Fit for fight? A cross-sectional study of union apathy in Norway

Abstract

The article synthesizes the literature on union commitment and union renewal through employing the concept of union apathy. This is done with a view to analyse the association between union strategies and union apathy, and the association between apathy and willingness to become a union delegate in Norway. To this end, the authors have run multivariate regressions on the YS Employment Outlook Survey. Results show that union strategy is significantly associated with apathy, and that apathetic union members are less willing to work for their union in the capacity of union delegate. The authors conclude that if unions are to counter apathy in an effort at renewal, and to ensure a sufficient pool of potential union delegates, they must mobilize the talents, ideas and energy of their members. Finally, the authors suggest that unions with an ‘organizing’ soul seem better equipped to counter apathy than unions bent on the ‘servicing’ model.

Keywords

Apathy commitment democracy empowerment membership unions

Introduction

Falling union density has been widely documented in most industrialized countries in recent decades (Cregan, 2005; Fiorito and Jarley, 2008; Gall and Fiorito, 2012; Johnson and Jarley, 2004; Mason and Bain, 1993; Peetz, 1998; Peetz and Pocock, 2009; Pekarek, 2009; Riley, 1997), causing some scholars to point to a crisis of unionism (Cregan, 2005; Gall and Fiorito, 2012). Efforts to stem the tide have led unions to try to devise effective responses, mainly by way of seeking strategies to recruit and/or retain members (Cregan, 2005). However, according to Gall and Fiorito (2012: 194, our emphasis), if unions are to renew, sustain and expand organizing, they need to ‘activate significant proportions of inactive members’. In our view, union renewal or revitalization is an ongoing process, in need of constant attention, and not an outcome (Behrens et al., 2004a).

Our point of departure is heeding Gall and Fiorito’s (2012: 203) call for analyses which synthesize the literature on union commitment and on union renewal, by, more specifically, studying the factors affecting union commitment, in an effort to arrive at ‘what unions can do to boost activism to further union renewal’. To this end, we employ Peetz’s concept of union apathy, because, in our view, it bridges the gap between the literature on strategies of renewal, on the one hand, and union commitment, on the other. Union apathy is defined as ‘a disinterest in union matters and a consequent lack of opinion on union-related issues’ (Peetz, 1998: 12). Clark (2009) argues that the strength or effectiveness of a union depends on the union-related behaviour of members, where behaviour is a result of attitudes and union strategies. Union apathy may negatively affect union-related behaviour. Peetz (1997) finds that respondents giving apathetic responses were more likely to leave the union than respondents who both gave more positive and more negative responses, and they were less likely to join. That is, apathetics had a markedly lower union propensity than non-apathetics (Peetz, 1997).

Norwegian unions have yet to experience a significant decline in union density, although a recent research report raised concerns about negative developments in the number of workers willing to join a union and the number of union members willing to become union delegates (Bergene et al., 2012). Norway is an interesting context for studying union apathy because of the broad political consensus on the merits of the tripartite ‘Norwegian model of industrial relations’ and the ensuing industrial peace, along with a stable union density. Norway is, in other words, interesting because it is different. These exogenous factors are common to Norwegian unions and their confederations, so our point of departure is that unions have strategic choice within the given context they operate in as they have ‘leeway in deciding … whether pursuing internal organizational or leadership changes is vital to their renewal efforts, and how crucial the membership dimension is’ (Behrens et al., 2004a: 24).

Apathy could, however, affect to what extent Norwegian unions are prepared for a harsher political and/or economic climate. As pointed out by Dörre et al. (2009), institutional power may compensate for, or even hide, a waning organizational power, and Norway could, potentially, end up with the same realization as German unions that they have ‘rested on their institutional laurels and lost touch with rank-and-file … concerns’ (Turner, 2004: 7).

Based on this, we analyse two research questions. The first is: to what extent is there an independent relation between union confederations with differing strategies and the apathy of their individual union members? In line with Oesch (2012) we presume that confederations differ in their approach to unionism and treat them as proxies for union strategies, and we thus use them as independent variables in our analysis. The second research question is: to what extent is there an independent relation between being apathetic and (un)willing to become a union delegate among individual union members?

Regarding the latter research question, we bear in mind Bartram et al.’s (2008) argument that it is difficult to assess whether union commitment is a cause or a consequence of delegate incumbency. Because our primary focus is on the union movement, we have chosen to study the association between union apathy and the willingness to become a union delegate, rather than how the holding of such an office is related to union commitment. We use multivariate logistic regression models and data from four waves of a nationally representative cross-sectional survey on Norwegian working life (2009–2012) to address our two research questions.

While the first part of this article is devoted to the determinants of apathy, the second half focuses on what may (not) convert individual experiences of apathy into union activity (here defined as becoming a union delegate). We would argue that this is a necessary supplement to the traditional focus in industrial relations on collective bargaining and its associated institutions (Johnson and Jarley, 2004). Without employing the concept of apathy, Gall and Fiorito (2012: 206) posit that commitment derives from inclusion in decision-making, effective communication with members, cultivation of a collective orientation, and mobilization, as the latter not only presupposes certain beliefs and attitudes, ‘but can change beliefs and attitudes as well’.

Our research contributes in several ways. When it comes to the first research question, only Peetz (1997, 1998) has analysed determinants of union apathy before us. Research on union joining and commitment has mainly focused either on the individual – arguably as a result of the continued hegemony of the servicing model (Cregan, 2005) – or, alternatively, on institutions and economic conditions in explaining union growth and decline (for instance the Bainian model; see Riley, 1997). Little research has focused on unions or confederations as strategic agents (Oesch, 2012). Oesch (2012) finds that internal differences in unions (e.g. strategies) have twice the explanatory power of external context when it comes to in- and outflows.

Regarding the second research question, while there is a large bulk of research on the determinants of union membership (for reviews, see e.g. Gall and Fiorito, 2012; Riley, 1997), the corresponding literature on the willingness to serve as a union delegate is rather dated, scant and incomplete (Flood and Turner, 1996), with a peak period in the 1970s in the UK (Pekarek, 2009). Bartram et al.’s (2008) study of union delegates in the Australian Nursing Federation is a notable exception. We have not identified papers estimating to what extent there is an association between apathy and the willingness to become a union delegate. A study of settings other than the Anglo-American should also be informative. Third, while many have studied specific workplaces, occupations, locations or unions (Gall and Fiorito, 2012; Gallagher and Strauss, 1991), we analyse a sample which is representative of all these subgroups of the Norwegian working population. Compared to the work of Peetz (1997, 1998), in addition to using data which are representative for a country rather than a single city (Norway vs Sydney), our net sample is much larger (10,000 vs 700 individuals) and it covers a more recent period (2009–2012 vs 1990–1991). The larger sample size makes it possible to detect statistically significant associations between smaller subgroups of the workforce and the outcome variables in question.

Theoretical framework

Unions facing decline can adopt a series of measures: they may tolerate an absolute decline if density remains high, they may merge or amalgamate, they may adopt retention strategies or they may start organizing drives (Mason and Bain, 1993). Behrens et al. (2004b) divide strategies of change into three predominant categories: changes to external structure, e.g. expanding the membership base; internal changes to union governance, e.g. increasing union democracy, representation and participation; and internal changes to union administration, e.g. allocation of resources and/or union management. These three categories of structural change may then be studied in light of the principal motivation for undertaking them: aggressive (expansive and accommodative), defensive (survival and accommodative) and transformative (strategic and substantive) (Behrens et al., 2004b). According to Behrens et al. (2004b) aggressive and defensive motivations for restructuring can account for the vast majority of union mergers. What sets the latter motivation apart is that transformative change is pursued with a view to revitalize the labour movement as a whole, meaning that it is not pursued ‘to protect the vested interest of leaders and current members, but rather to engage all of the union’s constituents and prospective constituents in the change process’ (Behrens et al., 2004b: 121).

Retaining and recruiting members may warrant separate strategies on the part of unions. Furthermore, it is important to bear in mind Oesch’s (2012) finding that what determines union growth or decline is recruitment efforts and not retention policies. One reason for this is that while organizing and recruitment are largely within the purview of unions, outflows are not necessarily within the reach of unions to influence as they can be the result of external conditions, such as changes in job circumstances (Oesch, 2012). A similar emphasis is found in Clark (2009) when it comes to union effectiveness. While several factors influence union effectiveness – such as conditions in the labour market, the state of the economy, technological development and government – member participation is the one over which unions themselves have substantial influence. As pointed out by Gall and Fiorito (2012), it is thus high time to supplement the call for union renewal with the insights derived from the union commitment/participation literature, so as to achieve a better understanding of how to activate members.

Two models of unionism

The union renewal strand has often operated with a division between two models: the servicing model and the organizing model. While retention is more associated with the servicing model, recruitment is increasingly argued best done according to the organizing model (Cregan, 2005). These two models of unionism should be understood as ideal types, i.e. any union will display aspects of both, and where the weight is placed could be understood as the ‘soul’ of the union (Budd, cited in Fiorito and Jarley, 2008). While unionism in the servicing model has been likened to insurance policy, in the organizing model, unionism is often portrayed as closer to social movements (Gallagher and Strauss, 1991).

To elaborate, the servicing model is characterized by employed union officials and staff being responsible for providing members with services and representing members’ interests, the idea being that offering good service will help retain members. This is done in a manner which resembles a business transaction. Hence, in the servicing model members pay their dues and, based on an economic exchange logic, expect grievances to be handled by union officials. There is, in other words, little need for active involvement (Gall and Fiorito, 2012; Johnson and Jarley, 2004; Peetz et al., 2002). This model prevailed and proved efficacious in the western world in the period of stable economic growth and union-friendly employers, often referred to as Fordism (Voss and Sherman, 2000).

Changing conditions in the post-Fordist era have, however, led several activists and scholars to call for a different approach, often labelled the organizing model, or organizing unionism (Crosby, 2005; Fiorito and Jarley, 2008; Gall and Fiorito, 2012). The organizing model implies a redirection of resources from providing services to recruitment activities, meaning that union officials spend most, if not all, of their time on organizing (Oesch, 2012). A major concern giving rise to the organizing model, alongside declining union density, is insufficient activism (Fiorito and Jarley, 2008), arguably partly as a result of apathy. Hence, as emphasized by Voss and Sherman (2000: 313), a shift from a servicing to an organizing model entails a transformation of ‘the role of current union members, promoting new levels of commitment and participation’.

The organizing model can be regarded as a ‘return to … self-help roots’, seeking to activate members so that they can themselves pursue union goals (Fiorito and Jarley, 2008: 196–197). This sentiment is captured nicely in the slogan of the Irish union SIPTU: ‘It’s your union – keep it strong!’ (Devine, 2009: 840). The organizing model thus seeks to ‘increase member identity with and participation in the union’ (Johnson and Jarley, 2004: 544), and it is defined as bent on ‘organizing non-union workers through deployment of union resources and employed and lay personnel, and activating existing members in furtherance of union objectives’ (Gall and Fiorito, 2012: 190). In the words of Bartram et al. (2008: 26), in the organizing model ‘members become active producers of unionism, rather than passive consumers’. Unions following the organizing model will, in other words, seek to ‘tie all their activities to the strategic aim of a long-term strengthening of their organisational power’ (Dörre et al., 2009: 44).

However, this transition necessitates a transformation of workers’ attitudes (Bartram et al., 2008), although, as pointed out by an organizer quoted in Voss and Sherman (2000: 322), participation itself means that members are ‘going to want to know a little more about the union’. We have now arrived at the lessons learnt from the union commitment literature.

Union commitment

As pointed out by Gall and Fiorito (2012: 193), a synthesized framework needs to pay heed to ‘loyalty to the union, responsibility to the union, belief in unionism, and willingness to perform unpaid work (WTW) for the union’. These are the four dimensions arrived at in Gordon et al.’s (1980) seminal work, and by which union commitment has largely been measured thereafter (Bartram et al., 2008). These four dimensions have, however, by some been reduced to two dimensions: belief in unionism and union loyalty have been clubbed together to form a moral or ideological dimension, while WTW and responsibility to the union have been regarded as behavioural intentions or propensity to act. This led some to argue that the attitudinal and behavioural dimensions are distinct (Gall and Fiorito, 2012). In so doing, there is also a more or less explicit assumption that there is a potential link from beliefs and attitudes to behavioural intentions, and ultimately to potential action (Gallagher and Strauss, 1991). Important for us is that WTW might be a dependent variable in these latter analyses rather than part of the ‘union commitment’ construct, although ‘commitment comprises not only beliefs and feelings but also expressed intentions to be active’ in much of the literature (Gall and Fiorito, 2012: 191).

A full review of the union commitment literature is, however, not possible here. Union commitment is broadly defined as ‘positive feelings toward the union’ (Newton and McFarlane Shore, quoted in Pekarek, 2009: 164), and is as such related to Peetz’s (1998) concept of union sympathy. Union commitment is a widely recognized antecedent of union participation (Pekarek, 2009), and is also identified as a strong predictor of becoming a union delegate. Using another union commitment scale, it has also been found that particularly ideological union commitment is a strong predictor of becoming a union delegate (Bartram et al., 2008; Pekarek, 2009). However, as Gall and Fiorito (2012: 193) argue, the union commitment literature does not explore different types of unionism, as regards aims and strategies, and, as a corollary, what kinds of union commitment are actually ‘sought and given’. According to Gall and Fiorito (2012) this is because the servicing model is unconsciously presumed.

Clark (2009: ix) argues that strategies for building stronger and more viable unions depend on ‘an increasingly active, involved, informed, and engaged union membership’, and emphasizes two key factors influencing member behaviour: personal characteristics and the union environment. Chief among the personal characteristics for union participation are attitudes towards unions. Pro-union attitudes have been found to be important antecedents of union commitment (Bartram et al., 2008). It is thus important to build and nurture a union culture which encourages and sustains positive attitudes and increased participation (Clark, 2009). We would argue that this also necessitates an engagement with the concept of union apathy: if unions strive to increase effectiveness and strength through increasing member participation, their best bet is to do so by raising the level of member commitment through altering overall attitudes about unions (Clark, 2009). Attitudes, beliefs and values are central to both the union commitment/participation and the union renewal literature, although the latter has treated these issues rather superficially (Gall and Fiorito, 2012).

Union apathy could be understood as the consequence of a certain union environment/culture on the attitudes (i.e. the chief personal characteristic) of members. Empirical studies have pointed out that members’ satisfaction, commitment and participation are dependent on perceived union support, i.e. the degree to which the union values members’ contributions in a democratic way and cares about members’ well-being (Fiorito and Jarley, 2008; Johnson and Jarley, 2004). Similarly, Peetz (1998) has argued that unions whose internal structures do not facilitate communication with members may face a higher level of apathy.

According to Gall and Fiorito (2012) unions would be close to achieving the goal of renewal if the so-called ‘activist problem’ is solved. However, increasing activism requires more than convincing members to participate in formal decision-making and/or policy-making processes, it also requires them to actively participate ‘in carrying out the work of the union, such as membership mobilization’ (Gall and Fiorito, 2012: 191). However, as observed by Nissen (1998: 149, emphasis in original), the number of activists is ‘way too small, even in the best cases’ although ‘there are a number of high-quality union members in locals who are motivated and ready to volunteer’. This brings us to the next important issue for the union movement: the willingness among members to do work for the union.

Willingness to work

There is a broad consensus that union renewal and an organizing strategy hinge on activity at the workplace level, and thus, by extension, revolve around the union delegate as the principal figure (Bartram et al., 2008; Cregan, 2005; Peetz et al., 2002; Pekarek, 2009). Union delegates are of vital importance for the union movement since they are, among other things, ‘the most effective recruiting agent’ (Mason and Bain, 1993: 336; Pekarek, 2009). It is thus emphasized that ‘to ensure an ample supply of aspirants for the shop steward’s job’ is a key concern for unions (Pekarek, 2009: 158), since this is ‘essential if unions hope to increase their membership figures’ (Bartram et al., 2008: 26). In the organizing approach, it is held that the power of a union is inextricably linked to its ability to mobilize workers in a wider sense than just during strike action, for instance through being willing to work for the union (Fiorito and Jarley, 2008; Gallagher and Strauss, 1991). A measure of this willingness can be found in the extent to which union members would agree to serve in the capacity of union delegate (Fiorito and Jarley, 2008).

However, our understanding of what influences members’ willingness to assume the role of union delegate is incomplete. As mentioned in the introduction, while research on this question surged in the UK in the 1970s, the question has received insufficient attention in recent decades and in other geographical contexts. In sum, the literature on union delegate incumbency identified several determinants, among them union commitment, attitudes and beliefs about unions in general, a collectivist ideology, a shared union identity, full-time employment and age (Bartram et al., 2008).

The union delegates in Pekarek’s (2009) study pointed out how they became delegates by virtue of some prior participation, and also how political socialization within the union had intensified their interest in union-related issues. Another influence Pekarek (2009) identifies is a feeling of reciprocity towards the union. Most importantly for our second research question, Pekarek’s (2009: 182) study also finds that the presence of an active union in the workplace served ‘as a means of creating opportunities for individuals who displayed a personal disposition for the role’.

The Norwegian context: Short history and characteristics of four confederations

The Norwegian industrial relations context is dominated by the so-called ‘Norwegian model’, entailing an institutionalization of centralized, tripartite cooperation between union confederations, employers’ associations and the state (Hernes, 2006). The model is justified by its facilitation of industrial peace, moderate wage increases and economic development through cooperation. Norwegian unions are thus, like most unions in industrialized countries were in the past, still able to exert pressure through the system of industrial relations through collective action and political influence. However, the question is, as posed by Wilson and Spies-Butcher (2011), whether they have the capacity to shape future industrial relations policy. If the focus of unions is mainly placed on influencing the state and employers’ associations through institutionalized cooperation, unions are rendered vulnerable to shifts in employer and/or state strategy, particularly if institutionalized cooperation is done at the behest of mobilization. This is particularly so if unions become ‘shallow’ through marginalization of workplace union organization (Peetz, 1998; Wilson and Spies-Butcher, 2011), which provides us with yet another reason for focusing on union delegates. As pointed out by Wilson and Spies-Butcher (2011: 310), being ‘enmeshed in law, formal politics and industrial relations’ tends to reduce the kind of mobilization associated with other social movements, and might thus, arguably, foster union apathy. Furthermore, as Bengtsson (2013) argues, holding a strong institutional position, defined in relation to political influence and collective bargaining strength, shifts the strategic choice of unions away from organizing.

In Norway, union density has been around 55% between 1999 and 2011, although it fell somewhat during the first half of the 1990s from 58.5% in 1990.¹ However, density levels vary considerably between different sectors of the economy, being highest in the public sector and manufacturing, and lowest in the private services sector, such as the tourism industry and retail, where union density is below one quarter of employees.² The dominant form of collective bargaining in Norway is negotiation of national collective agreements, in some instances supplemented by company-level negotiations and agreements. It is estimated that the overall collective agreement coverage is about 59% in the private sector and 100% in the public sector (Nergaard and Stokke, 2006). So far, Norwegian unions have opted for merger or amalgamation as a strategy for facing decline (Ebbinghaus, 2003), mostly motivated by survival and hence as a defensive strategy, but some have also shown a more aggressive motivation, especially among the larger affiliates of the Norwegian Confederation of Trade Unions (Landsorganisasjonen i Norge; LO). However, some, especially affiliates of Unio, have started to reveal an interest in the organizing model.

An important point of departure for our research questions is that confederations are not singular in character although they share the same external context, in our case the ‘Norwegian model’. They display differences in both function and structure, and also in the influence they have over affiliates (Behrens et al., 2004b). Norway is an example of a country where there are several (four) competing confederations, but as we shall see, they only compete to some extent on the basis of different political allegiances. Rather, they differ, in addition to their orientation to the Labour Party, in their historical legacy, in their level of centralization, their strategies, their size, the industries their affiliates organize and thus the occupational character of their individual members. The LO (organizing approximately 800,000 workers) is by far the largest confederation. The LO was established in 1899, and it has been the most important player historically. This means that the LO is important for many union members’ identity, for some verging on an equation between ‘union’ and ‘LO’. Although it struggled initially with the competing logics of collective action identified by Offe and Wiesenthal (1980) – between survival, which necessitates mobilization, and success, which to some extent hinges on good relations with employers’ associations and the state – it landed on opportunism as a strategy as early as 1935 with the signing of the Basic Agreement. The Basic Agreement was, however, entered into after a period of severe militancy (Bjørgum, 1985; Knutsen, 1985; Olstad, 2009), and has provided the LO with a strong institutional position. The history of the LO is thus closely bound to the development of the Norwegian model of industrial relations, i.e. of institutionalized cooperation, and the LO is highly centralized, at least in the sense that the confederation is more powerful than the sum of its affiliates. The LO thus has considerable power over its affiliates (Cumbers, 2004). Today the LO, and its affiliates, is largely bent on the servicing model. The LO has close relations to the Labour Party, historically also through collective party membership which was phased out by 1997 (Allern et al., 2007). Today the LO still provides financial support during election campaigns, and since the signing of the Basic Agreement, the LO has been directly involved in the regulation of employment relations (Cumbers, 2004).

The relationship between the LO and the Labour Party gave rise to an alternative confederation, YS (Yrkesorganisasjonenes Sentralforbund, Confederation of Vocational Unions, organizing around 220,000 workers), which in many ways mirrors the LO in structure and scope, but explicitly states its party-political independence. YS, established in 1977, has quite similar strategies to the LO, but lacks the historical clout of the latter, and recent disaffiliations might indicate that as a confederation it also lacks sufficient centripetal forces, apart from party-political independence, to tie the affiliates together. Compared to the LO, YS has historically had a larger proportion of white-collar workers.³ The affiliates which left YS, one organizing machinists and the other radiographers, joined Unio (Confederation of Unions for Professionals, organizing almost 320,000 workers), another independent confederation, but which restricts its scope to workers with higher education, often professionals such as teachers, nurses and police officers. Unio was established in 2001, and is thus the ‘youngest’ confederation and also the most activist, i.e. most willing to take industrial action, run campaigns and draw on militant discourses. As such, the other confederations regard Unio as too conflict-oriented. Furthermore, several of its affiliates have started to show an interest in the organizing model, where organizing is explicitly stated as the way forward instead of merger or amalgamation. One of its affiliates, Union of Education Norway, even explicitly disavows the servicing model in training programmes. Both YS and Unio are less centralized than the LO.

Lastly, Akademikerne (the Federation of Norwegian Professional Associations, organizing about 170,000 workers) also organizes workers with higher education, often on a disciplinary basis, with affiliates such as ‘The Social Scientists’ or ‘The Natural Scientists’, but also some professions such as lawyers, dentists and medical doctors. Akademikerne differs when it comes to union strategy compared to the three other confederations: it is explicitly in favour of decentralization of wage bargaining to the local or even individual level, and it is least activist. As can be seen, in English they have chosen the concept ‘association’ instead of ‘union’ in their title. Akadmikerne was established in 1997.

No Norwegian confederation or union has yet fully adopted the organizing model as such, arguably partly due to stable union density. However, as we hope has become clear, their strategies differ in a way that places them differently on the servicing–organizing continuum. Seen from the perspective of the power resources of unions (Dörre et al., 2009), it would be fair to say that while the LO has the greatest institutional power given its historical institutional embedding, Unio has the strongest organizational power, being the confederation which is most capable of acting.

We sum up the background information for the four confederations in Table 1.

Table 1.

Norwegian confederations by two strategic measures: strength of political orientation and model of unionism.

	Political orientation	Model of unionism
Unio	Moderate – case-oriented and independent from party politics	Bent on organizing, including several of its affiliates
LO	Strong – formal links to the Labour Party	Bent on servicing
YS	Weak – explicitly apolitical and independent	Bent on servicing, although one affiliate has recently shown some interest in organizing
Akademikerne	Weak	Bent on servicing, verging on affiliates being professional associations and not unions

In addition to the model of unionism, we include political orientation, as we assume that a strong political orientation might counter apathy through attracting politically engaged and conscious workers, who, arguably, should have a lower propensity to be apathetic.

Data and methods

Survey design

The data used in the analysis come from four waves (2009–2012) of a nationally representative survey of the working age population (18–68 years), the YS Employment Outlook Survey (n = 9857). This annual survey provides snap-shots of the current state of Norwegian working life, in comparison with other countries, and is an instrument to measure trends and changes, both of which have been presented in annual reports (e.g. Bergene et al., 2012). The YS Employment Outlook Survey covers a wide array of aspects of working life, such as union membership, wage formation, working environment, labour force participation, equal rights and labour market relations.

Data collection procedure

TNS Gallup conducted the surveys in June 2009–2011 and June–August 2012 by means of a web-based questionnaire. The respondents have been randomly sampled among the members of GallupPanelet. The members of GallupPanelet have been recruited randomly via surveys on other topics, and consisted of between 40,000 and 60,000 people in the years 2009–2012.

The sample

TNS Gallup was responsible for drawing the sample for the YS Employment Outlook Survey. In order to secure sample representativeness the gross samples have been stratified according to age, sex, education, place of residence, economic sector and the number of employees in the firm in which the respondent is working. When we compared the gross samples in each of the four waves with the population, we found that men and women with primary or secondary education aged 18–29 years were clearly underrepresented. This was not due to misspecification of the stratification techniques, but rather that GallupPanelet has not been able to recruit enough members with these characteristics (recruitment bias). On the other hand, the gross samples for each wave mirrored the population with respect to people above 30 years of age, sex (independent of age), those with secondary and tertiary education, geography and industrial sector.

Surveys are always hampered by non-responding. The ‘response rates’ (net sample in per cent of gross sample) in the years 2009–2012 were respectively 48.5%, 51.0%, 32.9% and 36.0%. The declining response rate shows that it has become increasingly difficult to recruit respondents.

When the net sample in the 2012 survey was compared to the gross sample, we found that men aged 18–29 years were underrepresented. This means that the underrepresentation of young men in the net sample is due to both the problem of recruiting young men to GallupPanelet and to non-responding. Some differences are worth noting when the net sample is compared to the population. First, union members are overrepresented in the net sample. The share of union members in our net sample is 65% compared to 55% in the population. Second, while members of the confederations LO and YS are overrepresented (LO only by 1.0 percentage point, YS by 6.2 percentage points), members of Unio and Akademikerne are underrepresented (Unio by 3.9 percentage points, Akademikerne by 3.3 percentage points).

In models not shown we included population level weights to correct for potential biases in the distribution of age, sex and industrial sector (this would also partly correct for the biases at the level of confederations) in our net sample. Results in size effects, signs and statistical significance generally did not change as a result of applying the weights.

Dependent and independent variables

Apathy is used as the dependent variable when analysing our first research question, and used as an independent variable when analysing our second research question. To answer our research questions an index of union apathy was constructed using a battery of 12 attitudinal questions regarding unions in general. Similar to Peetz (1997, 1998) we have defined an apathetic union member as a member who answered with neutral (‘neither agreeing nor disagreeing’) or no opinion responses (‘don’t know/not applicable’ or missing) to at least half of the 12 following questions: (1) ‘Is [your] union affiliated to one of the following confederations?’ (LO, YS, Akademikerne and Unio); (2) ‘Unions play an important role in ensuring wage increases’; (3) ‘Unions are important in securing co-determination at the workplace’; (4) ‘Unions play an important role in ensuring that wages are fair and equitable’; (5) ‘Without unions, working conditions would be significantly worse’; (6) ‘Unions are important for job security’; (7) ‘Unions are too powerful’; (8) ‘Unions cause conflicts at my workplace’; (9) ‘The workplace unions contribute in a constructive way to improvements at my workplace’; (10) ‘Unions hamper modernization/reorganization processes at my workplace’; (11) ‘At your workplace, how would you assess the importance of the unions over the past five years?’; (12) ‘In society at large, how do you assess the importance of the unions over the past five years?’

It is important that the apathy index has a high degree of both reliability and validity. In order to test the validity of our apathy index we cross-tabulated the apathy index and ‘not interested in trade union work’ as a reason for not being willing to become a union delegate. It is such disinterest in union matters which shows up in the subsequent lack of opinion (Peetz, 1998). This exercise showed that among the apathetic union members 33.3% of those who did not want to become a union delegate stated this as the reason, while only 14.3% of non-apathetics expressed the same. This result indicates that our index of apathy has some degree of validity.

In the analysis of the first research question, question (1) above is omitted from the index (since confederation is an independent variable in the analysis), leaving 11 questions in the apathy index. A union member is defined as apathetic in this case when answering neutral or no opinion responses on five out of the 11 questions (45.5%). The dependent variable in our analysis of the second research question, ‘would you be willing to serve as a union delegate?’, was coded 1 when respondents answered ‘yes, absolutely’ and ‘yes, possibly’, else 0 (the latter category included those respondents answering ‘no, not at all likely’ and ‘don’t know/not applicable’).

Question (1) above is coded 1 if respondents answered ‘don’t know’ on whether the respondent’s union is affiliated to any of the four confederations, else 0 (the latter category includes those who answered affirmative on membership in one of the four confederations and ‘union not affiliated to any confederation’). Questions (2) to (10) above had six options (‘strongly disagree’, ‘agree’, ‘neither agreeing nor disagreeing’, ‘agree’, ‘strongly agree’, ‘don’t know/not applicable’) and are coded 1 if answering ‘don’t know/not applicable’ or ‘neither agreeing nor disagreeing’, else 0. Questions (11) to (12) above had six options (‘much less important’, ‘a little less important’, ‘about unchanged’, ‘a little more important’, ‘much more important’, ‘don’t know/not applicable’) and are coded 1 if respondent answered ‘don’t know/not applicable’, else 0.

In the analysis of the association between the independent variable, ‘union confederation’ (LO; YS; Akademikerne; Unio; don’t know; not affiliated), and the dependent variable, ‘union apathy’, Unio is chosen as the reference category.

Control variables

Peetz (1997) found either weakly positive or non-significant independent associations of female gender with the risk of being apathetic, and that older workers and full-time workers have an independently lower risk of reporting apathy towards union-related questions compared to younger workers and part-timers respectively. Based on these findings we control for gender, age and full-/part-timers in addition to eight other variables. The 11 variables are: (1) time of interview/calendar year, operationalized as a continuous variable. (2) Age was coded in five 10-year categories: 18–29, … 60–68. The age group 60–68 was used as reference category. (3) Sex (women = 0, men = 1). (4) Education is a grouped variable with five levels: primary, secondary general, secondary occupational, tertiary lower and tertiary higher. Primary school is used as the reference category. (5) Gross household income in NOK is a grouped variable with five categories: below 200,000, 200,000–399,000, 400,000–599,000, 600,000–799,000 and more than 800,000. Below 200,000 NOK is used as reference category. (6) Sector is a grouped variable: private, public, state-owned company. Workers in the public sector are used as the reference category. (7) Where respondents live is coded in four regions: Oslo and Akershus, rest of Eastern Norway, Southwest, Mid- and Northern Norway. Oslo and Akershus is used as the reference category. (8) Working full- or part-time. Part-time is used as reference category. (9) Having a supervisory/managing position or not (the former was used as reference). (10) The number of employees in the company in which the respondent is working is coded in eight groups: 1–4, 5–9, 10–20, 21–50, 51–100, 101–250, 251–500, and over 500; 1–4 employees is used as the reference category. (11) When addressing the second research question we also adjust for whether a member has previously served as a union delegate. Those who have never served in the capacity of union delegate is used as the reference category.

Analytic strategy

We analyse the association between union membership and union apathy in the total population (disregarding union membership status), the association between confederation and union apathy, and the association between union apathy and the willingness to become a union delegate. Because the outcome variables are binary we use logistic regression models to perform the analyses. Statistical significance is set at a level of 5%.

Two models for each outcome are specified. The first examines the univariate relationship between the dependent and the independent variables. To check whether the relationship between union membership/confederation and union apathy is mediated by demographic and occupational characteristics in the second model we simultaneously control for all of our observable control variables. In the same manner, we check whether the association between apathy and the willingness to become a union delegate is mediated by the demographic and occupational characteristics. In the second model we simultaneously control for all of our available observable characteristics.

The design and data used in our analysis do not put us in a position to differentiate cause from effect. We therefore do not claim that the relationships between our dependent and independent variables are causally related.

Results

Association between union membership/confederation and union apathy

In the first step of our analysis we study union apathy in the total sample population, and to what extent union apathy differs among union members and non-union members. Almost one in four in the total population is apathetic, but the share of apathetics is much higher among non-union members (46.8%) than among union members (12.0%) (Table 2). In the univariate model the odds of being apathetic is more than six times higher for non-members relative to union members (OR 6.4, 95% CI 5.8–7.1). Although the control variables explain a substantial part of the difference between the two groups, the odds of being apathetic towards unions is still nearly five times higher among non-members than among members in the adjusted model (OR 4.9, 95% CI 4.4–5.5).

Table 2.

Results from logistic regression showing the risk associations (odds ratios with 95% confidence intervals) for being union apathetic by union membership status in the total sample of Norwegians, 2009–2012.

Independent variable	n	Apathetic (in %)	Unadjusted	95% CI	Adjusted^a	95% CI
Independent variable	n	Apathetic (in %)	Odds ratios	95% CI	Odds ratios	95% CI
Overall	9857	24.12
Union member	–	–
Yes (ref.)	6425	12.02	1.00	–	1.00	–
No	3432	46.79	6.44	5.82–7.12	4.91	4.38–5.51

Adjusted for time, age, gender, education, household income, full-/part-timers, supervisory/managerial position or not, sector, size of firm and where respondent lives.

In the next step, we study the association between union confederation and apathy (Table 3). As we reported in the previous paragraph a little over one in ten union members are apathetic. At first glance it seems that Unio (8.8%) may have a somewhat lower share of apathetic union members than the other confederations and among those who are not affiliated to a confederation, and that the share of apathetics is highest among those union members who do not know whether they are affiliated to a confederation (24.6%).

Table 3.

Results from logistic regression showing the risk associations (odds ratios with 95% confidence intervals) for being union apathetic by union confederation status among Norwegian union members, 2009–2012.

Independent variable	n	Apathetic (in %)	Unadjusted	95% CI	Adjusted^a	95% CI
Independent variable	n	Apathetic (in %)	Odds ratios	95% CI	Odds ratios	95% CI
Overall	6425	12.02
Confederations
LO	3197	10.60	1.23	0.95–1.58	1.02	0.75–1.38
YS	1134	12.96	1.54	1.16–2.05	1.41	1.01–1.95
Unio (ref.)	918	8.82	1.00	–	1.00	–
Akademikerne	422	13.51	1.61	1.13–2.31	1.52	1.02–2.25
Not affiliated	348	13.79	1.65	1.13–2.42	1.50	0.99–2.27
Don’t know	406	24.63	3.38	2.45–4.66	2.69	1.90–3.80

Adjusted for time, age, gender, education, household income, full-/part-timers, supervisory/managerial position or not, sector, size of firm and where respondent lives.

In the univariate and unadjusted model the union members affiliated to the confederations YS (OR 1.5, 95% CI 1.2–2.0) and Akademikerne (OR 1.6, 95% CI 1.1–2.3) have a significantly higher risk of being apathetic relative to the union members affiliated to the reference confederation Unio (Table 3). Members who answered ‘don’t know’ or claim that their union is not affiliated to a confederation have also significantly higher odds of being apathetic. Union members affiliated to the LO have a positive but non-significant association with apathy when other factors are not controlled for.

When we simultaneously adjust for all observable controls, the odds ratios for the confederations YS and Akademikerne are only slightly attenuated and they remain an independent and positive risk factor for union apathy (Table 3). The odds ratio for members who answered that they ‘don’t know’ whether their union was affiliated to a confederation was quite substantially reduced by including the controls, but remained an independent and positive risk factor for union apathy. In the adjusted model those members who are not affiliated to a confederation no longer have a significantly higher odds of reporting an apathetic attitude towards unions relative to members of Unio.

The independent associations between the control variables and apathy among union members are also worth reporting. The odds of reporting apathy is higher over time, for full-timers and those who work in the private sector, but decreases with age, income and size of the firm (not shown).

Association between union apathy and the willingness to serve as a union delegate

One in three of all union members who are currently not serving as a union delegate are willing to take such responsibility in the future (Table 4). However, the willingness to become a union delegate is higher among the non-apathetic (34.9%) than among the apathetic union members (20.9%). This willingness to work for the union measure is the sum of those union members responding ‘yes, absolutely’ and ‘yes, possibly’. The percentage of the members answering ‘yes, absolutely’ (apathetic: 1.6% vs non-apathetic: 4.6%) and the percentage of the willing who answered ‘yes, absolutely’ (apathetic: 7.5% vs non-apathetic: 13.2%) also clearly differ with respect to union apathy.

Table 4.

Results from logistic regression showing the risk associations (odds ratios with 95% confidence intervals) for the willingness to become a union delegate by union apathy status among Norwegian union members, 2009–2012.

Independent variable	n	Willingness (in %)	Unadjusted	95% CI	Adjusted^a	95% CI
Independent variable	n	Willingness (in %)	Odds ratios	95% CI	Odds ratios	95% CI
Overall	5486	33.76
Apathetic
Yes	1852	20.89	0.49	0.39–0.62	0.50	0.39–0.64
No (ref.)	3634	34.91	1.00	–	1.00	–

Adjusted for former union delegate status, time, age, gender, education, household income, full-/part-timers, supervisory/managerial position or not, sector, size of firm and where respondent lives.

In the multivariate analysis the dependent variable is the sum of members answering ‘yes, absolutely’ and ‘yes, possibly’ on the willingness to become a union delegate (Table 4). In the unadjusted model, the results show that the odds that an apathetic union member will become a union delegate are significantly lower (OR 0.5, 95% CI 0.4–0.6). The adjusted model shows that the association between union apathy and the willingness to serve in the capacity of a union delegate is not explained by any of the included control variables; the odds ratio and the confidence intervals for the apathetic are practically unchanged in the adjusted model.

The independent associations between the control variables and the willingness to become a union delegate are also worth reporting. The odds of reporting such willingness decreases over time, with age, and is lower for full-timers and workers in supervisory/managing positions. On the other hand, it is increasing by education and is higher for former union delegates, males and for those who work in the private sector (not shown). In a separate model (not shown) we also included union confederations as a control variable in addition to the demographic controls. We found no independent association between union confederation and the willingness to assume the role of a delegate.

Discussion

Comparison with prior research

Peetz (1997) found either weakly positive or non-significant independent associations between female gender and the risk of being an apathetic union member, and that older workers and full-time workers have an independently lower risk of reporting apathy on union-related questions compared to younger workers and part-timers respectively. Our results are not directly comparable to Peetz’s findings from Sydney, Australia. First, our data are representative for all sectors, occupations and union confederations in Norway. Second, although we use the same amount of union-related questions (12 vs 13), and used the same cut point on the number of neutral or no responses for defining an apathetic person (50% of the questions), we are not sure to what extent the union-related questions are similar. Third, we adjust for a larger number of selective observables, and we also partly adjust for different observables. However, in line with Peetz (1997), we also found that older union members are less likely to be apathetic, but we found a reverse effect for union members who work full-time; in Norway they are more likely to be apathetic than part-timers. We found no gender effect.

Possible explanations of main results

We set out to probe into two research questions, and we now discuss our findings in relation to these. The first research question concerns the relation between different union strategies, here studied at the level of union confederations, and the apathy of their individual union members independently of other factors. In our theoretical framework, we drew on literature from mainly two strands: union renewal and union commitment. In the first strand, it is posited that while the servicing model might have delivered for union members at a certain time in history or in certain contexts, unions need to renew themselves in an organizing direction if they are to counter union decline. We saw that one important ingredient in the turn to organizing unionism is activating members, which ultimately necessitates minimizing union apathy. Peetz (1997) also finds that apathetic members are more likely to exit the union than members who express an opinion on unions, either in positive or negative directions.

In our analysis we found, first, that the odds of being apathetic are nearly five times higher among non-members than among members, all other factors being the same (OR 4.9, see Table 2). This means that union membership may counter union apathy. However, we cannot rule out the alternative explanation that a higher share of non-apathetic persons become union members. Second, and more important for our first research question, we found that even when we adjust for all observables, the confederations YS and Akademikerne remain independent and positive risk factors for union apathy (ORs were 1.4 and 1.5, see Table 3). The same is also true for members who answered that they ‘don’t know’ whether their union is affiliated to a confederation (OR 2.7, see Table 3). We have argued that the confederations, and their affiliates, differ in their strategies in a way that places them somewhat differently on the organizing–servicing continuum and by their political orientation (see Table 1). As stated, YS and Akademikerne lack the strong political orientation and clout of the LO and, especially Akademikerne, are less activist in the sense of being willing to take industrial action, run campaigns and draw on militant discourses, compared to Unio, which was the reference category in our analysis. Unio, which had the lowest level of apathy among its members, has an independent and moderate political orientation, but is the only confederation bent on the organizing model. Although we have not directly measured union strategies, it nevertheless seems to be a relation between the different strategies on the part of union confederations and union apathy.

The organizing model emphasizes that the power of a union is inextricably linked to its ability to mobilize workers in a wider sense than just during strike action, for instance through being willing to work for the union (Fiorito and Jarley, 2008; Gallagher and Strauss, 1991). A measure of this willingness can be found in the extent to which union members would agree to serve in the capacity of union delegate (Fiorito and Jarley, 2008), which leads us to our second research question: apathetic union members are less willing to work for the union, in our case operationalized as less willing to serve in the capacity of union delegate.

Previous research has identified commitment as an antecedent of union participation (Pekarek, 2009), and Bartram et al. (2008) find that union commitment is a strong predictor of becoming a union delegate. In our analysis, we found that the odds that apathetic union members will become a union delegate are significantly lower than for non-apathetics. This result was not explained by any of the observables. In fact, the odds for union apathy remained unchanged in the adjusted model compared to the univariate model. In other words, the answer to our second research question is that union apathy is negatively related to the willingness to work for the union.

Limitations of our study

Our study has some limitations which may have consequences for the interpretation of the results. First, our data are somewhat biased for two reasons. Despite correct stratification of the randomly drawn gross sample, we pointed out that our data suffer from recruitment/selection bias due to the underrepresentation of certain demographic groups. The second source of bias stems from non-responding. Men under the age of 30 are underrepresented due to both recruitment bias and non-responding. Although our analysis showed that union members across all four waves were overrepresented in the net sample compared to the population, the most apathetic union members might be overrepresented among the non-responders because responding to a questionnaire on working life, which starts out with several questions on unions, requires being willing to at least engage somewhat with these issues (Gallagher and Strauss, 1991). Our data do not allow an empirical test of this hypothesis, but if this is the case our estimates of the prevalence of apathy among union members are conservative. Second, the design of our study, and the cross-sectional nature of our data, does not allow us to make any causal inferences from our results. Controlling for observable demographics substantiates the claim that different union strategies could explain (cause) the variation in union apathy (effect). However, even when controlling for selection factors assumed to be associated with apathy, which perhaps are also associated with becoming a union member, unobserved and unaccounted factors could also explain the variation in apathy (e.g. health, personality traits, family background). Third, it is a possible weakness that most of the apathy items address the respondent’s opinions about unions in general rather than about their own union or confederation in particular. Finally, lacking direct measures of strategies we used confederations as proxies for the assumed variation in strategies.

Conclusion

Falling union density has caused concern throughout the western world (Cregan, 2005; Fiorito and Jarley, 2008; Gall and Fiorito, 2012; Johnson and Jarley, 2004; Mason and Bain, 1993; Peetz, 1998; Peetz and Pocock, 2009; Pekarek, 2009; Riley, 1997). Unions have thus been challenged to devise effective responses or strategies in order to recruit and/or retain members (Cregan, 2005). According to Gall and Fiorito (2012: 194, our emphasis), if unions are to renew, sustain and expand organizing, they need to ‘activate significant proportions of inactive members’. As stated in the introduction, Norwegian unions have yet to experience a significant decline in union density, although there are signs of negative developments in the number of workers willing to join a union and the number of union members willing to become union delegates (Bergene et al., 2012). In this article we have also seen that union apathy has increased over time in the Norwegian union movement.

In this article we have explored to what extent union apathy is higher among members than non-members, which factors determine union apathy among members and to what extent union apathy is related to members’ willingness to work for the union. Not surprisingly, we found that the odds of being apathetic towards unions were significantly higher among non-members than among members. To analyse apathy among union members we addressed two research questions. First, we explored the independent relation between different union strategies, here studied at the level of union confederations, and the apathy of their individual union members. Second, we analysed the independent relation between being apathetic and the willingness to work for the union, in our case operationalized as the willingness to serve in the capacity of union delegate. After analysing data from the YS Employment Outlook Survey (2009–2012) based on a net sample of 9857 people of working age (18–68 years), we found that union confederation had an independent association with union apathy, and that union apathy had an independent association with the willingness among members to serve in the capacity of union delegate. On the other hand, we did not find an independent association between union confederation and the willingness to assume the role of union delegate. This strengthens our analysis because it means that it is union apathy, and not the confederation itself, which explains delegate incumbency.

What do these results mean? First, it is hard to disagree with Clark’s (2009: 1) contention that ‘one of labor’s greatest challenges is to mobilize the talents, ideas, and energy of its members’, both if unions are to counter union apathy in an effort at renewal, and, ultimately, ensure their future through having a plentiful, or at least sufficient, supply of potential union delegates. Second, it seems that unions with a more ‘organizing’ soul and/or a strong political orientation are better equipped to counter union apathy among their members than unions more bent on the ‘servicing’ model.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

Author biographies

Ann Cecilie Bergene is senior researcher at the Work Research Institute at the Oslo and Akershus University College of Applied Sciences, Norway. She holds a PhD in Human Geography, and her research interests are trade union strategies, employment relations and industrial relations. She is currently researching changes in employment relations, particularly the growth of fixed-term contracts and temporary agency work, how this affects the politics of work in general and trade union strategies in particular. Furthermore, she is also doing research on how unions navigate and manage professional interests balancing strategies of inclusion and exclusion.

Svenn-Erik Mamelund is senior researcher at the Work Research Institute at the Oslo and Akershus University College of Applied Sciences, Norway. He holds a PhD in Demography, and his research interests span the social and medical sciences, including epidemiology, vaccination, fertility, mortality, social inequalities and family sociology. He is currently doing research on labour market measures targeted towards young workers, evaluation of the core competencies in the Norwegian Labour and Welfare Administration and also how the latter deals with recruiting and placement of workers.

References

Allern

Aylott

Christiansen

(2007) Social Democrats and trade unions in Scandinavia: The decline and persistence of institutional relationships. European Journal of Political Research 46: 607–635.

Bartram

Stanton

Elovaris

(2008) The role of job representatives in an organizing strategy: The case of the Australian Nursing Federation. Journal of Industrial Relations 50: 25–44.

Behrens

Hamann

Hurd

(2004a) Conceptualizing labour union revitalization. In: Frege

Kelly

(eds) Varieties of Unionism: Strategies for Union Revitalization in a Globalizing Economy. Oxford: Oxford University Press.

Behrens

Hurd

Waddington

(2004b) How does restructuring contribute to union revitalization? In: Frege

Kelly

(eds) Varieties of Unionism: Strategies for Union Revitalization in a Globalizing Economy. Oxford: Oxford University Press.

Bengtsson

(2013) Swedish trade unions and European Union migrant workers. Journal of Industrial Relations 55: 174–189.

Bergene

Mamelund

S-E

Steen

(2012) Norsk arbeidsliv 2012: Svekket motstandskraft i gode tider. In: YS (ed.) YS Arbeidslivsbarometer. Oslo: YS.

Bjørgum

(1985) LO og NAF 1899–1940. Tidsskrift for arbeiderbevegelsens historie No. 2.

Clark

(2009) Building More Effective Unions, 2nd edn. Ithaca, NY: Cornell University Press/ILR Press.

Cregan

(2005) Can organizing work? An inductive analysis of individual attitudes toward union membership. Industrial and Labor Relations Review 58: 282–304.

10.

Crosby

(2005) Power at Work: Rebuilding the Australian Union Movement. Annandale, NSW: Federation Press.

11.

Cumbers

(2004) Embedded internationalisms: Building transnational solidarity in the British and Norwegian trade union movements. Antipode 36: 829–850.

12.

Devine

(2009) Organising History: A Centenary of SIPTU. Dublin: Gill and Macmillan.

13.

Dörre

Holst

Nachtwey

(2009) Organizing – a strategic option for trade union renewal? International Journal of Action Research 5: 33–67.

14.

Ebbinghaus

(2003) Ever larger unions: Organisational restructuring and its impact on union confederations. Industrial Relations Journal 34: 446–460.

15.

Fiorito

Jarley

(2008) Trade union morphology. In: Blyton

Bacon

Fiorito

. (eds) The Sage Handbook of Industrial Relations. London: Sage.

16.

Flood

Turner

(1996) Trade union activism: The characteristics associated with becoming and remaining a shop steward. Employee Responsibilities and Rights Journal 9: 103–117.

17.

Gall

Fiorito

(2012) Union commitment and activism in Britain and the United States: Searching for synthesis and synergy for renewal. British Journal of Industrial Relations 50: 189–213.

18.

Gallagher

Strauss

(1991) Union membership attitudes and participation. In: Strauss

Gallagher

Fiorito

(eds) The State of the Unions. Madison, WI: Industrial Relations Research Association.

19.

Gordon

Philpot

Burt

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

20.

Hernes

(2006) Den norske mikromodellen: virksomhetsstyring, partssamarbeid og sosial kapital. In: FAFO (ed.) Fafo-notat 2006:25. Oslo: FAFO.

21.

Johnson

Jarley

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

22.

Knutsen

(1985) Norsk Arbeidsgiverforening, revolusjonsfrykten og de korporative tendenser 1918–1920. Historisk Tidsskrift 64: 369–413.

23.

Mason

Bain

(1993) The determinants of trade union membership in Britain: A survey of the literature. Industrial and Labor Relations Review 46: 332–351.

24.

Nergaard

Stokke

(2006) Organisasjonsgrader og tariffavtaledekning i norsk arbeidsliv 2004/2005. Oslo: FAFO.

25.

Nissen

(1998) Utilizing the membership to organize the unorganized. In: Bronfenbrenner

Friedman

Hurd

. (eds) Organizing to Win: New Research on Union Strategies. Itacha, NY: ILR Press.

26.

Oesch

(2012) Recruitment, retention and exit from union membership: An analysis of member flows in Swiss union locals. British Journal of Industrial Relations 50: 287–307.

27.

Offe

Wiesenthal

(1980) Two logics of collective action: Theoretical notes on social class and organizational form. Political Power and Social Theory 1: 67–115.

28.

Olstad

(2009) LOs historie: Med knyttet neve, 1899–1935. Oslo: Pax forlag.

29.

Peetz

(1997) Why Bother? Union Membership and Apathy. CEPR Discussion Papers No. 357. Canberra: Centre for Economic Policy Research, Research School of Economics, Australian National University.

30.

Peetz

(1998) Unions in a Contrary World: The Future of the Australian Trade Union Movement. Cambridge: Cambridge University Press.

31.

Peetz

Pocock

(2009) An analysis of workplace representatives, union power and democracy in Australia. British Journal of Industrial Relations 47: 623–652.

32.

Peetz

Webb

Jones

(2002) Activism amongst workplace union delegates. International Journal of Employment Studies 10: 83–108.

33.

Pekarek

(2009) Why become a shop steward? Evidence from Australian unions. Labour and Industry: A Journal of the Social and Economic Relations of Work 19: 157–188.

34.

Riley

N-M

(1997) Determinants of union membership: A review. Labour 11: 265–301.

35.

Turner

(2004) Why revitalize? Labour’s urgent mission in a contested global economy. In: Frege

Kelly

(eds) Varieties of Unionism: Strategies for Union Revitalization in a Globalizing Economy. Oxford: Oxford University Press.

36.

Voss

Sherman

(2000) Breaking the iron law of oligarchy: Union revitalization in the American labor movement. American Journal of Sociology 106: 303–349.

37.

Wilson

Spies-Butcher

(2011) When labour makes a difference: Union mobilization and the 2007 federal election in Australia. British Journal of Industrial Relations 49: s306–s331.