Skills and Training in Hierarchical Capitalism: The Rise and Fall of Vocational Training in South Korea

Abstract

From an economic model in which education and growth reinforced each other, South Korea has developed a pathological equilibrium holding back economic and social progress. Low labour productivity and skills mismatch undermine the economic prospects of the country, and sharp rises in inequality in an ever more dualised labour market erode social cohesion. Governments of different political persuasion have recognised these challenges, and they have thought to reinvigorate vocational education and training (VET). However, this article shows that large employers – which were at the heart of a segmentalist coalition between business and government when collective skills formation of the Developmental State was dismantled – continue to undermine any efforts of meaningful vocational skills formation. It is argued that the country’s hierarchical production regime and, related to this, labour market dualisation provide the micro-foundations for successive failure in VET reform; and without challenging large employers’ dominant position in the Korean political economy and without addressing labour market dualism, the reform of VET policy can be expected to remain a futile endeavour.

Key Words:

• Dualisation
• hierarchical capitalism
• South Korea
• skills
• vocational education and training

South Korea’s export-oriented industrial strategy was a huge economic success. From the impoverishment of the Korean War (1950–1953), the country developed into one of the world’s leading economies, joining the club of rich nations, the Organisation for Economic Co-operation and Development (OECD), in 1996. Showing a remarkable “catch up” with advanced economies, Korea achieved, between 1960 and 2000, an average annual economic growth rate of 7.9%, with the greatest growth in the 1980s when an average growth rate of 9.3% was achieved (Lee 2016, 3–4). This economic performance is widely seen as resting on a “virtuous cycle … in which education and growth reinforced each other” (Cheon 2014, 209). In this era, vocational education and training (VET) and collective skills formation were promoted by government to drive the country’s late industrialisation. This allowed for remarkably egalitarian social outcomes, despite the absence of meaningful social protection mechanisms; until the 1997 Asian Economic Crisis, the country experienced a long-term decline in social inequality (Park 2010, 588–589).

Since 1997, Korea’s economic model has been showing “cracks.” Unsurprisingly, economic growth slowed down, as the country’s economy “matured.” After the Crisis, Korea achieved an average economic growth rate of 4.1% between 2000 and 2010 (Lee 2016, 4). Between 2010 and 2019, it declined to 3.3% (calculated using OECD.Stat). Although this is by no means a poor economic performance, concerns have been raised about Korea’s long-term economic prospects. The OECD (2013, 22), for instance, notes that recent economic growth relied primarily on increasing labour input, which is thought to hold back economic progress. Such a strategy, of course, has obvious limits in an ageing society such as Korea – in fact, the country is the fastest ageing society in the OECD with a projected old-age dependency ratio of around 70% in 2050 (OECD 2018b, 11). Yet, despite huge investments in education, Korea presents poor labour productivity by international standards, in particular in small- and medium-sized enterprises (SMEs) and in the service sector (OECD 2015, 37). Collective skills formation has been undermined, and commentators point out rising skills mismatch due to a lack of intermediate and vocational skills (see, for example, Park 2011). At the same time, there has been an over-investment in private tutoring and a higher education sector which struggles to supply the labour market with the “right” skills (Durazzi 2019, 1809; Fleckenstein and Lee 2019b, 181). In addition to these economic challenges, Korea faces a social inclusion challenge with sharp rises in inequality in an ever more dualised labour market after the Asian Crisis (Byeon et al. 2017, 8–9).

The Korean government, regardless of party-political orientation, has recognised these challenges – the need to improve labour productivity and social inclusion. In fact, strengthening vocational training is widely perceived to offer a “double dividend” of economic and social progress, especially by promoting the social inclusion of labour market outsiders. However, despite this recognition and considerable policy activism, the reform of VET remained largely unsuccessful. Korea continues to trail behind most European Union (EU) countries in training efforts, both in terms of training time per year and expenditure per head: 6.5 hours compared to the EU average of 10 hours, and €56 compared to €234 (Ban et al. 2018). The result is that the Korean economy continues to suffer from skills mismatch and poor labour productivity (OECD 2020b, 31–33).

What explains the decline of collective skills formation and the failure to reinvigorate VET in Korea? This article provides evidence of a segmentalist coalition of government and employers, with big business in the “driving seat.” Prioritising large employers’ interests at the expense of collective ones, this coalition dismantled collective vocational skills formation in the 1990s, following the country’s democratisation.¹ When government – recognising the shortcomings of the new skills formation regime – attempted to strengthen VET, it faced large employers that were not prepared to collaborate with government in a meaningful manner, and SMEs with little or no training capacity. In addition to extending the Europe-focused research on segmentalism in VET to a critical case in the study of East Asian capitalism where Korea is considered the “purest” example of the so-called Developmental State that characterises late industrialisation in the region (see Ringen et al. 2011), this article shows that business opposition to collective skills formation has its micro-foundations in Korea’s hierarchical production regime and dualised economy. The country’s economy is dominated by a small number of family-run business conglomerates (so-called chaebols), and they exercise considerable socio-economic and socio-political power in the Korean political economy. Big business has established extremely hierarchical inter-firm relations, and these allow them to exploit SMEs with detrimental effects for skills and labour productivity. In this context, large employers have no incentives to reconsider their business and skills strategies, and their extremely privileged position allows them to not engage with government initiatives for collective skills formation – effectively undermining any prospect of reforms that could have promoted economic and social progress. Also, fragmented business organisations associated with Korea’s hierarchical inter-firm relations undermine the overcoming of collective action problems in skills formation. Korean employers’ associations do not display the institutional capacity we know historically from co-ordinated market economies (CMEs). In other words, the inter-firm co-operation that Emmenegger and Seitzl (2019) identify as critical for collective skills formation is not observed in Korea; rather the opposite is true. Whilst sharing many similarities with the Japanese case where decisive government intervention was also critical in late industrialisation, it is important to note differences. Japan’s VET has been characterised by long-standing segmentalism in training and a corresponding firm-specific skills formation system (Thelen 2009, 480). In contrast, Korea’s Developmental State promoted collective, industry-specific skills formation that is associated with the country’s economic success story (Cheon 2014; Fleckenstein and Lee 2019b; Park 2013).

The article is structured as follows: first it reviews the rise and fall of developmental skills formation during Korea’s successful late industrialisation. It is shown how strong government, in an undemocratic context, was first able to impose its development strategy, which included the strengthening of vocational skills. However, large employers, as their economic and political power increased, displayed growing resistance to government’s expectations, and this started undermining collective skills formation in the 1980s. Their resistance turned into outright opposition after democratisation, and evidence is provided that large employers were at the heart of the segmentalist coalition that effectively terminated collective skills formation. This is followed by introducing government’s attempts to reinvigorate vocational skills, and an analysis of the micro-foundations of Korea’s vocational training policy showing that the country’s hierarchical capitalism made VET reform a futile endeavour.

Empirically, this article employs documentary analysis, using official reports, policy papers, and press releases to identify the positions and reasoning of government, employers, and trade unions. It also draws on 22 semi-structured interviews with carefully selected key stakeholders, including representatives from employers’ associations, trade unions, and public organisations (see the List of Interviewees in the Appendix). These interviews provide additional, in-depth insights into the micro-foundations of VET policy; in particular, the failure to engage employers, but also the lack of trade union interest in training and skills. With a focus on employers and their production strategies, 15 out of 22 nterviews are with employer representatives; and these come from both national and regional associations and chambers, in addition to interviewees representing different manufacturing and service sectors as well as firms of different sizes. Interviews with labour representatives allow both a capturing of workers’ perspective on skills and training and more generally their perception on employment relations in Korea’s increasingly dualised labour market. Lastly, public officials, with two coming from research organisations, can be considered as providing less biased assessments with some distance from workplaces and their conflicts between capital and labour. That said, it is recognised that all interviewees have conscious and unconscious biases. This underlines the critical importance of triangulation in qualitative social research, and thus this research draws on multiple methods and data sources for a comprehensive understanding of Korean VET policy.

The Rise and Fall of Developmental Skills Formation

Collective skills formation in Korea began in the 1960s under the authoritarian regime of Park Chung-Hee as part of the first five-year economic plan (1962 − 1966). The Economic Planning Board, the pilot agency that laid out development plans, projected a shortage of 165,000 skilled workers during the second five-year economic plan (1967−1971) period (Ministry of Employment and Labour and KRIVET 2012, 27). In the first instance, the government introduced vocational training legislation that subsidised employers’ workplace training. This led to an increase in the number of vocational trainees from 10,738 in 1967 to 30,558 in 1970, due to a rise in public training (from 1,502 to 11,840) and workplace training (from 3,890 to 13,483) while other forms of training declined in the same period. However, the employer subsidies were not deemed sustainable because of the pressure they put on public finances, which were under considerable strain at the early stages of industrialisation. When subsidies were withdrawn in 1972, workplace training dropped considerably (Ministry of Employment and Labour and KRIVET 2012, 30-31). This meant that a voluntary system relying on employers was not feasible, as it failed to provide a skilled workforce large enough to support the government’s Heavy-Chemical Industry (HCI) Drive industrialisation project, which was key to establish an export-oriented growth regime. With enhanced power after the 1971 presidential election, President Park Chung-Hee centralised decision-making for long-term economic planning. This placed the government in a position to establish, as part of the third economic plan (1972–1976), a compulsory training system that required large employers to provide workplace training; employers who did not meet expected levels of training provision had to pay a training levy, which was then used to fund the government’s vocational training provision. Also, assistance from the Asian Development Bank and the International Bank for Reconstruction and Development supported the government’s establishment of public vocational training schools to complement employer workplace provision. The introduction of the training levy translated into an extraordinary growth of workplace training: the number of workplace trainees rose from 10,799 in 1972 to 90,992 in 1979, which was more than three times the number of trainees in public training (28,488). Importantly, the compulsory training system expected large employers to train beyond their own skills needs; and “excess” skills were absorbed by SMEs, which lacked the resources for training in the workplace (Ministry of Employment and Labour and KRIVET 2012, 32–33).

In 1979, President Park was assassinated, but the authoritarian regime continued when General Chun Doo-Hwan seized control of the country by military coup. The economy experienced stagflation, a result of the Second Oil Shock (1978 − 1979) and over-investment in the HCI during the 1970s. Economic output dropped by 3.3% and inflation reached nearly 30% in 1980, putting enormous political pressure on the new government that needed to consolidate its rule. To cope with the crisis, the Chun government (1980−1988) implemented an Economic Stabilisation Policy through the fifth economic plan (1982 − 1986), with economic liberalisation policies that were aimed at moving the state-led economy to a more market-driven one, including the reduction of government spending, the relaxation of tight credit controls, the privatisation of state-owned commercial banks and partial permission for foreign investment in the stock market (see Haggard and Moon 1990). It is important to acknowledge that these economic reforms were not driven by employers, but by elite bureaucrats. In fact, chaebols and “their” Ministry of Commerce and Industry initially opposed economic liberalisation, as big business feared it might challenge the cosy relationship it had with the authoritarian state. Instead, the all-powerful Economic Planning Board and its think tank, the Korea Development Institute, led by neo-liberal economists who had earned their PhDs from US universities, were at the heart of the government’s liberalisation project. Critically, these deregulation measures undermined the dominant role of the state in economic governance (see Kim 1999; Yang 2017). Chaebols, which had grown dramatically through the HCI in the 1970s, had become influential in the economy, and they expressed their views more strongly than in the past. In this context, it is noteworthy that the Federation of Korean Industries (FKI), which represents big business, established a newspaper, the Korea Economic Daily, and a think tank, the Korea Economic Research Institute (KERI). These were important means for articulating chaebols’ interests in economic policy-making (Federation of Korean Industry 2011, 212–213, 429; see also Lee 2001).

The production regime incrementally changed from a labour-intensive mass production system to an increasingly capital-intensive and post-Fordist one (KRIVET 1998, 77; Ministry of Labour 2006, 93). Large employers began to recognise the importance of enhancing the skills of their existing workforce and prioritised firm-specific skills, whereas the government wanted large-scale industry-specific skills formation for new employees (Ministry of Labour 2006, 256). In the face of changing skills requirements, big business moved away from collective skills formation, prompting an emerging conflict between government and large employers. Rather than providing training for labour market entrants, an increasing number of companies chose to pay the training levy. In 1978, less than one third of employers paid the levy, but the proportion gradually rose to two thirds by 1986 (Lee 2005, 181; Ministry of Employment and Labour and KRIVET 2012, 43). In addition, large employers increased their pressure on the government to reform the levy system. Not only did they, represented by the FKI, call for reducing the number of labour market entrants they had to train, they also pushed for the upskilling of existing workers to be counted towards their quota (see Lee 1992; Park 1992). The Chun government responded by reforming the compulsory training system to the satisfaction of large employers (KRIVET 1998, 987–980; Ministry of Labour 2006, 252). The result was greater big company investment in training facilities to improve the firm-specific skills of incumbent workers (Interviews 9, 18). For example, in the steel industry, Posco, then one of the world’s largest steel producers, founded its own vocational training school, whereas Hyundai Steel established close relationships with a local school and university (Interview 16). In the electronics industry, large firms created their own training systems, being reluctant to share training-related information with their industry association in fear of potentially disclosing valuable knowledge to competitors in their industry (Interview 5).

Assessing the skills formation regime in this period, it is found that the Park government had established collective skills formation in a top-down manner, to which large employers only consented reluctantly. Business co-operated, but remained, in principle, sceptical and did not develop a genuine commitment to vocational training in the workplace. When the authority of the Developmental State declined and big business gained greater strength (as a direct result of the government’s industrialisation strategy), less compliance is observed in the 1980s with large employers increasingly withdrawing from industry-specific workplace training – even if that meant considerable financial penalties. So even before democratisation, the balance of power in the so-called Developmental Alliance between the authoritarian state and big business started shifting, indicating the limits of government in economic governance (Hundt 2009; see also Kang 2010). With large employers’ reduced engagement in skills formation, SMEs could draw ever less on “excess” skills from large workplaces and became more reliant on public vocational training schools, which were thought to produce poor skills; this caused a growing skills schism between big firms and SMEs (Ashton et al. 2002). Meanwhile, trade unions were not involved in the training system, as the authoritarian government, fearing that unions could grow into a political opposition, supressed them (Deyo 1987, 185). Under the authoritarian regime, workers were subject to military discipline on the shop floor, with large employers taking advantage of the state’s intervention in industrial relations to keep wages low and workers compliant (Interviews 1, 2, 22). Overall, with reluctantly consenting employers and suppressed unions, collective skills formation in Korea did not rest on a broad cross-class coalition as in the case of Germany, for instance, the archetypical CME (see Hall and Soskice 2001; Thelen 2004).

Skills Formation after Democratisation

Following democratisation in the late 1980s, large employers’ lack of enthusiasm for the government’s training policy turned into outright opposition with business campaigning for deregulation, including the abolition of the training levy. The Korea Employers’ Federation (KEF) argued the compulsory training policy was “outdated,” as it was introduced in the initial stage of industrialisation and “focuses on the training of simple manual jobs.” Instead, they advocated “autonomous and independent in-house job training” to enhance competitiveness and to facilitate the transition to high-value-added industries (KEF 1994, 235–236). The developmental skills formation regime came to an end when the conservative Kim Young-Sam government (1993–1998) abolished the training levy and, more generally, moved the state-led system to a voluntaristic one. These changes need to be seen in the context of a wider liberalisation strategy, which the government perceived as an imperative to cope with the economic pressures of globalisation. For instance, the government also deregulated the higher education sector to increase the supply of general skills for the knowledge economy, in addition to attempts to deregulate the labour market (see Kim and Lee 2006; Kong 2000).

When establishing the Employment Insurance System (EIS), the vocational training system was integrated into the new insurance scheme and its Job Skills Development Programme, which reimbursed employers the costs of employee training. As the coverage of EIS was extended to all workers in 1998, so was the coverage of the Job Skills Development Programme. To promote more market-led training provision, the government also privatised public training centres run by the Korea Vocational Training Management Corporation (a public organisation under the Ministry of Labour) by transferring the Corporation to the Korea Chamber of Commerce and Industry (KCCI) in 1993 (Ministry of Labour 2006, 267). For SMEs with limited capacity to train their own employees, the government allowed for-profit corporations to provide training services (Interview 6). When the government introduced the EIS, FKI argued for abolishing the training levy – essentially a tax in their view – as the employer contribution to the new insurance would “double” the employers’ training-related tax burden (Korea Economic Daily, January 21, 1997). KEF also called for an end to the training levy and made the case that “the government’s job training policy should only be applied to groups that the private market cannot support, such as the unemployed, disabled, elderly, and unemployed high-school graduates, funded by the government’s budget” (KEF 1997, 165). Even more aggressively, KERI argued for the government to withdraw from vocational training altogether (Choi 1997, 55). In the face of this pressure, the government eventually abolished the training levy.

Thus, in training policy, a new segmentalist coalition of the state and large employers emerged with big business having the upper hand – quite different from the previous Developmental Alliance, in which the state dominated. Critically, in democratic Korea, the conservative party relied hugely on donations from business to run its political campaigns, meaning it could not afford to upset business and chaebols in particular (Mo 2001, 218). By contrast, under conservative governments, trade unions had no meaningful role in training policy. That said, training did not feature highly on unions’ agendas (Yoon and Lee 2009, 165); instead, they focussed on wages, benefits and working conditions. Interviewees from Industry and Regional Skills Councils described unions as “not interested in training” (Interviews 5, 9, 14), “silent” (Interviews 8, 16) or “not proactive” on training issues (Interview 10), as training was not linked to the wage system and thus the remuneration of their members (Interviews 5, 10, 16). One interviewee suggested unions treated training and skills as “something employers should take care of” (Interview 11); and even representatives of both union federations, the Korean Confederation of Trade Unions (KCTU) and the Federation of Korean Trade Unions (FKTU), acknowledged the low priority of training within organised labour (Interviews 2, 20).

The absence of unions in training policy is not to argue that labour did not gain greater importance after democratisation – the opposite is true. In the Great Workers’ Struggle of 1987 and the general strike in 1996 opposing the government’s attempt to impose labour market deregulation, organised labour demonstrated that unions and workers flexing their muscles could make a considerable difference. However, the dominance of enterprise unions promoted a focus on wage increases and enterprise welfare in large workplaces (see Fleckenstein and Lee 2019a; Yang 2006). Thus, unlike the German experience (Busemeyer 2012), there was no coalition of “conservative” employers and trade unions that attempted to oppose the segmentalism pushed by large employers. SMEs did not have the capacity to challenge big business, and unions did not have an interest in skills and training. Obviously, unlike the Swiss experience described by Emmenegger and Seitzl (2019), large employers did not need to reach out to SMEs to form a broad business coalition to drive the reform of vocational training. Even more so than in the case of Germany, Korean chaebols – typically with strong export-oriented business divisions – were able to use their economic power to pursue their first-order preference of deregulating VET without compromise.

In response to higher labour costs and more frequent industrial actions, employers also developed new management and production strategies. They strengthened the line management on the shop floor and pursued company “culture campaigns” for better control of their workers, in addition to investing in automation, offshoring, and outsourcing. This allowed reducing reliance on “costly” insiders, but massively increased the use of irregular workers (Kim 2018; Yang 2017). Interviews (2, 15, 20) highlight large employers’ technology-focussed strategies that aimed at reducing their reliance on insiders through automation, including attempts to make production less reliant on “workers” but more “engineers-centred” to reduce unions’ disruptive potential (Interview 1). Also, research by a government think tank, the Korea Research Institute for Vocational Education and Training (KRIVET), found that employers prioritised automation and reducing their workforce over skills investment; as one human resource manager stated: “CEOs do not think that training makes a difference. And [they think] work is done by machines” (cited in Ban et al. 2018, 190). As a result, Korean manufacturing has an extraordinary level of automation with 855 industrial robots per 10,000 employees. For comparison, in the manufacturing powerhouses of Japan and Germany, the figures are just 364 and 346 robots, respectively, whilst the USA has a mere 228 industrial robots per 10,000 employees in the manufacturing sector (International Federation of Robotics 2020, 17).

The technology-focused “labour-saving” strategy – especially in knowledge-based, high-tech industries – accelerated the shift from firm-specific skills to general skills formation. In order to support rapid innovation in the face of growing global competition, not only did large employers invest in research and development, they also focused more on core workers with high levels of general skills, who would be recruited from the external labour market or universities (Jung, Lim, and In 2005; Lee 2011a). In addition to upgrading their in-house training centres, large employers such as Samsung, Hyundai, LG, and Daewoo, established their own corporate universities or MBA courses in close collaboration with universities to meet their skills requirements (Choi 1997, 44). Also, the investment in automation is seen as reducing the reliance on firm-specific skills with robots gaining ever greater importance in the manufacturing process (Park 1990, 139). As in training policy making, unions remained excluded from corporate decision-making with its profound implications for corporate skills formation systems (Park 1990, 130; Interview 1).

Large employers’ withdrawal from collective skills formation in the workplace had significant consequences for SMEs, especially in manufacturing, which relied more than ever on vocational schools and private training centres to supply them with young workers with industry-specific skills. However, private training corporations under the EIS tended to focus on general skills, which rather suited the service sector with government approving, for instance, new vocational schools for English-language training, cooking, hairdressing, and IT (Interview 6). Moreover, SMEs struggled with the largely general skills supplied by the growing higher education sector – these are of low utility in the SME sector, which requires industry-specific skills.

Re-inventing Collective Skills Formation: The Failure to Engage Employers

Acknowledging the huge training challenges SMEs faced and also recognising poor labour productivity and skills mismatch, the centre-left Kim Dae-Jung government (1998–2003), in 2001, established the Vocational Training Consortium Programme for SMEs. This high-profile programme provided grants for setting up training partnerships between SMEs and large enterprises, employers’ organisations or universities. SMEs are given access to existing training resources of their partners (typically the large enterprises whose supply chains the SMEs are part of). It is expected that large enterprises and SMEs design training programmes together, and the staff of large enterprises deliver training in their facilities (Ministry of Employment and Labour and KRIVET 2012, 52–53). The grants cover 80% or more of the costs, and up to 2 billion won (approximately, $1.75 million) for six years per consortium. The number of consortium training centres increased rapidly, numbering 142 in 2012, and these centres train an average of about 180,000 workers each year (Ministry of Employment and Labour 2021; The Republic of Korea 2021, 3). Despite expansion of the programme, its overall impact on training in the SME sector has been modest. In particular, the programme failed to reach firms with 50 or fewer employees that typically struggle to train their workers (Kim et al. 2007, 171). As a result, Korea still has the second largest training gap between small and large companies in the OECD, and with nearly 40% well above the OECD average of 25%; also only 30% of small firms with 10 or fewer employees engage in any training (OECD 2020a, 62).

Whilst continuing its support for the training consortia, the conservative Lee Myung-Bak government (2008–2013) introduced so-called “Meister Schools” as its flagship programme in vocational training policy. These state schools, drawing on Germany’s master craftsperson training, are thought to provide “industry-customized education” (Park and Chung 2013, 49). They aim at supplying the labour market with high-level specific skills as required by strategic industries. Meister Schools are given considerable autonomy in designing and developing curricula based on the needs of local industries, and in hiring experienced CEOs as principals and industry experts as teachers. Also, recognising the country’s pre-occupation with academic qualifications, it was the government’s objective to raise the status of vocational skills and make the vocational training track more attractive with Meister Schools. To recruit high-calibre students, incentives are provided, including job guarantees before graduation, free tuition and accommodation, and the postponement of compulsory military service. Graduates from Meister Schools can apply for university and a special route has been established for them, which increases their chances of getting into prestigious universities; but a minimum of three years of employment is required before higher education can be pursued (Park and Chung 2013, 46–52). Meister Schools enjoyed much prominence in the Lee government, and they have continued to expand after the Lee government with their number increasing from 21 schools in 2010 to 51 in 2020. Despite the quantitative expansion, however, the initiative has only had a rather limited impact, as only 1.3% of high-school students and 6.5% of vocational high-school students chose to attend Meister Schools in 2019 (Yu et al. 2020, 15). Moreover, according to a survey, still three out of four graduates of these schools wish to pursue tertiary education, suggesting that the initiative has not changed the culture of seeing a university degree as a “must” (Kim et al. 2015, 60).

Vocational training continued to enjoy a high profile in the conservative successor government of Park Geun-Hye (2013–2017). To address the mismatch between skills taught and skills needed in industry and to reduce youth unemployment, the Park government launched the so-called “Dual Work-Study Programme,” inspired by the German apprenticeship system. Companies hire apprentices and provide both on-the-job and off-the-job training, while vocational schools or junior colleges provide complimentary theoretical education. The duration of these apprenticeships ranges from six months to four years, with an average of 17 months. Upon completion of the programme, trainees receive a skills competency certificate approved by the government (Yoon, Hong, and Bae 2017, 69, 75, 89; Na 2013, 114–115). By 2019, almost 90,000 people participated in the Dual Work-Study Programme (Park et al. 2019, xiii), and, six months after programme completion, three out of four apprentices were employed at the companies that trained them (Yoon, Hong, and Bae 2017, 89). Despite an impressive quantitative expansion of the programme, however, the quality of training has been questioned given that 95% of the firms participating are small, which typically struggle to provide high-quality training. This has fuelled concerns that the programme subsidises labour costs in hard-pressed small companies rather than supporting genuine skills formation. Also, more than three-quarters of apprenticeships have a duration of only one year, being no more than one-third of the typical duration of a German apprenticeship. Such short duration is likely to compromise skills formation and undermine the consolidation of acquired skills (Chun and Lee 2015, 8, 22). The dropout rate is also fairly high, with nearly one-third of participants not completing the programme, raising the question as to whether these apprenticeships are perceived as offering promising career prospects (Kang 2016, 60).

Another innovation aimed at promoting greater co-operation between employers was the creation of Industry Skills Councils (ISCs) and Regional Skills Councils (RSCs), building on pilot schemes that date back to the Roh Moo-Hyun government (2003–2008). These councils were meant to be employer-led to better meet their needs. This initiative was informed by similar practices in liberal market economies, such as the UK, Canada, and Australia (Choi 2006). Eleven ISCs were established in 2015, and they expanded to 17 by 2018. The role of ISCs in vocational training is primarily that of a facilitator through the collection of skills-related information with the objective of identifying skills and workforce needs in their industries, and also the ambition to promote collaboration between companies (Interviews 5, 7, 8). To support the Park government’s National Competency System, ISCs have been furthermore tasked with providing skills and task knowledge needed for developing this new system, in addition to identifying companies for the Dual Work-Study Programme. Also, they have been providing some training, mostly in collaboration with RSCs. In 2018, 50 training programmes were developed through such collaboration (Industrial Skills Council 2018, 19). Yet, it remains doubtful that these councils play a meaningful role in vocational training. According to a survey among companies in the automobile, electronics, and shipbuilding industries, when asked whether sector councils contribute to skills formation, 60% of respondents answered, “neither yes nor no” (Yun, Kim, and Chung 2008, 244).

The government’s failures to develop an adequate vocational skills formation regime and to engage employers in that regime have their micro-foundations in the country’s extremely hierarchical production regime and the related dualism in the economy and labour market. Whilst hierarchical inter-firm relations have their historical roots in the HCI project of the 1970s that produced considerable economic concentration (promoted by the authoritarian regime), socio-economic changes since democratisation and especially the Asian Crisis have exacerbated the polarisation between large employers and SMEs, as well as the dualism between labour market insiders and outsiders.

Micro-Foundations of VET Policy I: Hierarchical Capitalism

In Korea’s industrialisation project, the state, with economic bureaucrats driving policy, was the critical facilitator – not only through the HCI drive but also because of its support for large companies that were “hand-picked” by the government to become “national champions” that allowed Korea to break into global markets. A focus on export was at the heart of the government’s growth strategy and favourable treatment, most notably preferential credit allocation through state-owned commercial banks was granted to those large companies that achieved the government’s export goals, and this in turn promoted the growth of these firms. This process was one of the foundations for Korea’s family-run business conglomerates (Amsden 1989; Woo-Cumings 1999). In its industrialisation strategy in the 1970s, the government effectively vertically integrated SMEs into large companies through sub-contracting arrangements. This was legislated in the 1975 Act on the Promotion of Co-operation between Large Enterprises and SMEs. Specifically, the main contractor (mostly chaebols) assigns the manufacture of parts to sub-contractors (mostly SMEs), in which sub-contractors typically enter an exclusive contract with the main contractor. If sub-contractors bring in new clients, this is widely regarded as a breach of contract (Interview 14). This relationship creates an extraordinary power asymmetry and skews the business relationship heavily in favour of the large companies. This practice in Korea is different from outsourcing where sub-contractors offer their product or service in the wider marketplace and engage with numerous firms. Although legislation recently banned such exclusive contracts, the practice persists due to lack of enforcement (The Hankook Daily, November 29, 2018; The Korea Industry Daily, January 30, 2020). The growth of chaebols promoted by government not only shifted the balance of power between large businesses and SMEs, but also between the state and business conglomerates – with the government’s co-ordinating capacity diminishing.

This co-ordination is different from the industry-based co-ordination in European CMEs, and Hall and Soskice (2001, 34) introduce the distinction of group-based co-ordination in light of the importance of business conglomerates in Japan and Korea, where we observe “sharp competition between companies in the same industry” rather than collaboration within industry sectors. However, even this image of East Asian political economies fails to fully capture the extremely hierarchical inter-firm relations in Korea. Drawing on the Latin American situation, Schneider’s (2009) concept of hierarchical market economies has relevance in the Korean context too. Certainly, there are differences between Latin America and East Asia, as Schneider points out, but not along all analytical dimensions of hierarchical market economies. In particular, the key features of family-run, diversified business conglomerates in Latin America have great similarity with the Korean situation, including unequal inter-firm relations “imbued with a hint of coercive hierarchy” (Schneider 2009, 560). Importantly, they also share an absence of institutions promoting employment relations that facilitate investment into skills and training. As already discussed, it was the power of the authoritarian state that allowed the emergence of the developmental skills formation regime and ensured that large employers shouldered the financial burden. But when state power dwindled in post-democratisation Korea, big business was quick to call for a more liberal VET policy, giving rise to a segmentalist coalition between big business and state that dismantled the developmental skills formation regime.

Evidence from interviews sheds light on how the widespread practice of sub-contracting in the hierarchical production regime has provided the micro-foundations for the demise of the previous regime and continued failure in the government’s effort to revive VET. Main contractors often exploit their sub-contractors, using the power asymmetry they gain from enforced exclusive contractual relationships. They demand confidential business information from their sub-contractors, including production costs and profits; and they use this knowledge to pressure sub-contractors into lowering prices. This allows main contractors to relax their efforts on training and skills development, as it is much easier for them to increase profits by “squeezing” their sub-contractors (Interviews 5, 20). Obviously, this leaves little incentive for SMEs to engage in training that might improve their productivity and profits, as any gains might be quickly absorbed by the main contractor:

For example, in the automobile industry, Hyundai Motors and Hyundai Mobis [Hyundai Motors’ parts producer] know the detailed financial structure of their sub-contractors like the back of their hand – even all production costs including labour costs. In this situation, sub-contractors do not have any incentive to increase their costs by investing in training (Interview 13).

This situation is at odds with arguments that Korean SMEs have flourished when large Korean firms, with support from government, reduced their dependence on foreign suppliers (Kim and Kwon 2017; Kim 2019). Admittedly, the SME sector has grown and some SMEs, mostly “first-tier” sub-contractors of chaebols, have benefited more than others in this process. However, this should not deflect from the primarily exploitative nature of “Korean-style” sub-contracting. This sub-contracting style is also used abroad, confirmed in a memoir by a former Hyundai Motors president, who states that working with Korean suppliers overseas is much preferred, as the company knows “the costs of producing parts very well, and thus, we are more likely to avoid price gouging by foreign parts suppliers” (cited in Kim and Kwon 2017, 524; see also Kim 2011). Foreign suppliers, producing for multiple clients, are unlikely to be forced into sharing confidential business information with their Korean contractors. Rather than being squeezed, these firms, notably Japanese ones, took advantage of Korean companies’ dependence on their products. Even Kim and Kwon (2017, 525) confirm that Korean companies are reluctant to do business with global suppliers that also work with foreign competitors, as this is believed to offer no competitive advantage; and instead, they take their Korean suppliers abroad to press prices. SMEs, even when offshoring does not offer any prospect of better contract terms or profits, cannot refuse their contractors’ demand to move abroad in fear of losing their entire business. Of course, those SMEs remaining in Korea are no less squeezed – an interviewee elaborates, with reference to the automobile industry, that “employees of third and fourth-tier sub-contractors are essentially only paid minimum wage” (Interview 8). In short, aggressive labour cost saving for price competitiveness is at the heart of non-collaborative sub-contracting practices in Korea and abroad, and this leaves very little room for training and skills to improve labour productivity.

In this business environment, SMEs, on tight profit margins, fear disruptions in production that would be caused by training – temporary cover for staff on training is either difficult to recruit or simply unaffordable (Interviews 3, 4, 8, 10, 13, 14, 15, 16, 20). Even if SMEs were keen on improving the skills profile of their workforce, they typically do not have the capacity for providing high-quality training. Whilst large employers have their own human resources departments or training centres, SMEs lack the personnel and resources for training (Interviews 4, 5, 7, 9, 13, 14, 17). This observation is confirmed by employer surveys: KRIVET’s Human Capital Corporate Panel and Korea Labor Institute’s Workplace Panel Survey show the differences in personnel and resources for training by firm size: in 2015 only 12% of SMEs with 30 − 99 employees had dedicated staff in charge of vocational training, while 58.1% of large employers with 300−999 employees and 88% of firms with more than 1,000 employees, have their own training division. In terms of training facilities, 86% of SMEs have no facilities, but 57% of firms with more than 2,000 workers have their own training facilities (Hwang et al. 2016, 175; Kim et al. 2018, 100). These differences in institutional capability are reflected in huge differentials in training efforts: firms with more than 1,000 workers invest nearly 15 times the resources of firms with 1–29 employees and more than seven times that of companies with 100–299 employees (Ministry of Employment and Labour 2018, 13). In the past, some SMEs could have relied on training provided by their main contractor. In shipbuilding, for instance, one interviewee reports that large shipyards have typically around 60% of workers coming from sub-contractors, but these workers are fully integrated in the shipyard’s main workforce. Sub-contractors offered very little training but relied on training offered by the shipyard’s training centre. Whilst every large shipyard has its own training centre, and thousands of workers were trained on-site every year, this effort has declined, in recent years, to a fraction of workers, leaving sub-contractors with significant skills gaps (Interview 18).

Furthermore, the extreme hierarchy in the business community undermines efforts to promote collective skills formation through ISCs and RSCs, which were established with the intention to promote co-ordination and co-operation between firms. Main contractors relying on firm-specific or high specialised skills show little willingness to co-operate with others within their industry, because they do not perceive any business interests in promoting skills beyond their company (Interview 22). For instance, interviewees of ISCs and RSCs report that main contractors do not share any information on future workforce needs or their job competency system, which they fear would expose confidential information to competitors within their industry (Interviews 5, 7, 15). At the same time, sub-contractors cannot engage with each other to collectively develop training systems. An RSC interviewee of an area well-known for manufacturing industries with extensive sub-contracting explains:

The rigid sub-contracting structure hugely undermines the autonomy of sub-contractors. For instance, they cannot afford not to please their contractors, and they must prioritise meeting the demands of contractors, … including the reduction of their unit price at the expense of investing into training for their future. At our council meetings, sub-contractors do not say a word if anyone from their main contractor is present (Interview 8).

The hierarchical business structure, in addition to the legacy of the government-led industrialisation of the Developmental State, has a similar effect on employers’ associations as a possible vehicle for co-ordination between companies in skills formation – an area where Korean intermediary organisations display little, if any, strategic capacity to overcome the challenges of co-ordination in skills formation. At the peak level, employers are fragmented into five associations: FKI, KEF, KCCI, the Korea Federation of Small and Medium Businesses (KBIZ), and Korea International Trade Association (KITA).

Whilst FKI and KEF are powerful actors in economic and social policy-making, they fail to encompass and mobilise a wide range of employers, but focus firmly on the interests of large employers and especially chaebols, which founded the two federations as voluntary organisations. Fully funded through membership contributions, FKI and KEF enjoy a great degree of autonomy from the government (Interview 4). However, they show little interest in training as their members already have sufficient training capacity, and thus do not rely on the state for training: “KEF, at the national level, speaks mostly for large companies. Because we [KEF] think that large companies can invest in training on their own, we have very little interest in training” (Interview 4; see also Interview No. 18).

By contrast, KCCI, KITA and KBIZ lack political clout and autonomy. Re-established as statutory bodies by law during the authoritarian Park government – KCCI and KITA had existed since 1945, and KBIZ since 1962 – they depend financially on government-imposed membership fees, government-funded projects, and/or government subsidies (Interviews 6, 22). For example, government subsidies account for 50.6% of KBIZ income, whereas government-independent membership fees make up a mere 4.3% of its revenue (KBIZ 2020, 50). This financial dependence on government undermines their capacity to formulate a collective strategy for their members, as they are incentivised to prioritise government interests rather than those of their members (Lee 2001). An interviewee representing SMEs describes their situation:

Because we [KBIZ] are subsidised, as much as half of the operating costs by the Ministry of SMEs and Start-Ups, government control and supervision are severe … Because KBIZ is a special corporation established by law, we cannot act as autonomously as FKI … If KBIZ becomes less reliant on subsidies or if subsidies are not jeopardised when we disagree with the government, we can, I think, represent more freely the interests of SMEs (Interview 17).

This dependence on government funding is also found in skills councils, which are supposed to be employer-led. A representative from an RSC explains that, at first, it relied exclusively on financial support from the Ministry of Labour, and about 80% of their budget continues to come from government (Interview 14).

The lack of strategic capacity of employers’ associations is even worse at the sectoral level. Historically, the authoritarian government developed and controlled industry sectors by legislating the establishment of industry business associations during the 1960s and 1970s (Interview 22; see also Kim 2003; Park et al. 2016). Six out of seven sectoral associations interviewed for this research are statutory organisations. Instead of membership fees, they almost exclusively rely on earnings from government commissioned projects, such as the certification of qualifications. For this reason, they tend to follow government views rather than challenge the government and speak out for their members (Interviews 1, 3, 4, 5, 6, 12, 19, 22). An expert from a government think tank sums up the dilemma: “employers’ associations … were ‘lower delivery systems’ of the developmental state for allocating funds for export support, or the import of raw material. They were not true member associations, but just delivery systems from the top” (Interview 1). The weak strategic capacity of employers’ associations hinders the development of collaborative training systems; especially industry-specific skills, which are important to SMEs. They typically pay no meaningful attention to training; much less important than, for instance, promoting their industry (Interviews 5, 11) or “income-generating business” to cover their operating expenses (Interviews 7, 11).

To conclude, the coercive and exploitive relations between large enterprises and SMEs, shaped by the hierarchical production regime, have created micro-foundations that are incompatible with a collective skills formation regime. Big businesses have in-house training systems to produce the firm-specific skills they (still) need, and they show little interest in investing in industry-specific skills from which their sub-contractors could benefit. SMEs have meagre resources and/or lack the incentives to train workers under the rigid hierarchical production regime in which higher labour productivity and profit are likely to only lead to price “squeeze” from the contractor. Organised business, reflecting the hierarchical nature of the production regime, is fragmented and lacks strategic capacity to promote a collective skills formation system.

The impact of hierarchical capitalism on Korea’s VET policy does not stop here. The next section shows that the hierarchical production regime has driven labour market dualisation that prevents both employers and employees from developing an interest in greater skills investment.

Micro-Foundations of VET Policy II: Labour Market Dualisation

The hierarchy in the production regime and inter-firm relations have huge impacts on the labour market and especially the dualisation between insiders and outsiders. Whilst the labour market was effectively dualised by the HCI projects of the 1970s, and more generally the government’s ambition to create “national champions” that could compete internationally, the authoritarian state’s wage guidelines, “enforced through the web of government, labor and industrial relations” (Lindner 1984, 6), had previously facilitated considerable social cohesion. However, wage inequality soon began to increase, with SMEs’ relative wage to that of large employers dropping from about 90% in 1980 to about 60% in the 2000s (Korea Labor Institute 2009, 80). In the face of pressure for higher wages and occupational benefits from often militant enterprise unions since democratisation, large employers focus on core workers, providing employment security, seniority wages, and firm-based training. At the same time, these firms have made increasing use of irregular workers. Unsurprisingly, they also pushed for labour market deregulation, which allowed them to make wider use of cheaper irregular workers, especially dispatched workers from SME sub-contractors with weak or no trade unions (Cho 2009, 66; Fleckenstein and Lee 2017, 169). Effectively, large employers transfer the cost of wage rigidity due to the seniority-based wage system to sub-contractors, thereby exaggerating wage inequality between workers and driving insider/outsider polarisation (see Woo and Cho 2016). Mass unemployment in the aftermath of the Asian Crisis entrenched the problem with business unionism in large workplaces promoting the somewhat narrow interests of insiders at the expense of outsiders. An interviewee from organised labour states:

Before the Crisis, large companies did not squeeze sub-contractors or shift cost pressures onto them. They sometimes even transferred technology to sub-contractors to increase product quality in a symbiotic relationship. However, after the crisis, the relationship between them became quite unequal, so that workers also have conflicts between themselves – between regular and irregular workers (Interview 20).

The huge differences in wages and other benefits between insiders in large firms and workers in SMEs undermine the latter’s commitment to their employer. Survey data, unsurprisingly, reveal a reluctance to work in SMEs because of poor salaries, lower job security and meagre, if any, occupational benefits (Kim, Kim, and Yoon 2012, 19). In 2015, half of workers (50.7%) in small firms (with less than five employees) had job tenure of shorter than one year, whereas only 12% of workers among larger firms (with 300 or more employees) had such short tenure (OECD 2018a, 40). In terms of average job tenure, small firms (3.3 years) did not even have half of the average tenure of 7.7 years in large firms in 2019 (Statistics Korea 2020). With their workers aspiring to jobs in large companies, SMEs and sub-contractors have little incentive to invest in their workforce’s skills. In fact, they fear the poaching of their skilled workers by larger companies (especially, their main contractor), with which they cannot compete (Interviews 4, 8, 14, 16, 17, 20). Of course, skilled workers of main contractors would not consider working for SME sub-contractors (Interviews 1, 2, 4, 13, 19). A trade union representative illustrates this with reference to the automobile industry: “Once over the fence of Hyundai Motors, the person cannot enjoy elsewhere a similar level of wage, company welfare and employment protection like those in Hyundai Motors. Nowhere. If going outside, they will fall off the cliff” (Interview 2). This situation is similar to hierarchical capitalism in Latin America where the “common fear of poaching discourages investment [in skills and training]” (Schneider 2009, 564). Certainly, this fear among Korean SMEs is not unfounded for after the Asian Crisis and the greater pressure from shareholder capitalism, chaebols paid more attention to short-term profitability, which makes the poaching of other companies’ skilled staff an attractive proposition compared to the costly investment in training (Interviews 1, 2, 4, 8, 16, 20).

Although workers with large employers still enjoy a very privileged position in the labour market, this is not to argue that they have not experienced increasing pressure and anxiety. Globalisation and the Asian Crisis have threatened insiders too as they have been affected by labour market deregulation since the late 1990s (see Fleckenstein and Lee 2019a; Jung and Cheon 2006). In this environment, insiders (especially, those with specific skills) have become more aggressive in protecting their employment (Interviews 20, 22). And, this has often translated into strong business unionism prioritising insider interests (job security, wage increases, and enterprise welfare) at the expense of outsiders, who are seen by some enterprise unions as a “buffer” between their insider membership and ever more difficult business environments. In other words, the “exploitation” of outsiders is perceived as protecting the secure and well-remunerated jobs of insiders (see Lee and Frenkel 2004; Lee 2011b). Whilst peak organisations, KCTU and FKTU, have increasingly recognised the short-sightedness of this strategy of their member unions and have pressed for greater solidarity with outsiders (including some progress at the public policy level), they have struggled to promote more inclusive unionism at the company level (Durazzi, Fleckenstein, and Lee 2018; Fleckenstein and Lee 2019a). The enterprise unionism, which was historically engineered by the authoritarian government to prevent working-class mobilisation by national labour federations and industry unions, has provided the institutional underpinning for the predominance of business unionism even after the collapse of the Developmental State; and with company unions prioritising job security, wages and working conditions on the shop floor, skills formation and developing workers’ qualifications through life-long learning do not feature in union strategy for collective bargaining and public policy (Interviews 1, 2, 11, 13, 18).

Critically, enterprise unionism favours chaebol unions, and low unionisation in the SME sector weakens the strategic capacity of industry and national trade union organisations. This obstructs collective bargaining beyond the firm and wage compression, which are indispensable for the development of industry-specific skills formation; and this leaves low-skilled and irregular workers in SMEs particularly vulnerable (Interview 17). Moreover, peak-level unions’ capacity to develop meaningful skills strategies is also undermined by enterprise unions’ reluctance to make funds available for policy staff dedicated to VET (Interviews 13, 22). Unions thus have little reason or capacity to make a meaningful effort to promote skills development. An analysis of the press releases of the two labour confederations reveals that KCTU has not produced a single one on training and skills since 1999, whilst FKTU could be argued to have at least started showing some interest. In 2006, it made an agreement with KRIVET to co-operate on vocational training for SME workers, and held a joint workshop to encourage union members to participate in training in 2015 (FKTU 2006; 2015; Interview 20). This difference between the two confederations might have to do with their membership composition. FKTU has a greater number of SME unions, whereas KCTU is dominated by unions in large manufacturing companies, such as automobile and shipbuilding (Choi 2011, 96). The argument that workers and unions have little interest in skills and training is supported by KRIVET research showing that insiders lack motivation for training, since there is no extra payment or future financial gain from participating in training. At the same time, unions display, the research argues, the strongest resistance to linking promotion to training and skills, as this is seen as undermining the seniority-wage system, which is still staunchly preferred by insiders. Employees in SMEs, due to their typically high workload, likewise do not welcome training, as it puts extra demands on their already pressed time. When asked by their employers to participate in training, well-organised workers “ask for a bonus instead of training” (Ban et al. 2018, 196).

In summary, labour market dualism derives from hierarchical capitalism, and it has become starker since the Asian Crisis and labour market deregulation in the late 1990s. The growing wage gap between insiders in large enterprises and outsiders in SMEs and the short job tenure of SME workers has made small employers reluctant to invest in training fearing their workers either leave soon or are poached by their contractor. Labour market deregulation and greater pressure from employers has made insiders more anxious about their job security, pushing their enterprise unions to focus even more firmly on protecting their jobs and wages. As business unionism has been reinforced at the shop floor level, labour federations have yet to develop serious interests in championing VET and especially training efforts that can benefit labour market outsiders.

Conclusions

Once praised for its economic model reinforcing education and growth, Korea has seen the demise of collective skills formation with huge impact on the labour productivity of SMEs and the service sector, in particular. This represents a major challenge to the Korean growth model, and in fact the country’s social model. With large employers in the “driving seat,” a segmentalist coalition of business and government first dismantled the developmental skills formation regime, and a shift in power between big business and the state undermined the government’s attempts to reinvigorate VET. Thus, rather than considering the government’s re-discovery of VET policy and collective skills formation as a return to developmentalism, this research has found that powerful employers flout government policy in VET.² The findings resonate with the literature emphasising the demise of the Developmental State or state-led co-ordination in Korea. For instance, Yeung (2016, 68) portrays big business as having “a free hand” in the national economy since the Asian Crisis; and with a similar emphasis, Hundt and Uttam (2017, 238) contend “the state [has] often played second fiddle to big business” in this new era. In this context, the Korean political economy is widely seen as having moved towards a much more liberal mode of market economy, as the state lost the institutional capacity to achieve developmental goals (see also Kang 2010; Pirie 2018).

Whilst liberalisation and deregulation are certainly important dimensions of institutional reform in Korea, we have shown that the country’s hierarchical production regime and inter-firm relations, established well before the Asian Crisis, have provided the micro-foundations for effective resistance to VET reform and, specifically, failed attempts to engage employers in meaningful vocational skills formation, rather than “classical” market failure that is often associated with unsuccessful VET policy and polarisation in liberal market economies, such as Britain (see, for instance, Fleckenstein and Lee 2018; Rainbird 2010). Exploiting their powerful position in the political economy, large employers prioritise their skills needs – take advantage of the expansion of higher education, invest in their incumbent workers, and poach the best workers from SMEs. They show no interest in portable, industry-specific skills. With little or no breathing space, SMEs have no capacity and indeed incentive to invest in the skills of their workforce, as their workers are most keen to move on to a larger employer for better wages, benefits, and working conditions. Workers and their unions do not show much interest in training either – the seniority-wage system does not provide any incentive to up-skill, and insiders do not allow any challenge to the seniority-wage system but they expect their enterprise unions to prioritise their wages and benefits. In other words, the hierarchical production regime has promoted a heavily dualised labour market and driven greater social inequality, which further undermine collective skills formation.

Clearly, the Korean political economy does not present the inter-firm co-operation that is needed for collective skills formation, and fragmented organised business do not possess the strategic capacity to promote non-exploitative relations between firms. Whilst Hall and Soskice’s (2001, 34–35) notion of “group-based co-ordination” captures the critical importance of chaebols in the Korean political economy, it fails to fully grasp the hierarchy in the production regime – with its profound consequences for skills and training, in addition to the labour market and industrial relations more generally. Here, Korea shares similarities with the Latin American experience (see Schneider 2009). Using their extraordinary socio-political power position, large employers have been pressing firmly for deregulation, including in the domains of the labour market and skills policy. This has produced an increasingly pathological equilibrium, holding back economic and social progress. Policy-makers have recognised the shortcomings of the “liberal” skills formation regime, and they have pushed reforms to reengage employers. However, the findings of this research suggest that, without more fundamental changes in the Korean political economy, reforms that are confined to skills policy are largely futile, as the very foundations of the production regime undermine effective VET policy. That said, radical political-economic reforms, challenging the current growth model, are of course a difficult endeavour, as the chaebols’ privileged position and their enormous power resources have promoted the emergence of the pathological path. Previously, only the repercussions of the Asian Crisis, which fuelled a perception of “crony capitalism” in a large part of Korean society, opened a brief window that allowed the power position of chaebols to be challenged (see Kang 2002). It remains to be seen when such window of opportunity reopens – business is the pivotal actor in training reform and political-economic reform more generally; and this seems to be true in Korea more than in most other countries.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the Laboratory Program for Korean Studies through the Ministry of Education of the Republic of Korea and Korean Studies Promotion Service of the Academy of Korean Studies (AKS-2018-LAB-1250002), and by Korea Foundation’s Support for Policy-Oriented Research programme.

Notes

1 For a discussion of segmentalism, see Thelen and Busemeyer (2013).

2 On the transformation and persistence of the Developmental State see Kim and Kwon (2017); Thurbon (2016); and Lim (2010).

Appendix.

List of interviewees

Table

Interviewee	Position	Date
1	Public Organisation	March 7, 2019
2	National Trade Union	March 14, 2019
3	Public Organisation	March 15, 2019
4	National Employers Association	April 8, 2019
5	Industry Association (Manufacturing), and Industry Skills Council	April 8, 2019
6	National Chamber of Commerce and Industry	April 9, 2019
7	Industry Association (Low-end Services), and Industry Skills Council	April 9, 2019
8	Regional Chamber of Commerce and Industry, and Regional Skills Council	April 10, 2019
9	Industry Association (High-end Services), and Industry Skills Council	April 11, 2019
10	Regional Chamber of Commerce and Industry, and Regional Skills Council	April 12, 2019
11	National Chamber of Commerce and Industry, and Industry Skills Council	April 12, 2019
12	Industry Association (Manufacturing), and Industry Skills Council	April 15, 2019
13	Industry Trade Union (Manufacturing)	April 15, 2019
14	Regional Chamber of Commerce and Industry, and Regional Skills Council	April 16, 2019
15	Regional Employers Federation, and Regional Skills Council	April 16, 2019
16	Industry Association (Manufacturing), and Industry Skills Council	April 16, 2019
17	National Employers Association	April 18, 2019
18	Industry Association (Manufacturing), and Industry Skills Council	April 19, 2019
19	Industry Association (High-end Services), and Industry Skills Council	April 24, 2019
20	National Trade Union	April 24, 2019
21	Public Organisation	April 25, 2019
22	Public Organisation	April 26, 2019