Voice within trade unions? A test of the voice and loyalty hypothesis

Abstract

Frustration–aggression theories, such as exit–voice and loyalty, have been tested using agents as voice but have not been examined with unions as the organization in question. Utilizing longitudinal panel data (n = 874), this article tests this theory with regard to voice within a trade union. This study does not find support for the exit–voice model; instead, it provides support for cost–benefit and interactionist theories. Further, the study finds that members with high union satisfaction are more likely to express voice when they perceive their leaders to be highly responsive to their needs and concerns.

Keywords

Exit–voice and loyalty trade unions

Introduction

Union participation is crucial for the survival of trade unions. Participation is a necessary channel of communication if members are to express their concerns to the leadership (Barling et al., 1992). Furthermore, participation can also be seen as a measure of union strength and efficiency (Fosh, 1993). Given the decline in unionization over the last few decades in most developed countries, research in the area of union participation is critical. Several authors (Fairbrother and Waddington, 1990; Fosh, 1993; Hurd et al., 2003; Hyman, 1984) have explored the notion that union renewal is possible through workplace activism. Fosh (1993: 589) contended that ‘local leaders, by their ability to lead in a way that encourages members to become involved … can build upon surges of participation and interest, thus increasing the strength of workplace unionism’.

Frustration–aggression theory suggests that members will either participate (exhibit voice) or leave the union when they are dissatisfied with the services provided by that union. In this article, we first examine the determinants of union participation through the lens of frustration–aggression theory by controlling for factors that have been associated with alternative explanations for union voice: interactionist and cost–benefit theories. The study involves the utilization of longitudinal panel data at three points in time and investigates the interaction between union satisfaction and two variables: union loyalty and leader responsiveness. These are unexplored relationships in the literature.

The article then proceeds with a review of the literature, followed by the methods and results sections. A discussion section examines the implications of the findings for theory and practice.

Theories of union participation

Barling et al. (1992: 99) define participation as ‘the extent of the individual’s active involvement in, and influence over, both formal and informal activities’. Informal behaviours are activities that reflect support, but are not necessary for, the unions’ survival (i.e. helping other members file grievances, reading union literature and talking about the union with friends) (Barling et al., 1992; Fullagar and Barling, 1989). Formal activities include becoming a union representative, participating in collective action, participating in meetings and filing a grievance.

This section discusses three main theories in relation to union participation. Although the aim of the article is to test the exit–voice and loyalty theory in relation to union participation, interactionist and cost–benefit theories are also discussed as potential explanations. This enables us to control for alternative explanations of union participation.

Frustration–aggression (exit–voice) theory

A number of early studies in union participation explored the notion that participation is a response to dissatisfaction or alienation, either at the workplace or with the union (Van de Vall, 1970). One theory that falls under the domain of the frustration–aggression tradition is Hirschman’s (1970) exit, voice and loyalty model. Although Hirschman’s model has been commonly applied in the context of labour management relations (Freeman and Medoff, 1984), trade unions have primarily been discussed as agents of voice, rather than as the organization in question. This has been an interesting omission, as Hirschman has argued that voluntary organizations (such as trade unions) provide ‘excellent examples’ of where exit and voice may coexist in a functional manner (Hirschman, 1970: 77). Hirschman (1970) argued that there are potentially two signalling mechanisms to inform management (or, in our case, union leaders) of organizational decline: exit and voice. Exit is simply the classical economic response of withdrawing services, support or membership from an organization or firm. Another mechanism of adjustment, voice, has been explored more readily in the political literature. Hirschman (1970: 30) defines voice as:

any attempt at all to change, rather than to escape from, an objectionable state of affairs, whether through individual or collective petition to the management directly in charge, through appeal to a higher order authority with the intention of forcing a change in management, or through various types of actions and protests, including those that are meant to mobilise opinion.

The application of this theory to a trade union is logical. In this article, voice is measured in terms of union participation (and so draws upon that literature). Indeed, a study by Leicht (1989) puts forward an argument that union participation is a measure of voice for members. Further, participation is a common proxy for voice in the human resource management and industrial relations literature (Dundon et al., 2004; Luchak, 2003; Strauss, 2006).

Interactionist theory

Klandermans (1986b) has argued that interactionist theories provide an important explanatory perspective of union participation. That is, participation is viewed as the result of the social context of members at work and in the union. There is a strong early literature supporting this view of the ‘union man’ (Perline and Lorenz, 1970; Strauss, 1977) and union culture (Goldthorpe et al., 1968).

More recently, Kelly (1998) has placed great emphasis on the interactionist model to explain how individuals potentially acquire collective interests and how this might be utilized by individuals to form grievances. Kelly (1998) has argued that the social movement literature (see especially Gamson, 1992, 1995; McAdam, 1988; Tilly, 1978) provides a useful theory in understanding collective action. The literature addresses the questions of how grievances become collective concerns and the conditions under which individuals will organize.

Mobilization theory helps to explain how and why collective action is frequently brought about by a small number of activists ‘who play a crucial role in promoting a sense of grievance, generating group cohesion and using and legitimising collective activity against a relevant target’ (Kelly, 1998: 213). In particular, transformational leaders are able to activate social identities and get subordinates to behave in a manner consistent with that identity. Fantasia’s (1988) study of wildcat disputes illustrates the importance of leaders in playing this role.

Cost–benefit theory/rational choice

Another influential theoretical tradition for explaining participation is that of rational choice, which focuses on the costs and benefits of participation. A key assumption underlying these theories is that individuals act to maximize their interests. Acting as rational agents, individuals do not take into account other agents’ behaviour.

Olson (1965) was a very prominent author in the rational choice school, whose main contribution was to explain the free-rider problem in unions by analysing the sociology of large groups. The free-rider problem exists because although all individuals within the group can enjoy the collective good, there is no incentive to provide the good. The problem is encountered twice over in unions: first, in contexts where individuals are able to experience the positive externalities of membership without actually joining; and second, where members are unprepared to participate in crucial activities to ensure the union’s survival (Kelloway and Barling, 1993). The suboptimal supply of the collective good, however, can be overcome when separate and selective incentives are provided to stimulate individuals to act to benefit the group. Large groups are therefore ‘latent’, as they have a latent capacity for action when members are mobilized by selective incentives or punishments for free riding.

Klandermans (1986a) also made the argument that individuals take into account the costs and benefits of collective action. He argued that ‘participation in a social movement is seen not as the consequence of predisposing psychological traits or states, but as the result of rational decision processes whereby people weigh the costs and benefits of participation’ (Klandermans, 1986a: 583).

Literature review and hypotheses

In this section, the empirical literature is reviewed and hypotheses are formulated. They are based on all three theories of union participation as we are drawing on both interactionist and cost–benefit theories, as well as the voice–loyalty perspective, to understand the causes of participation.

Union satisfaction

Fiorito et al. (1988) define union satisfaction as a function of the discrepancy between expectations and perceived outcomes on three major issues: bread and butter (wages, benefits and job security); quality of work life (giving members a say at the workplace and making the job more interesting); and leadership–membership communication (workers’ perceived influence in the union).

The relationship between union satisfaction and outcome variables such as union participation are not unproblematic. Gordon et al. (1980: 493) suggested that:

dissatisfaction with the union may produce opposite participative tendencies ranging from activism to an unwillingness to lift a finger on behalf of the union. Likewise, satisfaction with the union may serve as a catalyst that spurs membership participation or creates complacency.

In a study of professional engineers, Glick et al. (1977) found that union satisfaction and willingness to participate were positively correlated for those with strong ‘higher-order needs’ (for decision-making and growth), but negatively correlated among those for whom these needs were relatively weak. A study by Chacko (1985), using the 1977 Quality of Employment Survey, applied discriminant analysis to examine the relationship between union satisfaction and participation, finding an inverse relationship between the two variables. He suggested that dissatisfaction with the union heightens participation in union activities, such that union satisfaction may be more appropriately regarded as a manifestation of union democracy than of member support. Similarly, a study by Kuruvilla and Fiorito (1994) found that union satisfaction had a negative relationship with willingness to work for the union, which is a precursor to actual participation. The result implies that unions that meet the expectations of their members may be less likely to gain participation from their members.

Exit–voice theory suggests that the relationship between union satisfaction and participation will be negative. Hirschman (1970) argued that voice was a response to dissatisfaction and represented an attempt to improve organizational functioning. In other work-related settings, voice – or speaking up – has been associated with constructive efforts to challenge the status quo and improve workplace issues (LePine and Van Dyne, 1998). We would therefore expect dissatisfied union members to voice their dissatisfaction through greater participation in the union. Therefore, we hypothesize the following:

Hypothesis 1: Union satisfaction (i.e. delivers extrinsic benefits) will have a negative impact on union participation.

Procedural justice

Although the focus of this study is on testing frustration–aggression theory with the union as the point of interest, the antecedents of participation can also be examined in terms of the job. Frustration–aggression or exit–voice theory predicts that dissatisfaction with specific aspects of the job will have an impact on voice. Recently, workplace justice has been found by Johnson and Jarley (2004) to be a significant determinant of union participation. Further, a study by Redman and Snape (2004) found that pay inequity (as a justice perception in support of the frustration–aggression hypothesis) predicted participation. We expect that procedural justice – defined as the fairness of procedures designed to determine the outcomes an employee receives (Moorman, 1991) – will be inversely related to union participation:

Hypothesis 2: Procedural justice will have a negative impact on union participation.

Industrial relations climate

Industrial relations climate measures the extent to which the relationship between the union and management is perceived to be harmonious, productive and open (Dastmalchian et al., 1989). Kuruvilla et al. (1990) included union–management relations as a control variable in their study of union participation in Japan. They found the quality of relations between management and union to have a positive effect on participation. Two more recent studies by Fuller and Hester (1998) and Redman and Snape (2004), however, found an inverse relationship between industrial relations climate and participation.

In this study, it is expected that industrial relations climate will have a negative influence on participation. This is consistent with the frustration–aggression hypothesis as it relates to dissatisfaction with the employing organization (Redman and Snape, 2004). Poorer industrial relations climates will be consistent with higher levels of participation:

Hypothesis 3: Industrial relations climate will have a negative impact on union participation.

Family socialization

Spinrad (1960: 241) argued that individuals with a family background in activism would tend to be more active in the union. A number of studies have found support for a positive relationship (Dekker et al., 1998; Kelloway and Watts, 1994; Nicholson et al., 1981).

A study by Barling and colleagues (1991) examined a model of pre-employment predictors of union attitudes on young high school and university students. They hypothesized that as children observed their mothers’ and fathers’ participation in union-related activities, or listened to their parents talk about unions, their perceptions of their parents’ attitudes to unions develop. These perceptions affect children's attitudes towards unions (Barling et al., 1991). They found that young adults’ perceptions of their parents’ attitude to unions significantly predicted their own attitude; however, they did not find a relationship between perceived level of participation of parents and attitudes (although perceived level of participation of parents had a significant impact on perceived attitude of parents).

In this article, it is argued that family socialization is a social contextual factor and, hence, is relevant for interactionist theories. The following hypothesis was developed:

Hypothesis 4: Family socialization will have a positive impact on union participation.

Individualistic orientation

Spinrad (1960: 243) has stated that ‘the active unionist, then views his work group or the “working class” or both as a significant reference group. He emphasizes collective rather than individual efforts for improvement’. Wagner (1995: 153) asserted that ‘individualism–collectivism is an analytical dimension that captures the relative importance people accord to personal interests and to shared pursuits’. In other words, individualism is where personal interests are of greater priority than the group. An individualist, therefore, acts as though he/she defines their self as an entity consisting of a single person, whereas a collectivist is defined by an entity extending beyond the individual to the group (Wagner, 1995).

Numerous authors (Bassett and Cave, 1993; Brown, 1990; Cave, 1994; Storey and Bacon, 1993) have argued that there has been a general move from collectivist behaviour and attitudes to individualism. Brown (1990) stated that it was due to the increasing affluence of workers in the late 20th century. In the late 19th century, the working class was more collectivist because of common harsh conditions, such as poverty, job insecurity and unemployment, which were shared in communities. As communities have been broken up with the increased mobility of workers, this has led to a greater propensity of workers to perceive education and training as the main means for self-improvement, rather than through collective means.

Interactionist theories, such as mobilization theory, propose that union participation requires identification with group goals. Kelly and Kelly (1994) have argued that the impact of group identification will be greatest among groups characterized by a collectivist orientation. Social change strategies are therefore most likely among those individuals who identify strongly with the social group (Kelly and Kelly, 1994). Individuals who identify weakly with the group are expected to engage in individual action to improve their position at work. Therefore, we expect the following:

Hypothesis 5: Individualism will have a negative impact on union participation.

Union instrumentality

Newton and Shore (1992: 279) argue that instrumentality is ‘a calculative or utilitarian relationship with unions and is based on a cognitive assessment of the costs and benefits associated with union representation’. The relationship between participation and instrumentality is explained most readily through a rational choice framework of analysis. Strauss (1977: 222–223) argued that ‘for members to participate they must feel that such action will result in some sort of payoff’. Very early studies have found the relationship between instrumentality and participation to be positive (Anderson, 1979; Kuruvilla et al., 1990). Griffin and Benson (1987), however, found union performance to be an insignificant predictor of three measures of participation (industrial activities, union courses or union literature) and a study by Heshizer et al. (1991) found that union effectiveness (union instrumentality) had a negative effect on participation. Chacko (1985) found that union performance was the most important factor determining union participation. This leads to the following hypothesis:

Hypothesis 6: Extrinsic instrumentality will have a positive impact on union participation.

Union loyalty

Union loyalty is proposed to be an important variable that encourages voice in the exit–voice model. Exit and voice may coexist in an organization, given the existence of loyalty. Hirschman (1970) stated that ‘as a rule, then, loyalty holds exit at bay and activates voice’ (Hirschman, 1970: 78). That is, loyalty should encourage union participation or voice. Loyalty is functional because it impedes the tendency of the quality-conscious customers or members to be the first to exit. These individuals have some calculative belief that they can change the organization from within. Loyal members are also more likely to invest the time and effort to voice concerns and issues within the union and to present ideas and suggestions for change. Thus, loyal members who are dissatisfied with the union’s performance or activities could be expected to voice that dissatisfaction through participation or ‘speaking up’ within the union. As a consequence, ‘loyalty, far from being irrational, can serve the socially useful purpose of preventing deterioration from becoming cumulative, as it so often does when there is no barrier to exit’ (Hirschman, 1970: 79). Loyalty serves the same function as switching costs by raising the cost of exit. Further, the empirical literature has consistently found a positive relationship between union commitment and participation or voice (Fullagar et al., 2004; Goeddeke and Kammeyer-Mueller, 2010; Tetrick et al., 2007). Given exit–voice theory, our hypothesis is the following:

Hypothesis 7: Union loyalty will have a positive impact on union participation.

Leader responsiveness

The availability and style of the union representative (such as ability to communicate, individual consideration and accessibility) are crucial for both attitude formation and participation in the union (Barling et al., 1992). Perline and Lorenz (1970: 437) have stated that shop stewards are especially important because ‘many members identify with the stewards as the only liaison with the union organization’. Kelloway and Barling (1993) examined the effect of member evaluations of their shop stewards’ style, and argued that members’ perception of the transformational leadership qualities of their stewards would have a direct effect on both loyalty and participation. Transformational leadership style (Bass, 1985) involves pride in the organisation, stimulating followers and providing an encouraging workplace (Kelloway and Barling, 1993). In two samples, transformational leadership had a direct impact on participation. On the basis of this research, they argued that ‘interventions at the shop steward level would have a direct effect on increasing members’ part in union activities’ (Kelloway and Barling, 1993: 276). A study by Metochi (2002) also found a positive relationship between union leadership and participation. Further, a study by Hammer and colleagues (2009) found that union leadership was related to a number of important union outcomes. Based on the literature, we hypothesize the following:

Hypothesis 8: Leader responsiveness will have a positive impact on union participation.

Interactions

We propose two interactions. First, drawing on exit–voice theory (Hirschman, 1970), voice is likely to occur where loyalty is higher in an organization. Exit–voice theory suggests that loyalty inhibits exits and enhances voice because individuals want to change the organization rather than leave. Therefore, we expect that union loyalty will moderate the relationship between union satisfaction and participation such that members who are dissatisfied with the union will be more likely to participate when they are loyal to the union. We propose the following:

Hypothesis 9: Union loyalty will moderate the effect of union satisfaction such that low union satisfaction will be associated with greater participation when union loyalty is high.

Further, it is expected that leader responsiveness will also moderate the union satisfaction and participation relationship. This is based on the mobilization literature (interactionist theory) (Kelly, 1998), which argues that leaders are necessary to trigger perceived injustices at the workplace. This can then be extended to argue that union satisfaction will increase participation where there is high union leader responsiveness. Union satisfaction is expected to be high in this context as mobilization theory predicts that mobilization will occur where the union is perceived to be a good agent for change. This is more likely where there is high union satisfaction:

Hypothesis 10: Union leader responsiveness will moderate the effect of union satisfaction such that high union satisfaction will be associated with greater participation when union leader responsiveness is high.

Context of the study

The study was conducted in an organization employing approximately 20,000 full-time equivalent staff in Australia. Two-thirds of employees were members of the union. The union is one of the largest in Australia, with a membership of over 90,000 drawn from the insurance and banking sectors. It has branches Australia-wide, with national offices in Melbourne and Sydney. At the time of the study, the union had approximately 55% membership coverage in the banking sector and just under 25% in the insurance sector. It should also be noted that the banking sector is characterized by low labour turnover rates.

Historically, the union has had a harmonious relationship with the bank (Griffin, 1985). Banking employees have generally been considered ‘industrially conservative’ and reluctant to take industrial action (Davis, 1987; Hill, 1982). However, in the late 1990s the union began to question the cooperative character of its relationship with the banks and, at the same time, began to move from a servicing union model to an organizing model (Holland, 1999). The union began to open up communication channels with its members and run focus groups and surveys in an effort to better understand their members’ needs. Union stewards were also trained extensively within an organizing framework. Around two-thirds of the workplaces in the organization had union representatives. In their enterprise bargaining campaigns with the bank, the union attempted to mobilize members and build a better capacity for industrial action. This was difficult, however, because the bank was considered by employees to be a ‘reasonable’ employer and the union and the bank had had a long-standing cooperative relationship.

Qualitative data collected from interviews suggested that there was a range of ways in which members could participate in the union. One of the main forms of participation was attendance at union meetings. Members had the opportunity to attend a number of different types of meetings at the union. These included the Biennial Conference and the Branch Committee of Management, the latter of which represented all banks and insurance companies and reported on issues such as job security, occupational health and safety, and women’s issues. Another way in which members participated in the union was through sub-branch meetings based on enterprise. Information from the sub-branch committee was transmitted through to organizers, the Branch Secretary and the Branch Committee of Management. Other meetings that members attended and that provided opportunities for participation and voice tended to be issue-based in the workplace or revolve around negotiations during enterprise bargaining rounds.

Research methods

Research setting and data collection

This research was undertaken at an international bank based in Australia, which employed nearly 20,000 full-time equivalent staff across some 1600 branches and business outlets.

Qualitative data were collected for the contextual part of this study. Ten years of archival material consisting of meeting minutes, communications and policy documents were examined. Union meetings were also attended. In addition, interviews were undertaken with key informants and union representatives.

The quantitative data were collected in three waves over a three-year period. The first questionnaire of the study was distributed to 5978 employees in Time 1 via the internal mail of the employing organization. Each survey was coded with identification numbers in order to follow up respondents. Employees were made aware of the fact that although the information contained in the survey was confidential, it was not anonymous. Employees were also informed that their participation in the study was voluntary.

Of the 5978 surveys that were sent to the 624 branches, 3648 surveys were returned. Surveys that did not contain identification numbers and employees who completed their survey twice were discarded from the analysis. The useable sample was therefore reduced to 3554 employees, representing a 59.5% response rate.

The second wave of data was collected 12 months later (Time 2) and involved a similar procedure to the first mail out. Of the 3554 respondents who provided useable responses in time 1, 444 had left the organization. The second survey was sent to the remaining 3110 employees. Of these, 2388 surveys were returned. The data were then matched to the first wave, reducing the useable responses to 2309 (useable response rate of 74.2%).

The third and final stage of the data collection occurred at Time 3 (12 months following Time 2). The 2106 respondents who had participated in both the first and second stages of the project and were still employed by the organization (203 employees left) were sent a third and final survey. Of these, 1681 employees responded, with a subset of 1606 useable surveys. The useable response rate for the third and final survey was 76.3%.

Sample

The sample consisted of 874 union members (following the deletion of data for missing cases), of which 54% were female and 80% were full-time. The average tenure in the union (at Time 3) was 12.82 years (SD = 7.92). The final union sample of 874 was representative of the union percentage of the bank (χ²(1) = 2.23; p > .05). The insignificance of this finding indicates that the sample is not significantly different from the population.

Measurement

Participants responded to items on a five-point Likert-type scale (1 = strongly disagree to 5 = strongly agree), unless otherwise noted. Established scales were used where possible. The times at which the data were collected are indicated in the following. The outcome variable (union participation) was collected at Time 3; union loyalty and responsiveness were collected at Time 2; and the antecedents of union participation were collected at Time 1. The survey utilized longitudinal panel data.

Time 3

To measure union participation, this study utilized a measure from McLean Parks et al. (1995). The study utilized five measures (coded as ‘1’ if the member had participated and coded as ‘0’ if the member had not participated in the activity), including ‘Within the past year did you speak up at a union meeting?’. This measure was taken at Time 3. In order to address the non-linearity of the data, the log for participation was taken.

Time 2

Union loyalty is defined as the degree of pride and instrumentality in the union. The scale comprised three questions, such as ‘I feel a sense of pride in the union’ (Gordon et al., 1980).

Union leader responsiveness is defined as the degree to which the union representative at the workplace is sympathetic to the needs and concerns of individual union members. The scale is a shortened (two-item) version of that by Sverke and Sjoberg (1994). The scale consisted of questions such as ‘The union representatives at my workplace pay attention to demands put forward by members’.

Given the causal ordering with union participation (Bamberger et al., 1999), we measure both loyalty and leader responsiveness at Time 2.

Time 1

Union satisfaction is defined as the degree to which members are satisfied with their union. It is an adapted scale from Jarley et al. (1990) and contains six measures, such as ‘How satisfied are you with the union in terms of listening to the concerns of other members?’.

Procedural justice, defined as the fairness of procedures designed to determine the outcomes an employee receives, was operationalized by three items from Price and Mueller (1981, 1986) and included ‘The [bank] has procedures to ensure that decisions affecting staff are made with consistency’.

Industrial relations climate is defined as the extent to which organization–union relations are harmonious, productive and open (communication) (Dastmalchian et al., 1989). The modified scale that was used to measure this construct is well-established (Dastmalchian et al., 1989). Six items were used, including questions such as ‘Union and management have respect for each other’s goals’.

Family socialization is defined as an individual’s perceptions of their parents’ attitude to and participation in unions. The variable was operationalized using a single item that was a cumulative index of eight dichotomously measured items. These items were coded as ‘1’ if the respondent was aware that their mother and/or father had been involved in union activity (e.g. had ever been a union representative) and ‘0’ if they were uncertain or if the parents had not engaged in union activity. The range for this scale was therefore between 0 and 8. The scale is modified from Barling et al. (1991).

Individualism is defined as the degree to which members have an individualistic orientation to work. The measure is a three-item modified scale by Kelly and Kelly (1994). The scale includes questions such as ‘Only those who depend on themselves at work get ahead’.

Extrinsic instrumentality is defined as the calculative (costs versus benefits) relationship of being a union member. The variable is measured by three items and is a modified version of Youngblood et al.’s (1984) scale. The extrinsic instrumentality measure had questions such as ‘The union has improved my working conditions (e.g. wages and benefits)’.

Two control variables were included in the analysis. Sex (1 = female; 0 = male) and Full-time (1 = full-time; 0 = part-time) were coded as dichotomous variables.

Analysis

We conducted confirmatory factor analysis on the multiple-item measures using LISREL VIII (Jöreskog and Sörbom, 2006). The hypothesized nine-factor model (i.e. union participation, union loyalty, leader responsiveness, union satisfaction, procedural justice, industrial relations climate, family socialization, individualism and extrinsic instrumentality) fitted the data significantly better than both the null-(Δχ²(76) = 11665.94; p < .001) and the one-factor (Δχ²(36) = 2833.79; p < .001) models. The parameter estimates (factor loadings) of the best fitting nine-factor model were all significant (p < .05) and ranged from .44 to .68. The measurement model demonstrated acceptable fit: goodness of fit (GFI) = .92, comparative fit index (CFI) = .94 and root mean square error of approximation (RMSEA) = .04 (Hu and Bentler, 1998, 1999).

To test our hypotheses, we used hierarchical regression analyses and interaction tests were undertaken, as recommended by Aiken and West (1991). In order to address multi-collinearity we centred the interaction variables.

Results

Table 1 reports the descriptives and the correlations for all variables.¹ The means for union satisfaction, industrial relations climate, union loyalty and union leader responsiveness are relatively high, with all values greater than 3 on five-point scales. Participation is significantly correlated with procedural justice (r = −.13; p < .05), industrial relations climate (r = −.07; p < .05), family socialization (r = .18; p < .05), extrinsic instrumentality (r = .15; p < .05), union loyalty (r = .19; p < .05) and leader responsiveness (r = .21; p < .05).

Table 1.

Descriptives, correlations and reliabilities.

Variable	Mean	SD	1	2	3	4	5	6	7	8	9	10	11
Time 3
1. Participation (log)	1.18	.38	.63
Time 1
2. Union satisfaction	3.22	.62	.03	.84
3. Procedural justice	2.63	.79	−.13	.44	.83
4. Industrial relations climate	3.37	.52	−.07	.51	.41	.81
5. Family socialization	1.64	1.77	.18	.09	–.03	–.02	.74
6. Individualism	2.99	.91	−.01	.17	.26	.20	.01	.79
7. Extrinsic instrumentality	3.35	.64	.15	.65	.24	.40	.11	.14	.62
8. Sex	.54	.50	.01	.29	.18	.13	.05	–.10	.26	–
9. Full-time	.80	.40	.02	−.11	−.12	−.09	−.03	.01	−.11	−.43	–
Time 2
10. Union loyalty	3.18	.76	.19	.49	.20	.28	.15	.01	.54	.26	−.14	.87
11. Leader responsiveness	3.00	.90	.21	.34	.16	.18	.09	.08	.29	.07	−.09	.36	.93

Notes: n = 874; p < .05 for all absolute correlations greater than or equal to .06.

Table 2 reports on the hierarchical regression analyses. Step 1 indicates the results for the independent variable and control variables without the two interactions. Union participation or voice is significantly determined by procedural justice (β = −.133; p < .001), industrial relations climate (β = −.088; p < .05), family socialization (β = .130; p < .001), extrinsic instrumentality (β = .142; p < .01), union loyalty (β = .128; p < .01) and leader responsiveness (β = .185; p < .001). The adjusted R-squared was .119. That is, there is support for Hypotheses 2, 3, 4, 6, 7 and 8.

Table 2.

Regression results for union participation.

Variables	Step 1	Step 2	Step 3
Time 1
Union satisfaction	−.093	−.088	−.091
Procedural justice	−.133***	−.136***	−.133***
Industrial relations climate	−.088*	−.088*	−.088*
Family socialization	.130***	.127***	.129***
Individualism	.020	.017	.023
Extrinsic instrumentality	.142**	.146**	.150**
Sex	.001	−.005	−.007
Full-time	.043	.040	.044
Time 2
Union loyalty	.128**	.137**	.131**
Leader responsiveness	.185***	.185**	.190***
Interactions
Union satisfaction × union loyalty	.060
Union satisfaction × leader responsiveness	.073*
R ²	.129	.132	.134
Adjusted R²	.119***	.121***	.123***
Δ in adjusted R²	.119***	.002	.004*

Notes: n = 874; *p < .05; **p < .01; ***p < .001.

In step 2, the interaction between union satisfaction and union loyalty was entered and found to be non-significant (β = .060; p > .05). Finally, in step 3 the union satisfaction–leader responsiveness interaction was positive and significant (β = .073; p < .05), with an accompanying significant increase in the adjusted R-squared. To interpret this result, we plotted the interaction result using the process recommended by Aiken and West (1991) and present it in Figure 1. Consistent with Hypothesis 10, we found that responsive leadership did moderate the relationship between union satisfaction and union participation such that high union satisfaction was associated with higher participation when leaders were perceived to be highly responsive.

Discussion

The aim of this article was primarily to test whether frustration–aggression theories such as exit–voice explain voice or participation within trade unions. We did this by controlling for variables that measure components of other theories, such as interactionist and cost–benefit or rational choice theory. This is a novel approach as, thus far, unions have been perceived as agents of voice in employing organizations rather than the organization in question. Although participation may be motivated by other considerations, in this article, we have argued that union participation is an appropriate proxy for voice, particularly as we control for other reasons why members may participate, such as ideology (union loyalty).

In exit–voice theory, dissatisfied union members are expected to voice their dissent within the union by participating. This is especially the case where there is high union loyalty as this mechanism could be expected to encourage voice. This study, however, found no evidence that the exit–voice model explains voice within trade unions. Indeed, the interaction between union satisfaction and loyalty was not significant. Instead, this study found that members who reported that their union representatives at the workplace were responsive to their needs and concerns were more willing to participate in the union, even when they were highly satisfied with it.

Mobilization theory (interactionist theory) suggests that active, involved and responsive union leaders are necessary to encourage collective action. A responsive leadership can act upon perceptions of injustice in the workplace and encourage greater levels of participation (Kelly, 1998). This study lends some support to this theory. Interestingly, high levels of union satisfaction combined with leader responsiveness lead to individuals participating in their union. Framed within a cost–benefit analysis, this may be seen to reduce the psychological cost of participation and increase its benefits, particularly in the context of good leadership.

The study indicates that interactionist theory, where participation is viewed as the result of the social context of members, may have greater validity in explaining union participation than frustration–aggression theories (Kelly, 1998; Klandermans, 1986). Further, there is support for the rational choice or cost–benefit theory as extrinsic instrumentality was significant, indicating that individuals are more likely to express voice and participate in the union when they believe that their actions will have some pay-off.

This study is not without its limitations. The data were drawn from a single organization (albeit with several hundred branches), which restricts the generalizability of the findings. However, it should be noted that we employed a longitudinal design across three time periods, which limits the probability of common method bias (Podsakoff et al., 2003). We use participation as a proxy for voice. Although it is logical to assume that participation is voice (see Leicht, 1989), future research could usefully test this model with more direct measures of union voice.

In terms of practical implications, this article suggests that voice can be activated by the interaction of highly responsive union representatives and high levels of satisfaction. Unions should focus on creating effective leaders who are able to heighten perceptions of injustice at the workplace in order to increase voice within trade unions. They should also focus on improving union satisfaction. Some early literature suggests that union satisfaction is determined by performance of the union (Fiorito et al., 1988). Future research needs to revisit the causes of union satisfaction at the workplace. This is particularly important in the context of the debate over whether unions should apply the servicing or organizing model, where the impact of these models on union satisfaction is uncertain. Other practical implications are that unions should promote collectivism at the workplace and improve the benefits of being a union member, therefore activating voice. There are few practical policy implications regarding socialization as this is external to union control except that unions may promote family activities and raise awareness of family history in attempting to engage members to voice issues.

In conclusion, this article provides us with insights into the determinants of voice within trade unions. Although we failed to find evidence to support the frustration–aggression (exit–voice) theory of participation, our findings did lend weight to the importance of interactionist and cost–benefit theories. Within this context, future research is warranted to examine the pivotal role of active shop-floor representatives in improving voice, as well as union renewal.

Footnotes

Funding

This research is the result of an ARC grant awarded to the second and third authors.

Note

Biographical notes

Donna Buttigieg is a Professor at Victoria University in the area of Industrial Relations. She is Discipline leader of the Management group at Victoria University. Prior to this she worked at Oxford University, Melbourne University and Monash University. She also worked for a research think tank for three years with the State Government. Professor Donna Buttigieg's research interests are trade unions, OHS, vulnerable workers and strategic HRM. She has published widely in leading journals, such as the Journal of Applied Psychology, Journal of Management Studies, The British Journal of HRM, Work Employment and Society and Human Relations. She has also received$100,000 worth of funding in the past year. She is currently working on projects looking at union citizenship behaviour, union effectiveness and women in leadership.

Stephen Deery is Professor of Human Resource Management at Kings College, University of London. Prior to this he worked at Melbourne University and was Head of School for a number of years. His principal research activities lie in the areas of call centres, organisational performance, unions, employee turnover and absenteeism. He is also involved in the study of the psychological contract and employee stress as well as harassment and bullying. He has published in leading journals such as HRM, Journal of Applied Psychology, Industrial and Labor Relations Review, British Journal of Industrial Relations and Journal of Management Studies. He was also Chief Editor of Human Relations until recently. He has received hundreds of thousands in grant funding. Professor Deery was also joint recipient of Best Paper award from the Academy of Management twice.

Professor Roderick Dale (Rick) Iverson passed away suddenly on May 3rd 2012. He was a Professor at the Beedie School of Business at Simon Fraser University. His previous appointments were at Melbourne University and Iowa State University. Professor Iverson was the recipient of a number of teaching and research awards from Simon Fraser University and best paper awards at the Academy of Management. He has published in leading journals such as the Journal of Applied Psychology, Journal of Management Studies, Human Relations, HRM, British Journal of Industrial Relations, Academy of Management Journal, Industrial and Labor Relations Review and the Journal of Management. Overall, he published over 50 journal articles. He received a significant amount of funding in the form of ARC grants and Social Science and Humanities grants and before he died was setting up funding for an OHS Research Centre at Simon Fraser University.

References

Aiken

West

(1991) Multiple Regression: Testing and Interpreting Interactions, California, CA: SAGE Publications.

Anderson

(1979) Local union participation: A re-examination. Industrial Relations 18(1): 18–31.

Bamberger

Kluger

Suchard

(1999) The antecedents and consequences of union commitment: A meta analysis. Academy of Management Journal 42(3): 304–318.

Barling

Fullagar

Kelloway

(1992) The Union and Its Members: A Psychological Approach, New York, NY: Oxford University Press.

Barling

Kelloway

Bremermann

(1991) Preemployment predictors of union attitudes: The role of family socialization and work beliefs. Journal of Applied Psychology 76(2): 725–731.

Bass

(1985) Leadership and Performance Beyond Expectations, New York, NY: Free Press.

Bassett

Cave

(1993) All for One: The Future of the Unions, Fabian Pamphlet 559, London: Fabian Society.

Brown

(1990) The counter-revolution of our time. Industrial Relations 29(1): 1–14.

Cave

(1994) Managing Change in the Workplace, London: Kogan Page.

10.

Chacko

(1985) Members’ participation in union activities: Perceptions of union priorities, performance and satisfaction. Journal of Labor Research 6(4): 363–373.

11.

Cortina

(1993) What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology 78(1): 98–104.

12.

Dastmalchian

Blyton

Adamson

(1989) Industrial relations climate: Testing a construct. Journal of Occupational Psychology 62(1): 21–32.

13.

Davis

(1987) Democracy in Australian Unions: A Comparative Study of Six Unions, North Sydney: Allen and Unwin Australia.

14.

Dekker

Greenberg

Barling

(1998) Predicting union attitudes in student part-time workers. Canadian Journal of Behavioral Science 30(1): 49–55.

15.

Dundon

Wilkinson

Marchington

Ackers

(2004) The meaning and purpose of employee voice. International Journal of Human Resource Management 15: 1149–1170.

16.

Fairbrother

Waddington

(1990) The politics of trade unionism: Evidence, policy and theory. Capital and Class 41(2): 15–56.

17.

Fantasia

(1988) Cultures of Solidarity: Consciousness, Action, and Contemporary American Workers, California, CA: University of California Press.

18.

Fiorito

Gallagher

Fukami

(1988) Satisfaction with union representation. Industrial and Labor Relations Review 1(2): 294–307.

19.

Fosh

(1993) Membership participation in workplace unionism: The possibility of union renewal. British Journal of Industrial Relations 31(4): 577–592.

20.

Freeman

Medoff

(1984) What Do Unions Do?, New York, NY: Basic Books Inc.

21.

Fullagar

Barling

(1989) A longitudinal test of a model of the antecedents and consequences of union loyalty. Journal of Applied Psychology 74(2): 213–227.

22.

Fullagar

Gallagher

Clark

Carroll

(2004) Union commitment and participation: A 10-year longitudinal study. Journal of Applied Psychology 89(4): 730–737.

23.

Fuller

Hester

(1998) The effect of labor relations climate on the union participation process. Journal of Labor Research 19(1): 173–187.

24.

Gamson

(1992) Talking Politics, Cambridge: Cambridge University Press.

25.

Gamson

(1995) Constructing social protest. In: Johnston

Klandermans

(eds) Social Movements and Culture, London: UCL Press, pp. 85–106.

26.

Glick

Mirvis

Harder

(1977) Union satisfaction and participation. Industrial Relations 16(2): 145–151.

27.

Goeddeke

Kammeyer-Mueller

(2010) Perceived support in a dual organizational environment: Union participation in a university setting. Journal of Organizational Behavior 31(1): 65–83.

28.

Goldthorpe

Lockwood

Bechhofer

Platt

(1968) The Affluent Worker: Industrial Attitudes and Behaviour, Cambridge: Cambridge University Press.

29.

Gordon

Philpot

Burt

(1980) Commitment to the union: Development of a measure and an examination of its correlates. Journal of Applied Psychology 65(4): 479–499.

30.

Griffin G (1985) White-Collar Militancy: The Australian Banking and Insurance Union. Sydney: Croom Helm Australia.

31.

Griffin

Benson

(1987) Factors related to membership participation in public sector trade union activities. Australian Journal of Management 12(2): 263–276.

32.

Hammer

Bayazit

Wazeter

(2009) Union leadership and member attitudes: A multi-level analysis. Journal of Applied Psychology 94(2): 392–410.

33.

Heshizer

Martin

Wiener

(1991) Normative commitment and instrumental attachment as intervening variables in the prediction of union participation. Journal of Applied Behavioral Science 27(4): 532–549.

34.

Hill

(1982) From Subservience to Strike: Industrial Relations in the Banking Industry, St Lucia: University of Queensland Press.

35.

Hirschman

(1970) Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States, Cambridge, MA: Harvard University Press.

36.

Holland

(1999) Organising works: Meeting the challenge of declining trade union membership. International Employment Relations Review 5(1): 63–74.

37.

Bentler

(1998) Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods 3(4): 424–453.

38.

Bentler

(1999) Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling 6(4): 1–55.

39.

Hurd

Milkman

Turner

(2003) Reviving the American labour movement: Institutions and mobilization. European Journal of Industrial Relations 9(9): 99–117.

40.

Hyman

(1984) The sickness of British trade unionism: Is there a cure? In: Hyman

(eds) The Political Economy of Industrial Relations: Theory and Practice in a Cold Climate, London: Macmillan, pp. 166–187.

41.

Jarley

Kuruvilla

Casteel

(1990) Member–union relations and union satisfaction. Industrial Relations 29(1): 128–134.

42.

Johnson

Jarley

(2004) Justice and union participation: An extension and test of mobilization theory. British Journal of Industrial Relations 42(3): 543–562.

43.

Jöreskog

Sörbom

(2006) Interactive LISREL (8.80), Chicago, IL: Scientific Software International.

44.

Kelloway

Barling

(1993) Members’ participation in local union activities: Measurement, prediction and replication. Journal of Applied Psychology 78(2): 262–279.

45.

Kelloway

Watts

(1994) Preemployment predictors of union attitudes: Replication and extension. Journal of Applied Psychology 79(4): 631–634.

46.

Kelly

(1994) Who gets involved in collective action? Social psychological determinants of individual participation in trade unions. Human Relations 47(1): 63–88.

47.

Kelly

(1998) Rethinking Industrial Relations: Mobilization, Collectivism and Long Waves, London: Routledge.

48.

Klandermans

(1986a) Perceived costs and benefits of participation in union action. Personnel Psychology 39(2): 379–397.

49.

Klandermans

(1986b) Psychology and trade union participation: Joining, acting, quitting. Journal of Occupational Psychology 59(2): 189–204.

50.

Kuruvilla

Fiorito

(1994) Who will help? Willingness to work for the union. Relations Industrielles 49(3): 548–563.

51.

Kuruvilla

Gallagher

Fiorito

Wakabayashi

(1990) Union participation in Japan: Do Western theories apply? Industrial and Labor Relations Review 43(4): 374–389.

52.

Leicht

(1989) Unions, plants, jobs and workers: An analysis of union satisfaction and participation. The Sociological Quarterly 30(2): 331–362.

53.

LePine

Van Dyne

(1998) Predicting voice behaviour in work groups. Journal of Applied Psychology 83(6): 853–868.

54.

Luchak

(2003) What kind of voice do loyal employees use? British Journal of Industrial Relations 41(1): 115–134.

55.

McAdam

(1988) Micromobilization contexts and recruitment to activism. International Social Movement Research 1(1): 125–154.

56.

McLean Parks

Gallagher

Fullagar

(1995) Operationalizing the outcomes of union participation: The dimensionality of participation. Journal of Organizational Behavior 16(6): 533–555.

57.

Metochi

(2002) The influence of leadership and member attitudes in understanding the nature of union participation. British Journal of Industrial Relations 40(1): 87–111.

58.

Moorman

(1991) Relationship between organizational justice and organizational citizenship behaviours: Do fairness perceptions influence employee citizenship? Journal of Applied Psychology 76(6): 845–855.

59.

Newton

Shore

(1992) A model of union membership: Instrumentality, commitment, and opposition. Academy of Management Review 17(2): 275–298.

60.

Nicholson

Ursell

Lubbock

(1981) Membership participation in a white collar union. Industrial Relations 20(2): 162–178.

61.

Olson

(1965) The Logic of Collective Action: Public Goods and the Theory of Groups, Cambridge, MA: Harvard University Press.

62.

Perline

Lorenz

(1970) Factors influencing member participation in trade union activities. The American Journal of Economics and Sociology 29(4): 425–437.

63.

Podsakoff

MacKenzie

Lee

Podsakoff

(2003) Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology 88(5): 879–903.

64.

Price

Mueller

(1981) Absenteeism and Turnover of Hospital Employees, Greenwich, CT: JAI Press.

65.

Price

Mueller

(1986) Professional Turnover: The Case of Nurses, New York, NY: SP Medical and Scientific.

66.

Redman

Snape

(2004) Kindling activism? Union commitment and participation in the UK fire service. Human Relations 57(7): 845–869.

67.

Spinrad

(1960) Correlates of trade union participation: A summary of the literature. American Sociological Review 25(2): 237–244.

68.

Storey

Bacon

(1993) Individualism and collectivism: Into the 1990s. International Journal of Human Resource Management 4(3): 665–684.

69.

Strauss

(1977) Union government in the U.S.: Research past and future. Industrial Relations 16(2): 215–242.

70.

Strauss

(2006) Worker participation: Some under-considered issues. Industrial Relations 45(4): 778–803.

71.

Sverke

Sjoberg

(1994) Dual commitment to company and union in Sweden: An examination of predictors and taxonomic split methods. Economic and Industrial Democracy 15(4): 531–564.

72.

Tetrick

Shore

McClurg

Vandenberg

(2007) A model of union participation: The impact of perceived union support, union instrumentality, and union loyalty. Journal of Applied Psychology 92(3): 820–828.

73.

Tilly

(1978) From Mobilization to Revolution, Philippines: Addison-Wesley Publishing.

74.

Van de Vall

(1970) Labor Organizations, London: Cambridge University Press.

75.

Wagner

III (1995) Studies of individualism–collectivism: Effects on cooperation in groups. Academy of Management Journal 36(1): 152–172.

76.

Youngblood

DeNisi

Molleston

Mobley

(1984) The impact of work environment, instrumentality beliefs, perceived labor union image, and subjective norms on union voting intentions. Academy of Management Journal 27(3): 576–590.