Abstract
It has been observed that automotive firms in Japan follow a distinct approach of using digitalisation, especially in comparison to the German approach towards “Industry 4.0.” Theoretical explanations for the distinct way in which Japanese automotive firms deploy digital technologies in Japan will be explored and contrasted with the case of Germany. As firm strategy appears to differ between different locations, the Varieties of Capitalism (VoC) approach is one analytical lens that may help to understand different firm behaviour and different skill formation approaches towards the use of digital technologies in both countries. While VoC provides some insights, it is nevertheless concluded that more empirical research is needed to link theoretical constructs and actual use cases.
1 Introduction
It has been observed that automotive firms in Japan follow a distinct approach of using digitalisation (Holst et al. 2020; Mokudai et al. 2021). This particular approach is characterised by the usage of information generated by digital technologies to support shop floor operators and white-collar staff in joint kaizen
2 Varieties of Capitalism
The Varieties of Capitalism (VoC) framework is elaborating on a distinction between liberal market economies (LME s) and co-ordinated market economies (CME s). Both types have distinct sets of institutions, including financial, collective bargaining, vocational training, and welfare ones. One important difference from older literature on corporatism is that the VoC concept highlights employers’ interest in distinct institutions and even suggests that firm strategies are exploiting benefits that institutions provide. Thus, LME s and CME s are stylised to possess distinct efficiency in the production of different goods, the former in ones characterised by more radical innovation and the latter in ones characterised by incremental innovation (Hall and Soskice 2001: 36–44). A further aspect of the VoC concept is that institutions were described as complementary sets that have a self-reenforcing and perpetuating quality. Thus, a particular interpretation of the VoC concept stresses stability and suggests that countries have developed particular “national business models.”
Such a functional and seemingly rational explanation has received much criticism. One critical issue is that the VoC framework can only explain institutional change by pointing at exogenous forces such as globalisation (Howell 2003). Moreover, for a political economy framework, it paid little attention to political actors, political discourse, and (re-)formation of political or societal coalitions (Culpepper 2005; Schmidt 2009; Mahoney and Thelen 2010). More recent work has responded to this critique by looking more closely at actors, their strategies, and how their interaction with existing institutional structures has let to institutional change (Thelen 2014).
Complimentary relationships between institutions were for instance theorised for skill formation and welfare regimes (Estevez-Abe, Iversen, and Soskice 2001): by differentiating social protection into employment and unemployment protection,1 differing tendencies of skill formation were explained. For the case of Germany, at the time a country with high employment and unemployment protection, a mixture between industry and firm-specific skills was predicted. For Japan, a country with high employment but low unemployment protection, workers’ tendency to invest into firm-specific skills was regarded as a rational choice. This nexus between public welfare and skill formation must be regarded as less important than previously conceived: the German system of unemployment protection has been substantially eroded with the so-called Hartz reforms (Busemeyer and Trampusch 2013: 300; Baccaro and Howell 2017: 111–113) without undermining the attractiveness of vocational training (Thelen 2014: 86–90).2 However, the latter in Germany does suffer from dualisation, in this context meaning that, while (or because) the quality of vocational training is continuously updated to keep track with technological development and firm needs, the number of training opportunities in firms is decreasing. The latter trend is mainly caused by smaller firms retreating from offering apprenticeships as they lack the resources to implement training according to increasingly centrally defined demanding and encompassing standards.
Japan is a somewhat atypical CME.3 Similar to the German case, the transformation of the Japanese political economy since the 1990s is subject to debate. One group of scholars emphasises the resilience of co-ordination (Sako 2006; Thelen and Kume 2006; Witt 2006) but simultaneously acknowledges the decreasing social egalitarianism produced by said co-ordination. Another group (Miura 2008; Mouer and Kawanishi 2005; Rebick 2005; Shibata 2016) regards this as a sign that Japan underwent neoliberalisation. Kubo (2020) specifically links changing employment relations to financialisation. Lechevalier (2014) takes the middle ground by arguing that older forms of co-ordination, such as keiretsu
Whereas industrial relations characterised by centralised and co-ordinated collective bargaining are common among CMEs, co-ordination in the so-called shuntō is largely informal (Kume 1998: 87–96). While large firms from the metal industry were for a long time de facto pattern setters that influenced nationwide wage settlements, the pattern setting function of the metal industry declined and wage bargaining became increasingly decentralised (Weathers 2003). Whereas unions tried to revive shuntō by extending the scope beyond wage setting and towards topics such as working conditions and treatment of non-regular employees, this initiative largely failed (Weathers 2008; Ogino 2021: 21). Although wage setting is still negotiated during shuntō, its former function of achieving relatively egalitarian wage increases for Japanese employees has been eroded.
Further, the two most prominent features of Japanese industrial relations, i.e., lifetime employment (shūshin koyō
Employment by type in Japan, 1992–2017
Citation: Vienna Journal of East Asian Studies 15, 1 (2023) ; 10.30965/25217038-01501002
Source: Employment Status Survey, Ministry of Internal Affairs and CommunicationContractual constraints and employment rules
Citation: Vienna Journal of East Asian Studies 15, 1 (2023) ; 10.30965/25217038-01501002
Source: Modified from Marsden 1999: 33Employment status data clearly identify two trends. First, despite (or due to) aging population, the number of people working has expanded since the 1990s. There are several explanatory factors for this situation, chiefly the expansion of female employment and the growing number of people working past the official retirement age (Rebick 2005: 63–64). Second, while the number of regular employees4 has decreased by about 3.5 million persons between 1992 and 2017, the number of atypical employees doubled from 9.5 to 19.1 million. This observation is consistent with Kubo’s (2020: 149–151) review that identifies numerous studies that find that large firms maintain long-term employment, especially for university trained males, combined with increased use of atypical employees. Thus, there is a clear trend towards the expansion of atypical employment which is tied to lower wages, lower welfare benefits, and less job security. It can be concluded that employers shifted their utilisation of atypical workers from a buffer to adjust in order to demand fluctuations towards transforming fixed labour costs to variable labour costs (Sako 2006: 235–236). Japanese employment is increasingly dualised, i.e., lifetime employment still exists but it covers decreasing shares of workers.
Furthermore, the notion that Japanese industrial relations are distinctive and that their uniqueness is rooted in culture5 must be clearly rejected. Gordon’s (1985) historical study on the evolution of Japanese industrial relations demonstrates that relations between capital and labour in Japan during the Meiji era
Finally, the topic of employment deserves attention. Somewhat similar to VoC, Marsden’s (1999) theory of employment systems seeks to explain persisting national differences of employment practices. His theory adopts a rational choice perspective on transaction cost economics to explain employment relations. He argues that the latter rest on two key questions: first, how to align job demands with worker competencies, or what he calls efficiency constraint; second, how to formulate criteria for task assignment to workers, or what he calls enforceability constraint. Regarding efficiency, Marsden distinguishes between the production approach which adjusts the qualifications according to the job requirements and the qualification approach which adjusts the job environment to workers’ qualification. Concerning enforceability, he distinguishes between a task-centred definition, which means a narrow delineation of job contents, and a function-centred definition, which means a broad definition of contents, hence allowing for more flexible utilisation of workers. The combination of the constraints results in four employment rules (Table 1).
The “work post” rule which he describes as characteristic of French and US firms seeks to establish a relatively clear and narrow definition of job contents for each worker. This definition is based on the (process) requirements as developed by the firm. Training is typically confined to on-the-job training (OJT).
The “job territory” rule is characteristic of British firms and is equally narrow in the job definitions, but the delineations between jobs is not based on firms’ task design but on “traditional” crafts and their unions.
The “competence rank” rule is characteristic of Japanese firms. Here, workers are rather team members without a specific job. However, this does not mean that there is no hierarchy. Workers are ranked according to a job grade matrix, which assess workers’ competency in performing a variety of related tasks. It should be highlighted that this model presupposes that workers receive OJT and enjoy lifetime employment. A further characteristic of Japanese job relations is that individual accountability and performance is de-emphasised to foster co-operation within a team (Marsden 1999: 161–163) despite the usage of job grade matrices. Consequently, performance of the work team and individual workers’ contribution to team performance takes precedence over any individual’s performance.
Finally, the “qualification” rule is characteristic of German firms. It is highlighted that formal qualification certified through formal vocational apprenticeship is central for a worker’s status. Moreover, job design is rather adjusted to certified competences than the other way around.6 It needs to be highlighted that in comparison to the “job territory” rule, the competences are rather broadly defined and allow relatively flexible deployment according to management discretion, i.e., job categories tend not to constrain work allocation (Marsden 1999: 129). Simultaneously, management discretion is constraint because, if jobs contain mainly skilled tasks, then workers (directly through work councils and indirectly via unions) have grounds to argue that such jobs should be carried out by certified skilled workers.
The important point for our discussion is that Marsden demonstrates that these different rules are related to differing forms of performance assessment (Marsden 1999: ch. 6) as well as pay and incentive systems (ibid.: ch. 7). Thus, it should be expected that redefining job contents under different employment rules will have differing consequences when it comes to issues such as performance assessment and especially pay.
We follow Thelen’s (2014: 7–8) argument that VoC is more interested in co-ordination than in the social outcomes of said co-ordination, an aspect that many VoC critics focus on.7 Regarding the industry focus of our analysis, we emphasise findings of Doellgast and Greer (2007) for industrial relations in the German telecommunications and automotive industries as well as Haipeter, Jürgens, and Wagner’s (2012: 2029) study of this topic in the German automotive and banking industries that highlight significant liberalisation but conclude that industrial relations remain co-ordinated. We argue that the question of co-ordination is important to understand differing debates on the implications of digitalisation in Germany and Japan.
3 Skill Development in Germany and Japan: Responses to Digitalisation?
Regarding skill formation, differences between LMEs and CMEs as well as differences within CMEs are well explored by the VoC concept. Although Germany and Japan are both CMEs, the respective skill formation approaches differ.8
3.1 German Dualisms: Interaction between Industrial Relations and Vocational Training
While we will start with a formal description of how vocational training rules are developed in Germany, this account will be followed by a discussion that explores the relationship between skill formation and industrial relations, i.e., a more encompassing social institution. Since vocational training and industrial relations are characterised as dualisms, we will seek to clarify how these dualisms are interacting.9
Müller-Jentsch (1999) points out that the German law on occupational training is encompassing three elements, namely, initial vocational education and training (IVET), continuing vocational education and training (CVET), and retraining. Our discussion will focus on IVET, because it is based on strong legal guarantees10 and because CVET has been described as underdeveloped (Thelen 2014: 97–98). The German dual system of IVET is characterised by a combination of apprenticeship in a firm and vocational training at a public vocational school. Regarding training contents, employer associations, unions, and the state officially form a tripartite system that defines these contents since 1969 (Thelen 2004: 242).11
Following Müller-Jentsch, defining IVET rules can be described as a three-step process. First, at the federal level, the Federal Institute for Vocational Education and Training (Bundesinstitut für Berufsbildung; BIBB) is responsible for enacting training regulations (Ausbildungsordnung) for each officially sanctioned occupation. On paper, issuing training regulations does not require the participation of employer associations and unions, but it is widely observed that the federal government is basing its decisions on consensus reached between these two parties.12 In this context, it should be emphasised that national top-level organisations defer to sectoral organisations, such as employer associations and unions from the particular industry, for which training regulations are issued.
Second, at the state level, governments are responsible for enacting examination regulations (Prüfungsordnung) which are based on the more general training regulation. Again, the examination regulation is based on a consensus between state, employers, and unions, where the state is officially sanctioning consensus reached between capital and labour. Furthermore, whereas the examination of apprentices is administrated by local Chambers of Commerce, examination committees are constituted by a vocational teacher who is a state employee, a representative of the employer, and a union representative.
Third, the firm level is the place of practical apprenticeship. However, firms are not simply enacting vocational apprenticeship based on training regulations. Training regulations define minimum requirements which can be expanded by an agreement between management and works councils. Thus, IVET should not be understood as a top-down system where firms merely execute schemes developed at the federal and state levels.
German industrial relations are frequently described as a dual system (Thelen 1991; Visser and Van Ruysseveldt 1996; Hassel 1999) characterised by central collective bargaining between employer associations and unions, and co-determination rights at the firm and plant levels.13 Since the 1990s, central bargaining increasingly allowed deviations from central agreement, which has been interpreted as pressure towards decentralisation (Hassel 1999). Thus, despite central collective bargaining, capital and labour have some flexibility to find solutions that take local conditions into account. Regarding co-determination, works councils have a number of rights protected in the Works Constitution Act.
Works councils can utilise guaranteed rights to address issues related to technological change indirectly (Visser and Van Ruysseveldt 1996: 155). At the plant level, changes in working hours,14 job reclassification, and transfers require the approval of works councils. In her discussion of the role of dual system institutions in the introduction of automation in the 1980s, Thelen (1991: 213, 219) points out that the works council of VW Braunschweig linked its approval of changes in working hours, manning levels, and wage grades to its demands for broader jobs and group work. Additionally, VW Braunschweig works council members participated in talks with technology suppliers, viz. they were not excluded from the management’s search for new production technology. Hence, while management only needs to inform a works council about the introduction of new technology, the consequences of said introduction must be negotiated with works councils. For this reason, it appears unlikely that the management will try to implement new technologies on the shop floor without the works council’s approval.
It should be acknowledged that the linkage between industrial relations and skill formation is not systematic. Nevertheless, both topics clearly overlap (Müller-Jentsch 1999). Marsden (2015: 170) even declared that collective bargaining and workplace co-determination, along with skill formation, should be regarded as characteristic of German industrial relations. Researchers have highlighted several facets that illustrate the overlap of industrial relations and skill formation. For example, Sengenberger (1987: 255) emphasised that, whereas unions represent all workers regardless of qualification, union officials and works councillors usually hold vocational degrees. For this reason, he regarded skilled workers as the main union clientele. In their historical institutionalist analysis, Thelen and Kume (2001: 211–215) observed that the industrial unions’ pursuit of co-determination over vocational training was for a long time unsuccessful but never abandoned, suggesting that vocational training is an area of interest for unions. Concerning the linkage between industrial relations and skill formation, two points are worth emphasising: first, German unions have displayed a tendency towards maintaining the quality of training instead of limiting the quantity of skills supplied to the economy (ibid.: 206); second, in the context of changing technology, this tendency makes it possible for employers and unions to forge compromises that are based on enhancing training. If technological change requires additional and enhanced skills, reforming IVET to cultivate a supply of said skills satisfies employers and unions alike.
The preceding description may be summarised as follows: Formally, industrial relations and vocational training have only relatively weak linkages. Despite this legally loose relationship, especially labour found ways to link these two spheres in order to gain influence on the way new technologies are deployed on the shop floor and how training is adjusted to account for technological change. As this pattern has been described in studies that address automation – i.e., the third industrial revolution – it appears likely that a similar strategy can be observed for digitalisation which is already hailed as the fourth industrial revolution in Germany.
3.2 Germany’s Dual System of Vocational Training: Continued Co-operation to Deal with Digitalisation
Given that the revision of training contents is an activity co-ordinated between employers, unions, and the state, it appears natural that all parties engage in a debate about the best way to react to anticipated technological changes such as digitalisation. Furthermore, if one assumes that all actors have distinct goals and interests, a wider discourse about the impact on industrial relations and management is a logical consequence. For instance, if one limits the view to labour, unions have at least two goals. First, novel technology should not lead to substitution of labour to avoid redundancies. Simply put, unions seek to protect the interests of their members in continued employment and related economic security. Second, novel technology should ideally not result in de-skilling. Thus, unions have an interest that current workers have opportunities to learn how to use these technologies and that future workers are qualified to work using them. Therefore, unions have an interest to respond quickly to novel technology by increasing training and skill requirements. For instance, employers and unions agreed in 1984 to transform fifty-four distinct trades and related apprenticeship training in the metal and electronics industries into ten broader ones (Thelen 1991: 204), i.e., unions recognised that insisting on narrowly defined qualifications would in the long run undermine chances for employment and hence welcomed reforming apprenticeship. Another historic example for this tendency was the German unions’ reaction to increased automation occurring in the 1980s by promoting the increased use of skilled workers opposed to semi- or unskilled workers. While automation was commonly feared to result in de-skilling or outright substitution of work, this union policy led to an increased use of skilled labour in German manufacturing firms (Jürgens, Malsch, and Dohse 1993: 373). During the 1990s and 2000s, however, German carmakers followed divergent paths towards the skill requirements of their blue-collar workers in their specific interpretations of lean production. Whereas BMW, Opel, and VW maintained relatively high shares of skilled workers, Audi increased, and Mercedes-Benz decreased the share of skilled workers (Haipeter, Jürgens, and Wagner 2012). Despite the overall trends within a country, this highlights that unions and employers – especially large ones, such as carmakers – have agency in questions related to industrial relations, such as skill formation and extent of skill utilisation, in their production systems.
Regarding digitalisation, the continued broad support for vocational training in Germany described by VoC scholars suggests that training contents will be modified to prepare skilled workers for ongoing technological changes. Thus, one would expect that the “Industry 4.0” concept was closely followed by a debate on changing skill requirements in industrial work. In this context, research on the (potential) impact of digitalisation on the German dual vocational training system emphasises that work requirements are going to become increasingly complex, meaning more interdisciplinary and requiring a higher degree of social competences (Gebhardt, Grimm, and Neugebauer 2015: 121; Tenberg and Pittich 2017: 29–30). Indeed, training contents of eleven occupations belonging to the metal industry were already updated to reflect requirements of “Industry 4.0” in 2018 (BMWi 2018), which reflects the leading role these industries have for German manufacturing. BIBB initiated its pilot project to identify the impact of digitalisation on work and vocational training qualifications in co-operation with Volkswagen in 2014, which may explain why occupations from the metal industry were the first to be modified. Only three years after the forthcoming fourth industrial revolution was articulated at the Hanover Industrial Fair in 2011 did the central German agency in charge of vocational training initiate a programme to respond to the anticipated change, and first training contents were added seven years after this event. Thus, the German dual vocational training system needed very limited time to formulate a co-ordinated response to anticipated changes.
Added contents for all eleven occupations include topics such as data security and data analysis, portable data storage media and protection against malware, digital twin (systems for data-based assistance, diagnostics, and visualisation), and working in interdisciplinary teams. While on the one hand this is beneficial to keep vocational training attractive, it may on the other hand further “dualise” access to training opportunities. Addressing such a range of topics in vocational training may be increasingly difficult for smaller firms, so that the continued deepening and broadening of occupational profiles may effectively limit the pool of firms that can offer apprenticeship according to such requirements. Thus, this may have the consequence that firm-based training will be increasingly confined to large firms that have the resources to support workers’ skill formation. Indeed, some observers fear that increasing base requirements for apprenticeship will grow the number of youths who will first not qualify for apprenticeship and later will be regarded as unemployable to twenty or thirty per cent per age cohort (Tenberg and Pittich 2017: 43).
Additionally, it is important to point out that the change in occupations’ training contents can be regarded as a reflection of the German emphasis on IVET. Whereas IVET is recognised as a strength of the German vocational education system, CVET has been described as underdeveloped (Thelen 2014: 97–98). Therefore, one would expect that German firms limit support for CVET to few employees and that its duration is rather short.15 Interestingly, this tendency identified by VoC scholars contradicts the prescriptions of academic vocational training researchers in Germany who emphasise that CVET will become increasingly important to manage the anticipated transformation of work through digitalisation (Gebhardt, Grimm, and Neugebauer 2015: 122; Tenberg and Pittich 2017: 39), a view shared by BIBB officials (Padur and Zinke 2015: 31).
Adding qualifications under the German “qualification rule” must be expected to lead to demands for wage increases. Here it should also be pointed out that there is not a strict linkage between employee qualification and renumeration. Although qualification is important, wages depend on the classification of a job. In turn, the classification takes several variables into account, inter alia: vocational training, job experience (years of service), communication, and personnel management content of the job in question. Thus, whereas skills certified through vocational training are important, they are not the sole determinant of a worker’s wage. In the context of digitalisation, unions may point to added qualifications as an indicator that workers take over tasks that are presently performed by higher paid job classifications, or simply that their qualification is becoming more encompassing and thus deserving higher pay. Thus, labour may be willing to co-operate with capital to (re-)define broader job classifications in exchange for increased workers’ renumeration.
3.3 Japan’s Firm-Specific Training:16 Lifelong Employment-Skill Formation Nexus and Digitalisation
As skill formation in Japan is evolving around firm-specific skills, meaning it is not co-ordinated at the industry level,17 one would expect that firms develop internal solutions to confront digitalisation. As lifetime employment is still a widely shared expectation between employers and employees in large Japanese firms,18 the VoC concept suggests that firms give regular employees room to familiarise themselves with digital technologies and encourage them to find ways to apply these novel technologies in production.
Simultaneously, this firm-centric approach suggests why there is no debate on implications on industrial relations and management in Japan.19 Since Japanese industrial relations are fundamentally based on lifetime employment, the discussion about possible redundancies due to technologic development prevalent in Europe and North America (Briken et al. 2017: 2–5) may simply not take place because institutionalised protection from layoffs is central for motivating workers to invest into firm-specific skills in the first place. In other words, the narrow focus on technical topics such as AI and IoT is caused by the prevalent assumption that technical change will not cause firms to significantly reduce their (current) regular workforces. Concurrently, the increasing number of atypical workers may well be threatened by substitution. However, the confluence of the end of the “bubble economy,” IT revolution, and the aging society during the 1990s led to an increase in early retirement in Japan, precisely because it is basically the only way for firms to actually reduce the workforce (Thelen and Kume 2006: 29). Thus, if the introduction of digital technologies will lead to the drastic reduction in labour demand that some observers prognosticate, one would expect that Japanese firms would first try to retrain workers and then seek to hasten the exit of older workers through early retirement. Conversely, if digital technologies are not going to replace significant amounts of labour through advanced automation, the VoC concept would anticipate the aforementioned tendency to qualify existing workers to utilise novel digital technologies.
Regarding the topic of renumeration, it is well documented that the Japanese industry does not apply a strict job classification system that links the currently performed job to payment. Instead, Japanese workers are classified into job grades that basically reflect the range and depth of acquired skills (Koike 1994: 50–59). Thus, reassigning workers to perform other jobs is not opposed as it may not affect workers’ wages. According to Marsden (1999), Japan can thus be characterised as utilising a “competence rank” rule, according to which rank is not linked to the performed job but to the acquired and recognised competence of individual workers. Thus, it may be expected that firm management and company unions will engage in the discussion of how to incorporate skills related to digital technologies into existing firm-specific job grade matrices. Yet, this may be difficult, as such matrices usually assess the extent to which a worker has mastered an operation. In the German case, new job contents related to digitalisation are recognised as using different media and especially skills such as data analysis and co-operation across job qualifications. In Japan, soft skills – such as co-operation – may not be assessed through a job grade matrix but be part of an overall job evaluation. Further, data analysis is an essential part of kaizen activities. This suggests that what is defined as new skills in Germany may not be regarded as novel qualifications in Japan. Thus, we extend our discussion to the topic of kaizen and how it is practised in Germany and Japan.
4 Kaizen
4.1 Kaizen as Practised in Japan
Regarding kaizen as practised by Japanese firms, some important remarks are in order. First, whereas operator involvement is widely adopted, operators are not the main source of productivity improvement. The main motivation for operators being involved in kaizen is for instilling an improvement capability in the most capable workers who are promoted to team or group leader positions (Ishida 1997: 51). Thus, while operators are not a main source of improvement, kaizen is nevertheless practised as one criterion (among others) for promotion. Second, both group and team leaders are mainly responsible for planning and implementing activities that lead to major improvements (Nemoto 1992; Ishida 1997: 51; Aoki 2008: 528). Hence, the view that Japanese kaizen is building on worker involvement is only in so far correct in that firms encourage operators to engage productivity improvement.20 However, it is incorrect that workers are also the main source of productivity improvement.
Regarding industrial relations, Hyodo (1987: 268) observed that Japanese unions never attempted to challenge the character of kaizen as a managerial tool for productivity-enhancement. Instead of questioning the logic of rationalisation of work processes and the oft – following worsening working conditions, Hyodo concludes that unions only sought to ensure that workers would receive their fair share of financial benefits generated by realised productivity improvements. This observation is contrasted by Shimizu’s (1994) description of how Toyota changed its production system in the early 1990s. Virtually, Toyota management together with its company union concluded that production had to be humanised and enriched to solve the contemporary problems of workforce shortage and high worker turnover. Hence, Shimizu’s study highlighted that there was a shared shift in perception of the OEM and its union which resulted in reforming the pre-existing linkage between kaizen and work intensification in the sense that both parties sought to decouple these two topics to the extent possible.
It should be also mentioned that case studies on Japanese industrial relations in the automotive industry found that firms which significantly deviated from the Japanese approach, e.g., by hiring large numbers of temporary workers, suffered from deteriorating participation in kaizen activities (Aoki, Delbridge, and Endo 2014: 2567–2568). Sako (2006: 238) points out that management cannot expect temporary workers to be equally multi-skilled and problem-solving-oriented as regular workers, which makes it more challenging to accumulate, sustain, and transfer the know-how among shop floor workers. Miura (2008: 170) makes a similar point by emphasising that the expansion of atypical employment must be expected to undermine firm-specific skill formation. This suggests that the willingness to participate in kaizen is linked to lifetime employment (or other expressions of reciprocity) and that perceived undermining of this institution also leads to declining kaizen participation in Japan.
The idealised Japanese way of practising kaizen suggests that firms will seek to integrate novel technology into their improvement activities in an inclusive way for their regular employees. Precisely because operator kaizen is part of training and an important competence to qualify for promotion, novel technologies must be expected to be used in a way that allows operators to experiment with and gain an understanding of these technologies. Hence, one would expect that firms do not leave implementation and utilisation of digital technologies to IT specialists, engineers, and managers – instead, they encourage operators to engage in their utilisation. In effect, the preceding discussion on expanding atypical employment suggests that operators in atypical employment forms will probably be excluded from the experimentation with digital technologies, and that these workers will have little incentive to participate in productivity improvement activities.
4.2 Japanese Use Cases: Digitalisation as Lean Augmentation and Participatory Experimentation
Field research in Japan (Holst et al. 2020; Mokudai et al. 2021) provides some detailed context about digitalisation in Japanese firms.21 Case studies include five automotive suppliers (Firms A, B, D, E, and F) and two automotive manufacturers (Firms C and G). Firms C and G, despite being carmakers and hence large enterprises, display the tendency to gradually develop skills related to novel technology in a rather exemplary manner. In case of C, the carmaker so far had no AI specialists in their production departments, with firm engineers learning about AI by themselves and subsequently initiating its trial application. This illustrates the tendency to allow existing staff to learn about technology and apply it to firm-specific problems. However, it should be acknowledged that, in this case, opportunity to learn is confined to white-collar experts. Another remarkable feature of Firm C’s approach is that staff who study and experiment with AI are not doing so in isolation. Instead, the guiding principle is that AI’s capability for data analysis should be used to “hint” towards issues that are to be solved by existing processes. Thus, novel technology is used to compress the time needed for problem identification, but solution development is still assigned to humans. Firm G follows a similar approach by introducing co-operation between IT specialists and plant operations staff. Importantly, IT specialists do not take the lead and instead both sides try to learn each other’s “language.” While this learning process was described as difficult, it nevertheless highlights that both sources of knowledge are regarded as equally important, i.e., know-how related to novel digital technologies is not automatically assigned higher value than experience-based production know-how.
Firm F comes close to the outlined ideal type. Management encouraged team leaders to use digital technologies for production management. Conceptualisation and implementation were assigned to an operator team under a team leader. The team designed a production monitoring system that consisted of sensors, single board computers, and radio modules. All components were inexpensive and together functioned as an effective real-time monitoring system that records machine availability, cycle time, and output. Close to the described ideal, the firm did not utilise external expertise or deployed experts to achieve its goal. However, after the firm successfully deployed the monitoring system, it started a subsidiary consulting business that sells its system to other firms and – as part of the process – hired several IT specialists to manage the expanding IT infrastructure. Nevertheless, Firm F initially entrusted shop floor personnel with the task of utilising the technology. The operator team – of whom only one had tertiary education – effectively self-taught how to install sensors to existing machines, connect sensors to radios and computers, and programme the computers to perform the desired tasks. Thus, skills were developed by experimenting with technology, and the skills are only corresponding to the need of the specific firm – in this case the deployment of an improved production monitoring system. Evidence from Firms C, F, and G thus illustrates that Japanese firms are indeed able to utilise existing blue- and white-collar workers to deploy digital technologies. The cases suggest that skill development is organically evolving inside the firms. At the same time, it should be acknowledged that not all studied firms included blue-collar workers or deployment of digital technologies. Thus, while there may be a link between lifelong employment and firm-internal skill development for some employees, this connection is arguably stronger for white-collar and higher-ranked and/or skilled blue-collar workers who tend to belong to indirect production sections, such as tooling and maintenance.
Lifelong employment may however also become an obstacle for the adoption of digital technologies. In case of Firm E, the manager in charge of the digitalisation effort reported that shop floor and production line managers refused introduction of automated data acquisition to manage production-related information. Despite economic difficulties of this supplier, which led to a change from family ownership to the control by an investment fund, the manager pointed out that employees did not fear redundancies and lacked motivation to engage new technologies. Further, he mentioned that especially younger employees were leaving the firm. This suggests that lifelong employment in this case does not encourage operators and shop floor managers to learn to adapt to novel technologies. Rather, institutionalised employment protection in the form of lifetime employment here provides the opportunity to resist unwanted change.
A shared characteristic of most Japanese cases is an emphasis on current permanent employees, including blue-collar workers, and their qualification to utilise digital technologies in their workplace. This contrasts rather strongly with the German emphasis of redefining training contents for apprentices. Consistent with the logic of lifetime employment and firm-internal skill formation, Japanese firms display a tendency to offer opportunities for skill formation through experimentation. For the sake of comparison, one may label this as CVET. However, it should be emphasised that CVET suggests that training is conducted with a priori defined training content. In the Japanese cases investigated, it is evident that firms allow room for experiments to discover in actu applications.22
4.3 Kaizen as Practised in Germany
Studies on the introduction of lean production practices by German automotive firms found operator participation in kaizen to be limited and instead observed that white-collar experts, that is, engineers and managers, dominated improvement processes (Jürgens 1998; Gerst et al. 1999; Labit 1999). Despite this tendency, Roth (1997: 129–131) identified two distinct approaches to kaizen in the German automobile industry: on the one hand, expert-led kaizen, which is typically an expert-led, short-term effort that seeks to achieve predefined economic goals; on the other hand, group-led kaizen is an operator-led, long-term practice that includes improvement of economic and social aspects of work.23 Regarding kaizen at the Mercedes-Benz plant in Tuscaloosa, Alabama, USA, it has been observed that operators mainly focus the improvement efforts on non-value-adding activities, viz. not improving the value-adding activities (Pries 2004: 88).
As for industrial relations, Thelen (1991: 223 n. 49) notes that IG Metall changed their stance on kaizen activities – such as quality circles to lobby for union demands related to job design. Even though German unions did not question the efficiency enhancing logic of kaizen, they at least attempted to link questions of work efficiency and job design.
4.4 Hypothetical German Use Cases: Top-Down Implementation
This subsection seeks to formulate what kind of utilisation pattern (if any) one would expect to observe empirically against the background of the preceding discussion on VoC and skill formation. Hence, we describe hypothetical constructs which are consistent with these strands of academic research as well as past empirical studies on the German automotive industry.
Taking into account a rather strict Taylorist separation of work conceptualisation and work execution in the German automotive industry and the extension of this approach to kaizen, one would expect that digital technologies are mainly designed and utilised by white-collar experts rather than blue-collar workers. Thus, we expect a top-down initiation and implementation of digital technologies without significant bottom-up input from workers. The German debate on “Industry 4.0” emphasises the goal of autonomous self-regulation of machines and processes, which implicitly suggests an overarching importance of system designers (i.e., experts). This approach of relying on experts that use specialised tools to identify improvement potential from “big data” acquired by digital technologies would thus stand in the tradition of expert-led kaizen with limited operator participation.
Regarding the speed of introduction, we expect that the time required to deploy solutions will be longer compared to Japan. This is expected because the skill formation system in Germany is co-ordinated, whereas in Japan it is limited to (keiretsu) group standards or is not co-ordinated at all. Consequently, deployment in Japan should be faster due to limited need for co-ordination and the pursuit of narrower firm or group goals.
From the labour side, we expect to encounter resistance against the introduction of digital technologies only if they are not tied to opportunities for upskilling and eventually higher wages. As already mentioned, however, employers and labour unions are co-operating in the formulation of “Industry 4.0” job and associated training profiles, meaning we expect to see little resistance from organised labour. As in the case of Japan, however, unions are rather unable (or perhaps unwilling) to represent the interest of atypical workers. Given the strong legal rights at the plant level, we would expect that unions are seeking to utilise work councils to bargain over the deployment of digital technologies.
5 Conclusion
Approaches to the implementation of digital technologies in Germany and Japan seem to contrast rather strongly, with the latter being less reliant on experts and industry-wide job profiles in need of co-ordinated updating than the former. Situating the implementation of digital technologies within the VoC framework – and especially its analysis on skill formation systems – structural differences in employment and skill formation in Germany and Japan can partially explain the different digitalisation approaches.
In Japan, despite the firmly growing atypical employment, lifelong employment coupled with on-the-job training as well as promotions linked to meritocratic performance in the form of skill acquisition and participation in kaizen and related “intellectual skills” (Koike 1996) lead to opportunities for blue-collar workers to participate or even drive digitalisation projects. This results in solutions that are largely focused on resolving concrete shop floor issues along the lines of kaizen routines. Simultaneously, there is little co-ordination of skill requirements across the industry – at best this seems to be limited to keiretsu groups.
In Germany, the emphasis on certified qualifications has several consequences. First, digitalisation projects must be expected to be largely initiated and driven by white-collar experts in a top-down manner. In other words, digitalisation is construed by requiring academic qualifications, not vocational ones. Second, the still co-ordinated nature of vocational training and job descriptions means that the said descriptions can effectively only be changed with union consent. Due to the interests of organised labour, it must be expected that deskilling is rejected and instead upskilling is promoted. Third, the co-ordinated nature of the German skill formation system with comparatively strong industrial unions means that co-ordination – not to speak of the implementation of novel training contents – will require time.
Thus, systematic, structural differences identified by the VoC framework and skill formation systems research provide an analytical lens that helps to explain different approaches to digitalisation in Germany and Japan. However, it needs to be emphasised that more empirical research on use cases seems necessary in order to solidify or disproof the existence of systematic intra-country differences as an explanation. Research on the automotive industry provides strong evidence that intra-firm differences may be significant not only for intra- but also for inter-country comparisons (Freyssenet et al. 1998; Freyssenet 2009), suggesting that more inter-country, intra-firm comparative research of digitalisation approaches is required.
Acknowledgements
We would like to thank Ulrich Jürgens (WZB) for detailed and constructive comments on earlier versions of the manuscript.
Abbreviations
| AI | Artificial Intelligence |
| BIBB | Bundesinstitut für Berufsbildung (Federal Institute for Vocational Education and Training) |
| CME | Co-ordinated Market Economy |
| CVET | Continuing Vocational Education and Training |
| IVET | Vocational Education and Training |
| LME | Liberal Market Economy |
| OEM | Original Equipment Manufacturer |
| OJT | On-The-Job Training |
| VoC | Varieties of Capitalism |
References
Aoki, Katsuki. 2008. “Transferring Japanese kaizen activities to overseas plants in China.” International Journal of Operations & Production Management, 28 (6), pp. 518–539.
Aoki, Katsuki, Rick Delbridge, and Takahiro Endo. 2014. “‘Japanese Human Resource Management’ in Post-bubble Japan.” The International Journal of Human Resource Management, 25 (18), pp. 2551–2572.
Baccaro, Lucio and Chris Howell. 2017. Trajectories of Neoliberal Transformation. European Industrial Relations Since the 1970s. Cambridge: Cambridge University Press.
Bateman, Nicola and Nick Rich. 2003. “Companies’ Perceptions of Inhibitors and Enablers for Process Improvement Activities.” International Journal of Operations & Production Management, 23 (2), pp. 185–199.
BMWi. 2018. “Ausbildung und Industrie 4.0: Zupacken statt warten in der Metall- und Elektroindustrie.” Online: https://www.bmwk.de/Redaktion/DE/Pressemitteilungen/2018/20180608-ausbildung-industrie-4-0-zupacken-statt-zuwarten-in-der-metall-und-elektroindustrie.html (accessed: March 12, 2021).
Briken, Kendra, Shiona Chillas, Martin Krzywdzinski, and Abigail Marks. 2017. “Labour Process Theory and the New Digital Workplace.” In Abigail Marks, Kendra Briken, Shiona Chillas, and Martin Krzywdzinski, eds., The New Digital Workplace: How New Technologies Revolutionise Work. London: Red Globe Press, pp. 1–17.
Busemeyer, Marius R. and Christine Trampusch, eds. 2012. The Political Economy of Collective Skill Formation. Oxford: Oxford University Press.
Busemeyer, Marius R. and Christine Trampusch. 2013. “Liberalization by Exhaustion: Transformative Change in the German Welfare and Vocational Training System.” Zeitschrift für Sozialreform, 59 (3), pp. 291–312.
Culpepper, Pepper D. 2005. “Institutional Change in Contemporary Capitalism: Coordinated Systems since 1990.” World Politics, 57 (2), pp. 173–199.
Doellgast, Virginia and Ian Greer. 2007. “Vertical Disintegration and the Disorganisation of German Industrial Relations.” British Journal of Industrial Relations, 45 (2), pp. 55–76.
Dore, Ronald. 1985. “Authority and Benevolence: The Confucian Recipe for Industrial Success.” Government and Opposition, 20 (2), pp. 196–217.
Dore, Ronald and Mari Sako. 1998. How the Japanese Learn to Work, 2nd edition. London and New York: Routledge.
Estevez-Abe, Margarita, Torben Iversen, and David Soskice. 2001. “Social Protection and the Formation of Skills: A Reinterpretation of the Welfare State.” In Peter A. Hall and David Soskice, eds., Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press, pp. 145–183.
Freyssenet, Michel, Andrew Mair, Koichi Shimizu, and Giuseppe Volpato, eds. 1998. One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press.
Freyssenet, Michel, ed. 2009. The Second Automobile Revolution: Trajectories of the World Carmakers in the 21st Century. Houndmills, UK: Palgrave Macmillan.
Gebhardt, Jonas, Axel Grimm, and Laura M. Neugebauer. 2015. “Developments 4.0 – Prospects on Future Requirements and Impacts on Work and Vocational Education.” Journal of Technical Education, 3 (2), pp. 117–132.
Gerst, Detlef, Thomas Hardwig, Martin Kulmann, and Michael Schumann. 1999. “Group Work in the German Automobile Industry – The Case of Mercedes-Benz.” In Jean-Pierre Durand, Paul Stewart, and Juan José Castillo, eds., Teamwork in the Automobile Industry. Houndmills, UK: Palgrave Macmillan, pp. 366–394.
Gordon, Andrew. 1985. The Evolution of Labor Relations in Japan: Heavy Industry, 1853–1955. Cambridge, MA and London: Council on East Asian Studies, Harvard University and Harvard University Press.
Haipeter, Thomas, Ulrich Jürgens, and Karin Wagner. 2012. “Employment Relations in the Banking and Automotive Industries in Germany.” The International Journal of Human Resource Management, 23 (10), pp. 2016–2033.
Hall, Peter A. and David Soskice, eds. 2001. Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press.
Härtel, Michael. 2017. “Germany: BIBB.” In Colin Latchem, ed., Using ICTs and Blended Learning in Transforming TVET. Paris and Burnaby, BC: UNESCO and Commonwealth of Learning, pp. 59–72.
Hassel, Anke. 1999. “The Erosion of the German System of Industrial Relations.” British Journal of Industrial Relations, 37 (3), pp. 483–505.
Holst, Hajo, Katsuki Aoki, Gary Herrigel, Ulrich Jürgens, Takefumi Mokudai, Marvin Müller, Carsten Schaede, Martin Schröder, and Robert Sinopoli. 2020. “Gemba-Digitalisierung: Wie japanische Automobilunternehmen IoT-Technologien einsetzen.” ZWF. Zeitschrift für Wirtschaftlichen Fabrikbetrieb, 115 (9), pp. 629–633.
Howell, Chris. 2003. “Varieties of Capitalism: And Then There Was One?” Comparative Politics, 36 (1), pp. 103–124.
Hyodo, Tsutomo. 1987. “Participatory Management and Japanese Workers’ Consciousness.” In Joachim Bergmann and Shigeyoshi Tokunaga, eds., Economic and Social Aspects of Industrial Relations: A Comparison of the German and Japanese Systems. Frankfurt am Main and New York: Campus Verlag, pp. 261–270.
Ishida, Mitsuo. 1997. “Japan: Beyond the Model for Lean Production.” In Thomas A. Kochan, Russell D. Lansbury, and John Paul MacDuffie, eds., After Lean Production: Evolving Employment Practices in the World Auto Industry. Ithaca, NY and London: Cornell University Press, pp. 45–60.
Jürgens, Ulrich. 1998. “The Development of Volkswagen’s Industrial Model, 1967–1995.” In Michel Freyssenet, Andrew Mair, Koichi Shimizu, and Giuseppe Volpato, eds., One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 273–310.
Jürgens, Ulrich and Martin Krzywdzinski. 2015. Neue Arbeitswelten. Wie sich die Arbeitsrealität in den Automobilwerken der BRIC-Länder verändert. Frankfurt am Main: Campus Verlag.
Jürgens, Ulrich, Thomas Malsch, and Knuth Dohse. 1993. Breaking from Taylorism: Changing Forms of Work in the Automobile Industry. Cambridge: Cambridge University Press.
Kaneko, Motohisa. 2019. “The Formulation of Professional and Vocational Universities: Background and Challenges of a New Institutional Type in Japan.” Japan Labor Issues, 3 (13), pp. 30–42.
Koike, Kazuo. 1994. “Learning and Incentive Systems in Japanese Industry.” In Masahiko Aoki and Ronald Dore, eds., The Japanese Firm. Sources of Competitive Strength. Oxford: Oxford University Press, pp. 41–65.
Koike, Kazuo. 1996. The Economics of Work in Japan. Tokyo: LTCB International Library Foundation.
Kubo, Katsuyuki. 2020. “Changes in the Labour Market and Employment Relationships in Japan.” In Byoung-Hoon Lee, Sek-Hong Ng, and Russell D. Lansbury, eds., Trade Unions and Labour Movements in the Asia-Pacific Region. Oxon and New York: Routledge, pp. 148–165.
Kume, Ikuo 1998. Disparaged Success: Labor Politics in Postwar Japan. Ithaca, NY and London: Cornell University Press.
Labit, Anne. 1999. “Group Working at Volkswagen: An Issue for Negotiation between Trade Unions and Management.” In Jean-Pierre Durand, Paul Stewart, and Juan José Castillo, eds., Teamwork in the Automobile Industry. Houndmills, UK: Palgrave Macmillan, pp. 395–411.
Lechevalier, Sébastien. 2014. “Is Japanese Capitalism Still Co-ordinated?” In Sébastien Lechevalier, ed., The Great Transformation of Japanese Capitalism. Oxon and New York: Routledge, pp. 73–85.
Mahoney, James and Kathleen Thelen, eds. 2010. Explaining Institutional Change: Ambiguity, Agency, and Power. Cambridge: Cambridge University Press.
Marsden, David. 1999. A Theory of Employment Systems: Micro-Foundations of Societal Diversity. Oxford: Oxford University Press.
Marsden, David. 2015. “The Future of the German Industrial Relations Model.” Journal of Labour Market Research, 48 (2), pp. 169–187.
MHLW厚生労働省. 2018. 我が国の構造問題・雇用慣行等について [About Japan’s Structural Problems, Employment Practices, etc.]. Tōykō 東京: MHLW 厚生労働省. Online: https://www.mhlw.go.jp/content/11601000/000358251.pdf (accessed: April 2, 2022).
Miura, Mari. 2008. “Labor Politics in Japan During ‘The Lost Fifteen Years’: From the Politics of Productivity to the Politics of Consumption.” Labor History, 49 (2), pp. 161–176.
Mokudai, Takefumi, Martin Schröder, Marvin Müller, Carsten Schaede, Hajo Holst, Robert Sinopoli, Ulrich Jürgens, Gary Herrigel, and Katsuki Aoki. 2021. “Digital Technologies as Lean Augmentation: A Preliminary Study of Japanese Automotive Manufacturers.” International Journal of Automotive Technology and Management, 21 (3), pp. 228–249.
Mouer, Ross and Hirosuke Kawanishi. 2005. Sociology of Work in Japan. Cambridge: Cambridge University Press.
Müller-Jentsch, Walther. 1999. “Berufsbildung – eine Arena der industriellen Beziehungen?” In Klaus Harney and Heinz-Elmar Tenroth, eds., Beruf und Berufsbildung. Situation, Reformperspektiven, Gestaltungsmöglichkeiten. Weinheim and Basel: Beltz, pp. 233–248.
Nemoto, Masao 根本 正夫. 1992. TQC Seikō no Hiketsu 30 Kajō TQC成功の秘訣30カ条 [30 Secrets of TQC Success]. Tōkyō 東京: Nikagiren Shuppansha 日科技連出版社.
Ogino, Noboru. 2021. “What is Shunto?” Japan Labor Issues, 5 (28), pp. 18–21.
Olejniczak, Tomasz, Milosz Miszczynski, and Masato Itohisa. 2020. “Between Closure and Industry 4.0: Strategies of Japanese Automotive Manufacturers in Central and Eastern Europe in Reaction to Labour Market Changes.” International Journal of Automotive Technology and Management, 20 (2), pp. 196–214.
Padur, Torben and Gert Zinke. 2015. “Digitalisierung der Arbeitswelt – Perspektiven und Herausforderungen für eine Berufsbildung 4.0.” Berufsbildung in Wissenschaft und Praxis, 44 (6), pp. 30–32.
Pries, Ludger. 2004. “New Production Systems and Workers’ Participation: A Contradiction? Some Lessons from German Automobile Companies.” In Elsie Charron and Paul Stewart, eds., Work and Employment Relations in the Automobile Industry. Houndmills, UK and New York: Palgrave Macmillan, pp. 76–102.
Rebick, Marcus. 2005. The Japanese Employment System: Adapting to a New Economic Environment. Oxford: Oxford University Press.
Roth, Siegfried. 1997. “Germany: Labor’s Perspective on Lean Production.” In Thomas A. Kochan, Russell D. Lansbury, and John Paul MacDuffie, eds., After Lean Production: Evolving Employment Practices in the World Auto Industry. Ithaca, NY and London: Cornell University Press, pp. 117–136.
Sako, Mari. 2006. Shifting Boundaries of the Firm: Japanese Company-Japanese Labour. Oxford: Oxford University Press.
Sawai, Minoru. 2020. “Vocational Training and Vocational Education in Postwar Japan: An Overview.” Japanese Research in Business History, 37, pp. 1–10.
Schmidt, Vivien A. 2009. “Putting the Political Back into Political Economy by Bringing the State Back in Yet Again.” World Politics, 61 (3), pp. 516–546.
Sengenberger, Werner. 1987. “Vocational Worker Training, Labour Market Structure, and Industrial Relations in West Germany.” In Joachim Bergmann and Shigeyoshi Tokunaga, eds., Economic and Social Aspects of Industrial Relations: A Comparison of the German and Japanese Systems. Frankfurt am Main and New York: Campus Verlag, pp. 246–260.
Shibata, Saori. 2016. “Resisting Japan’s Neoliberal Model of Capitalism: Intensification and Change in Contemporary Patterns of Class Struggle.” British Journal of Industrial Relations, 54 (3), pp. 496–521.
Shimizu, Koichi. 1994. “Humanization of the Production System and Work at Toyota Motor Co and Toyota Motor Kyushu.” In Åke Sandberg, ed., Enriching Production: Perspectives on Volvo’s Uddevalla Plant as an Alternative to Lean Production. Aldershot, UK: Avebury, pp. 383–402.
Streeck, Wolfgang. 1984. Industrial Relations in West Germany: A Case Study of the Car Industry. London: Heinemann.
Tenberg, Ralf and Daniel Pittich. 2017. “Ausbildung 4.0 oder nur 1.2? Analyse eines technisch-betrieblichen Wandels und dessen Implikationen für die technische Berufsausbildung.” Journal of Technical Education, 5 (1), pp. 27–45.
Thelen, Kathleen. 1991. Union of Parts: Labor Politics in Postwar Germany. Ithaca, NY and London: Cornell University Press.
Thelen, Kathleen. 2004. How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan. Cambridge: Cambridge University Press.
Thelen, Kathleen. 2014. Varieties of Liberalization and the New Politics of Social Solidarity. Cambridge: Cambridge University Press.
Thelen, Kathleen and Ikuo Kume. 1999. “The Rise of Nonmarket Training Regimes: Germany and Japan Compared.” Journal of Japanese Studies, 25 (1), pp. 33–64.
Thelen, Kathleen and Ikuo Kume. 2001. “The Rise of Nonliberal Training Regimes: Germany and Japan Compared.” In Wolfgang Streeck and Kozo Yamamura, eds., The Origins of Nonliberal Capitalism: Germany and Japan in Comparison. Ithaca, NY and London: Cornell University Press, pp. 200–228.
Thelen, Kathleen and Ikuo Kume. 2006. “Coordination as a Political Problem in Coordinated Market Economies.” Governance: An International Journal of Policy, Administration, and Institutions, 19 (1), pp. 11–42.
Visser, Jelle and Joris Van Ruysseveldt. 1996. “Robust Corporatism, Still? Industrial Relations in Germany.” In Joris Van Ruysseveld and Jelle Visser, eds., Industrial Relations in Europe: Traditions and Transitions. London: Sage, pp. 124–174.
Walter, Andrew and Xiaoke Zhang. 2012. East Asian Capitalism: Diversity, Continuity, and Change. Oxford: Oxford University Press.
Weathers, Charles. 2003. “The Decentralisation of Japan’s Wage Setting System in Comparative Perspective.” Industrial Relations Journal, 34 (2), pp. 119–134.
Weathers, Charles. 2008. “Shuntō and the Shackles of Competitiveness.” Labor History, 49 (2), pp. 177–197.
Whitley, Richard. 1999. Divergent Capitalisms: The Social Structuring and Change of Business Systems. Oxford: Oxford University Press.
Witt, Michael A. 2006. Changing Japanese Capitalism: Societal Coordination and Institutional Adjustment. Cambridge: Cambridge University Press.
Employment protection means an institutionalised form of employment security, i.e., safeguards against being laid off. Lifetime employment is a typical form of employment protection. Unemployment protection means protection from income reduction caused by unemployment. Hence, welfare systems characterised by high replacement ratios, long benefit duration, and benefit administration that rather prioritises finding a “suitable job” (opposed to any job) characterise high unemployment protection.
It is noteworthy that researchers such as Busemeyer and Trampusch stress different conclusions, depending on whether they take a purely German (historical) perspective that stresses “exhaustion” of vocational training (2013) or a comparative perspective which finds that support for skill formation through vocational training is still high (2012: 12). Since we take a comparative perspective, we follow the conclusion that vocational training still enjoys broad support, but increasingly fails to provide enough apprenticeships for all those interested.
Due to issues with applying the VoC framework to developing countries or regions, alternative frameworks to study East Asian capitalism have been proposed (Walter and Zhang 2012). However, their alternative has almost nothing to say about skill formation, viz. the topic we are mainly interested in. Thus, we develop our discussion based on the VoC framework but acknowledge that it is not a perfect fit. In our view, a main issue is that many Japanese institutions are informal and thus more open to (re-)configuration.
The definition of regular employee in this survey encompasses both employees on indefinite contracts and those that have worked for the same employer for more than one year. Thus, the data set classifies long-term temporary workers as regular employees. Hence, the survey does not clearly distinguish between regular and atypical employees, and obscures the actual extend of atypical employment. Further, this number excludes executives, who were tracked separately until 1997 but are no longer reported by subsequent surveys.
Exemplary for this view is Dore’s (1985) attribution of Japanese industrial success to its industrial relations system and, by extension, Confucian culture.
An important caveat must be highlighted: Marsden’s model seeks to provide a national employment systems model, i.e., inter-industry differences within a country are not theorised. In their recent study on employment and internationalisation at Toyota and Volkswagen, Jürgens and Krzywdzinski (2015: 64) observe with reference to Marsden’s model that Volkswagen is better characterised by a strong focus on job requirements, i.e., training and job design are based on production requirements and firm needs, which may not be too surprising given the complex and capital-intensive nature of the production process.
Despite focusing on continued co-ordination between employers and unions, it is clearly acknowledged that co-ordination in Germany and Japan has promoted dualisation in employment, i.e., skilled core workers are protected against unemployment by the increased use of lower skilled temporary workers and/or outsourcing of formerly internally performed services (Thelen 2014: 7–8, 51).
Regarding the genesis of the German and Japanese systems, see Thelen and Kume 1999 and especially Thelen 2004.
We are indebted to Ulrich Jürgens for suggesting exploring these dualisms. He also suggested to discuss Germany’s dual study programmes, which we chose to exclude from our elaboration.
It is acknowledged that works councils gained rights concerning CVET in the 2001 reform of the Works Constitution Act.
Before the 1969 reform, vocational training contents in West Germany were defined by Chambers of Industry and Commerce, i.e., employer organisations.
Müller-Jentsch (1999: 235) point out that all seats on BIBB’s central decision-making bodies are equally occupied by four parties, namely, the federal government, state governments, employer associations, and unions. As the states are in charge of education, including IVET, they are represented at this federal level institution. In practice, representatives of the federal and state governments follow consensus negotiated by the social partners.
Formally, collective bargaining and co-determination have clearly separated spheres of influence, e.g., works councils are banned from engaging in collective bargaining which is the prerogative of unions. Regarding this dualism, however, Visser and Van Ruysseveldt (1996: 153) aptly remarked that “the formal independence of the unions and works councils is a fiction.” Streeck’s (1984) study on industrial relations at Volkswagen documents overlapping roles and mutual dependence of these two levels of interest representation.
Changing in working hour, i.e., short-time work or overtime, require the approval of the works council. This power has been identified as the single-most important tool for works councils to extend industrial interest representation to issues such as training or technological change, that is, areas in which works councils have formally weak co-determination rights (Streeck 1984: 110–114; Visser and Van Ruysseveldt 1996: 155).
Digital technologies offer new opportunities to support CVET and other forms of lifelong learning. However, recent case studies on the use of technologies and open information platforms (Härtel 2017) suggest that use cases are either concentrating on making IVET more effective, connecting firm trainers, vocational school teachers, and apprentices, or provide mobile access to data and contents for specific skilled trades. Thus, the opportunities for CVET may indeed continue to receive limited employer support and depend on individual initiative of employees.
It should be emphasised that this delimitation is in line with existing scholarship on VoC and similar comparative institutional analyses (Whitley 1999; Hall and Soskice 2001; Thelen 2004). Although vocational training directly organised or accredited by the Japanese state does exist (Dore and Sako 1998; Sawai 2020), it has received scant academic investigation. This may be ascribed to the fact that nearly all researchers who address this topic identify lower social status vis-à-vis firm-specific training conducted by large firms and tertiary education (Kaneko 2019: 35; Sawai 2020: 6). The 2017 decision to create a new institutional category labelled “professional and vocational universities” can be regarded as an attempt to increase the attractiveness of non-firm-specific vocational training (Kaneko 2019).
While there are industry level standards and related qualifications, earning qualifications is less linked to improving individuals’ marketability or improving labour market efficiency and more related to improving the efficiency of the organisation to which individual test takers belong (Dore and Sako 1998: 134). It follows that earning qualifications is driven by firm-specific motivations, with firms being relatively unconstrained by deciding what qualifications should be pursued by employees.
We chose this formulation to account for two issues. First, as an informal institution, lifetime employment is not enshrined in employment contracts. Hence, quantifying lifetime employment is rather difficult. One potential method is to track the average length of employment. However, official data are vaguely formulated, therefore complicating the interpretation. Data from the Ministry of Health, Labour and Welfare (MHLW 2018: 21) do not report average length of employment but rather use the category “Those who have joined the employer at a young age (meaning eighteen to twenty years for high school graduates, twenty-two to twenty-four for university graduates) and who continue to work for this employer.” This category does not provide concrete quantification of average length of employment and instead describes general trends. For university graduates, the share of individuals working for the same employer for a long time is roughly 50 percent, but for high school graduates, this share is only about thirty per cent. Second, data on average length of employment reveal strong differences between industries. For university graduates in the financial sector, long length of employment is observed for eighty per cent of employees, but this percentage in the medical sector is only forty per cent. For manufacturing industries, including the automotive industry, long-term employment is observed for nearly sixty per cent of employees.
Academic articles discussing the relationship between digitalisation and industrial relations are surprisingly rare in Japan. We conducted a systematic literature search for the scholarly articles from four Japanese management journals, namely, nihon keiei gakkasishi
Reviewing the development of kaizen activities (discussed as participatory management), Hyodo (1987: 264–265) observed that Japanese firms introduced these activities in the early 1960s to increase workers’ motivation to work and loyalty towards their employers.
This subsection uses and extends findings from a preliminary study on digitalisation in Japanese automotive firms (Holst et al. 2020; Mokudai et al. 2021). We shift the focus from the usage of digital technologies towards the following question: to what extend can the deployment of said technologies be regarded as consistent with the literature on VoC and skill formation systems? Regarding the methodology and findings of the cases studies, the interested reader can refer to those studies.
Comparing skill formation and certification of qualifications in Great Britain and Japan, Dore and Sako (1998: 158) observed that the British approach seeks to define standard occupational roles. Conversely, the Japanese approach to certification is akin to boy scout badges, i.e., there are various skills which can be combined in a near infinite number of ways. Thus, contrary to the British and German approach of defining standard roles which individuals with certified qualifications are ought to fulfil, the Japanese approach allows individuals to combine various certified skills in a way they or their employer deem(s) appropriate for their work. This absence of standard roles in Japan may thus explain why Japanese firms are letting employees experiment with new technologies to find firm-specific solutions.
Roth’s observations on the results of these two practices are noteworthy. The expert-led approach is said to be prone to failure as initial improvements cannot be maintained. Conversely, the opportunity to also address social aspects of work is said to lead to more sustainable improvements. Interestingly, these observations are remarkably similar to those by Aoki (2008) on the transfer of kaizen activities from Japan to China, in that successful cases allowed workers to address both economic improvements and advancements on how work was carried out. Finally, the two different kaizen approaches have been discussed under the labels of (short-term) process improvement and (long-term) continuous improvement by operations research (Bateman and Rich 2003: 186).
