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 “Andersson, Tommy (2017), Matching practices for elementary schools – Sweden, MiP Country Profile 24.”
Relevant country background
In Sweden, children have the right to start preschool from the age of six. Preschool is not compulsory but in practice around 98 percent of six-year-olds attend preschool. Municipalities are obliged to offer and arrange for preschool for all pupils that request to be enrolled. School becomes compulsory for children from the autumn term of the year they reach the age of seven, and compulsory school attendance ceases at the end of the spring term of their 9th school year (i.e., by the time children are 16 years old). Preschools are typically combined with primary schools but secondary schools are separate. As a consequence, most children have to change schools upon finishing primary school.
Private schools have always existed in Sweden but a new legislation in 1992 (Friskolereformen) made it easier to start and operate private schools.[1] Currently, 15 percent of the children are enrolled in private schools. The current legislation states that municipalities have to reimburse private schools for each pupil (the so-called Skolpeng). The exact amount is determined at the municipality level but it is regulated by law (Skollagen). Roughly, private schools should, for each pupil, be reimbursed at least 85 percent of the corresponding costs at the public school (the reimbursement level is higher if a pupil has special needs). The eligible costs include all the costs associated with teaching but also the costs of health care services, school lunch, etc. In general, both public and private schools must provide books, teaching materials etc. to the pupils free of charge and all school activities must be for free.
School seats in public schools are allocated based on Swedish law (Skollagen, ch. 9, par. 15) and in particular on the school ranking submitted by parents on behalf of their children and (relative) distance to the schools (Närhetsprincipen). The law leaves some room for interpretation and the central government has delegated some regulatory power to the municipalities (local self-government is part of the Swedish constitution since 1974). In this sense, the law dictates basic conditions but it is up to the local municipalities to interpret the law. Private schools are also regulated and supervised but follow a different system for admissions.
Summary box
Organization of education | The Swedish National Agency for Education (Skolverket) |
Stated objectives of admissions policy | Every pupil has the right to a seat. |
Who’s in charge of admissions? | Local authorities (municipalities). |
Admissions system in place since | 2010 with the main principles established by law. Interpretation and admission practices vary across local authorities. |
Available capacity | Upper and lower limits to class sizes imposed by local authorities. |
Timing of admissions | Starting date for registrations is set by local authorities. |
Information available to pupils prior to admissions period | Available schools and capacities. |
Restrictions on preference expression | No legal restrictions, but most municipalities allow children to rank at most three schools. |
Matching procedure | Non-algorithmic as long as legal and local guidelines are respected. |
Priorities and quotas | Priorities based on submitted preferences and (relative) distance. Local variations based on, e.g., sibling priority exist. |
Tie-breaking | Local variations. |
Description of current practices
Skollagen (the law, henceforth) dictates some minimal rules that the municipalities have to respect when assigning children to public schools and municipalities define further basic principles. Even if this means that municipalities can adopt in theory different rules, in practice almost all municipalities apply the same basic principles.
The law specifies that children should be placed at schools based on their ranking of the schools (submitted by the parents) and the priority assigned to the children by the schools. In practice, priorities are determined on the basis of relative distance to schools even if other factors, such as sibling priority, may also influence the priority of a child at a specific school. Parents are typically allowed to rank up to three schools but there are local variations.
Based on the rankings submitted by the parents, it is up to the municipality to make sure that school seats are allocated in such a way that the law and the local basic principles are respected. Often, this is done in a non-algorithmic fashion (i.e., not using specific matching software) even if there are exceptions (e.g., in Botkyrka). It should also be noted that several Swedish municipalities are considering to use algorithmic software (examples include Stockholm and Lund), and the author of this document is providing advice to some of these municipalities together with some colleagues. If a child cannot be placed in one of the schools ranked, the child is assigned to some school in the surroundings (typically within two kilometers walking distance). In practice, almost all children are placed in one of the schools that they rank (in Stockholm 2015, around 99 percent of pupils were placed in one of their top three ranked schools).
A key issue with the current Law is that the notion of relative distance is not precisely defined. A commonly adopted interpretation of relative distance (e.g., in the city of Stockholm) is to the difference between the walking distance to the ranked school and the walking distance to an alternative school (e.g., the school that is closest to the pupil’s home or the next school in the ranking). In other words, the more the pupil would lose by not getting a place in the chosen school, the higher priority for the pupil.
Admission practices for private schools differ from the practices of public schools. Private schools need not use the same criteria for determining priorities. Instead, the Swedish Schools Inspectorate (Skolinspektionen) has accepted selection criteria based on, e.g., application date, sibling priority, geographical proximity, and priority for pupils in associated preschools. Private schools need not base their priorities on all these criteria, but can choose one or more of them. These priorities must be made public. Private schools often have separate application platforms meaning that children that apply for private schools may have to submit multiple applications (one for each private school). Public schools use the same platform within a municipality.
Performance
There is no summary statistics available for Sweden, but it is often reported that between 85 to 95 percent of all pupils get their reported first choice, suggesting that the current system works reasonably well. In 2015, for example, it was reported that 83 percent, 96 percent and 93 percent of the children was assigned to their highest ranked school in Stockholm, Tyresö and Uppsala, respectively.[2]
Recent policy changes
The school law preceding the current legislation dates from 1986. Major efforts to initiate reforms were initiated in 1999 to address, among other things, the instability of the law at the time (the 1986 school law had changed over 70 times between 1986 and 1998, e.g., to adapt to new control systems and regulations). The Education Committee’s proposal was presented in December 2002 (SOU 2002:121), with the stated goals of promoting the quality of education and equality for different age groups. The proposal included a uniform and common school system for all types of schools, new rules and regulations for private schools and stricter supervision and sanctions. After this proposal, work began to develop a new school law. However, the negotiations in the parliament broke down and the proposal was postponed until after the parliamentary elections in 2006. In 2009 a proposal for a new school law was presented. This is the current law, known as Skollagen (SFS 2010:800).
Perceived issues
One perceived issue is that the law is not very specific about how to interpret the notion of relative distance (närhetsprincipen). This has resulted in multiple court cases. A recent court case is the city of Lund. This case concerns a specific school called P. Because this school had fewer seats than the number of students that ranked it as their first choice, the city of Lund decided to assign children to school P based only on walking distance from home. A child (called C) had 2.2 kilometers walking distance to his most preferred school (P) and 3.2 kilometers walking distance to the alternative school (A). C was not admitted to school P and all children that were assigned to school P had a shorter walking distance than 2.2 kilometers to the school. The parents of C argued that the walking distance for C had increased by 1 kilometer by placing him at A instead of P, and that the walking distance for some children that had been assigned to P only would have increased by only 650 meters if they were placed in their alternative school instead of school P. Hence, the parents of C argued that C should be given a higher priority at school P than some other children who were assigned to school P because C would lose more in relative terms by not being assigned to P. Two courts (Förvaltningsrätten and Kammarrätten) have ruled in favor of the parents.[3]
A second issue is transparency and accountability. Indeed, because school admissions are often conducted in a non-algorithmic way, it is not clear whether all local basic principles are respected. In fact, a recent analysis of the Stockholm school choice data from 2015 (Andersson et al., 2015) suggests that local basic principles are sometimes violated.
A third issue is the role of distance in increasingly segregated Swedish cities because it no longer guarantees that all children have access to high quality schools (even if they are allowed to apply to them). This has been recognized by Anna Ekström (The Minister for Upper Secondary School and Adult Education and Training in Sweden) and when she was chairing Skolkommissionen 2015, she actively pushed for controlled school choice in Sweden as a way to address this concern (SOU 2016:38).
Manipulability is a common concern in school choice mechanisms and there are at least three reasons to believe that this is a concern also in Sweden. First, it is well-known that school choice programs where parents not are allowed to rank all alternatives are open to strategic manipulation (Haeringer and Klijn, 2009; Romero-Medina, 1998). This is obviously an issue also in Sweden as parents are typically allowed to rank at most three schools. Second, because the notion of relative distance is not clearly defined, there may be room for manipulation. For example, if the alternative school is defined to be the next school in the ranking, parents obviously have a strategic incentive to rank their most preferred school first and then a school located far away from the home as their second choice. Third, in municipalities where priorities are based on reported preferences, parents have incentives to manipulate their reported ranking. Fourth, the bare fact that rules can be interpreted in different ways may also cause incentives to manipulate.
Finally, there is an ongoing debate about the fact that public and private schools use separate application systems and admission rules, see e.g. Lärarnas Riksförbund (2016).
Existing data
There is no public data available on school choice (neither for public school, nor for private schools). Tommy Andersson (Lund University) has access to data submitted for the Stockholm school choice in 2014 and 2015.
Legal texts
Skollagen (Skollag, SFS 2010:800, Utbildningsdepartementet 2010-06-23) describes minimal rules that the municipalities have to respect when assigning children to public schools.
http://www.skolverket.se/regelverk/skollagen-och-andralagar
Other resources and references
Andersson, T., Andersson, O., and E. Forsell (2015) Ett automatiserat skolval. Presentation för Utbildningsförvaltningen, Stockholm, 11 November 2015 (available upon request)
Haeringer, G., and F. Klijn (2009) Constrained School Choice, Journal of Economic Theory 144, 1921-1947
Romero-Medina, A. (1998) Implementation of Stable Solutions in a Restricted Matching Market, Review of Economic Design 3, 137-147.
The Swedish National Agency for Education (Skolverket) is the central administrative authority for the public school system, private schools and publicly organized preschools.
Municipalities have to define further basic admission principles. These principles are often available on their websites. For Stockholm, see:
http://www.stockholm.se/ForskolaSkola/Soka-skola/
For information about admission practices for private schools, see the website of The Swedish National Agency for Education (Skolverket):
http://www.skolverket.se/regelverk/mer-om-skolans-ansvar/val-av-grundskola-1.210176
SOU 2002:121. Skollag för kvalitet och likvärdighet
SOU 2016:38. Samling för skolan. Nationella målsättningar och utvecklingsområden för kunskap och likvärdighet
Lärarnas Riksförbund (2016) Friskoleurval med segregation som resultat.
Court ruling, HFD 2015, ref. 50, mål nr. 7748-13
Note:
[1] The Swedish term “friskola” is, in this document, translated to “private school” but the term “independent school” can also be used.
[2] See:
http://www.unt.se/nyheter/uppsala/skolval-bara-pa-papperet-3901650.aspx
http://sverigesradio.se/sida/artikel.aspx?programid=103&artikel=6498763
https://viityreso.se/artikel/skolvalet-sa-gott-som-alla-kom-in-pa-sitt-foerstahandsval
[3] A complete description of the case is available at (Court ruling, HFD 2015, ref. 50, mål nr. 7748-13):