Statskontoret (The Swedish Agency for Public Management) was tasked by the Commission on the municipalisation of the school system to analyse how the levels of costs and pupil-teacher ratio have varied over time in primary, lower-secondary and upper-secondary school between different municipalities, as well as the link between resources and pupil performance in primary and lower-secondary school.
The fact that a commission on the consequences of the municipalisation was appointed and that this assignment was given to Statskontoret is a result of the problems Swedish schools are facing today, where declining pupil performance and a decreasing equivalence have been noted.
The municipalisation of the school system
Our assignment relates to the municipalisation of the school system. The term municipalisation may cause confusion however. Ever since compulsory school attendance was introduced in 1842, municipalities and central government have had shared responsibility for school activities. The municipalisation that we are investigating involved a shift in this shared responsibility, from central government to the municipalities.
The shift in responsibility was done in two steps. In 1991 municipalities were given full employer responsibility for all school staff. Moreover, the specially allocated government grants were removed. The new government grants would still be intended for education, but would be designed as sector grants and municipalities would be in charge of deciding how to allocate the resources between different school activities.
The second step was a decentralisation reform in 1993 in which the sector grants were incorporated into a general government grant. The municipalities were then given full responsibility for making priorities of how to allocate the resources between primary, lower-secondary and upper-secondary school in relation to other municipal areas of responsibility.
In parallel with the municipalisation, a range of other reforms were introduced into the school system, such as increased opportunities for parents and pupils to choose schools (school choice system), the independent school reform and the upper-secondary school reform in 1991.
School resources in relation to other municipal activities
In the report we examine municipalities' total costs for childcare, primary, lower-secondary and upper-secondary school and for the care for elderly between 1980 and 2010. We have discovered that the municipalities' costs vary over time between their different areas of activity. While the municipalities' total costs for primary and lower-secondary school have not changed significantly since 1980, the costs for upper-secondary school have increased by 50 per cent in fixed prices. Both the costs of childcare and the care for elderly have increased by 70 per cent in fixed prices.
The costs for primary and lower-secondary school as a proportion of the municipalities' total costs for welfare have also decreased, while the costs of care for elderly today make up a larger proportion compared to primary and lower-secondary school.
The cost development in the examined areas can largely be explained by demographic changes and a change in the use of various services across the population, which is linked to political decisions. The large increase in cost for childcare and upper-secondary school may to a large extent be explained by the fact that more children are registered in childcare and that a greater number of young people attend upper-secondary school today than three decades ago.
Cost per pupil in primary and lower secondary school is on the same level as in the early 1990s
In the 1980s, the costs per pupil in primary and lower-secondary school were on a steady increase, as was the pupil-teacher ratio, until it peaked about 1990.
The cost per pupil in primary and secondary school has changed relatively little when comparing the period when the municipalisation was introduced to the current situation. The cost per pupil in primary and lower-secondary school in fixed prices was on the same level in the early 1990s as in 2010. In the 1990s however, there was a decrease, whereas the 2000s are characterised by a return to the levels of the early 1990s.
The pupil-teacher ratio is lower today than in the early 1990s. The pupil-teacher ratio in 1990 was 9.1 teachers per 100 pupils. Between 1992, the year before the shift to general government grants, and 2000, the pupil-teacher ratio fell from 8.6 to 7.6 teachers per 100 pupils. In the 2000s, the pupil-teacher ratio rose again andin 2011 it amounted to 8.3 teachers per 100pupils.
The difference between municipalities in terms of resources for primary and lower-secondary schools has decreased slightly
The difference in municipal principal organisers' costs per pupil in primary and lower-secondary school has decreased slightly between 1989 and 2011. There was a significant difference between municipalities even before the municipalisation and this still remains the case. Primary and lower-secondary education in 2010 cost between 40 and 70 per cent more per pupil (not including premises) in the ten municipalities where it was the most expensive compared to the ten municipalities where it was the least expensive.
In the 1990s the difference in pupil-teacher ratio between municipalities decreased significantly. In the 2000s there was a return to the same level as in the 1990s. Data for the 1980s show that the difference between municipalities in terms of pupil-teacher ratio was higher and equal to the level of 1991.
Cost development per pupil and pupil-teacher ratio over time correlate with local conditions in primary and lower-secondary school
Both the cost per pupil and pupil-teacher ratio show a correlation with the overall national number of pupils from 1980 to 2010. In periods where the number of pupils has been lower, the cost per pupil and the pupil-teacher ratio have been higher and vice versa. Our statistical analyses also show that the changes in cost per pupil and in pupil-teacher ratio on the municipal level correlate with changes in the size of pupil cohorts and the number of pupils per school. An increase in the number of pupils enables the municipalities to distribute the costs across more pupils.
There was a concern that the shift from specially allocated to general government grants in 1993 would result in wealthy municipalities spending more money on education and that the equivalence in primary and lower-secondary education would decrease. In an effort to create equal financial conditions for all municipalities a local government financial equalisation system was introduced to even out differences in income and other structural conditions within the municipalities. When examining the total amount of government grants and other local conditions however, our statistical analyses show that the change in cost per pupil, as well as the pupil-teacher ratio in primary and lower-secondary school, correlate with changes in tax capacity in the period that followed the municipalisation. There is no such correlation in the period with allocated government grants, which indicates that the allocated government grants may have been more effective in reducing the influence of tax capacity than the subsequent general grants. The correlation with tax capacity is significantly weaker in the last period examined, 2002 until 2011, and there is no such correlation with pupil-teacher ratio.
Our analyses indicate that certain local conditions may have had a greater influence in the 2000s, e.g. population density and the competition between different welfare areas that reflect the demographic development have become increasingly important.
Principal organiser cost data is only available for a few years before the municipalisation. Since the years following the municipalisation were characterised by an all-time low number of pupils in primary and lower-secondary education, a heavy economic downturn and a shift from allocated government grants to sector grants in 1991, there is a risk that the period examined does not accurately reflect the cost structure before the municipalisation.
The cost per pupil in upper-secondary school remains stable but the pupil-teacher ratio has increased
Upper-secondary school is different from primary and lower-secondary school in that the programme structure greatly influences the costs, the number of pupils in independent schools is greater and it has more of a regional character. The municipalities' total cost per pupil in upper-secondary school has varied little in the last thirty years. Between 1991 and 2011 it has decreased slightly in fixed prices.
The total cost per pupil for the municipal principal organisers was on approximately the same level in 2010 as in 1990. In 1991, when the cost peaked, the cost per pupil in upper-secondary school reached SEK 99,000 compared to SEK 96,000 in 2011. Upper-secondary school does not demonstrate the same fall in cost per pupil in the early 1990s, as in primary and lower-secondary school. There is however an apparent decrease in the 2000s when pupil cohorts increased significantly.
The pupil-teacher ratio for the municipal principal organisers has increased in most of the period from 1991 and 2011, from 7.4 to 8.5 teachers. The largest increase occurred in 1997 onwards. This increase has also influenced the teaching costs, as their share in the total cost per pupil in upper-secondary school rose from 42 per cent in 1997 to 49 per cent in 2011. The proportion of teachers with teaching qualifications decreased during that period however, from 94 per cent in 1992 to 81 per cent in 2011.
Changes in cost per pupil in upper-secondary school and local conditions.
The cost per pupil in upper-secondary school does not follow the same development on a national level as that of pupil cohorts in primary and lower-secondary school from 1980 up until today. One possible explanation is that upper-secondary education was extended in 1990s to include almost all young people of that school age. The upper-secondary school pupils are also able to travel further and the municipalities therefore have greater flexibility in the design and location of schools.
We have discovered that the cost per pupil in the local municipality correlated with the number of pupils in the municipality from 2002 to 2010, which means that an increase in the number of pupils results in a lower cost per pupil, once we have controlled for other local conditions which may be assumed to influence the cost. On a national level a decrease is also visible with regard to the cost per pupil, while there is a significant increase in the number of upper-secondary school pupils in the same period.
The local municipality's cost per pupil in upper-secondary schools also correlates with the programme structure in the municipality. We also find that an increase in the total government grants, as in the tax capacity, resulted in a higher cost per pupil for the local municipality between 2002 and 2010, once we had controlled for other local conditions. For upper-secondary education, we have not been able to make a comparison with the period before the municipalisation. Neither have we, aside from the number of pupils in the surrounding school market, been able to take into account any influence by other adjacent municipalities. This influence is probably more important for upper-secondary school than for primary and lower-secondary school.
The relation between resources and results in primary and lower-secondary school
From our knowledge overview, we have been able to establish that the equivalence in Swedish schools has declined significantly since the mid-1990s; the difference between the average results of schools has greatly increased. The difference between the results between municipalities has also increased, although to a smaller extent. The results from previous studies support the notion that the extensive changes in schools since the 1990s have had a significant influence on this development. The studies show that the increasing segregation in terms of housing and the school choice system play important roles in this context. Schools have become more segregated not only as a result of socio-economic background, but also of characteristics that are often not visible in the statistics. As an example, well-motivated pupils, regardless of socio-economic background, tend to apply to schools with other well-motivated pupils.
Conclusions from previous research supports the idea that there is a positive correlation between increased resources and better results in primary and lower-primary school. Increased resources seem to be more important for younger pupils and pupils with less favourable study conditions or with weak support from home. Studies show that the decrease in pupil-teacher ratio in the 1990s resulted in worse school performance and that the size of the resources are important primarily for pupils with low-skilled parents. There is also an indication that the quality of the resources matters. According to the research, the singularly most important factor in the pupils' performance is teacher competence.
The conclusion of our own analyses of school resources is that the decentralisation of resources in the schools have not led to any significant changes in terms of resource allocation per pupil. We do not find anything to support the notion that the shift from allocated to general government grants has resulted in greater differences between municipalities, or pupil-teacher ratio in primary and lower-secondary school. On the contrary, we have noticed a smaller difference. The cost per pupil and pupil-teacher ratio decreased in the 1990s and increased again in the 2000s. A contributing factor is likely to be the correlation we have seen between the cost per pupil and pupil-teacher ratio in primary and lower-secondary school and the change in pupil cohorts. Based on national data, this correlation appears to exist even before the municipalisation.
In our statistical analyses however, we only find little support for the idea that the development of resources over time correlates with the pupil's study conditions on a municipal level.
The many comprehensive reforms that have been implemented since the 1990s limit the possibility to discern the consequences of the municipalisation. For this reason, there will be no far-reaching conclusions about how the municipalisation has affected the cost per pupil and the pupil-teacher ratio and by extension, the results in primary and lower-secondary schools. The fact that the difference between schools and municipalities is on the increase, when there has been few changes to the allocation of resources after the municipalisation, suggests that there is a need for important changes in school resource allocation.
The compensatory task of municipalities is likely to have become more difficult. For example, the increased segregation of pupils between schools in terms of socio-economic background and motivation should place greater demands on municipalities' resource allocation. At the same time, we can confirm that the compensatory efforts of municipalities' resource allocation to individual schools are not very significant, even though they have increased slightly in the last few years. The compensatory efforts are also not very significant with regard to resource allocation to primary and lower-secondary school on the municipal level.