Thompson Work Organisations A Critical Introduction (3rd ed)

162 • W O R K O R G A N I S AT I O N S

If we return to the logic of the earlier claim about knowledge work, that it belongs to and moves with the employee rather than the organisation, this kind of interpretation can be contested on grounds of extent and character. Rather than seeing numerical flexibility as a management strategy, contracting out and homeworking can be viewed as a choice of individual employees, cashing in on their knowledge, and the dependency of organisations upon it. Hamel and Prahalad (1996: 238) refer to ‘a world of independent labour contractors’. While this may be something of an exaggeration, plenty of other commentators would regard it as a trend. Handy (1995) and Leadbetter (1999) link the emergence of a knowledge economy to the growth of portfolio workers, who will work either simultaneously or sequentially for a number of employers. In this context, short-term contracts no longer have a negative connotation, as they reflect the rapidly-changing business environment, with its emphasis on projects and teams.

Handy (1995: 6–7) does accept that there will be casualties in this restructuring of the boundaries between work and home, shortand long-term employment. In addition, there is wide recognition that employee adjustment to new realities is likely to be slow and painful. This is not just a question of changing employment status, but a transformation of the employment relationship itself. A quote from one of Heckscher’s interviewees sums it up with considerable clarity:

My basic mindset was, there’s an implicit contract, I expect that the company will provide me a career, development opportunities, and reasonable pay and benefits: and they, in turn, should expect from me that I’m willing to work very hard for them.When either one of us is unhappy with that situation, the contract is broken. And up until the past few years, as a corporation I had faith that that would occur. (Heckscher, 1997: 18)

In other words, the social, or what some commentators call the ‘psychological’, contract is no longer based on large, hierarchical companies providing stable employment, internal labour markets and long career ladders. Knowledge workers will increasingly be empowered to manage work and career themselves (Birchall and Lyons, 1995). While these discussions are pitched in the language of universality, such trends are predominantly about professional and managerial employers. After all, not that many routine manual and clerical workers had such conditions to be changed in the first place. On the surface this new social contract and the end of loyalty as a bond between organisation ‘man’ and the corporation seems somewhat lop-sided. However, there is a mutual gain as long as notions of opportunity and advancement are re-imagined. For Heckscher (1995) the new arrangements will have the benefits of being based on open negotiation and the promise of equality between organisation and individual, rather than paternalism and effort–security bargain. The individual will not have to subsume his or her personality in the organisation, and the breakdown of paternalism will allow more women and minorities to enter the ranks of management. Finally, there are new obligations: rather than provide employment, firms provide opportunities and transferable skills to help people become more employable, thus furthering a career across organisations and within occupational communities rather than up a single corporate ladder (Barley, 1996). We will return to some of these issues in Chapter 13 on corporate cultures.

If these, then, are regarded as trends in the labour market, what about work itself?

N E W E CO N O M Y, N E W O R G A N I S AT I O N S ? • 1 6 3

High performance and work systems: restructuring the division of labour

The labour process is a vital arena if the promises of the anti-bureaucratic revolution are to be fulfilled. Employees’ attitudes have to be transformed from grudging compliance to high commitment, while practices change from narrow, demarcated tasks to fully flexible, interchangeable labour. In the original ‘design chain’ of flexible specialisation, the last steps in the sequence were a high-trust strategy based on skill upgrading and worker autonomy. Similar themes reappeared in talk of the primacy of horizontal co-ordination in knowledge work, as well as the growth of functional flexibility. In more recent times they have been packaged in a broader guise of ‘high road’ strategies for workplace transformation and competition based more on quality than cost, with the ‘highs’ being commitment, involvement, skill, and performance, in addition to trust (Applebaum and Batt, 1994). We will return to claims of a high performance work system (HPWS) later, but before that we need to look at how such ideas and practices have travelled along this ‘road’.

Description or advocacy of participation, autonomy and collaboration is hardly new. In particular, on the surface the language and practices are continuations of the 1970s quality of working life (QWL) or work humanisation initiatives. QWL had long been a generic or umbrella term subsuming anything from job enrichment to participation schemes. Yet we know that despite the extensive publicity for highprofile cases such as Volvo Kalmar and some reports of gains in productivity and quality (Daniel and McIntosh, 1972), by the end of that decade most of the work reform initiatives had diminished or disappeared (Ramsay, 1991). It is therefore tempting to dismiss the new language and practices as a rebranding exercise. Yet the supporters of such approaches would argue that the speedy reappearance of new forms of employee involvement through quality circles, team briefings and other initiatives in the early 1980s, showed the persistence of attempts by employers to create new work paradigms that could move beyond the limits of Taylorism and Fordism.

This was not a direction chosen on the basis of values, but conditioned by a combination of opportunity and threat. New forms of involvement derived in part from technology such as flexible manufacturing systems and production processes, notably JIT (Sayer, 1986; Tailby and Turnbull, 1987). While it would be overdeterministic to ascribe ‘needs’, there is little doubt that both operate more effectively with higher levels of co-operation, knowledge and flexibility than previous configurations of technology and work organisation. Under JIT workers are expected to do on-the-spot problem solving. Indeed they have little choice given that reduced buffer stocks mean that subsequent activities would break down without such action. In turn this requires multiskilling and the incorporation of inspection, maintenance and support activities into production operatives’ tasks, or vice versa. These processes also require increased knowledge, both of details of the work process such as quality control, and a wider range of jobs.

That such systems are largely Japanese in origin indicates where the threat was coming from. The spread of quality circles and mid-1980s initiatives such as GM’s QWL and Ford’s Employee Involvement programmes in the US indicates the extent to which emulation of a perceived Japanese model was high on the agenda of Western manufacturers, though some, such as Rover in the UK, tried to avoid a ‘Japanese’ tag

164 • W O R K O R G A N I S AT I O N S

by relabelling initiatives as zone briefings and circles (D. Smith, 1987: 23). The comparatively short-lived nature of many such initiatives by no means ended the influence. Since that period the spread of Japanese transplants into the US and Europe has been extensive, and their apparent success has lead to further pressures that combine demonstration effect and emulation. The transferable ingredients of Japanese productive expertise have been repackaged as the system of lean production pioneered by Toyota, particularly in The Machine That Changed the World by Womack, Roos and Jones (1990). We discussed the issue of transferability in Chapter 6, but as a design template, lean production conforms to the anti-bureaucracy menu by stressing that in this brave new world, firms will employ teams of multiskilled workers whose jobs will be more challenging and productive, and carry more responsibilities.

The range of new involvement and communication practices had limited impact and sustainability, not simply because they were often imitative, but because ‘these schemes tended to be “bolted-on” rather than integral to the work process’ (Marchington and Wilkinson, 1998: 16). For example, many companies used quality circles as a catalyst for a period of organisational restructuring and dispensed with them once the goals had been achieved (Dunford and McGraw, 1987). This absence of a systemic character is addressed in some of the next phase of ‘new management practices’.

TQM programmes promised a broader, more holistic approach than localised quality circle techniques (Hill, 1991). In the pursuit of the goal of continuous improvement in the production of goods and services, TQM differs by starting from the top and cascading through the organisation. Flatter structures and reduced hierarchy develop as workers take increased responsibility. Hierarchy is also reduced by a network of interdependencies arising from the necessity for employees to treat each other as internal ‘customers’. Changes to job design are not the add-ons of old, but a genuine re-integration of conception and execution that is integral to the work process. Though TQM began by focusing on the ‘hard’ elements of capturing information about processes, it gradually shifted towards a ‘softer’ orientation to employee commitment and appropriate cultures (Wilkinson, 1992).

The consultant-driven (Hammer and Champy, 1993) emergence of business process reengineering (BPR) continued a holistic approach, promising that a focus on processes rather than functions would flatten hierarchies and produce complex jobs for smart people. Though BPR spoke the standard language of HRM and postbureaucracy, its route to competitive advantage lay in an integration of work and technical system redesign (Greenbaum, 1998: 136). Given the history of organisational change, it is hardly surprising that BPR initiatives squeezed out ‘slack’ with little consideration of the human dimension or consequences. Subsequent criticism and retreat meant that the management fad had reached burn-out in about five years (Buchanan, 2000: 26). Buchanan notes that, in contrast, though teamwork has been continually ‘rediscovered’ over three decades, the current wave is proving more durable.

It is not that teamwork is short on superficial prescription. The managerial literature is full of inflated empowerment rhetoric and exhortations to create self-managing teams so that ‘no boss is required’ (Dumaine, 1990: 40). There is, in fact, a wide consensus that current teamworking is distinguished from its 1970s predecessor by its much more instrumental, pragmatic orientation, with little concern for job enrichment and industrial democracy (Jenkins, 1994; Applebaum et al., 2000; Proctor and Mueller, 2000). At the same time it is seen as the core of work reform, and the ‘heart of the lean

N E W E CO N O M Y, N E W O R G A N I S AT I O N S ? • 1 6 5

factory’ (Womack, Roos and Jones, 1990: 99). Mathews argues that such elements of new production systems are not passing fads and differ sharply from QWL, which only grappled with the symptoms of workplace problems. Contemporary initiatives fundamentally reorganise the process of work: ‘In place of command and control structures designed to enforce rigidity and compliance, the new production systems call for management that offers facilitation, guidance and co-ordination between selfmanaging groups of employees who are capable of looking after the details of production themselves’ (1993: 7).

There is certainly considerable evidence for the spread of initiatives, as ‘teams have progressively replaced the individual as the unit of work organisation’ (Adler, 1997: 62). Adler notes that the proportion of leading US firms claiming to utilise selfmanaged work teams has risen from 6 per cent in 1979 to 28 per cent in 1987 and 68 per cent in 1995. In Britain, the WERS survey reported 65 per cent of all workplaces with over 25 people had teams (Cully et al., 1998). Interestingly there is also some agreement across managerial and radical commentators that management in an increasing number of companies is having success in devolving responsibilities to the group, getting employees to think and act like managers, and thus generating a genuine sense of shared values (Wickens, 1992; Barker, 1993; Sewell, 1998). In this way the workgroup can be used to redirect collective goals towards the company in a way that fulfils the old human relations dream; though for the critics, as we saw in Chapter 8, this is more like a prison of peer pressure and self-surveillance.

The evidence on value internalisation is critically evaluated in the following two chapters, but it is the case that both teamworking and TQM require attention to what Wood (1986: 432) calls ‘attitudinal restructuring’. In other words, it requires a much more explicit focus on shaping employee attitudes in spheres such as cooperativeness and self-discipline. As Sayer notes with reference to JIT, ‘profitability can depend quite heavily on the performance of workers who are technically unskilled or semi-skilled but behaviourally highly skilled’ (1986: 67). This helps explain the paradox that many companies are engaging in detailed and intensive selection and screening processes for relatively routine jobs, often recruiting young ‘green’ labour. When the NUMMI plant was established, all candidates for employment undertook three days of interviews, job simulations and discussion on the firm’s philosophy and objectives (Wood, 1986: 434).

Conclusion

For all the spread and significance of teamwork, not everyone would consider it the bedrock of organisational redesign. The idea of HPWS, developed primarily in the US (Osterman, 1994; Applebaum and Batt, 1994; Lawler et al., 1995), is based on the interlocking character of the management of work, organisation and employment. Work organisation, as already outlined, depends upon ‘the decentralisation of the gathering and processing of information to non-mana- gerial employees’ (Appelbaum and Berg, 1999: 1), with participation and problem-solving enabling front-line workers to contribute to operational decisions and performance. Work-based teams are supplemented by off-line problem and project groups are needed. Such practices are reinforced by skill formation strategies based on enhanced training in social and technical skills, plus more rigorous and selective recruitment, and incentives that make rewards contingent upon team or company performance, whether that be merit pay, profit sharing or group incentives.

166 • W O R K O R G A N I S AT I O N S

Finally, trust is enhanced by employees being encouraged to become stakeholders in the firm, primarily by the development of a mutual gains approach that, among other things, seeks to replace adversarial industrial relations by partnership and participation (Bluestone and Bluestone, 1992; Kochan and Osterman, 1994).

There are other versions of systemic interlocking, such as Huselid’s (1995) emphasis on the combination of employee skills, organisational structures and motivation, but the continuities are strong, and all versions claim that effectiveness resides in the ‘bundling’ of different practices that can lead to high commitment and positive employee behaviours (MacDuffie, 1995). Not only have such bundles or clusters been identified, a number of the studies have found a positive association with company performance (Huselid, 1995; Ichinowski et al., 1996; Applebaum et al., 2000). British research on similar lines has been more circumspect. Wood and de Menzes (1998) found a widespread adoption of high commitment practices, though of a diverse rather than ‘packaged’ nature; and also did not identify significant performance differences. However, the WIRS survey (Millward et al., 2000), and other evidence on European trends (Sisson, 1997), does confirm that there has been a spread and some clustering of a number of direct participation practices.

We have now worked our way right down the ‘design chain’ of organisational change, outlining claims and evidence about the specific dimensions, as well as the interrelations between them. Do these add up to a paradigm break: new organisations in a new economy? This is the theme of the next chapter.

12 Continuity and Change at Work

We have seen in the previous chapter that there are a number of overlapping organisational themes from the ‘models’ established through flexibility and other restructuring discourses. This chapter returns to each set of issues and uses contemporary research to subject the associated theory and evidence to a critical evaluation. In one sense this is not difficult. The quality of evidence in new economy, new organisation (NEO for short) arguments is often poor and frequently relies on relies on stereotypical polarities between old and new. Academics, consultants and a growing band of policy entrepreneurs have a vested interest in proclaiming and promoting the ‘new’, largely as a way of differentiating their own intellectual products. In fact a strong case can be made that leading purveyors of NEO, such as Drucker, have been making essentially the same argument about a rapidly changing economic environment compelling organisations to break up their bureaucracies since the 1950s (see Eccles and Nohria, 1992; Thompson and O’Connell Davidson, 1994). Similarly, knowledge economy perspectives sound remarkably similar to those put forward by writers such as Daniel Bell (1973) over a quarter of a century ago (see Smart, 1992). Yet each is predicated upon a supposed link between new market or technological conditions and new structures. Despite these weaknesses, it would be a mistake simply to deny the ‘new’ as frothy rhetoric. The really difficult task for organisational analysis is to grasp the patterns, dynamics and characteristics of continuity and change. This is our aim in this chapter.

Hierarchies, networks and knowledge revisited

There is no polarity that has been used more extensively than mass production and a variety of ‘flexible’ or knowledge-based successors. We focus on two of those successors – flexible specialisation and the knowledge economy – in this section, which deals primarily with macro market processes rather than organisational issues.

Though flexible specialisation is less fashionable these days, its account of changes in firms and markets remains typical of NEO arguments. The theory rapidly became a new work organisation paradigm and therefore a target for extensive critique (Smith, 1987). A number of studies, notably Williams et al. (1987), convincingly argued that the conceptual polarity between mass production and flexible specialisation is misleading. Traditional mass production is still widespread, particularly where semi-skilled women workers doing labour-intensive assembly and packing jobs remain a cheaper or more reliable option (Pollert, 1988a). In addition mass production does not always use dedicated equipment to make standardised products, but can handle diversification within flow lines. Wood (1989b) has shown that the world car – the archetypal example of mass production within the international division of labour – remains prominent, albeit in modified form. Economies

168 • W O R K O R G A N I S AT I O N S

of scale have fallen, but the break-even point is still high. Japanese and other companies have established variety within mass production with the use of JIT and not always with advanced technology (A. Sayer, 1986). Even the identification of flexibility with manufacturing itself neglects the service sector as the major growth area of such practices (Hyman, 1988). The flexible specialisation thesis and post-Fordists also made exaggerated claims for the use of programmable technology (Jones, 1997). Advanced technology such as FMS (flexible manufacturing systems) is very expensive, particularly for the small firms that should be in the forefront of customised production. Furthermore, not all new technology is either inherently flexible, or being used in a flexible way. The emphasis is more likely to be on control and co-ordination of the labour process, quality and routing rather than product flexibility (Wood, 1989b: 9).

Nor are mass markets necessarily saturated. There are a huge range of industries still based on mass and large batch production, and the pattern with goods such as colour TVs and video recorders is for production to be based on families of interrelated models. Fragmentation of demand, or what companies call ‘positioning strategies’, is more often an attempt to create a market rather than reflect new consumer tastes. Big firms can produce for mass and niche markets, as the ugly phrase ‘mass customisation’ indicates. Notwithstanding this hybrid capability, key industries, such as food, are actually destroying regional, craft production in the march towards global standardised products and marketing (Levitt, 1991; C. Smith, 1991). Indeed, it is ironic that the most explosive global growth is in fast food, which is ‘dominated by homogeneous products. The Big Mac, the Egg McMuffin, and Chicken McNuggets are identical from one time and place to another’ (Ritzer, 1993: 155). Ritzer may sometimes overdo the extent to which production and service has been McDonaldised, but the uniformity of high streets and shopping malls is testimony to the continued dominance of standardised commodities, mass markets and large firms. This is particularly the case in Britain, where large manufacturing firms have tended to ‘eschew the production of complex products’ in search of diversified production activities and standard products for mature markets (Ackroyd and Proctor, 1998: 166, 170).

The obsession with the Fordist stereotype also ignores inconvenient facts about the past. On the one hand the operations of Ford’s key plants, such as those as Highland Park, never corresponded to the description of inflexible Fordism (Williams et al., 1992b). On the other, the spread of Fordism has been exaggerated. Most plants in modern economies have never contained assembly lines, and Fordism never travelled far outside mass manufacturing (Williams et al., 1987; B. Jones, 1997). As Bryn Jones has noted, there is a persistent tension in production paradigms between Taylorist bureaucracy and Fordist integration, and pressures for product versatility (1998: 11). This tension has been reproduced in different ways within and across national economies. Japanese flexible manufacturing systems tackle capital and skill utilisation in ways distinct from European and American practices. But none resemble anything that resembles a flexible specialisation or post-Fordist model: ‘on the key measures of product-range, system-versatility and worker polyvalence, most FMSs operate within Fordist and Taylorist pre-conceptions’ (1998: 21).

This same observation, Jones notes, also applies to Italy, where flexible specialisation theory draws much of its inspiration. Even on their ‘home-ground’ in Emilia-Romagna, the majority of production is still Fordist in nature, while quality craft work is the preserve of the majority of middle-aged men: ‘Semi-skilled assembly

CO N T I N U I T Y A N D C H A N G E AT W O R K • 1 6 9

work, plastic-moulding, and wiring work is carried out by women, while heavy foundry and forging work is done by Southern Italian and North African workers’ (F. Murray, 1987: 88). Overall, as the work of the William’s team demonstrates, the statistical and case study evidence used by Sabel and Piore was frequently poor or non-existent.

There is a wider message about small firms here. As we noted in Chapter 11, there are sectors where innovative networks of such firms are playing a significant role in economic development. As usual, however, discontinuity perspectives have oversold the message. Across the globe, manufacturing and services remain dominated by large, indeed increasingly larger, firms. Counting the numbers of firms is misleading. It is true, for example, that in the US, 85 per cent of all individual enterprises in the computer industry employ less than 100 people. But the 5 per cent that employ 500 or more accounted for 91 per cent of employment and sales (Harrison, 1994: 5). Such trends are repeated sector by sector. In the UK, more than half of all employees work for organisations with more than 500 employees (Teasdale, 2000: 7). Even a persuasive advocate such as Perrow has to admit that the output of small firm networks on a global scale is ‘probably trivial’ (1992: 445). Moreover, small firms have been less technologically innovative than large ones, and their profitability often rests primarily on long hours and low wages (Amin, 1991; Harrison, 1994).

Knowledge in the economy

The basic pattern identified above, of stereotypical polarisation, limited evidence and neglect of diversity, tends to be reproduced in each new generation of macro arguments, and is certainly observable in relation to knowledge economy theories. As we saw in the previous chapter, the ‘weightless’ character of knowledge is said to make the economy fundamentally different from the old, hierarchical predecessor. Despres and Hiltrop argue that the crucial ingredient – knowledge – lies in the heads of producers: ‘it remains with the employee and in no real sense is it ever of the firm . . . it is impossible to separate knowledge from the knower’ (1995: 11). Relations inside and between firms therefore have to be of a horizontal, network kind.

However, this formulation is factually inaccurate and conceptually confused. What gives weight to an object or idea in a market economy is its status as a commodity. Its physicality or otherwise is wholly irrelevant. Knowledge can be calculated, rationalised and ultimately commodified (see Yakhlef and Salzer-Morling, 2000). A simple example is the emergent intellectual property regime in biotechnology. Global corporate players have been rushing to establish patents on genetic material, so that they can ‘own traditional knowledge and take advantage of the privatisation of life forms’ (J. Vidal, Guardian, 13 April 2000). Value can and is being assigned to these so-called intangible assets. Furthermore, though companies are dependent on scientific, knowledge workers to ‘mine’ this gene pool, the transformation into saleable commodities relies precisely on separating that knowledge from the knower. That, after all, is what ‘knowledge management’ systems are about. The language of knowledge economy theorists such as Nonaka and Takeuchi (1995) speaks of ‘capturing’ or converting knowledge as if that process is neutral and benign. As with our previous macroeconomic discussion, this is partly a temporal confusion. Autonomy and creativity are maximised during periods of pursuing knowledge and creating systems. Yet the commercialisation of the products of knowledge takes the

170 • W O R K O R G A N I S AT I O N S

form of standardised, highly-structured systems in areas such as software design or surveying. There are few detailed empirical accounts of how new knowledge management systems work. McKinlay (2000) shows the tension between creativity and control in his case study of Worldrug, a large pharmaceutical company. Mechanisms for the former, such as electronic cafes and intranets, sit alongside much more bureaucratic measures, such as data warehouses and other IT-led support tools with little dialogic character, to store and codify tacit knowledge. Knowledge management systems may also be geared towards standardisation to meet customer requirements:

For example, a lot of the engineer’s knowledge can be captured using knowl- edge-based (or expert) IT systems to map out design and installation processes, by following a conventional rule-based approach. The outcome is that customers get consistent specifications, price quotations and quality from any company in the group. (David Bell, Chief Knowledge Officer at Clyde Bowers, quoted in Scottish Business Insider, October 1999)

The impact of the valorisation of knowledge is not merely on employees. Corporate power places constraints on the process of independent, collegial knowledge production, as universities become ever more dependent on private sources to finance scientific research. For example, 37 per cent of biotechnology scientists who are members of the American National Academy of Sciences have industry affiliations (Rifkin, 1999). Nor in market terms do the fast growing knowledge intensive industries behave any differently from more traditional sectors. Rifkin details how, contrary to the network myth, the concentration of capital (and vertical integration) is unabated. Global pharmaceutical companies have recently spent $3.5 billion buying up biotech firms. When they are not buying them up they are tying them to licensing agreements (at the cost of $1.6 billion expenditure).

We can see similar trends with respect to the Internet. It is true that many new e- commerce companies have come into being, backed by venture capitalists and making a lot of money for their founders. However, the trends in what Schiller calls ‘digital capitalism’ are also towards concentrated ownership and vertical integration.

Smaller companies that specialised in what were initially niche markets at the frontier of the liberalisation process worked the new territory. When they succeeded, major traditional suppliers either snapped them up or rushed to develop similar applications of their own. (Schiller, 1999: 28)

Huge mergers such as that between Time-Warner and AOL merely confirm the direction of change. With this in mind, the association of knowledge production and acquisition with new forms of ownership is a mixture of wishful thinking and confusing the early stages of market formation in an emergent sector with its mature, normally oligopolistic characteristics. As their predecessors before them, knowledge economy theorists are unwilling or unable to explain such trends. In The Last Days of the Giants, Baldock (1999) asserts that the current merger wave is the ‘last throw’ of the big corporations as they try to fight off the threat from e-commerce. A decade earlier, Tom Peters (1992) had described mergers as ‘madness’. The fact remains that the global boom in mergers and acquisitions continues unabated. Despite relatively

CO N T I N U I T Y A N D C H A N G E AT W O R K • 1 7 1

poor returns for shareholders, United Nations figures show that cross-border mergers and acquisitions drove foreign investment volumes to a new record of $856 billion in 1999 (Mark Atkinson, Guardian, 10 October 2000). With market expectations of double-digit growth per year, such activity is the quickest way into mature industries. In addition, economies of scale and scope are more important than ever, particularly for knowledge intensive firms with huge research and development costs. Responses vary from strategic alliances to link-ups such as the £110 billion merger of Glaxo Wellcome and Smith Kline, but the message remains the same – big is still beautiful.

Finally, what about claims that knowledge work is the engine of growth in the new economy? As Warhurst and Thompson (1998) demonstrate, at the heart of NEO arguments are a host of conceptual and empirical confusions. Proponents exaggerate the scope of knowledge work. Occupations requiring ‘thinking’ skills will expand, but that expansion will be limited in scope and potential for employment growth. While there are certainly high tech clusters (Finegold, 1999), only 7 per cent of the fastest growing occupations in the US – the imputed model of economic transformation – could be classified as knowledge workers who manipulate symbols and ideas (Harwood, 1996). It is easy to conflate the production of high-tech goods with the need for a workforce of highly skilled workers (Keep and Mayhew, 1999). Even knowledge-intensive firms ‘actually require quite small numbers of highly skilled employees once the major information transfer technologies have been put in place. A very small élite of system designers and strategic managers is required, whilst the globally dispersed plants and offices can be serviced by workers with very much less training and knowledge’ (Ackroyd and Lawrenson, 1996: 157).

Further down the occupational hierarchy there is a further and consistent conflation of knowledge and information. Lists of ‘knowledge workers’ (for example, Handy, 1995: 4) consist of heterogeneous occupations who have little in common other than handling information. Financial services, for example, frequently require little more of workers than information transfer – the inputting of customer details onto pre-programmed screens and software. As even an exponent of the knowledge economy, Robert Reich, observes, ‘The foot soldiers of the information economy are the hordes of data processors stationed in back offices with computer terminals linked to worldwide information banks’ (1993: 23). Many of these foot soldiers no longer belong to Western armies. Locations in the Caribbean or the Philippines are promoted as ‘digiports’ with high tech infrastructures, and ‘skilled’ but low-paid staff.

Most employment growth, actual and projected, is in service work, and within that predominantly at the low end in serving, guarding, cleaning, waiting and helping in the private health and care services, as well as retail and hospitality industries – both in the US and the UK (Crouch, 1999). Much of this constitutes a growth in ‘no knowledge’ work, at least as measured in conventional terms. Recent skill surveys reveal that 41 per cent of personal services, 62 per cent of sales and 80 per cent of ‘other occupations’ require no qualifications for entry (Brown and Keep, 1999). When managers were asked what proportion of their workforce could be described as skilled (such as professional, associate professional, technical or craft), there were some startling results. For example, 40 per cent of managers in wholesaling/retailing and 57 per cent of managers in financial services said that they had no skilled non-managerial employees (Cully et al., 1998, and see Keep, 2000). There is an increasing emphasis from many employers on employee attitudes and even their ‘aesthetic’ qualities rather than formal knowledge and skills (Nickson et al., 2001). Such facts are almost