Research Perspectives: Paper Summaries
Market Pressures and Institutional Forces: The Early Years of the Quality Movement
Robert E. Cole, Haas School of Business, University of California, Berkeley
The author concentrates on the early years of the U.S. quality movement, roughly 1980–1985, a period in which the Japanese had shown great improvement in product quality across a broad range of manufacturing industries, but the Americans had yet to embrace quality improvement as a corporate objective. The Japanese market-based approach to quality focused on enhancing customer satisfaction, eliminating waste, and promoting efficiency. The Japanese methods proved effective, and they contributed to larger shares for Japanese companies in foreign markets. Despite the strong economic pressures, a broad range of institutional factors made American companies reluctant to abandon existing practices and learn from the Japanese. There was, in particular, a significant cognitive gap: the idea that costs could be lowered and quality could be raised simultaneously challenged the dominant American viewpoint. It was not until the late 1980s and early 1990s that American companies were able in any significant way to integrate adapted versions of quality practices into their own structures.
Achieving Organizational Quality: An Empirical Investigation of Quality Culture, Processes, and Outcomes
Kim Cameron, Marriott School of Management, Brigham Young University
Carole Barnett, University of New Hampshire
Although an extensive literature exists on quality techniques and procedures, the relationship between quality at the organizational level and actual organizational performance is still poorly understood. The literature on quality has focused almost exclusively on the process of producing goods or services, the nature of the goods or services themselves, the expectations and levels of satisfaction of customers, and the problems
with implementing quality programs in organizations. Although attention to quality and increased productivity, market share, and profitability seem to be linked, no real empirical evidence has established the basis for this claim. More important, it is necessary to distinguish between the adoption of one or another specific practice or technique and changes in the way in which problems are defined and solutions sought. In short, it is necessary to determine the extent to which an organization has embraced a quality "culture."
The authors attempt to conceptualize and quantify quality culture, a subset of overall organizational culture. Their model outlines three types of quality cultures: error detection, error prevention, and perpetual creative quality; these types represent a progression from a less to a more advanced culture. The model was empirically evaluated in a study of 68 strategic business units over 3 years. Business officials rated various scenarios, which typified each of the quality cultures, to characterize their own organization. The approach focuses on three major issues: (1) identifying the major predictors of quality culture differences; (2) examining the relationships between quality culture and organizational effectiveness; and (3) identifying the major predictors of changes in quality culture and organization effectiveness over time. Effectiveness measures included comparisons to industry averages, the performance of best competitor, customer expectations, and past performance. The overall findings suggest that advanced quality cultures are associated with organizational effectiveness and quality achievement.
Team-Based Incentive Pay and Worker Performance
Brent Boning, Graduate School of Industrial Administration, Carnegie Mellon University
Casey Ichniowski, Graduate School of Business, Columbia University
Kathryn Shaw, Graduate School of Industrial Administration, Carnegie Mellon University
In many manufacturing firms and increasingly in service-based firms, the output of the firm is a function of the team efforts of employees. Team-based incentive pay has been used by some firms as a means of improving worker performance, but this also introduces the "free rider" problem: as the size of a team increases, the return to individual effort
diminishes, and each employee has an incentive to depend on the efforts of other members of the team. The authors use data from production lines in the mini-mill segment of the steel industry to estimate whether team-based incentive pay does in fact raise the workers' performance levels. They examined one particular process in the steel industry that is comparable across different plants and companies and assess whether it is more productive when workers receive incentive pay. They also propose a theoretical framework that emphasizes the importance of positive peer pressure. They find that incentive pay works better when workers pay attention to each other's efforts and informally penalize those who do not perform to a certain standard, thus eliminating the "free rider" problem.
Agency Problems in Process Improvement Efforts
Nelson P. Repenning, Sloan School of Management, Massachusetts Institute of Technology
The author studies the problem faced by a firm that tries to induce its workers to reveal information leading to productivity improvements that may in turn lead to layoffs or "downsizing." He investigates the effect of different contractual and institutional assumptions on a firm's ability to implement quality programs. He proposes two hypotheses: (1) employees' ability to participate in binding side agreements—to write contracts with each other or to join a union—is a critical determinant of a firm's costs of implementing new programs; and (2) the program's perceived effect on the firm's survival strongly influences the firm's cost and the ability of employees to cooperate profitably. A firm's financial health has a large effect on its ability to successfully implement a total quality management (TQM) program: TQM is easiest to implement when a firm is either growing very quickly and can absorb the excess capacity generated by productivity improvements (providing job security) or when is it doing very poorly and may be forced to shut down. For a firm between the two extreme points, the timing and pace of improvement efforts are important. By matching the rate of productivity improvement with the rates of natural attrition and growth of demand, a firm can credibly be committed to job security and not face a decision to lay people off.
Misperceptions of Feedback and Self-Confirming Attributions in the Dynamics of Process Improvement Programs
Nelson P. Repenning, Sloan School of Management, Massachusetts Institute of Technology
John D. Sterman, Sloan School of Management, Massachusetts Institute of Technology
Although TQM and other process-oriented improvement techniques have increased the effectiveness and productivity of many organizations in the short run, they often fail in the longer run. Existing theory appears inadequate in explaining many of these failures, in part because operations research and management science focus on the technical aspects of process improvement and organization theorists focus on the behavioral issues. The authors develop a model that integrates the basic technical structure of process improvement with established theories on human cognition, learning, and organizational behavior to explain the dynamics of process improvement efforts. Through the development of stock-flow and feedback diagrams, they develop a representation of both the technical and organizational structures of improvement. They find that failures result from errors of attribution by managers in assessing the causes of inefficient processes or low outputs: if managers attribute low performance to the attitudes and disposition of the workers, they react in a manner that makes such an attribution self-fulfilling. Improvement is difficult since a large portion of the available resources are dedicated to correction efforts, and production pressure prevents the experimentation and adaptation needed for improvement. The authors' theory is grounded and validated in detailed, intensive field research of improvement programs at a large global manufacturing firm.
Research on Control and Experimentation in Total Quality Management: Theoretical and Applied Implications
Sim B. Sitkin, Fuqua School of Business, Duke University
Kathleen M. Sutcliffe, School of Business Administration, University of Michigan
Larry D. Browning, College of Communication, University of Texas
The singular emphasis on control that has characterized traditional approaches to TQM implementation is not well suited to conditions of high task uncertainty. A broader, more theory-driven perspective on TQM is needed in order to distinguish control from learning goals and to address limitations in the way TQM has been conceptualized and put into practice. Sitkin, Sutcliffe, and Schroeder (1994) examine the principles and practices that underlie TQM and offer a basis for predicting the conditions under which the use of different aspects of TQM should be more or less effective. Preliminary analysis based on several case studies of work groups that confront uncertain or nonroutine tasks suggest the importance of employing a contingency model that takes into account the nature of the work performed. Under conditions of higher uncertainty, learning goals that emphasize experimentation and allow for mistakes and failures are appropriate. Control models are appropriate only under conditions of routinized and stable tasks.
The authors also presented the results from an empirical study that provides support for this general contingency theory. The study was of a series of project-level quality efforts in which the task or problem was coded in terms of its level of uncertainty and the quality methods that were used. The authors found that traditional, control-oriented quality approaches were effective for routine, certain problems, and learning-oriented approaches were effective for nonroutine, uncertain problems. Mismatches between type of approach and type of problem resulted in lower levels of effectiveness.
Empirical Foundation for Examination of the Relationship Between Quality and Product Innovation Speed
Barbara B. Flynn, Department of Management, Wake Forest University
Roger G. Schroeder, Carlson School of Management, University of Minnesota
E. James Flynn, Department of Management, Wake Forest University
Susan D. Amundson, School of Business, Arizona State University
Traditional product development literature suggests that there is a tradeoff between innovation speed and product quality—that reducing time to market implies taking short-cuts in terms of attention to quality. Yet more recent quality management literature indicates that there may be a substantial amount of overlap between the practices associated with fast product innovation and those associated with quality management. It is reasonable to assume that this overlap may make it possible to improve both product innovation speed and product quality. The authors have examined eight projects, two in each of four firms in the electronics industry. On the basis of these cases, they propose a model that suggests that the majority of the practices associated with product quality are quite similar to those associated with fast product innovation. They plan to empirically test the model, using a questionnaire that will be distributed to a much larger sample of firms in several industries. Analysis of the data from this work will use causal modeling approaches.
Patterns in the Deployment of Total Quality Management: A Preliminary Analysis of Interviews with Twenty-Five Leading Companies
George S. Easton, Goizueta Business School, Emory University
Sherry L. Jarrell, Goizueta Business School, Emory University
The authors interviewed the senior quality executives of 25 companies that have implemented advanced TQM systems (as identified through