Shopfloor Apprentices under a Lean Labour Process: Insights from a British Car Plant

Abstract

This article considers how lean production changes the shopfloor learning experience of apprentices. Insights from labour process and communities of practice literature are used to interrogate the study. Evidence derives from apprentices’ experiences in a British car assembly plant across several decades. The study compares pre-lean apprentices’ experiences with those who learn under lean, considering how lean might reshape apprentices’ shopfloor interactions with the plant community of practice. It finds that lean appears unfavourable to apprentices’ integration into the community of practice by debilitating learning opportunities with mentors. The article concludes by theorising how inefficiency and waste may aid on-the-job learning for workplace apprentices.

Introduction

Lean production, and its impact on labour processes, is of continuing interest to scholars. However, within the lean debate (Carter et al., 2016; Vidal, 2022), little attention is given to experiences of workplace apprentices. Apprentices represent a distinctive cohort within the labour process. Via on-the-job learning, the novice apprentice acquires not just technical skillsets, but also customs associated with a craft, occupation or profession (Clarke et al., 2013; Lehmann and Taylor, 2015; Vickerstaff, 2003). Experienced mentors nurture that development at the point of production. Some conceptualise the process as socialisation into ‘Communities of Practice’ (CoP) (Grugulis and Stoyanova, 2011; Lave and Wenger, 1991). CoP literature seeks to understand knowledge sharing and learning within diverse social groups. Emphasis is on situated informal and experiential learning, particularly the process by which newcomers become full participants in the community over time by engaging in meaningful, albeit initially peripheral, activities. Newcomers progressively move from the ‘periphery’ to the ‘core’ of the community via interactions with established members who school them in the relevant domain (Brooks et al., 2020). Although the CoP literature is broad-ranging – Reddit users (Haythornthwaite et al., 2018) and anarchist collectives (Hemphill and Leskowitz, 2013) are some of the cohorts who come under its purview – it aligns with similar ideas prevalent in the sociology of apprenticeship literature (Penn, 1990). It can also contribute to labour process research: much labour process literature on lean is silent on how apprentices become immersed into a CoP of skilled trades, while literature on novices’ integration into CoP is silent on the material constraints of capitalist production.

This article considers apprentices’ experiences within a plant CoP of skilled trades under lean production. It addresses these issues via a case study of engineering apprentices’ experiences in a British-based lean car plant across time. While the article justifies a single case below, it uses interviews with over 60 plant respondents alongside plant documentary material across several decades to compare the experiences of apprentices before the pivot to lean with the later experiences of those who learn under lean.

The findings suggest the lean model is unfavourable to apprentices’ integration into the CoP of the plant’s skilled trades because it reduces learning opportunities. The article’s contribution is to unpack how various elements of lean contribute to such outcomes, proposing that inefficiency and waste may be conducive to supporting on-the-job learning. The article begins by reviewing the relevant issues from the literature more fully. It then describes the context of the case study and fieldwork methods before analysing the fieldwork findings. The article’s conclusion discusses theoretical and policy implications.

Literature

Lean serves as an important reference point in studies of labour processes (Smith and Vidal, 2021; Stewart et al., 2016). Within Marxian labour process traditions, scholars conceptualise lean as a ‘force of production’: the totality of means of production and labour power (Adler, 2007; Vidal, 2022). Vidal (2022: 10) outlines the ‘core package’ of lean as ‘just-in-time production (demand-driven production with low inventories), continuous flow organisation, waste elimination, teamwork and continuous improvement’ (see also Smith, 2000). However, labour process literature, with its predominant focus on front-line production workers under lean, follows two different trajectories. One, in ‘Bravermanian’ fashion, emphasises lean as degrading of labour, inducing deskilling and work intensification (Carter et al., 2013, 2016; Stewart et al. 2016). This perspective casts lean as an employer control structure deployed in the class struggle, stamping out ‘porosity’ in the working day via raising labour effort. Lean is also seen as an ideological reflection of ‘neo-liberalism’ in the workplace (Stewart, 2020).

An alternative perspective comes from Adler (2007) and Vidal (2019; 2022). Adler accepts lean can degrade workers but argues this is an implementation problem, not something endogenous to the model. Indeed, drawing on Classical Marxism, Adler (1990) sees a long-run technology-led upskilling trajectory, with lean production reflecting the progressive socialisation of the productive forces. Vidal (2022) distinguishes between lean as a ‘model’ and lean enacted (e.g. some managers adopt a toolbox approach). He argues that labour intensification is not innate to the package; it is contingent on how lean is applied by particular management strategies on the shopfloor. He argues lean can be implemented in a fashion akin to Friedman’s (1977) responsible autonomy to include participation and multi-skilling. That can occur, in some cases, without intensifying work. Nevertheless, he contends that capitalists often satisfice and fail to realise the potential gains of the model. Even when management overcomes such hurdles, efforts fall flat because they rely on alienated labour. In the Classical Marxist style, Vidal (2022: 14) argues that the ‘relations’ fetter the ‘forces’.

One commonality across labour process literature on lean is the relative neglect of experiences of skilled trades (there are, of course, labour process studies of skilled trades, like Cooley, 1987; Smith, 1987). A corollary is neglect of how apprentices become inculcated into skilled trades during developmental placements in labour processes. Apprenticeships vary widely across countries, but the commonality is training new generations of trade practitioners with shopfloor experiences alongside classroom study (Chankseliani et al., 2017). In lean literature, one finds limited references to apprenticeships. Teague (1997) theorises lean hollowing out apprenticeships by emphasising firm-specific practical know-how and soft behavioural competencies rather than generalisable theoretical knowledge. Jurgens (2008) notes that post-apprenticeship placements are lacking in the German automotive industry, attributed to lean rationalisation of indirect jobs. From a sociological perspective, the relative absence of research on apprentices’ socialisation within the lean labour process is a gap. As a cohort, apprentices embody a nexus of dynamics where workplace knowledge, skills, occupational norms, values and behaviours are transmitted and reproduced. Indeed, apprenticeship learning is not just technical skill acquisition but socialisation into occupational, craft and workplace customs (Clarke et al., 2013; Holmes, 2015). That occurs at the point of production when apprentices connect with established practitioners like senior craft workers who foster craft consciousness (Penn, 1990).

A helpful way to conceptualise that dynamic is the CoP literature, a heuristic lens to understand how newcomers are inculcated into a community by seasoned practitioners (Lave and Wenger, 1991). Unlike labour process theory, which primarily focuses on the control mechanisms workers experience when selling their labour power, the range of groups studied in CoP is broad. Literature on CoP explores themes like shared practices, learning and identity formation within diverse groups (Brooks et al., 2020; Grugulis and Stoyanova, 2011). CoP are informal and experiential learning contexts where members share a common interest, goal or practice. A key emphasis is social interaction in acquiring knowledge and skills (i.e. ‘situated learning’) (Lave and Wenger, 1991: 31). Learning occurs through participation in CoP, where individuals share knowledge and collaborate to solve problems, providing opportunities to observe, imitate and receive feedback. Instead of being passively transmitted from teacher to student, learning happens through engagement in activities: in a process termed ‘legitimate peripheral participations’ (Lave and Wenger, 1991: 29), novices begin with minor roles, before gradually taking on more complex tasks. Mentorship supports that transition.

While some studies find evidence of stunted growth opportunities for novices where short-term contracts in the labour market hinder learning processes (Grugulis and Stoyanova, 2011), the insights of CoP rarely appear in studies of the labour process. As Engeström (2007: 43) notes: ‘[o]ne looks in vain for discussions on the conditions of implementing communities of practice in highly rationalized hierarchical mass production organizations’. We do know firms’ structuring of labour processes is important for on-the-job skill use (Grugulis and Vincent, 2005; Lloyd and Payne, 2017). Of relevance is Tyler and Wilkinson (2007), who conceptualise slimmed-down, hollowed-out firms as weakening how valuable knowledge is transferred or retained. They theorise ‘anorexic’ firms who outsource and shed labour. This may result in the loss of established practitioners who act as a transmission belt for apprentice learning. The hollowing out of established cohorts could result in a loss of organisational memory (McBride and Lucio, 2011), disrupting novices’ ability to draw on an available reservoir of craft knowledge and traditions. Sociological meta-narratives support this idea that capitalist rationalisation tends to value output over the social processes of learning and collaboration (Sennett, 2009), eroding tacit knowledge (Braverman, 1974).

However, it is unclear which facets of the lean model, like those identified by Vidal above, are corrosive of apprenticeship immersion into CoP. Indeed, lean may not be detrimental to the apprenticeship experience; a propensity for lean to rely on teamwork may compensate for some of the adverse effects of rationalisation highlighted in the preceding paragraph by providing collective mentoring. Much remains unknown about how lean interacts with an existing CoP and the inculcation of new members. The remainder of the article explores these uncertainties by asking how might the lean labour process influence apprentices’ experiences within shopfloor communities of practice – what might it be about the lean labour process that changes apprentices’ experience, if anything?

The case

The article draws on a single revelatory case study (Yin, 2014: 38, 40), useful for exploring unaccounted-for dynamics (Ridder, 2017). The case was an English-based, foreign-owned automotive assembly plant operative for the best part of six decades in Britain. It employed approximately 10,000 workers throughout its first two decades. Recession during the 1980s saw the closure of some facilities, like an on-site foundry, and the loss of manufacturing capacity, like gearboxes, although, in the context of continued investment, new robotic technologies were implemented. That era heralded a downward trajectory in numbers employed, a process that has continued due to vertical disintegration and technological change. Given the research interests, the plant was relevant as an early adopter of lean. Company documentation from 1991, called ‘Becoming a Lean Organisation’, indicates 1990 as a kind of ‘year zero’ for lean implementation.

Consequently, the plant turned ‘leaner’, seeking to eliminate excess inventory not in circulation. Worker teams played a role in kaizen – or ‘Sack a Mate’ schemes, as known by the unions at the plant level, now amalgamated under Unite the Union. The turn to lean was part of a standardisation process enabling ‘whipsawing’ (Greer and Hauptmeier, 2016) with other international plants in the company, resulting in union concessions on shopfloor arrangements. The 2000s saw significant declines in employment due to losing press lines and production shifts. The total numbers employed were 1100 employees; just over 100 were apprenticed skilled workers, with an 80% decline in this cohort’s headcount since 1990. The plant can build 170,000 vehicles per year but was building 72,000 annually during the fieldwork period of 2016–2022 due to Brexit and Covid disruptions. The site’s labour process replicates those of volume assembly plants in the UK, including standard job design, technologies (3-axis robots), departmental shops of stamping and press, body welding, painting, final assembly and trim.

Concerning apprenticeships, the plant utilises the industry standard found in other car producers in the UK. The UK apprenticeship context is where private providers, like Further Education (FE) colleges, ‘sell’ government funding for various training schemes like apprenticeships to employers (Keep, 2018). Scholars highlight variation in the quality of UK apprenticeships and note limited co-involvement of ‘social partners’; the system is employer-led (Brockman et al., 2010). There is diversity in off-the-job training, with a mix of employer in-house training and external provider delivery. In the UK, most apprentices aim for the equivalent of lower secondary qualifications (Level 2) rather than craft or intermediate Level 3 awards (upper secondary equivalent). Engineering, however, is recognised as an area of strength in terms of rigour and prestige in the provision of UK apprentices (Fuller and Unwin, 2013: 27). In engineering, a Level 3 National Vocational Qualification (NVQ) includes a four-year initiative under the auspices of an apprentice training programme, formerly the Engineering Industry Training Board (EITB) and, as of 2015, the Automotive Trailblazer Programme. The Trailblazer requires learners to ‘apply Lean Manufacturing Principles’ via Behavioural Anchored Rating Systems (BARS), a type of soft skills assessment (Automotive Trailblazer: Level 3 Apprenticeship Mechatronics Standard: 2). The typical industry programme comprises on- and off-the-job training. The first year involves learning basic engineering skills at an FE facility. The three remaining years are spent on the job, interspersed with attendance at an FE college. While engineering apprenticeships are well respected, there have been long-standing criticisms of the NVQ approach to workplace learning: a top-down national approach to developing intermediate engineering skills is said to be inflexible in enabling workers to cope with organisational and technological change (Senker, 1996).

During the 1970s and much of the 1980s, the plant had 30 new apprentices each year. From 1990, six apprentices per year was the norm, and, in some years, the hiring of apprentices stopped due to a lack of company demand. Driving this decline were the aforementioned trends influencing drops in headcount. The range of apprenticeships also declined. At the start of the 1980s, the plant offered nine different apprenticeship types. From 1992, that compressed into two. One, a multi-skilled mechatronics apprenticeship, combining electrical and mechanical training – although in practice weighted towards one or the other – and a second, a tool and die maintenance apprenticeship. While multi-skilled apprentices rotate across departments, die maintenance apprentices are based in Stamping. ¹

Method

We focused on apprenticeships who had undertaken their training before lean and those who had undertaken their apprenticeships after the plant pivoted to lean, a demarcation explained below. The study drew a target list of 96 potential respondents who had apprenticed at the plant from management and union-level contacts. That sampling frame spanned a period of in-plant apprenticeship completion from 1977 to 2018. The research interviewed 48 of these targets. Interviews were recorded and lasted between 60 and 90 minutes.

Of the 48 respondents, 22 completed apprenticeships between 1977 and 1987. The case used material from these 22, alongside documentary material and additional context interviews, to construct a pre-lean ‘benchmark’. That ‘benchmark’ has a stylised demarcation: the pivot to lean was not overnight. However, analysis must start somewhere to tease out differences across time. The pre-lean respondents took a variety of apprenticeships, either mechanical, electrical or tool/die fitting. All 22 respondents were male, typically starting their training at 16, coming straight from Ordinary Level educational attainment. Two of the 22 undertook die maintenance apprenticeships, 15 electrical and five mechanical. That pre-lean cohort enabled one to establish a point of comparison in understanding if, and how, apprenticeship experiences later changed under lean. The remaining 26 interview respondents undertook apprenticeships at the plant from 1992 to 2015: the lean period. While these were ‘multi-skilling’ insofar as they incorporated a mix of electrical and mechanical elements, apprentices heavily majored in one over the other. Of the 26, just one was female and they typically started their apprenticeships at age 19, with 18 of the 26 already possessing a FE Business Technology Education Council (BTEC) qualification. Two respondents in the lean cohort undertook die maintenance apprenticeships, the remainder multi-skilling.

Interviews discussed apprentices’ experiences of initial entry onto the shopfloor, who they spoke to, what about and how often. We also examined the organisation of their shifts, the type of work and technologies they engaged with, and the assessment of support from skilled workers for completing learning logbooks. Respondents were questioned on their interactions with skilled workers and how they worked with them on shopfloor tasks. Interviews also asked about post-apprenticeship transition experiences, such as the extent of ongoing mentoring. As apprentices from the pre-lean period had later experiences supporting apprentices under lean, interviews discussed their comparative insights.

One concern with this method is recall. Poorly remembered details and the passage of time may bias reflections, although such recall methodologies are used in other studies of apprentices to good effect (Ergun and Sayfutdinova, 2021; Vickerstaff, 2003). We advance two points on the matter. First, the study found that respondents independently displayed consistency in their reported experiences. Second, respondents’ assessments were balanced: neither offering excessively ‘rose-tinted’ or wholly negative assessments. Reports had an intuitive face validity to the researchers who combined prior experiences researching and working in car assembly plants.

The study also interviewed 15 individuals based on their connection with managing apprenticeships at the plant. That included two senior union convenors who knew both periods, a sectoral union officer, an FE external assessor, and three HR managers who knew of both periods. The remainder were skilled workers who time-served elsewhere but acted as mentors to apprentices in the lean period. In total, the case study comprises 63 interview respondents.

The study also accessed documentation to inform case construction. Documents were from company circulars, the company newspaper archive, workplace agreements from the 1970s onwards, minutes of plant joint consultative committees, documents detailing sectoral apprenticeship schemes, national negotiation and consultation committee meeting minutes, company agreements, nine wage agreement books, plant negotiation and consultation meeting minutes and plant agreements. Some of the former local craft union branch archive, Amalgamated Union of Engineering Workers (AUEW), was also made available for the research. The documentary material helped add background and context.

Finally, evidence was analysed using NVivo v1.0. After uploading transcripts, we attached specific parameters to the ‘data’. Respondents were attributed tags, like ‘pre-lean’ or ‘lean’ groupings, and codes set up with relevant concepts like ‘opportunities for apprentice observation of skilled worker at work’, ‘opportunities for apprentice discussion with skilled workers’ and ‘apprentice assigned tasks’. Concepts were sub-themed: ‘opportunities for apprentice discussion with skilled workers’ were sub-coded into ‘frequency of interaction’, ‘duration of interaction’, and so on. When complete, the analysis mined the collated material for thematic links. The authors collectively discussed any patterns to ensure interpretations converged.

Findings

Apprenticeship experiences before lean

In the pre-lean context, apprentices’ interactions with the skilled community began in Year Two after the first year of classroom training. When arriving on the shopfloor, apprentices discussed their learning logbook with the most senior skilled worker in the department. Respondents recounted how, together with that senior tradesman, they would establish what work was available in the department and who the best-skilled man might be to assist in progressing a particular component of their logbook. This process repeated across the 12-week cycle of inter-departmental rotations. Shopfloor dynamics in this context involved apprentices following a skilled worker to complete a task like addressing equipment failure. The skilled worker would explain the breakdown problem at the relevant location and discuss the solution during or after completing the work.

Consistent with CoP literature discussed earlier, respondents indicated that initially they would be treated very much on the periphery, ‘as skivvies or gophers, doing the jobs that the skilled guys didn’t want to do’ (Respondent 1, Die Maintenance). Over time, however, as relationships built, and apprentices proved their ‘willingness to learn . . . the skilled man would become receptive’ (Respondent 2, Mechanical).

Those respondents taking their apprenticeship in very specialist trades, like die maintenance, resided in one allocated department for the entirety of their apprenticeship, enabling greater opportunities for developing relationships with mentors. Respondents reported how being confined to one department enabled working consistently with the same mentor across time, as opposed to those experiencing departmental rotation. Given the nature of their work, die maintenance and repair were often idiosyncratic and time-consuming for each unique die. A respondent commented:

The scheduling department would schedule all the jobs for the department. . . . they released dies to you for up to 28 days. So, you got to work with the same skilled worker for the majority of the time, learning from him about the best ways of repairing a die. (Respondent 3, Die Maintenance)

Respondents described a common exercise for skilled workers to bring relevant, unused or broken-down machinery or sub-components into the ‘crib’ (tearoom) or an offline departmental workshop for apprentices to work with. One respondent reported:

We had good support from most of the skilled lads; there was one electrician in the Paint Shop . . . who would make [electrical] control circuits [for the apprentices], and he would disconnect a couple of wires, and in the crib you had to fault find and explain what you were doing and why, then he’d ask you if there was a different way [to fix the fault], then he’d tell you the way he’d do it. (Respondent 4, Electrical)

When equipment was down, respondents recounted how the skilled workers would inform apprentices and encourage them to spend time experimenting with such artefacts. A respondent explained:

. . . when I was an apprentice, there were balancing machines in the tyre house [part of the Assembly] . . . when the balancing machine went down, you kept it down for a few hours and used the backup machine, and as an apprentice, you could get stuck into the problem-solving methods with the skilled lads. (Respondent 5, Mechanical)

Respondents outlined how interacting with skilled workers in cribs, workshops or on the shopfloor repairing faulty parts provided learning opportunities to understand how the part worked. Alternatively, if machine parts were irreparable, they were left idle. Skilled workers would ask apprentices to strip the kit down and use the opportunity to understand how the parts functioned. Often, opportunities to work on parts came from those left lying around in workshops, which skilled workers brought to the attention of apprentices. A respondent reported:

When I was an apprentice, the workshop was where you learned how to, say, strip down a motor, a pump, or a weld timer – there would always be something lying around for repair – something that you needed for your logbook. (Respondent 6, Mechanical)

Apprentice interactions with skilled workers emphasised the importance of socialisation in craft norms. Interactions with skilled workers were said to cultivate perceptions of trade status and demarcations. Respondents recounted how skilled workers taught apprentices to conceive of themselves as members of a distinctive trade. Moreover, skilled workers advised those training to be instrument technicians to wear a collar and tie to differentiate themselves from the other skilled workers and apprentices. A respondent explained:

. . . failure to wear a collar and tie meant that we were reprimanded by the skilled workers in the shop. It didn’t sit well with me as I was a ‘biker’ with long hair, but you had to toe the line. (Respondent 7, Mechanical)

Furthermore, respondents emphasised receiving advice from skilled workers on how apprentices should act regarding other trades and production line workers. Among electrical trades, apprentices reported how they were encouraged by senior skilled workers to see themselves as the elite of all the trades in the plants, with tool makers a close second in this worldview. Plumbers and painters were seen at the bottom of the trade hierarchy. Apprentices who failed to conform to the expectations of the skilled workers, such as not displaying sufficient interest in the trade, appearing lazy or not wearing the correct attire, would be ‘punished . . . left to fester in the crib until a shop steward had a word with them’ (Respondent 7, Mechanical). The existence of separate offline ‘cribs’ or tea rooms for the different maintenance groupings reinforced demarcated social relations across trades; thus, electrician apprentices socialised with electrician apprentices, mechanical fitters with mechanical fitter apprentices.

Upon completion of the apprenticeship, ² respondents reported how an older, skilled worker was explicitly assigned to guide and mentor the newly completed apprentice through the first year. That mentor, who played a crucial role in smoothing the transition from apprenticeship to fully fledged tradesperson, was seen by respondents as a guide and someone who extended the learning process, ensuring the continuity of knowledge transfer:

I can remember [coming out of my time, and] being put on . . . 12-hour nights . . . I was put with [one of the skilled lads]; it was a saviour, for me, because he taught me loads, I learned so much [during that first year] from him. (Respondent 8, Electrical)

Respondents reported minimal interaction with union representatives during their apprenticeship. Local stewards might only become a feature of the apprenticeship experience if there were disciplinary issues such as missing classes at college or continual lateness. A respondent commented: ‘. . . the union had bigger fish to fry, so they weren’t really bothered about apprentices’ (Respondent 9, Electrical).

The unions might also get involved if they had to challenge supervisors about not providing suitable training or if some skilled workers were reluctant to bring apprentices onto the shopfloor with them. Both issues could emerge as contention. The plant negotiating committee implemented a clause into a plant agreement to combat that disinclination: it specified that skilled workers were obliged to train others (1984 Plant Wage Agreement: 10). However, respondents reported that such reluctance from skilled workers was an occasional rather than pervasive problem – driven more by personality type than the structure of the labour process.

Apprenticeship experiences under lean

Between 1990 and 2018, plant documentation showed an approximately 80% decline in the skilled headcount. While there was a reduction in apprentice numbers to a maximum of six from 1991 onwards – and for seven years, recruitment stopped completely ³ – there were also fewer skilled workers available for mentoring. In the case of those skilled workers that remained, several respondents reported increased workload via preventative maintenance checks introduced from 1995 onwards. That increased workload stemmed from efforts to minimise ‘waste’, namely skilled worker idle time and line breakdowns. Management reduced idle time by increasing preventative maintenance checks on machinery to pre-empt and minimise line breakdowns.

One of the consequences of lean was the impact on apprentices’ ability to access skilled workers on the shopfloor. Although the three-year on-site structure remained the same, the reduced skilled headcount and increased workloads made it difficult for apprentices to access experienced workers. All apprentices reported that difficulty, except for die maintenance fitters. As one respondent commented:

When I first started my apprenticeship in 1992, there was the best part of 300 maintenance workers in the Body Shop alone. By January 2006, there was less than half this number – so, over time, it became more difficult for the apprentices to find skilled workers with the right skills. (Respondent 1, Multi-skilled)

Another respondent recounted difficulties developing programming skills on Paint Shop robots: ‘. . . there was only one person with appropriate software skills in the Paint Shop, so it was difficult for us apprentices to get time with him because he was always busy’ (Respondent 2, Multi-skilled). In addition, one respondent related:

I needed to get some advice on a project for my logbook from one of the guys in the Body Shop controls team, only to find he had left and the rest of the team had either also left or were now working in other parts of the plant, and on different shifts. (Respondent 3, Multi-skilled)

The cohort also reported difficulties in securing machinery to train on. With lean, there was an increased emphasis on eliminating machine downtime and idle machinery related to increased performance maintenance checks. Respondents reported that such emphasis resulted in management’s view of those machines not in use as redundant: they were then scrapped or sold off. There was no opportunity for skilled workers to stockpile kits in workshops or cribs and provide materials for teaching apprentices as in the pre-lean cohort.

Where machine breakdown occurred, management emphasised prompt repair, which limited the space for skilled workers to immerse apprentices in the workings of the technology. Aside from the pressure of performance maintenance checks, skilled workers’ capacity to instruct apprentices on technology became increasingly constrained by the adoption of plug-and-play technologies in the 2000s. Such technologies enhanced efficiency and reduced time lags in users’ tasks. ‘Plug-and-play’ parts related to electronic or electrical parts that in the past would require a ‘re-boot’ of machinery but were now pre-loaded with software, enabling the machinery to recognise the part when loaded into the machine. That helped minimise downtime, particularly on a moving assembly line, and meet the plant’s daily build schedule. With the increasing use of plug-and-play technologies, parts were either unrepairable – because they were inaccessible – or despatched to external vendors rather than restored in-house. Again, that minimised apprentices’ opportunities for experimenting on idle machinery.

Even with the machinery that was fixable on site, the lean period introduced timings within which repairs should be completed. An escalation process underpinned the timings: maintenance had to speak to their line manager if an online repair was not completed within five minutes. That could mean the repair was taken from the skilled worker and given to specialist engineers, often called in from outside firms. In the shops prone to machinery and tooling breakdown, pressure to get back online constrained opportunities to explain repairs to apprentices. As one respondent recalled:

The worst for me was Assembly, because . . . that is where you have the maximum pressure of any downtime of the line as it’s having an immediate effect on the cars coming off the end of it. So, there is severe pressure just to get that line running. . . . people were deeply unhappy. I was on the Trim line for a time and remember going to a breakdown and tagging behind them because they wouldn’t even tell me that there was a breakdown, and they’d just walk off. [So] I’d tag along, and then they’d fix it, and then I’d ask what they’d done to fix it . . . they’d go, ‘we’ve fixed it’ and walk off back to the crib. (Respondent 4, Multi-skilled)

While pressure to fix technology occurred in the pre-lean period, what was different under lean was:

. . . because of the reduction in heads you don’t have enough time to explain what has gone wrong either during, you can’t show them what’s gone wrong, or following a breakdown, you just don’t have the time to explain; you’re onto your next job. If there were two or three of you there, maybe you would be able to explain, and [there is] no chance of them fixing a breakdown, particularly if it directly affects production. (Respondent 5, Multi-skilled)

In Stamping, where apprentices did not report problems accessing skilled workers, there was a lack of spare parts in reserve due to just-in-time initiatives not to hold idle stock. Again, concerning repairs, the lack of spare kit meant pressure to have dies in use. That constrained skilled workers in terms of time to train apprentices on experimenting with parts. As one respondent recounted: ‘Before I left in 2018, you’d be lucky to get a die for four shifts [to work on a repair] . . . because it was going back in the press again’ (Respondent 6, Die Maintenance).

Lean also reduced the departmental workshops, where apprentices could work under the tutelage of skilled workers, and the skilled workers who operated them. Space diminished over time due to a lean project to maximise floor space for more value-added work to increase throughput and capacity utilisation. Consequently, respondents reported a lack of space to experiment on machinery. In the pre-lean period, respondents reported at least one workshop and sometimes two for each department, while in the lean period, there were only two for the entire plant. A respondent observed: ‘. . . there’s nowhere to go offline to do work with them [the apprentices], so they’re normally sat outside the office, and any job that comes in, they get sent on’ (Respondent 7, Multi-skilled). In most production areas, access to workshops for apprentices was limited, even given the relatively smaller numbers taken on. Thus, most interactions with skilled workers were confined to the ‘white heat’ of the production line, with all the pressures of timing that entailed.

Difficulties accessing skilled workers and a lack of technology and space for mentoring resulted in apprentices struggling to complete their logbooks. That issue led to intermittent interventions from the local union and the external examiners from the accrediting FE college to force management to make resources available. FE college assessors would intervene when apprentices were not receiving enough electrical training or where BARS reviews were not completed. BARS reviews were incomplete in some years because of a lack of interest among management and skilled mentors. Moreover:

When the apprentice manager left in 2018, it really went downhill as no one was doing the [BARS] reviews with us [in the departments] and the BARS reviews were fudged to make them seem like we had reviews. Which wasn’t great. (Respondent 8, Multi-skilled)

Interjections from the union or FE college were often sufficient to address the matter at hand, but the responses on the part of the plant were reportedly reactive and short-lived, with similar issues reappearing over time. By way of example, because the local union challenged the lack of space to work on machinery, a training room was provided for skilled workers and apprentices with robots and a small moving assembly line, which could be set up with faults to work on. However, management could not keep it operational because of the lack of skilled workers to support the activity. Respondents reported skilled workers struggling to get released from work tasks to support training. There was also a lack of spare parts for the training room due to the just-in-time initiative. Consequently, the room lay unused for several years before falling into disrepair as skilled workers, at the behest of production managers, cannibalised what equipment there was to keep production running on the shopfloor.

As the HR apprentice coordinator reported:

Some of the apprentices (in 2015) were getting that far behind, that the plant allowed them to go onto an afternoon shift, so not only doing a morning shift, but there would be a shift pattern where they would drop on to afternoons, a non-production shift, and then they would do all preventative maintenance tasks that would enable them to get back on track for their logbook. It was done at the behest of the union. It was the union saying to management: ‘Listen, they’re not getting the required work that would enable them to get through their NVQ’. So, the union was pushing for that.

That was supported by the plant Training Manager who noted: ‘. . . all the way through, I couldn’t have done my job with the apprentices without the trade union’s help . . . the union deputy convenor, I think, probably spent more time on apprenticeships than I did’. In response to apprenticeship experiences during this period, the union secured an ‘apprentice committee’ consisting of management and union representatives to discuss the progress of apprentices’ logbooks. However, documentation showed it met infrequently, with maintenance managers regularly not turning up due to other commitments.

Respondents also reported limited support for transitioning post-apprentice workers, particularly regarding mentoring in the year after completion. ⁴ The pressures on existing skilled workers reduced their time to assist those transitioning. As apprenticed workers from the lean cohort commented: ‘Now apprentices are expected to come out of their time and hit the floor running as there is no headcount to support them in the first year. That is the reality; it is a lot harder now’ (Respondent 9, Multi-skilled). Supporting this view were assessments from those completing apprenticeships in the pre-lean period:

When I came out of my time, I was in a team of nine, me, four electricians plus four fitters . . . you could spend a bit of time training on equipment that you hadn’t come across in your apprenticeship . . . there are . . . issues with the apprentices coming out of their time. There’s not many of us left to get help from because we are so spread out in the department. (Respondent 10 [Pre-Lean Cohort], Electrical)

Another skilled worker from the pre-lean period maintained:

I think the apprentices are more mature and confident than we ever were, because they are a bit older and already have some qualifications . . . they need to be because [they] haven’t got the numbers of skilled lads around them that we had, so it’s a lot harder now. (Respondent 11 [Pre-Lean Cohort], Mechanical)

The latter quote was instructive, as respondents reported that apprentices needed to become self-reliant and independent in completing logbooks. When a skilled mentor was unavailable to work on one logbook element, apprentices reportedly would try to compensate by working on those elements they could complete independently or undertake elements where a skilled mentor was likely to be available.

Finally, on-site placements for newly apprenticed trades emerged as a significant problem from the 2000s onward (National Negotiating Committee minutes, 29.10.2003; National Negotiating Committee minutes, 04.02.2010). Of the 52 apprentices recorded in the plant records as completing between 2011 and 2017, only nine secured permanent employment at the plant. Apprentices had to exit the plant or take semi-skilled jobs on the production line:

I was forced into the autonomous production role, which is something I didn’t join up for as I came here to be a maintenance technician. The job has been running for a couple of months now, and all I have been doing is filling up Kanban boxes, giving relief blows and basically spending not a lot of time doing maintenance on my group. (Respondent 12, Multi-skilled)

Discussion

The case reveals challenges confronting apprentices in the lean period. Reported obstacles like lack of access to mentors, lack of technological artefacts to experiment with, lack of spaces to work and lack of placements suggest overlap with the lean model as described by Vidal (2022). Just-in-time dynamics result in the unavailability of technology to use for hands-on experimentation. Parts are instead on-site as needed to minimise uncirculated capital and inventory storage costs. Continuous flow organisation, demanding synchronised production and quick identification of defects, reduces idle time. Under a taut production model, skilled workers become less available to apprentices, weakening their role as a transmission belt for on-the-job learning. Mentor–novice exchanges are debilitated, resulting in apprentice isolation and logbook struggles. The drive to maintain continuous throughput is associated with a pivot to plug-and-play technologies to enable reduced downtime (Smith, 2000: 29). Better-performing technology leads to less idle machinery for apprentices’ experimentation. Consistent with waste elimination, any space within a production setting not adding value is a cost without benefit. That perspective reduces workshop spaces, previously sites for mentor–novice interactions. Finally, the adoption of teamwork sees many maintenance tasks shift to direct workers on the line. If read alongside lean’s downgrading of maintenance by devolving some maintenance to semi-skilled workers, a generalised pivot to do more with less and more reliable technology, a perfect storm emerges for future apprentices. With fewer maintenance roles comes fewer placement opportunities.

How should the putative links between lean and apprentices’ experience be read from a CoP perspective? CoP literature suggests such challenges must debilitate situated learning and weaken peripheral participation (Hoadley, 2012; Lave and Wenger, 1991). The lack of continuity in CoP, via labour shedding and lack of placement, alongside the challenges confronting mentor–novice interaction, seem to hollow out the CoP. Do the barriers confronting apprentices hinder their acquisition of skill? By the measure of national accredited standards for the field, apprentices still acquire technical skill by virtue of completion, even if the route to knowledge acquisition is more challenging on the lean shopfloor. Research undertaken in this project shows the jobs continue to require skill; some new technologies necessitate upskilling, although some trades do appear subject to partial deskilling as diagnostic, fault-finding technologies improve. However, a learning process that places obstacles in the way of knowledge acquisition is inferior, where the obstacles are not the innate ‘desirable difficulties’ of the task (Bjork and Bjork, 2011). Rather, the evidence suggests obstacles arise from making learning resources challenging to access and that may be due to the lean model. The case suggests lean may weaken CoP and present undesirable challenges for apprentices accessing learning resources on the shopfloor. Given that access to mentors, technological artefacts to experiment with, spaces to work and placements seemed to be more readily available pre-lean, the case strongly suggests that lean impoverishes those aspects of the learning experience.

That raises the implications that inefficiency and slack in a production model may be more helpful from the point of view of on-the-job learning. The comparative pre-lean benchmark reflects some of the relative slack, technological inefficiency and managerial disorganisation prevalent in the industry during that period (Edwards and Terry, 1988). Such slack appears conducive to deeper novice participation in CoP. Taking time to explore multiple paths, experiment, fail or struggle with complex problems encourages deep cognitive processing, embedding knowledge more meaningfully; that is, it is slower to solve a mathematical question manually than via the efficiency of a calculator, but it strengthens problem-solving abilities. Although not all organisations are explicitly pursuing lean, many seek to counter perceived inefficiencies or waste. However, waste and inefficiency may help create spaces for important workplace learning. That relates to a second point. Lean in some quarters of Marxism is understood as a progressive development of the productive forces (Adler, 2007; Vidal, 2022). In this context, lean can be read as enabling, at a technical level, the social organisation of production at a higher degree of planned precision in a way that minimises waste. Nevertheless, the social relations of capitalism act, as Vidal argues, as ‘fetters’ (Marx, 1971a: 21; Vidal, 2022: 14). Here we see that firms may enact lean in a guise that retards learning and the training of a skilled cadre of labour, thereby undermining at least one of the conditions that might be necessary for the pursuit of genuine continuous improvement. That arises because of applying the forces of production under conditions of capitalist competition and the pressure for valorisation. Where those conditions are absent, say under hypothetical socialist arrangements, lean could be adopted to avoid the fettering of skill acquisition by making planned and balanced accommodations for collective learning in a way less achievable under the dull compulsion of capitalist competition (Marx, 1971a, 1971b: 689). That might also suggest a contradiction between forces of production – the capitalist force of lean as operationalised retards the acquisition of technical knowledge, itself a force of production.

Before referring to implications for apprenticeship policy, we stress limits to generalising from a single case and emphasise that the causal links are suggestive. Different plants can have variations of lean adoption, which may have different implications for apprentice on-the-job learning. Different jurisdictions, moderating institutions with state-backing and entrenched tripartite arrangements may inform different shopfloor experiences. That potential suggests the value of further research. The English case suggests that current policy emphasis on increasing apprenticeship numbers and completion rates (Department for Education, 2024) may miss out on processes that occur within the labour process. Of course, employers have not escaped the gaze of inspectors, although how robust these are is circumspect (Lewis and Ryan, 2009), and high completion numbers may ignore obstacles to meaningful learning within leaner labour processes. The informal triadic prompting and pushing in the case between FE providers, unions and employer does hint at the value of a more structured type of apprenticeship tripartism in helping to identify and mitigate such obstacles.