Corps de l’article

Dramatic increases in workplace restructuring and the attendant reorganization of work have created a new, more challenging environment for organized labour over the past two decades (Ackers, Smith, and Smith 1996; Kochan, Katz, and McKersie 1986; Walton, Cutcher-Gershenfeld, and McKersie 1994). No longer sacrosanct are strict lines of demarcation between jobs, job-determined pay, or workplace rules that can be fully specified in a collective agreement and policed through the grievance procedure. Rather, unions are having to grapple with issues such as multi-skilling, flexible and performance-based pay, and member involvement in on-going productivity and quality improvement teams (Ichniowski et al. 1996). Such change has brought about the need for continuous union-management negotiation throughout the life of the contract.

Increasingly, such tasks have fallen to local unions (Kelly 1996; Nissen 1997; Fairbrother 2000). Although bargaining has long been highly decentralized in North America, local unions often had surprisingly little formal bargaining responsibility. Rather, full time staff of the union, able to specialize in the highly technical knowledge associated with pensions, benefits, and the economics of an industry, often bargained on behalf of many locals (see for example, Lester 1958). In the new environment, however, such agents can not meet locals’ needs for on-going bargaining with management. Instead, locals find themselves having to engage with management, not only in contract bargaining every few years, but also in virtually continuous workplace negotiations.

On-going workplace change and restructuring have created the need for a new set of local union capabilities. In the past, local unions that could police the language of the collective agreement, fairly represent their members’ interests in the grievance process, and could periodically marshal member support for the union’s demands in contract negotiations were deemed successful. Now, however, the capabilities to deal with management and represent members’ interests have become more complex.

Recent research has focused on uncovering these capabilities. Frost (2000) identifies four—accessing information, educating and mobilizing the membership, accessing management decision making at multiple points, and balancing cooperation and conflict—that appear important in enabling local unions to negotiate over workplace change in ways that benefit themselves, their members, and their firms. Yet, little is known about where such capabilities come from and why some locals seem to possess them while others do not. Some preliminary suggestions, however, have been made.

Locke (1995) attributes variation in local union capacities to deal effectively with industrial restructuring to the socio-political context in which the local is embedded. Where institutions (such as employer associations, trade associations, or religious groups) effectively mediate the competing interests of labour and management, restructuring proceeds more smoothly and produces better outcomes for the parties involved. Frost (1994) highlights the importance of a local union’s own historical development in laying the foundation upon which capabilities can be built. Local unions organized from the bottom up by internal activists, she argues, are more likely to create the necessary internal network of union representatives for effective union-led restructuring (Frost 1994).

Unfortunately, however, unions have very little control over the socio-political context in which they find themselves or over the circumstances of their founding and subsequent historical development, leaving many unions struggling to manage the complexities they face in the new competitive environment.

One potential source of help for local unions may be the national union. National unions, however, face a twofold problem in seeking to develop capabilities within their local unions: while they seek to encourage the learning that will help develop vibrant, active, and capable local unions equipped to effectively negotiate with management over the need for on-going restructuring in the workplace, they must also ensure that the idiosyncratic local union agreements that often result do not jeopardize the strength of the union as a whole. This general problem is found in any network in which learning and knowledge sharing are critical to organizational success. It is found, for example, among organizational actors as diverse as franchisees and franchisers (Michael 2000), automotive assemblers and parts suppliers (Dyer and Nobeoka 2000), and alliance partners in biotechnology (Powell, Koput, and Smith-Doerr 1996). In all cases the parties need to reduce the costs associated with finding and accessing valuable knowledge while at the same time preventing free riders and local optimization that occurs at the expense of the system.

Sabel (1993) crystallizes the solution to this twofold problem in what he terms “learning by monitoring.” As others have noted, organizations learn by observing others and importing their practices (March and Simon 1958; Powell, Koput, and Smith-Doerr 1996; Levinson and Asahi 1996). Sabel’s (1993) work examines the relations between the state and firms in economies struggling to restructure. He argues that through the monitoring process (of ensuring agreements are upheld—for example, that alliance partners are fulfilling their mandate or that suppliers are delivering parts as required) the parties can learn from one another (Sabel 1993). I use this “learning by monitoring” lens to begin to conceptualize how a national union might address the double-edged need to develop independent and capable local unions while maintaining union solidarity.

In beginning to conceptualize the role of the national union in creating and sustaining local union capabilities, I focus on national union characteristics that may facilitate learning by monitoring and analyze the experiences of three unions: the Canadian Auto Workers (CAW); the United Steelworkers of America (USWA); and the American Flint Glass Workers Union (AFGWU). Table 1 contains descriptive data on the three unions. As this work is exploratory and involves the in-depth examination of each case, the use of a statistically representative sample is not feasible. Thus, the cases have been carefully selected.

In all three cases, locals of these unions have confronted the need for extensive workplace restructuring in the past decade and a half as management seeks to implement new technology, move to team forms of production, combine jobs, and involve workers in productivity and quality improving suggestion systems. Moreover, the unions under study here represent workers in a range of manufacturing industries—line-paced mass production (automobile assembly), process-oriented production (steel making), and batch processing (circuit boards and ceramic substrates). Thus, although not representative in a statistical sense, the experiences of these locals ought nonetheless to provide insights that are likely generalizable to a broader cross section of unions.

Table 1

Characteristics of the Unions under Study

Characteristics of the Unions under Study

-> Voir la liste des tableaux

As well, the sample has been chosen to show variation on the dependent variable. In two cases, the local unions appear to possess the necessary capabilities to effectively manage on-going workplace change. In the other, the local union appears unable to marshal any resistance to management-led restructuring that is imposing significant costs on the union’s members. By sampling on the dependent variable, I hope to be able to advance some preliminary propositions about the factors responsible for these differences.

Finally, the cases are drawn from both sides of the Canada-U.S. border. The CAW only represents Canadian workers, while the AFGWU only represents workers in the United States. The USWA, in contrast, is an international union representing workers on both sides of the border. Although important institutional distinctions in labour law are often credited for the generally more favourable outcomes enjoyed by organized labour in Canada, in this study favourable outcomes occur on both sides of the border. Thus, institutional differences are unlikely to be a central factor in explaining the outcomes observed here.[1]

This study is based largely on field research conducted at the plant sites and in the local union offices. The one exception is the case study of local union restructuring in the automotive industry which has been drawn from extensive and detailed secondary sources in addition to interviews with the CAW national union president and regional representatives. Each of the more than twenty interviews for the other case studies lasted between one and one and a half hours. In each instance, I took detailed notes that were later sent to company and union representatives for verification.[2] My informants in each case included first-line supervisors, plant managers, IR/HR managers, national union officers, regional union representatives, local union executive members, union shop-floor representatives, and union members. In all cases the interview data have been supplemented with archival materials including copies of collective agreements, memoranda of understanding, and company supplied performance data.

Creating and sustaining local union capabilities: learning by monitoring

The point of departure for this research is the central concept of local union capabilities. Before turning to investigate the role played by the national union in creating and sustaining such capabilities, however, we first need to understand what capabilities local unions now require for success in the new competitive environment.

Local Union Capabilities

Drawing on the experiences of four local unions dealing with management-initiated restructuring in the integrated steel industry, Frost (2000) identified four specific local union capabilities. First is the ability to access information—information resident in members such as knowledge of the technology in use, the work process, or important work norms; as well as information external to the local in the national union, in other locals, or in industry bodies—so that local unions are able to formulate effective strategies regarding workplace reform. Second, Frost (2000) identifies the ability of the local union to educate its members about, and mobilize them in support of, union-led workplace reform. Without the backing of its members, the union’s vision will likely not be heeded by management. Third is the ability of a local union to access management decision making regarding work reorganization at multiple points. By having multiple points of access, the union ensures its vision and concerns are heard and taken seriously. In addition, management gets more information of greater accuracy than if the union had access to management only at a single point or had no access at all. Fourth, and perhaps most critical, is the capability of a local union to balance cooperation and conflict. Despite undertaking initiatives of joint interest that demand cooperation, the local union needs to retain its role as the protector of members’ interests where they run counter to those of management. Without the ability to protect members’ best interests, local union leaders run the risk of being perceived as coopted by management and are unlikely to be re-elected.

In the next section, I examine the experiences of three national unions and one of each of their locals in coping with the need to negotiate over workplace change. Despite facing highly similar competitive challenges and prospective workplace changes, the three locals have fared very differently. In the article’s third section I revisit the cases to identify the characteristics of the three national unions that I argue have facilitated (or blocked) the development of the local union capabilities needed for dealing effectively with workplace restructuring.

Unions and workplace restructuring

The Canadian Auto Workers: CAW Local 88 and CAMI

The Canadian Auto Workers was founded in 1985 after breaking away from the United Auto Workers over disagreements about the appropriateness of concession bargaining and more collaborative relationships with management. In 1999, the CAW was Canada’s largest private sector union and has organized broadly throughout the Canadian economy, representing workers as diverse at flight attendants, fishery workers, and McDonald’s employees (Human Resources Development Canada 1996).

The CAW National Office has evolved several key institutions to manage this diverse membership. Twice a year the CAW Council brings together over 700 elected local union delegates from across the country for two and a half days of discussion and debate. The CAW National Office also maintains a separate research department known as Work Organization and Training that conducts original research concerning the impact of new forms of work organization and new technology on workers and their union.[3] Further, the CAW National Office devotes considerable time, attention, and resources to union education, for both CAW members and their families, that focuses on the union’s role in workplace change, and the kinds of change that are good for workers.

CAW Local 88 represents production and maintenance workers at CAMI, the GM-Suzuki joint venture located in Ingersoll, Ontario. Opened in 1989, GM-Suzuki did not oppose the unionization of the CAMI plant. After 90 percent of the workforce signed union membership cards, the union was duly certified by the Ontario Labour Relations Board making CAMI one of only four organized Japanese transplants in North America (CAW Research Group on CAMI 1993). As part of the agreement negotiated with GM-Suzuki management, the CAW negotiated access for its research team to document at six month intervals over two years workers’ experiences in a lean production plant.

In addition, the CAW realized that the collaborative labour relations envisioned by CAMI would test the CAW’s previous opposition to cooperative relations with employers. As a way to cope with this new relationship, the CAW national office released its Statement on the Organization of Work (CAW 1989) that would serve as a union guidepost to implementing the CAMI system and its attendant values of teamwork and labour-management collaboration.

After a rigorous selection process and a significant training and acculturation period, CAMI workers began to work on the line with high expectations for involvement on the job (Rinehart, Huxley, and Robertson 1997). At the first survey opportunity in March 1990, workers reported high levels of satisfaction with the CAMI production system: 43.3 percent reported that they were enthusiastic and excited about the special kind of production system in place at CAMI; 56 percent reported perceiving CAMI as a democratic wokplace; and 56.9 percent perceived CAMI as cooperative and helpful rather than competitive and stressful (CAW Research Group on CAMI 1993).

However, as the plant ramped up to full production over the next few months, worker satisfaction turned to disillusionment. Under production pressure, workers had little input to changes in the system. Line speed-ups and work redesign were overseen by the plant’s industrial engineering staff. Increasingly, workers found themselves stretched to the limit—additional jobs such as housekeeping, “kaizening” production, or scheduling were not being done and workers struggled simply to keep up with production (Rinehart, Huxley, and Robertson 1997).

Subsequent visits by the CAMI research team documented this discontent. By the time of the team’s last research visit in November 1991, only 11.5 percent of workers agreed that CAMI was special and were excited and enthusiastic about working there—fully 88.5 percent felt CAMI was no different than other corporations; those who thought CAMI a democratic workplace had fallen to 25.9 percent of the workforce; and only 16.7 percent continued to see CAMI as cooperative and helpful—in contrast, 83.3 percent of workers saw CAMI as competitive and stressful (CAW Research Group on CAMI 1993). Overall, workers reported CAMI was “just another car factory” (Rinehart, Huxley, and Robertson 1997).

As it became clear that the lean production system imposed considerable costs on the workforce in terms of repetitive strain injuries, high stress levels, and frustration with management, the union began to intervene on workers’ behalf. After a grievance campaign waged over workload and a number of refusals to work prompted by the rising number of repetitive strain injuries, CAW Local 88 made North American auto industry history by becoming the first union to strike a Japanese transplant. In September 1992, CAW Local 88 struck CAMI for what would become five weeks.

The national union president was involved in the negotiations to end the strike. Of central importance to the national union was the issue of wage parity with the rest of the Big Three in North America. Also of importance to the local union was the negotiation of improvements to life on the line in a lean production plant.

The eventual agreement narrowed the wage gap between CAMI and the Big Three. The eventual settlement also met many of the strikers’ demands regarding the humanization of the Japanese production system. Importantly, provisions were included that limited line speedups and placed floors on staffing levels to ensure greater regulation of workloads. Workers also won the right to elect team leaders, whose duties were now specified by contract, rather than have them appointed by management. The contract also expanded the union’s role in workplace governance, providing for five additional full-time union representatives, a broader mandate for the Health and Safety committee, and establishing joint training, employment equity, and environmental committees. Finally, Local 88 secured additional funding from management of the CAW’s Paid Educational Leave program, the purpose of which was to strengthen Local 88 members’ connection to the larger union and to reinforce the local union’s independent stance towards management.

CAW Local 88’s experience dealing with the deleterious effects on its members of the lean production system demonstrates the ability of a local union to negotiate successfully with management to represent its members’ best interests. By involving itself in workers’ struggles, CAW Local 88 not only won important contractual victories through the exercise of collective power by its members, but it also established itself as a countervailing power to management, protecting and promoting workers’ distinct rights and interests (Rinehart, Huxley, and Robertson 1997).

The United Steelworkers of America: USWA Local 1010 and Inland Steel’s Indiana Harbor Works

The United Steelworkers represents workers in a broad range of industries in North America. Although originally organized to represent workers in the U.S. steel industry, the union quickly moved north of the border to organize Canadian steel workers as well. Today the USWA represents workers in a broad array of settings including hotels, call centers, and community colleges. Because of its broadly diverse membership, the USWA operates in a relatively decentralized manner encouraging local unions to be active and responsive in representing their members. The USWA puts great emphasis on active shop floor representation, encouraging the election rather than the appointment of several layers of shop floor representatives. Further, the union is divided into distinct geographic regions each headed by a District Director who serves as a resource to local unions, providing them with information, strategic guidance, and negotiation skills.

The national union’s research department is also a resource for local unions, providing them with information on the changing workplace such as the pros and cons of new ownership structures (such as ESOPs or employee buyouts) and new governance structures (such as the 1993-negotiated “Partnership”[4]) and their effects on the union and its members. The USWA’s education department also plays a large role in managing the union’s diverse membership. It provides union-focused training on workplace change: the kinds of change that are good for workers, the role the union has to play in such processes, and how to negotiate effective enabling language that protects members’ interests.

USWA Local 1010 represents production and maintenance workers at Inland Steel’s sprawling, century-old Indiana Harbor Works in East Chicago, Indiana. In 1988, Inland Steel management identified its bar-making operation as uncompetitive and faced the stark choice of substantial restructuring or imminent closure. Management approached the USWA District Director to negotiate a new contract that would allow it to lower costs and justify new investment. USWA Local 1010, however, demanded that instead its leadership be involved in any such negotiations, arguing that Local 1010’s members needed access to the decision-making process in any matters that would affect them. Backing up these demands was a work-to-rule campaign waged on the shop floor to pressure management into allowing local union input to the restructuring negotiations.

Management responded with a list of specific demands including a separate (and inferior) contract to cover the bar mill, wage give-backs, and head-count reductions. USWA Local 1010 responded in turn with its own demand: namely that management share its need with the union, not its solution. Management announced that it needed to reduce costs by $40 per ton to make the bar mill competitive. USWA Local 1010 turned to external consultants to assess whether the reduction of $40 per ton would indeed make the mill competitive again. Given several USWA-led employee buyouts of ailing steel mills in the 1980s, the union had developed good working relationships with industry consultants who could provide such information in a credible form. Once the consultants had verified that the mill could indeed successfully compete with a cost reduction of $40 per ton, the parties sat down to find ways to reduce costs.

The parties set up a multi-level committee structure to oversee the restructuring process. USWA Local 1010 has long enjoyed an active shop floor presence with three layers of shop-floor representatives—grievers, assistant grievers, and stewards—all of whom are elected, the great majority of them in contested elections. Drawing on this resource, Local 1010 and plant management created twenty-three design teams on the shop floor, staffing them with union activists and first line supervisors. The design teams then redesigned work by combining jobs in ways that made sense from an operating and skill requirement point of view. Drawing on national union sponsored research on new forms of work organization, the teams sought to combine the narrow Taylorist jobs into clusters of jobs that incumbents would learn and rotate through. As well, union members sought to increase worker autonomy by replacing supervisors with bargaining unit leaders. The plans then had to be approved by departmental committees co-chaired by the area’s griever and the department manager. Finally, the overarching joint committee co-chaired by the president of USWA Local 1010 and the vice president of manufacturing had to give its approval.

However, even after approval by the joint committee, the plans had to be ratified by the affected membership before any changes could be implemented. In 1988, a majority of workers in each area voted to embark on a five-year trial of the new model of work organization. In 1993, the workers voted heavily in favour of maintaining the system. The 1988 mutual agreement was then rolled into the Inland Steel-Local 1010 collective agreement.

The empirical evidence suggests that the restructuring produced benefits for all. Management achieved the cost reductions it needed to justify new investment and workers gained skills and additional pay while at the same time supervision decreased as supervisors were replaced by bargaining unit leaders. The union as an institution also benefited as the bar mill model became the model to oversee restructuring throughout the plant. Thus, the union is now perceived as an important vehicle by which workers have access to important decisions about their jobs, their role in the workplace, and the future of their work lives.

The American Flint Glass Workers Union: AFGWU Local 1000 and Corning’s Erwin Ceramics Plant

The AFGWU is a small union representing workers employed in the glass-making industry. For many years, the AFGWU was dominated by members employed by one company—Corning, Inc.—where, in 1999, close to 40 percent of AFGWU members were employed. Historically, Corning has been a leading-edge, technology-driven company that has enjoyed de facto monopoly status in many of its markets. It has been highly profitable and able to grant generous wage increases to its organized workers, providing the AFGWU a benign environment for many years.

The national union’s research department focuses on providing local unions with financial and economic data to support contract negotiations. In the mid- to late-1980s, however, the AFGWU entered into a number of “Partnership” initiatives with Corning management to undertake the joint redesign of several manufacturing facilities. To support this initiative, the AFGWU national office, jointly with Corning management, provided literature and local union leadership development concerning what Partnership was to be.

Located in upstate New York, Corning’s Erwin Ceramics Plant manufactures the ceramic cores used in automotive catalytic converters. AFGWU Local 1000 represents workers at the Erwin Ceramics Plant as well as workers in nine other Corning plants in the area. Each plant elects a “union executive” (an in-plant Local 1000 representative) to the Local 1000 executive board. The executive board then oversees operations in all Local 1000-represented plants but one.[5] Workers in each plant also elect a set of shop floor representatives to represent their immediate day-to-day interests. However, many of these positions go unfilled due to a lack of interest.

In 1988, cost pressures in the automotive industry began to mount and the Erwin plant found itself squeezed by the Big Three. In 1991, plant management and the local union began to look jointly at ways to reduce costs and improve productivity in the plant. Through a process of “Partnership”, negotiated between Corning and AFGWU Local 1000, the parties agreed on a combination of investment in new technology and redesign of work. Collapsing job categories, removing three layers of management, and replacing shift supervisors with “coaches” enabled Corning to decrease staffing and costs considerably.

Another drop in market demand came in 1995 and Erwin was again forced to reduce costs by 2 to 3 percent. Working with the local union leadership, the plant manager introduced cost-cutting process improvements and in June 1995 announced the layoff of 200 workers as the oldest production line (and the only one that had not adopted the 1991 model of work organization) was eliminated. At the same time, management announced the introduction of new technology to dramatically shorten the work-in-process period, improve scrap rates, and automate material handling eliminating a significant number of positions. As a result, employment at Erwin fell dramatically from 1996 levels of approximately 800 employees to 285 production and maintenance staff in 1999.

Concurrent with the layoffs, the new plant manager began to radically shift the culture of the Erwin plant. She perceived “Partnership” to be out of control, requiring negotiation over operational decisions that ought to have been management’s alone to make. To remedy this situation, she put supervisors back out onto the shop floor and insisted they enforce the contents of the collective agreement—especially with respect to job descriptions, break times, starting and ending times, and attendance. Managers reported that workers were now fully engaged for the duration of their shifts and that workloads had increased dramatically.

There was considerable resistance to these changes from the workforce. The in-plant union representative, however, perceived that they were necessary if the plant were to survive. Working closely with plant management on a daily basis, he supported the changes and helped to implement them even though they inevitably led to the loss of jobs. In the words of the plant manager:

What the union leadership has allowed us to do here has been just fantastic. They have shown real leadership and have really been great for us to work with. They could have been fighting us all the way, but the union didn’t. We are lucky to have a very mature union. We can agree to disagree on occasion, but the union will continue to go along with us. This union is really dedicated to the business’ success—even if it means the loss of jobs.

The restructuring at the plant—the downsizing, the reassertion of managerial control, and the implementation of new technology—all paid dividends for Corning management as the Erwin plant enjoyed superlative financial performance in 1999. As a result, management announced that Erwin would be the site of a new multi-million dollar developmental manufacturing line that would provide several dozen new jobs. However, because of the highly skilled nature of the new jobs, more stringent selection criteria would be in place making most regular Erwin employees ineligible for these jobs.

This news only served to further exacerbate the levels of discontent already experienced by workers in the plant. Moreover, more downsizing lay ahead with the announcement that the quality inspection portion of the main operation was to be automated. In the words of one operations leader:

The upcoming downsizing makes the last one look like a walk in the park. People here are fed up with the changes over the last ten years. People are working harder here than ever before, are rising to challenge after challenge put before them to save this plant and then there is another reduction in force!

It appears that the restructuring of the Erwin plant has produced large, positive gains for Corning management. However, it has significantly reduced employment opportunities for Local 1000 members, increased the workload for members who remain at the plant, and greatly heightened their insecurity about their continued prospects of employment. Furthermore, the management-dictated restructuring has considerably weakened the union on the shop floor. The current in-plant Local 1000 representative is perceived as coopted by management and able to do nothing to prevent management doing exactly what it wants to do. Apathy towards their union characterizes many members’ feelings towards Local 1000.

AFGWU Local 1000 and Corning’s Photonics Plant

In contrast to the Erwin plant, that has faced considerable competitive pressure and undergone significant restructuring, the Photonics plant is part of Corning’s high growth, highly profitable fibre optics division. Opened in early 1997, the Photonics plant manufactures circuit boards to connect optical signals in fibre optic networks. Voluntarily recognized by Corning management, AFGWU Local 1000 represents the plant’s 550 workers. Rather than restructuring or downsizing as at Erwin, the union’s focus at Photonics has been on the management of growth and the successful implementation of the high performance work system management envisions for the plant.

From the beginning, the Photonics plant has had a special relationship with AFGWU Local 1000. Whereas all other Local 1000 plants elect a member to the Local 1000 executive board and the executive board in its entirety oversees the operations of all plants, Photonics is overseen only by its own elected in-plant representative and the Local 1000 president. In addition, the Photonics plant is covered by a separate Local 1000 agreement, rather than the valley-wide contract that covers all other Local 1000-represented plants.

The separate Photonics contract allowed management to make three important breaks with the standard Local 1000 agreement. First, Corning put into place a very rigorous selection process. Other Local 1000 members who wished to transfer to the new and growing business were frustrated to learn that very few current Corning employees had the necessary qualifications to be hired at Photonics. Of the 550 who worked at Photonics in July 1999, fully 450 were new to Corning. Second, starting wages for Operations Associates (or OAs as they are known) were set at $9.66 per hour, rather than the $11.00 per hour paid at other Local 1000-represented plants. Wages increased in recognition of the significant training investment workers were asked to make, rising to $10.02 after 6 months and to $10.72 after the completion of the full 320 hours of training. Third, management had secured the right to staff up to 20% of the OA positions on average with contract employees. These contract employees also were paid $9.66 per hour, but they received no benefits, nor were they members of Local 1000. During a site visit in July 1999, close to 40% of the OAs were contract employees.

Because of its separate contract and governance structure, the Photonics plant had little contact with the rest of Local 1000. Instead, the elected in-plant union representative at Photonics focused largely on his own plant and its needs. The union representative worked closely with plant management on a daily basis and was seen frequently on the shop floor with his management counterparts. Despite his personal visibility, the in-plant union representative recognized that the union as an institution was not highly visible in the plant as it had a thin shop floor presence and he wanted to increase employees’ awareness of their union. To help give purpose to the union in the plant, the union representative formulated a list of union objectives for 1999 that specified ways the union could add value to the Photonics business. The in-plant union representative sought to build a more significant union presence by getting more shop floor union representatives in place and integrated into working with the team leaders (supervisors) on the shop floor. However, the item receiving top priority was helping management make and ship product to customer specifications.

With a highly profitable product, with a huge market potential, management wanted to continue to expand the Photonics plant. Corning was especially pleased about the nature of the relationship it enjoyed with the AFGWU Local 1000 president and the Photonics in-plant union representative. Knowing the centrality of fibre optics to the company’s continued financial success, a work stoppage at a fibre-related plant could spell quick disaster for the company. Employee relations management at Corning headquarters regarded the current president of Local 1000 as the best president Corning had worked with. Having served three consecutive terms in office, the local union president was credited with never having stood in management’s way and with always understanding what was best for the business. Similar praise was forthcoming regarding the Photonics in-plant union representative. In the words of one manager:

[Name of in-plant representative] is outstanding. He is a real proponent of making the business work. He could easily be a salaried staff person.

Despite such praise for their union leadership, workers at Photonics were increasingly dissatisfied. Photonics workers found themselves working side by side with increasing numbers of contract employees. In addition, they perceived themselves to be underpaid, both compared to other AFGWU Local 1000 members and to similar others in the external labour market. Given the extremely tight labour market in 1999, many such frustrated Photonics employees simply left Corning for better opportunities elsewhere, resulting in turnover of approximately 66 percent. In the meantime, the union at Photonics was largely powerless to do anything except continue working with plant management and hope somehow to create a more vibrant union presence on the shop floor.

The role of the national union in creating and sustaining local union capabilities

The experiences of the local unions examined here indicate significant differences not only in the process of restructuring, but also in the outcomes experienced by the unions and their members. What helps to explain the variation among these local unions I argue are critical differences in national union characteristics. Although not wholly determining a local union’s ability to negotiate effectively with management over the need to restructure, characteristics of the national union provide a set of resources available to affiliated local unions. Table 2 summarizes differences in the three national unions along several key dimensions. These dimensions appear, from the cases described above, to allow the union to learn through monitoring, facilitating the development of local union capabilities.

As can be seen in Table 2, the CAW and the USWA are broadly similar across the five dimensions noted, while the AFGWU differs starkly from the other two unions. Despite the similarity between the CAW and the USWA on these national union characteristics, however, the relationship of each national union to the local union under study here is quite different largely, I argue here, due to the existing capabilities of the locals themselves. In the case of USWA Local 1010, the local union has existed since the early 1940s, was organized from the ground up by dedicated internal union activists, and over its history has developed and nurtured a set of capabilities that have well equipped it to deal effectively with current workplace restructuring (see Frost 1994). The role for the USWA national union in dealing with Local 1010 is largely one of providing resources to a local union that knows its needs and seeks to fill them out.

In contrast, CAW Local 88 has a very different history from USWA Local 1010. Voluntarily recognized by GM-Suzuki management before CAMI was opened, CAW Local 88 came into existence as soon as the first workers were hired and had signed union membership cards. As a result, at founding, the union existed in little more than name. The role of the CAW national union thus became one of much more active intervention in the development of the capabilities needed by the local union to deal effectively with on-going workplace change.

Table 2

National Union Characteristics Facilitating the Development of Local Union Capabilities

National Union Characteristics Facilitating the Development of Local Union Capabilities

-> Voir la liste des tableaux

The first national union characteristic that I argue is important is the national union’s representational breadth, by which I mean the range of industries and employers across which it represents workers. A union monitoring locals dispersed across a broad array of industries and a range of different firms (large and small, facing competitive challenges of different kinds) has the opportunity to learn from a large number of diverse experiences. It then has the ability to leverage that knowledge to support local unions confronted with similar challenges. Walsh’s (1994) work examining unions in the airline industry supports this notion.

The experiences of the local unions profiled here also appear consistent with this. Both the CAW and the USWA represent workers across a broad range of industries and employers. The experience of USWA Local 1010 with the restructuring of Inland Steel’s bar mill illustrates the learning potential a broad network affords a union. In negotiating with management, USWA Local 1010 drew on the expertise of consultants known to Local 1010 through previous union contacts to assess the bar mill’s likely future.

Similarly, the experience of CAW Local 88 in successfully negotiating a new collective agreement with CAMI management drew on the knowledge gleaned by the national union from negotiations in the broader automotive industry. The CAW national office was aware of pay levels and working conditions throughout the auto industry and by providing that information to the local union, CAW Local 88 was able to negotiate a softening of the lean production system as well as comparable contract provisions.

In contrast, the AFGWU is a small national union representing workers in a single industry, with a significant proportion of its members employed by a single company. Because much of its membership was employed at Corning and because the company was traditionally highly profitable, Corning had long served as the AFGWU’s pattern setter. Once Corning got into financial difficulty and demanded changes in the union-management relationship, the AFGWU found itself poorly poised to respond effectively. As Corning experienced financial difficulties, downsizing negatively affected the AFGWU’s membership and its financial position. After relying so heavily for so long on the largesse of Corning, the AFGWU found itself without a broad range of experience to draw upon and without the resources to gather information or seek help from outside. The lack of resources due to its size, its single industry status, and its historical reliance on a single company all served to restrict the AFGWU’s learning opportunities that might otherwise have helped AFGWU Local 1000 negotiate more proactively and successfully with Corning management.[6] This leads to the first proposition:

Proposition 1: Broader representational coverage (across firms and industries) by a union provides more exposure to learning opportunities through national union monitoring of its locals’ activities.

The second characteristic of the national union that appears important in promoting union learning is the focus of, and resources devoted to, the union’s research department. National union monitoring of conditions its local unions are confronting alerts the union to current needs. By conducting research on the changing workplace—new forms of work organization, new technology, or new forms of labour-management governance—the national union is creating knowledge and distilling learning that can be accessed by or provided to its affiliated local unions.

Eaton’s (1990) study of local union control of union-management participation programs cites the importance of the union’s research department in providing guidance to local unions about how best to manage their interactions with management in joint programs. The three national union cases described in this paper provide evidence consistent with this. For example, as the national office of the CAW negotiated the first contract with CAMI management, it recognized the need for guidance regarding lean production and collaborative union-management relations. The research that resulted in the Statement on the Organization of Work (CAW 1989) provided not only CAW Local 88 with learning and guidance, but the findings have also been used by other locals confronting similar changes in their workplaces (CAW Research Group on CAMI 1993).

Similarly, as the USWA confronted the need to reopen its contracts with employers in the early to mid-1980s during the depth of the steel recession, the union’s research department developed a set of guidelines to govern union-management workplace negotiations. These included joint representation, protective enabling language, and worker ratification of any proposed changes. These guidelines were evident in the restructuring of Inland Steel’s bar mill negotiated by company management and USWA Local 1010.

In contrast, over the past decade AFGWU Local 1000 has faced tremendous pressure for restructuring in all its plants. The pressures for change are also diverse, including demands for new technology, job redesign, and contracting out. Yet, the national union’s research department has provided scant guidance for local union leaders in how to manage such change or what options might be available to counter management’s demands. Instead, in-plant union representatives are largely left on their own to fend for themselves, taking direction on how to proceed from their management counterparts. Failure on the part of the national union to monitor conditions and challenges faced by its local unions has resulted in the absence of activity to support local unions with relevant learning. This leads to the second proposition:

Proposition 2: National union research on the issues and challenges for its local unions raised by on-going workplace change provides learning that can be provided to or accessed by local unions.

The third feature of the national union that appears important in enabling learning by monitoring is the union’s training and education function. Monitoring of local union actions can be done both directly and indirectly: directly through observation and reporting systems and indirectly through the structuring of union member expectations (Offe and Wiesenthal 1980).[7] Union training and education are important for both. In direct terms, training and education help unions learn how best to manage the new demands for on going negotiations over workplace change (Eaton 1990; Ellinger and Nissen 1987). Indirect monitoring is aided by union education regarding the range of behaviours and activities that are accept- able. With this training, local union behaviour is more likely to correspond to these guidelines making the national union’s monitoring job easier (Sabel 1981).

All three national unions studied here provide local union leaders with training in the basic skills of grievance handling, contract interpretation, and running effective meetings. In addition to this traditional union education and training, the CAW and the USWA have expanded their educational offerings to include training for local union leaders and the general membership about on-going workplace changes including the union’s role in workplace change, the kinds of changes that are good for workers (and those that are not), how to negotiate effective enabling language for joint undertakings, and how to monitor joint initiatives to ensure members’ interests are protected. Such education plays a large role in aiding the national union’s monitoring role. No such education takes place in the AFGWU. What little training does occur is jointly designed and delivered with Corning management.

USWA Local 1010 drew on Steelworker training to guide its redesign of work in the bar mill. Changes in work organization that were worker-friendly (upskilling rather than deskilling; increasing autonomy by removing supervision) came from union education. CAW Local 88 drew on national union-provided training to promote worker interests (solidarity, safe work, and an independent union voice) in its negotiations with management to soften the lean production system. This leads to:

Proposition 3: National union-sponsored training and education aids union monitoring by outlining behaviours and activities that are acceptable (or not) to the union. In addition, national union-sponsored training and education can be provided to or accessed by local union leaders and activists to support their negotiations with management about workplace restructuring.

The fourth feature central to a union’s ability to learn by monitoring is effective communication channels. These enable the national union to monitor the activities of its local unions to detect transgressions, recognize local unions needing assistance, and to identify those that might be sources of learning for the rest of the union. Further, such channels enable the transmission of knowledge gained by the national union to its constituent locals and allow local unions to access knowledge resident in the national union. As well, such channels promote communication across locals, enabling them to learn from one another.

Locke (1992) highlights the importance of effective communication between the national and its local unions in defining roles and responsibilities. When each party’s role was clear, restructuring in the Italian automobile industry proceeded more smoothly and with better outcomes for all involved than when the converse was true (Locke 1992).

The CAW and the USWA have established such effective communication channels while the AFGWU has not. The CAW Council provides the CAW with a highly effective communication channel. The semi-annual meeting of over 700 elected local union delegates provides three-way flows of information—from the national union to the locals, from the locals to the national union, and between local unions. The USWA accomplishes much of the same through the institution of the District offices and the role of the District Director. Located in the regions, the District Directors are close to the workplaces in which members are confronted by the on-going need for restructuring. Thus, they can observe the issues local unionists are grappling with and can provide resources as needed. The AFGWU has no such comparable institutions. The only opportunity for intra-union communication on any scale comes at the union’s triennial convention. Yet, because of poor finances, many locals cannot afford to send representatives. This leads to:

Proposition 4: Union communication channels enable the national union to monitor the key issues confronting its affiliated locals while at the same time ensuring local unions have access to or can be provided the learning of the national union and other locals.

The structure the national union imposes on its locals is the final national union characteristic that appears important in enabling union learning through monitoring. Two dimensions seem especially important: whether the local represents workers at a single workplace or across several; and whether the local union has the active participation of a dense network of shop floor representatives.

Leaders of local unions that encompass several workplaces have a more difficult time monitoring and overseeing the specifics of change in each setting. The difficulties are compounded the greater the number of workplaces and the more disparate the nature of the sites’ businesses. Such difficulties may be overcome, however, if a national union structures local representation to include dense and accountable shop floor representation (Wells 1987). A dense and effective shop floor presence enables a local union to assess its needs and to implement what it has learned. Rubinstein (2000) credits the ability of UAW Local 1853 to effectively manage the joint quality program at Saturn to the richness of that local’s in-plant representation.

Both CAW Local 88 and USWA Local 1010 represent workers in a single plant.[8] In contrast, AFGWU Local 1000 represents workers in ten different Corning plants. In the single plant locals of the CAW and USWA studied here, dense networks of elected shop stewards monitor workplace relations. In the multi-plant AFGWU local, a single elected in-plant representative oversees each plant with the help of a handful (or fewer in many cases) of elected shop floor representatives.

Thus, CAW Local 88 and USWA Local 1010 were able to monitor the changes taking place in a single plant in which the local union’s leadership was active on a daily basis. The AFGWU Local 1000 leadership however, was spread thinly across ten plants, unable to effectively monitor change for the local union as a whole. Compounding the difficulty was the fact that the ten plants made different products, competed in different markets, confronted different competitive challenges, and faced different issues with respect to the changes management wished to make on the shop floor. For some it was the implementation of new technology, for others it was job combinations, and for still others it was the relocation of work to Mexico.

Moreover, the CAW and USWA locals each had a dense network of elected shop floor officials[9] in place to monitor management action and respond appropriately to it. In both these unions, norms require the election of such officials and the staffing of a relatively dense network of them in each local. Thus, holders of these positions must campaign for their posts, know the issues of concern to workers, and be prepared to promote workers’ interests in negotiations with management.

The lack of such activists or even a meaningful union presence on the shop floor in the case of AFGWU Local 1000 meant that that union had a much more difficult time responding to management-initiated change. In most cases, the elected union official within each plant was solely responsible for deciding how to respond. Not being well integrated into the rest of the union also meant that the union representative had few resources to draw upon. Thus, management’s vision for change was largely passively accepted and implemented with the assistance (rather than the resistance) of the union representative. This leads to:

Proposition 5: A local union structure in which a local union represents just a single site and includes a dense network of shop floor representatives, will be more likely to enable a local union to monitor workplace change and seek out and implement appropriate learning.

Summary and conclusion

The purpose of this paper has been to further our conceptualization of local union capabilities by seeking to understand where such capabilities come from. I have argued here that characteristics of the national union—its representational breadth, the extent of its education and training focus on new workplace issues, the resources it devotes to research on new workplace practices, the presence of multiple communication channels, and its structuring of local union representation—play an important role in creating and sustaining those capabilities. These five characteristics of the national union together aid a local union in developing and sustaining the capabilities needed to successfully respond to management-initiated workplace change: locating information, educating and mobilizing their membership regarding a union strategy for workplace reform, accessing management decision making at several points, and balancing cooperation and conflict.

The evidence from these case studies also suggests that three categories of local unions exist. There are those local unions such as CAW Local 88 that are initially lacking the capabilities necessary to be effective in the current competitive environment whom the national union can actively help—through training and education, through leadership development, through the provision of appropriate information. There are also those local unions like USWA Local 1010 who have been able to develop the internal capacities to deal effectively with management-initiated restructuring on their own and who themselves access the resources of the national union to seek information, to gain additional training, or to locate a given type of expertise. Finally, there are those local unions like AFGWU Local 1000 who have not developed the needed capabilities to cope effectively with the changes proposed by management either due to apathy, a lack of resources, or internal disarray. Not only does the local union not actively seek out help, but even if it did, the national union is ill-prepared to provide it.

These conclusions, however, are based on just three case studies of national unions. However, consistent findings have come from research drawing on a much different methodology and a much larger sample. Using Canadian survey data collected from over 900 leaders of local unions affiliated with one of the major labour centrals in Quebec, the CSN, Murray et al. (1999) find that local unions who actively confront the need to restructure at the workplace are more highly integrated into their national union. The case study evidence presented here adds considerable detail in support of the argument these authors make. For example, the authors write:

These results suggest that when the union is closely articulated to the larger union structure, which creates greater access to the technical and human resources of the larger union and facilitates the transfer of experiences between local unions, the local union is most likely to espouse an offensive stance towards work reorganization. Such local unions are closely tied to their members as a result of both a structure of delegates at the shop-floor level, whichincreases union leaders’ day-to-day communication with their members and a strong sense of identity with the union on the part of the members.

Murray et al. 1999: 185

Given their cross-sectional data, the authors make no claims about the causal relationship between a local union’s ability to successfully negotiate over restructuring and its integration into the national union. My work suggests that affiliation with a national union possessing the five characteristics highlighted in this paper provides the resources local unions require to develop and/or sustain the capabilities needed to successfully cope with the challenges of workplace restructuring. In one scenario, the national union actively seeks to build a capable local union. In a second scenario detailed here, the local union itself accesses the resources of the national union to support its on-going negotiations with management over workplace restructuring.

By seeking to further develop the work begun with the identification of local union capabilities, this paper has sought to broaden our understanding of variation in the process and outcomes of workplace restructuring. By shifting attention now to understanding where these capabilities come from we have shed further light on the heretofore largely unexplored variation in local union activities.