Development paradigms in the institutional configuration of vocational education and training in Chile (1964-2005)

ABSTRACT Mainstream approaches in comparative education have paid more attention to trends in donors’ priorities than to policy processes in recipient countries, overlooking the agency of national governments in the institutional configuration of vocational education and training (VET) in developing contexts. Drawing on constructivist political economy and historical institutionalist frameworks, the paper analyses the role played by global development paradigms and ideas in the VET policy reforms adopted in Chile during the period between 1964 and 2005. The study shows how the selective adoption of international development ideas in Chile was mediated by the ideological orientation of national government administrations in each historical period. It also shows how the main goals of VET policies evolved as the economy tertiarised and the demand for education grew among lower social classes. Finally, the paper identifies the military dictatorship as a path-shaping moment in the institutional configuration of VET in Chile, as it was able to impose a radical transformation of the system from central planning to decentralised market competition. This profound transformation generated institutional path-dependencies that restricted the capacity of future democratic governments to reverse the market model in VET.


Introduction
The institutional configuration of vocational education and training (VET) systems in developing countries has received little attention from comparative education studies and still remains largely under-theorised (McGrath 2012). The most salient effort to provide an overarching explanatory framework of the emergence of differentiated VET models in developing contexts was proposed by Ashton, Turbin, and Sung (2000), which argued that these institutional variations were caused by the limited capacity of governments in developing countries to resist the influence of multilateral financial organisations and respond to the challenges posed by economic globalisation. More recent comparative studies (Maurer 2012) have shown how policy actors in some developing countries actually deviated from the priorities in VET set by international organisations, and how local demands emerging from educational expansion and economic development dynamics better explain changes in national VET systems.
Our study aims to contribute to these international debates by analysing how the interplay between global and national drivers of policy change shape the institutional configuration of VET systems in developing countries. Drawing on the combination of constructivist political economy and historical institutionalist frameworks (Hay and Wincott 1998), our study specifically investigates the role played by global development paradigms and ideas in the VET policy reforms adopted in Chile during the period between 1964 and 2005. The period covers the military coup of 1973 and the transition to democracy after the national plebiscite of 1988, which constituted critical historical junctures for the development path of the country. The case of Chile is particularly relevant to these debates because it was originally characterised by Ashton, Turbin, and Sung (2000) as a paradigmatic example of the neo-market model of skills formation that emerged in Latin American countries as a result of neoliberal globalisation. Whilst more recent policy studies have analysed current policy efforts to reform the VET system (Zancajo and Valiente 2018), no research has been conducted so far to explain the historical processes leading to the institutional configuration of the market model of VET in Chile. 1 At a methodological level, our research mainly relies on a review of academic publications and grey literature. Grey literature during some of these periods is especially rich because of the professional mobility of national experts from university to government, and from government to international organisations and think tanks. In addition to the secondary sources, we analyse the content of 12 policy documents that offer first-hand direct insights on the dominant policy ideas and discourses during each period under study.
The paper starts by reviewing comparative approaches to the institutional configuration of VET systems in developing countries and outlines the framework adopted in our research. The following three sections analyse VET policy developments in Chile during three distinct historical periods. The first section analyses the influence of manpower forecasting ideas, under the intellectual leadership of the United Nations Economic Commission for Latin America and the Caribbean (ECLA), in the education reform of 1965. The second section covers the adoption of the neoliberal paradigm during the authoritarian regime of Augusto Pinochet, which completely transformed the governance of the system from central planning to decentralised market competition. The third section focuses on the influence of the 'growth with equity' ideas, again promoted by the ECLA, in the post-transition democratic governments. The conclusions reflect on the theoretical implications of our analysis for comparative studies of institutional change of VET systems in developing countries.

The configuration of VET systems in developing contexts
For many years mainstream approaches in comparative education (i. e. World Culture Theory and International Political Economy) paid more attention to global drivers than to the policy process in recipient countries, assuming that governments in developing countries had no or little capacity to resist external pressures in the adoption of global development ideas. World Culture Theory (WCT) tried to explain, through a neo-institutionalist lens, the apparent convergence of education systems around global models. For WCT, the global spread of Western ideals of progress and modernity trigger legitimation pressures over policymakers in developing countries that force them to demonstrate their conformity with these ideals, for example, through the institutionalisation of mass education (Ramirez and Boli 1987).
Although VET has not attracted much attention from WCT scholars, there are some important exceptions. 2 In the early 80s, Benavot (1983) predicted a decline of secondary VET in developing countries caused by the global spread of the egalitarian ideal of comprehensive schooling. For the author, the 'limiting and unidimensional' training offered by vocational schools was incompatible with a strong individualist culture and the expansion of access to secondary education as a citizenship right (73). Contrary to what Marxist structural functionalists would expect, the satisfaction of economic imperatives would not protect VET from its inherent incompatibility with modern education values and the legitimation needs of governments in developing countries. Interestingly, the thesis of the decline of VET systems rightly predicted the fall from grace of the sector among international organisations like the World Bank (Ziderman and Van Adams 1991), but failed to explain that the causes of this policy shift were not in the spread of liberal and egalitarian ideals but in changes in the economic development paradigms of these organisations (Heyneman 2003).
International Political Economy (IPE) has provided a more adequate framework to explain global drivers of education policy reforms in developing countries than the one offered by WCT. IPE authors argue that the liberalisation of international trade and the advance of global capitalism since the late 80's have put pressure on national governments to reform their education systems and to align them with the neoliberal competitive strategies promoted by international organisations (Dale 2000). These international organisations influence national education policy through different mechanisms, ranging from more coercive (to more persuasive ones, depending on the level of autonomy of countries from these external pressures (Dale 1999). The most significant contribution from IPE scholars in the field of VET was made by Ashton, Turbin, and Sung (2000). They showed how governments responded differently to the challenges posed by economic globalisation depending on the role of the State in economic development and the geopolitical position of the country. Particularly interesting is their analysis of the transition of Latin American countries from a highly-centralised skill formation system to a neo-market model in the 80s (Ashton, Turbin, and Sung 2000, 20). Under import substitution industrial strategies, Latin American countries developed emergent State-owned industries that demanded qualified manpower (technicians, managers); a manpower which at that time was trained by public institutions. With the deregulation of the economy and the adoption of export-orientated development strategies, governments decentralised their education systems and privatised training institutions that now had to compete among themselves to satisfy the demand from foreign-owned industries connected to global markets. This radical transformation could not be explained without the external imposition of stringent neoliberal development agendas by multilateral financial institutions like the World Bank and the IMF, and the repression of social contestation to the reforms by authoritarian regimes.
The importance attributed by IPE scholars to global economic drivers of policy change contrasts with the focus on national drivers and institutional pathdependencies in the work of historical-institutionalist authors. For these authors, the scope of options available to policy actors in different skill formation regimes will be limited by path-dependencies that have been historically configured through critical junctures and reinforced by recurrent complementarities between educational, economic and political institutions (Busemeyer and Iversen 2012). Maurer (2012) made the most significant contribution to this tradition in our understanding of the configuration of VET systems in developing countries. He criticised Ashton, Turbin, and Sung (2000) for explaining VET policies as functional adaptations to global economic changes, and showed how governments in Sri Lanka and Bangladesh deviated from the priorities set by international organisations and continued investing in their VET system even after opening their economy to global markets. Drawing on the structural elaboration framework of Archer (2013), his analysis of the two cases shows how the political struggle for social mobility within the education sector creates its own social demand independently from its economic function. Although VET policies in the two countries were initially more closely linked to business needs, they became more embedded into the institutional logics of the education sector as the access to education expanded and modern economic sectors grew (Maurer 2012, 500). According to him, the different political management of these educational and economic dynamics by national policy actors over time is what explains variations in the institutional configuration of VET in the two countries.
The structural elaboration version of historical institutionalism proposed by Maurer (2012) is possibly the most accomplished framework to explain the interrelationship between strategic policy actions and institutional contexts in the configuration of VET systems in developing countries. However, recognising the strategic character of national policy decisions, requires necessarily to problematise the formation of perceptions and ideas of national policy actors (Hay and Wincott 1998). Recent constructivist turn 3 in political economy studies has tried to address this problem by incorporating the causal power of ideas and economic imaginaries in the analysis of policy changes (Jessop 2010). Its application in comparative education has led to a growing interest in the interplay between ideational and material drivers of adoption of global education models in different political economy contexts (Verger 2014).
In the following sections we combine constructivist political economy and historical institutionalism into the analysis of VET policy developments in Chile during the period between 1964 and 2005. Our approach suggests that historical institutionalism requires a constructivist social ontology to explain the role of ideas in the formation, evolution and transformation of VET systems. Although historical institutionalism recognises that policy change is the consequence of strategic policy actions in selective institutional environments (Maurer 2012), this assumption only addresses the tendential structuralism of institutional approaches partially. Policy actors are thrown into a world full of meanings and ideas, and they strategically appropriate this world through perceptions of what is feasible, legitimate, possible and desirable (Hay and Wincott 1998, 956). As we will show for the case of Chile, combining social constructivism and historical institutionalism allows us to better analyse how institutional settings and global development paradigms shape the range of policy options available to policymakers in VET in each historical period.

Structuralism: 1964-1973
The broad electoral victory in 1964 of Eduardo Frei, the candidate of de Cristian Democratic party, supposed the beginning of the end of the 'State of Commitment', a political alliance between the right and the centre-left which ruled Chile from 1938. Beyond the political consequences of this electoral victory, it also supposed the end of a period of pro-liberal political and economic reforms that was replaced by a structuralist development strategy characterised by the increasing role of the State and central planning (Taylor 2003).
The adoption of this structuralist development paradigm in Chile was strongly influenced, among other factors, by its promotion across Latin America by the UN agency ECLA. 'Latin American structuralism' was mainly based on the idea that the neoclassical approach to development might be adequate for northern countries but not for peripherical economies (Kay 2009). To overcome the limitations of the neoclassical approach, the ECLA strongly advocated for import-substituting economic policies as a means to switch from the 'outward-looking development' (desarrollo hacia afuera) to an 'inwarddirected' one (desarrollo hacia adentro) (Kay 1989, 27). For the ECLA, this reorientation of development strategies would lead to an important process of industrialisation of Latin American economies (Kay 1991).
The most relevant institutional change produced during this period was the 1965 educational reform, which was based on some of the principles established in the 1962 ECLA conference on education and development (ECLA 1962). This policy initiative expanded primary education from six to eight years, made it mandatory and free for everyone, and promoted active learning approaches (Celis 2004). In the case of secondary education, the reform created a VET track that, for the first time, allowed its participants to obtain a secondary education certificate that gave them access to higher education (Miranda 2003b). This reform widened the access to primary and secondary education among the poorest populations, reducing the number of early school leavers and the percentage of primary school-age out-of-school children (Cox 1986).
The 1965 educational reform adopted by the Cristian Democratic government has been qualified by different authors as an important turning point for the Chilean education because of its democratiser effect (Núñez 1984). Nevertheless, Ruiz (2010) considers that this reform did not necessarily aim to democratise education but to align the education system with the requirements of the new development paradigm.
[. . .] this process does not primarily aim to expand democracy. It is based on a very specific conception of education that is more related to global conceptions of development, and more precisely to economic development. Its relationship with the democratisation of the Chilean educational system exists, but clearly subordinated to the issue of development (Ruiz 2010, 84). Secondary VET was designed as an educational offer that would train the new technical elite required under the economic development strategy of the country 4 (Schiefelbein 1971). In 1966, the National Institute of Training was created. This State institution was in charge of planning and delivering training courses for the workforce (Núñez 1990) and guaranteeing the alignment of training programmes with the development requirements of the country.
The implementation of the 1965 educational reform was continued by the socialist government of Salvador Allende that came to power in 1970. The new government formulated an ambitious education reform project (Unified National School) that was not finally approved because of the coup d'état of 1973 (Núñez 2003). The educational reform of Allende sought to eliminate the institutional and curricular differences between academic and vocational tracks in secondary education, as well as to strengthening the links between education and the world of work (Cox 1986). This proposal aimed to combine both the productive and the socio-political objectives of education, favouring a new educational experience where manual/intellectual integration was its guiding principle (Núñez 2003).

Neoliberalism: 1973-1989
The coup d'état led by Augusto Pinochet in September 1973 against the socialist government of Salvador Allende caused a drastic shift in the Chilean development paradigm. The military dictatorship that ruled the country between 1973 and 1989 imposed a neoliberal model of economic development that directly affected social policies, including education. Different authors agree that the absence of a pluralistic setting explains the capacity of the new government to dramatically alter the development paradigm and the orientation of public policies (Salazar and Pinto 2002). The authoritarian character of the government and the ideological adoption of the neoliberal paradigm explain why Chile has been considered an extreme case internationally, as other countries adopted these neoliberal ideas in a more 'pragmatic' way (Fourcade-Gourinchas and Babb 2002).
The model and the strategy of development imposed by the military dictatorship were based on three main components. Firstly, a policy approach to economic development based on monetarist principles. Broadly speaking, this entailed a reduction in the role of the State in economic activity in favour of the private sector, which became the main driver of development. Secondly, the privatisation of State assets and the involvement of private actors in the provision of public services. Third, the establishment of market mechanisms of coordination of supply and demand in different social sectors, and the deregulation of these sectors (Calcagno 1989).
In this context, education was reoriented towards the generation of human capital and enhancing the productivity of the economy, leaving aside other social objectives and the structuralist development strategy that characterised the reformist governments of the previous period (1965)(1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973). The next quote from the military government's white book ('el ladrillo') 5 elaborated by the Chicago Boys 6 summarises the role given to education in the development strategy of the country, as well as the aim of making the education system more responsive to economic and production needs.
The human factor is, without a doubt, the most important productive resource of any country, which progresses in quality through, fundamentally, the educational processes. Our overly humanistic and encyclopaedic general (primary and secondary) education conspires against the necessary process of modernisation of the country.
[. . .] It also engenders the idea of the uselessness of education as an instrument of productivity, thereby hindering the access of education to the workers who need it most. There is an evident need to reform general education by giving it more practical and technical content. (CEP 1992, 46-47) The adoption of the neoliberal paradigm of development and the human capital approach in education had clear implications for education policy (Arnold and Krammenschneider 1999). In basic education, an extreme version of the education market was adopted. This meant, amongst other pro-market reforms, the implementation of a universal voucher system for both public and private schools, in an effort to facilitate the school choice capacity of families and to incentivise competition dynamics between educational providers. In parallel, the management of public schools was decentralised from the State to municipalities and teacher labour status was deregulated, aligning them with private sector workers (Cox 2003). These measures sought to promote competition among public and private providers, to reduce the power and influence of teachers' unions, and to limit State intervention in education to a subsidiary role (Verger, Fontdevila, and Zancajo 2016).
Secondary VET experienced important transformations during this period. First, the government transferred the management of 71 vocational education schools to companies and business associations' private foundations, trying to align the supply of education with economic and labour market needs (Corvalán and Santibánez 1987). VET schools were given autonomy to design their own curriculum and to adapt it to the requirements of local and regional economies (Núñez 1984). These pro-market reforms laid on two main assumptions. On the one hand, that students and families would choose their VET studies based on instrumental calculations of the expected labour and economic returns of their educational investment. And, on the other hand, that the competitive environment created by the per capita funding scheme (voucher) would force VET providers to offer specialisations that were aligned with the productive requirements of the economy, which at the same time would maximise their attractiveness of their schools to students and families. Nevertheless, the autonomy given to educational providers and the high level of deregulation of the system generated a dramatic increase in specialisations offered by VET institutions, many of them completely irrelevant to the labour market. A study carried out in 1998 by the Ministry of Education showed that, as a result of the reforms adopted in the 80s, there were at least 1,595 different VET plans of study in the country (Miranda 2003a).
As it had happened in secondary education, the reforms in higher education aimed to diversify the supply of education and to introduce competitive mechanisms in a sector that was considered 'closed and virtually a monopoly' (Cox 1989). The reforms led to an exponential expansion of the number of higher education institutions in just eight years, from 8 to 26 universities, 30 professional institutes and 123 technical training centres. Working social classes could not benefit from this expansion because of the predominant role of private providers in tertiary VET. As in the case of secondary VET, the high level of deregulation also generated curricular specialisations barely aligned with the requirements of the labour market (Miranda 2003a).
Between 1981 and 1987, the enrolment in secondary VET decreased significantly as a result of two main factors. First, the per capita funding scheme (i.e. voucher) had provided the same amount of public funds for schools to offer vocational and academic tracks, but the actual cost of offering vocational specialisations was higher for schools than in the case of the academic ones.
Second, the dramatic expansion of private schools was at expense of VET, as these schools were more likely to offer an academic secondary education (Supanc 1995).
It was not until 1987 that enrolment in VET increased significantly. This expansion of VET during the military dictatorship is not explained by the increasing level of deregulation, but by the fact that VET became the only accessible offer to the increasing demand for education amongst socioeconomically disadvantaged students. As Messina and Weinberg (1996) point out, the raise in VET enrolment during this period does not necessarily has to be read as an increasing public policies interest in this educational track, but as the acknowledgement from lower social classes that the Chilean university system was not an attainable option for most of them.

Growth with equity: 1990-2005
The 1988 plebiscite ended the military dictatorship in Chile. One year later, presidential elections took place, and the centre-left candidate came to power in 1990. This first victory started a period where the centre-left coalition would be electorally hegemonic for 16 years. The restoration of democracy inaugurated a period where the development strategy of the country tried to combine both economic growth goals and equity improvements. This approach became to be known as 'growth with equity ', or what Taylor (2003) defines as a 'particular form of neoliberalism with a human face' (p. 27). The reasons for the centre-left governments not to reverse the market model imposed by the dictatorship in education are diverse, 7 but what is clear is that the inheritance of these institutional settings restricted and significantly influenced the policy orientations adopted by the new democratic governments. For the education system, the 'growth with equity' development paradigm had two main consequences. On the one hand, the continuation of the promarket model adopted during the military dictatorship with the same high levels of private provision of education. And, on the other hand, the development of several compensatory initiatives and policies orientated to improve the deteriorated levels of equity in the education system. This combination of market structures and greater State involvement is what Falabella (2015) has denominated the 'equilibrium mantra'. This equilibrium mantra assumed that the market mechanism guaranteed individual freedom and efficiency, while the State should support educational institutions so that, through the market, the achievement of objectives of 'human capital development, citizen training and greater educational equity' were reached (Falabella 2015, 708).
As in the period previous to the dictatorship, the regional diffusion by the ECLA of the need to adopt a development paradigm which combined a productive transformation with higher levels of social inclusion, played an essential role in influencing the approach adopted by the centre-left government. Between 1980, the ECLA adopted what Sikkink (1997 considers a paradigm shift in the development ideas for Latin America. Broadly speaking, and amongst other issues, the core idea of this new approach was that equity improvement would lead to economic growth. This is in direct contrast with previous policy orientations in which economic growth was conceptualised as a driver for the reduction of social inequalities and poverty alleviation (Sikkink 1997). Likewise, under this new paradigm, the ECLA stressed the importance of international competitiveness as an essential factor of development (ECLA 1990), which would overcome the inward-directed development promoted during the 60s and 70s.
This new approach in the development priorities and policy orientations had clear consequences for the education system and particularly for VET. In 1992, the ECLA and UNESCO published a policy document entitled 'Education and knowledge: basic pillars of changing production patterns with social equity'. This document considered that education should play a central role in this new economic development strategy by transforming comparative advantages into competitive ones: To reconcile social equity with international integration, productivity must be constantly increased, and all of the institutional capacities of countries must be improved, in a process in which education and dissemination of technical advances play crucial roles. The proposal contained in the document Changing Production Patterns with Social Equity clearly pointed in this direction, as efforts are under way to develop this approach in the areas of education, training and development of national capacities for innovation. (ECLA-UNESCO 1992, 29).
One of the main consequences of this approach is the focus on human capital formation as a key driver of productivity improvements. The public policies developed by the new democratic governments emphasised the need to increase and improve the human capital of the country as the main factor to foster economic development but also to promote social inclusion. This led to higher investment and resources dedicated to expanding access to all education levels, from preschool to higher education (Arellano 2001).
During this period, the value of VET for development was called into question at the international level by the World Bank, as it was considered students in this sector were not getting the expected returns to investment (Ziderman and Van Adams 1991). This very same idea was shared by one of the most influential policy documents elaborated in Chile at the beginning of the 90s, known as the 'Brunner report'. 8 Cariola (2003) considers that the VET policies adopted after the restoration of democracy were based on the idea that this educational path was low quality, reduced the chances of its graduates to continue studies, and was only valid as a way to address the extreme precarity and limited educational opportunities available to the most disadvantaged in society at that time.
Despite the secondary role given to VET by the new democratic governments, the 1998 educational reform brought an important rationalisation of the curricular specialisations offered by the sector. This curricular reform responded to the idea, particularly emphasised by the 1992 ECLA-UNESCO policy document, that there was a significant mismatch between the supply of VET and the requirements of the labour market, as the next quote summarises: The sector most affected by this deterioration has been technical training, which has become increasingly irrelevant in the face of ever more numerous and diversified requirements for the occupations the students are preparing for. The fragmentary studies available coincide in pointing out the inefficiency and high cost per student in this education. Criticisms include its lack of communication with enterprises, rigidity in responding to their needs, resistance to change rising from the inertia of existing programmes and equipment and inadequate teacher training (ECLA-UNESCO 1992, 44).
The 1998 educational reform assumed the diagnostic summarised above regarding the division between academic and vocational tracks and the need to update curricular approaches to improve the relevance of VET. In this respect, the main objectives of the reform were (Miranda 2003b), 1) improving the alignment of VET with the requirements of specific productive sectors, 2) deferring the separation of students between academic and vocational tracks, as well as favouring the access to general and basic skills, 3) restructuring the education system to concentrate student specialisation in the last two years of secondary education.
Despite the objectives stated above, the curricular reform did not solve the problems of low labour market relevance of VET and the low social prestige of this educational option (Espinoza 2008). Likewise, despite the policy initiatives adopted in this period, they were not sufficient to articulate a coherent and strong VET system, which would improve the alignment between secondary and tertiary VET and the coordination between demand and supply of training. In part as a result of these shortcomings, the demand for higher education continued increasing dramatically, particularly university education. Between 1990 and 2009 university undergraduate students raised from 127,133 to 535,643 (321%), and the enrolment in tertiary VET institutions increased from 117,780 students to 299,604 (154%) (Espinoza and González 2013).

Conclusions
The analysis of the historical configuration of the VET system in Chile helps us to explain why the WCT authors failed when predicting the decline of this sector in many developing contexts. As IPE authors had rightly argued (Ashton, Turbin, and Sung 2000), the neoliberal turn in international development paradigms in the late 80s did not lead to a decline but to a transformation from state-led to market models of VET in many Latin American countries. However, rather than being the result of an external imposition, the influence of these international development ideas was heavily mediated by the neoliberal ideological orientation of the authoritarian regime in Chile. In fact, Chile was a paradigmatic case of ideological and purposeful adoption of these neoliberal ideas in the education reforms of the military regime, which created a quasi-market model of unregulated provision that resulted in large educational inequalities. This path-shaping moment in the institutional configuration of the VET system was not reversed by the subsequent democratic governments. Despite the egalitarian development ideas of these governments, their reform efforts to promote more equitable education did not challenge the fundamental structures of the market model. Interestingly, as Maurer (2012) had showed for other developing contexts, the expansion of VET during the democratic period in Chile cannot be explained just as a functional response to the evolving needs of businesses and the economy, but as the result of political strategies to reconcile the economic objectives of VET with the growing educational aspirations from lower social classes. The political management of these economic and social goals is what led to the institutionalisation of VET as the default education pathway for the most disadvantaged in society.
The combination of social constructivism and historical institutionalism in the analysis of TVET policy developments in Chile provides important insights for comparative studies of institutional change of VET systems in developing countries. These insights refer to the mediated influence of global development paradigms in national policy changes, to the competing institutional logics that national governments embrace in different phases of their industrialisation, and to the difficulties for radical institutional change in periods of political pluralism. We develop these three ideas below.
Firstly, global and local drivers of VET policy changes do not operate separately but are reciprocally interrelated. The influence of international ideas and actors on education policies in Chile has been mediated by the agency and the ideological orientations of national governments. While progressive democratic governments adopted the more egalitarian and humanistic development paradigms promoted by ECLA, the right-wing governments of the dictatorship opted for the neoliberal recipes promoted by multilateral financial organisations like the World Bank. As constructivist political economy approaches in comparative education have rightly argued (Verger 2014), the selective adoption of international development ideas by national actors shows how important is to problematise the formation of policy ideas for the explanation of the influence of international organisations on national VET policy reforms.
Secondly, in the re-contextualisation of these international development ideas into concrete VET reforms, national policy actors have to accommodate competing social demands that are evolving over time. The Chilean experience illustrates how VET policies tend to be captured by the productive requirements of the economy during industrialisation efforts, while the education logic tends to dominate with the steady tertiarisation of the economy and the expansion of access to education. Specifically, in the case of Chile, the economic and academic barriers faced by the lower social classes to access the academic routes fuelled the expansion of VET both at secondary and tertiary levels. This is a good example of how the historical reasons that explain the initial formation of a VET system are not the same that explain its subsequent institutional continuity and expansion.
And finally, radical transformations of institutional structures in the configuration of VET systems are rare, even in the case of critical historical junctures as the transition from dictatorship to democracy. The experience of Chile shows how the military dictatorship that ruled the country between 1973 and 1989 was able to impose a radical transformation of the VET system from a state-led to a market-oriented model precisely because of the authoritarian nature of the Chilean State and the lack of political pluralism. The profound transformations in economic, political and social domains during the dictatorship generated institutional path-dependencies that restricted the capacity of future governments to overturn the market model despite the political hegemony of centre-left administrations in the transition to democracy. This pattern has been repeating itself even after the student protests of 2008/2011 and the second government administration of Michele Bachelet in 2014-18 (Zancajo and Valiente 2018), which did not manage to fundamentally alter the institutional configuration of the education system in Chile.

Notes
1. The VET sector in Chile is a very important component of the initial education offer, with around 40% of upper secondary education students attending VET schools and around 31% of tertiary education students enrolled in VET institutions (Centro de Estudios Mineduc 2017). For an analysis of the current institutional design and enactment of the market model of VET in Chile, see Valiente, Zancajo, and Jacovkis (2020). 2. Although the work of Jakobi (2012) does not focus on VET specifically, her analysis of the global diffusion of lifelong learning and knowledge economy imaginaries helps to explain why, since the late 90s, international organisations like UNESCO or the OECD have been promoting policy reforms that remove institutional barriers in VET and facilitate mobility of learners from secondary to tertiary education, from vocational to academic routes, and from initial to continuous education. 3. The 'constructivist turn' in political economy has led to the integration of analytical tools from social constructivism into the examination of the role of ideas in policy changes in different political economy contexts (Abdelal, Blyth, and Parsons 2010). A core premise of constructivist political economy is that policymakers interpret the context in which they act through the lens of intersubjective understandings or paradigms (Hall 1993) that specify the outcomes that policies should pursue and the most feasible and effective options to achieve them (Hay 2006). 4. It is important to bear in mind that, in the late 60s and early 70s, the access to secondary education was far from becoming universal. In 1970, the enrolment ratio was 39% (Ponce de León 2012), and therefore the concept of 'elite' refers to those who were able to access secondary education. 5. Although it is not clear when exactly 'el ladrillo' was elaborated, it was at some point between 1969 and 1972. However, the reference presented in this paper corresponds to the first public edition published by the Centre of Public Studies in 1992. 6. Chicago Boys is denomination frequently use when to refer a group of Chilean economists trained at the University of Chicago in the sixties and seventies, that later assumed strategic positions as ministries or senior advisers in the military government (Valdés 1995). 7. Different authors, particularly in the field of education policies, have identified diverse factors to explain why the new democratic governments did not reverse pro-market reforms. Some of them are: the adoption of market principles by the centre-left coalition, the lack of a necessary parliamentary majority to reverse some of these policies, the political and military power maintained by relevant actors of the dictatorship, the pressure of middle and upper-middle social groups, or even the conditionalities imposed by the World Bank and the global hegemony of neoliberal ideas during the nineties (Gauri 1998;Bellei and Vanni 2015;Cox and Avalos 1999). 8. The Brunner report is the output of the National Commission for the Modernisation of Education, a panel of experts and key stakeholders organised by the second government after the restoration of the democracy to define the long-term challenges of the Chilean education system (Bellei and Vanni 2015).

Disclosure statement
No potential conflict of interest was reported by the authors.