Dual Vocational Education and Training Systems’ Governance Model and Policy Transfer: The Role of the European Union in Its Diffusion

Abstract

Some southern member states have undertaken the reform of their vocational education and training (VET) systems so as to promote dual apprenticeship, such as that which is promoted in Germany. During this process, the European Union (EU) has exercised an extensive influence. This paper advances the analysis of the role exercised by the EU in this stage of cross-national attraction by analysing the model of VET governance promoted by European institutions. The methodology consists of a reflexive thematic analysis of the EU VET texts. A total of 35 texts from the EU institutions was analysed by using NVivo. This produced three themes under the overarching theme of fostering a more pluralistic governance system: promoting cooperation among stakeholders in the design, management and financing of the system, enhancing social dialogue and strengthening responsiveness to change in the world of work. This analysis concludes that the EU, as an agent of the policy transfer process, encourages the implementation of a cooperative governance model of VET systems, such as that present in the dual VET systems of certain member states with collective skill formation governance regimens such as Germany, Austria and Denmark.

1. Introduction

Some member states (MS) of the European Union (EU) have undertaken reforms of their vocational education and training (VET) systems during the last decades. This process began with the signing of the Copenhagen Declaration of 2002, which gave rise to the Copenhagen process and to wide-ranging documentary production by EU institutions over the years, including their reforming proposals in the form of conclusions and recommendations for the member states (MS). During this reform cycle, the European level played a crucial role by influencing Spain and other southern countries. As stated by Zaunstöck et al. (2021):

The Copenhagen Declaration, the process leading to it and its aftermath were ultimately relevant for Spanish stakeholders in terms of discussing the intentions and directions of Spanish VET. Spain has not only adopted the wording behind the Copenhagen Declaration but also aligned itself in the same direction: the foundation of its current VET system can be clearly correlated with the Copenhagen principles and pillars.

(Zaunstöck et al. 2021, p. 15)

Policy makers, in order to meet the challenges of the national VET systems, “try not to reinvent the wheel, and therefore look for models and best practices that have worked elsewhere” (Maurer and Gonon 2014, p. 17). Throughout this reform cycle inaugurated by the Copenhagen Declaration, the German system of dual VET, in conjunction with other European dual VET systems, has been the most relevant model of good practice in the public imaginations of the southern reforming countries (Martín-Artiles et al. 2019). This implies a process of policy transfer where the EU is involved as a dynamic agent. In order to shed light on this stage of cross-national attraction (Phillips and Ochs 2004) in dual VET policy transfer, this research analyses the EU discourse on the promotion of dual VET throughout this policy transfer process. More precisely, this paper advances the understanding of the model of VET governance which has been promoted by the European Union.

This research is contextualised within a research topic, VET governance, which has experienced relatively recent growth. This interest in VET governance stems from two fundamental aspects. Firstly, its “significant impact on the functioning of a VET system and also on its way to reform itself” (Barabasch 2010, p. 225). Secondly, the large number of actors concerned in VET decision making, implementation and management (central and regional public administrations, training centres, companies, social partners, etc.) and the relevance of their coordination for the proper functioning of the system (Oliver 2010), especially in dual systems. On the one hand, traditionally, research on VET governance has been particularly extensive in countries with dual VET systems, especially in the German-speaking world. Its focus has been placed on the examination of the tradition, organisation and function of collective skill formation (Bürgi and Gonon 2021; Busemeyer and Trampusch 2012; Greinert 1998; Maurer 2022). Within the same trend, some authors have focused on the comparative analysis of the different national models of dual VET with the aim of detecting the strengths and weaknesses of each governance model (Emmenegger and Seitzl 2020; Rauner and Wittig 2010; Trampusch 2010). On the other hand, due to the key influence of governance on the success of VET systems, a significant part of VET governance as a research topic has consisted of the analysis of possibilities relating to policy borrowing from the dual VET model of best practices. From this perspective, we identify two streams of researchers: that focusing on the analysis of the importation of the dual system by the reforming countries (Martín-Artiles et al. 2019; Maurer and Gonon 2014; Valiente and Scandurra 2017; Vogelsang et al. 2022) and that focusing on the export potential of the models of best practice (Davoine and Deitmer 2020; Lassnigg 2015; Salameh et al. 2012).

This research is close to the VET governance policy borrowing perspective, as it analyses the cross-national attraction phase of the dual VET policy transfer process by focusing on the agency role exercised by the European Union. Furthermore, the aim of this paper is based on previous empirical evidence which showed the prominence of the dual model of vocational education and training as a model of good practice in the minds of the local actors of the reforming countries (Echeverría 2016; Martín-Artiles et al. 2019). At the same time, the theoretical framework of this paper is based on the main findings of authors who focused on the analysis of the organisation and function of the dual VET models. These findings are key elements for the discussion of the results. The following points elaborate on those theoretical issues that are fundamental for the discussion of the results.

Accordingly, the questions which guided this research were: What model of VET system governance is promoted by the European Union? What examples of good practices in respect to VET governance are sponsored by the European Union?

1.1. Policy Transfer and Europeanisation

Part of the theoretical framework of this research is based on educational policy transfer (Phillips and Ochs 2004; Steiner-Khamsi and Waldow 2012) and Europeanisation (Lawn and Grek 2012) theories. Policy transfer terminology and focus are wide-ranging, but the whole area is “concerned with the process by which knowledge about policies, administrative arrangements, institutions and ideas in one political system (past or present) is used in the development of policies, administrative arrangements, institutions and ideas in another political system” (Dolowitz and Marsh 2000, p. 5). Evans (2009) states a complex and comprehensive approach to policy transfer analysis:

Policy transfer analysis is a theory of policy development that seeks to make sense of a process or set of processes in which knowledge about institutions, policies or delivery systems at one sector or level of governance is used in the development of institutions, policies or delivery systems at another sector or level of governance.

(pp. 244–45)

Cowen (2006) argues that the basic challenge in the study of any policy transfer process is the identification and analysis of transfer, translation and transformation. Transfer, the stage on which this research focuses, is defined by Cowen as “the movement of an educational idea or practice in supra-national or trans-national or inter-national space: the ‘space-gate’ moment, with its politics of attraction and so on” (Cowen 2006, p. 566).

Li and Pilz (2021) concluded from their literature review of the discourses about VET policy transfer that “VET transfer research is also an important topic within Comparative education science—in particular international educational transfer and its possible forms” (6). The attention of this research area has been traditionally drawn to the import of ideas, strategies, measures and concepts between countries in order to achieve an effective solution for local educational problems. Steiner-Khamsi (2014) differentiates, within comparative education science, between applied normative researchers and analytical researchers, although she acknowledges that it is difficult to find authors who stick exclusively to one field. The former set of researchers actively advocate policy borrowing, using comparison as a means of identifying and transferring good practice from the best-performing education systems. In contrast, the analytical comparative researchers are more interested in understanding when, why and how policy borrowing occurs, while examining the impact of such imports on existing policies and power constellations. They subscribe to foundational research and opt for comparison as a methodological tool to better theorise the policy process. They seek to provide answers to questions that reveal imbalances of diffusion power, the conditions that fuel the dissemination of a certain practice and the beneficiaries and losers of the transfer process (Steiner-Khamsi 2014). This research belongs to this second category; it attempts to abolish the dichotomous trap of deciding whether transfer is good or bad and seeks to discover what happens during transfer in terms of translation and transformation of the educational phenomenon in its new place (Larsen 2010).

Over the last decades, different analytical frameworks have been proposed to facilitate the reading of the policy transfer process. Phillips and Ochs (2004) suggested a circular model (Figure 1) based on a detailed analysis of the four stages of the policy borrowing process, understood as the “conscious adoption in one context of the policy observed in another” (Chicago Manual 17th edition, 774).

The first stage of dual VET policy transfer, the cross-national attraction where this research puts its focus, has two components: impulses and externalising potential. Impulses are those factors that motivate or drive the transfer and create the conditions that make it possible to look for examples or models of success elsewhere (Phillips 2004). These factors may include political change, the systemic collapse of the present system, negative external evaluations, the creation of new international alliances or extreme events such as war or natural disasters. The potential for externalisation, on the other hand, constitutes those elements of the foreign system that could theoretically be borrowed by the importing country. These elements, which make up a total of six, are known as focal points of attraction (Ochs 2002, p. 29) and are based on the structural typology of transnational attraction (Ochs and Phillips 2002): guiding philosophy, ambitions or goals, strategies (financing, further training), enabling structures, processes and techniques.

Additionally, educational transfer occurs through a variety of more or less direct pathways or mechanisms. Indeed, there is no diffusion or reception, lending or borrowing without agency (Steiner-Khamsi 2014). The question of agency implicitly suggests that we must take into account multiple actors and their different roles in transfer processes and refuse previous approaches where the state was the only actor (Rappleye 2012). As Li and Pilz (2017) concluded from their literature review of the discourses about policy transfer, “when investigating transfer activities, scholars working in comparative political science often focus on the relationship between the actors involved and the impulses that result from this relationship” (4). Transnational organisations, such as the EU and its respective agencies, have become a hub for the dissemination of ideas, programmes and institutions (Jakobi 2012). These organisations influence national policy makers directly, through their policies and funding, and indirectly, through the information and policies disseminated in their conferences and reports (Dolowitz and Marsh 2000).

Meanwhile, Europeanisation has become a key concept in understanding how the EU has transformed the way member states make policy. It is used by various authors to describe the impact of the EU on the political life, legislative policies and governance of member states (Ante 2016). EU actions are clearly aimed at promoting greater cooperation between member states, but also at strengthening the community’s political impact in the field of education and, hence, vocational education and training. To this end, two lines of action have been defined: the dissemination of best practices and the promotion of joint reflection (Novoa 2002). As an area where hierarchical intervention by the EU is not possible, a horizontal mechanism is used through soft frameworks where the EU serves only as a stage. In these policy areas, European policy is based on cooperation and agreement between member states (Zaunstöck et al. 2021). The EU works with minimum or non-binding directives or regulations that provide solutions to a specific problem and which are then introduced into the national debate with the potential to change the perception of the problems of national actors and, thus, trigger learning processes and the legitimisation of alternatives (Zaunstöck et al. 2021). As a consequence, the EU’s role influences and could even activate the agency role of local actors in the policy transfer process.

In order to converge European VET policies, the EU uses what Cort (2009) calls the EU Governance Triangle. This governance triangle consists of three mechanisms covering different means of regulating the behaviour of MS: the community method, the programme method and the open method of coordination (OMC). This research analyses the main texts resulting from the EU Governance Triangle in order to interpret the model of VET governance promoted by the EU as an agent of the policy transfer process (Steiner-Khamsi 2014).

1.2. VET Governance

Governance is an abstract and diffuse concept. Barabasch (2010) sheds light on that construct by combining Benz (2004) and Benz et al. (2007) statements:

It (governance) represents all forms and mechanisms concerned with the coordination between various social actors, whose actions are interdependent and affect or support each other (Benz et al. 2007). Governance refers not only to activities and operations of governing, steering or coordinating, but also to the nature and ways in which these tasks are performed. That also includes procedural, structural, functional and instrumental aspects of governing.

(Barabasch 2010, p. 225)

According to Sanz (2017), the study of governance focuses on the mechanisms that, in complex social systems, serve to coordinate the activities of various institutions and actors, both public and private and located at different levels (transnational, national or regional), which have interdependent relationships (Mayntz 2004).

This paper is contextualised within the research on the governance of VET systems. As highlighted previously, the study of VET governance has experienced recent growth, justified by the fact that the proper functioning of dual VET systems relies on the correct coordination and integration of the different actors involved in the training and decision-making processes (Oliver 2010). Within this area of research, two fundamental currents can be distinguished for the configuration of the theoretical background of this paper. One of these currents is constituted by the authors who have focused their research on the examination of the tradition, organisation and function of VET (Bürgi and Gonon 2021; Busemeyer and Trampusch 2012; Greinert 1998; Maurer 2022). Greinert (1998) based his theories on the Weber categories of social regulation and proposed, in terms of the allocation of roles and responsibilities, as well as the underlying logic of agency, a classification of VET governance models consisting of three categories: state-driven, market-controlled and corporatist models. The state-driven model is based on the predominant role of the state. The system is concisely regulated by the administration. It is based on the logic of school operations. Companies do not have an institutional role and are limited to providing internship opportunities (e.g., the French system). On the other hand, in the market-controlled model, the system direction is in the hands of employers and the labour market. Training is geared to the needs of enterprises and takes place on the job and in private institutions (e.g., the Anglo-Saxon system). The corporatist model of VET governance is presented within the mixed systems of cooperative governance in which the regulation of vocational education and training takes place in a pluralistic network consisting of state bodies, companies or business associations, as well as trade unions or professional associations (Rauner and Wittig 2010). The functioning of these governance networks is based on social dialogue (Hippach-Schneider and Huismann 2016) as a tool for reaching agreements between the different actors of the networks. Among the main characteristics of the corporatist model, the extensive influence exercised by the training companies and chambers stands out. This corporatist model gives a fundamental role to the “the corporate bodies or associations that represent the business community at the local or regional level” (Rauner and Wittig 2010, p. 3). In the same vein, Busemeyer and Trampusch (2012) define four forms of skill formation governance (segmentalist, liberal, statist and collective). The segmentalist form corresponds to the existence of a strongly segmented labour market, and training companies must implement their training activities individually. In the liberal one, we find a similar segmentation, accompanied by a low level of business and public commitment to VET where large companies monopolise training (Martín-Artiles et al. 2019). The statist model (France, Sweden) is based on strong public support for in-company training. Collective skill formation is a form of governance in which both public authorities and social partners are strongly committed to VET. These collective skill formation models (Germany, Austria, Switzerland, the Netherlands and Denmark) are based on a strong partnership between the social partners within the framework of a coordinated economy at the macroeconomic level (Martín-Artiles et al. 2019). The responsiveness of this model to changes in the world of work is high (Busemeyer and Trampusch 2012) thanks to the articulation of systematic collaboration across the different actors of the governance networks and the extensive use of different strategies for detecting these needs (Hippach-Schneider and Huismann 2016). This model of VET governance is a reality in countries such as Austria, Switzerland, Germany and Denmark, although they present big differences in the way it is articulated. Comparative research about these differences has focused on aspects such as the general orientation of the system (Emmenegger et al. 2020), the role of the different stakeholders (Bürgi and Gonon 2021; Emmenegger and Seitzl 2020; Trampusch 2010), the coordination between the different agents with their respective internal logic and input or output orientation (Rauner and Wittig 2010).

Furthermore, a fundamental current for the theoretical contextualisation of this paper inside VET governance research focuses on the analysis of the policy borrowing possibilities of the dual VET models of best practices. Certain authors have analysed the importation of the dual system by the reforming countries. For example, Valiente and Scandurra (2017) and Maurer and Gonon (2014) analysed the challenges facing the transfer of the dual vocational training system, including its governance model, to countries whose tradition is based on other types of model more closely linked to state or liberal control. Vogelsang et al. (2022) focused their research on the analysis of the role played by the local actors in the transfer and import of dual training approaches. From a similar perspective, Martín-Artiles et al. (2019) analysed the discourse of institutions and social actors from importing contexts, including aspects such as governance and social dialogue. Other authors focused their research on the export potential of the models of best practices. For example, Salameh et al. (2012) focused on the driving and inhibiting factors in VET exports from the perspective of German providers. Dvoine and Deitmer (2020) discussed “the characteristics and specifics of the German apprenticeship model and presents the examples that illustrate the innovative practices of German companies in China, the United States and Mexico” (1), and Lassnigg (2015) reflected on the exporting myth of employment transition of dual VET systems.

2. Methods

The methodology of this research consisted of a reflexive thematic analysis (TA) (Braun and Clarke 2021) using NVivo as the qualitative data analysis computer software. Braun and Clarke (2021) define TA as “a method for developing, analysing and interpreting patterns across a qualitative dataset, which involves systematic processes of data coding to develop themes” (4). This TA followed the principles of the fully qualitative or Big Q approach, which “involves both the use of qualitative tools and techniques and qualitative values, norms and assumptions” (Braun and Clarke 2021, p. 287). Furthermore, the term reflexive assumes the active role of the researcher in the coding process and the development of themes. Reflexive also means accepting the author’s subjectivity and the relevance of the researcher’s reflection about his assumptions and practices and how they shape and delimit their data analysis.

The corpus of this TA was created by exploring the European Union repositories, with a special focus on the documentation related to the Copenhagen process and the enhancement of the member states’ VET systems. Specifically, the document search process was carried out in the EUR-Lex repository, which provides official and comprehensive access to EU legal documents. The criteria for the search were limited to documents in English published between the signing of the Copenhagen Declaration (November 2002) and June 2022 containing ‘vocational education’in their title or text. Results were limited as well to legal acts. After that, a manual selection of texts was developed, excluding texts that do not deal with the configuration or reform of the VET systems. This manual selection process excluded legal documents associated exclusively with aspects that were far from the focus of the research, such as convergence in professional qualifications, migration or labour regulations. The same process was replicated using the term ‘apprenticeship*’. The selection was complemented by snowballing from the most relevant references found in a first reading of the texts selected in the previous two searches. The documents resulting from this snowballing process came from two sources: the Publication Office of the European Union and the Document Library of the European Education Area.

After a thorough reading of the selected documents, a total of 35 documents was selected for this TA because of their close relation to the thematic issue of the reform of VET systems in the MS and aspects related to VET governance. As a result, the final corpus of the research consisted of 35 texts (Appendix A-

Table A1. Corpus of the research.

Author	Year	Nº	Title
Council of the European Union	2009	1	Council conclusions of 12 May 2009 on a strategic framework for European cooperation in education and training
Council of the European Union	2013	2	European Alliance for Apprenticeship Council Declaration
Council of the European Union	2014	3	Council conclusions of 20 May 2014 on quality assurance supporting education and training
Council of the European Union	2018	4	Council Decision (EU) of 16 July 2018 on guidelines for the employment policies of the Member States
Council of the European Union	2018	5	Council Recommendation of 15 March 2018 on a European Framework for Quality and Effective Apprenticeships
Council of the European Union	2020	6	Council Recommendation of 24 November 2020 on vocational education and training (VET) for sustainable competitiveness, social fairness and resilience
Council of the European Union	2021	7	Council Resolution on a strategic framework for European cooperation in education and training towards the European Education Area and beyond (2021–2030)
Council of the European Union and Representatives of the Governments of the Member States	2006	8	Conclusions of the Council and the Representatives of the Governments of the Member States, meeting within the Council, on the future priorities for enhanced European cooperation on Vocational Education and Training (VET)
Council of the European Union and Representatives of the Governments of the Member States	2009	9	Conclusions of the Council and of the Representatives of the Governments of the Member States, meeting within the Council, on the future priorities for enhanced European cooperation in vocational education and training (VET)
Council of the European Union and Representatives of the Governments of the Member States	2010	10	Conclusions of the Council and of the Representatives of the Governments of the Member States, meeting within the Council, on the priorities for enhanced European cooperation in vocational education and training for the period 2011–2020
Directorate-General for Employment, Social Affairs and Inclusion	2012	11	Apprenticeship supply in the member states of the European Union
Directorate-General for Employment, Social Affairs and Inclusion	2013	12	Apprenticeship and Traineeship Schemes in EU27: Key Success Factors. A Guidebook for Policy Planners and Practitioners
Directorate-General for Employment, Social Affairs and Inclusion	2020	13	The renewed European Alliance for Apprenticeships Action plan 2020–2021
Directorate-General for Employment, Social Affairs and Inclusion	2020	14	The renewed European Alliance for Apprenticeships: Key Objectives
Directorate-General for Employment, Social Affairs and Inclusion and ET 2020 Working Group	2016	15	High-performance apprenticeships & work-based learning: 20 guiding principles
Directorate-General for Internal Policies of the Union and European Parliament	2014	16	Dual education: A bridge over troubled waters?
European Commission	2010	17	Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions. A new impetus for European cooperation in Vocational Education and Training to support Europe 2020 strategy
European Commission	2012	18	Communication from the Commission to the European Parliament, the European Council, The Council, the European Economic and Social Committee and the Committee of the Regions. Rethinking Education: Investing in skills for better socio-economic outcomes
European Commission	2016	19	Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions. A new skills Agenda for Europe. Working together to strengthen human capital, employability and competitiveness
European Commission	2020	20	Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions. European Skills Agenda for sustainable competitiveness, social fairness and resilience
European Commission	2020	21	Communication from the Commission to the European Parliament, the European Council, the Council, the European Central Bank, the European Economic and Social Committee, the Committee of the Regions and the European Investment Bank. Annual Sustainable Growth Strategy 2021
European Commission	2021	22	Commission Recommendation (EU) of 4 March 2021 on an effective active support to employment following the COVID-19 crisis (EASE)
European Commission, European Social Partners and Lithuanian Presidency of the Council of the EU	2013	23	Declaration of the European Social Partners, the European Commission and the Lithuanian Presidency of the Council of the European Union. European Alliance for Apprenticeships
European Ministers for Vocational Education and Training, European Social Partners and European Commission	2002	24	Declaration of the European Ministers of Vocational Education and Training, and the European Commission, Convened in Copenhagen on 29 and 30 November 2002, on Enhanced European Cooperation in Vocational Education and Training. ‘The Copenhagen Declaration’
European Ministers for Vocational Education and Training, European Social Partners and European Com-mission	2004	25	Maastricht Communiqué on the Future Priorities of Enhanced European Cooperation in Vocational Education and Training (VET)
European Ministers for Vocational Education and Training, European Social Partners and European Commission	2006	26	The Helsinki Communiqué on Enhanced European Cooperation in Vocational Education and Training
European Ministers for Vocational Education and Training, European Social Partners and European Commission	2010	27	The Bruges Communiqué on enhanced European Cooperation in Vocational Education and Training for the period 2011–2020
European Ministers for Vocational Education and Training, European Social Partners and European Commission	2015	28	Riga Conclusions on a new set of medium-term deliverables in the field of vet for the period 2015–2020, as a result of the review of short-term deliverables defined in the 2010 Bruges Communiqué
European Ministers for Vocational Education and Training, European Social Partners and European Commission	2020	29	Osnabrück Declaration on vocational education and training as an enabler of recovery and just transitions to digital and green economies
European Parliament	2010	30	European Parliament resolution of 6 July 2010 on promoting youth access to the labour market, strengthening trainee, internship and apprenticeship status
European Parliament	2011	31	European Parliament resolution of 8 June 2011 on European cooperation in vocational education and training to support the Europe 2020 strategy
European Parliament	2013	32	European Parliament resolution of 22 October 2013 on Rethinking Education
European Parliament	2014	33	European Parliament resolution of 17 July 2014 on Youth Employment
European Parliament	2016	34	European Parliament resolution of 19 January 2016 on skills policies for fighting youth unemployment
European Parliament	2017	35	European Parliament resolution of 14 September 2017 on a new skills agenda for Europe

) whose authorship corresponded solely, and in some cases jointly, to the European Commission, the European Ministers for Vocational Education and Training, the Council of the European Union, the Representatives of the Governments of the Member States, the European Parliament, the Directorate-General for Employment, Social Affairs and Inclusion and the ET 2020 Working Group

Furthermore, the methodological process of the reflexive TA involved six phases, as recommended in Braun and Clarke (2021). First, the dataset familiarisation stage, where an initial approach to the European documentation collected was carried out, detecting the first focuses of interest and pouring the corpus of texts into the NVivo software. Secondly, the data coding phase was developed from a mixed deductive and inductive approach. Similarly, the analysis captured meanings at both the semantic and latent levels. Thirdly, the initial generation of themes was developed by looking for patterns of meaning and trying to find the ones that could answer the research questions. After that, the theme development stage was carried out. During this phase, the creation of themes was reviewed to finally name and define them. The final objective of this stage was relating each one of them to the global history of the analysis. Finally, this paper was written.

In order to increase the comprehensibility of the results of this paper, an identifying number was assigned to the different analysed texts, which can be checked in the Appendix A section. The assigned numbers of the texts whose fragments configure each theme and pattern of meaning appear in brackets next to each of them. Thus, it is recorded which documents contributed with their content to the elaboration of each theme or pattern of meaning beyond the proposed quotations.

3. Results

This analysis produced three themes under the overarching theme or macro-theme of fostering a more pluralistic governance system, giving rise to the following distribution:

• Promoting cooperation among stakeholders in the design, management and financing of the system;
• Enhancing social dialogue;
• Strengthening responsiveness to change in the world of work.

The improvement of the governance of the system and the inclusion of new actors are recurrent themes in the European texts. Generally, governance issues are implicit when addressing aspects such as decision making, design or management of the vocational training system. However, certain European texts (8, 26) refer explicitly to this term:

Q1—In improving the attractiveness and quality of VET, more emphasis should be placed on good governance of VET systems and providers in delivering the VET agenda;

(Council of the European Union & Representatives of the Governments of the Member States 2006, p. 1)

Q2—(MS should) Support sustainable partnerships for the governance of vocational education and training, in accordance with national context and, where relevant, through public-private partnerships.

(Council of the European Union 2020, p. 7)

These concepts of good governance, sustainable partnerships and public–private partnerships are developed across a wide range of texts, the analysis of which led to the four key themes which we explore below.

3.1. Promoting Cooperation between Stakeholders in the Design, Management and Financing of the System

The issue of the inclusion of different stakeholders as active participants of the decision-making process has been present in EU texts (3, 5, 6, 8, 9, 10, 11, 17, 19, 20, 25, 26, 28, 29, 31, 32, 35) since the beginning of the Copenhagen process in 2002. However, at the origin of the process, the responsibility of these actors was limited to aspects related to competences and qualifications:

Q3—The social partners play an indispensable role in the development, validation and recognition of vocational competences and qualifications at all levels and are partners in the promotion of an enhanced cooperation in this area.

(European Ministers for Vocational Education and Training et al. 2002, p. 3)

Already, in the Maastricht Communiqué of 2004, we can observe a qualitative leap with respect to the calls for integrating stakeholders in the governance system, advocating for a larger allocation of responsibilities among all the actors in the field of VET. Two excerpts help to illustrate the fundamentals of this leap:

Q4—Priority should be given to the increased relevance and quality of VET through the systematic involvement of all key partners in developments at national, regional and local level, particularly regarding quality assurance;

(European Ministers for Vocational Education and Training et al. 2004, p. 3)

Q5—All actors in the field of VET—providers, employers, trade unions, branch organisations, chambers of commerce, industry or crafts, employment services, regional bodies and networks, etc.—are invited to take their responsibilities and to contribute to making effective the implementation of the Copenhagen process at all levels.

(European Ministers for Vocational Education and Training et al. 2004, p. 4)

Linking the relevance and quality of VET to the involvement of key actors is an important change in the governance model. Firstly, it implies progress in the competences assigned to these actors in decision making and responsibility, not only in terms of professional qualifications, but also in terms of quality and relevance of the system. Secondly, the inclusion of the term systematic entails a regulated involvement beyond the mere consultative role exercised occasionally or at the expense of the will of the policy makers. Cooperation between stakeholders in the design and management of the system is reinforced in successive texts (3, 5, 6, 8, 9, 10, 11, 17, 19, 20, 25, 26, 28, 29, 31, 32, 35) and is even presented as a guarantee of efficiency:

Q6—It (VET) is a shared responsibility which closely involves the Member States, the social partners and the sectoral organisations in all stages of the process—a necessary precondition for the quality and efficiency of VET…(It would be desirable)—ensuring that the social partners and economic stakeholders are properly involved in defining and implementing VET policies;

(Council of the European Union & Representatives of the Governments of the Member States 2009, pp. 3–4)

Q7—Involving, where appropriate, relevant stakeholders, including staff, learners and employers—through quality assurance—in strategic decision-making, in qualification design, and in programme development, delivery and monitoring, with a view to ensuring continuous quality enhancement within education and training institutions.

(Council of the European Union 2014, p. 3)

From 2010 onwards, references to vocational training as a general concept decreased in favour of references to dual vocational training systems. Thus, instead of VET or vocational education, terms such as dual system or apprenticeships appear:

Q8—Encouraging national partnerships with social partners in the design, implementation and governance of apprenticeship schemes, together with other relevant stakeholders;

(Council of the European Union 2013, p. 2)

Q9—(Member State should) Promote the active involvement of social partners in the design, governance and implementation of apprenticeship schemes, in line with national industrial relations systems and education and training practices.

(Council of the European Union 2018, p. 5)

At this point, it is worth questioning who is part of this group (2, 5, 6, 28, 29, 34, 35) which the EU refers to as key actors, social actors or stakeholders. The Council Recommendation of 24 November 2020 on Vocational Education and Training (VET) for Sustainable Competitiveness, Social Fairness and Resilience defines more precisely the stakeholders and key actors which should be integrated into the governance system. To this list of actors, in line with the following EU objective, the different areas of government involved in VET must be added (10, 20, 27, 29, 34):

Q10—Support sustainable partnerships for the governance of vocational education and training, in accordance with national context and, where relevant, through public-private partnerships. Involve social partners and all relevant stakeholders, including vocational education and training institutions, industries and businesses of all sizes, public and private employment services, VET teachers and trainers and their representatives, intermediary bodies such as chambers of industry, commerce and crafts, professional and sectoral organisations, national coordinators for the Youth Guarantee, ESF and other EU initiatives, the information technologies sector, Centres of Vocational Excellence, clusters, learners’ and parents’ organisations, as well as local, regional and national authorities. Promote such partnerships at regional and sectoral level;

(Council of the European Union 2020, p. 8)

Q11—Intensifying cooperation between VET policy and other relevant policy areas—Member States and the European Commission should intensify cooperation between VET policy and other relevant policy areas, such as employment, economic affairs, research and innovation, social affairs, youth, sport and culture.

(Council of the European Union & Representatives of the Governments of the Member States 2010, p. 7)

Within the theme of promoting cooperation between stakeholders for the design and management of the system, co-financing is a recurring pattern of meaning (2, 8, 18, 25, 26, 27, 31, 32). The financing of the VET system is a major challenge, both in times of economic growth and in times of greater difficulty:

Q12—Competitive business environments and strained national budgets pose challenges for ensuring necessary investments in skills;

(European Ministers for Vocational Education and Training et al. 2006, p. 3)

Q13—The economic downturn should not lead to reduced investment in VET. Budgetary constraints will force us to come up with innovative solutions to secure sustainable funding for VET and to ensure that resources are efficiently allocated and equitably distributed.

(European Commission 2010, p. 4)

The EU presents this as a key issue for the reform of vocational training systems and makes both public and private actors jointly responsible for financing the VET system:

Q14—Implementation of these reforms will not be successful without increasing the efficiency of funding in education. In order to address this complex challenge, the Commission calls on Member States to stimulate national debates on ways to provide sustainable funding mechanisms to enhance stability and efficiency, while channelling support towards those who tend to participate less;

(European Commission 2012, p. 15)

Q15—Public and private investment in VET should be improved through the further development of balanced and shared funding and investment mechanisms.

(European Ministers for Vocational Education and Training et al. 2006, p. 3)

In the same line but from a different perspective, as far as apprenticeships are concerned, the texts (2, 5, 30, 35, 27) emphasise co-investment aimed at ensuring decent working conditions for the apprentice and at making the system attractive to all actors:

Q16—Involving both employers and public authorities sufficiently in the funding of apprenticeship schemes, whilst ensuring adequate remuneration and social protection of apprentices, and providing appropriate incentives for all actors to participate.

(Council of the European Union 2013, p. 2)

Within the texts analysed (2, 18, 25, 26), we found different formulas to strengthen this co-responsibility in investment without harming the efficiency of the system and guaranteeing its sustainability. The European Ministers for Vocational Education and Training (2006) call for “the effective use of structural funds to support VET reforms at national level” (10). Furthermore, there is an invitation for MS to develop a system of fiscal incentives and subsidies in order to guarantee the involvement of the private sector:

Q17—Improving public and/or private investment in VET, including by public–private partnerships and, where appropriate, by the “training incentive effects of tax and benefit systems” as recommended by the Lisbon European Council.

(European Ministers for Vocational Education and Training et al. 2004, p. 3)

Finally, within the theme of promoting cooperation between stakeholders in the design and management of the system, the promotion of cooperation at the micro level between schools and enterprises involved in training configures another pattern of meaning (5, 8, 10, 27, 29, 31):

Q18—Governments, social partners and VET providers should make the necessary arrangements to…create opportunities for enhanced cooperation between VET institutions and enterprises (profit and non-profit).

(European Commission 2010, p. 10)

Most of the statements (8, 10, 27, 29) related to this pattern of meaning focus on the promotion of the cooperation of the teacher with the company and the cooperation of the trainer with the VET school. The aim is to ensure the acquisition of certain competences by VET school teachers (work-related competences) and in-company trainers (pedagogical competencies):

Q19—Authorities in the Member States—at national, regional, or local level—should create opportunities for enhanced cooperation between schools and enterprises in order to improve teachers’ knowledge of work practices on the one hand and trainers’ general competences on the other.

(Council of the European Union & Representatives of the Governments of the Member States 2010, p. 8)

However, in the case of apprenticeships, the promotion of company–school cooperation goes further. In this case, texts (5, 31) focus on strengthening the cooperation of the company trainer with the VET school in order to achieve greater efficiency and coherence in apprenticeship training:

Q20—In-company trainers should be designated and tasked to cooperate closely with vocational education and training institutions and teachers to provide guidance to apprentices and to ensure mutual and regular feed-back.

(Council of the European Union 2018, p. 4)

3.2. Articulating and Strengthening Social Dialogue

The articulation and strengthening of social dialogue is another of the themes produced by this TA. The EU gives a positive assessment of the social dialogue between the social partners, stakeholders and the state within the vocational education and training system:

Q21—We highly value the existing dialogue and cooperation with social partners and other relevant stakeholders, such as chambers and various competent institutions;

(European Ministers for Vocational Education and Training et al. 2015, p. 1)

Q22—We build this Declaration on underlying principles that include the due consideration for social dialogue.

(European Ministers for Vocational Education and Training et al. 2020, p. 5)

Different texts (10, 13, 14, 17, 27, 28, 29, 30, 31, 32) show the EU’s confidence in social dialogue as a tool to enhance quality and improve the internal functioning of the VET system, calling for its reinforcement. This promotion of social dialogue is also included among the objectives and priorities of the European Alliance for Apprenticeships as of its renewal in 2020:

Q23—Calls for all stakeholders, especially educational institutions, employers, employees and unions, to engage in formal dialogue with a view to ensuring that vocational education is of high quality and geared to the current needs of the labour market;

(European Parliament 2011, p. 5)

Q24—Strengthening social dialogue through more active involvement by national social partner organisations.

(Directorate-General for Employment, Social Affairs and Inclusion 2020, p. 27)

3.3. Strengthening Responsiveness to Change in the World of Work

The joint response capacity of the different actors involved in the governance of the system is a key issue in this TA of European texts (1, 2, 3, 4, 5, 6, 9, 10, 11, 17, 18, 20, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35).The mismatch between the skills covered by the vocational training system and those demanded by the current and future labour market is one of the most frequent concerns (2, 9, 10, 17, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 35). The EU stresses the importance of convergence of vocational training policies and societal demands:

Q27—Whereas skills mismatches are a worrying phenomenon affecting individuals and businesses, creating skill gaps and skill shortages and is one of the causes of unemployment;

(European Parliament 2017, p. 4)

Q28—Skills mismatches may impede productivity, growth and competitiveness. It is necessary to anticipate skills needs and shortfalls at all levels of qualification, and to transfer the results into policy and practice, with a view to improving the match between VET provision and the needs of the economy, citizens and society at large.

(Council of the European Union & Representatives of the Governments of the Member States 2010, p. 2)

However, the ambiguous statements of the previous fragments are made more concrete in others (2, 9, 17, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33), where the responsiveness of the VET system to the demands of the economy, citizens and society is replaced by adaptation to the demands of the labour market:

Q30—These challenges include adequately reflecting changes in the labour market which have an impact on the nature of developments in the field of VET;

(European Ministers for Vocational Education and Training et al. 2004, p. 3)

Q29—Considers it extremely important to adapt the education and training system to the rapidly changing labour market and the demand for new professions.

(European Parliament 2010, p. 8)

The European texts (2, 9, 10, 17, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 35), therefore, urge member states to reform their capacity to respond to these changes in the world of work:

Q31—This calls for better governance of training systems and responsiveness to the changing skill requirements of the labour market—training should be more demand-driven;

(European Commission 2006, p. 3)

Q32—Member States should ensure that the education and training provision meets the needs of the labour market. In particular, vocational education and training (VET) programs should offer a balanced mix of vocational skills and competences and create work-based learning and apprenticeships opportunities, with a focus on young people.

(European Commission 2021, p. 7)

The EU links the responsiveness of the VET system to more pluralistic governance. The theme of promoting and highlighting the cooperation of actors in detecting and adapting the system to the demands of the world of work permeates EU documentation (1, 3, 4, 9, 10, 20, 25, 26, 27, 30, 31, 32, 34, 35):

Q33—Participating countries should promote partnerships between social partners, enterprises, education and training providers, employment services, public authorities, research organisations and other relevant stakeholders, in order to ensure a better transfer of information on labour market needs and to provide a better match between those needs and the development of knowledge, skills and competences;

(European Commission 2010, p. 9)

Q34—Calls for all stakeholders, especially educational institutions, employers, employees and unions, to engage in formal dialogue with a view to ensuring that vocational education is of high quality and geared to the current needs of the labour market.

(European Parliament 2011, p. 7)

In this respect, the EU focuses on the development of a cooperative system based on mutual feedback and joint reflection that allows the creation of a flexible curriculum (3, 8, 10, 11, 17, 18, 20, 24, 25, 27, 29, 31, 32, 33, 34, 35) which can be reactive to changing labour needs:

Q35—VET curricula should be outcome-oriented and more responsive to labour market needs. Cooperation models with companies or professional branch organisations should address this issue and provide VET institutions with feedback on both the employability and employment rates of VET graduates;

(European Commission 2010, p. 10)

Q36—Calls upon the Member States to improve cooperation and partnerships between businesses and the education sector at all levels, including social partners and employers, and students and youth organisations, in particular with regard to the planning of curricula, the provision of guidance and the provision of education, training and specialisation, with a range of curricula which better meet the demands of the labour market and contribute to finding a sustainable solution to the problem of skills mismatches.

(European Parliament 2013, p. 7)

At this point, it is worth questioning how this process of joint reflection would be developed, which is another of the issues raised by our analysis. On a general level, the texts (1, 5, 9, 10, 20, 25, 27, 31, 32, 35) deal with the creation of knowledge partnerships between the different actors, including training providers, social partners, employment services and researchers:

Q37—With specific regard to the knowledge triangle, special attention should be paid to the synergies between education, research and innovation, as well as to complementarity with the aims of the European Research Area;

(Council of the European Union 2009, p. 5)

Q38—Member States should promote partnerships between social partners, enterprises, employment services, public education and training providers, authorities, research organisations and other relevant stakeholders, in order to ensure a better transfer of information on labour market needs and to provide a better match between those needs and the development of knowledge, skills and competences.

(Council of the European Union & Representatives of the Governments of the Member States 2010, p. 8)

The European texts (3, 20, 28, 29, 30, 34) also encourage the development of anticipation tools and systematic methodologies with the aim of understanding the shortcomings of the system and anticipating future needs:

Q39—The labour market relevance of VET can be strengthened by the development of forward planning tools to match skills and jobs. Based on such matching, VET providers in cooperation with local labour market representatives should be able to adapt curricula accordingly to reflect skills shortages, surpluses, skills gaps or obsolescence. Improvements of methodologies are needed to make anticipation tools coherent and comparable;

(European Commission 2010, p. 7)

Q40—Develop national and regional skills intelligence systems including skills anticipation and graduate tracking; enable social partners, decision-makers, stakeholders and providers to adapt and update VET programmes, curricula and guidelines in a timely and effective manner.

(European Ministers for Vocational Education and Training et al. 2020, p. 6)

Most texts mention monitoring and its mechanisms in generic terms. However, more recent texts (20, 29) point to the problems posed by the slowness of traditional mechanics or digital forms of skill monitoring. Therefore, they propose going a step further and integrating the use of new technological advances such as artificial intelligence or big data:

Q42—Stresses the importance of regular monitoring of future skills needs and therefore encourages the Member States and all relevant stakeholders to share good practices in this regard and to further develop monitoring and forecasting tools;

(European Parliament 2016, p. 7)

Q43—However often skills intelligence comes too late to inform choices. Accessible, easily understandable, targeted and up-to-date skills intelligence is necessary. Besides graduate tracking surveys and administrative data matching, artificial intelligence and big data analysis have a great potential. AI and big data can be applied to defining new job profiles in different sectors based on the specific skill sets required.

(European Commission 2020, p. 9)

As this last excerpt shows, the EU proposes the use of technology 4.0, not only to adjust the curriculum to the real needs of the world of work, but also to collaborate in the creation of new professional profiles thanks to its capacity to integrate data.

4. Discussion

With regard to the first research question (What model of governance of the VET system does the European Union promote?), the quotes analysed at the beginning of the results section can be interpreted as the enhancement of a cooperative governance model to the detriment of the other two single-agency forms of regulation and governance of VET systems. According to Greinert (1998), in terms of the allocation of roles and responsibilities to the different actors, as well as the underlying logic of agency, we can find the state, market and corporatist models of control. As was outlined in the theoretical framework, Busemeyer and Trampusch (2012) define four forms of skill formation governance (segmentalist, liberal, statist and collective). From the analysis of the European texts, it can be inferred that the corporatist model (Greinert 1998) and the collective skill formation regime (Busemeyer and Trampusch 2012), both based on public–private collaboration, are being promoted to the detriment of any single-agency models such as the market or state model.

The theme of promoting cooperation between stakeholders in the design, management and financing of the system further suggests the furtherance of the VET corporatist regulation (Greinert 1998) and collective skill regimes (Busemeyer and Trampusch 2012) due to the fact that, among the main characteristics of these models, the extensive influence exercised by the training companies and chambers stands out. That is, “i.e., the corporate bodies or associations that represent the business community at the local or regional level” (Rauner and Wittig 2010, p. 3). According to Rauner and Wittig (2010), this type of governance is present within mixed systems of cooperative governance in which the regulation of vocational education and training takes place in a pluralistic network consisting of state bodies, companies or business associations, as well as trade unions or professional associations. As the results of the analysis show, the EU invites member states to involve different stakeholders in the processes of defining, implementing and financing VET systems for the sake of this mixed system of cooperative governance. This form of stakeholder involvement in strategic policy making is a reality in dual VET countries such as Austria, Switzerland, Germany and Denmark (Bürgi and Gonon 2021; Emmenegger and Seitzl 2020; Trampusch 2010).

The functioning of the governance networks referenced before is based on social dialogue (Hippach-Schneider and Huismann 2016) as a tool for reaching agreements. Precisely, the promotion of dialogue between actors as a tool for improving the quality of the system is another of the themes configured during this TA of the European texts. Likewise, the academic literature names the variety of stakeholders involved in the governance of cooperative governance systems (Alemán-Falcón and Calcines-Piñero 2022; Emmenegger et al. 2020; Hippach-Schneider and Rieder 2021) which coincide with the extensive list of stakeholders proposed by the EU: different areas of government, VET institutions, industries and enterprises, public and private employment services, VET teachers and trainers and their representatives, intermediary bodies such as chambers of industry, commerce and crafts, professional and sectoral organisations, trade unions, national Youth Guarantee coordinators, the EU and the actors resulting from its initiatives, the IT sector, centres of vocational excellence, clusters and learners’ and parents’ organisations, as well as local, regional and national authorities.

Furthermore, the third of the major themes detected in this TA (strengthening responsiveness to change in the world of work) coincides with the extensive responsiveness to changes in the world of work which dual VET models present (Busemeyer and Trampusch 2012). This responsiveness is the result of the articulation of systematic collaboration and the use of strategies for detecting these needs (Hippach-Schneider and Huismann 2016). Precisely, the strengthening of collaboration to detect the needs of the world of work, as well as the introduction of new tools for monitoring labour market needs, is part of the patterns of meaning configured under the scope of this theme.

Regarding the second question guiding this analysis (Which good practice example of VET governance is sponsored by the European Union?), it can be noticed that the analysed texts do not establish a specific governance system as a good practice model. Our interpretation indicates that, while encouraging a system of cooperative governance, the recommendations are general enough to accommodate any country which makes use of such a VET governance model. This means that the dual VET models of Austria, Denmark, Germany and Switzerland, where we can find variations of mixed systems of cooperative governance, fall within the governance good practice model outlined by the themes configured through the analysis of the European documentation. The differences present in these models cannot be captured within the EU recommendations on promoting a more pluralistic governance because they are general and encompass the different models present in these countries. As outlined in the theoretical framework, comparative research has underlined differences between the different MS models of dual VET related to aspects such as the general orientation of the system (Emmenegger et al. 2020), the role of the different stakeholders (Bürgi and Gonon 2021; Emmenegger and Seitzl 2020; Trampusch 2010), the coordination between the different agents with their respective internal logic or the input or output orientation (Rauner and Wittig 2010), among other issues.

By delving deeper into the policy transfer perspective, the results reached under the guidance of these two research questions accredit the European Union with a role as an agent of the transfer (Steiner-Khamsi 2014) of dual VET policies to the reforming MS. The results lead us to conclude that this actor has exercised agency in the dissemination of the collective skill system governance regime (Busemeyer and Trampusch 2012) as a model of good practice during the cross-national attraction phase of policy transfer (Phillips and Ochs 2004). As explained before, this VET governance model is traditionally associated with dual VET systems at the expense of single-agency models such as the market or state model (Busemeyer and Trampusch 2012; Greinert 1998). Such a promotion fosters the national debate of the reforming MS, starts a learning process and fosters the legitimisation of alternatives (Zaunstöck et al. 2021). This fact becomes very relevant in the beginning of the process of policy transfer. During the reception phase defined by Cowen (2006) and the cross-national attraction phase defined by Phillips and Ochs (2004), the transnational movement of ideas depends on a series of impulses and the potential for externalisation. The role of the EU increases the internal impulses for reforming the systems, not only by pointing out the shortcomings of the present system but also by the promotion of the benefits and advantages presented in the model of best practices. This practice coincides with the tool described in Novoa (2002) which is used by the European Union to keep its influence among non-hierarchical areas (as vocational education), i.e., the dissemination of best practices and the promotion of joint reflection. Simultaneously, the EU fosters the potential for externalisation by overlapping its VET governance recommendations and those governance patterns presented into the dual VET models. Again, it is worth noting the absence of detailed specifications which explicitly associate the proposed patterns with the present one in any single country or region. As outlined before, there are certain differences between the governance models of the different national dual VET systems. However, the set of general recommendations configures a governance model of best practice during the policy transfer process which is close to the ones found in the dual VET systems of certain MS, such as Germany, Denmark or Austria.

5. Conclusions

This paper sought to explain the role played by the EU in the promotion of the dual VET model as the reference for the southern, reforming MS during the policy transfer stage of cross-national attraction (Phillips and Ochs 2004). In doing so, this paper provided a deep analysis of the main texts related to VET from the EU Governance Triangle (Cort 2009) in order to interpret the VET governance model promoted by the EU since the signing of the Copenhagen Declaration of 2002.

This TA concluded that the EU encourages the implementation of a cooperative governance model of VET based on an extensive network of actors, which is traditionally associated with the dual model of VET. According to the themes shaped through this TA of the EU documentation, the EU promotes a model of VET governance based on public–private partnerships to the detriment of single-agency models such as the market or state model (Busemeyer and Trampusch 2012; Greinert 1998). The promotion of cooperation between stakeholders in the design, management and financing of the system, in conjunction with the strengthening of the system capacity to respond to changes in the world of work and the strengthening of social dialogue, is a recommendation that mimics the basic characteristics of the collective skill systems of VET governance which are associated with dual VET (Busemeyer and Trampusch 2012; Rauner and Wittig 2010).

From a policy transfer perspective, these results are a testament of the role exercised by the EU in the dissemination of dual VET as a model of good practice, setting itself up as an agent of policy transfer (Steiner-Khamsi 2014) during the cross-national attraction phase (Phillips and Ochs 2004). The role of the EU increases the internal impulses for reforming the governance of the VET system among certain MS whose VET system has been traditionally based on single-agency models such as the market or state model (Busemeyer and Trampusch 2012; Greinert 1998). This role has been played not only by pointing out the shortcomings of the present system but also by the promotion of the benefits and advantages presented in the model of best practices.

However, it cannot be concluded that the EU appropriates on an exclusive basis any VET governance model of a single member state’s dual VET system, as the EU recommendations are general enough to accommodate any country developing a mixed model of cooperative governance (Austria, Denmark, Germany, etc.). According to Rauner and Wittig (2010), the mixed model of cooperative governance varies between national contexts in terms of orientation and fragmentation. Furthermore, other authors point out the differences present among national dual VET governance systems with regard to aspects such as the general orientation of the system (Emmenegger et al. 2020) or the role of the different stakeholders (Bürgi and Gonon 2021; Emmenegger and Seitzl 2020; Trampusch 2010). On the basis of these theoretical explanations, the dual VET governance models of Austria, Denmark, Germany and Switzerland, where we found variations of the cooperative governance model, coincide with the VET governance patterns promoted by the EU.

In short, this paper concludes that the EU exercises an agency role during the cross-national attraction phase (Phillips and Ochs 2004) of the dual VET policy transfer which has been taking place in Europe since the signing of the Copenhagen Declaration of 2002. During this process, the EU texts favour the establishment of a more pluralistic system of governance. This is a system of cooperative governance similar to that found in the Central and Northern European dual VET systems characterised by a collective skill formation regime. However, these recommendations are so open-ended that they only convey the general philosophy of the model, leaving it up to the member states to specify its detailed configuration. In summary, the EU indicates a VET governance reform path based on cooperation, extensive networks and social dialogue which is close to the German approach, but also to that present in other dual VET systems, such as those in Austria or Denmark.

Future research will have to delve deeper into the European level to elucidate whether the concreteness of the dual VET model is more present in the rest of the EU mechanisms of soft power (conferences, subsidies, contests, agencies) (Dimitrova et al. 2016; Dolowitz and Marsh 2000) which promote the reform of VET systems. At the same time, further research is recommended for the analysis of the voice of EU institutions regarding the rest of the aspects that configure the VET systems (e.g., the training model, lifelong learning, etc.) in order to determine if European institutions advocate for dual VET systems among them.