Corps de l’article

Introduction: The revitalizing promise of union newcomers

Two features largely characterize membership of mainstream trade unions in most western European countries—indeed, across the global North—and in turn influence union members’ willingness to pay membership dues and act collectively. The first is that most people tend to join a union due to instrumental motives and less for ideological reasons (Ebbinghaus et al., 2011; Scheuer, 2011; Waddington, 2015). Instrumental motivation relates to the security provided by unions, to different degrees, at workplace level and across economic industries and countries, which is shaped by institutional frameworks. From a historical perspective, the insurance function of membership, shielding workers from (labour market) risks, has been a core feature of unions from the outset. Some scholars claim, however, that when unions merely provide a range of individual services, especially if they are external to the workplace, such a calculative member-union relationship “is less likely to retain members than is a strategy that emphasizes a broader appeal” (Snape and Redman, 2004: 869). The second feature, underscoring the dominance of this kind of service-oriented unionism, is that a passive interpretation of membership prevails today (Gumbrell-McCormick and Hyman, 2013: 6). Although members are willing to pay dues, they may be apathetic towards the union itself, leading to the danger of unions no longer having the active support necessary to be “fit for fighting” (Bergene and Mamelund, 2017) against employers’ union-busting tactics, or against political authorities taking anti-union measures or fostering the re-commodification of labour.

A sociological understanding of unions entails the idea that, besides the external environment, agency-based, internal dynamics can also be a source of organizational change, and thus trigger a shift away from a service-oriented unionism (Heery, 2005). Burgeoning literature that studies under-represented groups in unions suggests that young people might be such a source, with their engagement and participation a possible driver for union transformation, injecting creative energy, fresh ideas and imagination into the labour movement (Hodder and Kretsos, 2015; Tapia and Turner, 2018; Vandaele, 2018). While specific personal characteristics, skills and knowledge of young people may make them stand out from their older counterparts, age or cohort differences should not be exaggerated. Although the labour market experience of young and older new members likely differ, both share the characteristic of being new to the union since they have not been influenced by previous direct union experience or socialization. Thus, the revitalizing promise of young new members for the union movement might be overstated. Therefore, such a generational viewpoint is de-emphasized here in favour of a focus on first-time union members, regardless of age. They can be colloquially labelled ‘newcomers’. Their newness offers opportunities for stimulating a joint process of organizational learning (Hyman, 2007), combining the newcomers’ possibly innovative approaches and the union’s use of ‘socialization tactics’ for facilitating ‘adjustment’ to the labour movement. Based on an e-survey conducted among newcomers of a Belgian union, this article will analyze their willingness to voluntarily engage in union activities in an empirical manner. Apart from being future-orientated, with a methodological focus on ‘planned behaviour’, the research approach is also characterized by its explorative nature, whereby union participation is conceptually linked to specific tasks imbued with an ‘organizing soul’ (Fiorito and Jarley, 2008). In this regard, Belgium is an interesting case for exploring how member-union ties can be enhanced, as it is a quintessential example of a high-union-density country, with instrumental and passive membership predominant (Van Gyes, 1999).

Contributing to the debate on a conceivable vanguard position for newcomers in union revitalization, this article is also novel in three other ways regarding how it conceives of union membership. First, inspired by the developmental perspective on union participation (Fiorito et al., 2015: 572), and unlike most studies, voluntary work for the union is not conceptualized here as the intent to become a union delegate or representative, but rather as the willingness to perform several less-demanding tasks. This follows the argument that a gradual approach, via the introduction of various levels of union participation, might be more fruitful for stimulating union activism and appealing to a larger membership base than creating a basic distinction between passive and fully active members. Second, it focuses on existing members in order to examine the appeal of activism based on organizing unionism. This emphasis distinguishes it from most other studies on the topic, which instead assess the outcomes of organizing campaigns targeted at potential members and membership growth. Third, while research on union activism is typically focused on English-speaking countries, the article highlights the potential of organizing unionism for unions with a more all-encompassing membership, i.e. including non-working members like the unemployed, which also opens the door to working on issues beyond the workplace.

Hypotheses and planned behaviour theory as a framework for examining unions

Although organizing unionism is not the only union revitalization strategy (Frege and Kelly, 2004), it has been claimed for quite some time now to be an important one (de Turberville, 2004), particularly for achieving a greater impact in advancing the interests and needs of union members and workers in general, in both the workplace and society as a whole. Apart from accentuating recruiting and mobilizing over servicing members—a contrast that should not be exaggerated, as both are essential to unions—the organizing philosophy also promotes increasing the active participation of the rank-and-file. It is claimed, however, that current union practices tend to consider organizing unionism as only recruiting new members—and sometimes mobilizing them (Simms and Holgate, 2010; Simms et al., 2013). Yet its “qualitative dimension” (Ibsen and Tapia, 2017: 11) of increasing members’ activism and union commitment is at least equally important (Bergene and Mamelund, 2017; Fiorito et al., 2010; Gall and Fiorito, 2012; Smith et al., 2019; see Hickey et al., 2010 for a more skeptical view). This is the case for at least three reasons. First, unions can directly influence this dimension (Behrens et al., 2004), which is also of importance as a retention-centred strategy for newcomers, as the first years of membership are crucial for encouraging members to stay in the union (Leschke and Vandaele, 2018). Second, by partially redistributing routine-style union work through involving ordinary members and encouraging their self-sufficiency in finding solutions for workplace or other issues, it is believed that already hard-pressed union representatives and full-time officers will be able to allocate more time to work on union policies of strategic importance (Nissen, 1998). Third, the unpaid volunteer efforts of members could help enhance a union’s organizational power and mobilization capacity (Clark, 2009). For understanding the latter, it is helpful to conceptually distinguish between three approaches for rebuilding collective power based on the agent of change: advocacy, mobilizing and “deep organizing” (Holgate et al., 2018; McAlevey, 2016).

Union member participation is nearly absent in advocacy, which is instead run by professional union staff. In contrast, ordinary union members are involved in mobilizing and deep organizing, but to different degrees. While mobilizing does entail the involvement of union activists, its occasional nature implies that this approach only focuses upon those members who are already supportive, and also largely depends on staff for leadership, decision making, and problem solving. In the bottom-up organizing approach, the base of activists is widened through the day-to-day engaging and mobilizing of ordinary workers for a transformative change via mass pressure. Union commitment studies (Snape and Redman, 2004; Smith et al., 2019), as illustrated in Figure 1, help to further understand how unions relate to their members in each approach and how this explains the multidimensionality of union participation.

Figure 1

Linking union strategies with member-union relationships and their operationalization

Linking union strategies with member-union relationships and their operationalization
Source : Inspired by McAlevey (2016) and Snape and Redman (2004).

-> Voir la liste des figures

Thus, in the advocacy strategy, the economic exchange between members and unions is central to realizing immediate gains and benefits, while mobilizing involves a social relationship, as “members are seen as potentially active participants” (Snape and Redman, 2004: 857). Deep organizing goes beyond the exchange motives driving advocating and mobilizing; it is characterized by covenantal orientations, i.e. based on shared values on organizing unionism (Fiorito et al., 2010) and ideological identification and congruence between the political beliefs of members, activists and union leaders (Upchurch et al., 2012). Taking into account those intrinsic motivations, two hypotheses are proposed:

  • H1 Strong shared values on organizing unionism influence intent to engage in organizing-oriented activism positively.

  • H2 Strong ideological identification and congruence influence intent to engage in organizing-oriented activism positively.

Union members’ willingness to perform direct, tangible and ‘small’[1] union tasks, inspired by organizing unionism, could be considered a specific aspect of the union commitment construct, especially regarding union participation (Gall and Fiorito, 2012; Smith et al., 2019). Therefore, the literature on union commitment is of value for exploring the behavioural intentions behind future union activism. The theory of planned behaviour offers a conceptual framework for linking intentions with behaviour via three types of beliefs identified as psychological antecedents of union activism (Fiorito et al., 2015). First, these intentions are influenced by union members’ beliefs regarding union activism, to which perceived union instrumentality and general beliefs about unions or pro-union attitudes are related, with the latter being more strongly related to activism. It is questionable, however, whether union instrumentality is relevant for every type of union participation and the importance of pro-union attitudes and union instrumentality for union participation in general is debatable (Fiorito et al., 2010, 2015). Second, views on union activism held by families, friends or co-workers have an impact on an individual’s willingness to engage in such activism. Third, beliefs about one’s behavioural control, or the conviction that one’s personal involvement will make a difference, also contribute to union activism. In light of such context, the following hypotheses can be formulated:

  • H3 Union instrumentality has no influence on intent to engage in organizing-oriented activism.

  • H4 Strong pro-union attitudes influence intent to engage in organizing-oriented activism positively.

  • H5 Strong pro-union socialization influences intent to engage in organizing-oriented activism positively.

  • H6 Strong behavioural control influences intent to engage in organizing-oriented activism positively.

Apart from these types of beliefs, previous research has identified several other antecedents of activism. Past union participation, with a distinction made between passive and active participation (Fiorito et al., 2015; Smith et al., 2019) and union loyalty (Buttigieg et al., 2014) are positively associated with a higher probability of union activism, while the converse is true for union apathy (Bergene and Mamelund, 2017). These antecedents will also be examined.

Now is the time to experiment – the Belgian context

Union density stabilized at around 55% in Belgium from the mid-1990s until 2013. Unions’ considerable mobilization capacity, including large-scale demonstrations and one-day general strikes, in combination with their institutional embeddedness in the labour market and social security system (which has so far undergone incremental adjustments without any major policy changes), are possible explanations for this high and stable density (Faniel, 2010). Two examples illustrate how the member-union relationship is underscored by an economic exchange. First, union involvement in the administration of unemployment insurance schemes is a particular incentive for workers to join a union or retain their membership (Van Rie et al., 2011). This pertains not only to young people, usually at higher risk of unemployment, but also to older workers, as they are often disproportionally hit by corporate restructuring and could be entitled to early retirement benefits paid out by the unions. About one in three union members are not in employment, chiefly due to this ‘quasi-Ghent system’.[2] The system thus tends to deliver many passive consumers of individual union services outside the workplace, like career counselling and training. Second, member retention is also encouraged via a ‘union premium’. Subject to compliance with the ‘social peace’ clause in collective agreements, this additional benefit for union members only is paid out from social funds at the industry level.[3] The premium not only acts as a selective incentive for preventing free-riding, but it can also substantially reduce union dues in most but not all industries.

For a long time, Belgium was considered an exception to the trend of declining unionization, but the unions’ problems have increased. While the quasi-Ghent system results in a small youth-adult gap in unionization, the median age of members has increased over time (Vandaele, 2018). Moreover, membership has recently started to decline. Consequently, there is a growing awareness among some unions to revitalize among the membership dimension, in particularly focusing on recruiting more young people. Indeed, a recent initiative temporarily lowers the union subscription for them, but its success is not unequivocal (Delespaul and Doerflinger, 2019), illustrating the limits of a member-union relationship solely based on economic exchange. Also, Belgian unions have frequently protested against the austerity measures adopted in the aftermath of the 2008 financial crisis, but without much direct impact, indicating a loss of political influence. Pointing to deficiencies in the member-union relationship based on social exchange, within union circles and beyond, it is often stated that only the ‘usual suspects’, a network of middle-aged committed activists and union representatives, are sufficiently motivated to take part in union-staged demonstrations and industrial action (Faniel, 2012). Following the dominant servicing unionism, these activists run the risk of being overwhelmed by members’ demands (Liagre and Van Gyes, 2012). Developing rank-and-file support could in fact help to mitigate feelings of burnout.

Tactics associated with organizing unionism are not unfamiliar to Belgian unions. In particular, the period prior to the quadrennial social elections for works councils in the private sector offers them an opportunity to reach out to workers and to identity future activists, i.e. those who want to run as candidates on the election lists (Op den Kamp and Van Gyes, 2009). Nevertheless, the focus often tends to rest upon existing activists than identifying newcomers as potential future ones. Unions seem to think that the latter (and young people in particular) are difficult to convince to stand as candidates in these elections. Furthermore, the elections only take place in companies with at least 100 employees.[4] Consequently, although union representatives in some industries may be active in companies with twenty employees or more, a considerable number of workers have no direct exposure to unions at their workplace; this is especially true for young workers as they are more likely to work in small companies. Finally, less attention is paid to members in the four years between elections: targeted member recruitment campaigns are lacking and there are generally no dedicated (follow-up) programs available for candidates who have not been elected. Therefore, de-emphasizing servicing unionism, by prioritizing innovative practices to involve members more in union work, could be a viable strategy for reigniting organizational power and regaining membership growth among Belgian unions (Van Gyes, 1999), especially as they still have considerable financial and personnel resources as well as vast numbers of members whose potential remains largely untapped.

Data collection and sample

Cross-sectional data were obtained from a 2015 web-based survey commissioned by Local Y of Union X, as part of a revitalization project accepted by its congress and sent to its 6,039 members with a known email address. [5] Like other Belgian unions, Union X has several local units responsible for providing union services and communicating with its members. Local Y is one of the largest locals within Union X, but membership has been declining slightly (by 0.3% in 2015 and 0.7% in 2016).[6] Despite being one of Belgium’s largest unions, neither Union X nor its Local Y studied here had formal youth structures in place at the time of the survey. All respondents were first-timers in Local Y and had not been members for longer than seven years. This cut-off point was chosen because the likelihood of turnover is especially high in the early years of membership (Leschke and Vandaele, 2018). Union activists and representatives have been excluded from the data membership file as the research focus is on members not yet formally active.[7] Newcomers in Union X receive a welcome brochure and additional information about union services, but there are no formal procedures for any additional follow-up, nor does any retention policy exist. They are, however, usually welcomed by shop stewards in companies with union representation during the induction period.

A total of 1,528 usable questionnaires were received, yielding a response rate of 25.3%.[8] The sample is representative in terms of gender (χ2 (1, n=7.626)=1.84, p=0.17), with women accounting for 66.3% (n=1,013) of respondents; a percentage reflecting the industries in which Union X is located. Average membership duration stands at two years and five months (n=1,528, SD=1.83). Confirming earlier studies (for an overview, see Vandaele, 2018), Figure 2 shows that the vast number of newcomers join Local Y at a rather young age, with women joining at a younger age than men, which is, again, linked to the industries in which women are employed (being associated with a more practical education). A quarter of Local Y’s members had already joined by the age of 23, half by 28 and three quarters by 41. The average age of respondents with no prior union history was 35 (n=1,088, SD=12.16), indicating that workers also unionize for the first time at an older age, especially men; the quasi-Ghent system could be a particularly strong incentive in these cases. Following case-wise deletion of missing data, 518 observations were available for analysis.

Figure 2

Entry age of new union members of Local Y

Entry age of new union members of Local Y
Source: Survey results

-> Voir la liste des figures

Measures and tests

Table 1 provides an overview of the used constructs and reliabilities, where relevant; the items and exact wording have been translated from Dutch. Established scales were used where possible, but several measures were non-established due to the exploratory nature of the survey. Exploratory factor analysis, using maximum likelihood, confirms the convergent and discriminant validity of all used constructs.[9] The scales’ reliability is satisfactory as measured by Cronbach’s alpha coefficient except for the passive participation construct. The dependent variable measures the organizing-oriented activism intent based on union tasks, reflecting one-on-one organizing tactics for reaching out to potential members; some of these tasks are not necessarily confined to the workplace. The activism intent should be distinguished from the willingness to become a union representative (Fiorito et al., 2010), a question targeting those respondents with a job, as the estimated correlation between those two variables is significant but only moderately strong (rs (303)=.56, p<.00). A critical mass of 41.3% (n=214) is (greatly) interested in at least one task. Newcomers with a strong willingness to do so are significantly more prepared to be contacted by Local Y for further developing their union involvement (M=3.06, SD=.96) than members with a lower intent (M=2.28, SD=.88) (t(513)=-8.96, p<.00). Certainly, not all newcomers are willing to participate in the union’s work. A loss of membership due solely to an organizing approach is, however, quite unlikely, as the activism intent is only weakly correlated with leaving LocalY “for reasons of a too active role expected from ordinary members” (r s (518)=.34, p<.00).

Table 1

Constructs, their Likert-type items and reliabilities

Constructs, their Likert-type items and reliabilities

-> Voir la liste des tableaux

Four focus variables were introduced using the conceptual framework on planned behaviour. Given that Local Y organizes members with and without a job, a scale measuring union instrumentalism through workplace performance is less relevant here, as members without a job are unable to assess this. As an alternative, given the predominance of service unionism, union instrumentalism is measured by a four-item Likert-type scale measuring respondents’ satisfaction with union services as an example of an ‘economic exchange’ between Local Y and its members. The respondents’ pro-union attitudes are also based on a four-item Likert-type scale. The pro-union socialization context could only be indirectly measured for all Local Y members by the answer ‘because a lot of my colleagues, friends or family members are a union member’ to the question “Why did you become a union member?”. It is assumed that this initial pro-union context, either at the workplace or beyond, is almost intact or fairly sustained. The perceived self-control, as an intrinsic motivation, is based on two items and a grouped variable, with four categories (‘weak’, ‘moderate’, ‘strong’ and ‘I don’t know’) (rs (474)=.57, p<.00).[10]

Regarding the independent variables, a conceptual distinction is made between passive and active participation for measuring prior union participation, as each type is linked to other antecedents, because they require different levels of effort, with each of them based on four activities (Fiorito et al., 2015; Smith et al., 2019). The attitudes towards organizing unionism are based on a three-item Likert-type scale. Such attitudes are significantly positively related to members’ desire for transparent bargaining, i.e. to be more and better involved in every phase of collective bargaining at workplace (rs(289)=.49, p<.00) or industry level (rs (283)=.41, p<.00). Feelings about ideological congruence are based on a single item, ‘I agree with Union X ideologically and politically’; this is a group variable with four categories (weak, moderate, strong and ‘no opinion’). Union loyalty comprises four Likert-type items. Respondents are considered apathetic when providing neutral or ‘no opinion’ responses to at least five items out of the eleven union-related questions (see Bergene and Mamelund, 2017). Their share stands at 16.0% (n=83), with the youngest age category being significantly more apathetic (χ2(3, n=518)=9.18, p=.03), while there is no significant association with the duration of union membership (t(516)=-.60, p=.55). Five control variables are introduced: gender (1=male, 2=female); age is a grouped variable with four categories (≤24 years, 25-34 years, 35-54 years and ≥55 years); education is a grouped variable with four levels (basic and lower secondary education, upper secondary, tertiary non-university and university level); labour market position is a grouped variable with four categories (full-time, part-time, unemployed and other), with ‘other’ including students, occupationally disabled members, members on a leave scheme, house husband or wife members, and (early) retired members (all sharing a certain detachment from the labour market); and with the duration of union membership as a continuous variable. Regarding the age variable, the maximum age limit for the first category is based on the social elections: there is a separate candidate list for young workers in companies employing at least 25 workers below the age of 24. This age category is also in line with youth studies. The maximum age of the second category is used in Union X for defining a young worker. Accordingly, respondents less than 35 years old are considered ‘young’ throughout this article (see Tapia and Turner, 2018). Finally, a dummy variable is added as the data file did not allow a prior distinction to be made between first-timers and members who had left another local of Union X or another union and joined Local Y; newcomers without a history of union membership dominate the sample with 68.9% (n=357) and they are significantly younger (M=34.9, SD=11.6) than new members with a union history (M=41.9, SD=11.2) (t(518)=-6.36, p<.00).

Analysis and findings

Descriptive statistics and correlations are presented in Table 2 for all variables. The pairwise correlation results demonstrate how the focus and independent variables correlate significantly and positively with the organizing-oriented activism intent, except for union apathy. This is especially the case for perceived self-control, organizing unionism attitudes and both forms of union participation. The correlations suggest no problems of multicollinearity between the independent variables. While union participation occurs more passively than actively, the correlation between these types is moderate, confirming that these are different forms of participation. There are some noteworthy correlations. Union loyalty correlates positively in particular with ideological congruence, pro-union attitudes, passive participation and satisfaction with union services. Pro-union attitudes are similarly positively associated with satisfaction with union services and ideological congruence. Finally, perceived self-control is positively linked with attitudes towards organizing and, again, ideological congruence.

Table 2

Descriptives and correlations, n=478

Descriptives and correlations, n=478

Note: Correlations with an absolute value of 0.12 are statistically significant at the 0.01 level.

Source: Survey results.

-> Voir la liste des tableaux

Table 3 provides step-by-step results of ordered logit regressions to estimate the willingness of newcomers to take on organizing unionism tasks. The coefficients (b) signify the proportional odds ratios for the models. Model 1 includes only the control variables. The model suggests that, compared to members aged 55 or older, newcomers between 25- and 34-years old show higher odds of performing organizing unionism tasks. In contrast, more longstanding members display a lower intent to take on organizing tasks. Models 2 and 3 respectively introduce the independent and focus variables to assess their effect and to what extent they contribute to the model fit. Model 2 indicates that the younger age categories (below 35) are more likely to perform organizing unionism tasks than older newcomers (above 54). Members with a tertiary non-university level have a lower intent to take on such tasks compared to newcomers with upper secondary education. The weaker willingness of more longstanding members to do so is no longer significant. Hypotheses H1 and H2, which postulate that the willingness to conduct organizing tasks is positively related to strong attitudes towards organizing and strong ideological congruence with Local Y, respectively, are supported in this model. It is also worth noting that members who have ‘no opinion’ (yet) about their congruence also have higher odds compared to members with a weak ideological congruence. Additionally, in this model, newcomers with high prior passive or active union participation and those with outspoken union loyalty all demonstrate higher odds of taking on organizing tasks. Introducing the independent variables contributes to a better model fit, as is also the case for the focus variables in Model 3. Hypotheses H3, H4, H5 and H6, all linked to the theory of planned behaviour, are tested in this model. Hypothesis H3 theorized that union instrumentality will have no impact on the willingness to conduct organizing tasks, which is confirmed since the satisfaction with union service is not significant. H4 is falsified: strong pro-union attitudes are not positively related to performing union organizing tasks. Hypotheses H5 and H6 are supported, as members strongly embedded in a pro-union context, and especially those newcomers with moderate or strong perceived self-control, display higher odds of taking on organizing tasks. Furthermore, a strong passive participation and union loyalty are still both significant. There are no significant differences between the age categories and education levels in Model 3, however, while the significant associations of newcomers with a strong active participation and with a moderate or strong ideological congruence (H2) also disappear. A member’s gender, labour market position and history of union membership attain no statistical significance in any of the models.

Table 3

Ordered logit regression for intent to carry out organizing tasks

Ordered logit regression for intent to carry out organizing tasks

Note: +p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001.

Source: Survey results.

-> Voir la liste des tableaux

Running the full model for each of the four single items of the composite scale measuring the organizing unionism activism intent as dependent variables confirms that there are common antecedents: strong attitudes towards organizing unionism (H1), the pro-union context (H5), moderate and strong perceived self-control (H6), passive union participation and union loyalty are all significantly positively associated with the intent to inform, recruit or mobilize friends, family or colleagues or to distribute flyers in public spaces; union loyalty is not significant for the latter task.[11] Several variables are insignificant for all four tasks: this is the case with gender, satisfaction with union services (H3), pro-union attitudes (H4) and for those members with no knowledge about their perceived self-control (compared to a weak one). There are also some antecedents that differ according to the task. Members with basic or lower secondary education, and members with a moderate or strong ideological congruence (H2), or with no opinion about it, display higher odds of informing others about the demands of Local Y. Additionally, members with no opinion about their ideological congruence display higher odds of recruiting members compared to those with a weak congruence. Newcomers displaying strong active participation display higher odds of mobilizing, while ideological congruence is insignificant. Finally, surprisingly, apathetic members and members who have no opinion about their congruence with Local Y display higher odds of carrying out union leafleting. These results point to several ramifications for unions if they pursue policies for enhancing the engagement of newcomers via organizing-oriented tasks: the explicit ideological embeddedness of the tasks within the union objectives, the diversity of the tasks and their developmental character all need to be considered.

Discussion

Although the ideological congruence between newcomers and Local Y is no longer significant in the full model, there is some support for the view that boosting member-union ties via organizing tasks is more successful when underpinned by an articulated political vision and long-term objectives (Simms and Holgate, 2010). In particular, ideological congruence plays a significant role in informing others about the demands of Local Y. Stimulating worker self-organization or empowerment does therefore not only require training, but also mentoring and education, and a particular targeting of the youngest age category who tend to display weaker or more moderate congruence (χ2(9, n=518)=21.25, p=.01).[12] At the same time, rather strikingly, newcomers with no opinion about their congruence are significantly keener to devote time and energy to union tasks than members with weak ideological congruence. Similarly, apathetic members are also associated with the intent to distribute flyers in public spaces. Both findings hint as to what extent a union has room for manoeuvre when initiating organizing unionism with the intent to reduce ideological unawareness and feelings of indifference among newcomers through experience and learning (Bergene and Mamelund, 2017).

The regression models for each separate organizing task indicate that some antecedents vary. As tasks could differ “time-wise, skill-wise or stress-wise” (Fiorito et al., 2015: 575), they might appeal to different member categories. Member diversity should thus be considered when implementing specific tasks. Regarding age, the younger age categories display a stronger activism intent, possibly explained by their “biographical availability” (Milkman, 2017) or by the very nature of the tasks (Smith et al., 2019), which may be more attractive to them, although this is not always significant in the models. In this context, young members should be offered genuine opportunities to carry out specific tasks they find appealing and older union activists should put aside stereotypes about the younger generation (Vandaele, 2018). It is crucial for youngsters’ continued involvement that they “feel that their contribution can make a difference” (Byford, 2009: 237). At the same time, older newcomers, over 54 years old, display a weaker willingness, likely pointing to greater economic exchange motives for union membership, at least in the Belgian context. In addition, newcomers, irrespective of their age, are probably more prone to leave the union than longstanding members (Leschke and Vandaele, 2018). Therefore, in looking for innovative dynamics that they can bring to unions, it might be worthwhile lessening age categorization and to widen sights to encompass all newcomers, including older ones, albeit without denying the importance of age-based union networks and representation structures for young members. Educational levels and labour market positions seem to be of little relevance for most but not all tasks. The latter finding is not only relevant for Belgium or other Ghent-system countries (Denmark, Finland, Iceland, and Sweden), where unions organize the unemployed, but also for unions in other countries that count members among their ranks who are fairly detached from the labour market. The union tasks considered here appeal equally to men and women, although significant gender differences have been recurrently found regarding the willingness to become a union activist or representative, with women less likely to display such an interest (Gall and Fiorito, 2012: 198).

The results suggest not only that union participation is a multidimensional construct, but also that there is a need for a developmental perspective. From the viewpoint of newcomers, supporting the union through small tasks on an intermittent basis might help lower perceived thresholds to union participation, and be more appealing to a larger pool of (younger) members than taking on the arduous responsibilities of union representatives and the tasks associated with it. Therefore, it might be worthwhile for unions to investigate how hard-pressed (older) representatives could feel more supported by involving ‘union advocates’, as this kind of union participation is termed by the Public and Commercial Services Union (PCS, 2015) in the United Kingdom. In line with the understanding of unionism as an ‘experience good’ (Vandaele, 2018), such informal induction into union work might be a ‘stepping stone’ to deeper union involvement (Fullager et al., 2004). Therefore, ‘union advocacy’ could help ‘occupy’ the no-man’s-land between inactive, service-prone members and over-stretched union representatives. This in turn could create a prosperous virtuous circle: “in the process of engaging in lower-demand activities, (…) awareness grows of other activism opportunities—along with the chance to observe positive impact from voluntary efforts. Over time, this effect likely enhances new members’ perceived control, eventually leading to taking on more demanding roles” (Fiorito et al., 2015: 575). It remains to be seen if mobilizing newcomers can have spillover effects on longstanding members, and how this will influence workers as prospective members—a suggestion for subsequent research.

Turning to the more practical implications then, like with any other organizing initiatives, internal tensions and resistance might arise when introducing and implementing organizing-based tasks; there is no guarantee of success. Therefore, prior debate is needed, for instance about the required resources, the relationship with other types of non-voluntary union work and their embeddedness in existing union organizational structures and the overall union strategy. First, as the tasks demand coherence, coordination and steering, a re-allocation of resources for building effective ‘union advocacy structures’ is required. Second, while utilizing the experience and skills of newcomers could help distribute the workload of union representatives, this implies a rebalancing of the relationship between those two categories of members, especially as established union representatives might get the feeling that their status is somehow being affected as the member-union relationship would shift from one based on economic and social exchange to a covenantal one. Consequently, the appeal and success of organizing-based tasks might be greater in industries with a weaker union presence than in industries with well-established union structures and entrenched cultures. Lastly, organizing-oriented activism via small tasks might be still marked by an overly top-down character, facilitating control by bureaucratic and hierarchical union structures (Hurd, 2004; Simms et al., 2013).

‘Managed activism’ (de Turberville, 2004) might entail the risk that certain tasks are only half-heartedly implemented or that novel tasks proposed by ‘union advocates’ are stymied when they are perceived as alien to the predominant union culture, traditions and practices. The opportunities to anchor organizing-oriented tasks in union strategies might therefore work better in a union that has already switched to an organizing culture and wants to develop this further via ‘union advocates’, as is the case with the PCS (see Hodder, 2014). Alternatively, as proposed here, introducing organizing-inspired tasks might be a ‘soft way’ to import organizing unionism into institutionally embedded, large and mainstream unions (like in Belgium) to strengthen their organizational power. In a similar vein, in Germany, unions like ver.di and IG Metall deploy policies for strengthening union democracy and participation, especially in the build-up to collective bargaining negotiations, for instance through member surveys or a ‘mitgliederorientierte Tarifpolitik’, which sets a minimum union density that is required in a particular industry before unions can start negotiations. In any case, it is an empirical question to what extent a transformative leadership together with a critical mass, bringing together newcomers as ‘union advocates’ with existing, dynamic union activists, can shift away from a top-down organizing approach and evolve towards bottom-up unionism and a worker-led horizontal network of ‘union advocates’.

Conclusion

Two variables, derived from planned behaviour theory, significantly influence organizing-based union tasks: the pro-union context and behavioural self-control, with the latter being an especially strong, intrinsic antecedent. The importance of the pro-union context confirms the association between newcomers’ embeddedness in social networks, in and outside the workplace, and union activism (Johnson and Jarley, 2005). This calls for a network-based focus on organizing unionism tasks, with the union as the orchestrator (Heckscher and McCarty, 2014). Satisfaction with union services, as a proxy for union instrumentalism, and pro-union attitudes are not significant, indicating that economic or social interaction with Local Y is not driving the organizing tasks intent. Instead, the significance of positive attitudes towards union organizing points to a covenantal relationship that is driving deep organizing (Holgate et al., 2018; McAlevey, 2016). While providing evidence of a “nature of activism [that] is qualitatively different from most existing activism” (Gall and Fiorito, 2012: 194; Smith et al., 2019), this is a rather self-evident explanation and further research is needed to better understand the background of such attitudes. For instance, is it possible that members preferring a covenantal exchange with the union have experience in community organizations or other progressive movements? The intent to take on organizing-based tasks, as a socialization tactic, is also clearly driven by union loyalty, possibly indicating that loyal members “have some calculative belief that they can change the organization from within” (Buttigieg et al., 2014: 10). A developmental view on union activism is backed by the finding that members with a history of passive union participation seem to be significantly interested in taking on union organizing tasks. Previous active participation seems a less robust precursor to further activism in Local Y, indicating its slightly different nature or certain limits to expanding the base of activists, although it remains an empirical question at which level of activism (fatigue) this would occur.

The findings presented here have several limitations, which stem from the research design. Some variables are single-point items undermining the reliability and validity of attitudinal measures. Measurement context effects might have occurred, as the survey was conducted in a period when Union X was preparing to protest against the government. The idiosyncratic character of a single-union survey also implies that replicated research into other unions and countries is needed to assess the generalizability of the findings. Obviously, only relationships, attitudes and intentions towards union activism have been studied; no claims could be drawn regarding actual behaviour. Furthermore, solely social-psychological factors were considered, but no economic or other contextual factors (Angrave et al., 2017); similarly, additional variables like a member’s ethnic background, job satisfaction, or the support (Tetrick et al., 2007), culture or leadership of the union were not taken into consideration. Despite these caveats, if there is one key message to take away from this article, it is that there is a critical mass of newcomers (among them young members) interested in performing organizing tasks, which is currently underutilized. The actual size of this group of potential internal reformers should be a subject of empirical investigation in each and every union. Union activism is often sparked by opportunities or (unexpected) events like perceived unfair treatment by management or company restructuring. The generally low degree of activism among union members might not just reflect a lack of opportunity, however, but also indicate an absence of dedicated union programs to stimulate less-demanding forms of activism. If interested members can be identified and addressed, this could be an opportunity for unions to enhance (in a sustainable way) their power and capabilities in the workplace and beyond, and thus to revitalize their organization.