Printer Friendly

Education and youth employment.

The Swedish labour market functions well for core workers, but the inclusion of youth could be improved. The unemployment rate for youth is four times higher than for prime-age adults, reflecting both deficiencies of the education system and some specific features of the Swedish labour market. Learning outcomes in compulsory schooling are above-average in reading, but not in mathematics and science. In secondary school, the programme structure is fragmented, and youth completing a vocational programme appear to lack important skills: they are not "job ready". This is problematic in the context of a labour market characterised by high minimum wages, set in collective agreements, and stringent employment protection rules: those with low productivity have little chance to find a job, and employers are cautious about hiring youth whose skills are often hard to assess, in particular youth with an immigrant background. This chapter reviews recent proposals for education reform. It also examines how recent tax and benefit changes might be supplemented to ensure that job-search is genuinely attractive without deterring youth from entering and completing further studies early on.


Education and youth employment must be seen together in a wider economic perspective. This holds especially in an economy like Sweden's where high income taxes and generous benefits shape not just labour market outcomes but also the relative attractiveness of different educational choices. For example, the most effective solutions to skill shortages resulting in unemployment spells for adults might be to improve the early stages of education. Increasingly skill-intensive labour demand requires larger investments in education to create a matching supply. But if this were to be all met via current arrangements, relying on public funding, it would add to the fiscal challenges of ageing and drive up taxes. Those taxes, however, are already an important reason behind labour market exclusion and generate some of the adverse incentives in the education system. Therefore, this chapter analyses education and youth employment with an emphasis on the economy-wide linkages (Box 4.1). After disentangling labour market outcomes for youth, it reviews learning outcomes and policies in compulsory education, i.e. age 7-16. The chapter then turns to upper secondary education and thereafter higher education. Finally, the role of labour market institutions, including minimum wages, employment protection rules, activation programmes and benefits is analysed.

Human capital and youth employment outcomes

By international standards, educational attainment is high in Sweden. Over 90% of the younger cohorts have completed upper secondary education, a share among the highest in the OECD and above the other Nordic countries (Figure 4.1, panel A). The share of adults having completed tertiary education is closer to the international average, but the share of each youth cohort graduating from tertiary education is now rising rapidly following strong expansion of higher education since the 1990s. In 2006, the number of new graduates was equivalent to 40% of a youth cohort; more than the OECD average, but less than in the other Nordic countries (Figure 4.1, panel B). But with around 80% of each cohort entering higher education in the most recent years, the supply of graduates is set to rise further.

Changes in workforce skill composition are shaped not just by the educational attainment of the young flowing into the labour market, but also by those retiring. There is a sizeable gap between the 91% upper secondary attainment rate of the 25-34 year olds who recently joined the labour market and the 73% rate of the 55-64 year olds who will retire from the labour market over the coming ten years (Figure 4.1). This gap means that the average educational attainment of the labour force is growing, enabling Sweden to benefit from technological change that raises skill-intensive labour demand. However, the gap is set to narrow in the future: the upper secondary attainment rate is stable for the inflow, being virtually the same for the 35-44 year olds as for the 25-34 year olds, but the cohorts moving towards retirement in the coming decades are considerably more qualified than those retiring today. At that horizon, it could become challenging to continue raising the average educational attainment of the labour force at the current pace.
Box 4.1. Education and youth employment policies in a Swedish
economic context

Official Swedish education policy documents mostly state the
learning objectives in the various areas of education. Meanwhile,
to analyse the economy-wide linkages that are relevant for
education and youth employment, it is useful to lay out the core
objectives and interactions that seem central in an economic
perspective. The central policy objectives might be summarised as:

* Equal opportunities. Realising the potential of all to be
employable in a modern workplace gives people the possibility to
improve upon their situation via working. Being employable is also
a key step towards social integration and participation, including
for immigrants. And it gives individual independence, not least for

* Economic growth. Human capital and skills in a wide sense
are essential for economic growth, since they are necessary in
order to adopt technological advances and complementary
organisational changes. Hence, the educational achievement of youth
cohorts entering the labour market is one of the most important
areas of government policy influence over business competitiveness.
Indeed, companies typically cite human capital as a major
determinant of international location choices.

Meanwhile, some features of the Swedish economy, combined with a strong
preference for equal opportunities and equity, mean that certain
interactions warrant particular attention, possibly more so than in
some other OECD countries:

* The composition of labour supply. Each person's
choice of skill formation affects the relative earnings of others
via the labour market equilibrium. This matters because of the
equity preference and the labour market institutions which limit
the availability of low-skilled jobs. If those who have the
cognitive capabilities to study abstain or wait too long while
taking low-skilled jobs, it will be harder to find employment for
those who don't have the cognitive capability to study. This holds
provided there is some complementarity between high-and low-skilled
labour demand or that consumption includes demand for non-tradable
low-skilled services. Similarly, boosting average hours worked for
the high-skilled would raise demand for low-skilled labour.

* Fiscal externalities. Choosing education with good labour market
prospects has a large positive impact on public finances by
boosting the tax base and reducing outlays on income-replacing
benefits. By contrast, if more prospective students shift towards
education choices driven by consumption aspects (leisure and other
non-pecuniary benefits), the capacity to finance a welfare state
would be weakened.

Youth unemployment is a cause for concern. In 2006, more than one in five 15-24 year olds participating in the labour market were unemployed, a proportion four times higher than for the 25-54 year olds. Among OECD countries, apart from Finland, youth unemployment rates above 15% are only found in Central, Eastern and Southern European countries. By contrast, in Denmark, Iceland, the Netherlands and Norway, where social preferences are much the same as in Sweden, the unemployment rate of the 15-24 year olds is just a third of the Swedish rate. Moreover, among the OECD countries with high youth unemployment rates, the ratio between the unemployment rates of young and prime-age adults is nowhere more skewed to the disadvantage of youth than in Sweden (Figure 4.2). On the back of buoyant economic growth, youth unemployment has declined in recent years from 22.8% in 2005 (3.7 times the rate for 25-54 year olds) to 19.2% in 2007 (4.3 times the rate for 25-54 year olds), but that is still high compared to other countries. Even excluding full-time students, the unemployment rate for 15-24 year olds stood at 14.3% in 2007. (1)


Short duration mitigates the economic and social costs associated with youth unemployment. In fact, two thirds of Swedish youth unemployment is due to spells of three months or less (Figure 4.2). On that score, youth unemployment may be considered less of an issue than in Germany and the United Kingdom, even if the youth unemployment rate is higher in Sweden. By contrast, the incidence of youth unemployment is higher across all durations when compared to Denmark and Norway. (2)

At the core of the youth unemployment problem are the long-lasting unemployment spells affecting certain groups. Immigrants and youth who fared poorly in school are particularly at risk (Box 4.2). Morover, unemployment spells early in life may affect motivation and self confidence or be seen as a bad signal by future potential employers: even spells shorter than three months following graduation from Swedish upper secondary education give rise to long-run negative effects persisting for at least five years (Skans, 2004).

While a fairly high proportion of youth are unemployed, few are inactive. Indeed, the share of youth that are neither in employment, education or training (the so-called NEET rate) is a bit below the OECD average (OECD, 2008a). (3) This suggests that lack of education or training places is not the main source of the Swedish unemployment problem. The root cause might rather be inadequate quality in education and training or labour market distortions--the issues reviewed in the rest of this chapter.


Compulsory education: ensuring a strong start

Giving children a strong foundation in compulsory school is obviously important, because without basic literacy it is hard to manage in virtually any modern workplace or to pursue post-compulsory education. Before turning to secondary education and the school-to-work transition, this section therefore reviews learning outcomes and policies to improve the early stages of education.

Learning outcomes

Swedish learning outcomes are not as good as they used to be. At age 15, student proficiency measured by PISA is above the OECD average for reading but not for mathematics and science (Figure 4.3). Moreover, learning outcomes in 2006 had deteriorated relative to the in-depth assessments of reading in 2000 and mathematics in 2003. (4) In the light of Sweden's ambitions, and as public spending on compulsory education is relatively high, these outcomes are not satisfactory--certainly not when compared to Finland.
Box 4.2. Immigrants and school drop-outs are most at risk of
long-term unemployment

An econometric "hazard" model can estimate the effect of various
characteristics on the likelihood that an unemployed person will
find a job ("job-finding rate"). Thereby, groups who are at risk of
drifting into long-term unemployment can be identified (Table 4.1):

* Immigrants and youth born in Sweden with immigrant parents have
lower job-finding rates. The more recent the immigration
experience, the lower the job-finding rate. This may partly be due
to lack of skills, not least as regards proficiency in Swedish.
However, there is also evidence that employers tend to discriminate
against applicants with foreign-sounding names (Carlsson and Rooth,

* Very few unemployed youth have children, but those who do have
much lower job-finding rates.

* Compulsory school performance is a strong predictor of
job-finding rates. This holds throughout the skill distribution. A
complete education raises the chances of finding a job,
irrespective of the level of education. Graduates from vocational
programmes have slightly higher job-finding rates than graduates
from academic upper secondary school programmes. Those who do not
study beyond compulsory school have the lowest job-finding rates.

* The difference between a complete and an incomplete upper
secondary education is large. It is of about the same magnitude as
going from the high end to average in terms of compulsory school
performance, or as big as the difference between natives without
immigrant background and recent immigrants. Those who struggle at
school are likely to have both poor grades in compulsory school and
to end up with an incomplete education. Thus, job-finding rates are
likely to be very low for this group.

In the domestic debate, it is sometimes argued that what really matters is schools' capacity to improve the weakest students' abilities. Indeed, given the features of the Swedish labour market, including high minimum wages, this is vital as the low-skilled are at a high risk of unemployment. But even on this score, Sweden does not rank much better, based on the share of 15-year olds left behind with poor learning outcomes (Figure 4.3): 15% of a cohort only attain proficiency level 1 in reading, i.e. they are able to locate explicitly-stated information but struggle to identify information that is not prominent, to deal with competing information and to understand relationships. The share of students at this lowest level of proficiency is three to four times higher than in Finland across all three subject areas--reading, mathematics and science. (5)


Science is a weak spot both in terms of proficiency and student interest. Given Sweden's traditionally strong position in engineering and science-related industries, such as electronics and pharmaceuticals, this is of particular concern. For average student proficiency, Sweden ranks in the middle of OECD countries, but what may be most worrying from an employer perspective is that the elite is quite narrow: the share of 15-year olds attaining the highest proficiency levels 5 and 6 reaches 13% in mathematics and merely 8% in science, compared with 24% and 21% respectively in Finland, which has a similar industrial structure. To change this pattern, schools may need to do more, building on childrens' natural curiosity, to nurture the interest to learn more and acquire specific skills and proficiency in science. Sweden is below the OECD average on virtually all measures used in the PISA assessment for student's aspirations and attitudes towards science. Swedish 15-year olds indicate some general interest in learning about science, but less interest when it comes to specific topics, such as acid rain. Only little time is spent on science-related activities outside school. Most remarkably, Swedish youth are among those indicating the least awareness of the science-related aspects of environmental problems and they are rather unconvinced that science promotes economic and social progress (Figure 4.4; OECD, 2007a). (6) This pattern is problematic, as there is clear evidence that early interest in science is a strong predictor of lifelong science learning and/or a career in a science or technology field (OECD, 2006a). Indeed, relatively few Swedish 15-year olds indicate an interest in working with science later in life. The current moderate difficulties for employers to recruit engineering and science graduates might worsen in the future if children's natural capacity and potential to learn science fails to flourish in school. (7)


In the traditional Swedish welfare state, compulsory school plays an important role as an institution building social cohesion. The capacity to engage with people of all backgrounds is also beneficial from an economic perspective, facilitating market flexibility. While variation in learning outcomes is not much smaller than in other countries, it is found within rather than between schools, indicating that there is little segregation between good and less good schools and neighbourhoods (Figure 4.5).


The integration of the rapidly-growing immigrant population is particularly important in this context, but the outcomes are difficult to interpret. There is a wide gap between the learning outcomes for immigrants and natives, but that may be hard to avoid given the language barriers. Few immigrant children arrive with prior knowledge of Swedish, which is used as the instruction language in all subjects. The second generation, i.e. children born in Sweden from immigrant parents, appear to do much better than the first, the difference being considerably greater than in other OECD countries (Figure 4.6). However, it is hard to know if this reflects a genuine improvement from one generation to the next, or differences in composition, since today's first-generation immigrant families typically have a less favourable background than the immigrants who arrived in Sweden some decades ago.


Policy initiatives and school efficiency

A number of policy measures have recently been introduced to improve learning processes and monitor outcomes. A new reading-writing-arithmetic initiative was announced in 2008. National tests were introduced some years ago and, from 2009, they will be advanced to start from the third year of compulsory education. National tests are applied again at the fifth year where they now become mandatory also in English. Finally, science tests will be introduced at the ninth year of compulsory education. Moreover, a legislative proposal has been presented with a new scale for marking students to be introduced from the academic year 2011/12.

Swedish schools appear to be at some distance from international best practice in terms of how efficiently they use available resources (Figure 4.7). The use of teachers' time can often be an Achilles heel of efficiency at schools, but a lack of data makes it hard to compare Sweden internationally. Teachers' total statutory working time is above the average for the OECD countries for which there are data. However, total instruction time for students seems to be relatively short, and the student-to-teacher ratio is relatively low (OECD, 2008b). Together, that would indicate that teachers in compulsory school spend less of their time teaching than in other OECD countries. Consequently, there might be scope to employ teachers' capacity more effectively, and reward teachers better when they put in extra teaching effort.


The renewed Swedish policy focus on teacher competence is therefore most welcome. Following the Scottish example, a recent government report has proposed that teachers should go through an accreditation process after graduation (SOU, 2008a). During a probationary year, the new teacher would be introduced to the profession by a mentor. Only if assessed as suitable for the profession, would she or he then be granted accreditation and be able to apply for permanent employment. Additional information about each teacher's profile is to be used by school management to ensure that teachers are competent in the subject areas and functions they handle. To encourage continuing professional development, new advanced professional levels are proposed, accessible via further study or well-documented practical development activities. By contrast, serious neglect would lead to the accreditation being withdrawn. Persons currently working as teachers without a degree would be required to complete a degree. Otherwise they would have to leave the profession when an eight-year transition period expires.

Once the accreditation scheme is well established, greater wage flexibility could reward the best teachers. Some flexibility already exists since 1996 when a fixed pay ladder was replaced by individualised pay and strict working time regulation relaxed. Thereby, salaries are determined in discussions between each teacher and the school principal or in negotiations involving the local teacher trade union. Pay rewards are supposed to be linked to municipal school objectives, even if in practice they are more based on competence, effort etc. The dispersion of teacher earnings is not wide, though, and it actually narrowed following the 1996 reform, perhaps reflecting that individual-pay dispersion has been offset by less dispersion based on age, as the starting salary was raised significantly (Strath, 2004). Still in 2007, the upper quartile of earnings was just 20% above the lower quartile for teachers in compulsory education. When the accreditation scheme is in place, it might offer a basis for stronger rewards to the best teachers and the ones taking on additional duties, classes or longer working hours. Compared with other countries, teacher salaries are not particularly high (OECD, 2008b). (8)

In 1992, Sweden introduced a system of choice that is quite remarkable from an international perspective: independent schools are guaranteed public funding on an equal basis with municipal schools in the same area. By contrast, independent schools are not allowed to charge tuition fees, and they cannot reject students from within the municipality. Funding for public schools responds automatically to changes in attendance. Choice among public schools across municipalities is also allowed. Since choice was introduced, the share enrolled in independent schools has risen rapidly, reaching 9% in 2007/08 (National Agency for Education, 2008). In other countries, the shares of students attending public versus private schools is typically very stable, reflecting tradition. Another remarkable factor is that independent schools can be organised as private limited companies, in addition to the forms of teacher or parent co-operatives, associations and non-profit foundations seen in other countries (Lundsgaard, 2003).

Competition from independent schools is found to improve learning outcomes in public schools. Systematic studies find that test scores and final grades in key subjects are better, ceteris paribus, in areas where many attend independent schools (Sandstrom and Bergstrom, 2005; Ahlin, 2003; Bohlmark and Lindahl, 2007), even if the exact size of the effects is subject to debate (Bjorklund et al., 2006). (9) The model appears to be a clear success.

To sum up, learning outcomes at age 15 are not as good as they could be, and they should be improved. However, shortcomings in compulsory education alone cannot explain high youth unemployment: Denmark and Norway, for example, display similarly mediocre PISA scores but have much better youth employment outcomes. Therefore, upper secondary and tertiary education as well as labour markets also need attention.

Upper secondary education: preparing better for labour-market entry

Sweden has few problems getting youth to join upper secondary education, but the latter does not fully succeed in making students "job ready". The high incidence of short unemployment spells after upper secondary school might indicate a mismatch between the needs of the workplace and what has been learned at school. A striking example is the media programme, where only a small fraction ends up working in media jobs; the period of unemployment after completing the media programme is a process through which many realise that they have to seek work elsewhere, such as in long-term care for the elderly. These persons might have been better off if, from the outset, they had chosen the programme giving the skills needed in elderly care.

Against this backdrop, an initiative was recently launched to overhaul upper secondary education. A government commission identified two broad problem areas (SOU, 2008b):

* Programme structure. The vast range of programmes on offer makes it difficult for students to gain an overview of the alternatives. Consequently, many students (12%) switch programmes during their first year at upper secondary school. Those who receive school leavers (universities, companies and organisations, etc.) find it difficult to assess their capabilities, which may make it more difficult to find a first job.

* Results, goal fulfilment and efficiency. Both clear result indicators and an improved evaluation system are needed. For example, there are no data on the extent to which the guaranteed teaching time is satisfied. The Commission has developed a quality measure, namely the share of students achieving basic eligibility for entering higher education in three years; only 62% of all students currently achieve this goal. Streamlining the system towards fewer different alternatives might also lead to lower costs and higher efficiency as, currently, small class sizes increase costs. This is particularly important in a context where student cohorts are set to fall within a few years.

There is also a growing consensus that Swedish upper secondary education could be improved if vocational programmes had closer workplace links. Indeed, this is the main focus of the recommendations in the recent OECD Policy Review of Vocational Education and Training in Sweden (OECD, 2008d). (10) International research shows that close co-operation and participation of companies in upper secondary education provide good opportunities for youth to enter the labour market (SOU, 2008b). In principle, 15 weeks of workplace-based training (Arbetsplatsforlagd utbildning, APU) are included in all upper secondary vocational programmes. In practice, however, this is not the case everywhere as many schools seek dispensation to allow workplace-based training to take place at school. Against this background, a government commission has recently proposed a revamped three-track upper secondary system, including a new apprenticeship route alongside modernised versions of the traditional academic and vocational routes (Box 4.3). Indeed, workplace learning should be seen as a necessary component of a vocational programme. This is especially important for immigrant and second-generation youth for whom workplace attachment may give not just professional skills but also a wider understanding of relations in a Swedish workplace (Lemaitre, 2007).

For the new apprenticeships to function well, workplace activities should involve learning, not just work, and the competencies acquired should be of general value rather than specific to narrowly-defined professions. The experience from countries with long apprenticeships traditions (e.g. Austria and Switzerland) is that while youth with apprenticeships find employment more easily, they may later on experience more difficulty if having to change profession than similar persons with school-based vocational training (Ludwig and Pfeiffer, 2005; OECD, 2004a and 2006b). The heart of the problem is that if apprenticeships are too closely linked to existing professions, they may end up providing youth with excessively narrow skills that over time become redundant. Ways to avoid these pitfalls have been identified by the Commission:

* Apprenticeships would be anchored into the structure of the 14 national vocational programmes, sharing the common upper secondary subjects.

* The content of workplace learning would be clearly defined in blocks tied to the programme-specific subjects for the vocational programme in question. Apprentices would thereby acquire sufficiently general skills to ensure a capacity to change towards working in other areas, if needed later in life.

* Quality is to be guided and monitored via local agreements drawn up between the school governing body and the firms taking apprentices.

* Firms would be obliged to allocate a skilled employee to supervise the apprentice. The supervisor should first have undergone a course arranged by the school governing body.

If these Commission proposals are implemented, the benefits of extensive workplace learning during an apprenticeship are likely to outweigh the potential drawbacks.
Box 4.3. Proposals from the Upper Secondary Reform Commission

The curriculum with its approach to knowledge would be preserved in
its present form: all study paths in upper secondary schools would
continue to prepare students for active participation in society.
The reforms would increase the distinction between the academic and
vocational tracks in terms of selection, curricular content and
routes into tertiary education. Following the report issued in
March (SOU, 2008b), preparations are now underway for legislation.

All upper secondary programmes would continue to last three years,
with a common basic structure. However, the contents of each of the
proposed 14 vocational programmes and five higher-education
preparatory programmes would be more targeted than currently,
notably by adapting the focus of common upper secondary subjects
(Swedish, math, English, etc.) to what is relevant for each
particular programme. This should allow vocational programmes to
stretch further than today's upper secondary school, leading to a
vocational diploma that constitutes a recognised skill
qualification and enhances employability.

As a major novelty, the Commission proposes to introduce
apprenticeships as an alternative to school-based vocational
programmes, leading to the same qualification--a vocational
diploma. Upper secondary education would then have three
alternative routes:

* School-based vocational programmes. A third of the three years
would be spent on common upper secondary subjects, individual
options and a diploma project, with the remaining two thirds spent
on specialisation. There would be 14 national programmes (building
and construction; farming and natural resources; electricity and
energy; aircraft; vehicle and transport; trade and administrative
service; hairdressing and handicraft; hotel and tourism;
manufacturing; children and preventive healthcare; restaurant and
food; maritime; heating; water and sanitation; health and long-term
care). In each of these programmes there would be two to five
nationally-defined branches, and schools would have the option to
include further specialised in-depth components for up to a third
of the three-year programme duration. At least 15 weeks of
workplace learning would be compulsory.

* Apprenticeships. These so-called upper secondary apprenticeships
would be introduced in each of the 14 national programmes. Students
would have the same common upper secondary subjects, individual
options and diploma project as for school-based vocational
programmes, but most or all of the two years' programme-specific
learning would happen at a workplace.

* Higher-education preparatory programmes. There would be five
national programmes (economics and business, arts and humanities,
social science and media, natural science, technology), each with
three to five nationally-defined branches.

The emphasis on standardised core content would be backed by the
introduction of a diploma based on specific objectives for each
programme. There would be two types of diploma: a vocational and a
higher-education preparatory diploma. A clearly-defined diploma
would also help document competencies when going abroad, and assist
each industry in adapting its further education and training for
possible certification. Students on vocational programmes should
have clear options to achieve basic eligibility for higher
education within the upper secondary school system, possibly via
extended studies.

Finally, a new National Council for Education would be created with
representation of those receiving youth once they have completed
secondary school, i.e. tertiary education institutions and
employers, to identify development potentials and satisfy the needs
of future industries which currently lack established structures of
co-operation. For each of the 14 national programmes, there would
also be a Programme Council responsible for content. These councils
would involve employers, trade unions, relevant national
standard-setting agencies, and they should also make use of
students' experiences and views of the programmes. Moreover, local
programme councils should be set up for co-operation between local
employers and school-governing bodies.

The extent of pay received under an apprenticeship is to be decided locally. The school and the participating employers have to agree. Monitoring how this evolves will be important for successful implementation: having firms interested is crucial to be able to offer enough apprenticeship places, and low labour costs are typically essential to secure this interest (Quintini et al., 2007).

With the 2009 Budget, considerable allocations were set aside for an expansion of vocational education. The number of apprenticeship places will be raised from 4 000 in the pilot phase of the academic year 2008/09 to 6 000 in 2009/10. The number of vocational training places in adult education will be increased and vocational higher education will be introduced.

School choice has expanded in upper secondary education

Private independent schools have become an important element in upper secondary education. The same regulations apply for compulsory as for upper secondary schools, but the growth in attendance has been most spectacular for upper secondary: 17% were enrolled in independent upper secondary schools in 2007/08, up from just 3% ten years earlier; for compulsory education the share had reached 9% (National Agency for Education, 2008). At times it is argued that independent schools have grown because they offer music or sport programmes, attracting students with an emphasis on leisure more than learning. Whether this is true in general is hard to ascertain, but there have been specific cases where independent schools did not deliver what they were promising. For potential students to make an informed choice, more transparency is needed about the labour market prospects associated with alternative education routes, such as by publishing data for employment outcomes of recent school leavers (OECD, 2008d). Aside from this, the Swedish model for school choice is a clear success.

The entry of independent schools can reveal valuable information about shortcomings in the established system. For example, it could be questioned whether municipalities are large enough to offer public secondary schools the conditions to thrive. Although about a quarter of all students in public upper secondary education cross municipal borders when going to school, municipalities are often criticised for trying to avoid that youth leave the local community. For the theoretically-oriented programmes, this may not be an issue, but for vocational programmes a larger size may be needed to provide students with modern equipment and build relations with the bigger companies. In Norway, upper secondary schools belong under the counties and in Denmark under the central government, not under the municipalities. In this context, per-student funding, introduced in Denmark in 1991, initiated a process where each public vocational upper secondary school increased its attention to strategic management, often closing courses with too few students to cover costs. (11) It could be interesting to analyse the consequences, in Sweden, of moving the responsibility for upper secondary education to central government, or perhaps the counties, while giving each institution more independence based on per-student funding. It would give public schools more room to merge, re-organise and specialise where that is desirable. Such an analysis could also look at other levels of schooling.

Higher education: shortening delays in study completion

Swedish higher education has a number of strengths. University (12) management has developed significantly over recent decades, with ample representation of external members on university boards, and the implementation of such quality assurance methods as regular evaluations. Higher education is easily accessible following strong capacity expansion--a near doubling of the student intake--since the early 1990s, including life-long learning offers.


Yet, there are challenges as well, as identified in the context of the recent OECD Thematic Review of Tertiary Education (National Agency for Higher Education, 2006; OECD, 2008e). (13) Specifically, the allocation of core research funding could be based on more transparent criteria, including performance indicators. Knowledge transfer via contacts with business could be deepened to nurture innovation. Retirement of the ageing academic staff will be a challenging transition for some institutions. And the institutions providing higher education could be given more freedom to develop their own strengths. With the 2009 Budget, the allocations for publicly funded research are being raised considerably for the 2009-12 period to reach 1% of GDP. One third of the increase will come via a new form of strategic investments in the fields of medicine, technology, climate/environment and finally interdisciplinary, humanity and social sciences. The other two thirds will be channelled via the general budgets of higher education institutions, research councils, etc.

A particular challenge relates to the heavy reliance on public funding. Tuition is free and the government offers students grants and loans to cover their living costs. Consequently, prospective students are not troubled by liquidity constraints: the cost of studying, measured relative to available individual funding, is among the lowest in the OECD (Figure 4.8). Meanwhile, wage premia for persons with higher education are low. Private internal rates of return from studying are also relatively low: while boosted by public financial support, they are held down by high and progressive income taxation (Boarini and Strauss, 2007). Two problems require attention: there are incentives for students to start and complete higher education late, and the prohibition of tuition fees constrains the development of the higher education sector as a whole. These two issues are discussed below.

Delays in entering and completing higher education

The median age for entering tertiary education is high in international comparison--a full year higher than in Finland and Norway, and almost three years higher than in the Netherlands, which shares many societal preferences with Nordic countries (Figure 4.9). Following recommendations in previous Surveys, clear improvements are now being made in the admission criteria: notably, the possibilities to improve grades by repeating secondary school courses in adult education have been reduced (Annex 1.A1). Yet it might be worth reviewing if there are still unintended factors that cause students to defer and delay tertiary education.


The high median starting age would not be a problem if it simply reflected that many participate in life-long learning, thereby boosting higher education entry late in life. However, it is clear that the high median starting age also reflects the large share of youth entering higher education after a fairly long gap following the completion of upper secondary education (Figure 4.10). At age 19 and 20, participation in education drops to 40% and 35%, well below the level in comparable OECD countries. At age 21, the share in education rises--a pattern not seen in any other country, except for Denmark and, to a smaller extent, Norway (Figure 4.11). While many youth appreciate taking a break between secondary and tertiary education--travelling abroad, working in cafes, etc.--this leads to lower income for many years: ceteris paribus, two years postponement reduces the present value of lifetime earnings by nearly one half of one year's peak earnings (Holmlund et al., 2008). With high and progressive income taxes, a considerable part of this income loss is carried by the public purse in the form of foregone tax revenue. This creates an adverse incentive for youth to take more gap years than they might have desired, had they themselves retained the full income gain from fewer gap years.



A considerable part of youth unemployment consists of persons who have completed an upper secondary course directed at entering higher education, but then wait a few years. Among 19 and 20-year olds, 22-24% of unemployment is experienced by youth who will proceed to tertiary education within three years (Figure 4.10). To avoid that these people take up low-skilled jobs and displace those who lack the capacity needed to study, it would be desirable to encourage less delay before proceeding into higher education.

Sweden's taxes and benefits encourage youth to start and complete studies late--and increasingly so with recent reforms. In a system where marginal tax rates rise with income, there is a general incentive to smooth income over life. Moderate differences in marginal rates are not problematic, but recent reforms have magnified the gap between the marginal tax wedge faced as a graduate, and that faced before and while studying. In particular, the introduction of lower social contributions for youth up to and including 25 raises the opportunity costs of focusing on studies, not work, before age 26. (14) The effect is big enough to warrant attention, because it comes on top of the special excemption of study grants from personal income taxation. While most income benefits (unemployment, sickness and disability benefits, etc.) are taxed as personal income, this is not the case for the public grants available for students to cover part of their living costs. Consequently, it is not uncommon for students to face a zero marginal income tax rate, and with the in-work tax credit, introduced in 2007, this zero marginal rate now extends further up the income scale. With the pending 2009 enlargement of the employer contribution rebate for youth, the initial tax wedge resulting from social contributions and personal income taxation combined is only 27% for students compared to 63% after graduation when many face the state income tax. From a societal perspective it would be beneficial that students sometimes take loans and focus on completing their studies, to make their graduate skills available on the labour market, but from an individual perspective things now look different: unless the employer is willing to pay at least twice as much per hour after graduation, the large gap in marginal tax wedge implies that the initial net hourly earnings are higher from student jobs. (15) Effectively, late completion of studies is encouraged. This is an important reason why the general youth employer contribution rebate should be reconsidered, as discussed below in the context of labour market policies.

To offset these adverse incentives, changes in the financial support students receive to cover living costs should be considered. Already today, the public study grant available for students to cover part of their living costs can only be received at a full-time rate if studying full time. However, some countries go further and convert debt to grants if the studies are completed within a stipulated time, or give a bonus for starting soon after secondary education. Another possibility would be to change the composition of financial support for students further towards loans, possibly with a sliding scale giving a considerable share as grants early on but moving towards a larger share of loans when studies get prolonged. (16) The justification of the special youth housing benefit might also be reconsidered: it offers extra public money for those students who choose a "city-centre lifestyle" with expensive housing. (17)

Introducing tuition fees

Free tuition and generous grants for students' living costs have not achieved their original aim. Traditionally, generous public funding has been motivated by equality of opportunity: a desire to make higher education accessible for students from all family backgrounds alike. More recently, educational subsidies via free tuition have been described as the "siamese twin" offsetting the disincentives to human capital formation arising from redistribution via high marginal taxes (Bovenberg and Jacobs, 2008). (18) However, the model has not succeeded in significantly reducing differences in the likelihood by which people from different social backgrounds pursue higher education (SOU, 2008c). Indeed, experience from other countries indicates that, so long as ample loans are available to overcome liquidity constraints, moving towards letting students pay part of the tuition costs does not affect the social composition of the student body much (OECD, 2008e).

Moreover, free tuition may be at odds with growing international mobility of graduates. Compared to other countries, Sweden is not receiving that many foreign students, but the number is growing rapidly (Figure 4.12). Persons moving abroad after having studied for free don't contribute to the tax revenues on which the financing of their education relies in Nordic Countries (Welfare Commission, 2006). Previously the magnitude of this problem has, in practice, been small, but that is starting to change: more and more apply from all over the world to benefit from the free tuition, and as universities are obliged to select based on academic criteria, foreign enrolment is rising. (19) Against this background, the Swedish government is planning to introduce tuition charging for students from outside the European Union. (20)

Gradually introducing tuition charging also for Swedish and EU students would give stronger market signals for universities to develop courses in high demand and for students to reflect more on how and what to study. Public funding is allocated based on student numbers, implying some demand signals, and universities have considerable autonomy in terms of budget allocation, staff matters and other areas of input flexibility (Figure 4.13). Output flexibility is more limited as, in a few fields, a numerus clausus is imposed. In general, universities are only allowed to develop courses to be paid for privately, if there is one organisation contracting and paying for the course or programme; universities are not allowed to offer a programme and then charge tuition from students who apply. It would seem natural to give universities more independence and allow them to develop training paid for by tuition fees, with due consideration for the accountability mechanisms that would be a natural counterpart to increased flexibility. As knowledge itself increasingly becomes a tradable service, Swedish universities are unduly disadvantaged as they are largely prevented from developing education offers based on charging (Globalisation Council, 2008).


Financing tuition charges via loans with income-contingent repayment can ease access for students from less-advantaged backgrounds as they may be relatively risk averse. Such arrangements exist for tuition costs in countries like Australia, New Zealand and the United Kingdom (OECD, 2005, 2008e and 2008f). However, this would only work in a Swedish context if the high marginal tax wedges are reduced. The reason is that if income-contingency can lead not just to deferral of repayment, but to remission of debt, then it ultimately resembles a marginal tax. In countries with low or intermediate marginal tax wedges, this is not a problem, but in Sweden, where it comes on top of a very high marginal tax wedge, the resulting incentive problems would be large. In fact, Sweden previously had loans for living costs with income-contingent repayment, but these were changed to standard loans for this reason. Introducing loans with income-contingent repayment for tuition charges would therefore only be advisable if marginal income taxes are lowered substantially. This in turn is desirable also for a number of other reasons, as discussed in Chapter 3.

Labour markets: easing the impediments to youth employment

Many of the Swedish labour market's salient features constitute impediments to the entry of youth. In countries with very flexible wage formation, employers can offset the costs of the necessary training in entry jobs for young people by offering lower wages. Where firing is easy, the risk of hiring young people with unproven skills is relatively small. But in Sweden, the costs and risks may make it unattractive to recruit young people since their wages are pegged to those of older workers and they are hard to lay off from regular contracts if the particular person turns out not to be the right one for the job. Even with the best imaginable education system, Swedish youth unemployment can hardly be brought down to the level of the leading countries without tackling the core rigidities. Minimum wages, employment protection rules, benefits and activation policies are therefore reviewed in turn below.


High minimum wages--can their adverse effects be offset by cuts in employer contributions?

Sweden has no legislated minimum wages, but collective agreements between unions and employers stipulate minimum wages for nearly all workers. These are high by international standards (Skedinger, 2007), following strong increases over the past ten years. Workplaces that are not covered by collective agreement can, in principle, set wages freely, but in practice this seems not to matter much. Unlike in most OECD countries, unions are allowed to take collective action against firms without collective agreements even if the workplace has no union members. Moreover, generous income transfers put a floor under workers' reservation wages.

High minimum wages are likely to inhibit access to employment for youth who are still learning the trade. The effects are mitigated where minimum wages are differentiated depending on experience, e.g. increasing after one year of work in the industry, profession or firm. Most agreements also have special arrangements for those younger than 18, either with a lower minimum wage or no contractual minimum wage at all. However, the age limit for eligibility for the adult minimum wage is rarely higher than 20 (Skedinger, 2007). (21) This shapes the earnings dispersion: the ratio between the median and the lowest paid for each age group is higher at age 19 than at age 20 (when the adult minimum wage usually kicks in), whereafter it grows with age (Figure 4.14). Apparently, minimum wages are a real constraint for how flexibly wages can adjust to the productivity of young workers. Quantifying the extent to which this hampers youth employment is difficult given the institutional complexity of Swedish minimum wages. For individual sectors, however, research suggests that minimum wages have had detrimental effects on employment (e.g. in hotels and restaurants, see Skedinger, 2006). Other countries, such as the Netherlands, are now making more room for lower starting wages for new graduates/ young workers in collective agreements (OECD, 2008g).


As it has no control over the minimum wage, the government is seeking to boost labour demand via reductions in employer contributions. The first step, introduced in July 2007, reduced gross wage costs for 18-24 year old employees by 8.3% through halving social contributions, except for pensions. A pending proposal would remove an additional quarter of the normal non-pension-related contributions, taking the cumulative reduction in gross wage costs to 12.1% in 2009 (Ministry of Finance, 2008).

Lower employer contributions for youth can mitigate the negative employment effects from high minimum wages, but it may also reduce wage restraint on the union side, leading to further increases in minimum wages. The three-year collective agreements concluded in 2007 imply strong annual pay rises of 6-7% for hotels and restaurants (though half for personnel younger than 20) and 5-6% for inexperienced retail workers (National Mediation Office, 2007). (22) This will attenuate the benefits of the contribution cuts and could spill over to other unemployed marginal groups, such as immigrants, who would become more expensive to employ (they would not benefit from a reduction in contributions).

Another difficulty with across-the-board cuts in contributions is that they are poorly targeted. The payroll tax cut covers all youth, and most of those covered would have found work anyway. The targeted New-Start Job scheme, introduced in 2007, is more promising. (23) For youth, it implies a subsidy lasting for six to 12 months, depending on the length of the jobless spell. (24) Formally, the worker has to apply in order for the employer to get the reduction. However, in contrast to preceding policies, this reduction does not require consent from an employment service case manager. There are no Swedish evaluations of employment subsidies for young workers, but evidence for older unemployed workers suggests that the effects are positive, although there can be substantial displacement effects with subsidies paid to jobs which would have materialised anyway (Calmfors et al., 2004; Forslund et al., 2004). Adding this to the adverse incentive to start and complete studies later discussed above, it would seem worth reconsidering the general contribution rebate for youth in favour of more focused measures.

Relaxing employment protection legislation

Strict employment protection rules are known to be an important obstacle for youth entering the labour market. Theoretical considerations as well as empirical studies suggest that strict employment protection legislation may be neutral with regard to aggregate employment and unemployment insofar as the effects of fewer layoffs and less frequent hiring balance each other out. However, less job turnover complicates and prolongs the search process for new entrants to the labour market, such as youth. Moreover, if regulations make it hard to lay off someone who turns out not to be the right person for the job, employers will tend to restrict hires to those whose capabilities are easier to assess, for example because they have an employment history. Consequently, youth unemployment is higher where employment protection regulations are strict (OECD, 2004b and 2006c; Skedinger, 2008).

The recent extension of the maximum standard duration for temporary contracts allows employers to better "try out" a young person, but also runs the risk of increasing labour market dualism. The so-called general temporary employment contract has now been introduced in legislation, raising the maximum duration of standard temporary contracts from 12 to 24 months. (25) It is now in the process of being implemented in collective agreements. These steps continue the reform approach followed over recent decades. While Swedish regulations for regular employment have remained unchanged, the restrictions on fixed-term contracts and temporary work agencies were eased between the late 1980s and late 1990s.

The downside of this approach is that it may further entrench insider-outsider problems. The incidence of temporary contracts has increased strongly since the early 1990s, in particular for young workers. As a result over half of all youth employment in Sweden is based on temporary contracts, compared with around a quarter in Denmark and Norway, and even less in the United Kingdom (Figure 4.15). It has been suggested that high unemployment rates during the deep economic crisis of the 1990s may have contributed to this development (Holmlund and Storrie, 2002). However, temporary contracts account for a larger share of youth employment now than in the 1990s, indicating that structural changes also play an important role. In particular, today there are more students looking for part-time work and willing to take a temporary contract: enrolment in tertiary education has grown substantially, and reductions in grants to cover living costs may encourage more students to seek work. (26) But in the end, such factors cannot explain why a larger share of employment is on temporary contracts than in otherwise comparable countries: growing duality of employment protection rules seems to be a real problem.


Employers may resort to temporary forms of employment to circumvent strict rules, resulting in people taking a series of temporary contracts, rather than moving to a permanent contract. Indeed, looking at labour market transitions, the most common destination for those leaving unemployment is temporary employment, with only few going directly into regular employment (Figure 4.16). Unemployment is more persistent for prime-age adults than for youth, but when leaving unemployment, youth are less likely to get a permanent contract than prime-age adults. Moreover, youth from less advantaged backgrounds may have more difficulty finding their way in a dual labour market, and they may be more vulnerable to screening by employers who become risk-averse when regulations make it difficult to lay off someone turning out not to be the right person for the job. (27) In the event of an economic downturn, youth and others on temporary contracts may have to bear the brunt of adjustment as they are the least difficult to lay off (OECD, 2004b). Operating within the labour market as it is, individuals are better off taking a temporary job than remaining unemployed, (28) but entry into the labour market for youth in general would be helped substantially by easing the rules for permanent contracts.

The best for youth would be to improve the path to regular employment contracts. To that end, reforms should not deepen the difference between permanent and temporary contracts further; it is anyway hard to think of further marginal changes that could be made. Rather, reforms should address the core of employment protection rules (Box 4.4). In particular, the stringency of the criteria for what is a fair dismissal under a regular contract might be reconsidered, with a view to making employers less hesitant to offer youth a regular contract. In order to reduce the gap between temporary and regular contracts, the trial period for regular contracts could also be lengthened. (29) The complexity of these issues would call for a review of the broader implications, including with respect to collective wage bargaining.


Labour market activation and benefits

Active labour policies

Sweden's active labour market policies were reshuffled in 2007, inter alia with the introduction of a Job Guarantee for youth: after three months of unemployment, persons below 25 are offered intensified study--and job-coaching; if they don't find a job after another three months, they are oriented to workplace practice, i.e. subsidised employment, or training. Half of the participants are expected to be catered for by private contractors. By February 2008, 26% of the registered unemployed youth were participating in a labour market programme of some form, down from 38% in 2006.

The design of the lob Guarantee reflects available research evidence, notably in terms of the emphasis on job-search assistance early in each unemployment spell The frequent short unemployment spells make youth susceptible to "lock-in effects", i.e. reduced job-finding rates due to lower search intensity when enrolled in a labour market programme (Larsson, 2003). In this sense, the lob Guarantee is an improvement relative to its predecessor, the Youth Guarantee, where municipal activation programmes set in only after five months of unemployment. Evaluations found that transitions into jobs were increased by the introduction of the Youth Guarantee, but only during the first five months of unemployment, suggesting a pre-programme deterrence effect: youth knew that they would be sent into the programme if they remained unemployed and this seemed to increase their job-finding rates. No positive effects could be found after the programmes actually started, suggesting that the skills acquired during training were not very useful (Forslund and Skans, 2006a and 2006b). (30)
Box 4.4. What elements of the strict Swedish employment protection
rules may hamper youth entry?

Regulations giving job security for those who have a permanent
contract are relatively elaborate in Sweden. In fact, France is the
only high-income country with stricter employment protection
legislation. The various aspects of this legislation are here
compared to Germany, having similar overall strictness, and Denmark
and the United Kingdom, which are generally more liberal (Figure
4.17). Differences are most remarkable for procedural
inconveniences and difficulties involved with dismissal of an
employee on a regular contract. For example, worker capability or
redundancy of the job are adequate and sufficient grounds for
individual dismissal in Denmark and the United Kingdom, whereas in
Sweden and Germany, a transfer and/or retraining to adapt the
person to different work must be attempted prior to dismissal. (1)
Moreover the trial period during which the employer can easily
terminate a contract must end after six months in Sweden, like in
Germany, whereas Denmark and the United Kingdom allow longer trial
periods of up to 10 1/2 and 12 months respectively. By contrast,
notice periods for individual dismissals are not much longer than
elsewhere, and Sweden has no severance pay. For collective
dismissals, collective agreements may prolong the notice period,
but the other additional requirements are not stricter than


Sweden's last-in-first-out rule for regular employment has been
particularly controversial. Introduced in the 1970s, it remains a
nationally-legislated rule stipulating that employers lay off staff
in reverse order of seniority within each professional segment. (2)
In the public sector the employer is defined so narrowly that a
teacher or childcare worker will lose their position in the
last-in-first-out hierarchy if they shift to work in another
municipality, which is very detrimental to labour mobility. The law
allows collective agreements between employers and unions to set
aside the last-in-first-out rule. Firms undergoing large
restructuring therefore often say that the last-in-first-out rule
is not a hindrance, as they would anyway have a close dialogue with
unions in an effort to facilitate the restructuring. In parts of
the labour market, the main effect of the last-in-first-out rule may
therefore be to elevate union bargaining power and ensure high
union membership, as collective agreements may set the rule aside
also for employees that are not union members, at least in
principle. For youth, the last-in-first-out rule implies more
frequent layoffs, as youth are more likely to have joined the firm
recently. However, the rule is more likely to deter employers from
hiring a 52--than a 27-year old: the employer might be concerned
about how the former would gain seniority and be difficult to lay
off when his/her work capacity is declining. Such a concern would
not matter for the 27-year old, whose work capacity is likely to

1. According to the Swedish Employment Protection Act, a
termination of an employment contract may be deemed unlawful if it
had been reasonable to require the employer to provide other work
in his or her service for the employee. There is a very well
developed case law from the Labour Court concerning this rule. In
principle, an employer is not required to "create" work and/or make
substantial changes in the organisation to meet this requirement.
Nor is the employer required to offer work within the organisation
that the employee does not have the right qualifications for.
However, if the employee can gain the right qualifications through
training within reasonable limits, based on the circumstances in
the individual case, the employer must offer the work to the
employee before termination of the contract.

2. The rule does not protect outright incompetent workers, though,
as these may be dismissed on personal grounds, in which case the
last-in-first-out principle does not apply. Moreover, the
last-in-first-out rule takes into account education and skills as
an employee with longer employment time only has priority over
other employees if he or she has satisfactory qualifications for
continued work.


Starting from comparatively generous income-replacing benefits, a number of changes have been made since 2007 with the purpose of making work pay. Most affect all age groups: the large in-work tax credit, the declining time profile of replacement rates in unemployment insurance benefits, and the so-called New-Start Jobs waiving of employer contributions for persons coming back to work after prolonged absence. As discussed in the previous Survey, these policies can be expected to have a lasting positive effect on the level of employment (OECD, 2007b).

Age-differentiated rules have been introduced for employer contributions, as discussed previously, but only to a small extent for income-replacing benefits. Just two youth-specific benefit changes have been made. First, while the gross replacement rate in unemployment insurance benefits declines from 80% to 70% after 40 weeks for adults, this now happens after only 20 weeks for youth. Second, graduation from secondary or tertiary education no longer gives access to unemployment assistance. This concerns some 10% of the unemployed 19-24 year olds.

In light of the positive experience with differentiating the benefits available for youth and prime-age adults in Denmark, further steps in this direction might be considered. Offering the same benefits across all age groups is likely to cause unemployment and inactivity traps for youth, as they have not yet attained the productivity level of prime-age adults. Moreover, exposing youth to clear economic incentives might appear less problematic from a welfare perspective: youth are less settled and therefore better able to reorient to a new profession or move to another part of the country to take up work. Last, but not least, avoiding benefit dependency among youth is key as they are forming habits that are likely to follow them through adult life. Three issues might warrant attention:

* Unemployment insurance benefits are effectively more generous for youth than for prime-age adults. The formula combines a high gross replacement rate with a ceiling at 75% of average full-time earnings. As youth typically have lower earnings, they are less likely to be facing the ceiling, and they are unlikely to face the high marginal tax rates that prime-age adults may do if they are unemployed only part of the year. Consequently, effective net replacement rates will often be somewhat higher for youth than prime-age adults, even though they might be better able to bear an income reduction as they are less likely to have dependent children. About 5% of the 19-year old registered unemployed receive unemployment insurance benefits, rising to about half of those aged 24. A lower starting replacement rate for youth, say 65%, would give clearer incentives to seek a new job quickly; it would allow the employment service to focus activation efforts on those in genuine need of coaching and training.

* For those who drop out from upper secondary school into inactivity, going back to complete schooling can entail a short-term income loss. This situation arises because from age 19, youth are eligible for the same social assistance as older adults: for a single person, SEK 3 550 per month plus coverage of reasonable expenses for housing, electricity, insurance, etc. On going back to school, the person would, often, no longer be eligible for social assistance, but only for the study grant worth SEK 1 050. (31) The resulting incentive not to return to complete school is problematic not least because persons with incomplete secondary education have a higher risk of unemployment (Box 4.2). (32)

* Withdrawal of social assistance generates a 100% effective marginal tax rate that can discourage taking up temporary or part-time work. How far it extends up the income scale depends on the recipient's housing and family situation. About a third of the registered unemployed aged 19-24 receive social assistance, and this share is likely to grow as an estimated 10% of the registered unemployed 19-24 year olds have lost access to unemployment assistance following the change of rules for school and university graduates.

Social assistance involves complex trade-offs and therefore changes in that area might best be done following a wider review. One could envisage differentiating the level of social assistance for persons below the age of, say, 23 by lowering the standard rates or by introducing a parental means test. The two thirds of the 19-year old social assistance recipients who live with their parents, declining to one third by the age 22, don't need as much public support as an older adult.

Youth entering disability benefits is a highly worrying phenomenon, which has emerged as a major issue in recent years. In 2007, 10% of the persons entering disability benefits were aged 19 to 30, compared with just 4% three years earlier and a stable 2% in the more distant past. One reason for this sudden increase may be that youth attending special schools due to handicaps have gained entitlement to disability benefits; when leaving school, they tend to remain in the system. To break this pattern, consideration should be given to paying handicapped youth in special schools a study grant and removing the automatic entitlement to disability benefits.

Co-ordination between municipalities and central government agencies

Many countries are moving towards integration between the agencies administering benefits and those providing activation, as well as integration between different types of benefits related to unemployment, ill health and social problems. Both Norway and Denmark are following this route, and their experience shows that the way benefit outlays are split between different agencies and levels of government matters a lot for how well activation instruments are used and how many people are granted various benefits. In Sweden, fiscal equalisation compensates municipalities that are likely to have high social spending because of the composition of their population, but costs or gains when actual outlays on social assistance are higher or smaller are fully borne by each municipality. Unemployment insurance benefits are administered by union-organised funds financed by central government subsidies and membership fees. Finally, unemployment assistance, sickness and disability benefits are administered by the public social insurance agency, which is fully financed by central government. These arrangements raise familiar fiscal incentive problems and the fragmentation of service delivery results in users complaining that they are "falling between two stools" as they are sent from one agency to the next without being adequately helped anywhere.

Recent reforms lack clear direction in this respect. The restriction of access to unemployment assistance implies that more youth will come to their municipality for social assistance. However, as the responsibility for youth activation has been shifted to the Job Guarantee scheme run by the central-government Employment Service, municipalities have now lost an important instrument to curb the number of social assistance recipients. A few years ago, the local social security offices were moved from municipalities into a central-government agency. But these offices, which bear the costs of unemployment assistance, sickness and disability benefits, are not linked locally to the Employment Service responsible for activation or municipalities responsible for education. Developing the local coordination between these agencies will be an important challenge.


To tackle high youth unemployment, Sweden needs a broad-based approach touching upon education as well as labour market institutions, taxes and benefits (Box 4.5). A series of education reforms, which virtually all move closer to international best practice, are currently being discussed. These proposals now need to be implemented. The wide-ranging labour market reforms put in place from 2007 to make work pay help, but they could go further for youth by amplifying the incentives to seek work while avoiding encouraging delays in the start and completion of studies.
Box 4.5. Summary of education and youth employment recommendations
Compulsory education

* Strengthen compulsory schooling, not least in science, and
continue to develop the use of tests to better monitor progressions
during school.

* Introduce accreditation of teachers and give more weight to
teachers' specific competences when allocating tasks among staff.
Once the accreditation scheme is well established, greater wage
flexibility could reward the best teachers.

Upper secondary education

* Introduce apprenticeships anchored into the same competence
structure as school-based vocational programmes so as to allow
hands-on learning while ensuring that apprentices acquire the
general skills they will need if later in life they have to
reorient professionally. Monitor how pay for apprentices is set
locally to ensure ongoing employer interest.

* Increase transparency for potential students about the labour
market prospects associated with alternative education choices,
e.g. by publishing data for employment outcomes of recent school

* Analyse the possibility of moving the responsibility for upper
secondary schools from municipalities to central government, while
giving each institution more independence based on per-student
funding. This would allow public secondary schools more room to
merge, reorganise and specialise, helping vocational programmes in
particular. Such an analysis could also look at other levels of

Higher education

* Base the allocation of core research funding on more transparent
criteria and give universities and university colleges (hogskolor)
more freedom to each develop their own strengths.

* Adjust the grants students receive to cover living costs:
consider introducing rewards for early entry and on-time completion
of studies or changing the composition so that loans become a
larger share and grants a smaller share of total financial support
when studies get prolonged.

* Consider moving gradually towards a system where not only non-EU
students, but also Swedish and EU students are charged for tuition,
while extending the government loans available for students to
finance tuition costs. If today's high marginal income taxes are
reduced, repayment of these loans could be made contingent on
graduates' future income.

Labour market institutions, social contributions and benefits

* Reconsider the general reduction in employer contributions for
youth in favour of measures that are more narrowly targeted at
those with difficulties getting employment.

* Avoid further dualism in employment protection rules. One
approach would be to create better in-roads to regular employment,
such as by easing the definitions of what is a fair dismissal and
lengthening the trial period for regular contracts.

* Focus on job-search coaching in the early phase of youth
unemployment spells, as done with the lob Guarantee for Youth, so
as to avoid lock-in effects where participation in activation
programmes reduces job-search intensity.

* Consider differentiating unemployment benefits by age: lower the
initial gross replacement rate in unemployment insurance for
persons below 25. Conditions for social assistance might also be
reviewed, including for youth living with their parents.

* Pay a study grant, not disability benefits, to handicapped youth
in special schools.

* Improve local co-ordination between municipalities,
central-government employment services and social insurance offices
and unemployment insurance funds.


Ahlin, A. (2003), "Does School Competiton Matter? Effects of a Large-Scale School Choice Reform on Student Performance", Working Paper No. 2003:2, Department of Economics, Uppsala University.

Alstadsaeter, A., A.-S. Kolm and B. Larsen (2008), "Money or Joy: The Choice of Educational Type", European Journal of Political Economy, Vol. 24.

Aslund, O., R. Erikson, O.N. Skans and A. Sjogren (2006), Fritt intrade?--Ungdomars och invandrades vag till det forsta arbetet (Free Entry?--Youth' and Immigrants' Way to the First Job), SNS Forlag, Stockholm.

Bjorklund, A., M. Clark, P.-A. Edin, P. Frederiksson and A. Krueger (2006), The Market Comes to Education in Sweden: An Evaluation of Stoeden's Surprising School Reforms, Russell Sage Foundation Publications.

Boarini, R. and H. Strauss (2007), "The Private Internal Rates of Return to Tertiary Education: New Estimates for 21 OECD countries", OECD Economics Department Working Paper No. 591.

Bohlmark, A. and M. Lindahl (2007), "The Impact of School Choice on Pupil Achievement, Segregation and Costs: Swedish Evidence", IZA Discussion Paper, No. 2786.

Bovenberg, A. and B. Jacobs (2005), "Redistribution and Education Subsidies are Siamese Twins", Journal of Public Economics, Vol. 89, No. 11-12.

Calmfors, L., A. Forslund and M. Hemstrom (2004), "The Effects of Active Labour Market Policies in Sweden: What Is the Evidence?", in J. Agell, M.J. Keen and A.J. Weichenreider (eds.), Labor Market Institutions and Public Regulation, MIT Press, Cambridge, MA.

Carlsson, M. and D. Rooth (2007), "Evidence of Discrimination in the Swedish Labor Market Using Experimental Data", Labour Economics, Vol. 14, No. 4.

Confederation of Swedish Enterprise (2008), Akademi eller verklighet? (Academy or Reality?), Stockholm.

Danish Ministry of Education (1998), Rapport om taxameterstyring (Report on per-student funding), Copenhagen.

de Luna, X., A. Forslund and L. Liljeberg (2008), "Effekter av yrkesinriktad arbetsmarknadsutbildning for deltagare under perioden 2002-04" (Effects of Vocational Training for Participants During 2002-04), Rapport 2008:1, IFAU.

Forslund, A., P. Johansson and L. Lindqvist (2004), "Employment Subsidies--A Fast Lane From Unemployment to Work?", Working Paper No. 2004:18, IFAU.

Forslund, A. and O. Skans (2006a), "Swedish Youth Labour Market Policies Revisited", Vierteljahrshefte zur Wirtschaftsforschung, Vol. 75, No. 3.

Forslund, A. and O. Skans (2006b), "(Hut) hjalps ungdomar av arbetsmarknadspolitiska program for unga?" (How are Youth Helped by Labour Market Programmes for Youth?), Rapport 2006:5, IFAU.

Globalisation Council (2008), Globaliseringen och den hogre utbildningen, Stockholm.

Holmlund, B. and D. Storrie (2002), "Temporary Work in Turbulent Times: Swedish Experiences", Economic Journal, Vol. 112, No. 480.

Holmlund, B., Liu Q. and O. Skans (2008), "Mind the Gap? Estimating the Effects of Postponing Higher Education", forthcoming in Oxford Economic Papers.

Kahn, L. (2007), "The Impact of Employment Protection Mandates on Demographic Temporary Employment Patterns: International Microeconomic Evidence", Economic Journal, Vol. 117, No. 521.

Larsson, L. (2003), "Evaluation of Swedish Youth Programs", Journal of Human Resources, Vol. 38, No. 4.

Lemaitre, G. (2007), "The Integration of Immigrants into the Labour Market: the Case of Sweden", OECD Social, Employment and Migration Working Paper No. 48.

LO (2008), Etableringen pa arbetsmarknaden--bur klass, kon och etnicitet styr ungdomars villkor (Establishing on the Labour Market--How Class, Gender and Ethnicity Determines Youth's Conditions), Landsorganisationen i Sverige, Stockholm.

Ludwig, V. and F. Pfeiffer (2005), "Abschreibungsraten allgemeiner und beruflicher Ausbildungsinhalte", (Depreciation Rates for General and Vocational Curricula), ZEW Discussion Paper No. 05-36.

Lundsgaard, J. (2003), "Competition and Efficiency in Publicly Funded Services" OECD Economic Studies, No. 35.

Ministry of Finance (2008), Kraftfullare nedsattning av socialavgifter for unga (Stronger reduction in social contribution for youth), Promemoria, 15 May, Stockholm.

National Agency for Education (2008), Barn, elever och personal (Children, Students and Personnel), Report No. 315, Stockholm.

National Agency for Higher Education (2006), OECD Thematic Review of Tertiary Education--Country Background Report for Sweden, Stockholm.

National Mediation Office (2007), Avtalsrorelsen och lonebildningen 2007 (Collective Agreements and Wage Formation 2007), Stockholm.

NIER (2007), Lonebildningsrapporten 2007 (Wage Formation Report 2007), National Institute for Economic Research, Stockholm.

OECD (2004a), OECD Economic Surveys: Germany, Paris.

OECD (2004b), "Employment Protection Regulation and Labour Market Performance", OECD Employment Outlook, Paris.

OECD (2005), OECD Economic Surveys: United Kingdom, Paris.

OECD (2006a), Assessing Scientific, Reading and Mathematical Literacy: A Framework for PISA 2006, Paris.

OECD (2006b), OECD Economic Surveys: Germany, Paris.

OECD (2006c), Boosting Jobs and Incomes--Policy Lessons from Reassessing the OECD Jobs Strategy, Paris.

OECD (2006d), Jobs for Youth: Spain, Paris. OECD (2007a), PISA 2006--Science Competencies for Tomorrow's World, Paris.

OECD (2007b), OECD Economic Surveys: Sweden, Paris.

OECD (2008a), "Off to a Good Start? Youth Labour Market Transitions in OECD Countries", OECD Employment Outlook, Paris.

OECD (2008b), Education at a Glance, Paris.

OECD (2008c), Going for Growth, Paris.

OECD (2008d), Learning for Jobs: The OECD Policy Review of Vocational Education and Training in Sweden, Paris.

OECD (2008e), Tertiary Education for the Knowledge Society, Paris.

OECD (20080, OECD Economic Surveys: Australia, Paris.

OECD (2008g), Jobs for Youth: The Netherlands, Paris.

OECD (2008h), OECD Economic Surveys: Norway, Paris.

Oliveira Martins, J. et al. (2007), "The Policy Determinants of Investment in Tertiary Education", OECD Economics Department Working Paper No. 576.

Quintini, G., J. Martin and S. Martin (2007), "The Changing Nature of the School to Work Transition Process in OECD countries", IZA Discussion Paper No. 2582, Bonn.

Quintini, G. and S. Martin (2006), "Starting Well or Losing their Way?: The Position of Youth in the Labour Market in OECD Countries", OECD Social Employment and Migration Working Paper No. 39.

SACO (2008), Kundskapens lage 2008 (The State of Knowledge 2008), Stockholm.

Sandstrom, M. and F. Bergstrom (2005), "School Vouchers in Practice: Competition Will Not Hurt You", Journal of Public Economics, Vol. 89, No. 2-3.

Skans, O. (2004), "Scarring Effects of the First Labour Market Experience", Working Paper No. 2004:14, IFAU.

Skedinger, P. (2006), "Minimum Wages and Employment in Swedish Hotels and Restaurants", Labour Economics, Vol. 13, No. 2.

Skedinger, P. (2007), "The Design and Effects of Collectively Agreed Minimum Wages. Evidence from Sweden", IFN Working Paper No. 700.

Skedinger, P. (2008), Effekter av Anstallningsskydd (Effects of Employment Protection), SNS Forlag, Stockholm.

SOU (2006), Studieavgifter i hogskolan (Tuition Fees in Higher Education), Swedish Government Official Reports, No. 2006:7.

SOU (2008a), Legitimation och skarpta behorighetsregler (Registration and Stricter Qualifying Rules), Swedish Government Official Reports, No. 2008:52.

SOU (2008b), Framtidsvagen--en reformerad gymnasieskola (The Future Path--A Reformed Secondary School), Swedish Government Official Reports, No. 2008:27.

SOU (2008c), Valja fritt och valja ratt (Choosing Freely and Making the Right Choices), Swedish Government Official Reports, No. 2008:69.

Statistics Sweden (2006), "Intradet pa arbetsmarknaden" (Entry into the Labour Market), UF 86 SM 0601, Stockholm.

Statistics Sweden (2007), "Sambandet mellan utbildning och yrke--En studie om hogskoleutbildningar" (The Link between Education and Profession), Befolkning and Valfard, No. 1, Stockholm.

Strath, A. (2004), Teacher Policy Reforms in Sweden: The Case of Individualised Pay, Paper submitted by the Swedish Ministry of Education and Science to the International Institute for Educational Planning under UNESCO.

Welfare Commission (2006), Fremtidens velfoerd--votes valg (The Future Welfare--Our Choice), Final report of the Danish Welfare Commission, Copenhagen.


(1.) It is sometimes questioned whether the international definitions are applied in exactly the same way across countries, in particular when it comes to how widely students looking for part-time or holiday employment are included. A simple gauge is to compare the share of full-time students being counted as unemployed and the share of students among all unemployed in the Labour Force Survey. Applying this test to the Sweden-Denmark comparison does not indicate any real problems. The full-time students among the 15-74 year olds were distributed as follows in Sweden, 2006:158 500 were employed, 88 100 were unemployed, and 535 000 were outside the labour force, meaning that 14.1% of the full-time students not working were counted as unemployed. For Denmark, the same share was 11.4% in 2005 and 11.7% in 2007. Full-time students accounted for a steady 26% of all unemployed (age 15-74) in Sweden, 2005-07, where the economy moved from a slight negative to a slight positive output gap. For Denmark, full-time students accounted for 19% of all unemployed (age 15-66) in 2005, when the economy was close to equilibrium, and 24% in 2007, when there was a clear positive output gap.

(2.) The short duration of most unemployment spells may partly reflect that the persons concerned enter education or a labour market programme, or simply stop searching. For those registered with the Public Employment Service, it is possible to exclude such cases statistically: for youth, the median duration of unemployment slightly exceeds four months for the spells that end in a job, as against five months for persons in their 40s.

(3.) Thus NEET treats jobs and education equally on the one side and job search and inactivity equally on the other. It is important to note that participants in all forms of activities, including various labour market programmes, are treated as being active. A consequence is that countries with large volumes of labour market programmes will have lower NEET rates, which may explain part of the more positive picture emerging from these statistics. The position relative to Finland, Norway, the Netherlands and Denmark remain the same as for unemployment: Sweden and Finland appear to be the worst-performing countries in this group.

(4.) For mathematics, the fall in average PISA score from 2003 to 2006 is statistically significant at the 90% confidence level. For reading, there is a noticeable quantitative fall in the average PISA score from 2000 to 2006, but it is not statistically significant.

(5.) In Sweden most students with special needs attend ordinary schools, but this should not bias the result as the PISA study generally includes students enrolled in special educational institutions.

(6.) Given that Swedes are typically known for environment-friendly habits, it may surprise that 15-year olds are not to a larger extent indicating to be aware of the science aspects of important environmental issues. It might reflect a discrepancy between the value attached to environmental protection and the quality of science teaching. Within each country, there is a strong correlation between each student's self-reported awareness of environmental issues and the science score obtained in the PISA test.

(7.) Employers often point to recruitment difficulties, but as labour is a scarce resource, the key question from an economy-wide perspective is whether labour shortages are more pronounced for certain professional groups than others. A recent study classified 25-54 years olds with tertiary education based on whether they work in the area and at the competence level matching their education (Statistics Sweden, 2007). The traditional longer engineering degree (civilingenjor) was among the educational groups having the highest match rate indicating relative scarcity. For natural science graduates and the shorter engineering degree introduced in the early 1990s (hogskoleingenor), however, this was not the case.

(8.) Gross statutory salaries for teachers are, at first glance, among the lowest in the OECD. In 2006, gross salaries for teachers with the minimum training required and 15 years of experience were 88%, 91% and 98% of GDP per capita in primary, lower secondary and upper secondary education, respectively, compared with OECD averages of 122%, 126% and 134% (OECD, 2008b). However, supplements to base salary based on competence may have a larger extent than in most countries. Moreover, employer contributions (24.5% of wage costs) are higher than in most OECD countries (ranging from 0.0% to 29.6%), diminishing gross pay relative to employers' total wage costs.

(9.) Due to data availability, the studies are all based independent schools in compulsory education. However, there is no particular reason to presume that the effects would not be similar at upper secondary level.

(10.) The Learning for Jobs OECD review is part of a series of country review of vocational education and training prepared by the OECD Education Directorate. Aside from Sweden, reviews are carried out for the United Kingdom, Hungary, Australia, Norway, Mexico, Korea and Switzerland between late 2007 and late 2008. Together with cross-country analytical work, the country reviews will feed into a final comparative report to be published in early 2009.

(11.) This development is best documented in an evaluation where school managements was asked about what changes had been initiated by per-student funding (Danish Ministry of Education, 1998).

(12.) Throughout this Survey, the word "university" refers to institutions of higher education in a broad sense. In a Swedish context, it includes universities (universitet) as well as university colleges (hogskolor).

(13.) No country-specific report on Sweden was prepared by the OECD Secretariat as part of this thematic review. However, a seminar was organised on 26 May 2008 discussing inter alia the challenges mentioned in the text.

(14.) Strictly speaking, if wages were completely and uniformly fixed by collective agreements, changes in employer contributions for youth would not affect the hourly earnings of youth. In practice however, youth working on-and-off before and while studying typically take up work in areas with considerable de facto wage flexibility, implying that changes in employer contributions would feed through to individual pay. Moreover, as lower contributions makes it more attractive for shops and restaurants etc. to hire youth relative to low-skilled prime-age adults, it will be easier for youth and students to find work, raising the effective earnings net of implicit job-search costs, transport, etc.

(15.) If the study grant for living costs was liable for income taxation, like most income benefits, the 27% versus 63% gap in marginal tax wedge would only apply for earned income up to SEK 13 000, not SEK 45 000. Perhaps the special tax excemption for study grants is less warranted today than it was before the introduction of the in-work tax credit and the employer contribution rebate for youth.

(16.) The value of the study grants has decreased relative to market wages, from 75% of industrial wages net of taxes to 55% between 1992 and 2005 (LO, 2008). Nevertheless, students are able to attain considerable income: they are allowed to earn SEK 102 500 annually before withdrawal of the study grants starts.

(17.) This benefit has not been touched in recent reforms: persons aged 18-28 without children are offered compensation for 75% of monthly housing costs in excess of SEK 1 800, followed by 50% of costs in excess of SEK 2 600 up to a ceiling at SEK 3 600. Consequently, the maximum monthly value of the benefit is SEK 1 100. The key question might be whether the grant is warranted at all in its current form: by subsidising a "city-centre lifestyle" it may well prolong study duration. Moreover, the high compensation percentage risks encouraging landlords to raise rents when these are liberalised (Chapter 1).

(18.) Free tuition and generous grants may attract potential students, but the high marginal tax rates encourage students to emphasise consumption aspects when choosing when and what to study while paying less attention to labour market prospects (Alstadsaeter et al., 2008).

(19.) For example, over half of the students admitted to the new masters programme at Gothenburg Business School come from outside the European Union.

(20.) A proposal was prepared two years ago (SOU, 2006). In June this year, the government announced its intention to move forward, although more detailed information about how tuition fees for non-EU students would be implemented have not been given yet. EU rules require that students from other EU member states should be treated in the same way as nationals. Consequently, EU students can only be required to pay tuition if Swedish students are.

(21.) The exception is some white-collar agreements with an age limit of 24 (National Mediation Office, 2007).

(22.) Already prior to these increases, minimum wages affect many workers in these service sectors: 30% of all employees in areas covered by the hotels-and-restaurants agreement receive the minimum wage; for the retail agreement the share is 20% (NIER, 2007). Youth are heavily affected, as they are overrepresented in these two areas: 29% of employed upper secondary school graduates work in retail trade or hotels and restaurants three years after graduation (Statistics Sweden, 2006). More generally, many young workers find their first lasting jobs in these and other parts of the service sector (Aslund et al., 2007).

(23.) It should be noted that New-Start Jobs can be combined with the reduction of employers' contributions.

(24.) The spell can be longer if joblessness was due to sickness.

(25.) Under 2006 rules, fixed-term contracts could be used without particular time limits under certain circumstances, for example contracts concerning a specified season or specified task, temporary substitute employment, and cases where the employee has attained the age of 67. But aside from these cases, it was possible for an employer to conclude a fixed-term contract with one and the same employee for at most a year within a three-year period (so-called agreed fixed-term employment). An employer was never allowed to have more than five employees with agreed fixed-time employment at the same time.

(26.) According to Swedish Labour Force Survey estimates, on average 117 000 students aged 15 to 24 worked in 2007, of which 102 000 were enrolled full-time. This means that on average 23% of full-time and part-time students aged 15-24 were employed in a given month during the year. Those who worked did so for on average 14 hours per week. This means that student employment amounts to 43% of all temporary employed youth, or if only counting full-time students, 37%. These numbers only include youth working while studying, and not summer jobs taken in-between studies. The fraction of students on leave among temporary employees during summer is likely to be considerable. Thus, the impact of the supply of students willing to work is likely to have been a major contributor to the growth of temporary contracts.

(27.) The incidence of temporary employment has been found to be higher for those with lower cognitive ability in countries with stricter employment protection rules (Kahn, 2007).

(28.) Taking the features of the Swedish labour market as given, the limited available empirical evidence on the consequences of individual behaviour suggests that temporary contracts are beneficial for those taking them up (Larsson et al., 2005).

(29.) A similar recommendation has been made for Spain (OECD, 2006d).

(30.) In a longer run, labour market training may have positive effects, and these effects may have improved in recent years (de Luna et al., 2008).

(31.) There is also an extra assistance of at most SEK 855 for individuals from households with low income. In order to be eligible for social assistance the recipient must search for work; studying is not considered a valid exception. Parents are responsible for children attending compulsory and upper secondary schooling until the age of 21. From the age of 20, individuals attending upper secondary education are eligible for the same financial support as students attending higher education, that is, grants and loans combined to SEK 7 492 per month (4 weeks).

(32.) In principle, a similar problems arises if dropping out from higher education, but it matters little in practice because the study grant for tertiary education is higher, at SEK 2 572 per month plus loans.
Table 4.1. Estimated percentage differences in job-finding rates

                                              Estimate      error
  19 (39%)                                   -0.183 ***    (0.017)
  20 (24%)                                   -0.046 ***    (0.018)
  21 (15%)                                        0.025    (0.018)
  22 (11%)                                        0.019    (0.019)
  23 (10%)                                        (ref)

  Female (47%)                                    (ref)
  Male (53%)                                 -0.055 ***    (0.009)

Immigrant background
  Arrived at age 12 or after (2%)            -0.273 ***    (0.040)
  Arrived before age 12 (7%)                 -0.270 ***    (0.019)
  Second generation with both parents
  being immigrants (5%)                      -0.131 ***    (0.021)
  Natives (86%)                                   (ref)

Family situation
  Living with parents (65%)                  -0.042 ***    (0.010)
  No children, not with parents (31%)             (ref)
  Children and spouse (2%)                   -0.160 ***    (0.032)
  Single parent (1%)                         -0.277 ***    (0.047)

Compulsory school grade decile
  1 (12%)                                    -0.518 ***    (0.026)
  2 (12%)                                    -0.407 ***    (0.025)
  3 (12%)                                    -0.308 ***    (0.024)
  4 (12%)                                    -0.267 ***    (0.024)
  5 (11%)                                    -0.229 ***    (0.024)
  6 (11%)                                    -0.184 ***    (0.023)
  7 (10%)                                    -0.161 ***    (0.024)
  8 (9%)                                     -0.122 ***    (0.024)
  9 (7%)                                     -0.076 ***    (0.025)
  10 (6%)                                         (ref)

Highest completed education by end
of the year
  Compulsory (14%)                           -0.363 ***    (0.016)
  Some vocational upper sec. (10%)           -0.231 ***    (0.016)
  Complete vocational upper sec. (41%)            (ref)
  Some academic upper secondary (4%)         -0.233 ***    (0.023)
  Complete academic upper sec. (24%)         -0.026 **     (0.012)
  Some tertiary (4%)                         -0.223 ***    (0.025)
  Complete tertiary (3%)                          0.046    (0.028)

Note: Data are from Public Employment Service registers (Handel)
and the Louise database. Results are from a Cox proportional hazard
model, N = 77 921. The model is estimated on registered
unemployment spells starting in 2005. Jobs are either temporary or
regular. Participation in subsidised jobs for 30 days or more is
considered finding a job. A random half of those with lost contacts
are considered to have found a job. Other exits are censored.
Percentages in parentheses indicate the share in the total set of
observations. *** (**) indicates significance at the 99% (95%)
COPYRIGHT 2008 OECD Publications and Information Centre
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Chapter 4
Publication:OECD Economic Surveys - Sweden
Geographic Code:4EUSW
Date:Dec 1, 2008
Previous Article:Taxation and growth: what direction should Sweden take?
Next Article:Privatisation: the next phase.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters