[ Giæver Home ] [ Henrik Giæver CV ] [ TQM & Innovation ] [ Quotes ] [ The Giævers ]
[ Sociology ]

Does Total Quality Management restrain innovation? -a discussion of the effect of TQM in Knowledge Intensive Organisations

Oslo, 5 May 1998; Henrik Giæver (e-mail)
Høvik, 29 Nov 1999 Approved as DNV report no.: 99-2036

CONTENTS

1. Summary
2. Introduction

3. Total Quality Management

4. Learning in Organisations

5. Discussion: TQM and learning

6. Conclusion, Closing Remarks
7. Literature

Appendix 1: Deming's 14 Points for Management

1. Summary

This study is based on analyses of literature and interpretation, to establish the organisational implications of Total Quality Management (TQM) compared to theories of learning. Learning theorists stress the critical importance of explorative learning for organisations that are exposed to highly competitive and unpredictable environments. The study identifies a harmonious link between TQM and incremental, small steps learning, and shows how it can provide the metrics for managing knowledge based assets. It also concludes that some practices of classical TQM restrain innovation.

2. Introduction

2.1 Background; Motivation for the Study

There is an increasing awareness that knowledge is becoming the key factor in improving productivity and creating growth. "Knowledge now has become the real capital of a developed economy" (Peter Drucker 1989: 180). This shift may place new demands on how organisations should be designed and controlled.

With a few exceptions, (e.g. purchase of a brand name), assets based on knowledge are not reported in the financial statement regardless of their financial importance. It is an increasingly evident paradox that formal reporting mechanisms are becoming less and less relevant for the service industry, which is the fastest growing segment of the economy. Says Steven M.H. Wallman of U.S. Securities and Exchange Commission:

Total Quality Management (TQM) in various forms has for decades proved successful in improving productivity; continuous improvement and learning being essential tools. For knowledge intensive organisations it is of interest to investigate whether the successful tools of Quality Management are also suitable for them.

TQM, as presented by the European Foundation for Quality Management (EFQM) in the criteria of the European Quality Award, suggests solutions to the insufficiencies of present financial planning and reporting systems (EFQM 1996). Much in line with a "Balanced Scorecard" approach it requires planning and reporting of intangible assets such as identification of critical processes, people satisfaction, customer satisfaction, impact on society and business results other than financial (EFQM 1996, Kaplan and Norton 1992).

Knowledge being an important asset, the rate of learning becomes a key factor. Leading theorists suggest that there are two significantly different types of learning:

• Small steps improvements, incremental learning: adaptation without asking the fundamental questions.
• Radical change: change based on a fundamental questioning of the reasons and motives behind the present situation.

Chris Argyris raises a flag of warning about uncritical application of Total Quality Management (TQM). He agrees that TQM has built-in mechanisms of single-loop learning that will bring about cost reduction and other improvements to established processes, but questions whether the TQM approach inspires the radical changes that is also needed (Argyris 1994: 47). James G. March points out that refined exploitative learning may be self-destructive for the organisation in the long run (March 1991).

Det Norske Veritas (DNV) is a knowledge intensive organisation that serves the international marketplace with a broad spectre of quality and management related services. Det Norske Veritas has committed itself to TQM as its own Management Philosophy, and is presently also working to provide the concepts, methods and tools for managing its intellectual capital, i.e. the intangible assets. With the popularity and proven capability of TQM in producing efficiency and results for a wide range of very different organisations, and the present awareness of the importance of knowledge for the survival and success of organisations, there is a need to investigate relation of TQM to scholarly organisational theory, thereby getting an improved foundation to assess its usefulness and limitations for knowledge intensive organisations.

The link between TQM and theoretical organisational theory is loose. Collins "Dictionary of Sociology" describes Total Quality Management as: "Managerial technique for the pursuit of continuous improvement through strategic, processual and cultural change in organizations (…)" (Collins 1995: 692), however micro sociological theories; e.g. about bureaucracy, system theory, decision making, contingency and bounded rationality, offer models and tools for interpreting the management and business practices associated with TQM.

What is now recognised as TQM, has evolved a great deal since it was first articulated by its main advocates, W. Edwards Deming, Joseph M. Juran, and Kaoru Ishikawa in the 1950s. TQM has become so differentiated that it may be difficult to establish what exactly it is, and whether it still has a core. (Hackman 1995).

2.2 The Objective of Study

The primary objective of this study is to investigate the effect of Total Quality Management (TQM) practices, as taught by W. Edwards Deming, on innovative learning, to evaluate whether TQM in fact restrains innovation.

The study will also discuss TQM effects on other types of learning, and discuss the suitability of TQM in knowledge intensive organisations.

2.3 Methodology

TQM may escape the critical attention of researchers, because it is not firmly founded in traditional sociological or organisational theory. To better achieve the objective of this study, an attempt is made to link TQM to scholarly organisational theory, so that knowledge about these theories may give new insights on possibilities and limitations of TQM in relation to learning. My ambition is to discuss whether TQM fits into one of the five configurations that Henry Mintzberg presents in his book: "Structures in fives: Designing effective organisations" (Mintzberg 1983). Richard Hackman's article "Total Quality Management: Empirical, Conceptual and Practical issues" has been particularly useful in pursuing this end (Hackman 1995). Based on the study of TQM literature and research, I will interpret and suggest the core characteristics of TQM.

TQM has evolved to mean many different things, slogans sometimes winning over practices (Hackman 1995). In order to identify a core of TQM, I have studied how the quality concept has developed in the social setting of modern industrialisation. The quality awards (Deming prize in Japan, Malcolm Baldridge Quality Award in The United States, European Quality Award) receive a great deal of attention, and they all have traces to the teaching of W. Edwards Deming. Hence I have chosen to study the writings of W. Edwards Deming and two of his close associates: Henry Neave and William W. Scherkenbach. Main sources of information are Det Norske Veritas course material (DNV 1995-98), Deming's own book "Out of the Crisis" (Deming 1986), Henry R. Neave's book "The Deming dimension" (Neave 1990) and William W. Scherkenbach: "Deming's road to continual improvement" (Scherkenbach 1991). Some examples of TQM practices are drawn from Det Norske Veritas' TQM program. (DNV CMS 1997).

Chris Argyris is very articulated and influential in business practices on the topic of organisations and learning. His article "Good communication that blocks learning" pinpoints the objective of this study (Argyris 1994). To acquire a broader perspective of the function of learning in organisations, I have studied James G. March's article "Exploration and Exploitation in Organizational Learning (March 1991) and other writings by Argyris, March and other writers. (Argyris 1994, March and Olsen 1988, Kim 1990).

I then proceed to discuss the TQM core characteristics in relation to these leading theories of learning to conclude about the compatibility of TQM and learning.

3. Total Quality Management

3.1 The Development of Quality Management

There are a number of definitions of quality:

• Philip Crosby: Quality is conformance to requirements (Schultz 1994: 94)
• Joseph M. Juran: Quality is fitness for use (Schultz 1994: 52)

Interestingly W. Edwards Deming does not define quality, except indirectly on statements such as: "It will not suffice to have a customer that is merely satisfied" (Deming 1986: 141)

In 1994, the International Standardisation Organisation's definition of quality was expanded to include other entities than products and services, and satisfaction of implied needs: "Quality: (2.1) totality of characteristics of an entity (1.1) that bear on its ability to satisfy stated and implied needs NOTES (...)" (ISO 8402, 1994). An 'entity' may for instance be: a product, a service, a process, an organisation, a system or a person.

The need for formal control of the quality of products arose with industrial mass production. Specialised production as well as the detachment and geographic distance between manufacturers, called for more formal communication about the requirements to and nature of the product. Legal issues inspired new ways of verifying that the product conformed to agreed requirements.

Driving factors in the introduction of more sophisticated mechanisms to control quality were:

• increasing specialisation, trade and competition
• increasing product complexity
• the manufacturing of products where consequences of faulty product could be fatal (e.g. defence industry)
• an increased understanding of the profitability of producing right quality first time

The groundwork for TQM was done by American consultants assisting in the rebuilding of Japan after the second world war. This blend of American scientific approach with Japanese holistic traditions of the oneness of humanity and nature, the oneness of body and mind and the value of interaction between self and other (Nonaka and Takeuchi 1995: 27), have produced a management system theory that is one of the causes of the Japanese economic success in the decades after World War II . (Neave 1990: Preface)

In the 1950s Deming worked as a consultant for the Japanese industry. In his honour, the Union of Japanese Science and Engineering (JUSE) instituted annual Deming Prizes, first awarded in 1951 (EFQM 1996: I, 12).

National and international product standards have evolved alongside industrial development. The International Standardisation Organisation published the first edition of the Quality Management standards (ISO 9000 series) in 1987, with a revision in 1994. In 1998 more than 170.000 organisations world wide have been certified according to the ISO 9001 or ISO 9002 standards, in Norway approximately 1.000 (Sources: Det Norske Veritas Flash Report February 1998, and Mobile Services Company Ltd, December 1997).

In 1991 The European Foundation for Quality Management launched The European Quality Award.

3.2 "Theory of Profound Knowledge" and "14 Points for Management"

A foundation in Deming's teaching is his Theory of Profound Knowledge, that appears in four parts:

• Theory of Systems.

Figure 1 Demings flowdiagram, showing his view of the organisation as part of a system. Source: Deming, Out of the Crisis, 1986. Redrawn by author.

• Theory of Variation.

• Theory of Psychology.

• Theory of Knowledge.

Figure 2 Deming was introduced to the Plan, Study, Check, Act by the statistician Walter A. Shewhart. Deming later replaced the word "Check" by "Study". Source: Louis E. Schultz, "Profiles in Quality", 1994. Redrawn by author.

Being closely associated with the economic success of Japan, Deming eventually also gained an audience in the United States. He was awarded The National Medal of Technology in 1986:

Part of the medicine he prescribed to transform American industry was formulated in what Deming called "14 Points for Management" in Out of the Crises 1986 (Deming 1986, see also Appendix 1)

3.3 TQM in Practice

American surveys report that the most commonly used TQM practices are

• problem solving teams
• training as a change effort
• top-down implementation
• development of relations with suppliers
• collecting customer information.

Most quality writers and practitioners would agree to these principles as common denominators amongst a variety of applications of TQM. Other characteristics would be teamwork and employee involvement and a holistic approach. Benchmarking (learning from other organisations) has through the Malcolm Baldridge Quality Award and European Quality Award criteria, become strongly associated with TQM in the United States and Europe. It is a worrying (seen from a Deming viewpoint), that a large majority of the TQM organisations use some form of modified performance measurement and reward systems contrary to Deming's strong advice (Hackman 1995).

Det Norske Veritas has in its TQM implementation chosen to focus on these principles:

• Customer Focus
• Process Orientation
• Continuous Improvement
• People Empowerment

(Det Norske Veritas, 1997)

3.4 In Search of a Theoretical Foundation for TQM

Neither the leading Quality theorists, nor institutions such as the European Foundation for Quality Management (EFQM), have made serious attempts to identify the relation of TQM to scholarly organisational theory. Researchers' view range from TQM as a late version of Scientific Management, to a cultural phenomena. (Thomsen 1998, Hackman 1995).

It is an interesting fact that both W. Edwards Deming, and Joseph M. Juran were employed at Western Electric's Hawthorne Plant in Chicago. Deming had summer jobs in 1926 and 1927, Juran was there from 1924 to 1941. They both acknowledge the importance of the statistical training they received under Walter A. Shewharts supervision. However, neither of them mentions the famous Dickson/ Roethlisberger or the Mayo experiments at Hawthorne (Perrow 1986). An influence may be traced in that they both advocated a human centred management approach.

• Rejecting Scientific Management but rejoicing scientific method

• TQM: an open system view, compatible with socio-technical theory

• TQM and Contingency theory

• The strategic control subsystem
• The operational subsystem
• The human subsystem
• The managerial subsystem (Burrell and Morgan 1978: 168-170)

Figure 3 An attempt to apply the contingency theory subsystem to the European Model for Total Quality Management.

3.5 A Machine Bureaucracy?

Below is a table (Figure 4) presenting the five configurations of organisations that Henry Mintzberg presents in his book "Structure in fives" (Mintzberg 1983). The underlining is my attempt to indicate the corresponding TQM characteristics.

Figure 4 Mintzberg's five main designs put into a matrix. The author's interpretations of the corresponding TQM characteristics are indicated by underlined font effect. Italics indicate weaker correspondance.

• Prime TQM co-ordinating mechanism: Standardisation of work processes

• Key part of TQM organisation: Technostructure and Strategic apex

• Main design parameter in TQM: Behaviour formalisation

• Typical TQM situational forces: Control in a stable environment

3.6 Summary of TQM Core Ideas

Based on the above considerations, the following elements represent the basic ideas of TQM for further discussion in this study; reference is made to Appendix 1, Deming's 14 point for management.

• TQM is designed for long term operation and survival of organisation in stable environments (1)
• TQM assumes that employees care about the performance of the organisation (10, 11)
• TQM regards the organisation as holistic, open system: Theory of system, systems within the system. The survival of the organisation depends on how well it serves its customers and other stakeholders (5, 9)
• TQM stresses continuous improvement for the benefit of all that are related to the organisation. (5)
• TQM focuses on training and education (6 and 13)
• TQM focuses on work processes and cross functional teams (3, 9, 11)
• TQM advocates management by fact; interplay of theory prediction and collection of data, systematic data collection and interpretation to monitor processes and the understanding of variation
• TQM appreciates that people are whole individuals, with individual learning styles, emotions (fear, pride etc.). (7, 8, 9, 10, 11, 12, 13, 14)

In short: a typical TQM organisation has likeness to Machine Bureaucracy configuration in Mintzberg's sense, but with an obsession with the customer, fact based management and continuous improvement.

4. Learning in Organisations

4.1 The Rationality of Individual and Organisational Learning

There are a number of diverse definitions of learning. One of the learning experts, Charles Catanya, in his book "Learning", simply refuses to define it. (Catnya 1984: 2) For this study, I will use James March's definition: "Potential change in behaviour because of experience" (March and Olsen 1988). Note that the definition covers desirable as well as not desirable learning.

Understanding how organisations learn is essential in order to improve their performance. According to March, organisations learn from their members and accumulate this collective knowledge in the form of procedures, rules and norms: the organisational code. By various forms of instruction, indoctrination and exemplification, individuals are socialised to organisational beliefs. Individual beliefs and organisational code converge over time. (March 1991).

March proposes a simplified cycle of choice which is also a theory of organisational learning, illustrating the interaction of the individual, the organisation and the environment (Figure 5).

Figure 5 The complete cycle of choice and a theory of learning. The circles represent the four learning constraints; possible breaks in the learning cycle: 1 Role constrained, 2 Superstitious, 3 Audience, 4 Ambiguity. Source: March and Olsen 1988. (Modified by author by combining the original figures).

With perfect rational calculations and perfect learning from experience, performance would automatically improve: Based on his cognition and 'models of the world' (D), the individual takes action (A). This affects the organisation's actions and choices (B), leading to a desired effect on the environment which responds (C). The response (C) is observed by the individual, who then gets his old views reinforced or renews his 'model of the world' (D).

March has in earlier works shown that organisations are far from super-rational (March and Simon 1958). Elaborating on the bounded rationality of organisations, he therefor acknowledges that between every stage (A, B,C, D) in the learning cycle, an irrational break may occur. The circles (Figure 5) indicate these 4 possible breaks in the learning cycle:

1. Role constrained experiential learning: Standardised operating procedures, organisational norms or group pressure, prevents the new individual insight (D) in affecting individual action (A). Hence a modified organisational behaviour (B) is also inhibited.
2. Superstitious experiential learning: The modification of individual beliefs (D) is based on an environmental action (C) that had nothing to do with the organisational action (B). In this case, the individual observes an environmental response, but he misunderstands the causality (B).
3. Audience experiential learning: The individual action has insignificant effect on organisational action; there is no adaptation of the organisational behaviour (B).
4. Experiential learning under ambiguity. The environment responds (C) to the organisational action (B), but the environmental responses are seen only dimly by the individual, so they do not affect his beliefs and cognition (D). In this case, there are environmental responses, but the individual is blind to them: His beliefs do not change, although they should.

(March 1991: 346-350). Similarly, Argyris shows how defensive reasoning and personal self-protection hinder especially 'double-loop' learning (Argyris 1994).

In other words; under certain conditions, learning from experience is desirable, but under other conditions experiential will be misleading. Social psychologists support this view: People are not scientific in the way they perceive; they are inclined to see that which induces familiar feelings; they tend to make assumptions about causality. Examples: People remember and attribute most importance to experiences and explanations that are given in effective language (salience bias). People remember and attribute most importance to experiences and explanations that are recent (Hewstone, Stroebe, Codol and Stephenson 1990: 125-134). Giddens puts these mechanism into a context of man's need to keep the fundamental anxiety that threatens pride and self-esteem under control (Giddens 1991).

4.2 Exploring the New or Exploiting the Old?

The learning theorists distinguish between incremental learning and learning in the form of radical change. They use different terms, but the essential division in two is the same.

Figure 6 Table showing how the terms of Argyris, Kim and March correspond. Sources: Kim 1990, March 1991, Argyris 1994.

Exploitative learning has to do with refining existing technology, leaner production, improved implementation and execution, more efficiency.

Explorative learning represents experimentation, playfulness, innovation, risk-taking and uncertainty; searching for new possibilities.

There is a danger that exploitative learning, with its lower risk and more certain short term pay-back, may take attention away from explorative (renewal) learning. While exploitation would normally lead to superior performance in the short term, there is the danger that lack of exploratory search may be self-destructive in the longer term (March 1991).

4.3 Slow Learners and Turnover Increases Exploration

When discussing the fine balance between explorative and exploitative learning, March introduces simulations of organisational learning by a mathematical model of external reality, organisational members beliefs about the external reality, the continuous modification of beliefs (socialisation), and the dynamics of the organisation's code (March 1991: 74). He obtains support for his hypothesis that rapid learning is not always desirable (they adapt so quickly that the organisational code does not absorb the new knowledge) (March 1991: 76). He also concludes that a moderate level of personnel turnover can increase exploration and improve aggregate knowledge thus safeguarding against knowledge degeneracy (March 1991: 76).

4.4 Slack and Smart Foolishness

During periods of success, a typical organisation will direct some of its efforts toward activities which cannot be justified easily in terms of goals; slack. At first glance this looks like waste and private sub-optimisation for the organisation (March and Olsen 1988: 4). Slack may have a highly desirable effect though: It will protect individuals and groups from normal controls, thus leading to the kind of foolish behaviour which is a prerequisite for new ideas (March and Olsen 1988: 179).

5. Discussion: TQM and learning

5.1 Organisational Learning and Focus on Work Processes

Organisational code (habits, procedures, standards, methods, recipes etc) is the memory of the organisation (March 1991). In a TQM company the creation of this memory is systematised and assisted by tools (process modelling and monitoring, product descriptions, standard operating procedures, formal training, inspections and improvement etc.). TQM's bureaucratic character ensures that a lot of the organisational memory can be found in the documented and implemented management system, files and records; facilitating long term organisational memory, but hindering the 'forgetting' which is necessary in 'double-loop' learning.

5.2 Learning, Continuous Improvement and Winning the Race

"As work is standardised, as techniques are learned, variability both in the time required to accomplish tasks and in the quality of task performance is reduced". (March 1991: 83). Compare Deming: "If I had to reduce my message for management to just a few words, I'd say it all had to do with reducing variation" (Deming cited in Neave 1990: 57).

However, March finds that improvement of organisational performance often seems to take place in two ways: a) reducing the chance of finishing last without affecting the chance of finishing first (amongst competitors) b) increasing the average performance even if the organisation does not finish first (March 1991: 83). In other words, there is a danger that exploitative learning does improve the performance, without improving the competitive position.

Knowledge intensive products (e.g. software) typically requires long development period and heavy investments in design and initial construction. The subsequent mass production and distribution is quick and low-cost (e.g. CD-ROM or net-based distribution), with a possibility of "winner takes it all". If a company manages to create a "path-dependency" or "lock-in" situation for a product, this will generate huge revenue. If finishing first is of great importance in such an environment, March suggests that a better strategy may be to increase variability to get the winner, even if it is at the expense of performing lower in average performance (March 1991: 83).

5.3 Learning from Experience, and Management by Fact

March's learning circle (Figure 5) and the PDSA circle of TQM (Figure 2) comply in their assumptions of how an entity adapts to the environment, in a repeating circle of action and reflection. Deming seems to be aware of the learning constraints; his focus on understanding statistical variation (page 7) is a warning that causality is not always as it appears. Misconceptions inspires action that only distorts the process; e.g. when the shower water is too cold, we turn up the hot water too much, it gets too hot and we over-react again (Scherkenbach 1991: 54). This is exactly what March illustrates (on organisational level) as learning under ambiguity (Figure 5). Deming's teaching is a crusade for rationality and scientific method; March has elaborated on the obstacles of irrationality management is up against.

5.4 Explorative Learning in a Machine Bureaucracy

TQM in the machine bureaucracy is tuned for elimination of waste, optimisation and rationality. March and Olsen show that slack, even if it is considered highly undesirable by top management, may have positive effects for explorative, innovative learning. (March and Olsen 1988). Argyris shows that 'double-loop' learning requires tougher questions, a more committed personal accountability and less linear reasoning than typical TQM. (Argyris 1994).

Deming speaks against personnel turnover (Deming 1986: 98, 120), March and Olsen shows how turnover may be important to organisational learning and adaptation to the environment (March 1991). This seeming disagreement between TQM and organisational learning may be insignificant; Deming's worry about top management job hopping and the less controversial word 'recruitment' can substitute 'turnover' in March's article (March 1991)

6. Conclusion, Closing Remarks

With its focus on continuous improvement, management by fact, training and education, appreciation of the organisational members motivation and learning style, TQM is as facilitating for exploitative learning as an organisational philosophy can possibly get. There is a good match between TQM's fact based management style and March's theory of superstitious learning. This seems to be true also for deviant Quality Management practices, such as those based on ISO 9001/2 or the European Quality Award.

However, questions can be raised concerning TQM's innovative capability. Its machine bureaucracy configuration, continually improving the processes already mastered well by reducing variation, may hinder innovation. March has shown that increased variation and lower average performance may be desirable to get a winning position in highly competitive situations.

Employee involvement in improving work-processes may be two-edged: It has proven to give productive results, not least because of the enthusiastic implementation support the changes are likely to get from the involved employees. But what happens if there is a drastic change in the environment (new technology, substitute product arriving on the marketplace, dramatic increase in prices of purchased goods)? The very feeling of process ownership by the employees, may obstruct a radical redesign of processes. Also Argyris argues that the TQM focus on employee satisfaction may be counterproductive; hindering the identification of fundamental problems and their

solutions (Argyris 1994). Perrow refers to numerous studies that indicate that the common belief that happy employees increase productivity needs to be refined (Perrow 1986: 86-88).

Recent editions of TQM, such as The European Model for Quality Management, incorporates elements of strategic considerations and planning. For organisations threatened by rapid changes in the environment, radical changes of direction and operations may become necessary to ensure survival. To this end, a flexible, more contingent management approach must replace or supplement traditional TQM. Large companies, such as Det Norske Veritas, may need to differentiate the strategic management approach so that units in product/market segments that are most exposed to threat may use other means than those prescribed by traditional TQM.

TQM in Deming's sense does not provide the metrics needed for managing intangible assets. However, a planning and reporting system in line with The European Model for Quality Management (see Figure 3) can be developed to contain appropriate indicators of intangible assets such as: explorative learning capability, customer satisfaction, customer relations, customer loyalty, employee satisfaction, process improvement.

This study finds that traditional TQM may have shortcomings for managing knowledge intensive organisations that operate in highly competitive environments. Their success depends on their ability to innovate and change quickly to win new markets early.

end of text

7. Literature

APPENDIX 1:

The following is excerpted from Chapter 2 of Out of the Crisis by W. Edwards Deming (Deming 1986). 14 Points for Management

http://www-caes.mit.edu/products/deming/home.html

Other Deming resources:

The W. Edwards Deming Institute (WEDI) http://www.deming.org/

The W. Edwards Deming Institute is a non-profit organization founded by W. Edwards Deming in 1993. Its aim is to focus on learning and sharing ideas and foster understanding of The Deming System of Profound Knowledge to advance commerce, prosperity, and peace.

Oslo, 5 May 1998; Henrik Giæver (e-mail)