Skills, enrolment and hybridity in the Dutch skill regime : a cross-sector analysis of qualifications in the upper secondary vocational education system

(1)

Skills, Enrolment and Hybridity in the Dutch

Skill Regime

a cross-sector analysis of qualifications in the upper

secondary vocational education system

Joshua Verkerk (5897815)

Sociology

First supervisor: dr. V. Di Stasio

Second supervisor: dr. J. De Deken

July 2014

Master‟s Thesis in Sociology,

track: „Comparative Organisation and Labour Studies‟

University of Amsterdam

(2)

(3)

3

Summary

This research has analysed differences in skill content and enrolment between manufacturing- and service sector qualifications in the upper secondary vocational education system. The Netherlands is considered to have a collective skill formation system, in which social partners are involved in the provision of skills. In the Dutch case the collectivity can be observed both by the involvement of employers in the determination of qualifications and the fact that firms offer apprenticeships. Therefore, the Netherlands is considered to be a hybrid case, combining features of a school-based and workplace-based skill regime. In the upper secondary vocational education system two tracks exist in which the prime venue of training is respectively school (BOL) and firm (BBL).

A shortcoming to earlier research on collective skill formation is the fact that it tends to overlook differences between the manufacturing and service sector herein. In order to provide more insight on this issue, this research answers the following question: What are the differences in the skill content and enrolment between the manufacturing- and service sector qualifications in the Netherlands on the upper secondary vocational education level?

In order to answer this a content analysis is performed on qualification documents of the upper secondary vocational education system in the Netherlands, combined with an analysis of the enrolment in the two tracks this system contains.

The results show that qualifications in the service sector rely more on theoretical knowledge and social skills, whereas the manufacturing sector contains more manual skills. General cognitive knowledge is also important, but this is equally distributed across the two sectors. Service sector qualifications have a high enrolment in BOL track whereas manufacturing qualifications have a high enrolment in the BBL track. This result shows that the Netherlands is indeed a hybrid case and that this hybridity is associated to sectors. The service sector has more similarities with the school-based skill regime, whereas the manufacturing sector has more similarities with the workplace-based skill regime.

(4)

(5)

5

Introduction

Vocational education is an important feature of the education system in the Netherlands. In 2012 79,5% of the students were enrolled in vocational program, as opposed to 20,5% of the students that are enrolled in a tertiary academic program (CBS, 2014). With 43,6% of the students enrolled in an upper secondary vocational education program, this form of education is the most popular one (ibid., 2014).

The Dutch upper secondary vocational education system may be characterized as a collective skill formation system. Collective skill formation refers to a system in which vocational education is organised collectively by the State, firms and intermediary associations (e.g. employers‟ associations) (Busemeyer, 2012: 4). An important feature of such systems is that training partially takes place within firms and provides portable certified skills (ibid., 2012:4). As a result, students that have obtained a degree can more easily move from one employer to another. Because cross-national differences and commonalities among such systems have been widely discussed (cf.: Estevez-Abe et al., 2001; Martin 2012).

Within the Dutch vocational education system all qualifications are based on qualification documents. These documents are collectively agreed upon by school- and labour market representatives (i.e. employer and/or employee organizations). These qualifications may be offered in two different tracks (MBO Raad, 2009). In one track students receive the largest share of their education in a school environment, which is complemented with internships (Beroepsopleidende leerweg – BOL). Following the other track students are apprentices in a firm, which is complemented with a smaller share of their education in a school environment (Beroepsbegeleidende leerweg – BBL). In principle all qualifications can be attended in both tracks, and students would receive the same degree. The cooperation of education- and labour market stakeholders points out the collective character of skill formation in the Netherlands. In both tracks, part of a students‟ education takes place at the workplace. Moreover in both tracks an authoritative certified degree is provided.

While much research is done on cross-national differences in (collective) skill formation, it has been argued that also existing differences between sectors within the same collective skill system must also be accounted for. In post-industrial economies the service sector has an increasing share of the entire economy; the differences between the manufacturing- and service sector have become increasingly important (Anderson & Hassel, 2008). The Netherlands is a particular country in this sense, traditionally it is characterized more by trade than the industrial sector, which indicates the service sector has an important role in the Dutch economy (ibid., 2008). Moreover, it is considered a hybrid case (ibid., 2008) in which both a school-based and workplace-based track are offered. In principle all qualifications can be offered in both these tracks (Van Lieshout & Scholing, 2009).

(8)

8 In previous literature (cf. Estevez-Abe et al., 2001; Soskice & Finegold, 1988), it is assumed that on the upper secondary vocational level asset-specific skills are needed. However, some elements must be taken into account. First, the need for such skills differs strongly per country (cf. Busemeyer, 2009) - not all countries with a collective skill formation system are strongly oriented towards specific skills. Second, the need for such skills may differ per sector; within the service sector specific manual skills are less in demand by employers (Anderson & Hassel, 2008). This results into the criticism that many studies on specific skill formation primarily account for the provision of manual skills, and not skills on the intermediary level in general.

Contributions that attempted to account for these issues usually aimed at providing more insight in institutional differences among countries (Anderson & Hassel, 2008; Busemeyer, 2012). However, cross-sector differences in skill provision remain unclear. The aim of this research is to provide more insight in this component by analysing qualification documents and the differences herein between the service- and manufacturing sector. Therefore I will research the following question:

What are the differences in the skill content and enrolment between the manufacturing- and service sector qualifications in the Netherlands on the upper secondary vocational education level?

Including the service sector into the research on skill provision is important for two close related reason. First, in the contemporary economy there is an increase in service related occupations, rather than manual work. Second, such an analysis is in particular of importance for the Netherlands because of its tradition in trades rather than industry (Anderson & Hassel, 2008). Research that focusses on industrial foundations of skill provision (cf. Streeck, 1989; Finegold & Soskice, 1988; Estevez-Abe et al.,2001) might thus be less applicable to the Dutch situation, than it does to economies with an industrial orientation (e.g. Germany).

In order to answer this research question I will compare the skills in manufacturing- and service sector qualifications in the Dutch upper secondary vocational education system, as well as the enrolment in either a school-oriented or workplace-oriented track.

In the following section I will discuss existing literature regarding skill formation and the potential differences between skill formation in the manufacturing and service sector. I will focus on the varieties of capitalism literature and the literature regarding skill regimes in coordinated market economies. The latter I will use to discuss what kind of skill regimes the Netherlands has, and what is unclear about the skill provision in the Netherlands. Following, I will elaborate on the concepts of high skills equilibrium and diversified quality production which are assumed to be underlying premises to collective skill formation but seem to less fit collectively provided service sector skills. Furthermore, I will go deeper into to the existing literature on collective skill formation and to what extent service

(9)

9 sector skills do and do not fit this theory. Finally I will outline the Dutch upper secondary vocational education and place the theoretical concepts discussed above into this context.

After discussing the existing literature on skill provision I will outline the research design and the process of data collection. Hereafter the results will be presented and conclusions will follow.

Theory

Varieties of capitalism and collective skill formation

Skill formation is related to a variety other institutions in society. Firms have an interest in a skilled population from which they can hire employees. When firms invest in training, however, there is the risk that employees leave for another employer and the firm will not reap the investment they have made, so-called „poaching‟ (Streeck, 1989; Acemogule & Pischke, 1999). To understand how the issue of training investment is coped with, the varieties of capitalism literature stresses the importance of institutional complementarity (Hall & Soskice, 2001). We should understand the national political economy as „the way in which firms resolve the coordination problems they face‟ (Hall & Soskice, 2001:8). The varieties of capitalism literature distinguishes between liberal market economies (LMEs) and coordinated market economies (CMEs), which differ in the extent to which firms rely on either market or non-market related modes of coordination. In LMEs coordination between firms takes place primarily via „hierarchies and competitive market arrangements‟ (Hall & Soskice, 2001:8) whereas in CMEs „firms depend heavily on non-market relationships to coordinate their endeavours‟ (ibid., 2001:8). Thus, in the latter, many non-market institutions play a part in ensuring coordination, such as trade unions, employer organizations and the State.

Within the varieties of capitalism literature, the Netherlands is characterized as a CME. In order to analyse the Dutch skill regime, the theoretical focus should lie on the non-market modes of coordination. In the Dutch collective skill formation system, such coordination consist of the cooperation of employer associations, trade unions and schools to develop qualifications - which are authorized by the State and must underlie the programs that are offered in the upper secondary vocational education system1. Moreover, firms offer apprenticeships to students. These programs are also based on the qualification documents.

In order to understand the way skills are acquired, Estevez-Abe et al. (2001) argue we should distinguish between general skills on the one hand, skills that are „recognized by all employers‟. And on the other hand firm- and industry-specific skills; skills that are acquired respectively through „on-the-job training‟ and „apprenticeships and vocational schools‟. The dimension on which these skills differ is „asset-specificity (i.e. portability)‟ (ibid., 2001:148). By portability is meant the extent to

1

It is assumed that this strategy causes students to obtain credentials in the education system that are based on the needs of the labour market (SBB, 2014a)

(10)

10 which employees are able to take skills from one employer to another. The academic debate on the sufficiency of this distinction is often aimed at whether „specific skills‟ exist as such (cf. Busemeyer, 2009; Streeck, 2011; Culpepper, 2007). Three forms of critique may be distinguished.

The first set of criticism is aimed at the assumption that only a one dimensional skills distinction exists. It has been proposed that the differences in skills must be explained in a multi-dimensional model, in which not only an economic but also a substantive dimension is of importance (Streeck, 2012). In the economic dimension skills are either portable or non-portable. The substantive dimension refers to the content of skills: whether someone has acquired a narrow or broad set of skills. This argument by Streeck is meant to show that skill specificity can be understood in different ways and that in previous contributions these dimensions are assumed to be similar: narrow skills would be non-portable and broad skills portable, while this does not need to be the case (Streeck, 2012). Broad skills may very well be non-portable, as well as narrow skills could be highly portable. In the varieties of capitalism approach, skill portability is indirectly addressed (Busemeyer, 2009) by measuring the enrolment in upper secondary vocational education (cf. Estevez-Abe et al., 2001), but skill content remains an overlooked component. This component I will address in this study by analysing qualification documents.

The second set of criticism is aimed at the disregarded institutional variety across coordinated market economies. Within CMEs there are still relevant differences in the way vocational education is organized and the sort of skills it provides. Rather than approach coordinated market economies as one homogenous cluster, three separate skill regimes within CMEs may be identified (Busemeyer, 2009). „[At least] two dimensions are necessary to map the variety of training regimes in CMEs. One relates to the way vocational skills are authoritatively certified within and through educational institutions, and the other captures the degree of involvement of firms in skill formation and the workplace-relatedness of vocational training.‟ (ibid., 2009:384). These dimensions led Busemeyer to develop a typology of three skill regimes: First, a segmentalist regime, in which skill formation is primarily achieved within firms and without collective agreement. Second, an integrationalist regime, in which skill formation is primarily done within vocational schools, and thus school-based. And finally, Busemeyer identifies a differentiated skill regime in which skill formation is determined by the social partners and workplace-based. The key difference, thus, between the segmentalist en differentiated skill regime is that the latter is collectively organised, whereas in the segmentalist regime firms provide training for their own employees. The conceptualization of such skill regimes is meant to show how the relationship between education, labour market institutions and industrial relations is shaped within a country. The Netherlands is a hybrid regime (Anderson & Hassel, 2008), which is situated between a school-based (integrationalist) skill regime, and workplace-based (differentiated) skill regime. The most important reason for this is the combination of two tracks which respectively primary offer training in schools and in firms (Lieshout & Scholing, 2009). Less,

(11)

11 however, is known about how these tracks are distributed across sectors. This research will contribute to the literature by focussing on these sector differences, by focussing on the manufacturing and service sector.

Literature regarding skill regimes is primarily aimed at integrationalist versus differentiated regimes. It might be argued that segmentalist regimes are not as collectively organized, because training takes place within firms, without State regulation or sector determination. Also within this research the focus will lie on the integrationalist and differentiated approach, because these are most relevant to the Dutch case. As I will argue below State commitment is of importance in the Netherlands, as well as the collective character of skill formation in general.

What Busemeyer calls school-based and workplace-based regimes could be considered „collective skill regimes‟. The indicators used to measure the extent to which skill provision is collectively organized are the employer involvement and the public commitment (Busemeyer, 2012:12). The commitment of firms is expressed in the extent to which firms are willing to invest in the formation of skills. By the public commitment of skill formation is meant the public investment, as well as the extent to which skills are certified and standardized. These commitments result into systems that vary in the venue of training, type of specific skills, portability and the levels of state subsidy (Martin, 2012:43). In the Netherlands the venue of training could either be within the firm or in schools, the type of specific skills either (or both) industry- or firm-specific. Portability is the extent to which employees are able to take skills from one employer to another2. State subsidy is the budget for the purpose of training (Anderson & Oude Nijhuis, 2012). In the Netherlands especially the public commitment is of importance (measured by the relative size of public investments in skill formation) and is also high in international perspective. Employer involvement, although of importance, is average in international comparison3 (Busemeyer & Schlicht-Schmälze, 2014).

The characteristics that are addressed by Martin are related to each other. A key difference between firm- and industry specific skills is the fact that industry specific skills are transportable to other employers within the same industry (Estevez-Abe et al., 2001). Industry specific skills are therefore related to the certification of skills. Certification of skills indicates that it is clear to employers what kind of skills someone possesses and thus can signal whether they are useful within the firm. The fact that all credentials in the upper secondary vocational education system underlie a collectively authorized qualification document could be considered the kind of certification Martin (2012) refers

2_{Portability and specificity are not the same per se, as portability is also influenced by other factors such as the}

extent to which employees have opportunities to switch employers. For a detailed examination of this issue see Streeck (2012).

3

It must be noted, however, that the measurement of Busemeyer & Schlich-Schmälze (2014) is primarily focussed on the employer involvement in regard to offering apprenticeships. Although other forms are included as well, extra weight is given to countries with a high degree of apprenticeship takers. This seems to

(12)

12 to, because from the credentials that students obtain employers are able to signal the skills that students possess. For the upper secondary vocational education system in the Netherlands this is an notion is important, because it shows why it is necessary to maintain an institutional framework that makes certification possible. Also, the fact that in the Netherlands firms are involved in the certification of skills may thus be an important feature that influences the position credentials have on the labour market.

So far I have discussed two shortcomings to the varieties of capitalism approach that have been discussed in previous literature. First, the conceptualization of skill, and second, the fact that it overestimates the resemblances between skill formation systems across CMEs. In the next section I will discuss a third shortcoming to the varieties of capitalism literature. Namely that it is too much focussed on the manufacturing sector and therefore neglects issues that are particular to skill provision in the service sector.

High skill equilibrium, diversified quality production and service sector

occupations

A premise to the theory of institutional complementarity are the concepts „high skill equilibrium‟ and „diversified quality production‟. In this section I will clarify the concept and discuss its relevance for this research in light of the above noted limitations to the varieties of capitalism approach. In this section. The purpose of this discussion is to show how skill formation in the manufacturing sector is considerably different from skill formation in the service sector, which will also be discussed here.

A merit of the varieties of capitalism approach is that it shows the close relatedness of skill formation and the political-economic context in which it takes place. For the Netherlands, this means that the non-market modes of coordination provide a solution to the collective action problem of skill formation. In their „The Failure of Training in Britain‟ Finegold and Soskice (1988) conceptualize this context as a „high skill equilibrium‟ which they describe to be the following:

„Education and training are seen to play a crucial role in restoring or maintaining international competitiveness, both on the macro-level by easing the transition of the work force into new industries, and at the micro-level, where firms producing high quality, specialized goods and services require a well-qualified workforce capable of rapid adjustment in the work process and continual product innovation‟ (Finegold & Soskice, 1988:21).

Maintaining a high skill equilibrium could be considered a commonality of CMEs, regardless of their collective skill formation system. By producing certified intermediate skills with a system in which employers are willing to make investments (Culpepper, 1999), a common pool of skilled workers is created of which the same employers are able to hire skilled workers from. Skills, then, are not an

(13)

13 investment of one employer in one (future) employee. Rather, it is a collective good that employers contribute to and benefit from (Streeck, 1991).

To maintain a high skill equilibrium it is important that societal- and state institutions interact with each other (Finegold & Soskice, 1988). Neither State nor market are very successful in providing skills by themselves (Streeck, 1989), rather non-market coordination is necessary in which both State and firms are involved. The core idea of this form of coordination is the employers are encouraged to participate in the provision of skills in an institutional framework that limits the risk of poaching, or other ways by which employers would not reap the entirety of the profit of investing in skills4. This could be considered a prerequisite for collective skill formation. Long-term investments such as training are particular to CMEs. The investments made in non-market relationships (training, in this case) do not have immediate returns (Culpepper, 1999). Strong coordination of skill formation is therefore of importance.

The comparative advantage of the diversified quality production lies in the fact that employers are able to produce large volumes of products although still from high quality. This allows firms to compete on other factors than price alone, the latter being the case in Fordist production lines (Streeck, 1991). In order to achieve this, employees with a sufficient level of technical skills are needed (Culpepper, 1999). Previous literature has primarily focussed on the workplace-based skill regimes that provide such skills, and the German apprenticeship system in particular (Culpepper, 1999; Culpepper & Thelen, 2008; Estevez-Abe et al, 2001). Busemeyer (2009) did make an important contribution by proposing that within CMEs multiple pathways of collective skill formation are possible which all could contribute to diversified quality production. As argued, however, less is known about how occupations that rely less on diversified quality production would fit this theory.

Initially consequences of the rising service sector have been a neglected feature of the varieties of capitalism literature, whose primary focus was firms in the manufacturing sector (Anderson & Hassel, 2008). Recently more authors have been focussing on this segment of the labour market (cf. Culpepper & Thelen, 2008; Anderson & Hassel, 2008; Anderson & Oude Nijhuis, 2012; Busemeyer, 2012). To the service sector in Germany, which could be considered a very clear case of a workplace-based skill regime (Busemeyer, 2009), the workplace-workplace-based system is less attractive as „many elements of the service sector depend less on the broad technological training that is a prerequisite for diversified quality production‟ (Culpepper & Thelen, 2008: 35). With diversified quality production is meant the kind of production in which firm-specific and industry specific skills are combined in order to pursue high-quality production combined with large volumes of production (Estevez-Abe et al., 2001). This form of production, however, is not typical of the service sector, and is often not even

4

(14)

14 require. Culpepper and Thelen have made this argument for Germany. Anderson & Hassel (2008:24) argue that the „higher status and larger role for vocational school in the Netherlands (…) means that general skills occupy an important part of the curriculum in [upper secondary vocational education].When discussing the German case, they argue that the service sector needs „general‟ skills, rather than the „service-specific‟ skills that are currently provided by the education system. However, little clarification of these concepts makes it difficult to understand how these forms of skills are different.

Previous contributions (Culpepper & Thelen, 2008; Busemeyer, 2012) on the expanding service sector imply that workplace-based skill regimes address the needs of a declining number of firms. This trend is supported by an empirical analysis of Berribi-Hoffman et al. (2010) which shows the declining the interest of employers in upper secondary skills and related that the institutional complementarity is much weaker in the IT sector than elsewhere. Lower levels of unionization, less implementation of work councils and a weak collective bargaining (ibid., 2010:100) seem to imply that the institutional complementarity between vocational education and labour market institutions in Germany (Busemeyer, 2009) is at stake in the IT service sector. Although it is not clear whether this is equally the case fo the Netherlands, Anderson and Hassel (2008) argue that the service sector is not in need of the broad technological skills that workplace-based skill regimes could provide. Culpepper and Thelen (2007) are more pragmatic about the matter. They argue that for a successful continuation of the workplace-based regime in Germany the service sector must be convinced to adapt to the apprenticeship system. The authors signal the issue that within the same skill regime different preferences are apparent, but to a lesser extent address the issue of whether this is problematic or not. The difference between these authors is that Culpepper and Thelen (2007) built their argument around the question of sustainability of a particular skill regime whereas Anderson and Hassel (2008) are interested in what skill regime would best suit the service sector. Especially in the Netherlands, this issue is of importance because the Dutch labour market has a large share of workers in service sector occupations (Anderson & Hassel, 2008).

In this section I have discussed the concepts of high skill equilibrium and diversified quality production as they are presented in previous literature, and why this is peculiar in the service sector. In the next section I will elaborate further on why it is important to take this into account when analysing collective skill formation in the service sector.

Service sector (soft) skills

In the discussion above it has been argued that literature on collective skill formation has insufficiently taken into account the particularity of the service sector and the fact that employers in the services are assumed to have other skill needs than employers in the manufacturing sector. In this

(15)

15 section I will discuss what is particular about service sector skills, and specifically soft skills. First, I will address service sector skills, after which I will discuss how these fit into systems of collective skill formation. The purpose of this section is to show how service sector skills differ from skills in the manufacturing sector and why it is important to study these differences.

From the kind of skills that were associated with the service sector by Busemeyer and Trampusch (2012), social- and managerial skills are arguably the most peculiar in respect to whether they would fit the literature of collective skill formation. Especially given the fact that they are less clearly definable as manual skills. These skills, often referred to as „soft skills‟ (cf. Hurrell et al., 2012; Grugulis & Vincent, 2009; Lloyd & Payne. 2009) are „non-technical and not reliant on abstract reasoning, involving interpersonal and intrapersonal abilities to facilitate mastered performance in particular contexts‟ (Hurrell et al., 2012:162) These soft skills are very different from the kinds of skills that are referred to by Finegold & Soskice (1988) and Estevez-Abe et al. (2001), whose primary concern are skills of an asset-specific nature, less portable to other occupations or industries.

The criticism aimed at the conceptualization of soft skills, primarily discusses the fact that it is unclear to what extent soft skills might be considered actual skills (Grugulis & Vincent, 2009). Scholars whom have argued this, point out that social competence must not be confused with analytical or practical skills (Lloyd & Payne, 2009), as they are best considered as „personal attitudes, dispositions or behaviours‟ (Hurrell et al., 2012). Next to this substantive criticism, there has also been expressed the methodological concern that conceptualizing soft skills as skills would blur the meaning of the concept skill (Lloyd & Payne, 2009). A final potential problem with providing soft skills through the education system is that the use of such skills is more dependent upon the employer one will eventually work for. Grugulis & Vincent (2009) argue that from firm to firm it will differ what kind of soft skills are desirable, because each firm has its own „local language‟.

An example of a sector in which soft skills and regulating emotions is of large importance is the hospitality sector. Baum (2002) argues that „there is little about hospitality work and the skills it requires that is unique to the sector. There is, however, a studied argument that it is the context and combinations of those skills that does generate unique attributes (ibid., 2002:345). Summing up the concerns regarding conceptualizing soft skills, it may be argued that it is less clear to what extent soft skills must and can be taught in the education system.

Whether or not there is a necessity to teach such skills in a collective skill formation system is difficult to answer. It would depend on the extent to which the „local language‟ between employers differs. However, a shortcoming to the above discussed literature on „soft skills‟ is that it does not question „local languages‟ as a obstructing factor to learn basic principles of certain soft skills (e.g.

(16)

16 the key difference between greeting customers in a supermarket or luxury boutique will likely be using a different tone of voice).

By opposing soft skills to technical skills, authors such as Hurrell et al.(2012) oppose „real‟ skills to „soft‟ skills. Regardless whether there are or are not differences between the two, in the analysis of skill provision the most important question is probably whether employers find enough commonalities on which a qualification can be developed which would envelop the entire industry. In the Netherlands it can be seen that indeed employers do indeed expect these commonalities to exist, since a great deal of the upper secondary vocational education system consists of students in a service-oriented programme.

Another issue that is important to address here is the extent to which the service sector is indeed a homogeneous group that could be studied as such. In earlier contributions problems regarding „the service sector‟ seem to imply that the sector indeed shares some key characteristics (cf.: Busemeyer, 2012; Culpepper & Thelen, 2008; Andersen & Hassel, 2008). A commonality of service sector occupations is its different needs for skills as opposed to the manufacturing sector: „whereas practical talents and manual work are valued in an industrial setting, social and management skills as well as more theoretical knowledge are more important in service-sector occupations‟ (Busemeyer and Trampusch, 2012:28).

In this section I discussed the particular nature of service sector skills, as opposed to manufacturing skills, specifically the issue of soft skills and whether or not they fit into collective skill formation system. As argued before, in earlier studies the particularity of the service sector has been neglected. In this study I will provide more insight into the differences of the manufacturing and services sector. In the sections above I have shown the different skill regimes within coordinated market economies. This section shows why cannot only expect that different CME skill regimes would differ, but also that within the same skill regime differences may exist between sectors. In the introduction I already touched the point of why the Netherlands is an interesting to case in this respect. In the following sections I will elaborate further on the Dutch upper secondary vocational education system.

Upper secondary vocational education in the Netherlands

In this section I will discuss the context of the Dutch upper secondary vocational education system. In figure 1 I have outlined the Dutch education system and how the upper secondary vocational education system fits into this. Thereafter I will discuss some key concepts of the education system and connect to these to the above discussion of collective skill formation.

(17)

17 Figure 1: structure of the Dutch vocational education system5

The Dutch education system is characterized by high a degree of sorting (cf. Kerckhoff, 2001). This may be seen in the fact that tracking starts at the age of twelve. The transition from primary to secondary education already tracks students in Voorbereidend Middelbaar Beroepsonderwijs (VMBO, secondary vocational education), Hoger Algemeen Vormend Onderwijs (Havo, pre-tertiary vocational education) and Voorbereidend Wetenschappelijk Onderwijs (VWO, pre-academic tertiary education). Within VMBO there are, again, multiple tracks distinguished which lead to different levels of upper secondary vocational education.

Standardization refers to „the degree to which the quality of education meets the same standards nationwide.‟ (Allmendinger, 1989:233). Especially within the Dutch education system it is of importance to distinguish between the two forms of standardization. On one hand standardization of input. Which refers to „the extent to which schools have limited control over the input in education. Examples of such standardization are a lack of autonomy to decide who they hire, what they teach and which books they use‟ (Bol & Van de Werfhorst, 2011:8). On the other hand standardization of output which „describes the extent to which educational performance (the output) is tested against external standards. It tells us the extent to which schools are held accountable for their performance‟ (ibid.,

5

This table is taken from Eurydice:

(18)

18 2011:8). This argument by Bol and van de Werfhorst is based on lower secondary education, in which output is standardized by central examination (Bol & van de Werfhorst, 2011).

Although there is no central examination in upper secondary vocational education, the concept of standardization of output provides a tool to create an understanding of the extent to which it is standardized. Within upper secondary vocational education the output is standardized through the aforementioned qualification documents, as these documents are used to determine what skills students must possess in order to obtain a degree (SBB, 2014a). This could be considered a form of authoritative certification, as the credentials in the upper secondary vocational education system all meet the same requirements, and thus allow employers to signal the skills potential employees possess based on their degree.

Within upper secondary vocational education standardization of output, thus, comes about in another way than in the secondary general education system. It is standardized because the requirements for obtaining a degree are determined on the sector level by intermediary organizations of schools and labour market actors (i.e. employers- and employee representation) (SCP, 2006:34). , and not on school- or firm level. As Busemeyer and Schlicht-Schmälze (2014) have pointed out, public commitment is an important feature of the Dutch upper secondary vocational education systems. This reflects in the degree of standardization. The importance of intermediary organisations would reflect the involvement of employers. For the Netherlands it is of importance to stress that employers may be involved in other ways than in offering apprentices. As will be further elaborated below, employer associations (and trade unions) are formally involved in the determination of curricula.

The input is standardized in a less profound way, as schools themselves are responsible for how they arrange their education (SCP, 2006:34). There are limits, however, to this freedom of schools. Schools must comply with centralized regulation regarding, for example, „teacher/pupil ratio‟s and salaries, credentials when hiring or awarding tenure to staff or hiring priorities‟ (OECD, 2008:61). The OECD, nevertheless, has reported a tendency towards output oriented accountability of the government, rather than regulating the input.

The upper secondary vocational education system (Middelbaar beroepsonderwijs) has the most clear connection between the education system and the labour market, because of the apprenticeships that are offered by firms, and the aforementioned qualification documents. These documents, consisting of one occupation or a cluster of similar occupations, are validated by representatives of schools and social partners on the sector level (SCP, 2008:91).Although the Netherlands has a prevailingly school-based skill regime (Busemeyer, 2009), it contains features of the workplace-school-based regime as well,

(19)

19 such as an apprenticeship system (Anderson & Hassel, 2008; SCP, 2008:57) and could therefore be considered a hybrid regime.

In this section I have discussed how the upper secondary vocational education system relates to other parts of the Dutch education system and the extent the system is standardized. In the following section I will go deeper into the upper secondary vocational education system itself, by discussing the different tracks it offers and why this is important to understand the way skills are provided.

BOL and BBL

An important feature of the Dutch upper secondary vocational education system is that it is a system with two tracks: BOL (=Beroepsopleidende leerweg) and BBL (=Beroepsbegeleidende leerweg). The venue of training for the BOL track is the school, students enrol in a school, follow classes and take part in internships. A part of the BBL track takes place in school as well (about one day a week). In the past, craft schools had a very dominant position in the education system, with as a consequence that the apprenticeship system has never taken a large share in the entire enrolment in upper secondary vocational programs (Anderson & Hassel, 2008). It is known that over the years the structural share of enrolment is stable and that particular differences within the enrolment in BOL and BBL are conjunctural (Van Lieshout & Scholing, 2009); in economic downturn, less BBL places are offered and students that otherwise would have applied for a BBL program then enrol in a school-based BOL program.

An essential difference between BOL and BBL students is that the latter group are in fact employees that work for a firm, may it be in a learning position, and are paid a salary. Whereas BOL students receive (in most cases) a compensation for their internship and have the status of an intern rather than an employee (Van Lieshout & Scholing, 2009). As mentioned in the previous section, the BOL/BBL tracks are where the Dutch skill regime shows to be hybrid. It must be stressed that regardless of the „mixed‟ nature of the skill regime, the BOL and BBL tracks share important similarities as well. Programs are often offered in both tracks, and moreover students are required to obtain the same skills regardless the track they are enrolled in. The development of qualifications and the governance of the programs is structured in a similar way.

Now I have outlined the upper secondary vocational education system, I will further elaborate on the governance of vocational education in the Netherlands and how this relates to previous contributions on systems of collective skill formation.

Structure of the education system

In the upper secondary vocational education systems schools and the labour market both have influence in the process of skill provision, as well as the State. In this section I focus on the way this coordination takes place on the national- and sector level.

(20)

20 On a national level the Ministry of Education, Culture and Science co-operates with the Samenwerking Beroepsonderwijs Bedrijfsleven (Co-operation Vocational Education and Labour Market). The board of this organization consists of representatives of vocational education (MBO raad, AOC raad), employer organizations (VNO-NCW, MKB-Nederland, LTO Nederland) and trade unions (CNV Vakcentrale, FNV Vakcentrale, FNV Jong). Furthermore the board consists of a representative of non-public education and the union for teachers (SBB, 2014d). Every other year a representative of the schools or employers is the chair of the board (SBB, 2013a).

On sector level every sector has a Kenniscentrum voor Beroepsonderwijs en Bedrijfsleven (knowledge centre for vocational education and labour market) in which, again, representatives of schools and employers are seated in the board. The centres „jointly represent more than forty different branches of industry and constitute around 80 percent of economic activity in the Netherlands (Hoffman, 2011)‟. In practice the knowledge centres develop the qualification documents. From an institutional perspective it is especially important that all knowledge centres have a partite board that consists of representatives of the sector whom have to comply with the content of qualification documents. In the Education and Vocational Training Act (Wet Educatie en Beroepsonderwijs) specifically is described what the tasks of a partite board are and what the minimum requirements are for representation by stakeholders:

“To a knowledge centre vocational education and labour market is a partite board vocational education and labour market connected. Half of this board consists of representatives of the labour market and the other half consists of representatives of institutions that offer programs within the sector the knowledge centre operates. Among the representatives of institutions are at least one or more teachers. The partite board has the task to find agreement among the representatives of the labour market and the representatives of institutions on the qualification documents that are submitted to the Minister of Education6.” (WEB, 2014 [1995])

In practice most of these boards are tripartite and consist of education-, employer- and employee representation (for examples see: Aequor, 2014; SH&M, 2014). But even if boards only comply with the minimum requirements, qualification documents are still agreed upon by several stakeholders. The governance of the upper secondary vocational education system is thus very much aimed at the co-operation between education and employers. Employers play a large role in the upper secondary vocational system (Hoffman, 2011). Even when students choose to follow a program in the school-based track – which the majority does – they follow a program validated by the responsible partite board. Not only are employers involved in the development of the programs themselves, students in both tracks follow a part of their education in a firm through internships. Employers that choose to

(21)

21 take on interns must be certified to do so: „the accreditation is intended to certify that the company is a safe learning environment and that trainers agree to participate in a course before working with and supervising young people‟ (Hoffman, 2011: 92). This accreditation is in line with Busemeyer‟s (2012) notion that authoritative certification is a vital part of collective skill systems. This form of authorization results in the fact that the firms where students follow internships or apprenticeships meet predefined demands which reinforces the authorization and standardization of the education. Employer participation is important, but so is the influence of schools. Students that participate in a program on the upper secondary level always attend a school-taught program but it depends on the track they are enrolled in what the extent is to which their education takes place at school. In international perspective the employer participation is average. As argued before the Dutch collective skill formation system has a stronger emphasis on standardization than employer participation (Busemeyer & Schlicht-Schmälze, 2014). However, the paragraphs above clearly show that employers still have a great deal of influence. The fact that the BOL track is larger than the BBL track, however, that this influence is to an extent only indirect. For many students the main venue of training is not the workplace but a vocational school. However, the content of curricula in these vocational schools is still influenced by employers while being part of partite boards of knowledge centres, and being represented in the board of SBB.

Concluding the description of the Dutch upper secondary vocational education system, we may argue that authoritative certification is warranted by the qualification documents. Although a central exam is absent, through these documents it remains clear what the meaning of credentials is, what students have learned and are capable of. The absence of a central exam is in line with the statement that in the Netherlands schools have a relatively large freedom, however the qualification documents warrant this is not at cost of the standardization of the education system. Firm involvement is, even though partially indirect, warranted by the fact that firms are involved on multiple levels in the institutional framework of the upper secondary vocational education system. Moreover, even though less in demand a BBL track still exist in which the venue of training is the firm.

Although this discussion has shown that the Netherlands fits the concept of collective skill formation. It is, however, less clear to what extent the skill regime is applicable to different sectors in a similar fashion. This research will provide more insight into to this issue. In the following sections I will discuss the aim of this research and the way I will conduct it.

Aim of this research

There have been several attempts to analyse the challenges regarding skill formation for service sector occupations. The focus of such research was mainly on the differences in the institutional framework, both in cross-national (Busemeyer, 2009; Culpepper, 2007) and cross-sector perspective (Culpepper & Thelen, 2007; Anderson and Hassel, 2008). Therefore, much is known about the institutional

(22)

22 foundations of skill formation. The consequences these foundations have for the skills that are provided within the education system remain relatively unclear. It has not yet clearly mapped out what the differences are in the skills that are provided, and whether this takes place in a fashion that may be expected from the difference in the needs of manufacturing- and service sector employers. The Netherlands is a case that has been studied very little in this perspective. This might be due to the fact that it less clearly fits the ideal types of the aforementioned skill regimes. Also in a more general sense does it less clearly fit the ideal type of a CME as the often-studied case of Germany does. For this study this particularity of the Dutch case makes it interesting to study, because the traditional importance of the service sector provides a good context to analyse the way service sector skills would fit collective skill formation systems.

It has been argued by Anderson & Hassel (2008) that school-based regimes might better adjust to the needs of post-industrial economies in which the service sector has a very central role. An important feature of these regimes is that they are better able to provide general skills. It however, remains unclear to what extent they do so in practice and especially how this is different from the manufacturing sector. As Anderson and Hassel argue: the Dutch skill regime has characteristics of both a school-based and workplace-based regime. It remains unclear if this mixed regime has different outcomes across sectors, or rather the „mix‟ of school-based and workplace-based learning is equally distributed across these sectors.

This research will provide more insight in differences in content, rather than the often researched issue of transportability by certification. The variety of cross-country differences between CMEs has been addresses in the past (Busemeyer, 2009). This research will contribute to the discussion of skill regimes, by providing more insight into the differences in skill provision that may be present. To do so I will compare qualifications in the service- and manufacturing sector on the upper secondary vocational level. If there are differences in content, they will be reflected in the qualification documents that have been developed by stakeholders of the particular sector, which also included employer- and employee-representation. Besides this I will look at the differences in enrolment in either the school-based BOL or workplace-based BBL track, in order to see whether the hybrid regime is either more workplace or school oriented on the sector level.

Hypotheses and sub questions

In the typology of skill regimes, as presented by Busemeyer (2009), a distinction is made between collective skill formation systems in which skills are provided in a school- or workplace environment. When training takes place in a workplace environment the collective character could be because of a „workplace-based occupational skill system‟ (e.g. Germany) in which skills are provided through an apprenticeship system. The Dutch collective skill formation system has been argued to combine the workplace-based apprenticeship system with a school-based system (Anderson & Hassel, 2008). In

(23)

23 order to understand differences in skill formation between sectors it is thus important to include enrolment, as the BOL and BBL tracks relate to aspects from a different skill regime (namely the venue of training), these tracks may both be observed within the Dutch skill regime. In order to analysis this, I will answer the following question:

What differences can we observe in enrolment in school-based and workplace-based oriented programs across sectors in the Netherlands?

Anderson and Hassel (2008) have argued that the workplace-based track is substantially smaller than the school-based track. It is, however, not yet clear if this is still the case when we look at these sectors apart from each other. One might expect that from the students in a workplace-based track, a larger share is prepared for an occupation in the manufacturing sector. This could be the case because it has been argued by Busemeyer and Trampusch (2012) that employers in the service sector are more interested in skills social-, and managerial skills which are less necessary to learn in a workplace environment than manual skills.

The second sub question is more closely related to the content of skills itself. I will look at the composition of qualifications in terms of skills:

What are the differences in skill content between the services and manufacturing sector in the Netherlands?

The Netherlands, as a mixed case of a school-based and workplace-based regime, could be considered a country in which both regimes are combined. It is, however, not yet clear in what way these hybridity appears. On the one hand the fact that the system is assumed to be very school-based, assumes that it also oriented more towards general skills such as social skills, general cognitive skills and managerial skills. On the other hand, the fact the public commitment is high (Busemeyer & Schlicht-Schmälze, 2014) would assume that there would be an overall orientation towards such skills. Moreover, following Busemeyer and Trampusch (2012), it might be expected that even though the Dutch system is prevailingly oriented towards general skills, qualifications in the manufacturing sector will contain more manual skills, as this is argued to address the needs of such firms. It is not yet clear as to what extent the different needs of employers in the manufacturing- and service sector also results into differences in skill provision in these sectors.

Data Collection and Methods

In order to answer this research question I have performed a content analysis in which I compared qualifications based on the kind of skills the entail. I look at 31 qualifications, of which 15 in manufacturing and 16 in services. To measure the manufacturing and services sector I will use the

(24)

24 ISCO-08 classification (ILO, 2008). I look at Dutch qualifications that prepare for occupations within the „Clerical Support Workers‟, „Services and Sales Workers‟ „Craft and Related Trades Workers‟ and „Plant and Machine Operators and Assemblers‟.

The information regarding enrolment is accessible through Dienst Uitvoering Onderwijs (DUO)7, the executive organization that is responsible for the registration of students and the coordination of scholarship payments and tuition fees. This information of how many students are enrolled in a qualification and whether they are enrolled as a BOL or BBL student is registered and published per qualification. Qualification documents themselves are published online by SBB.

The chosen cases were selected based on their popularity. Using the most recent International Standardized Classification of Occupations (ISCO-08) of the International Labor Organisation (ILO) I was able to determine whether qualification suited the sectors I want to analyse. By comparing the ISCO-08 database with the enrolment data I was able to choose the relevant cases. The reason to choose the most popular cases is that it makes the findings more relevant, because it applies to a large student population. Although it is not necessary that popular qualifications provide skills that are high in demand, a large share of the students at the upper secondary level will obtain the skills that are analysed here.

The ISCO-08 database consist of 10 major groups, which again are divided into subgroups and units. From these major groups, most relevant for this research are: Craft and related workers, Plant and Machine Operators and related workers, Clerical support and related workers, services and sales workers. These four are the most relevant because they consist of consist of occupations that are likely to exist on the upper secondary level and relate to the manufacturing- and service sector. The managerial and professional groups consist primarily of occupations that are more likely to exist on the tertiary level.

A content analysis is the best method to answer my research question because it allows drawing conclusions from the documents that have been developed by the representation of stakeholders. It would therefore give insight into the actual requirements that are the result of the co-operation between employer-, employee- and school representation. This content analysis is complemented with data regarding the tracks students are enrolled in. The measurement of portability in the studies discussed above is often done indirectly. Content, however, is measureable in a more direct way through a content analysis.

7

http://www.duo.nl/organisatie/open_onderwijsdata/databestanden/mbo_/Onderwijsdeelnemers/ Onderwijsdeelnemers/mbo_deelname8.asp; the data I used is a version of this dataset already aggregated to all enrolled students per qualifications (filtering out gender, province and institution), received from SBB.

(25)

25 I selected the cases on basis of popularity: out of every occupational group I took eight qualifications in which most students are enrolled. To analyse the qualifications I categorized the skills in a classification scheme. Using this scheme I am able to visualize the composition of the skills. The dataset with the analysed qualifications will look the following: Qualification X will consist of y% manuals skills, z% social skills, etcetera. This classification will allow me to proxy the content of the skills. The database that I have made from this consists of information of the skills broken down into categories and expressed in percentages of the total number of skills that are included in a qualification document.

I counted the amount of skills in each document, and to what category these skills belonged. If a skill was mentioned twice, I also counted it twice. I did this under the assumption that skills that are mentioned more often are also more important. I will to this for all the qualifications and will make comparison between the qualifications in the manufacturing sector on the one hand, and the qualifications in the service sector on the other. Doing this, I can show whether there are indeed differences in content between these sectors. This part of the analysis is done with ATLAS.ti

The explanation for these difference will be that I can show whether the Dutch skill regime, on the basis of the content and enrolment, has different preferences in skills between both sectors, or we may conclude that the cross-sector differences are not so large.

In order to analyse enrolment I will look at the data that is published by DUO. Here I also analyse the differences between the manufacturing and service sector, as it allows me to see whether the education mostly takes place in a school- or firm environment. I have connected the enrolment data to the data of the content analysis. By doing so I can analyse whether the kind of skills in a qualification document are related to the enrolment in one of the two tracks. The argument in the result section will be built around the connection between content and enrolment data.

Operationalization

In this section I discuss the concepts that I use and how I use them in this research. These concepts being: qualification documents, skills, the manufacturing- and service sector, the BOL/BBL-tracks and I discuss how I correct the data for the relative size of the qualifications in terms of enrolment. Qualifications

Qualifications are descriptions of what a labour market entrant is supposed to know and able to do. It is not the same as a diploma. Schools are able to combine several qualifications in one credential (SBB, 2014b). The requirements for the separate qualifications will still remain the same. In my research I will focus only on qualifications, since this is the unit that is agreed on by stakeholders of education and labour market (ibid., 2014b).

(26)

26 Qualifications are described in qualification documents. These documents contain the requirements for one or more qualification. The mentioned requirements are descriptions of what a student must know and be able to do in order to obtain a degree. These requirements are described as what an employee of a certain occupation does when he starts his job without any prior work experience (SBB, 2014a). A qualification document is subdivided into tasks. These tasks are a combination of several groups of actions. For example, within ICT there is a task: „Installing hard- and software‟. This task is, again, subdivided into processes: „Assembling systems‟, „Disassembling systems‟, „Installing and configuring systems and (standard) applications‟ and „realising the cable infrastructure‟ (SBB, 2014b). These processes describe concrete actions that have to be taken in order to perform these processes. Next to these processes a task also consists of a description of the complexity of the actions and the extent to which the employee is responsible for his/her own work or whether a supervisor is ought to keep an eye out on him/her. Lastly, there is a description of skills and knowledge one must possess in order to be able to perform the task.

I analyse the content of the „skills and knowledge‟ section in each qualification document. This is a comparable element of the documents as it is all described in relatively similar terms across all qualifications. Moreover, there is a clear link between the education system and the „skills and knowledge‟ section because these are elements that a student needs to know in order to be able to perform his or her occupation. By collecting data from these documents we know what skills someone is supposed to possess when leaving the education system. However, this data cannot account for whether these skills are provided within the education system.

Qualifications thus serve as a proxy for the skills that have been taught in the upper secondary vocational education system. Important to note is that, especially in the realm of soft skills, it might be the case that not all skills that students obtain are listed, some of them are unintentionally learned on the shop floor or in the classroom (e.g. planning, problem-solving). Therefore, what can be measured are the skills that are determined students should obtain, not the skills that that students actually will obtain8.

The analysed qualifications are based on the 2014-2015 cohort, which means that these are currently the most recently approved documents. For the enrolment data I have used the data of students that were enrolled in the year 2012-2013. Because the qualifications document might undergo minor changes every year this means that some students are enrolled in programme that is based on a qualification that was validated two or three years earlier. I do not expect this to influence the data,

8_{This difference will not be very strong. In fact, it is likely that students will learn more rather than less, because}

the Inspectorate assesses the qualities of schools among others by looking at the extent to which qualification documents are implemented appropriately.

(27)

27 changes between the 2013-2014 and 2014-2015 cohort are very small and moreover did not affect qualifications that are included in the sample (SBB, 2014e)

Skills

Skills are an important feature of this research. It has been argued that a dichotomous general-specific distinction is not sufficient (Streeck, 2012), and rather a two-dimensional framework should be used in which specificity could either be an economic dimension (portable versus non-portable) and a substantive dimension (broad versus narrow). This is an important assumption of how to approach skills that I will have to take into account when discussing the Dutch skill regime, but as mentioned before the focus of this research will lay on the content of skills.

In order to measure skills I will use a number of categories. The categorizations I use are based on the skills that Busemeyer has mentioned as differences in skill needs between manufacturing and service sector employers.

Manual skill Non-manual skills Knowledge

- Manual skills - General cognitive skills - Theoretical knowledge

- Social Skills - Safety regulations

- Managing skills - Competence

- Occupation-specific skills

Table 1: categories for the content analysis

The below mentioned categories relate to each other as outlined in table 1. The categories that are used in the content analysis are the cursive labels. The labels, in their turn, belong to a larger category. This way, the described skills could be considered either manual, non-manual or knowledge. The subdivision is of particular importance in the non-manual skills section, because these categories entail a wider variety of non-manual skills.

- Manual skills: Under this label I will categorize those skills that require working with machinery and tools. The importance of these skills is at the core of the theory on collective skill formation and diversified quality production. In the theory section I have discussed the emphasis on manual skills in previous research.

- IT skills: I have taken working with personal computers as a separate skill. Under this label the skills are categorized that require working with computers, the use of software and understanding of computers.

- Social skills: Grugulis et al. (2004) argued that particular about service sector skills is that it entails many „behavioural‟ elements (e.g. being polite). Busemeyer and Trampusch (2012)

Skills, enrolment and hybridity in the Dutch skill regime : a cross-sector analysis of qualifications in the upper secondary vocational education system