The decline of incentive pay in British manufacturing


  • James Arrowsmith,

    Corresponding author
    1. School of Management, Massey University, New Zealand
      James Arrowsmith, Massey University, Private Bag 102904, North Shore City 0745, Auckland, New Zealand; email:
    Search for more papers by this author
  • Paul Marginson

    1. Industrial Relations Research Unit, University of Warwick, Warwick, UK
    Search for more papers by this author

  • James Arrowsmith is Professor at the School of Management, Massey University, New Zealand and Paul Marginson is Professor at the Industrial Relations Research Unit, University of Warwick, Warwick, UK.

James Arrowsmith, Massey University, Private Bag 102904, North Shore City 0745, Auckland, New Zealand; email:


Motivation theories and the strategic pay literature envisage that the management of employees can be well-served by financial incentives and other forms of pay flexibility. Traditionally, UK manufacturing has made extensive use of variable payments systems (VPS), notably piece-work and bonuses, but these have declined at the same time as managerial control over pay-setting has increased. Evidence from six case studies suggests that a focus on pay is only part of the picture. Increased competition and change makes the design of VPS more complex, and new forms of work organisation become the focus of performance. In this context, firms have (i) abandoned individual incentive pay and (ii) aggregated VPS in support of broader objectives.

‘Cash Payment has become the sole nexus of man to man.; and there are so many things which cash will not pay’

Thomas Carlyle1


It is commonplace to observe that pay is at the heart of the employment relationship. This is not just tautological, in that employment is almost universally defined in terms of paid labour. It is also because of the importance of pay as a means of recruiting and retaining workers, its significance for employers' costs, and the potential for conflict that the ‘cash nexus’ introduces into the collective endeavour that is work. This conflict is most evident over levels of pay, but it is also manifest in the sense that employers often have to navigate potentially contradictory goals concerning pay systems design. On the one hand, employers want to keep control of pay costs and are sensitive to constraints on ‘ability to pay’; on the other, they are aware of external ‘going rates’ and may use pay as a means of motivating employees.

A traditionally important means of reconciling what Flanders (1975: 72–73) termed the ‘managerial’ as well as ‘labour market’ functions of pay has been to make some part of pay contingent on output or other performance measures. Workers' interest in higher pay is thereby aligned with management concerns relating to output, productivity or profits. A range of variable-payments systems (VPS) are available to this end. These are usually differentiated into three groups (van het Kaar and Grünell, 2001): essentially output-related bonuses, such as piecework or payments-by-results (PBR); pay based on managerial appraisal, such as individual ‘merit pay’, often used for ‘white-collar’ jobs where individual output cannot readily be measured; and pay linked to overall company performance, such as profit-share schemes. Smaller firms also favour informal and ad hoc bonus arrangements (Gilman et al., 2002).

The focus here is on the operation of VPS in manufacturing companies, a sector that has traditionally made extensive use of various forms of variable pay and in particular incentives-based systems (Bryson and Forth, 2008). However, nationally representative statistics suggest that manufacturing employers are increasingly rejecting these forms of VPS. In the next section, we briefly discuss the theory and practice of VPS, especially incentive pay, which has strong foundations in economic theory and work psychology. The following section outlines the research methods used in the study, which is based on case-study research. The results section is organised into three sub-sections covering, respectively, the demise of piece-work; the use of company performance bonuses, and employee appraisal and merit pay. The discussion section then explains these developments in terms of a wider context of organisational change, before a brief conclusion.

1.1 The theory and practice of VPS

The principal theoretical explanations of VPS are familiar and include agency theory, expectancy theory and equity theory. Agency theory locates VPS in the risks associated with the incompleteness of the employment contract, which may be corrected by aligning employees' financial interests to managerial goals (Eisenhardt, 1989). The ‘principal-agent’ problem has been analysed most exhaustively in terms of executive remuneration, but the basic model has also been applied to other employees in the form of incentives theory. As Behrend (1957) observed in her analysis of the ‘effort bargain’, such accounts rest on assumptions that effort can be varied, can be translated into outputs and that money is the most important work motivator. Hence, it resembles expectancy theory, which suggests a virtuous circle linking effort and performance to a reward that is valued by employees, though this need not be financial in the model (Vroom, 1964). Other theories pay closer attention to the process of employee motivation and especially considerations of fairness. According to equity theory, for example, employees judge the fairness of a pay scheme in terms of how pay relates to their own effort inputs and by comparing this to the inputs and outputs of other employees (Adams, 1965). The implication is that employees may perceive VPS as potentially more legitimate than standardised pay systems, provided they are procedurally robust and deliver what is perceived to be an acceptable return to differentiated levels of performance.

What these perspectives have in common is the suggestion that VPS can be an important influence on employee behaviour, and that schemes with a clearer ‘line of sight’ between effort, performance and reward are more effective in this regard. The implication is that, where practicable, management would favour individualised and localised schemes such as PBR or team bonuses over aggregated arrangements, such as factory bonuses or company profit-share schemes. The so-called ‘New Pay’ literature differs to some extent in its prescription. This advocates a pay-systems strategy that exposes employees to a greater element of risk in order to better respond to an increasingly dynamic business environment and requirements for greater organisational flexibility (Lawler, 1990; 2000; Schuster and Zingheim, 1992). As Heery (1996: 56) noted, ‘for new pay writers it is axiomatic that if pay is to function as a tool of strategy it must be variable’. However, the New Pay writers also recommend implementing multiple schemes at different levels, each under strict management control.

There is certainly a wide use of different forms of VPS in British workplaces. A useful data source is the 2004 Workplace Employment Relations Survey (WERS), which questioned managers about the use of the three distinct types of VPS:

  • • PBR, defined as ‘any method of payment where the pay is determined by the amount done or its value, rather than just the number of hours worked. It includes commission, and bonuses that are determined by individual, establishment or organisation productivity or performance’. Subsequent questions explored whether measures used reflected ‘performance/output’ at the individual, team, workplace, organisation or other levels.
  • • Merit pay, defined as pay ‘related to a subjective assessment of individual performance by a supervisor or manager’.
  • • ‘Profit-related payments or profit-related bonuses’. Data on use of share schemes was also gathered.

Overall, the results showed that VPS was extensive, with a half of all private sector workplaces (with five or more employees) having some form of scheme in place—just over a third used individual PBR, extending to 43 per cent of employees; 26 per cent used group-based PBR, covering 30 per cent of employees; 23 per cent had profit-related pay schemes, which applied to 29 per cent of employees; and 16 per cent used appraisal-based merit schemes, covering 26 per cent of employees (Bryson and Freeman, 2008: 24). Furthermore, results from the 1998–2004 panel dataset also suggested that VPS was increasing; the use of ‘performance-related payments’ (merit pay and individual or group-based PBR) in continuing workplaces with 10 or more employees rose from 20 to 32 per cent over this period, ‘suggesting that there has been a substantial increase in the use of performance-related pay schemes since 1998’ (Kersley et al., 2006: 191).

However, a somewhat different picture seems to emerge when we look at manufacturing specifically, where PBR schemes have long been a feature (Brown, 1973; Smith and Boyns, 2005). First is that the use of such systems significantly declined in the 1990s (Table 1); there was also some retreat from collective PBR after strong growth in the late 1980s. This picture of declining use of PBR is supported by data from other sources. According to New Earnings Survey (NES) data, 29.4 per cent of manufacturing workers were in receipt of ‘incentive pay or bonuses’ in 1994, but this fell to one in five (20.9 per cent) by 2001. For those actually receiving such payments, the proportion of earnings accounted for in this way also significantly declined, from 19.0 to 13.7 per cent, over the same period. Although the NES was discontinued, other series suggest a continuation of both trends in recent years. According to the Labour Force Survey (LFS), the proportion of manufacturing workers in receipt of bonus payments fell from 8 per cent in 2001 to 4 per cent in 2008, while the Annual Survey of Hours and Earnings shows a fall in the proportion of manufacturing earnings accounted for by incentive pay across all workers from 5.1 per cent in 2002 to 4.5 per cent in 2008. Yet as Table 1 also indicates, the use of profit-related schemes continued to grow, even after the withdrawal of tax advantages in 2000, with 42 per cent of manufacturing workplace utilising such schemes in 2004. On the face of it, and contrary to incentive-based theories, manufacturers have been substituting collective for individual forms of VPS.

Table 1. Use of PBR and profit-related pay in British manufacturing workplaces (%)
  1. Note: Merit pay data not available due to inconsistencies in research instruments.

  2. Source: WIRS/WERS series data in Bryson et al. (2008).

Individual PBR323621
Collective PBR202825
Profit-related pay163342

There are three main, and related, potential explanations for the relatively recent accelerated decline in incentive pay. First is the steep decline in manufacturing employment, which fell from 5.2 million workers in 1987 to 3.2 million in 2007, according to the LFS. Though all sectors were affected, the biggest decrease was in those employing large numbers of lower-skilled workers, such as textiles, where piecework was particularly widespread. Second is reduced trade union presence and workplace strength. Manufacturing trade union density fell by 5 percentage points, to 22.2 per cent, in the six years to 2006 (Grainger and Crowther, 2007). Unions have also become more defensive due to the threat of corporate restructuring and capital mobility, which arguably makes it easier for management to intensify work effort without having to link it to pay.

A third possibility relates to the changes in the organisation of work. There is some evidence for the growth in recent years of forms of ‘high-performance’ work systems (HPWS), based on teamwork, functional flexibility and employee involvement (Kersley et al., 2006: 97). This has been linked to technological change and in particular greater use of sophisticated information technology systems (Brynjolfsson and Hitt, 2000). These HPWS are less suited to individual VPS and output-based measures (Wood, 1999), but may favour collective schemes for two reasons. First, group-based pay can reinforce collaborative effort within work teams (Appelbaum, 2002: 125; Freeman et al., 2008). Second, broader-based schemes, including those operating at the establishment or enterprise level, such as profit share, may be utilised to impact on workplace culture. This occurs when schemes are used to support the regular communication of business information, and where they serve as ‘recognition and acknowledgement by the company of the worker's stake in the firm’ (Appelbaum et al., 2000: 44).

This is speculative, however, as very little research has been conducted into the decline of PBR in British manufacturing. In particular, little is known about trade union approaches or management considerations concerning what other forms of VPS, such as group-based bonus schemes, might be put in its place (Cox, 2005). The present research therefore explored the dynamics of VPS as it affected shopfloor workers in unionised manufacturing settings.


The research began in 2005 with a series of sector-level interviews in order to map the broad terrain and refine the research instruments for subsequent company case studies. These involved a senior national representative of the manufacturers' association (the EEF),2 three regional EEF officials (a Director of Employee Relations (ER), a field ER specialist, and a payment systems consultant), a regional secretary of Amicus, and a regional industrial organiser and a national secretary of TGWU. This was followed by six case studies conducted over the period 2005–06. These were selected as unionised firms of different size that had some experience of VPS. The specific focus was the machinery and equipment sector, which has a large ‘blue-collar’ workforce, and, according to LFS data, a level of union density, collective bargaining and VPS similar to that of manufacturing as a whole.3 In the two largest companies, interviewees included representatives of group-level management, manufacturing and plant general management, senior HR managers, and trade union senior lay officials. In the other cases, interviews were conducted with full-time union officials and shop stewards, the managing director or works manager, and HR managers where these were in place. The companies were anonymised at the request of participants and are briefly described in Table 2 as M1 to M6.

Table 2. Case study (CS) business and industrial relations context
CSBusinessEmploymentTrade unions and industrial relations
M1Industrial engines (NACE 29.11). Part of US MNC (acquired 1997); 30% of products sold within MNC. Sales growing but running annual losses of c. £78m.4,000 employees, 3,600 at the CS site. Hourly paid = 1,700 (incl. 350 agency workers); 95% are male. Outsourcing components manufacture reduced headcount; automation led to deskilling (200 ‘skilled’ workers remain, primarily in maintenance and tooling). Teamwork is extensive (c. 10 per team); double shift is main working pattern. Labour = 12% total costs.Site has 29 HR staff incl. 4 senior managers. Site-based bargaining; 25 shop stewards, shopfloor density 90%. Single-union/ partnership agreement with Amicus (AEEU) from 2000 (staff union de-recognised 1999 due to falling membership). Relations with Amicus ‘very solid … enormously positive’ (HR mgr.); ‘we're all on the same side … a responsible trade union brings an awful lot to the business’ (Manufacturing Coach). Last reported dispute 1990.
M2Aero components (29.11). Part of UK MNC. Site faces internal and external competition. Fluctuating markets but profitable (c. £450m business-level; half of group profits).Over 20,000 UK employees. CS is greenfield site opened 2003–05; 918 employees plus 300 agency workers (due to upturn in orders and as a ‘10% buffer’ policy). Most workers (1,011) blue-collar, 90% ‘skilled’, almost all male. Relatively high pay, minimal turnover. New working practices on relocation. Plant expanding, partly due to in-sourcing; 75% workforce on 4 shifts.Site has 7 HR staff. Mainly site-level collective bargaining, with some at business level. Site has 2 recognised unions (overall 98% density): Amicus (was AEEU for shopfloor, MSF staff) and GMB (c. 100 members in functions such as polishing). Single-status since 2004 and single-table bargaining (8 union reps). Partnership agreement since 2002. TUs closely involved in negotiation of ‘modern working practices’ to secure replacement facility.
M3Heavy equipment for paper industry (29.55), mostly to order; 80% for export. Part of German (private) MNC since mid-80s. Profit £1.5mParent employs 24,000 around the world and has 3 manufacturing sites in UK. CS site a stand-alone profit centre with 94 employees, 38 shopfloor (all but 3 skilled, almost all male). Workforce reduced by 40% in 1999.No specialist HR function on site. Amicus (formerly AEEU and MSF) recognised. 60 union members (90% of shopfloor workers and 25% staff. One shop steward (‘there used to be 2 but no-one else wants to do it’). IR focused on annual pay claim in which FTO involved.
M4Diesel engine pumps (29.12). MNC owned by UK venture capitalists (1999). New management 2002; restored profitability (company c. £6.8m, site c. £0.5m). 60% product exported to the United States. Sold to Swedish group 2008.Group employs 500 worldwide, including 320 in UK (240 shopfloor plus 30 agency workers). Mainly semi-skilled workers; almost all male, ethnically diverse. Outsourcing basic components to China and India (led to closure of the second unit at the UK site in 2005). Cellular manufacturing and teamwork. Functional flexibility, except for highly skilled machinists. Long-service employment and low turnover.New senior management team appointed 2003, including HR manager (who heads a team of 4). GMB the largest union (113 members, but rep position vacant); T&G (20, 1 steward) represents some workers (including logistics) and Amicus (54 members in office staff (1 rep) and skilled engineers, with 2 stewards). Single-table bargaining since 2005. Restructuring under new management team has strained relations (3 ballots in favour of industrial action 2002–04, last one realised). Current relationship ‘arm's length’ (GMB FTO).
M5Gas pressure regulators (29.13). Only UK site of a German MNC, acquired 1990. 75% UK market share and 30% products exported. Profitable since restructuring in 1995 (c. £2.5m)Redundancies in 1998, but growth since (131 employees 2003, 200 in 2006). 80% workforce is shopfloor, 70% in assembly. High volume, continuous process and mainly semi-skilled (65%) and unskilled (25%) work, Unskilled workers paid national minimum wage. Long-service employment and low turnover. Most workforce female (and 30% part-time). Labour = 15% operating costs.No specialist HR function. TGWU recognised; 40% employees are members (all but 2 shopfloor). 3 stewards. IR focused on pay deal. New stewards less adversarial; negotiated long-term pay deals and not made recourse to FTO in 4–5 years. One strike in 10 years (2001 one-day action over pay award). Firm ‘has always been a good company to work for’ (steward); ‘I think we have a good working relationship; we don't always see eye to eye but we don't expect to’ (works manager).
M6Healthcare products (33.10). US MNC owned by private equity group since 2000. UK losses late 90s, fluctuating performance since.4,000 employees worldwide, 350 in UK (650 in 2000), 170 shopfloor (mainly semi-skilled), 65% male. Outsourcing and narrower product range since 2000. Low labour turnover (<5%) due to competitive pay and residential location.HR team of 3. Tight control of pay budget from parent. TGWU recognised since 2001. FTO describes company as in his best 5 for openness and ‘give and take’.

Each of the case study sites and companies are well unionised and form part of larger multinational concerns. This was part of the international dimension to the research design, based on matched comparisons of subsidiaries across countries, but the presence of multinational companies (MNCs) is not unusual in manufacturing. The proportion of UK manufacturing employment accounted for by foreign-owned MNCs stood at 27 per cent in 2005, and UK-based multinationals are estimated to employ a further one-third of the manufacturing workforce (Marginson and Meardi, 2010). However, it is acknowledged that the research focus might introduce some contextual implications that are peculiar to subsidiary companies of MNCs—for example the relevance of transfer pricing and exchange rates to investment decisions and reported profitability.

The case study firms also share a number of other features. For example, voluntary labour turnover was low due to relatively high pay and/or limited opportunities for similar employment locally. Most had recently introduced forms of teamwork and (especially M1, M2 and M4) cellular production. Each company also had its share of trauma in the years preceding the research, including acquisition (M1 to another MNC; M4 and M6 by private-equity firms), site relocation (M2), new senior management (M4, M6) and outsourcing (M1, M4), which led to redundancies (M2, M3, M5, M6) or otherwise reduced headcount (M1, M4).

At the same time, there were important differences. The first is establishment size. The smallest is M3, with a shopfloor workforce of 38; conversely, M1 and M2 employ 1,700 and 1,200 blue-collar workers, respectively. The other cases employ between 160 and 270 shopfloor workers. Second, skill requirements (as reported by company grading systems) also varied from the predominantly high-skilled nature of work in M2 and M3; mainly semi-skilled work in M1, M4 and M6, and lower-skilled work in M5, which was the only company with significant female shopfloor employment. Third is profitability, which varied across the cases. M1 was experiencing operating losses, although business volumes were improving. M2 was moving back into profitability after a sharp downturn in business and subsequent relocation to a new site. M4 was moving in the same direction, following financial difficulties. Financial performance at the other three companies fluctuated, with margins tightest at M6, though each had faced financial difficulties around the turn of the decade. Finally, there were also differences in industrial relations, from the close and formal partnership of M1 and M2, to the more traditional but still good working relationships of M3, M5 and M6, and the adversarial experience of M4, where a new management team had sought rapid restructuring in order to immediately improve performance.


The context for each of the companies was therefore organisational change, which had major implications for collective pay bargaining and VPS. Four of the six firms had moved from annual pay negotiations to the conclusion of long-term agreements (LTAs), for three years' duration in M2 and M5 and two years in M1 and M3. This was because annual collective bargaining had become, for management, ‘a ritual’ (works manager, M5) or ‘a game’ (managing director, M3) that absorbed a lot of their time. A senior manager at M1 also said that the LTA removed some of the ‘edginess’ associated with negotiations and introduced greater stability, which facilitated the negotiation of broader changes to work organisation and working time. For example, the two-year deal reached in 2005 introduced a new shift pattern to enable continuous working and simplified overtime payments around a standard 50 per cent premium. Similarly, in M3, the collective agreement introduced a system of banked hours in 2004. This broader focus went furthest in M2, where the collective agreement set three successive ‘milestones’ for the introduction of ‘modern working practices’, including increased working time flexibility, ‘a big removal in demarcation lines’ (AEEU-Amicus convenor), and shopfloor workers moved to salaried status. However, managers also reported a risk associated with LTAs in committing to future pay awards, which were normally slightly above the change to the retail price index (RPI), given it was difficult to accurately forecast company performance.

The two companies with annual negotiations experienced poor or fluctuating results and had implemented below-inflation increases in recent years. In M6, the budget for pay awards was tightly controlled by the parent company. The last two settlements were below inflation, and both were accompanied by redundancies. New working-time arrangements were introduced to improve productivity and cost performance, notably a banked hours arrangement in 2002. The company was trying to extend this in the 2006 negotiating round in order to make Saturday working compulsory. In M4, organisational change was not linked to collective bargaining; relations were strained after the appointment of a new management team in 2002. Their turnaround strategy included outsourcing basic components manufacture to new sister plants in China and India, together with investment in technology, which helped eliminate many low-skilled jobs and increase functional flexibility.

Hence, change was extensive in each of the companies in response to intense and increasingly international product-market competition. In M2, these external competitive pressures were exacerbated by heightened internal ones. Arguably, competitive pressures might be expected to both contain pay rate growth and encourage VPS such that management became concerned to tie additional earnings to improved productivity or other performance measures; in this way, VPS may link a part of pay costs to ability to pay or even generate a ‘self-financing’ element. At the same time, as union priorities in this context were maintaining basic real pay and defending jobs, so managerial discretion over VPS schemes (outside basic pay) might increase. Tellingly, the two industrial disputes reported in the years prior to the research (a one-day stoppage in M5 in 2001 and a series of strikes in M4 in 2005) were both in protest at base pay offers below inflation; bonuses were less of a union concern.

In the event, we found three sets of results that suggest that VPS is indeed linked to organisational change, but less for its direct incentive or cost effects than to support broader processes of work reorganisation and culture change. The first is that output-based individual VPS hardly features anymore. Two companies abolished piecework arrangements while the others have never used them (Table 3). Only in M5 did management claim the VPS scheme was designed to directly incentivise employees. Second, four companies have an aggregate, company performance scheme in place covering the entire workforce. Payments are usually based on company and/or divisional financial performance (M1–M4). A fifth (M5) has a factory-level bonus scheme which applies to the entire production workforce. The sums involved vary, from 4–5 per cent of earnings in M3 and M4; 7–10 per cent in the two largest companies (M1 and M2, which is the only case to have two schemes in operation); and 25 per cent in M5, where most workers were engaged in lower-skill assembly work. Only in M6, where a company performance bonus scheme was withdrawn as a cost-cutting measure, was VPS of marginal significance for earnings. Third, four of the six companies (M2, M3, M4 and M6) have performance appraisal arrangements for blue-collar workers, but these schemes are not linked to pay, except in M6. We now explore each of these results, utilising the qualitative data from the case study interviews.

Table 3. Overview of blue-collar payments systems
CSBasic pay structure and awardsForms of VPS
M15 skill-based grades until 2006 when highest-skilled divided into 4 points for recruitment and retention purposes; union agreed as a means to improve pay 2-year agreements (2003: year 1 = 3%, year 2 = RPI + 0.5%; 2005: same 2-year formula)Business incentive plan since takeover in 1998. Based equally on company and divisional performance (earnings per share, profits, quality measures linked to Six Sigma). Paid annually. Restructured 2006 to allow payments for each component target. Shopfloor workers receive 10% bonus if all targets reached; usually hit 80–90% = 7% payment. Criteria and thresholds solely management decisions. Also has an ‘Employee Investment Plan’: voluntary share save scheme with company matching 50% free shares. Shopfloor take-up around 50%.
M27 grades (including team coaches) linked to skill. ‘All-inclusive’ (salaried) pay since Jan. 2004, with shift workers on 30% higher salaries in each grade. 3-year agreement 2002: year 1 = 2.5%, year 2 = RPI + 0.5%, year 3 t.b.c. (was 3%) plus 2% each year (and further 5% in year 3) for reaching flexibility ‘milestones’All-Employee Bonus Scheme. Introduced 1999. Maximum 2 weeks pay (minimum 1 week) split between company and divisional performance (i.e. max. = c. 4% of earnings). Latter mainly uses a product cost index (an internal profit measure) plus some quality and delivery measures. Company half has paid every year; divisional all but one. Facility (site) bonus. Planned 2002 (but delayed until 2005 to achieve inter-plant consistency), partly as overtime pay removed. Flat-rate payment of a maximum of £500 per quarter or c. 6–7% earnings. Simplified in 2006 from 16 to 3 indicators: 70% decided by profit and 15% each quality (returns) and delivery. Agency workers excluded; around 1,000 employees eligible. No payment until Q1 2006 (£227).
M35 skill-based grades. All workers are salaried 2-year agreement 2005: year 1 = 4% (or £10 per week if higher), year 2 = 3% or RPI + 1% (was 3.5%)‘Company bonus’ introduced 2005 in response to union demands for a more formal/systematic scheme. Form of profit-share with a 3.5% pot above a certain threshold. Paid each Christmas; 60% a flat-rate sum and 40% linked to salary, with slight adjustment if <100% attendance. Has paid every year except 2003. Typical payments vary between £400–£550 (union rep) or average 4% of earnings (according to management).
M4• 5 grades, hourly-paid
• 2005 = 2.8% (below-inflation)
• Piecework abolished 2003; profit-related company bonus introduced 2004. Pays quarterly if profit targets reached, only one quarter missed. Capped at 5% paybill. Flat-rate sum of up to £200, or 5% of earnings for production workers.
M5• 7 grade sets linked to skill
• 3-year deal 2004: years 1 & 2 = 1.5%, year 3 = 3.5%
• Piecework abolished and replaced with a monthly ‘factory bonus’ (for shopfloor workers) in 1998. Based on average efficiency performance, as against work study benchmarks; starts at 68% performance (6p per percentage point above this) with no cap. Varies between £1.50–£2 per hour on base pay (20% skilled and 30% unskilled earnings).
M6• 5 grades, salaried pay linked to introduction of hours bank.
• 2005 pay award = 2%
• ‘Tier-system’ of service-related/ competency pay replaced with appraisal-related pay in 2002. Only operates if pay pot sanctioned by parent company is over RPI (occurred in first 2 years but not last 2). Link to pay loosened when appraisal system standardised across the group in 2006.
• ‘Global bonus scheme’ linked to group financial performance revoked 2002 on orders of new US owners. Triggered only 4 times in its last 10 years; paid 2% salary bill or £340 for shopfloor workers.

3.1 The demise of piecework

Piecework was not a feature of pay systems in the case-study companies, and in two firms, it had been recently withdrawn. This is indicative of the transactional costs and technical problems associated with piecework in a context of changing work requirements. However, in one company (M5), the logic was to refine the incentive by collectivising it; in the other (M4), the objective was to institute a link to profits to underscore the importance of broader business measures of performance. These two cases are now discussed in turn.

The end of piecework in M5 occurred in 1998 in a context of poor business performance and was part of a wider package of change, including a programme of rationalisation, the restructuring of back-office work and introduction of new working methods on the shopfloor. The works manager explained that individual pay was unsuited to the high-volume assembly line production that was increasingly used due to kit assembly (‘it's not possible to identify who's working hardest; we can only identify what comes off at the end’). There were also problems of administration and equity because some product lines involved more difficult jobs than others, and workers were required to move between jobs:

(collective) is the only fair way of doing it because some of the things we do are difficult to make. So if you find yourself doing a difficult job which is more complicated and the incentive is individual, you wouldn't have an opportunity to get the incentive. You drew the shortest straw. Other people have got a simple job with more opportunity to earn more. We don't think that is fair. We think it should be shared by everybody.

Additionally, the incentive effect of the piecework scheme was seen as weak, since it accounted for less than 5 per cent of earnings overall, though the company still had to employ two payroll staff to administer the scheme. The new bonus simplified the data-collection process, which was calculated almost automatically from the daily input of job data by supervisors. However, in contrast to the other cases, a key management priority for the new scheme was to reinvigorate the incentive effect to increase productivity and, potentially, to offset part of any fall in orders by a decrease in labour costs. Management argued that the change was essential to maintain the viability of the company, while the union was concerned that the new scheme would not generate sufficient bonus earnings to cover the reduction in base pay that was also linked to the introduction of the new scheme. The manager candidly explained that there was a contrived element to the new definition of job and skill requirements underpinning the changes to basic and variable pay:

we briefed (the consultant) what we needed to pay for these jobs and he went out there and determined what skill level they were and what the reasonable pay rates for them were. we were keen to have some sort of neutral person to come in and look at it for us and not just say that this is our sort of idea. This is in order to get real credibility because it was quite difficult to negotiate a pay reduction.

After protracted negotiations, and following trials of the new scheme to demonstrate that it could generate sufficient earnings for employees, the proposals were narrowly passed in a ballot of union members. The new factory bonus was linked to average efficiency as compared with work-study targets, and paid monthly. Payments start once a threshold of 68 per cent performance is surpassed, with workers receiving six pence per hour for each additional percentage point reached for that month's average. Average bonus earnings immediately fell from £1.20 to £1.00 per hour until ‘the mix of work and demand’ (volumes and complexity of work) changed in workers' favour.

The piecework scheme in M4 applied to 210 production workers and was dismantled in 2004. This was accompanied by a 15 per cent increase in base pay to consolidate the piecework premium into basic rates. The new bonus was paid on a quarterly basis to all employees if certain profit targets are reached; following the first payment, and at the instigation of the union, workers balloted in favour of switching from a percentage of salary to a flat-rate payment, which management conceded. Payments (up to £200 per quarter) represented around 5 per cent of earnings for production workers, which was a ‘cap’ agreed with the unions. As profits increased, management proposed raising this limit but the unions were reluctant to accede in case it constrained real basic pay growth. Indeed, implementation of the scheme was protracted, as it took some time to convince the workforce they would not be worse off. This reflected an unfavourable context marked by closure of the final-salary pension scheme, pay settlements below RPI, reduced shift working, which impacted on earnings, and the threat of redundancies.

The HR manager outlined two main reasons for this initiative. First, the old scheme was not working well; the negotiation of a number of ‘safety nets’ or allowances over the years for downtime, attendance, meetings, scrap and re-work and the like meant that much of the bonus payment, which was as high as 20 per cent of earnings for some workers (15 per cent on average), had become guaranteed. This benefited some groups of workers more than others;

there were some jobs that due to the machinery or products it was impossible to achieve the target … some groups were getting hefty bonuses and others with the best workers in the world couldn't achieve it.

In addition, operation of the scheme was contentious and costly to run—two office workers were employed to administer it and deal with the succession of queries and challenges that accompanied the delivery of pay slips every Thursday.

Second, managerial goals had changed in response to increasing product diversity and pressure for customisation of orders. Quality and reducing re-work costs were increasingly important (‘we would rather make fewer but have them right’), and this was linked to the introduction of self-inspection and functional flexibility (teamwork, multi-skilling and flexible deployment). In particular, a recent collective agreement concerning labour mobility was undermined by having different pay rates for different jobs. The company also argued that a move to higher base pay with a collective bonus introduced greater certainty over earnings for employees, and was fairer to workers employed on dispatch or support tasks that were excluded from the previous piecework scheme. These arguments were accepted by the GMB full-time official who led the negotiations from the union side.

Line management also believed the new bonus was vital to help communication and education around broader business performance, and emphasise quality and customer satisfaction rather than simply ‘volumes that went out the door’. For management, replacing piecework with the profit-share bonus helped symbolise a shift to a team-based, ‘right first time’ philosophy as much as the physical investment that had been made in new high-technology computer numerical controlled (CNC) machinery on the shopfloor. Despite the difficult context, the GMB official also stated a belief that the new arrangements were indicative of a much more ‘professional’ and less ‘old-fashioned’ managerial approach on the part of the new team. Piecework was seen as indicative of a low-trust culture where employers believed workers needed to be financially incentivised to do the job, and workers acted accordingly. In sum, then, changes in management and work organisation, where teamwork and task flexibility became a priority, defocused the ‘line of sight’ between effort, output and reward, and this was reflected in changed VPS. In M5, which retained a more traditional production/management regime, incentive pay was not abandoned but rather standardised around collective measures.

3.2 The use of company performance schemes

Collective bonuses linked to profit and other business metrics were used in M1, M2 and M3, as well as M4 as described above; M6 had a scheme until it was terminated in 2002. This had paid a fixed percentage of salary to all employees if a set of annual financial targets were reached, but it delivered only four times in the preceding 10 years, and sums were relatively small, averaging around 2 per cent of earnings or £340 for shopfloor employees. The HR Director explained that the decision to end the scheme was taken unilaterally at headquarters under instruction from the parent company, a private-equity firm that acquired M6 in 2000. It was reported that the decision reflected a renewed cost concern and a belief on the part of the new owners that such programmes offered little incentive to employees.

In contrast, the bonus scheme in M1 was a corporate programme introduced by the parent company following its takeover in 1998. The ‘business incentive plan’ was seen as a means to encourage employees to identify with the group and to focus on quality. It was also accompanied by an employee share-save scheme that management said was designed to focus employees on general business performance. Management and union representatives confirmed that the incentive plan was not a matter for consultation, let alone negotiation with the union (‘it's a discretionary scheme—we don't have to give it’, HR director; ‘it's a “given” from the company’, AEEU-Amicus rep). The criteria involved a mix of financial and quality indicators divided equally between the division (which are specified locally and not by the parent company) and the group as a whole. This group element was linked to overall business performance metrics for the group; M1 constitutes just one of 27 divisions organised into four major streams. The scheme was revised in 2006 so that profit-related targets are no longer prerequisites for payment on other measures, and payments have been at the 7–10 per cent level even when the company has experienced difficulties. The HR director, who also sits on the global compensation council, observed that:

the 50–50 split is probably good for employees. It's unlikely that (we) would both have a crap year. If we have a bad year, we may not deliver on our bits—or on only some of them—but there's a good chance the corporation will deliver. So you will still get some chunk of the money

Essentially, the use of group as well as divisional measures ensures workers receive a bonus payment even if the division is going through difficulties. This was the case in M1, where the introduction of automation, teamwork, and headcount reduction was a feature of restructuring efforts to improve productivity. The senior HR manager put it another way; the bonus was less about employee motivation (since ‘workers probably did not see the relationship between their day-to-day performance and the bonus payment they received’), but perhaps a means to avoid their de-motivation. This view was implicitly echoed by union representatives, who said they had difficulty understanding the scheme, but that employees were (for the time being) buoyed by the payments through a period of great uncertainty:

It's absolutely marvellous. We don't know anything about it, but for the last two or three years we've been getting very good pay(outs) and the company is running at a loss. (We) understand that it's split into four different parts like health and safety, Six Sigma, quality … so we understand that, but actually how they work it out we don't know … and we ain't asking while it keeps paying [AEEU-Amicus shop steward].

Divisional management conceded that the mechanics of the bonus are fairly complex, and the mix and weighting of targets varies from year to year as priorities change:

In terms of divisional metrics, we change them every year. We debate them each year at the [divisional] board. We try to profile issues we want to focus on [HR director].

This complexity reflects the negotiation of different local and central priorities in a rapidly changing environment, but evidently incurred some cost in terms of efficacy as a means of business communication with employees.

The scheme in M3 was a legacy of profit-share arrangements introduced in 1995 to benefit from the then-favourable tax regime. The managing director also said it was a response to union pressure, though it was not actually negotiated as such:

the unions had complained about the former ‘grace and favour’ loose bonus scheme; so the new scheme structured it as well as taking advantage of the tax regime of the time.

The withdrawal of tax advantages in 2000 did not prompt a revision of the scheme, though its continuation was reviewed in consultation with the union (which unsuccessfully proposed a higher flat-rate weighting). The managing director explained that this was because it was valued as a vehicle for communication and awareness raising around wider managerial concerns:

I wanted to get away from the culture associated with long-service employees that are not exposed to the outside world, taking stability of employment for granted, and wanting more (pay). It gives the shopfloor information about the business … (and) is linked to getting employees to think about the business as a whole, such as where we are exposed to variation in steel prices, not just their own requirements … it ties the attention of the workforce into the profitability of the company. In the past the workforce would look at the number of (units) going out the door as the benchmark of how well we are doing, but we could actually be losing money on certain strategic contracts … it really drives some glue between employees, managers and the company into how the company is doing.

Workers, of course, were more concerned with what the scheme delivered. A sum amounting to 3.5 per cent of site profits over a certain threshold (then £500,000) is distributed at Christmas with 60 per cent, a flat-rate sum, and 40 per cent linked to salary. Three per cent of the amount is also ring-fenced for employees with 100 per cent attendance. The average payment for a typical employee amounts to 4 per cent of earnings (£800) according to management, though union estimates were lower (£500). The scheme paid out every year except for 2003, when one union representative recalled the effect on workers:

they were all moaning and a bit gutted. The rep said to people ‘but it's not part of your pay; you never know, next year it might pay a thousand’. Although it has never paid a thousand … 

In M2, a new ‘facility’ (i.e. site) VPS scheme formed part of a wide package of change, including the introduction of flexible working. The context was a planned relocation from a 65-year old site, possibly abroad, and a more immediate sharp fall in orders after the September 11, 2001 attacks, which prompted 440 redundancies in the plant. The company eventually decided against a proposed move to eastern Europe so that it could continue to benefit from the skills of the existing workforce, but made this conditional on an agreement relating to the transformation of working practices. The idea was to ensure optimal use of the £85 million capital investment in the new plant or, as the HR manager put it:

it's about lean methodologies, vertical integration, flexibility and multi-skilling, team philosophies … so that we can fully utilise capital, people, machines (to) get the most efficient cost-effective way to deliver the product to high spec when the customer needs it. When you bundle all that together it sounds quite easy if you say it very quickly, but considering where we came from, it was a huge transition.

A ‘partnership agreement’ was therefore concluded with the unions in 2002, and the new factory opened in 2003, although the full transfer of operations took over two years to complete. A three-year LTA was also reached in 2002, which provided for ‘modern working practices’ and a move to salaried status. Shift payments were consolidated, and a 5 per cent pay increase was implemented to recognise increased working-time flexibility. The idea was to improve delivery performance and eliminate the costs of overtime by ‘incentivising to achieve within normal working time’ (HR manager). The changes were accompanied by three additional ‘milestone’ payments of 2 per cent on base salary each year.

The agreement also made provision for a local ‘self-generating facility bonus’, and a site-based plan was devised by a joint working party in 2003. This was seen as ‘a key part of the (working practices) package because once you put people on base salary they have no other way of increasing their earnings’ (HR director, employment policy). The site HR manager described it as ‘very much about win-win’, with employees getting a financial return if targets are met. However, implementation was delayed until 2005, as the parent division decided to centralise bonus arrangements in order to ensure consistency across plants. The final design was therefore not negotiated locally, and revisions implemented in 2005 were also made unilaterally. The criteria were simplified from 16 to three ‘key performance indicators’ (KPIs) based on quality, product cost and delivery, and potential payments were increased. The twin concerns were to make the bonus more transparent and achievable, not least as it had not yet paid out. The general manager said ‘it was a huge boost to our credibility and the scheme's credibility’ when the first payment was made in 2006.

In the meantime, the parent group scheme, which was originally introduced in 1999, did generate payments. This originally paid a maximum of one week's pay linked to company results, and an additional week's pay was soon added based on divisional performance. Both company and divisional criteria were linked to profit measures, with the latter also taking some account of quality and delivery, so it bears some resemblance to the facility bonus scheme in terms of the metrics utilised. The group scheme paid out in full each year up to the time of the research, except in its initial year when the divisional element did not pay. As with the plant bonus, management objectives were articulated around ‘softer’, cultural issues, as well as a focus on quality; according to the group HR director, employment policy, there is a ‘reward’ rather than ‘incentive’ rationale:

The rationale is that employees should share in the success of the company … It is certainly not something that people think, oh I must put in a bit of extra effort today so that I can get my whole employee bonus … The way most people look at it is I won't bank on it.

3.3 Employee appraisal and merit pay

Individual performance appraisal for blue-collar workers emerged in the case-study companies with the development of teamwork and multi-skilling. Appraisals usually focus both on ‘objective’ job requirements, such as skills development and also behavioural elements related to employee commitment. Appraisal evaluations or scores are linked to pay only in the case of M6, though it has been proposed, or informally mooted, at three other companies; barriers to implementation normally reflect less any technical considerations to do with measurement criteria than limited managerial enthusiasm and active union opposition, as indicated below.

Performance appraisal was introduced for production workers in M4 in 2004. Management proposed basing the distribution of the following pay award on the appraisal scores in order to better ‘reward the high performers and incentivise the poor performers’, according to the HR manager, but ‘it was rejected out of hand’ by the trade unions. Senior management also recognised that supervisors were not keen on linking the appraisal assessments they made of subordinates to pay, and the idea was shelved. Similarly, in M1, management had ‘floated’ the idea of extending the matrix-based individual merit pay system from the white-collar to the manual workforce, according to the unions, ahead of recent pay talks. This was rejected by union representatives because of perceived difficulties, including potential abuse, which was identified with ‘subjective’ assessment. The system was, however, implemented for the blue-collar workforce in a joint-venture business at an adjacent location to M1 from the outset of its opening in 1996. Although the union had consultation rights, recognition did not extend to bargaining, and managers attributed these differences in industrial relations and pay systems to the preferred style of the Japanese joint-venture partner. Union representatives at M1 seemed satisfied that the merit-pay arrangement was ‘ring fenced’ to the separate company.

In M3, the appraisal (or ‘employee dialogue’) scheme extended to all employees, but was only linked to pay for managers. The managing director expressed some support for union concerns by arguing that a link to pay might undermine the communication and development aspects of the performance review and bring its own problems:

with appraisal-type (pay) systems it is difficult for some people to accept; there are reservations over whether and how you get 6 out of 10 or 4 out of 5 and whether it is subjective—and if this affects bonus payments it can be negative. It can be self-defeating. The formal sit-down dialogue is good but not the link to pay as this can actually undermine the dialogue. There is still nervousness but I tell staff we all have them (targets and appraisals), we are all appraised every day—this is just formalisation and provides you with input.

In contrast, the three-year collective agreement reached in M2 in 2003 provided for the immediate introduction of an individual performance management system and envisaged that this would be linked to pay from 2005. Management believed that this would complete the overhaul of the traditional ‘basic plus overtime’ pay model to a salaried payments system in which collective and individual contributions would be recognised through the vehicles of the site bonus and merit pay. Unions saw individual payments as re-establishing an opportunity for additional earnings that had been circumscribed by the abolition of overtime. However, this was blocked by head office. Corporate management felt that the individual scheme would be an unnecessarily complication to remuneration given the priority granted to the bonus; they were also aware that unions (and managers) in other plants might be alarmed at what the HR manager acknowledged was a ‘radical’ development for an engineering workforce.

Accordingly, as noted above, provision for individual performance pay was found only in M6, though even here it was not normally triggered. Management in the company emphasised that the main objectives of the appraisal and merit pay scheme were to help clarify employee roles and identify development needs through regular two-way communication. It believed that a link to pay might also help reinforce the underlying performance focus. In practice, the appraisal ratings were not always used for pay purposes, and when they were, the differentiation effect was small, except for the small number of low-rated workers. This was because the budget for the pay award was strictly determined by the parent company, and the union agreed to merit pay only on the basis of, and in order to underscore, the principle that annual settlements at or below RPI would be implemented on a standardised basis. Management agreed and across-the-board settlements were applied in 2004 and again in 2005, when a settlement of 2 per cent was imposed following the loss of an £8 million order. In 2003, when appraisal-based payments were most recently triggered, a pay budget of 3 per cent was distributed as zero for anyone judged ‘unsatisfactory’ (around 5 per cent of workers) and 1 per cent for a ‘needs improvement’ rating (also around 5 per cent). Most workers (60 per cent) either received 3 per cent for ‘meets expectations’, 3.2 per cent for ‘exceeds job requirements’ (20 per cent) or 3.4 per cent for ‘outstanding’ (10 per cent). At the time of the research, the union said it had concerns over inconsistency, favouritism and subjectivity in the operation of the scheme, mitigated to some degree by the fact that it had had no recent bearing on pay outcomes. The union (and management) also envisaged further difficulties within the appraisal scheme itself after the parent company decided to roll out a global appraisal template in 2006, as this was much more behaviourally related than the preceding ‘home-grown’ scheme.


The dominant economic and psychological theories of variable pay emphasise the important role of incentives, which have greater purchase at the individual level. Statistics suggest, however, that manufacturing companies are increasingly rejecting incentive-based schemes, primarily individual arrangements, but also at team level, in favour of ‘aggregate’ bonus schemes linked to establishment or enterprise results. Contrary also to the prescription of ‘new pay’ advocates, the case-study companies were also generally not implementing VPS at multiple levels, which is also borne out by WERS (Bryson and Freeman, 2008: 24). The results of the case studies shed some light on these developments.

First, the results offer support for our third possible explanation, raised at the outset, for the decline of incentive pay and widespread use of aggregate, profit-related schemes. This is that the growth of high-technology production systems utilising teamwork and flexible labour render individualised and incentive-based schemes more problematic; yet they promote broader-based VPS as part of a renewed managerial concern with the commitment of employees and their perceived understanding of the business context. Essentially, changes in the competitive environment and technologies used by the firms had prompted a change in production and management regimes, emphasising (i) teamwork and functional flexibility in terms of work organisation and (ii) greater communication and employee involvement in terms of management style; albeit allied to apparently technical forms of monitoring and control through statistical data management techniques and deployment of ‘KPIs’. The increased sophistication of computer-aided manufacturing systems and machines facilitated changes to work organisation, such as cellular production, teamwork and multi-skilling, and also permits more accurate and easier measurement of efficiency and quality. It therefore renders individual bonuses inappropriate but is suited to aggregate multi-factor schemes.

These innovations clearly undermine the principle and practice of traditional PBR schemes, such as piecework. As Thomas Carlyle recognised at the birth of industrial capitalism, and has been well documented in the British workplace since (Brown, 1973; Edwards, 1987), exercising control via a simple cash nexus such as piecework is a recipe for opportunism and antagonism. Such arrangements were characterised by a very experienced senior EEF official as ‘complex to administer, expensive and caused disputes with the trade unions’. Traditional bonus schemes introduced conflict over work-study timings and rates; encouraged ‘wage drift’, with incremental changes in job requirements or working methods enhancing earning capacity over time; and risked disincentive effects where workers judged the ‘wage-effort bargain’ to be either too lax or too demanding. Both management and trade union respondents alike therefore linked the decline in such schemes to a ‘professionalisation’ of management, linked to quality and employee involvement approaches such as TQM, Kaizen and Six Sigma. As a regional TGWU official explained, under PBR:

if production stopped, the pressure would come from the shopfloor to the managers—‘where is the work, where is the work, where is the work?’; under the daywork system, the pressure is reversed … It really puts more and more responsibility on management to make sure that there are the logistics, supply line and the number of operators required.

Allied to this is a concern with direct communication in order to encourage greater customer awareness and responsibility for quality on the part of shopfloor workers. In addition to regular team briefings by supervisors, senior managers in most of the case study companies introduced regular meetings with workers to brief them on business and financial performance. In M2, for example, direct communication takes the form of ‘dialogue sessions’ and team briefings. The former is a company-wide initiative designed so that employees meet with senior site managers on a quarterly basis to discuss market conditions and the performance of the business. There is a dialogue session at the site practically each morning involving 20 employees and two senior managers. Team briefings are used to discuss team objectives and performance, notably around the KPIs. According to the HR manager, ‘we are getting very positive feedback (and) the level of business understanding in our employees has probably doubled within the past year’, and the general manager said that ‘it's demanded by employees now’. This was confirmed by one of the Amicus convenors, who remarked:

you go down on the shop floor and you are actually starting to hear ‘management speak’; they would never have bothered with KPIs, time, and deliveries before. Now they know them inside out, they look at that board and they understand what they mean—which is good.

Broad-based bonus schemes are seen as helping underpin this approach, as the metrics are linked to criteria as quality indicators and overall business performance. This has some theoretical basis in reinforcement theory, with some researchers suggesting that financial incentives might undermine the positive effects on performance associated with job redesign and improved employee feedback and recognition (Luthans and Stajkovic, 1999). Changes in VPS therefore come as part of a wider process of organisational change that embraces efforts to shift organisational culture, as well as actual working practices.

A second important finding is that this is not always part of some rational strategic plan, as anticipated in the ‘new pay’ literature (Lawler, 1995; Trevor, 2009). Changes in pay systems have accommodated to, and act in support of, wider changes in the organisation of work rather than been driven as performance strategies in their own right. The connection between pay systems and these other elements is not necessarily explicit, except, as in the case of piecework, where they were problematic for these other goals—or in the case of M2, where an event such as relocation is predicated on wholesale change. Neither is VPS necessarily seen as the most important part of the process; with one exception, managers articulated the objectives of VPS less in terms of direct incentive effects than in supporting the communication of business results and wider changes to company culture. The exceptional case was M5, where an aggregate form of incentive pay was introduced (in place of piecework) for a mainly low-paid, female workforce engaged in assembly-line work. Here, gender is also a consideration; international studies consistently demonstrate that PBR is generally most likely found in establishments with a high proportion of women workers (Heywood and Wei, 1997).

The various changes to pay and work systems had a number of ‘triggers’. Uncertain and turbulent trading conditions removed some of the inertia that often governs pay systems. In particular, there were ‘critical incidents’, such as acquisition or (threatened) disposal, new management teams, and acute financial difficulties. Increased competition and leaner supply chains also meant that customers became more demanding, placing a premium on quality (‘right first time’) objectives and delivery to schedule not just cost or output per se. Working-time flexibility is thus increasingly important, and this is usually linked to pay; indeed, one of the biggest changes in payments systems for blue-collar workers has been the consolidation of shift and overtime earnings into all-inclusive salaried pay. With opportunities for overtime reduced by initiatives such as ‘hours banks’ and new shift systems, bonuses become the primary opportunity for workers to achieve additional earnings.


This study suggests that the decline of PBR and growth of profit-share bonuses is linked to a ‘modernisation’ of manufacturing workplace practices and management systems, that is the progressive implementation of a new production paradigm ‘increasingly defined in terms of flexibility, quality and productivity’ (Bélanger et al., 2002: 156–157). Terms such as ‘modernisation’ and ‘professionalisation’ were frequently used by trade union, as well as managerial respondents to contrast new working practices with traditional work systems featuring clear demarcations, close supervision, and pay systems relying on overtime and incentive bonuses to supplement hourly-based pay. This is not to say that the new work practices were simply endorsed or treated as unproblematic; trade union respondents in particular were well aware of issues raised in the critical literature (e.g. Knights and McCabe, 2000; Vallas, 2003) to do with potential work intensification, disputes within teams, and tensions concerning worker autonomy and managerial control. Unfortunately these themes cannot be pursued here given the primary focus of the research on variable pay rather than the perceived outcomes of ‘high performance’-type systems per se.

Essentially, then, management favoured simplified, aggregate forms of bonus over individual PBR as part of a wider change agenda. This echoes other studies, which suggest that the significance of VPS depend ‘less on their effect on motivation than on their direct association with a radical change in performance norms and renegotiation of job design and work rules’ (Grimshaw and Rubery, 2010: 361). Trade unions also made a clear, if less direct, contribution to this development. Though each of the bonus schemes were fundamentally a matter of management prerogative, it was the unions' vigorous defence of real basic pay, together with agreements such as salaried pay and flexibility over working time and labour deployment, that helped shape the parameters for the framework of VPS pursued. Unions generally have fewer problems with ‘on top’ profit-related pay schemes than either PBR or merit pay (Pendleton, 1997). Within a wider context of changes to basic pay systems, work reorganisation and working time change, such bonus arrangements, using broader criteria that do not impact on basic earnings, retain an appeal. Similar results were found in other country contexts, with piecework in decline and management favouring aggregate, profit-related schemes (Nergaard et al., 2009).

Of course, no claims can be made for the generalisability of the case-study results. First, these sets of pressures and responses are more likely to be associated with larger than smaller firms, and with skilled workforces rather than those involved in basic manufacturing and assembly (Barth et al., 2008; Brown, 1990; Nickell, 1995). In contrast, lower-skill work may well still be incentivised—as the changes at M5 indicate. Here, incentivisation was retained, but on a collective basis, reflecting a move to simpler kit-based assembly work. Second, the study examined only unionised workplaces; the absence, with one exception, of appraisal-based individual payments schemes might, for example, reflect this focus, as unions had opposed management proposals to introduce such schemes within two of the case study companies. However, the use of merit pay in manufacturing is relatively rare according to WERS, with only 13 per cent of workplaces using such arrangements (Arrowsmith, 2008). In most of our case studies, management viewed establishing a link between performance appraisal and pay as problematic, especially given the team-working involved in contemporary production systems. Hence, with an emphasis on flexible working, rather than flexible pay as such, the results suggest that many would not embrace such individual pay schemes even if they had clear opportunities to do so.


The constructive comments provided by Referee 1 are appreciated and helped improve the draft. We also acknowledge the contribution of Molly Gray to the fieldwork process.


  1. 1 Thomas Carlyle, 1840: Chartism. London: James Fraser. pp. 61, 66 (digitised by Google).

  2. 2 The EEF was originally an acronym for the Engineering Employers' Federation, but now that it represents the broader manufacturing constituency it is self-styled as ‘EEF—The Manufacturers’ Association' while retaining the shorthand use.

  3. 3 This sector was also chosen to facilitate comparative analysis of companion projects in three other countries; see Nergaard et al. (2009) for comparative results.