Job Design and Technology

Taylorism vs Anti-Taylorism

Hans D. Pruijt

• Publisher: Routledge
• Year: 1997
• Topic: Business
• eBook ISBN:
  9781134725090
• Subtopic: Business General
• Index: Business

Despite global competition and the need for speed, flexibility and quality, trends such as lean production and McDonaldization show that Taylorism remains alive and well in the contemporary workplace. There is however a countermovement, particularly in North-West Europe, where successful alternatives are being pursued. Job Design and Technology fil

1

Introduction

This book is about Taylorism and, above all, about the fight against it. Taylorism is taken as management strategies that are based on separating thinking from doing: managers think and plan, workers basically execute orders.

On the surface, Europe seems to adopt its organization of work entirely from other parts of the world: first by deepening the division between thinking and doing according to the principles of classical Taylorism, later by embracing neo-Tayloristic trends like lean production from Japan and McDonaldization from the US.

There is, however, a countermovement. In some organizations, innovative managers, staff personnel and union groups are pursuing anti-Tayloristic¹ strategies. They receive support from researchers who, combining consultancy and research, accumulate expertise on changing working life and feed this back into organizations. National unions have supplemented their bread-and-butter activities with involvement in the organization of work. So have politicians, who, in several countries, have pressed for working life legislation that covers work organization as much as health and safety. And in Scandinavia and Germany, there are state-sponsored research and development programs that explore alternatives to Taylorism. These programs include efforts to develop technologies that do not have a Taylorist bias and that could support change in organizations.

This book addresses the following questions:

• What types of initiatives does this countermovement consist of?
• To what extent can these types of initiatives be successful in bringing about durable change at the shop-floor level?
• How can differences in the effectiveness between these types of initiatives be explained?

There are several reasons why these questions are interesting:

• Taylorism proved to be more persistent than many had expected.
• Anti-Tayloristic initiatives can give insight into the range of options for the application of information technology and job design, and the possibilities for planned change in this area.
• Neo-Taylorism needs an alternative.

Taylorism Proved to be More Persistent than Many had Expected

In the 1980s there were signs—at least on paper—that in some sectors Taylorism was coming to an end. Kern and Schumann (1984: 19) reported a rise of “new production concepts” in the core sectors of German industry—the automobile, machine tools and chemical industries. These new production concepts revolved around greater respect for skill and worker involvement. This meant that in the core sectors, the polarization between elite and routine workers was coming to an end. Product quality, flexibility and enabling technology were given as reasons for this change. Routine workers would become “system regulators”—system regulators monitor the actions of automated machinery and intervene when they deviate from their programmed course. Piore and Sabel (1984) claimed that flexible technology and fragmentation of markets would usher in an era of “flexible specialization.” However, it seemed that the change process was not universal. Kern and Schumann, for example, specifically excluded the food sector.

Subsequent scrutiny showed that—even in those core sectors—change was less than expected, maybe even non-existent. System regulators remained few—around 2 to 5 percent.² A large-scale study on information technology in the European service sector (Child and Loveridge 1990: 360) concluded that banks rarely exploited the new possibilities of broader access to pooled information and knowledge and that “modifications to existing hierarchies are modest.” Huys, Sels and Van Hootegem (1995) found that Belgian car plants did acquire the capability for flexible adaptation, but not through discussions among workers. On the contrary, it was sophisticated planning in advance that produced flexibility. Furthermore, management was pushing towards enhanced predictability and uniformity of actions. Finally, a mammoth trend study on rationalization in the German car, chemical and machine tool industries (Schumann et al. 1994: 659) concluded that the division between conception and execution proved stronger than had been predicted a decade earlier (in Kern and Schumann 1984). Professionalization was not a sufficient condition for emancipation. Moreover, the study found that a substantial gap remained between direct and indirect sectors, between “new production specialists and specialized specialists.” A telling point is that in the car sector, new assembly lines continued to be installed.³

Studying the ups and downs of those organizations that did take the plunge to depart from Taylorism might help explain why the transformation did not materialize to the extent that many had hoped for. No discussion of work is complete without an attempt to answer the question of whether the reality described is the only possible reality; social reality reveals itself best when one tries to change it.

The Principles of Taylorism

“Taylorism” (after Frederick Winslow Taylor, management expert, 1856–1915) has become a catchword for management strategies that lead to impoverished jobs. (Impoverished in the sense that skill content and autonomy are going down and that workers are allowed less influence on the work process at large.)

The central feature of Taylorism is the separation of conception from execution. Managers achieve this by applying three principles. The first principle of “scientific management,” as Taylor called it, is the decoupling of the labor process from the skills of the workers: “The managers assume … the burden of gathering together all of the traditional knowledge which in the past has been possessed by the workmen and then of classifying, tabulating, and reducing this knowledge to rules, laws, and formulae” (in Braverman 1974: 112).

The second principle is: “All possible brain work should be removed from the shop and centered in the planning or laying-out department” (in Braverman 1974: 113).

The third principle is that management should not leave it to the workers to decide how they go about their tasks. Instead, management should prescribe exactly how, and how fast, the tasks must be performed. The main reason for doing this is to prevent workers from holding their output down.

Taylorism is a refinement of the management strategy of detailed division of labor. It is important to draw a sharp distinction between detailed division of labor and specialization. Through specialization, people can develop themselves further in their crafts or professions, whereas detailed division of labor reduces people to performers of routine tasks. Detailed division of labor entails analyzing a production process and breaking it down into a multitude of tasks performed by different workers. In this way, a craft-based labor process, that was once controlled by the workers themselves, falls to pieces. Then, managers put the pieces together to create a process that is under management’s control. The financial advantage of this strategy is that it becomes possible to hire less well-paid workers. This principle was already clearly stated by Charles Babbage in 1832—the same Charles Babbage who devised the first computer architecture (Braverman 1974: 188). Taylorism tends to carry the detailed division of labor to new extremes, where work cycles are measured in seconds.⁴

Taylorism implies low-trust relations between employer and employees. Therefore, direct control is needed to ensure that labor power bought is turned into labor performed. This control question urges managers to find ways of imposing on workers what they should do, in what way, within which limits and at what pace, and to evaluate work performance and apply sanctions. Taylor’s prescriptions amounted to reliance on a raft of supervisors, but the Ford Motor Company proved that a mechanically-paced assembly line is a more efficient control system. The assembly line functions as a system of technical control, which means that the entire production process, or large segments of it, are based on a technology which regulates the working pace and controls the labor process (Edwards 1979: 112–113). The assembly line supplants the direct conflict between worker and foreman. Scientific management, as an explicit method, became unfashionable after the 1930s. Its principles, however, have continued to have an impact on the design of jobs, up to this day (Braverman 1974: 119; De Sitter 1981: 21; Merkle 1980: 3; Schumann et al. 1989: 67). Examples are:

• In banks and insurance companies there is a tendency to permanently analyze and standardize the work. The routine tasks are relegated to a category of workers who are only engaged in “bulk” work, and who have to work according to strictly specified rules. An elite group handles the more complicated, incidental cases. The work of this elite group is permanently analyzed to see if parts can be formalized and transferred to the bulk group (Doorewaard 1986).
• In the early 1990s, a labor shortage in the Dutch health care sector prompted managers to tap the reservoir of people with little formal education.⁵ To create special jobs for this category of workers, they split off simple tasks from nurses’ jobs. Subsequently, a Labor Party committee took this as a model solution for the problem of creating jobs for the disadvantaged.⁶
• Finally, the rise of the McDonalds-type firm (“McDonaldization”) (Ritzer 1993), saturated by predictability and controllability, testifies to the vitality of the Tayloristic organization concepts.

Anti-Taylorist IC Initiatives can Give Insight into the Range of Options for the Application of Information Technology and Job Design, and the Possibilities for Planned Change in this Area

Some, for example, Blauner (1964) and Woodward (1958), saw automation and information technology as factors that tend to upgrade work and put workers back into control of the labor process. Often it does not come out in this way.

• Depending on how it is used, information technology facilitates Taylorist division of labor. An example of an application that can strengthen the separation of conception from execution is Numerical Control (NC). Another current example of detailed division of labor is the creation of data-entry work, leading to dead-end jobs, particularly for women. Information technology makes this kind of division of labor more feasible, because transporting information becomes easier. This enables firms to tap the labor markets of homeworkers and the labor markets in low-wage countries.

Numerical Control

An important technology in the discussion about Taylorism is Numerical Control (NC). A changeover from hand control to numerical control facilitates a division between, on the one hand, planning and programming, and, on the other hand, unskilled operator work (Braverman 1974: 197–206; Noble 1983). NC is an example of the way in which Tayloristic views can influence the development of technologies, that in turn influence the organization of work. In his study on the coming into existence of the NC technology, Noble (1983: 86) emphasizes that technology has a social history. The design process involves choice—the selection or discarding of alternatives. Social relations affect these choices. This is not a clear-cut process; the outcome of technological development is characterized by a large proportion of unintended consequences.

Originally, two competing principles guided the development of automatic machine tools:

1. The “record playback” principle, i.e. that a skilled worker uses hand control to produce a first piece of work, which is then automatically copied to produce the required number of products. This technology depends upon the skilled worker.
2. The NC principle in which the instructions for the machine are delivered on tape. This system makes a division of labor between programming and operating possible.

The NC principle prevailed, to a large extent, because of financial support from the defense sector. The military wanted to avoid the presence on the shop floor of blueprints of military equipment.

NC makes the division of labor between programming and operating the most likely work organization. However, there is no complete determination: apart from programming in a separate department programming on the shop floor itself is possible.

Further development brought the CNC (Computer Numerically Controlled) machine. CNC technology makes shop-floor programming easier to realize (Gill 1985: 77; Sorge et al. 1983).

The next step in the development is the DNC (Directly Numerically Controlled) machine. In DNC, there is a direct connection between the CAD (Computer Aided Design) system and the machine tools. This implies the disappearance of programming and, with it, the opportunity for shop-floor programming.

• Through automatic data collection and feedback loops, the introduction of computer technology widens the possibilities for computer surveillance systems. Management does not actually have to use the computer-collected information to control the work force, but it reserves the power to do so. This is the “panoptical” principle: all subjects know that they can be observed from a central point, but they have no way of knowing whether, at a given moment, actual surveillance takes place (Foucault 1975: 197–229).⁷
• Logistical information systems are in use that issue instructions to workers and then check whether the instructions are executed within the specified time.⁸
• Advanced automation can entail the incorporation of human skills and decision-making into machinery, and thus the removal of brainwork from the shop floor—a key element of Taylorism.⁹

Expert Systems

A subdiscipline of computer science, Artificial Intelligence, is partly devoted to enabling computers to take over skilled tasks. In contrast with traditional automation, Al deals with problems for which no algorithmic solutions are known (i.e. that not all steps leading to the solution are known). One of the most flourishing areas is that of expert systems. An expert system is a program for solving decision or judgment problems on the basis of built-in expertise that was gained from human experts. The principle is that a “knowledge engineer” interviews experts and tries to make their expertise as explicit as possible. This process of “elicitation” is difficult because, to a large extent, expertise consists of intangible rules of thumb and heuristics; reluctance of experts to give away valuable knowledge may be a complicating factor.

One way to represent knowledge in an expert system is to use “production rules”—rules with an IF … THEN structure. The expert system includes a knowledge base of production rules and inference engine that selects rules and takes appropriate action. Possible actions could be: asking the user—who is consulting the system—for information, performing calculations, looking up information in a database, etc.

The yardstick for success is whether or not the expert system matches or improves upon the accuracy of human experts. This line of development in computer science is more oriented towards competing with humans, than with supporting them.

An example of a successful expert system is Digital Equipment Corporation’s XCON. XCON’s purpose is to configure the VAX line of computers. Each VAX is an often unique combination of thousands of parts. A first abortive attempt was to automate parts of this process in a traditional way. A big handicap was that the project meant “aiming at a moving target,” because of the constant introduction of new parts and options (Scown 1985: 114). An expert system, growing to around 4,200 production rules, proved accurate in about 95 per cent of the cases.

An addition to XCON was XSEL, an expert system designed for assisting salespe...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Dedication
• Contents
• List of illustrations
• Preface
• Acknowledgments
• 1. Introduction
• 2. Inside Consensual Alternatives to Taylorism
• 3. Enlightenment: Showing Managers the Way
• 4. Consensual Alternatives: Achievements in Job Design
• 5. Management Power and Efficiency as Constraints
• 6. Alternative Alternatives
• 7. A Final Note on Anti-Taylorism
• Appendix
• Notes
• Bibliography
• Index