submitted are used to create the order at any round.
 Cf. decree 29/2013 from April 2013: http://www.bocm.es/boletin/CM_Orden_BOCM/2013/04/12/BOCM-20130412-1.PDF