2

THE MACROPOLITICS OF ORGANIZATIONAL CHANGE: A COMPARATIVE ANALYSIS OF THE SPREAD OF SMALL-GROUP ACTIVITIES

Robert E. Cole

Beginning in the early 1960s a number of movements emerged in a wide variety of national settings for various kinds of worker “participation” in shopfloor activities. These movements seemed to promise to go beyond simply human relations and to reverse the historic process of deskilling that had characterized developments in the first half of the twentieth century (Bendix, 1974; Braverman, 1974; Nelson, 1975; Stone, 1975; Montgomery, 1976; Edwards, 1979). Usually initiated by management groups, these post–World War II movements involve efforts to change patterns of shopfloor organization from the specialization and compartmentalization that have characterized modern factory life. Often tagged as a form of “participative management” by their proponents, these new work structures tend to afford workers more participation in the execution of decisions than in their formulation. They usually include a stress on small-group activity, stimulating worker involvement in more facets of shopfloor operations and problem solving, and a more active mobilization of worker involvement and commitment than is common in most industrial regimes. To focus the research effort, I concentrated on small-group activities rather than on those that expand the range of tasks a worker performs or those that involve indirect forms of participation, such as membership on boards of directors.

Comparative analysts of participative organizational forms typically compare the structural characteristics and the value rationality underlying these altered forms of organization (e.g., Dachler and Wilpert, 1978; Rothschild-Whitt, 1979). Their agenda for research is to specify those matters over which employees are to exercise some control, compare the mechanisms through which workers are to participate in these decisions, measure the amounts of control these forms delegate to employees, compare the different value systems underlying different programs of participation, and evaluate the relative effectiveness of various strategies in terms of productivity and job satisfaction (see Tannenbaum et al., 1974; Katzell, Yankelovich et al., 1975; Bernstein, 1976; Cummings and Molloy, 1977).

The comparative agenda here is different. The purpose of this analysis is to explain why movements to popularize these new work structures have been sustained in some countries and not in others and why, when they spread, they are maintained to the degree they are. This involves exploring the processes through which diffusion occurs and the issue of whether these movements represent faddish episodes or have the capacity for more long-lasting impact. Most generally, the questions being asked are, Why and how do organizations change? The answers to these questions are important in their own right, because the phenomenon of work-reform movements has generated considerable interest across several social science disciplines and through much of the world. Moreover, many of the considerations grow out of the issues raised in the human relations school, traditionally constituting one of the largest if not the largest body of empirical literature in organizational analysis (Perrow, 1979: 109).

An explanation of the differing outcomes of these national work-reform efforts is of theoretical import as well. Because the level of analysis in this study is cross-national and because the units of analysis are industries and national environments, the findings are directly relevant to two concerns that have drawn a great deal of attention in recent literature. The first is the impact of power and politics on organizational form. Much recent theorizing has stressed these phenomena as explanations for the course of change in organizations (i.e., Harvey and Mills, 1970; Perrow, 1970; Zald, 1970; Pfeffer, 1978; Burawoy, 1983). The bulk of theory and research, however, has focused at the level of the organization as a setting for politics—the “micropolitics” of organization. The cross-national analysis presented here, by contrast, points to the importance of politics and loci of power at the level of industries or nations, as well—the “macropolitics” of organizational change. Here, I also join those recent efforts to delineate more specifically the nature and effects of organizational environments on organizational change. These objectives parallel recent efforts by Burawoy (1983), though our research strategy and conclusions are often strikingly different.

By focusing on macropolitics, one can distinguish the forest from the trees in a way that much contemporary theorizing and empirical research fail to do. Thus, in the case of research on small-group activities, researchers will often report factors within the organization that inhibit or enhance the probability of success, with success being measured in terms of survival of the innovation or contributions to improved quality, productivity, or increased participation or job satisfaction. These critical factors may include the mode of innovation, the amount of authority given to group sponsors and leaders, the makeup of the group, its leadership, the group’s procedures and objectives, the nature of technology, how the small-group activities fit with wider organizational practices, the extent to which they threaten managerial prerogatives, and so forth. A pattern seen in other domains of organizational research is beginning to emerge in which we are left with a variety of often contradictory studies showing how under different conditions one or another variable contributes marginally to affecting outcomes.

Perrow (1979: 96–112) documented just such a pattern in his scathing critique of the human relations literature. He showed how the accumulated literature failed to demonstrate a clear link between leadership and morale, on the one hand, and productivity, on the other. The application of increasingly complex research methodologies and causal models only resulted in a loss of applicability and theoretical power. The variables became so numerous that we could hardly generalize to organizations or even to types of organizations.

By shifting from the microanalysis of a phenomenon to an examination of the macropolitics, we create the opportunity to get a grip on the broader environmental factors driving innovation. As Perrow (1979: 110) put it, “it may be, hopefully, that any theory that has the power to explain a good deal of organizational behavior will have to deal with more general variables than leadership and small-group behavior.” Perrow further suggested that in treating these macrovariables, we may assume that the specific micro variables associated with small-group behavior are randomly distributed and thus have little effect when a large number of organizations are being studied.

In keeping with this vision, I will show in this study of small-group activities the importance of understanding the position of the national business leadership toward such innovations and how this interacts with national labor-market conditions and the positions of national union leaders. The overall hypothesis is that this will produce a far better and more generalizable predictor of success, as measured in the spread of small-group activities across and within firms, than an analysis of the micropolitics of small-group activities would allow.

The Study

The analysis is limited to three countries: Japan, Sweden, and the United States, and focuses on the period 1960–80. These are countries in which the author has linguistic competence and has carried out field and archival research based on primary source materials and interviews. In the case of Sweden, in addition to the voluminous scholarly literature in Swedish on the subject of small-group activities (semiautonomous work groups), the newspaper coverage was also extensive, and this allowed a different vantage point on developments. These data sources were supplemented with interviews of key informants and recognized experts in the unions, companies, universities, the employer federation, and research institutes. Volvo and Saab personnel were particularly receptive to discussions; because they were key innovators, their cooperation was especially useful.

For the United States, in addition to the scholarly literature on direct participation on the shop- and office-floor level, the author has been an active observer of the diffusion process of quality circles. My survey of some 176 early adopters of quality circles in the United States also yielded important background information (Cole and Tachiki, 1983).

In the case of Japan, a significant scholarly and management literature exists in Japanese on the subject of quality circles. Selected company visits and interviews supplemented this literature, as did interviews with employer federation officials and union leaders. The Japanese Union of Scientists and Engineers (JUSE), a key actor in the diffusion process, was particularly forthcoming in its cooperation and specifically in arranging for a variety of meetings with key informants. In addition, my U.S. counterpart survey of 302 early adopters of quality circles yielded important background data.

These three nations span a range of variation favorable to the comparative task. The United States is an example of limited change relative to the other two. Despite considerable advocacy of work reforms in academic and government circles, actual changes in the United States were rather limited through the 1970s. Of the other two nations, both of which have experienced significant although varying degrees of institutionalization of small-group activities, Sweden may be characterized as social democratic and Japan as liberal capitalist. These different political systems encapsulate considerable variation in the dimensions: group interest, labor-market conditions, managerial problems, political organization, and institutional mobilization.

Introduction of Small-Group Activities: Three National Experiences

The United States. In the early 1970s there were extensive discussions in the United States in academic, government, and foundation circles about the need to “humanize work” and raise the quality of worklife for employees. The proposed programs for change carried various labels, such as job enrichment, job enlargement, job redesign, and job humanization. By increasing employee participation in workplace decisions, increasing job variety, and making more effective use of worker potential, it was argued, not only would the quality of worklife be enhanced but organizational efficiency and worker productivity would be improved. The central argument made by advocates was that new work structures should be designed to allow employees to control aspects of their work that most directly affected their everyday lives. Small-group activities were central to that vision.

Efforts in the 1960s and 1970s to institutionalize these new work structures met with limited success. Generally speaking, most job redesign in the 1970s took place in new plants—often nonunion—with very little being accomplished in established ones (Davis and Cherns, 1975). Even in reputedly successful experiments, the innovation did not diffuse widely to other plants within the company (Walton, 1978). Widely publicized programs were often discontinued or scaled down a few years after their initiation, with little fanfare (e.g., quality control circles at Lockheed Air Missile). With few exceptions, American union leaders have been distinctly cool to proposed innovations affecting the boundaries and content of established jobs (e.g., Industrial Union Department, 1978). The movement received consistent support only from a scattered group of concerned academics.

Major institutional actors—management, unions, government—displayed little interest in introducing new work structures. The result through the 1970s was isolated, piecemeal experimentation. To be sure, not all such activities are widely reported nor easily counted. Some companies, particularly in newer industries such as electronics (e.g., Hewlett-Packard, DEC), prided themselves on developing a more participative management style than in the old-line manufacturing industries, without the adoption of specific programs. It must also be recognized that one’s assessment of the degree of diffusion in the United States may reflect more one’s initial value premises and expectations than it does any empirical evidence. Notwithstanding, as a general statement, it seems reasonable to conclude that relative to the changes taking place in the two other countries under consideration, movement toward adoption of small-group activities was slow in the United States.

The growing internationalization of the U.S. economy, and particularly the powerful competitive threat posed by the Japanese in recent years, has forced U.S. management and labor to reexamine small-group activities. Although data from a 1981 national survey of firms show high rates of adoption and trial on the part of individual companies, there is far less evidence of institutionalization (New York Stock Exchange, 1982). The same survey shows that quality circles had the highest rate of growth of human-resource innovations and had become synonymous for American managers with the quality-of-worklife movement. Yet, at the same time, there is ample evidence of strong management resistance to accepting quality circles at all levels of management (Cole and Tachiki, 1983).

Japan. The Japanese movement began in the early 1960s and accelerated after the middle of that decade. To be sure, there were prewar cultural models as manifested in the use of small work groups for solving productivity problems. But overall, strong authoritarianism and hierarchy were the central prewar legacies.

The key organizing principle for the postwar innovations is known as “small group-ism” (shōshūdanshugi). Management focused on the decentralization of decision making. Employees were to assume increasing responsibilities for a variety of everyday functions—most notably maintenance, safety, and quality control—that previously had been the province of management representatives. The vehicle for this increased responsibility was small-group organization. By the late 1970s various Japanese surveys estimate that some 50 percent of Japanese firms with more than thirty employees were practicing some form of decentralized decision making based on small-group activities (Cole, 1979: 135). Quality circles were among the most popular forms, with 1.7 million registered members accounted for by 200,000 registered circles; the number of unregistered circles is conservatively estimated to be four times that number.

Quality circles typically meet on the average of one hour a week and are relatively autonomous units composed of a small group of production or clerical workers from each workplace. The workers are taught elementary statistical techniques and modes of problem solving and are guided by their leaders in selecting and solving job-related quality problems and improving methods of production. Although workers participate in decision making through these kinds of small-group activities, it is clear that this approach does not threaten the hierarchical structure of management authority. First-line supervisors have not been significantly threatened by the reforms, in part because they typically serve as circle leaders.

Sweden. Swedish efforts to develop new work structures began in the late 1960s, with an emphasis on autonomous self-steering work groups (självstyrande grupper) as the basic unit of production organization. These ideas spread rapidly, accompanied by a long and intensive public debate on the need for democratized industrial organization in Swedish society. Though the diffusion of such practices has clearly been more limited than in Japan, the actual extent to which these practices have been institutionalized is difficult to specify. This is, in part, because there has not been an extensive system of formal registration and surveys as in Japan. The ideal of the autonomous work group, with workers making all their own decisions regarding work allocation, recruitment, planning, budgeting, and purchasing (Agervold, 1975: 46–65) is rare in actual practice. Yet, modified versions of this decentralized system of work organization appear in most large Swedish firms. One contrast with the Japanese case is that early Swedish efforts—in part because of the political setting of social democracy—explicitly challenged the hierarchy of managerial authority, particularly that of first-line supervisors. Consequently, middle managerial opposition was more common than in Japan, leading to a less complete diffusion of small-group activities.

General Observations on the Cases

The contrast of the experiences of the United States with those of Japan and Sweden is striking. In each of the latter countries over the past two decades small-group activities in one form or another have been adopted and sustained to a considerable degree. Despite wide differences in culture, economy, and politics, the Japanese and Swedish initiatives all have emphasized the small group as the core unit of workshop organization and participation in these small groups as a source of work motivation. In comparing the Swedish and Japanese experiences, one could say that the Swedes tried more and accomplished less while the Japanese tried less and accomplished more. By contrast, the Americans tried still less and accomplished very little.

The dependent variable—the spread and institutionalization of small-group activities in the three nations—makes it clear that reporting on the three national experiences is like comparing apples and oranges. In practice, the availability of data and the nature of the three national experiences led us to compare the evolution of quality circles in the United States and Japan and semiautonomous work groups in Sweden. Quality circles are “off-line” activities, that is, workers engage in these problem-solving teams separate from their “normal work” activities. Semi-autonomous groups, however, are designed to give teams of workers more control over the work process itself.

In Japan, the language used to describe the key innovations was decentralization of responsibility, and the term participation seldom appeared. In the United States the debate was dominated by the terms participation in management, QWL (Quality of Worklife), and EI (Employee Involvement). In Sweden the notions of joint influence and democratization of the workplace punctuated the early debate.

This different language is not a random phenomenon. It tells us a great deal about objectives and who were the central actors controlling the debate in each country. In the case of Japan, focusing the debate on decentralization of responsibility tells us that management was in charge; it could impose its own categories and labels on developments. In Sweden the focus on industrial democracy tells us, first, that management did not have full control of the agenda of solutions and definition of problems. Moreover, the Swedes had a highly centralized labor-management decision-making system. This meant that advocates of small-group activities within labor, management, and academic circles could argue forcefully and believably that there was something missing at the shop- and office-floor level in terms of democratic decision making and that semi-autonomous work groups could fill this vacuum. In the United States, unlike many western European countries, the labor movement had an active shopfloor presence. This preempted to a large extent the industrial democracy theme at the shop floor level; instead, there is a rather more modest focus on participation, leadership, and employee involvement.

It would be an artificial and sterile exercise to try to winnow down these different approaches to small-group activities to the lowest common denominator so as to assure an identical dependent variable across countries. Rather, the tack taken here is to acknowledge these differences and take them into account in explaining the extent to which small-group activities diffused in the three countries and how and why they took the different forms they did.

Although management has been the primary initiator of these new work structures, there is another driving force. Underlying the initiative in Sweden, and to a far lesser extent in the United States, is a focus on industrial democracy. Fuller participation in the decisions of the firm and over one’s daily activities is seen as contributing to the democratization of the firm. Burawoy (1983: 602–603) ignored this theme, though many unions, most notably in Sweden, stressed industrial democracy as a basis for cooperating in the initiation and shaping of these new activities. Even in the United States union supporters have heavily stressed the theme of industrial democracy (Bluestone, 1978: 21–24). How this additional driving force interacts, complements, or opposes management initiatives is a matter for empirical examination in each nation, industry, and firm.

Managerial Interests and the Incentives to Innovate

Improving the quality of working life, reducing worker alienation, and enhancing self-actualization are the benefits of small-group activities most often stressed by academic proponents. Industrial administrators, however, are less motivated by these sorts of concerns, and it is they who must introduce and institutionalize these innovations. Management is concerned more with performance and economic payoff. To understand the diffusion of small-group activities, we must look more closely at the forces driving management to change existing practices; we must look for something other than a liberal reforming impulse.

Sweden and Japan. Extensive interviews with Japanese and Swedish managers and an examination of the relevant literature in these countries make clear that the driving force for innovation was quite similar in both cases. Both countries have faced severe labor-shortage problems over the last two decades.

In Japan, industrial employers confronted an increasingly tight labor market in the late 1960s and early 1970s. It became more difficult for the major manufacturing firms to recruit and retain those select employees they desired (Minami, 1973). Compared with the United States, labor turnover in Japanese industry was still quite low, but it occurred in the context of a serious shortage of recent male graduates entering the labor force, especially middle-school graduates. Replacing existing workers was thus both difficult and costly because there was no pool of workers willing to take the most disagreeable jobs in the manufacturing sector. Political and economic leaders considered and later rejected Southeast Asian immigration as a desirable option.

Rapidly rising educational levels in Japan, further, led to an increasing proportion of labor-force entrants who were unwilling to accept the least demanding jobs. The educational system was producing an increasing number of high school graduates who had come to expect white-collar jobs commensurate with their educational levels. Instead, an increasing number were being assigned to blue-collar jobs.

By the late 1960s and early 1970s the labor shortage was most serious for those manufacturing sectors that had the most standardized and routinized jobs and were characterized by hard physical labor under difficult conditions. Surveys reported extremely high turnover among new recruits to these industries (Ministry of Labour, 1974: 72). These circumstances constituted an important motivational force for Japanese managements to search for solutions to minimize their problems. Small-group activities appeared to be one possible strategy to make firms more attractive to well-educated potential recruits, thereby reducing turnover and labor unrest and contributing to productivity and quality goals. There was, however, little assurance of the outcome, and innovators operated under great ambiguity and uncertainty. Furthermore, other strategies were pursued simultaneously, such as locating labor-intensive production facilities offshore, in Southeast Asia. Firms in industries that suffered most from recruitment and retention problems—notably automobiles and steel—were some of the most active in early efforts at innovation. All this occurred in a context of growth in internationalization of the economy and resultant concerns that managers must make better use of their human resources if they were to survive in world competition.

The labor situation facing Swedish managers was similar but even more serious than in Japan. In Sweden a severe labor shortage had developed by the mid-1960s, educational levels had also risen rapidly, and labor-force entrants were increasingly unwilling to accept routine or arduous jobs. Swedish employers sought initially to alleviate these problems by acting to expand the labor force. They began more actively and quite successfully to recruit economically inactive women. A second strategy was to increase the employment of migrant aliens. The number of aliens taking jobs in Sweden increased steadily to 20 percent of total employment in the mining and manufacturing sectors (Statistiska Centralbyrån, 1978: 195).

By the late 1960s Swedish manufacturers became increasingly disenchanted with these strategies. It was still difficult for them to find enough workers to fill the least desirable jobs. Moreover, they were having to rely on what they considered to be labor of the lowest quality, and the large influx into Sweden of foreigners, concentrated in the least desirable jobs, came to be seen as a threatening social problem. Absenteeism and turnover, further, swelled to unprecedented proportions. Annual employee turnover reached 50 percent during 1970 in a number of metropolitan industries—notably auto manufacturing. Sweden’s legislated unemployment and sick-pay benefits had made turnover and absenteeism relatively costless to employees.

All of these factors spurred management to consider strategies for restructuring work to bring native male Swedish workers back into the factories and cut turnover and absenteeism to manageable proportions. As in Japan, major innovations in work structures took place in precisely those industries suffering the most severe recruitment, turnover, and absenteeism problems.¹ Volvo Corporation, the largest Swedish private firm, experienced some of the most severe problems and later achieved fame as an innovator in creating new work structures. Multiskill training and teaching employees a fuller range of job tasks were part of the approach to small-group activities in Sweden. These practices facilitated a workfloor adjustment to high rates of absenteeism and turnover.

The United States. It is instructive to contrast the United States with Japan and Sweden. A large pool of unemployed labor exists in the United States to fill the most disagreeable jobs. Despite high rates, by world standards, of labor turnover, replacements have been readily available through this pool. This constituted a major barrier to the diffusion of small-group activities in the United States in the 1970s—there were no obvious economic incentives to interest managers in innovation. Until Japanese competition became impossible to ignore in the late 1970s, no pressing managerial problems triggered searches for new methods (Wool, 1973: 38–44).

The differences between the United States and Sweden and Japan are most striking in the auto industries during the early 1970s. Swedish auto officials found that some U.S. auto plants had turnover figures similar to those of Swedish auto firms—about 50 percent per year. Surprised that most American managers did not perceive this as a serious problem, they concluded that U.S. managers were accustomed to greater job mobility of workers and reasoned that they could train people for simple jobs so quickly that the turnover figures did not matter (Gyllenhammar, 1977: 7–9). Indeed, the assumption that large quantities of unskilled labor would circulate annually through the firm reinforced the strategy of using industrial engineers to increasingly simplify job content.

Although the original General Motors–United Automobile Workers (UAW) agreement on QWL was signed in 1973, little occurred until the late 1970s, despite selected “showcase” plants. Innovation that did occur was often piecemeal, rather than through the more systemic approach developed in Sweden (Tichy and Nisberg, 1976). Although Kanter (1983:221–347) documented high-level management concerns with participative management as a solution to problems of turnover and absenteeism within GM in the early 1970s, it is striking how long such discussions were confined to the corporate level.

It was not until the Japanese threat of the late 1970s and early 1980s that U.S. auto manufacturers began broadly to operationalize solutions addressing the costs of high rates of absenteeism and turnover. They came to the conclusion that they were suffering a competitive disadvantage because they failed to mobilize their human resources through a participative strategy. Data from our 1981 survey of the universe of early adopters of quality circles reveal that the greatest number of adopters were in automotive parts and manufacturing and the electronics industries (Cole and Tachiki, 1983). Early adopters tended to be in manufacturing sectors characterized by highly labor-intensive industries, industries in which quality has been a high priority in the past, and those in which quality and costs have received new emphasis as a result of Japanese competition. Auto and electronics both partially fit this characterization, although they differ in degrees of unionization, rate of growth, and the like.

Ideology and Organization: Mobilizing for Structural Change

If our preceding descriptions are accurate, we would expect the relative interests of major institutional actors to be reflected in the emergent national infrastructures for diffusing small-group activities. Specifically, one would predict that when at least one of the major parties to the labor market—especially management, given its leadership position in market economies—becomes strongly committed to small-group activities as a solution, a well-organized national infrastructure will be created. Through definition and publicizing of “best practices” and providing legitimacy, such organizations provide the national leadership critical to the emergent movement as it develops at the plant level.

In the cases of Japan and Sweden, the major incentive for management to develop this commitment to innovate was an evolving set of managerial problems relating to labor shortage that made earlier strategies ineffective. Motivation, however, is a necessary but insufficient condition for change. Managerial groups have to mobilize to implement these innovations. This has involved, in each case, the articulation of an ideology that justifies these changes and serves to persuade other members of the industrial community either to support or not to resist them.² It also involves the creation of an organized infractructure of communication and organization with which the resources of the innovating groups can be mobilized and changes introduced and institutionalized in firms. The emergence of a national infrastructure reflects the consensus of top management that it has identified the direction in which it would like to move; its leaders articulate the ideology that legitimates the new movement. It is this interplay of ideology and organization that is at the core of the rapid spread of new organizational practices. Without such an interplay, the emergent practices are more likely to assume the status of fad.

The United States: Late and Weak Managerial Mobilization

American managers had little incentive to reform existing work structures in the 1960s and 1970s. As a result, organization and mobilization for the diffusion of participative work structures was left to mavericks, who tried to convince managements that such innovations were in their interest. Most of these advocates were from loose networks of academics, management experts, and, occasionally, business and union leaders. On this basis we would expect that the organized efforts to spread small-group activities would differ markedly from the same processes in Japan and Sweden.

Failing strong interest by the national business leadership, “outside” organizations like the Ford Foundation sought to stimulate institution building in the 1970s. However, achievements have been quite modest relative to what occurred in Sweden and Japan. Government-encouraged initiatives proved to be of modest value because of sharp fluctuations in the level of government support and because the government could not or would not provide management the long-term incentives that were necessary.

With the upsurge of experimentation in the United States in the 1980s, there has been a notable increase in the activities of various research and training organizations focusing on small-group activities, in particular, the American Society for Training and Development, the American Management Association, and the OD Network (Organizational Development Network). These kinds of organizations are not likely to create a national infrastructure, but they do supply much of the raw material for supporting its emergence.

Although there were a number of possible contenders for leadership of the small-group activity movement in the United States, none individually or collectively have come to play the same extensive role as key organizations did in Sweden and Japan. Existing national organizations reflecting top-management interests, like the National Association of Manufacturers, the Chamber of Commerce, and the Business Roundtable, saw the locus of their activities as Washington, with lobbying as the chosen method to achieve their ends.

Other organizations interested in the participation movement, such as the Work in America Institute, focused on improving labor-management relationships in the unionized sector, primarily through labor-management committees. This calls attention to a major problem in building an integrated national infrastructure in the United States. Those who focus on working with the unionized sector ignore 80 percent of the labor force. Those who include the nonunionized sector invite the suspicions of the unions that the real agenda is union busting. The failure of big business to accept unions as a legitimate force in the United States has worked against building a national infrastructure for small-group activities and has led, rather, to a splintering of efforts.

The American Society of Quality Control (ASQC) was a different kind of contender. It is a voluntary professional association composed primarily of middle-management specialists in quality. ASQC members did not, however, have a great deal of status in their firms. Just as important, ASQC proved to be so tied up in knots by their bureaucratic structure of rules regulating organizational innovation and limited by their reliance on volunteer labor that they failed to provide leadership for the small-group activity movement.

The American Productivity Center, established in 1977 with top-management support from key business organizations, was a clear candidate for providing nationally recognized leadership for the small-group activity movement. For the most part, especially in its early years, however, its high-level corporate sponsors and board members saw the path to productivity improvement in terms of traditional approaches, like enhancing availability of capital at favorable terms, increasing technology investment, and improving taxation policy. Again, these individuals were more at home negotiating and lobbying for solutions in Washington than in the factories and offices of America. Union leaders on the board were just as comfortable focusing on the “big issues” associated with traditional labor-management concerns. Yet, support for small-group activities in terms of their productivity and quality payoffs requires the acceptance of a view that many small changes, in the aggregate, yield big changes.

The closest thing to a national organization designed to promote small-group activities is the International Association of Quality Circles (IAQC), begun in 1977 by two consultants; it served at least indirectly to promote their own business. That in itself is typically American. The organization officially reported some 7,000 members by the mid-1980s, and ninety chapters. It conducts a great deal of introductory training in circle activity and promotes national conferences.

While these are not insignificant achievements for a volunteer organization, IAQC continues to experience severe problems in carrying out its mission, moving from one crisis to another. The problems it has had illustrate well the differences between the U.S. experience and the Swedish and Japanese experiences. IAQC difficulties stem, above all, from an elected board of directors, composed now primarily of middle- and lower-ranking management officials (including quality-circle facilitators), most of whom have no experience running a complex national organization. Lacking national recognition, they were unable to bestow the mantle of legitimacy on the organization as the leader of the national small-group activity movement. Operationally, they have great difficulty tapping the coffers of major businesses and foundations, and financial instability has been a continuing problem. They have also had to rely heavily on volunteers, with all the problems of follow-through that this entails. Finally, consultant influence, though no longer dominant, has continued to create problems, as consultants have pushed to keep the organization out of activities where IAQC would be in competition with “their own members”—as the consultants like to phrase it. This has limited organizational initiatives. IAQC has yet to show that it will be able to legitimate and provide resources for the growth of the small-group activity movement in a way that occurred in Sweden and Japan.

Organized efforts in the United States to effect the implementation of small-group activities have emphasized careful research, measurement, and prior scientific demonstration of their effectiveness. By contrast, experience in Sweden and Japan shows that when management has a motivation to innovate, funding is forthcoming, organized infrastructures develop quickly, and efforts to institutionalize new work structures move ahead full speed with relative inattention, sometimes even self-conscious disinterest, in effectiveness research.

Japan: Early and Effective Managerial Mobilization

Japanese managers are far more likely than their American counterparts to think of solutions to managerial problems as coming from the area of human relations rather than technical relations. There are a number of historical and cultural precedents for this (Cole, 1979: 108–111). This view has been reinforced by the rapid translation and absorption of the American human relations literature in Japan. For these reasons, Japanese managers, when faced with the labor-supply problems of the late 1960s, readily turned to restructuring work arrangements and organizing workers in small-group activities designed to arouse in workers a sense of loyalty to the firm and an internalization of company goals. These activities have been carefully controlled, and involvement is not always voluntary for workers. These arrangements are premised on considerable sharing of information with workers on production schedules and plant objectives and performance (Cole, 1979: 224–50).

Not surprisingly, high-level management groups provided crucial leadership for the implementation of small-group activities. The Japan Federation of Employers’ Associations (Nikkeiren) was especially important. Nikkeiren, though analogous in function to the National Association of Manufacturers in the United States, is more specialized in labor and personnel matters and more prestigious and powerful than its American counterpart. Many of Nikkeiren’s early small-group activities can be dated from 1966. By 1970 the chairman of Nikkeiren, believing he perceived a trend toward “all-employee management,” advocated support for this trend at a top-management seminar. Since that time Nikkeiren has engaged in a wide range of publicizing activities designed to explain and spread such approaches (Nakayama, 1972).

Specialized organizations such as the Japan Union of Scientists and Engineers (JUSE) also developed departments designed exclusively to propagate specific kinds of small-group activities. JUSE is a national nonprofit organization dedicated to providing services to participating Japanese companies in the areas of quality and reliability. It is composed of university professors in engineering and science, as well as engineers from leading firms, but is closely tied to business circles. Since 1948, shortly after its founding, the chairman of JUSE has either been the chairman of Keidanren (Japanese Federation of Economic Organizations), the most powerful business association in Japan, or a former chairman. Furthermore, the intellectual leadership of JUSE has been provided to a great extent by Ishikawa Kaoru, the secretary general. His father was the first chairman of Keidanren to sit simultaneously as chairman of JUSE. These linkages provided legitimation for JUSE initiatives; JUSE assumed the major leadership role in developing and diffusing the concept and practice of quality-control circles. Having corporate sponsors, JUSE could employ professional staff and not rely on volunteers for its core administrative functions. Central to its mission was gathering company experiences, defining “best practices,” and feeding the information back to the field in its conferences and publications. Specific mechanisms included publishing a low-priced magazine for first-line supervisors and circle members, developing a plethora of training programs and publications, and establishing a quality-control-circle registration system. JUSE also created quality-control-circle conventions for companies to share circle experiences.

In summary, we see a firm linkage between the emergence of a national infrastructure designed to spread the innovation and top-management leadership and support in the private sector. It is this linkage that makes less likely the outcome of short-term faddish commitments and increases the probability that institutionalized firm-level structures will develop.

Sweden: Early Joint Management-Union Initiatives Followed by Weakened Commitment

In Sweden management was attracted by ideas that originated at the Tavistock Institute in England (Emery and Trist, 1969) and were propagated in Sweden by the Norwegian scholar Einar Thorsrud (Thorsrud and Emery, 1969). Tavistock ideas emphasized the development of the organization as a sociotechnical system, the interaction of social and technical factors in the organization of the workplace, and prescribed the development of small, cohesive work groups, which would maintain a high level of independence and autonomy. Jobs would be enriched, individual responsibility increased, learning possibilities enhanced, and, most importantly, from the standpoint of Swedish management, jobs would be more attractive to desirable recruits and labor turnover and absenteeism could be reduced.

The Swedish Employers’ Confederation (Svenska Arbetsgivareföreningen—SAF) made a high-level decision in the mid-1960s to support small-group activities on the shop floor. The Tavistock ideas provided a coherent ideological framework and a practical direction for reform efforts. SAF, committed to Thorsrud’s ideas, entrusted its program for research and popularization to its technical department, which began a large-scale training and publications program. Another organizational offshoot of SAF, the Personnel Administration Council (Personaladministrative Rådet), guided several important projects and played a key liaison role between academic research and the world of business (e.g., Björk, Hanson, and Hellberg, 1973).

In Sweden, unlike Japan and the United States, the strength and the orientation of organized labor were such that these management initiatives for the changes in the workplace had to be taken in cooperation with labor unions. In 1966 SAF, together with the white-collar (TCO) and blue-collar (LO) union federations, established the Development Council for Cooperative Questions (Utvecklingsrådet for Samarbetsfrågor) to facilitate labor-management cooperation.

The Development Council set up a special group for research, called URAF. Its general task was to make clear which factors stimulated workplace democratization and which factors hindered it. Its specific purpose was to initiate and supervise projects and conduct research. These efforts were anchored in development groups at the company level in which both management and labor were represented. URAF sponsored ten key pilot projects during 1969–1970. These pilot projects were well reported, usually in a favorable light, by both labor and management groups (e.g., Landsorganisation, 1976), and the reports served to stimulate further initiatives.

However important URAF was in the crucial early periods of the reform movement, this kind of cooperative union-management activity fell off markedly by the early 1970s, for two reasons. First, the unions began to articulate a competing ideology of worker control. LO wanted not only sociotechnical participation in an atmosphere of cooperation but more power for workers and the unions in the firm. Second, the employers’ confederation, SAF, felt shackled by cooperative activity, especially with the increased militance of the unions, and preferred to proceed on its own with a freer hand (as its counterparts in Japan were able to do).

As LO turned increasingly toward legislation to further its aim of enhancing industrial democracy, SAF shifted its main initiatives to its own technical department and away from cooperative activity with the unions. URAF was finally abolished in 1978. Similarly, SAF’s interest in small-group activities weakened with the passage of the Co-Determination Law and the threat of the “wage earners fund”;³ this shift in focus came to be reflected in the spinning off of the technical department as a separate enterprise in the early 1980s. These developments, in effect, resulted in a weakening of the national infrastructure and thereby contributed to a loss of momentum in the movement itself.

Interests and Orientation of Organized Labor

The development of widespread managerial interest in innovation and the creation and use of organized infrastructures for national communication and mobilization do not uniformly assure successful diffusion. Other loci of interests, organization, and power must be entered into the political calculus. In this case, the strength, interests, and orientation of labor are crucial in determining the shape and outcome of the diffusion process. In each of the three nations labor has exhibited a different orientation toward shopfloor reforms, and this has served to shape the outcomes of the diffusion process in different ways.

The United States: Union Suspicion

With the notable exception of the UAW and later the Communications Workers of America, advocates of new work structures in the United States generally have failed to gain strong union-wide support for their programs. The major spur to across-the-board activity came in the early 1980s from the internationalization of the economy. In the auto industry there were some impressive achievements in the early 1980s with widespread establishment of employee-involvement groups and employee-participation groups, as the circles were known respectively at Ford and General Motors. Notwithstanding, in the eyes of many union officials and workers, many of these initiatives became negatively associated with concessions in bargaining. General UAW support for the QWL programs showed signs of evaporating by the mid-1980s (cf. Thomas, 1984). Deep division within the top leadership of the union as to the proper position toward QWL made forward movement difficult.

The leaders of most other major unions have generally been even more suspicious of management motives, seeing small-group activities as a management scheme to get more productivity from workers without sharing the increased economic rewards (Industrial Union Department, 1978). Another concern is that these approaches often break down established job boundaries, thereby resulting in “speedups” and reducing manpower requirements. They are also seen as a threat to existing collective-bargaining structures (Sorge, 1984). Finally, many union leaders, not incorrectly, see small-group activities as a device to avoid or weaken unionization by building loyalty to management. The result has been active union distrust of such reforms.

This union orientation arises from specific features of American industrialization and of the tradition of unionism that arose during the period, as a subsequent comparison with Japan will highlight. One of the most notable features of American unionism historically has been its orientation toward control of job opportunities (Perlman, 1968); unionization took place often under conditions of labor shortage, and American workers failed to assert themselves collectively as an organized force in national politics. It was decisively shaped, further, by the struggle between craft unions and management over control of the production process and the forms of remuneration involved, as management moved to reorganize the production process around the turn of the twentieth century (Stone, 1975; Montgomery, 1976). Behind the movement to reorganize production known as Taylorism (Braverman, 1974; Nelson, 1975) was a managerial strategy prompted by comparatively high wages in American industry, by contemporary world standards—wage levels that were the product of a labor shortage in industrializing America and relatively strong craft unions in older industries. To compete in international markets, U.S. managers adopted a high-wage, low-cost strategy. Since labor was relatively expensive, there were strong incentives for American managers to economize on its use and cut unit costs—thus the aggressive strategy of substituting machinery for labor and subdividing, analyzing, and reorganizing the labor process.

In the face of these managerial initiatives, unions were faced with two options. First, they could continue to struggle for the maintenance of craft control over the production process. Second, they could accept the shifting overall trend of power in industry and push for quantitative improvements in worker rewards within the emerging new patterns of labor relations—in effect, a rearguard action against scientific management and Taylorism. The first option was rejected in the face of management and government power arrayed against the unions. Instead, the unions generally adopted the more limited second strategy, where collective bargaining came to legitimate an increasingly extreme division of labor (Fox, 1974: 201, 204–205): In order for unions to control job opportunities, standardized rules of wage determination and job allocation had to be developed where none existed before. To negotiate and conclude labor contracts on this basis assumes the determination of pay rates on the basis of explicitly defined jobs. In pursuing this strategy, it became in the union’s interest to set clear, precise job boundaries and negotiate pay rates for each job. These practices became institutionalized in American work structures as industrialization proceeded, the scale of enterprises grew, and internal labor markets assumed increasing importance. There evolved a complex set of industrial relations rules in which the equity principle—predicated on the ability to identify and compare sets of clearly defined jobs—became a central basis for wage determination. In this context contemporary American unions are suspicious of small-group activities because they often threaten vested union interests in existing job structures, pay scales, and collective-bargaining arrangements.

Japan: Union Indifference and Uninvolvement

Japan’s industrialization process gave rise to different forms of unionism and to markedly different union strategies than those that arose in the United States. First, Japan experienced labor surplus rather than labor shortage throughout much of its industrialization, and wages were quite low by international standards. Second, union organizations were suppressed or severely restricted in their activities in the pre-World War II period and again in the 1950s (Ayusawa, 1966: 302–304; Cook, 1966: 97–98). Japan’s history of industrial development was such that the Japanese skipped the craft stage of union organization (Dore, 1973), precisely the stage at which the model of precise job specification was most carefully worked out in the American experience. Under these conditions, Japanese industries could compete in foreign markets, because labor was cheap. Unions, relatively weak and facing a large pool of potential replacement workers, found job security a more viable goal than control of job opportunities. Given these conditions, when scientific management principles were adopted in Japan, time-and-motion studies were used not to determine jobs but simply to determine “correct” job procedures. Managers were interested in developing a disciplined workforce motivated to take on cost reduction and other management goals as a personal challenge. This desire, in interaction with union interests in job security, led to the institutionalization of “paternalism”—the emphasis on principles of permanent employment and promotion and pay according to a seniority system.

Japanese management’s control of worker training under the seniority wage system, combined with the lack of a national labor movement in the course of industrialization, led to an acceptance of management’s prerogative to decide unilaterally on work assignments, job demarcation, and the restructuring of job boundaries, along with changes in technology. Even today, management generally has a discretionary right to move its workforce about without union interference (Okamoto, 1975). In this context, neither management nor labor saw great benefit in tying wage determination to specific job performance. The interplay of union and management interests that led to sharp job demarcation and union commitment to these work structures in the United States led to the opposite outcome in Japan.

This different historical pattern led to Japanese labor’s relative indifference to the introduction of small-group activities. Its agenda for participation focused instead on cooperation with management to set up an elaborate system of labor-management consultation committees at all levels of the firm. Because small-group activities are within the historically agreed-upon realm of managerial prerogatives and because they are consistent with union goals of creating greater economic prosperity, Japanese unions have maintained their relative uninvolvement in these changes.

Sweden: From Cooperation to Contention

Swedish trade unions, unlike those of Japan and the United States, took an active role in organized efforts to establish small-group activities in industry. The nature and effect of this involvement, however, have been quite complex and have had a mixed impact on the diffusion process. As an organized national force, Swedish unions are extremely potent, with organization rates of about 80 percent of the entire workforce (with almost similar rates of organization in blue- and white-collar sectors). These levels are far higher than the American and Japanese rates. Nationally, moreover, the organization of the Swedish trade-union movement is highly centralized, and through forty-four years of Social Democratic government that ended only in the 1976 elections, organized labor has been a much stronger force in national politics than its counterparts in the United States and Japan (Korpi, 1978: 74–75). The Social Democrats returned to power in 1982.

Historically the Swedish union movement has been hostile to the idea of direct workshop participation, not unlike most American unions. This orientation was not due to a vested union interest in existing job definitions, but rather to the unions’ commitment to centralized national bargaining. The unions feared that increasing the decision rights of small groups in the shop would undermine this.

The union strategy for expanding workplace power in the mid-1960s focused on company-level works councils. A series of threats to the organizational integrity and discipline of the unions, beginning around 1970, changed the union leadership’s orientations and moved them to support small-group activities as part of a strategy too democratize Swedish firms. In late 1969 a wildcat strike took place at the state-owned LKAB iron mine. The participants were well-paid miners who were protesting inept and unresponsive union and state officials as much as poor working conditions and lack of wage equity (Korpi, 1970; Dahlström et al., 1971; Hammarström, 1975). The LKAB strike, followed by a wave of wildcat strikes throughout Sweden during 1970 and 1971 (Berglind and Rundblad, 1975), set in motion a national debate on the meaning of work, the proper trade-union role, what employees had a right to expect from their work, and the possibility of restructuring work and authority relationships in the workplace. Media coverage of these issues was intense (Karlsson, 1969).

It was during this period that Thorsrud’s ideas about autonomous work groups were being propagated among management circles. Direct workplace participation by workers through small-group activities appealed to top union officials as one way to accommodate the disaffected and rebellious rank and file while moving toward democratization of the firm.

This crucial early period of cooperation, however, was short-lived. The momentum of political events required changes that were beyond the scope of the initial experiments with small-group activities. LO in its 1971 convention endorsed a historic reversal of policies. Where previously the union had been indifferent or hostile to worker participation, it now committed itself fully to the worker-participation movement, demanding power for workers at all levels of the company. It demanded that this be guaranteed not merely through cooperation or bargaining with management to support small-group activities, but through national legislation. The LO policy, endorsed also by the white-collar federation, TCO, led to the passage of the Co-Determination Law.

The law represents a crucial shift in Swedish union strategy. Traditionally, Swedish trade unions were notable for their ability to agree with management on certain areas in which mutual interests could be served through cooperative action. Initially, this was the case for worker participation, as unions cooperated with management in the setting up of works councils and later through participation in URAF with experiments stressing small-group activities. They expected this activity to lead to greater worker and union input at all levels of decision making within the firm. When they found that management resisted efforts to extend worker participation beyond small-group activities at the shop and office floor, pressure grew for the shift in union policy that eventually took place in 1971. In this watershed year, the previously cooperative policy gave way to a new strategy of having the Social Democratic government enact legislation that would facilitate the democratization of the firm through union negotiation on the basis of these laws. The unions saw the Co-Determination Law as a collective-bargaining model for controlling participatory activities rather than bringing about direct participation of workers. This vision worked against the development of small-group activities, and the shift from cooperation to contention served to slow the initially rapid progress of their diffusion, since there now was a basic tension over the direction and scope of reform.

Implicitly, this shift in union position had another consequence that slowed the progress of shopfloor innovations. Small-group activities now became just one of several levels on which worker participation and union power were being pursued. In 1971 the LO congress also endorsed various representation schemes, including board of director membership for employees (passed into law in 1972) and expanded works councils, but there has been considerable union infighting over the priorities assigned these various forms of participation.

It remains to be seen whether the norms to be worked out to operationalize the Co-Determination Law will revitalize the small-group activity movement. The lack of a clear idea on how to concretize the legislation, combined with the shift to a non–Social Democratic government in 1976, led to a paralysis of the small-group activity movement. With the return of the Social Democratic government in 1982, new life seems to have been breathed into small-group activities (Mellbourn, 1984: 16). Slogans of the new movement are worker participation in “development” and “change” and “local solutions for local problems.” In the private sector, this translates into a more flexible approach on the part of the unions. In early 1986 Volvo, Sweden’s largest private employer, announced plans for rebuilding its largest plant as well as for a new plant. Both plans contained a strong emphasis on small-group activities.

The Macropolitical Dimensions of Work Reform

From the case-by-case comparison summarized in Figure 2-1, we can extract and highlight the macropolitical dimensions of work reform. To the extent that there is contention over the implementation of small-group practices—when groups have different interests with regard to change, articulate competing ideologies, and act on their respective interests—implementation becomes politicized. To the extent that different groups have congruent interests and cooperate in such efforts or to the extent that one group is indifferent to the change, the process is, in our terms, nonpoliticized. Whether the process of change becomes politicized or not has an important impact on the outcome, but this impact is not a simple one. A comparison of the three cases shows that the effect of degree of politicization is contingent on other variables at work in each national situation.

Nonpoliticized Patterns: The United States and Japan

In the United States neither unions, management, not politicians exhibited strong interest in small-group activities: unions because they would challenge vested interests in existing job structures and disturb long-standing strategies for the pursuit of their interests through collective bargaining, management because there were no acute problems with existing practices, and politicians because there was no manifest constituency. Unlike Sweden, there was no impact of the issue upon public opinion or national political discourse. With such a lack of commitment on the part of all major institutional actors, a strong national infrastructure failed to develop. The result was that no one contender or group of contenders effectively promoted the small-group activity movement in the United States.

In Japan, also, the process of change has been nonpoliticized. Unlike the United States, however, there has been a very widespread application of small-group activities. In Japan management became interested in such reforms because of difficult labor-market problems it faced in the mid-1960s and concern about utilizing an increasingly educated labor force. With the development of such a commitment, employers used existing organizations to create a national infrastructure for the diffusion of solutions such as small-group activities. Japanese unions, unlike their American counterparts, had no interest in opposing these changes in the workplace. Similarly, the ruling Conservative government had little incentive to take positions or action on this matter.⁴ Therefore, in the absence of involvement by government, unions, and political parties, the implementation of these new work structures proceeded as a well-coordinated case of managerial innovation. There were no conflicts over reforms, as in Sweden, that threatened to push them in directions feared by managers, and there was no entrenched union opposition to such changes in traditional work structures, as in the United States. As a result, small-group activities diffused rapidly and thoroughly, management being able to control carefully the direction, scope, and content of reform in a way that Swedish management found impossible.

Politicized Pattern: Sweden

In Sweden, as in Japan, strong management interest in small-group activities resulted from serious turnover and recruitment problems due to changes in the domestic labor market. Union orientation toward such reforms, however, shifted from disinterest to cooperation and then to contention in a relatively short period of time. This rapid shift in union orientation coincided with the quick politicization of the issue. The issue of democratization of industrial life came to the center of public attention, with high media coverage, and spurred the Social Democratic government into action. More importantly, it resulted in a rapid shift in union orientation and strategy, as the centralized union hierarchy was threatened with challenges from below. Initially hostile to shopfloor participation and preferring centralized control, the Swedish union confederations quickly moved toward support of shopfloor reforms and eventually came to contend with management over the direction of reform by pursuing legislation to enforce changes. This meant that Swedish managers, unlike their Japanese counterparts, were unable to control carefully the direction, scope, and content of reform after the initial period of management-union cooperation. Later moves by the unions to extend democratization to worker involvement in decisions relating to capital investment and the like threatened traditional management prerogatives and authority. As a result of this basic tension, the initially rapid diffusion of small-group activities slowed as management became wary of changes that it could not completely control. The less complete diffusion of small-group activities in Sweden, despite the consensus of all major institutional actors that such reforms were desirable, is due to this basic management-labor contention over the direction of reform. This contention was heightened by the subsequent loss of power of the Social Democrats to the conservatives in the 1976 election.

The Macropolitics of Organizational Change

The explanation of why change took place in some countries and not in others has two essential components. The first is the motivation of management to introduce small-group activities, rooted in perceived managerial problems. In two of the cases examined in the period 1960-80 (Japan and Sweden), such motivation was present and arose from the characteristics of a changing national labor market; such motivation, however, also derives from more generalized concerns as well. The interaction of these generalized concerns with specific motivators poses enormous problems for social science as we ask what kinds of data will provide meaningful explanations of management behavior. We are dealing with an environment composed of numerous problems and solutions with various degrees of saliency at different times, acting and reacting upon each other in a variety of ways over time. Although I have identified labor-market factors as critical to precipitating out small-group activities, these factors are part of a broad constellation of factors in the total social system. How do managements’ conceptions of problems and solutions actually crystallize in relation to competitive ideas? Certainly, longitudinal analysis is necessary. Questionnaire surveys alone are inadequate to the task. A variety of research tools prove useful, including participant observation, nondirective interviews that allow interviewees to explore the full social framework in which decisions are made, and surveys of management journals to see what managers are saying to each other. This suggests that a good social science answer is likely to be a work of art, as one sews together disparate pieces of information from various data sources. That is a disturbing answer from a positivistic perspective, but it has the potential to be a far better one, at least for the kinds of questions being asked here. This paper, still exploratory in many ways, is an attempt to do some of the initial stitching.

The second and determining factor in the spread of small-group activities is the political process that surrounded the change: the efforts of organized managerial or labor groups to enforce their preferences with regard to the change and their ability to call resources into play to do so. Crucial in this regard is the congruence or divergence of management and labor interests with regard to change, especially the orientation of labor unions, given their traditional strategies and their respective abilities to enforce their preferences. These fluid political processes are summarized in Figure 2-2.

This explanation of cross-national differences in the diffusion and survival rates of small-group activities does not rely on organizational level or group-level variables as much as it does on macropolitical processes. We can predict with some confidence that if a national business leadership is committed to small-group activities as a solution to perceived problems, and this is uncontested by labor, a national infrastructure to legitimate and promote small-group activities will come into place. This is independent of any small-group effects or organizational variables that might affect the process in a given company. It is this kind of analysis that allows us to see the forest for the trees. Similarly, if we take as our dependent variable cross-national differences in some of the most often-studied organizational variables—degree of task specialization and degree of differentiation of roles in the shopfloor—we find a similar situation. That is, our analysis suggests the importance of macropolitical variables in explaining these outcomes; such explanations are rarely seen in the literature.

These observations clearly go well beyond the study of small-group activities and provide support for Perrow’s (1979) observations about moving away from microanalysis of many questions to broader explanatory factors in order to expand the generalizability and applicability of our theories.

It may indeed be that the most profitable route to studying the impact of macropolitical factors is to see these factors as a subset of a still broader set of environmental factors (cf. Freeman, 1978; Meyer, 1978). The differential survival and success rates of organizations and the creation of organizations are currently much studied from a population ecology perspective (Aldrich, 1979). It is apparent from this analysis of the variation in the type and form of national infrastructures for the diffusion of small-group activities that the explanations of such outcomes could benefit substantially from the introduction of macropolitical variables. Conditions affecting the ability of organizations to extract resources from the environment (as exemplified by the resource-dependence model) have long been a concern of those stressing the impact of environments on organizations. The analysis here suggests, however, that the role of macropolitical factors in resource acquisition has been underplayed in much previous research. Conversely, integrating the study of macropolitical factors into environmental research frees the study of macropolitical factors from the constraints of its primarily Marxist framework, which simply cannot do justice to the varied empirical outcomes.

The macropolitical perspective can also be useful in the study of innovation. Studies of factors affecting the adoption and diffusion of organizational innovation typically stress organization-level variables (Kimberly, as cited in Aldrich, 1979: 98), with at best some attention given to selected economic variables. Yet, it is apparent from this analysis that the adoption and diffusion of managerial innovations, e.g., small-group activities, depend in large part on the macropolitical factors outside the organization.

It might be asserted in response to this study that the changes described are not “normal” instances of organizational change, that they constitute special cases of well-defined national movements, and therefore the key explanations are necessarily macropolitical. The movements analyzed here, however, are the direct counterparts of previous movements that transformed industrial organization early in this century—the movements for the open shop, for scientific management, Taylorism, and Fordism. All these movements involved the same kind of interplay between management and labor, the same kind of managerial response to perceived problems, and the same kind of organization and mobilization for change through organized infrastructures. These are not abnormal instances of change. Social change usually takes place through these broader, macro-political processes, and their effects on the micro-organization of factories can be quite profound.

Of course, macropolitics does not explain everything, or even most things, about the way factories and other enterprises are organized. We already have a rich literature telling us what these other variables at lower levels of social organization are. A great deal depends on the questions being asked. For the questions raised in this article, why small-group activities developed differently in the three countries, organization-level variables are not the most important ones. Comparative sociology makes a contribution here by highlighting factors that are always at work in any national setting and historical context. Because they exert relatively constant effects across a range of smaller units of organization, such factors are not clearly perceived without being studied in a comparative framework.

Notes

Acknowledgment: The data for Japan were collected during my tenure as Fulbright Research Scholar, 1977-78. I am indebted to the Fulbright Commission for its support and to the Japan Institute of Labour for providing research facilities. I also express my appreciation to the German Marshall Fund, which provided a research grant for collection of the Swedish data. Research facilities in Sweden were provided by the Department of Applied Psychology at Gothenberg University. Data collection for the U.S. was made possible by a grant from the Henry Luce Foundation, Inc. I am especially indebted to the many business organizations and associations in all three countries that gave freely of their time in an attempt to educate me in the intricacies of organizational change. While I am grateful to the institutional benefactors, they are not responsible for my findings.

My greatest intellectual debt is to Andrew Walder. Our joint working paper, “Structural Diffusion: The Politics of Participative Work Structures in China, Japan, Sweden, and the United States” (Cole and Walder, 1981), provided much of the framework and language that led to this article. Finally, reviewers and editors for the Administrative Science Quarterly were most generous with high-quality advice.

1. Burawoy (1983) did not include labor shortage as a motivating factor for workplace innovations. Although he referred to market forces and the labor process as factors in the emergence of “hegemonic despotism,” a reputedly new stage of capitalism, he did not allow for differences in labor supply that affected the form of new work structures in the four nations he examined (U.S., England, Japan, and Sweden). His analysis, which compared one workshop in England with another in the U.S., was designed to hold labor process and market factors constant. This method highlights the macropolitical factors but entirely excludes the labor-market characteristics that are an important empirical link between the macropolitical factors and the organization of work.

2. This important concept cannot be treated adequately in the space available. Central to the analysis is Bendix’s (1974) notion of “managerial ideology”; my use of the concept focuses on organized management efforts to promote change.

3. The Democracy of Work (or Co-Determination) Law (Medbestammandelagen), passed in 1976, provides Swedish workers and their unions with legal authority to bargain collectively over a wide range of management decisions that previously had been strictly managerial prerogatives.

4. There seems to be little basis, other than that it fits the Marxist perspective, for Burawoy’s contention that state involvement in the Japanese industrial relations system is quite low. Similarly, his description of Japan as closely approximating “the despotic order of early capitalism in which the state offers little or no social insurance and abstains from the regulation of factory apparatuses” (Burawoy, 1983: 599) is innocent of all facts (see Shirai, 1983; Gould, 1984).

References

Agervold, Mogens. 1975. “Swedish Experiments in Industrial Democracy.” In Louis Davis and Albert Cherns, eds., The Quality of Working Life, 2: 46–65. New York: Free Press.

Aldrich, Howard. 1979. Organizations and Environments. Englewood Cliffs, N.J.: Prentice-Hall.

Ayusawa, Iwao. 1966. A History of Labor in Modern Japan. Honolulu: East-West Center Press.

Bendix, Reinhard. 1974. Work and Authority in Industry. Berkeley: University of California Press.

Berglind, Hans, and Bengt Rundblad. 1975. “The Swedish Labor Market in Transition.” Unpublished paper. University of Gothenberg.

Bernstein, Paul. 1976. Workplace Democratization: Its Internal Dynamics. Kent, Ohio: Kent State University Press.

Björk, Lars, Reine Hanson, and Peter Hellberg. 1973. Okat inflytande i jobbet (Increased Influence on the Job). Stockholm: Personaladministrativa Rådet.

Bluestone, Irving. 1978. “Human Dignity is What It’s All About.” Viewpoint 8(3): 21–24.

Braverman, Harry. 1974. Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century. New York: Monthly Review.

Burawoy, Michael. 1983. “Between the Labor Process and the State: The Changing Face of Factory Regimes under Advanced Capitalism.” American Sociological Review 48: 587–605.

Cole, Robert E. 1979. Work, Mobility, and Participation: A Comparative Study of American and Japanese Industry. Berkeley: University of California Press.

Cole, Robert E., and Dennis Tachiki. 1983. “A Look at U.S. and Japanese Quality Circles: Preliminary Comparisons.” Quality Circles Journal 6(2): 10–16.

Cole, Robert E., and Andrew Walder. 1981. “Structural Diffusion: The Politics of Participative Work Structures in China, Japan, Sweden, and the United States.” Working Paper Series no. 226, Center for Research on Social Organization, University of Michigan.

Cook, Alice. 1966. An Introduction to Japanese Trade Unionism. Ithaca, N.Y.: Cornell University Press.

Cummings, Thomas, and Edmond Molloy. 1977. Improving Productivity and the Quality of Work Life. New York: Praeger.

Dachler, H. Peter, and Bernhard Wilpert. 1978. “Conceptual Dimensions and Boundaries of Participation in Organizations: A Critical Evaluation.” Administrative Science Quarterly 23: 1–39.

Dahlström, Edmund, et al. 1971. LKAB och Demokrati. Stockholm: Waldström and Wedstrand.

Davis, Louis, and Albert Cherns (eds.). 1975. The Quality of Working Life, 1. New York: Free Press.

Dore, Ronald. 1973. British Factory—Japanese Factory. Berkeley: University of California Press.

Edwards, Richard. 1979. Contested Terrain. New York: Basic Books.

Emery, Fred, and Eric Trist. 1969. “Sociotechnical Systems.” In Fred Emery, ed., Systems Thinking: 281–96. London: Penguin.

Fox, Alan. 1974. Beyond Contract: Work, Power and Trust Relations. London: Faber and Faber.

Freeman, John. 1978. “The Unit of Analysis in Organizational Research.” In Marshall Meyer and Associates, eds., Environments and Organizations: Theoretical and Empirical Perspectives: 335–51. San Francisco: Jossey-Bass.

Gould, William. 1984. Japan’s Reshaping of American Labor Law. Cambridge, Mass.: MIT Press.

Gyllenhammar, Pehr. 1977. People at Work. Reading, Mass.: Addison-Wesley.

Hammarström, Olle. 1975. “Joint Worker-Management Consultation: The Case of LKAB, Sweden.” In Louis Davis and Albert Cherns, eds., The Quality of Working Life, 2: 66–82. New York: Free Press.

Harvey, Edward, and Russell Mills. 1970. “Patterns of Organizational Adaptation: A Political Perspective.” In Mayer N. Zald, ed., Power in Organizations: 181–213. Nashville, Tenn.: Vanderbilt University Press.

Industrial Union Department, AFL-CIO. 1978. “The Quality of Working Life.” Viewpoint 8(3): 1–29.

Kanter, Rosabeth. 1983. The Change Masters. New York: Simon and Schuster.

Karlsson, Lars Erik. 1969. Demokrati pa Arbetsplatsen (Democracy at the Workplace). Stockholm: Prisma.

Katzell, Raymond, Daniel Yankelovich, et al. 1975. Work, Productivity, and Job Satisfaction. New York: Psychological Corporation.

Korpi, Walter. 1970. Varför Strejkar-Arbetarna? (Why Do Workers Strike?). Stockholm: Tidens Förlag.

——. 1978. The Working Class in Welfare Capitalism. London: Routledge & Kegan Paul.

Landsorganisation. 1976. Arbetsorganisation (Work Organization). Stockholm: Tidens Förlag.

Mellbourn, Anders. 1984. “MBL Inte så Dålig som sitt Rykte” (The Co-Determination Law Is Not as Bad as It’s Rumored to Be). Dagens Nyheter, April 29: 16.

Meyer, John. 1978. “Strategies for Further Research: Varieties of Environmental Variation.” In Marshall Meyer and Associates, eds., Environments and Organizations: Theoretical and Empirical Perspectives: 352-68. San Francisco: Jossey-Bass.

Minami, Ryoshin. 1973. The Turning Point in Economic Development: Japan’s Experience. Tokyo: Kinokuniya Bookstore.

Ministry of Labour. 1974. Koyō kanri shindan shiyō (Indicators of the Conditions of Employment Administration). Tokyo: Employment Security Office.

Montgomery, David. 1976. “Workers’ Control of Machine Production in the Nineteenth Century.” Labor History 17: 485–509.

Nakayama, Saburo. 1972. Zen’in sanka keiei no kangaekata to jissai (All–Employee Management Participation: Viewpoints and Practices). Tokyo: Japan Federation of Employers’ Associations.

Nelson, Daniel. 1975. Managers and Workers. Madison, Wis.: University of Wisconsin Press.

New York Stock Exchange. 1982. People and Productivity. New York: New York Stock Exchange.

Okamoto, Hideaki. 1975. “Jizen Kyōgisei no Ronri” (The Logic of the Prior Consultation System). In Nihon Rōdō Kyōkai, ed., Haichi Tenkan o Meguru Rōshi Kankei: 43–93. Tokyo: Japan Institute of Labour.

Perlman, Selig. 1968. A Theory of the Labor Movement. Originally published in 1928. New York: Augustus Kelly.

Perrow, Charles. 1970. “Departmental Power and Perspectives in Industrial Firms.” In Mayer Zald, ed., Power in Organizations: 59–79. Nashville, Tenn.: Vanderbilt University Press.

——. 1979. Complex Organizations: A Critical Essay, 2d ed. New York: Random House.

Pfeffer, Jeffrey. 1978. “The Micropolitics of Organizations.” In Marshall Meyer and Associates, eds., Environments and Organizations: Theoretical and Empirical Perspectives: 29–50. San Francisco: Jossey-Bass.

Rothschild-Whitt, Joyce. 1979. “The Collectivist Organization: An Alternative to Rational-Bureaucratic Models.” American Sociological Review 44: 509–27.

Shirai, Taishiro (ed.). 1983. Contemporary Industrial Relations in Japan. Madison, Wis.: University of Wisconsin Press.

Sorge, Marjorie. 1984. “UAW Lists Top Priorities for Talks in Canada.” Automotive News (Detroit: Crain Automotive Group), April 23: 4, 8.

Statistiska Centralbyrån. 1978. Arbetsmarknads-statistisk Arsbok 1977 (Yearbook of Labor Statistics 1977). Örebro: National Central Bureau of Statistics.

Stone, Katherine. 1975. “The Origins of Job Structures in the Steel Industry.” In Richard Edwards, Michael Reich, and David Gordon, eds., Labor Market Segmentation: 27–84. Lexington, Mass.: Lexington Books.

Tannenbaum, Arnold S., Bogdan Kravcic, Manachem Rosner, Mino Vianello, and Georg Wieser. 1974. Hierarchy in Organizations. San Francisco: Jossey-Bass.

Thomas, Robert. 1984. “Participation and Control: New Trends in Labor Relations in the Auto Industry.” CRSO Working Paper no. 315. Ann Arbor, Mich.: University of Michigan, Center for Research on Social Organizations.

Thorsrud, Einar, and Fred Emery. 1969. Medinflytande och Engagemang in Arbetet (Participation and Engagement in Work). Stockholm: Utvecklingsrådet for Samarbetsfrågor.

Tichy, Noel, and Jay N. Nisberg. 1976. “When Does Work Restructuring Work? Organizational Innovations at Volvo and GM.” Organizational Dynamics 5: 1 (Summer): 63–80.

Walton, Richard. 1978. “Teaching an Old Dog Food New Tricks.” Wharton Magazine, Winter: 38–47.

Wool, Harold. 1973. “What’s Wrong with Work in America?” Monthly Labor Review 96(3): 38–44.

Zald, Mayer. 1970. Power in Organizations, Nashville, Tenn.: Vanderbilt University Press.

© 1985 by Cornell University. Reprinted with minor revisions from “The Macropolitics of Organizational Change: A Comparative Analysis of the Spread of Small Group Activities,” by Robert E. Cole, published in Administrative Science Quarterly 30, no. 4 (1985), by permission of Administrative Science Quarterly.