Professor Bainbridge...on law, politics, religion, culture & food

Quality Circles in Education

Gordon Smith offers up a long blog post on his plans to use quality circles in his business associations class. I’m fascinated by his plans, because I did a lot of research on quality circles and other forms of participatory management back in the late 1990s, and came away a skeptic. Go read Gordon’s post and then come back.

In my article, Privately Ordered Participatory Management: An Organizational Failures Analysis, I wrote that: Quality circles commonly consist of a small group of employees (typically under 15) meeting on a regular basis on company time to discuss problems in their work environment. The emphasis is on productivity and product quality.  Members of a given quality circle usually come from a single department, in contrast to the plant or firm-wide committees typical of QWL. Employee enrollment in a quality circle program is usually voluntary, with participation by 25% of a plant’s work force being typical.

Unlike QWL’s broad jurisdictional purview, discussions within quality circles are typically limited to production processes. A common task is the design of workplace procedures. Recall that under Taylorism, management developed standardized work procedures and enforced employee compliance with them. At NUMMI (GM’s joint venture with Toyota), quality circle-like groups of workers designed their own standardized procedures. As one worker explained, “we all work out the objectively best way to do the job, and everyone does it that way.”

Quality circles are often associated with so-called Total Quality Management, perhaps the best known management fad of the last decade. In fact, however, TQM’s proponents usually draw a sharp distinction between the employee involvement aspects of TQM, which involve the use of cross-functional problem-solving teams, and quality circles. Despite their insistence on these denominational distinctions, TQM teams and quality circles are sufficiently similar to justify treating them as variations on the same theme.

Quality circles have undergone many changes in recent years. One important trend is the demise of permanent quality circles and their replacement by temporary problem-solving teams.  An important sub-set of quality circles has evolved in the direction of self-directed work teams, which are discussed below. In such cases, participation in the circle is usually mandatory and typically involves the entire work force.

…

Quality circles are quite explicitly directed at gathering input from line employees about production processes. At Otis Specialty Paper, Inc., for example, ad hoc circles are convened to solve production problems and provide input on changes in the production process.  Recall also the NUMMI example in which quality circles were used to design standardized work procedures.

Interestingly, Gordon plans to use them as a two-way flow of information.

Many features of quality circles confirm their information gathering function. Quality circle training focuses on training both individuals and groups to process and communicate information.  Once implemented, quality circles (like problem-solving QWLs) encourage employees to communicate more freely with supervisors.  In at least some firms, management also reviews meeting minutes or directly observes meetings.

Interestingly, management-initiated quality circles solve more problems and solve them faster than employee-initiated circles.  This suggests that circles are most useful when put to management-intended purposes. At the same time, however, manager-dominated quality circles are not as successful as those in which managers play a less dominant role.  This evidence supports my hypothesis: if the point of quality circles is to give management access to employee-held information, that purpose is negated when managers take over.

I will be interested to see reports from Gordon on the extent to which he led the discussion, versus simply listening. It might be interesting for Gordon to experiment in the future with allowing the QC to meet outside his presence, with someone taking detailed minutes he (and the class?) could review later.

The information gathering function is especially explicit in TQM, which places great emphasis on collecting data and solving problems.  The goal is continuous learning about and improvement of the production process.  Quality circles are an integral part of this management style. Under TQM, quality circles are sometimes divided into two types. One is the “steering arm,” which identifies the organization’s main problems. The other is the “diagnostic team,” which tries to find solutions to the identified problem. In both cases, teams are composed of those “people who can provide access to the data necessary for testing potential solutions and who are critical to implementing the solutions developed.” Once the teams are formed, three techniques are used to extract production information from the team members: Brainstorming is used to generate lists of ideas, with a strong emphasis on creative thinking. Flowcharts of the production process are drafted by team members to help eliminate inefficient steps in the process. Cause and effect diagrams are used to match problems to the probable causes.

What information asymetrically held by students is Gordon trying to extract from the class via the quality circle? Why is that information impacted in such a way that makes a quality circle the best way of extracting it? The post identifies the following benefits:

Quality circles give constructive criticism of your teaching when you can do something about it.
Quality circles let you counter students’ criticisms.
Quality circles let you explain where you’re coming from.
Quality circles help with diversity issues.
Quality circles give you a mechanism for criticizing your students.
Quality circles give students a sense of ownership of the class.

For experienced teachers, it seems unlikely that you need a quality circle to get a handle on these issues. Plus, the academic literature on quality circles doesn’t suggest any reason to think they would be particularly adept mechanisms at resolving most of these issues.

As my hypothesis predicts, however, quality circles do not devolve real decisionmaking power to employees. To the contrary, the managerial literature on quality circles strongly emphasizes the need for management to retain ultimate decisionmaking control.  After discussing a problem, a quality circle may make recommendations to management, but typically has no power to direct that the recommendation be implemented.  In contrast, management retains a considerable degree of control over the quality circle itself; management readily can punish quality circles that tackle the wrong problems or make poor recommendations.  As one critic of quality circles observed, they provide “an ideal structure for controlling decision making while management’s power to implement decisions is maintained.”

…

Employees who work together acquire at low cost information about one another, which is not readily available to managers.  In cases in which this information asymmetry is especially pronounced, traditional Taylorist monitoring structures may be rendered wholly ineffectual. If such employees can be motivated to monitor one another, however, peer pressure may substitute for managerial monitoring. Some forms of participatory management, most noticeably self-directed work teams, are well-suited for creating substantial peer pressure.

As with self-monitoring, peer pressure takes advantage of widely-shared traits among U.S. workers. The Families and Work Institute’s survey is again instructive. Thirty percent of the respondents defined success as earning the respect or recognition of their supervisors and peers.  Such employees are likely to engage in self-monitoring, but are also vulnerable to peer pressure. Indeed, as to such employees, the peer pressure generated by operational participation programs should prove a powerful motivating force.

This prediction is borne out by Guillermo Grenier’s well-known case study of a quality circle program in which workers were given responsibility for making discipline decisions with respect to their fellow employees. “Although this responsibility was largely illusory,” as my hypothesis predicts, it proved an effective means of generating peer pressure for greater productivity.  Workers were required to provide written evaluations of their fellow team members, which were used to determine raises.  The circles also identified and openly discussed individual employees who had problems constituting “counterproductive behavior.”

Because the firm’s workers had some of the forms of decisionmaking power, while ultimate authority remained in management’s hands, the quality circles studied by Grenier closely resembled self-directed work teams. As such, Grenier’s work points towards an explanation of the ambivalent empirical data relating to quality circles. Studies of employee attitudes find that quality circles positively affect program-specific attitudes (such as the perception of influence), but also find a negative impact on general employee attitudes (such as job satisfaction and organizational commitment).  About half of the studies of the effect of quality circles on firm productivity find positive effects, but half find no effect.  Many favorable studies could not show causal links between participation and productivity.  Only one study tried to measure changes in product quality resulting from a quality circle program; it found no effect.

Although the effectiveness of quality circles remains open to question as a matter of solid empirical data, there is a growing body of anecdotal evidence suggesting that long-term quality circles generally are not successful.  Estimates of the failure rates for quality circles range from a low of one-third to a high of 60%.  Whatever the cause of these failures, a question on which we shall speculate in a moment, quality circles arguably have flopped as a management tool.  According to some observers, they are an unstable organizational structure waiting to self-destruct.

…

[Q]uality circles fail because they lack a monitoring component or put no teeth into team members’ ability to monitor one another.  In such cases, quality circles can only serve information transmission purposes. We have known at least since Taylor’s time that informational asymmetries about production processes arise precisely because workers have strong incentives to withhold information from management that might prevent shirking or otherwise be used to the workers’ disadvantage. As noted above, this was the rationale for Taylorism’s emphasis on discouraging workers from thinking about their job or its place in the production process. Quality circles may not offer sufficient incentives to overcome the strong bias against voluntary disclosure of information asymmetrically held by workers.  Quality circles therefore ought to be subject to displacement by programs that can both more effectively resolve informational asymmetries and provide enhanced monitoring mechanisms to prevent shirking, which is precisely what my analysis suggests self-directed work teams have done.

Gordon (and Eric Orts on whose work the idea is based) seem focused solely on information gathering and transmission; specifically, two-directional feedback about how the class is going. To the extent quality circles have any utility in the classroom, it would seem useful to direct them mainly at trying to get a handle on how well students are learning the material.

Count me an interested skeptic