Union Presence and Work Team Autonomy in the Aluminum Industry

The introduction of teamwork constitutes a radical change for managers and workers because it involves flexible forms of work organization which foster the individual and collective empowerment of team members. Two models of work organization can be distinguished regarding the degree of autonomy that characterizes work teams. The lean production model gives a dominating role to managers in decision-making related to team-based work organization, whereas the socio-technical model lends much more importance to decisions made by team members in managing their work.

Several studies confirm that labour-management cooperation in managing organizational changes contributes to employee empowerment and involvement in work teams (Cohen-Rosenthal and Burton, 1993; Cooke, 1994; Cutcher-Gershenfeld, Kochan and Verma, 1991; Havlovic, Kroll and Bushe, 1993). Other studies show that the presence of a strong and independent union involved in work organization is a determining factor in the autonomy of work teams and the success of organizational changes (Cohen-Rosenthal, 1997; Frost, 2000; Lapointe, 2001; Lévesque et al., 1996; Wright and Edwards, 1998). These studies thus support the hypothesis that trade union presence has a positive effect on work team autonomy. This hypothesis is backed up by numerous studies conducted at aluminum plants in Quebec which underline the beneficial effect of trade union presence on employee job security and teamwork (Bélanger, 2001; Bélanger, Dumas and Monette, 1995; Edwards, Bélanger and Wright, 2002; Lapointe, 1991; Socher, 2000).

However, this hypothesis runs counter to other research studies which establish that trade union presence is negatively associated with employee autonomy in work organization (Cappelli and Sherer, 1989; Kirmeyer and Shirom, 1986). According to these studies, the negative impact of trade unions on employee autonomy could be explained by the rigidity of collective agreements in terms of work organization due, in particular, to the great number of job categories and task descriptions which restrict employee multiskilling and operational flexibility.

Our research aimed first to confirm the hypothesis of a positive effect of trade union presence on work team autonomy, and if such proved not to be confirmed, to highlight other factors likely to contribute to this autonomy. Thus, we conducted a study in two aluminum plants that used similar technology, one of which is unionized and the other is non-unionized. In each plant, we first conducted two sessions of direct observation in the pot-rooms as well as semi-structured interviews with HR managers and members of the union executive in the unionized plant, and non-structured interviews with pot operators during the direct-observation sessions. Second, a questionnaire was distributed to members of two teams of pot operators in each of the plants in order to assess the degree of autonomy of work teams in each plant.

The hypothesis that trade union presence has a positive effect on the autonomy of work teams is not confirmed by our research results. Indeed, the team members in the non-unionized establishment express a higher degree of autonomy than those in the unionized plant, for most indicators of our operating model, and the differences for each of the six dimensions of this model are significant at the 10% (2 dimensions), 5% (2 dimensions), or 1% (2 dimensions) levels. Moreover, these assessments are in line with the information obtained during our interviews with managers and pot operators, and confirm our direct observations of how work teams operate in the two plants, highlighting the greater flexibility and autonomy of work teams in the non-unionized plant.

These results thus lead us to consider other factors likely to explain the differences observed regarding the degree of autonomy of work teams. The first factor refers to the model of team-based work organization, which relates to the lean production model in the unionized aluminum plant, whereas in the non-unionized plant, it is closer to the socio-technical model of work organization. In addition, the management style applied in the non-unionized plant is much more participatory than that in the unionized plant where, since the start-up of operations, management has exerted more rigid control over work organization.

The striking difference in the climate of labour relations in the two establishments adds to the complexity of an analysis of the link between union presence and work team autonomy. At the time of our survey, the climate of labour relations was considered to be highly positive by employer representatives and members of work teams in the non-unionized plant whereas it was considered to be conflictual by employer and trade union representatives and pot operators in the unionized plant. Due to the conflictual climate of labour relations which has existed in this plant since its unionization, labour-management cooperation has not been able to develop and thus employee involvement in teamwork has not been fostered.

Our research results therefore show that the hypothesis of a positive relation between trade union presence and work team autonomy should be qualified. These results and other studies consulted show that the main explanatory factors of work team autonomy are forms of individual and collective participation integrated into the managerial approach to teamwork and the climate of labour relations. These two factors play a determining role both in the initial implementation of teamwork and at different stages of its development (Bourque, 1999; Cohen-Rosenthal, 1997; Harrisson, Laplante and St-Cyr, 2001). Union presence proves to be a moderating variable rather than a causal variable of work team autonomy because it can positively or negatively influence employee involvement in work teams according to whether it is associated with a cooperative or conflictual work climate.