A small-N cross-sectional study of British unions’ environmental attitudes and activism – and the prospect of a green-led renewal

Abstract Unions understand the environmental agenda as a technocentric one but also believe it can function as a vehicle for renewal. It is developing slowly, with unions behaving cautiously—resources are scarce. Although popular with members, there is limited evidence that it is effective as a recruitment tool and whilst employers are willing to work in partnership with unions on it, this may confer only phony insider status. Overall, the agenda has limited appeal to the types of employees and employers unions must recruit in order to grow. Identifying a clear environmental premium for members may help.


Introduction
British unions-still haemorrhaging members-are under attack again, this time by a majority Conservative administration insistent on curbing unions' abilities to mount effective political campaigns and industrial action. What is to be done? There is little talk of a "magic bullet" to address unions' current decline. Instead, the renewal literature is characterised by complexity, speculation

PUBLIC INTEREST STATEMENT
Environmental issues constitute an important new organising, negotiating and campaigning agenda for British unions. Unions appear to be motivated by genuine environmental concern but also hope the agenda can function as a vehicle for union renewal. However, the research suggests that as practised, it has limited use as a recruitment tool and neither does it appear to generate any pro-union shift in relative bargaining power. Unions seeking to exploit the agenda in this manner must be prepared to exercise a more muscular green agenda featuring a clear environmental premium incentivising membership-but this will not be easy to achieve in the types of workplaces unions need to penetrate most in order to grow. Nevertheless, unions' participation in the environmental agenda widens their remit and confers legitimacy and may help construct the conditions under which growth may eventually occur. and hedge-betting, offering no quick fixes but, rather, a stultifying menu of options for reform focused on enhancing union efficiency, governance and use of resources (Levesque & Murray, 2006) which offer no easy-to-follow blueprint for change. Even old certainties, such as the need for unions to recruit new members and increase lay activism, are now routinely and fashionably questioned (Hickey et al., 2010, p. 57;Simms & Holgate, 2010, p. 165); something that would surely leave the majority of actual trade unionists mystified. And amidst this confusion, over the years, one too many authors appear to have rather relished spreading doom and gloom-as the Transport and General Workers' Union's former General Secretary Jack Jones once remarked about one researcher: "like so many other academics he wants to be on the side of the workers, but with friends like him, workers do not need enemies-and we have enough of those too! … You do not help workers by writing off the very organisations that those workers, purposefully struggling for a more humane life, have created for this purpose" (Beynon, 2013, p. 26). Academic reputations might be made (and lost) on the ability to construct ever-more complex theories of union decline-but that is not what unions need-or deserve-from us. They need practical advice expressed in clear, understandable and operationalisable terms.
So let me get right to the point. Unions have the potential to play a positive role in tackling climate change and it is right that they develop this. But "the environment" as currently practised by British unions does not appear to be a particularly useful vehicle for union renewal. To any unions contemplating its adoption solely or mainly for this purpose, I say "try something else"-or at least "do it differently". The rest of this article explains why.
The article is based on a survey of 22 unions and summarises British unions' contemporary environmental activism, explores how unions understand the environment, identifies which understanding(s) of the environment associate with environmental activism and evaluates the agenda's potential contribution to union renewal. The environment is one of UK unions' newest bargaining, campaigning and organising agendas, emerging in the mid-1990s alongside equality and diversity and learning and skills. Like these, the environment can be understood as a developmental agenda concerned with "quality of work" issues requiring ongoing maintenance and dialogue with employers. It has, however, developed comparatively sluggishly, possibly attributable to the lack of supportive legislation (as per equality and diversity) and government funding (as per learning and skills). Although Farnhill (2013) understands unions' inchoate environmental agenda by investigating the institutions and resources that unions create to prosecute it, these can be seen as products of unions' environmental activism, not the cause. This article takes a step back and instead investigates how unions' attitudes towards the environment-and the expectations they attach to it-affect the likelihood that they will be environmentally active as well as how they operationalise the agenda. I provide an overview of unions' environmental activism before introducing the rationale and methodology of the survey and the findings-which suggest that unions operationalise the agenda as a technical matter but understand it as a renewal opportunity; and that this self-interest operates alongside environmental concern. This "framing" influences the speed, manner and theatres in which the agenda is adopted and may ultimately be responsible for the limited benefits unions are deriving from it. Further, despite growing evidence that employee awareness of and interest in environmental matters is generally increasing unions' own experience of the agenda as a vehicle for renewal suggests this interest is differentiated, e.g. according to age, gender and occupational status. The requirement to resolve such disagreements here is, however, obviated by the unfortunate truth that the agenda (as practised) appears incapable of generating the orthogonal relations with employers necessary to incentivise union membership. Neither does it provide unions with the right bargaining or organising strategic opportunities necessary to re-establish their centrality in the theatres they need to penetrate most in order to grow: slow-moving accretive agendas-even ones popular with members and encountering little opposition from employers-are insufficient to guarantee growth. But although unions' environmental activism appears to benefit organisations (and their environmental performance) less problematically than unions themselves, it would still be odd for them to fail to engage with arguably the greatest challenge facing this generation.

Situating the article within the literature
As an object of study in itself, labour-environmental relations (LER) theorists typically focus on: the origins and class composition of each movement; their strategies and positioning within the political system; ideological differences; unions' policy-making mechanisms (specifically how they facilitate and constrain environmental policy-making); and sectoral specificities to explain the contingency and patterning of different unions' environmental attitudes (Siegmann,1985;Silverman, 2004). A geographically specific LER has also been treated as a dependent variable within a case study approach investigating LER-specific and generic conditions influencing the conception and maintenance of coalitional behaviour (Diani, 2002;Doherty, n.d.;Hojnacki, 1997;Miller, 1980;Obach, 1999Obach, , 2002. Growing awareness of global climate change has also prompted scholars to increasingly adopt an international perspective and explicitly partisan approach advocating greater union involvement in environmental campaigning and decision-making (Hampton, 2015;Rathzel & Uzzell, 2013).
However, relatively little research investigating unions' environmental activism has emanated from the labour and industrial relations disciplines-in stark contrast with analyses of the impact of unionism and specific configurations of employee voice and representation mechanisms on other variables, including pay, job security, and health and safety (Hayes, 2000;Pe'rotin & Robinson, 2000;Stuart & Martinez Lucio, 2005;Walters & Nichols, 2007). Here, the literature is mainly concerned with assessing unionisation and union engagement in a specific field on that field (such as measuring the union wage premium and/or its effect on wages generally).
In contrast, this article, uniquely, seeks instead to understand the impact of environmental activism on unions. Although the subject matter may therefore be of interest to LER scholars, it is chiefly designed to contribute to the literature on union renewal and in particular the efficacy of developmental agendas as vehicles for collectivism and their behaviour as employee relations negotiables. Farnhill's (2014a) research investigating environmental policy-making at the Trades Union Congress (TUC) 1 since 1967 argues that ideological, demographic and programmatic differences between labour and environmental movement organisations (EMO) have been exaggerated and pro-environmental policy-making has been a growing feature of British trade unionism since at least the 1970s. Nevertheless, throughout the 1980s, unions failed to consistently operationalise their environmental policies and sustained only arm's-length relationships with EMOs because they did not see the green agenda or partnership with environmentalists as facilitative of an industrial rapprochement with the government and were busy concentrating on their core agendas and stemming membership loss Farnhill (2014b). Farnhill argues that as the 1990s progressed, a weakened, introspective and increasingly moderate trade union movement encountered a de-radicalised 2 environmental movement and this facilitated greater collaboration. This new, improved, LER reflected unions' modernisation agendas including social partnership (the desire to form strategic relationships with other key civil society actors); the organising model (the devolution of power to sub-national union tiers to empower increasingly diverse memberships); and workplace partnership (replacing adversarial systems of employee relations with systems predicated on trust and mutual gains, suitable for progressing unions' developmental agendas)-all to facilitate unions' rehabilitation within the UK's policy-making milieu.

The environment-left-wing or right-wing?
Throughout the 1980s, left-wing unions continued to view non-class-related issues a distraction from the socio-economic, industrial and political reforms they considered necessary to advance their members' interests, cementing the environmental agenda's association with the movement's right-wing. However, at the end of the 1980s (in a move interpreted by some as an attempt to kill-off the nuclear sector rather than evidence of genuine environmental concern 3 ), the influential leftwing National Union of Mineworkers (NUM) embraced greater investment in renewable energy and clean coal technology: all of a sudden the environment became a left-wing issue. The pro-nuclear General Municipal and Boilermakers Union (GMB) and the Electronic, Electrical, Telecommunications and Plumbers Union-each representing large memberships in the nuclear sector-were right-wing unions, consolidating the agenda's left-wing status. The nuclear issue exerted considerable influence over unions' attitudes towards the environment at this time, conflating environmental, armsrace, safety considerations and membership interests-a tricky combination promoting inaction. To make progress, in the mid-1990s, the nuclear issue was, according to John Edmonds (former General Secretary of the GMB, TUC Council Member and first co-Chair of the Trades Union Sustainable Development Advisory Committee 4 ), "put in a box" and the GMB emerged as the most environmentally active union. Since the GMB was also closely associated with union modernisation at this time, the environment then became associated with union fashionistas.

Jobs vs. the environment
During the 1970s and 1980s, EMOs viewed unions as a conservative force and were critical of their continued prioritising of economic growth (as the most effective means with which to maintain their members' living standards), despite growing awareness of environmental degradation and the limits to growth thesis. For their part, unions saw EMOs as more concerned with protecting hedgerows than jobs and dominated by the middle classes which contrasted with unions' blue-collar powerbase. Despite the greens' labour-intensive normative future, unions were convinced that "going green" cost jobs, rather than created them. Unions now believe the opposite, but it is difficult to establish when this conversion occurred. Robinson (1992) believes it might have been 1986, when Dr David Clark, the Labour Party's spokesperson on environmental protection, published "Jobs and the Environment", a "smart political serenade to the trade unions" of the impact of environmental policies on the labour market. Certainly by the early-to-mid-1990s, unions therefore also understood the environmental agenda as capable of stimulating employment.

The environment as a developmental issue and union renewal
Production issues are those primarily concerned with how a job is done. Distributional issues are those concerned with the allocation of scarce resources (such as pay). Developmental issues are predicated on a wide set of concrete and abstract needs and values, the maintenance of which valorises ongoing dialogue between unions and employers (Stuart & Martinez Lucio, 2005). Unions' developmental agendas-including the environment, equality and diversity and learning and skillshave complex origins. British Social Attitudes Survey data show that throughout the 1990s, employees believed the most important union functions were the non-developmental issues of protecting jobs and improving pay and conditions of service, but that their importance was already in decline, whilst that of developmental issues remained relatively static. The election of New Labour in 1997 gave a temporary boost to all union bargaining agendas-particularly non-developmental onesbut the trends "corrected" themselves very quickly afterwards, albeit leaving the gap between unions' developmental and non-developmental issues significantly reduced. By the early 2000s, the content of employee relations was just as likely to comprise developmental agendas as productivist and distributive ones.
Unions' developmental agendas therefore emerged at the intersection of new types of concerns and unions' own modernisation policies which created spaces for memberships to articulate them. Alternatively, some elements of the weakened British trade union movement viewed developmental issues--particularly those underpinned by supportive legislation-as ones that they could still campaign around and win on. The counter view is that developmental issues have undermined collectivism by privileging personalised and time-consuming relationships between members and representatives, with the former cast as recipients of expert union advice and advocacy (Amoore, 2002, p. 45), often focused on upholding statutory rights. It is also possible that the agendas are more facilitative of personal (instrumental), rather than collective (covenantal), values, leading some (Donnelly & Kiely, 2007;Mason, 1999, p. 153;Moore, 2011, p. 75) to ask whether adherents represent a new "breed" of trade union activists. Although my analysis (below) suggests that as a motive for going green the recruitment of activists cannot discriminate between environmentally active and inactive unions, my survey shows that nine unions believe the agenda is "Effective" or "Very Effective" at this. This might be because it has intrinsic appeal to certain members (e.g. ethical employees) enabling them to operationalise their personal concerns and validating the private/instrumental thesis. However, Farnhill (2013) argues that the agenda is attractive to inexperienced first-time activists because it is not characterised by conflict and/or because branches adopting it must make concerted efforts to enlist new personnel to do so successfully.
Can the environmental agenda function as a vehicle for union renewal? Danford, Richardson and Upchurch (2003, p. 11) identify three renewal options. Unions can recruit more members and dynamise existing structures (internal expansion); recruit new employers and members in greenfield industries and/or develop new bargaining agendas (external expansion); and/or strengthen their links with civil society (social unionism). The environment provides unions with opportunities to expand internally (by creating a new genre of representative-the union green representative [UGR]) and externally (through the creation of a new bargaining/campaigning/organising agenda to be operationalised in greenfield theatres). It also provides unions with opportunities to develop a progressive campaigning agenda through strategic alliances with non-union actors (such as EMOs) beyond the workplace (e.g. communities).
The environment therefore has the potential to function as a vehicle for union renewal and is capable of being understood differently-as a technocentric or a politically charged one. The key question behind this research is: Which of these understandings is most closely associated with environmental activism?

Unions' contemporary environmental activism
Since 2002, the TUC has organised semi-regular surveys to gauge unions' interest in the environment and their environmental activities. These reveal memberships' continuing concern about the environment, a belief that government is not doing enough, and that environmental protection cannot be left to private enterprises. Although only a minority of respondents report receiving facility time for environmental work, growing numbers of workplaces feature regular management/union discussions on the environment (although much of this is informal, with just 7% of workplaces reporting joint environmental agreements), with unions often the originators of policies. But the results also suggest an implementation deficit regarding operationalisation of members' and activists' enthusiasm for the environment. The overall increase in unions' environmental workplace activism combined with the absence of facility time and rare formal negotiating arrangements suggests that those responsible for the agenda locally are squeezing-in environmental work alongside better established union responsibilities and that much negotiating on the environment remains informaland possibly dependent upon a benevolent management.
My research formed part of a wider project which confirms a consistent, incremental increase in unions' green agenda with their activism now characterised by a diverse action repertoire, including: institution building; workplace greening; lobbying; campaigning; joint meetings with employers; and strategic relations with external environmental actors. According to the wider research, neither is there a clear relationship between industrial sectors and environmental activism, although unions in some sectors may encounter more industry-related environmental issues and this may stimulate wider activism. Larger, multi-sector unions are generally the most active and most activity occurs in large and/or public sector workplaces where the union is already well established. Sectorally, although the salience of "the environment" may increase the likelihood of unions encountering the agenda, some active unions were from industrial sector where "the environment" presses relatively lightly. Sectoral differences in systems of employee relations may therefore also be relevant: where well-established systems exist, unions may find it easier to establish a new workplace agenda; where employee relations are sub-optimal, they may struggle to do so (and remain focused on productivist agendas).
The wider research also revealed that members and local activists appeared interested in the agenda. National union leaderships also appeared interested, but there is a variation in their commitment and unions' sub-national environmental agenda is presently seriously underdeveloped. Overall, few unions evidenced widespread and/or regular engagement.
Inactivity can be blamed on variations in the provision of resources by union headquarters, (perhaps only paying lip service to the principles of the organising model) because union leaderships exert significant control over union's expenditure and strategic direction (which undoubtedly includes whether or not to adopt new agendas). Siegmann (1985), Mason (1999, p. 177), Norton (2004, p. 207) and Yates (2004, p. 349) acknowledge the considerable influence of senior national union figures on policy and strategy. However, despite variation in union leaderships' commitment to the agenda, most appeared interested and approximately half of the unions investigated in the wider research provided environmental training and other resources to local representatives. 5 Low take-up may reflect the failure of branches to engage. Assuming memberships are interested, this suggests the problem is one of the reluctant local executives. This may reflect enviroscepticism and already crowded local agendas. Additionally, Farnhill's (2013) study of workplace greening in three large organisations (a government office, a hospital and a large telecommunications office) found that established local hierarchies could be resistant to new agendas.
Employers' willingness to facilitate unions' participation in the agenda may be important. But although unions themselves cite the absence of facility time as an obstacle to deeper engagement, they frequently cope without it (or with insufficient allocations) for all manner of activities, so it cannot be seen as deterministic. Further, according to unions, employers are relatively content to submit their environmental performance to employee scrutiny.
Where active, unions' environmental agenda takes five broad forms. First, unions are engaging with key environmental arguments and campaigns, helping them acquire legitimacy and traction. Second, unions use environmental arguments to bolster traditional union demands such as greater investment in public transport. Third, unions attempt to enhance various policy domains' environmental component, such as the insertion or strengthening of environmental objectives in the UK's transport and planning regimes. Fourth, unions seek to inject their collectivist values into the environmental agenda itself, by insisting on a fair distribution of the costs and benefits of environmental policies, and democratic systems of environmental decision-making. Finally, unions are engaged in "workplace greening" which union grey materials suggest forms the bulk of their green agenda (Farnhill, 2013). This comprises the site-or employer-specific bottom-up interventions involving union branches-ideally working in partnership with employers-conducting environmental audits to secure measurable energy savings and improvements in organisations' recycling, waste, water usage and procurement policies. It seeks to provide employees with collective opportunities to behave ethically at work, to showcase unions' commitment to the environment and demonstrate their contribution to improving employers' environmental performance. Overall, unions' environmental activism suggests unions can understand the environmental agenda as a progressive and politically charged one and as a practical, technocentric one.

The survey
My survey departs from institutional explanations of their activism and instead seeks to understand unions' environmental activism as a product of their attitudes towards the environment, including the rewards they associate with participation.

Respondents
The questionnaire was administered by post to the EPO of all unions affiliated to the TUC and Scottish TUC. The 22 respondents comprise private and public sector unions and represent 13 industrial sectors, excluding four unions describing themselves as multi-sectoral (Table 1).

Independent variables
Respondents were asked to award a mark (0-10) reflecting how strongly they agreed with 11 reasons for engaging with the environmental agenda. The score (μ) of each reason constituted the independent variable and correlational analysis was undertaken to measure the strength of the relationship between these and unions' overall Environmental Activism Score (EAS). So if the score for The environment is a vehicle for attracting brand new activists increases, does unions' overall activism increase (+ve), decrease (−ve) or stay the same (0)? The reasons themselves were derived using content analysis of 30 personal interviews with EPOs, UGRs and union elites (including union General Secretaries and National Presidents, the former General Secretary of the TUC and the TUC "Green Workplaces" Project Manager).

Dependent variables
EAS is the sum of scores to 18 questions in the survey and is a measure of individual union's overall activism. The questions addressed a range of factors indicative of environmental activism ( Figure 1 and Table 2).

Limitations
The response rate is relatively low, representing just 43% of the 51 unions affiliated to the TUC, although the respondent unions represent over 80% of Britain's 7 million trade union members, a wide range of industrial sectors and a mix of environmentally active and inactive unions. This is a small-N study problematising generalisability (although the correlational analysis does reveal several strong relationships). But this is still a worthwhile enquiry because British unions continue to haemorrhage members and evaluating the contribution of any (relatively) new agenda to union renewal is important. Its main value is that it investigates a seriously under-researched field of union activity for future research to build on. Additionally, the questionnaires were completed by EPOs and it is possible that responses may reflect the thoughts of individuals rather than wider thinking across the union. 6 They may also be calibrated to present their respective organisations in a favourable light. These are occupational hazards and may also be indicative of something corporeal. For example, the findings suggest a "gap" between respondents' stated reasons for going green (unconditional environmental concern) and what the statistical analysis suggests may be happening in reality (the most active unions are those pursuing the agenda for instrumental purposes). To better understand unions' environmental activism, it is recommended that future research incorporates a temporal dimension and secures a larger number of respondents including unions' sub-national environmental actors.

Results
The mean scores for each reason are shown in Figure 2. The most obvious feature is that the three most explicitly instrumental reasons for becoming involved score the lowest, whilst unconditional environmental concern scores the highest.
Something interesting happens when unions' reasons for going green are correlated with their actual activism (Table 3). Pure environmental concern-We've got to do our bit to save the planet, full stop-radicalism-Environmental issues are international and allow us to critique the excesses of capitalism and globalisation …-and the link to health and safety-The Environment is an extension of our health and safety functions-are all NS, whilst the three lowest scoring and explicitly instrumentalist attitudes all correlate particularly strongly with EAS. Three things are suggested. First, unions' engagement is not simply motivated by environmental concern-the more unions believe that the environmental agenda can recruit members and activists and enhance their influence with employers, the greater the likelihood that they will be environmentally active (and vice versa). Secondly, although unions' well-established anti-capitalist and/or international solidarity agenda could quite easily accommodate environmental matters, such radicalism does not appear to be behind their current activism. Thirdly, despite obvious links between environmental issues and health and safety, upon operationalisation, the environment's unique characteristics may emerge.
The "gap" between the raw data and the correlational analysis may reflect respondents' wishes to downplay vested interest and instead emphasise the union "sword of justice" (indeed, other "sword of justice" reasons relating to the "ethical employee" and communitarianism also scored well). Also, Farnhill's research (2013) showed that unions involved in workplace greening can be reluctant to visibly exploit the agenda as a recruitment tool, fearing employer resistance if union interest is interpreted as a recruitment strategy (not just environmental concern) and because some unions' local activists currently believe that to achieve traction, workplace greening must be de-unionised, accommodating members and non-members alike.
Eight reasons, then, appear to be related to union environmental activism and all are significant (p < .05): (1) Environment is a vehicle for recruiting new members (2) The Environment is a growing policy area … employers and/or (potential) members must see us capable of engaging (3) Environment is a vehicle for attracting brand new activists (2) Only the highest relevant mark was awarded, and only once. So, a union which spent "Significant Amounts of Time" on several environmental issues and "Fair Amounts of Time" on several others would still only receive 1 mark. This is designed to avoid discrimination against single-sector unions which may be less likely than multi-sector unions to encounter multiple environmental issues.

Question Options Available mark Cumulative maximum mark available
Employs one or more staff with environmental policy responsibility (4) Environment agenda will continue to develop with or without us-we have to be "in there" influencing it as much as possible (5) Environment is a vehicle for improving or initiating relations with employers (6) We must get involved to maximise "at-work" opportunities to be green and help satisfy people's increasing desire to be environmentally responsible in all aspects of their lives (7) We must get involved to help employers navigate a complex agenda (8) Environment is an opportunity to connect with local communities, young people and those about to enter the workforce Reasons 1, 3 and 5 are examples of "hard" instrumentalism-concrete and immediate gains unions expect from their environmental efforts. Reasons 2, 4, 7, 6 and 8 are examples of "soft" instrumentalism. Reason 2 is concerned with unions' need to be seen to be modern, professional and capable. Reason 4 reflects concerns to retain "insider" status in the environmental policy domain. Reason 7 reflects a unitarist approach to employee relations. Reason 6 suggests unions are convinced of the popularity of the green agenda across society (and their memberships) and the concept of the "ethical employee". Reason 8 reflects unions' communitarianism and ambition to extend their influence to new constituents.
Which, if any, of these eight variables is most able to discriminate between levels of environmental activism? Multiple Analysis of Variance was performed for which unions were divided into three groups: Low EAS; Medium EAS and High EAS. The F-ratio for all four multivariate tests is significant, suggesting attitudes may have a significant effect on which EAS Group unions belong to. 7 There was a statistically significant difference between EAS Group membership and attitudes, V = 1.28, F (16, 22) = 2.42, p < .05. 8 Separate univariate ANOVAs on the outcome variables (EAS Group) revealed the following three attitudes to have a statistically significant effect on activism:

Figure 2. Mean scores of unions' reasons for getting involved in the environment.
(1) The Environment is a growing policy area … employers and/or (potential) members must see us capable of engaging-F (2, 17) = 35.50, p ≤ .005 (2) Environment agenda will continue to develop with or without us-we have to be "in there" influencing it as much as possible-F (2, 17) = 34.72, p ≤ .005 (3) Environment is a vehicle for recruiting new members-F (2, 17) = 32.79, p ≤ .005 The above was followed-up with multiple comparisons 9 , yielding the following results: (1) The Environment is a growing policy area … employers and/or (potential) members must see us capable of engaging …: there was no significant difference between membership of the We must get involved to maximise "at-work" opportunities to be green and help satisfy people's increasing desire to be environmentally responsible in all aspects of their lives Pearson Correlation .420 * Sig.

N 21
Low EAS Group and Medium EAS Group associated with a belief in this statement, although it is lower in the Low EAS Group (p > .05). There is no significant difference between membership of the Low EAS Group and High EAS Group associated with a belief in this statement, although it is lower in the Low EAS Group. There is no significant difference between membership of the Medium EAS Group and High EAS Group associated with a belief in this statement, although it is lower in the Medium EAS Group (p > .05) (2) Environment agenda will continue to develop with or without us-we have to be "in there" influencing it as much as possible: there was no significant difference between membership of the Low EAS Group and Medium EAS Group associated with a belief in this statement, although it is lower in the Low EAS Group (p > .05). There is no significant difference between membership of the Low EAS Group and High EAS Group associated with a belief in this statement, although it is lower in the Low EAS Group (p > .05). There is no significant difference between membership of the Medium EAS Group and High EAS Group associated with a belief in this statement, although it is lower in the Medium EAS Group (p > .05) (3) Environment is a vehicle for recruiting new members: there was no significant difference between membership of the Low EAS Group and Medium EAS Group associated with a belief in this statement, although it is lower in the Low EAS Group (p > .05). There is a significant difference between membership of the Low EAS Group and High EAS Group associated with belief in this statement, which is higher in the High EAS Group (p < .05). There is no significant difference between membership of the Medium EAS Group and the High EAS Group associated with a belief in this statement, although it is lower in the Medium EAS Group (p > .05) The tests therefore reveal eight attitudes which correlate strongly and positively with activism; a mixture of "hard" and "soft" instrumentality. Of these, three were capable of functioning as an indicator of how environmentally active a union is likely to be, although only Environment is a vehicle for recruiting new members could accurately discriminate between groups-and only then to differentiate between the least and most active unions. In sum, unions want to be seen as capable of dealing with environmental issues by members, non-members and employers. They want to become "go to" environmental actors. And they want to use the agenda to recruit new members.

Evaluating the impact of unions' reasons for engaging with the environment on their activism
Unions did not just lose members in the 1980s and 1990s; they lost much of their status as key civil society actors. The environment presents unions with a huge cross-cutting policy milieu to engage with, in the world of work and beyond. Unions have spent over two decades attempting to regain their status as "worthy to listen" and "worthy to speak" and it would be odd for them to fail to engage positively with the greatest challenge of this generation. However, the first two reasons probably explain unions' current technocentric approach-unions striving to appear "professional" and seeking "insider" status are likely to eschew "deep green" thought and campaigning, preferring to depoliticise and mainstream their environmental praxis. Thus, an overwhelmingly practical workplace greening agenda comprises the cornerstone of unions' environmental activism. Since the subnational actors and structures required to advance the agenda take time to mature, this helps explain the agenda's sluggish development.
Because unions also want the agenda to recruit new members, it is mainly focused on workplaces (not communities), targets employers (not government) and is technocentric (with the union variously occupying advisory, co-ordinating and policing roles). This also explains the agenda's piecemeal expansion. First, because resources are precious and unions typically approach novel renewal agendas cautiously. Secondly, because the process of convincing busy branches to adopt a new agenda and ensuing capacity-building is labour intensive. Thirdly, because early results suggest the agenda is not a particularly effective recruitment tool (see below). And fourthly, because unions are currently limiting and/or disguising their use of the agenda in this manner anyway, fearing resistance and suspicion amongst employers and non-members if union participation is interpreted as self-interest (Farnhill, 2013).

Evaluating the environment's renewal potential-influence with employers
The wider research asked unions to award a mark (0-10) to various agendas' tendency to promote "Conflict" (0) or "Consensus" (10) in the workplace. "Environment" scored 6 compared to "Staffing/ Workloads" (3.72); "Pay" (4.11); "Equality and Diversity" (6); "Learning and Skills" (6.39); and "Health and Safety" (6.83). In fact, out of the 16 respondent unions that had attempted to use the environment as a vehicle for partnership, three unions described it as "Ineffective", five as "Neither effective nor ineffective", seven as "Effective" and one as "Highly effective". These relatively positive results may be because unions are currently prosecuting an undemanding, non-adversarial technocentric agenda which employers are comfortable with and the increasing pressure employers themselves are under to go green.

The environment's limitations as a partnership negotiable
The TUC views partnership as the employee relations model of choice. The benefits for unions of partnership-which seeks to replace adversarial systems of industrial relations with a more conciliatory approach-are widely contested (Edwards & Wacjman, 2005;Fernie & Metcalf, 2000;Huzzard, Gregory, & Scott, 2004;Marchington, Wilkinson, Ackers, & Dundon, 2001;Stuart & Martinez Lucio, 2005). The literature asks whether unions can use partnership to increase their influence with employers, or whether employers are using partnership to weaken unions by, for example, opening their doors to rival bodies; using partnership to bring about a dilution in pre-existing consultation/ negotiation practices; informalising consultation/negotiation channels; or eschewing all indirect representation and dealing with staff directly. Partnership pivots around the conduct of the relationship between employers and employees, geared towards the construction and maintenance of a unitarist philosophy-how unions fit in to this modus Vivendi is really a sub-issue (but a substantive one for unions). They may be welcomed (or tolerated) as stakeholders. If, however, unions are perceived to be too weak or adversarial to co-construct dual commitment, they may be marginalised.
Although the environment is not a classic bargaining territory, according to Farnhill (2013), employers recognise the importance of unions in helping to sell and monitor compliance with the behavioural change required of employees in order to achieve many environmental objectives. The TUC notes that most workplaces featuring unionised workplace greening achieve or exceed their environmental targets (TUC, 2014) and this can be further understood through an examination of the literature assessing the effect of contrasting systems of employee representation and participation on the implementation of various discrete environmental initiatives and processes (examples include Kornbluh, Crowfoot, & Cohen-Rosenthal, 1985;Bunge, Cohen Rosenthal, & Ruiz-Quintanilla, 1995;Fredriksson & Gaston, 1999;and Lund, 2004). Whilst a unionised workplace greening agenda is focused on saving employers' money, unions are likely to remain "worthy to speak" and "worthy to listen". But although employers are content to submit their shopfloor environmental practices to union scrutiny and partnership, this does not necessarily result in spillover-an enhancement of union influence in other key agendas-and there is less evidence that employers are willing to submit their wider operations (e.g. product design, logistics and customer services/support) to such arrangements. And Farnhill (2013) notes that whenever unions seek to expand the agenda to include gainsharing initiatives and/or seek to formalise it (by, for example, asking for facility time or attempting to negotiate formal environmental agreements), employers are resistant. This looks very much like phoney insider status incapable of stemming unions' "declining instrumentality" (Levesque & Murray, 2006, p. 7) via a substantive redistribution of power in the workplace and across to unions' key non-developmental agendas.

An unavoidable joint agenda?
Unions believe that the environmental agenda will develop "with them or without them" and they need to be "in there influencing it as much as possible". Environmental concern is certainly pressing harder on employers. Writing in 1998 Doyle and McEachern (1998, p. 136) identified three main types of UK employers-environmental rejectionists, accommodationists and environmental businesses-and that employers practised rejectionism if they could get away with it. By the early 2000s, however, rejectionism was no longer business' default orientation due to the maturation of the ecological modernisation discourse which privileged relatively affordable technical fixes to environmental problems and acknowledged not only businesses' contribution to environmental degradation, but also their status as a repository of the resources and expertise to ameliorate it; and it quickly became the preferred modus operandi (Blair & Hitchcock, 2001, p. 80;Carter, 2007;Mol, Sonnenfeld, & Spaargaren, 2010, p. 34). Since the 1990s, there has also been a rapid increase in the numbers and success of environmental damage limitation and repair businesses and of environmental business services.
There are various indicators of the environment's increasing importance in the workplace since the 1990s and of employers' changing attitudes. The Federation of Small Businesses' annual survey reports declining (but still substantial) levels of dissatisfaction with environmental regulation between 1995 and 2004 (Carter, Mason, & Tagg, 2004, p. 83); the number of UK firms adopting environmental management schemes increased significantly throughout the 1990s and 2000s 10 (Chen, 2004;Kolln & Prakash, 2002) and so too did firms' environmental reporting and corporate governance practices (Gray, Kouhy, & Lavers, 1995, p. 57;Solomon, 2004, p. 52), reflecting the firm's desire to "strategically manage a new and emerging issue with its stakeholders whilst attempting to assess the extent of the power of those stakeholders" (Gray et al., 1995, p. 66). Contact between businesses and EMOs also increased (Janicke & Jorgens, 2010, p. 159). A complex of state regulation, co-operative intervention and self-regulation now exists, preventing employers from practicing rejectionism with impunity and many firms now equate good environmental practice with business growth.
Nevertheless, there remains significant variation in engagement, patterned sectorally, geographically and on a firm-by-firm basis-it is by no means inevitable that unions will encounter the agenda and/or be able to initiate or engage with it. Many small and medium size enterprises (SMEs) undoubtedly still practice rejectionism and/or struggle with environmental regulation (Carter et al., 2004, p. 84;Carter, Mason, & Tagg, 2006, p. 70;Spence & Rutherfoord, 2004). Large firms also vary in their participation, but unions may encounter different problems here-many large firms have been engaging (reluctantly or willingly) with environmental regulation and management schemes for decades and their managements are often significantly more knowledgeable than unions. Here, unions are likely to occupy only peripheral roles in shaping organisations' environmental practices. Generally, according to Rivera Alejo and Martin Murillo (2014), because "the environment" presses unevenly on different sectors, it may be perceived as a threat or an opportunity. When it is understood as a threat, it may be harder for unions to respond positively [although there are notable exceptions including the "green bans" in the 1970s at a time of high unemployment in the construction industry (Elliott, Green, & Steward, 1978) and the GMB union's support for unleaded petrol in the 1980s which impacted negatively on their members in the lead additives industry]. In addition, the problem may not be simply one of engaging with new (environmentally friendly) technologies and processes but, rather, disengaging from well-established (environmentally damaging) existing ones.

Overall evaluation
So here is the problem. With employers engaging positively with the agenda, it is not automatic that unions will be afforded privileged roles, whilst with employers practicing rejectionism unions will struggle to even get the agenda off the ground. SMEs-responsible for 43% of industrial pollution in England and Wales and 60% of all commercial waste (Vickers, Vaze, Corr, Kasparova, & Lyon, 2009, p. 15)-are more likely to practice environmental rejectionism. SMEs are where unions need to penetrate most in order to grow 11 , but SMEs are notoriously anti-union (and their employees the least receptive to unionism in general). Although environmental interventions aimed at SMEs are therefore essential to the transition to a low carbon economy, the green agenda is not easily asserted there and unions may not be the best actors to assert it. Even if unions eschew SMEs and pursue green-led growth in larger private sector organisations, there is no guarantee that unions can easily occupy influential roles-particularly in sectors where the agenda is viewed as threatening and whilst unions are still capacity-building. And although some unions consider the agenda suitable for partnership, the limitations and disadvantages of partnership identified earlier remain. From a renewal perspective, the environment thus appears to be strategically problematic. Unions believe the "environmental agenda will continue to develop with or without us-we have to be "in there" influencing it as much as possible". Actually, in many workplaces, it may hardly develop at all and unions' ability to influence it may be minimal and contingent.

Evaluating the environment's renewal potential-recruiting new members
Unions were invited to award a mark (0-10) to 10 recruitment arguments (Figure 3). The environment trails unions' traditional reasons for joining-those related to personal representation and protecting jobs and pay-and lags behind equality and diversity and learning and skills. It is also, surprisingly, considered to have less appeal than the provision of free or discounted financial and legal services. Eleven unions have experience of using their environmental agenda as a recruitment tool and have found it to be of limited utility, with only two describing it as "Effective". This sits uneasily alongside the findings reported earlier, showing that unions did view the environment as a vehicle for recruiting more members, although some of the difference might be explained by unions with no experience of using the environment in this manner anticipating its effectiveness and those that have (but found it wanting) remaining optimistic.
Unions will attract new members if: there is a perceived injustice; it is experienced by a relatively large number of employees; it is considered to be serious; there is an identifiable source of redress (typically senior management); and the union is perceived to possess the resources and skills to address it (Kelly, 2005). The wider research revealed that employees generally support union participation in environmental matters, but the agenda fails to generate the sufficiently sharp differences between employees and employers required to incentivise membership.
The survey also asked unions with experience of using the environment as a recruitment tool which groups of non-members it most appeals to (Figure 4). Respondents claimed that the environment appealed most to young workers, skilled and professional employees and women. These findings can be analysed from three directions. First, why do these groups respond more positively to unions' environmental agendas? Secondly, are these the types of non-members that unions need in order to grow? Thirdly, how well placed is the environment as a vehicle to achieve this?

The appeal of the environment to different groups of workers
Unions consider workers in low-skilled, low-paid and/or vulnerable employment to be less enthusiastic about the environment. Although young people, as new entrants to the workforce, may also occupy such jobs, they are, in contrast, thought to respond positively to the agenda, possibly because they have grown up with environmental issues. Women, too, may occupy low-paid vulnerable employment, but some essentialist ecofeminist discourses argue that women "naturally" care more for the environment, typically linked to their roles as mothers and care providers [see Mellor (1992) for a comprehensive, but ultimately unsympathetic account of the arguments]. Well-educated, professional employees, meanwhile, may be constructed as most capable of understanding the science behind it (Witherspoon & Martin, 1993) and/or may occupy jobs where "the environment" is especially salient.
Occupational and class-based analyses of environmental concern are confusing. In a seminal study, Lowe and Goyder (1983, p. 12) claimed that environmental concern was generally smoothly distributed across society. Yet, others have insisted that support for environmentalism is strongest amongst the socially detached (including young people; the unmarried; the unemployed; and students) and certain counter-cultural groups-none of which (then, or now) typified union memberships-and a particular stratum of the middle classes (those who are financially comfortable but feel excluded from the benefits of liberal capitalism) (Byrne, 1997, p. 67;Cotgrove & Duff, 1980). However, Rootes (1995, p. 235) argues that environmentalism's appeal to the socially detached has been exaggerated and it is not that the middle classes are especially environmentally conscious, but, rather, better placed to operationalise their concern because they possess more resources and reside in communities characterised by established traditions of civic engagement.
Alternatively, Burningham and Thrush (2001, p. 2) observe that poor and disadvantaged groups construct and understand "the environment" differently from most green activists-indeed, they rarely use the word. Disadvantaged communities are concerned about climate change even if they experience difficulty with the science of the environment. However, they are more likely to prioritise  the loss of green play-spaces, graffiti and dog shit in the local park. Burningham and Thrush argue that EMOs have historically failed to engage effectively with such groups.
Class' impact may be more pronounced on individuals' capacities to be green, the extent of the "sacrifices" they are prepared to accept to be green and in shaping opinions regarding the nation's spending priorities (Guber, 2003, p. 177). The contribution of "class" to both environmental activism and consciousness is at the very least contested, and cannot be resolved here. I merely state what unions with experience of using the environment as a recruitment tool report happening.

What types of members must unions recruit in order to grow?
Unions certainly need to recruit new entrants to the workforce just to stand still. But for growth to occur, unions must penetrate the private sector and, in particular, SMEs, which employ large numbers of low-and semi-skilled workers on vulnerable contracts (including large numbers of women and black and minority ethnic employees). Although hard-to-reach non-members in SMEs are considered to respond more positively to organising and bargaining strategies centred on non-developmental agendas, some research suggests that employees generally are increasingly interested in acting ethically at work (The Work Foundation, 2002, p. 14) and a survey of 300 UK small business leaders in 2005 cited employee pressure as the third most common source of pressure to go green, behind "Regulation" (37%) and "Customers" (24%); level with "Shareholders/Investors" (18%) and ahead of: "Local Community" (16%); "Environmental Groups" (12%); "Media" (12%); "Banks/ Insurance" (11%); and the "Public" (11%) (ENDS Report 250, 2005 Novmber). Nevertheless, according to unions, low-and semi-skilled employees and those on non-traditional contracts are the least interested in environmental matters. Again, the environment appears problematic strategically as a vehicle for renewal. Danford et al. (2003), Haiven (2006) and Waddington (2000Waddington ( , 2006 insist that unions must respond positively to new potential sources of collectivism and collective identities by creating efficient structures and systems of governance through which they can be expressed. Although the concept of the "ethical employee" may have the potential to function in this manner, environmental concern may not be strongest amongst the employees (or employers) unions need to recruit in order to grow. Further, as practised, it does not prompt the orthogonal relations between employers and employees required to incentivise membership and apart from any "feel good" factor associated with belonging to an organisation that takes its environmental responsibilities seriously, there is no clear product for employees. There could be-unions could, for example, demand that savings from reduced energy bills be used to improve wages, thereby mainstreaming environmental considerations into their broader pay and reward strategies-but unions are currently reluctant to exploit the agenda like this, fearing employer resistance if they interpret union participation to simply be a recruitment strategy and motivated by anything other than genuine environmental concern.

Overall evaluation
This need to downplay self-interest may be behind unions' conservative technocentric agenda which, at branch-level, accommodates non-members (qua management-appointed "green champions", for example) and/or non-union bodies (such as carbon clubs)-casting the union as the coordinator (of environmental reform) rather than its beneficiary and/or as one actor amongst several (Farnhill, 2013). It is difficult to imagine unions compromising ownership of their productivist and distributive agendas-or even their other developmental agendas-in this way.

Conclusion
What, then, of the apparent gap between EPO's reasons for unions' adoption of the agenda (unconditional environmental concern) and the statistical evidence suggesting the most environmentally active unions are those motivated by instrumental factors?
The most obvious explanation is that respondents were downplaying union instrumentality in order to present their organisations in a favourable light. This may be all (or mainly) about union renewal. Indeed, at a time of continuing decline, much union activity is. Therefore, when new agendas come along, it should not come as a surprise to find that they end up being understood and operationalised in instrumental terms. Still, there is no prima facie reason to suppose that genuine environmental concern and instrumentality cannot coexist. Branch-level actors, for example, enjoy significant levels of Lipskian autonomy and the patterning and slow spread of workplace greening initiatives suggest that it is only adopted by branches that are committed to the agenda-in other words, environmental concern precedes (and operates alongside) the agenda's exploitation as a vehicle for growth. This is not to argue that different union actors at different levels of the organisation all understand the agenda in precisely the same way. Crudely, for target-driven headquartered EPOs, the agenda looks successful if it is being widely adopted. For busy branch activists however, a successful agenda is one which aligns with local organising and negotiating priorities. The observations of one frustrated, mid-level TUC project worker, responsible for collaborating with individual unions to roll-out workplace greening, are instructive: "we shouldn't have people travelling hundreds of miles around the fucking country to speak to non-union members about how they can reduce their employer's electricity bills. I'll not do it. This should be about capacity building and increasing the influence of unions in the workplace in a growing agenda" (personal interview). Positioned "in-between" union EPOs and UGRs, the project worker was clearly not at all motivated by pure environmental concern and simultaneously disenchanted with the purpose and manner in which the agenda was being rolled-out and implemented. All of this serves as a reminder that unions are polyarchic, with multiple centres of authority.
This article has not attempted to understand the causes and patterning of variation in environmental activism that exists-why some unions are more active than others. Instead, using a broad brush, I have attempted to understand how unions' attitudes towards the environment affects their environmental activism. Certainly, the way unions perceive particular agendas and the expectations they have for them influence the theatres and manner in which they are operationalised. Unions can understand the environment as a cross-cutting politically charged agenda necessitating radical industrial reform and as a conservative technocentric one focused on carbon management policies in individual workplaces. The latter is dominant and clearly considered more compatible with unions' renewal agenda. However, the environment appears to be a curate's egg-although popular amongst employees, there is limited evidence that it is effective as a recruitment tool and although employers are willing to submit their environmental performance to union scrutiny and work in partnership with them, this may only confer phony insider status. Further, because unions want the agenda to function as a vehicle for growth, they are behaving cautiously when deciding how quickly to adopt it-resources are scarce and unions already expect a lot from under-pressure lay representatives. In fact, unions appear to have decided to eschew implementing a particularly muscular environmental agenda with employers, fearing employer resistance if union participation is seen to be pursued in order to grow the union. Accretive, popular agendas, facilitative of good employee relations, clearly do not automatically guarantee growth.
Identifying and prosecuting a clear environment-related premium for members-less "sword of justice" and more "vested interest"-might help a bit here. But unions' "framing" of the environment needs to be reversed-by understanding the agenda as a renewal opportunity whilst operationalising it as an apolitical and largely technical one, unions (still relatively inexpert in the field) risk remaining peripheral environmental actors and simultaneously limit the agenda's potential to generate attractive products for members. Alas, for unions, the agenda may be strategically unrewarding anyway-by their own admission, it does not seem to appeal strongly to the members they need to attract in the workplaces unions need to penetrate in order to grow.
These results, then, confirm an inchoate environmental agenda that has developed in less than ideal conditions: unions other faster-moving and more successful developmental agendas benefitted from supportive legislation and generous government funding and the absence of these must surely explain (some of) unions' piecemeal adoption of the environmental agenda, the manner and theatres in which they are operationalising it and the (limited) success they are having. The agenda is moving relatively slowly and it is taking time to implement and refine the training, resources, governance and operational methodologies required to exploit its effectiveness as a renewal strategy. One suspects that if any trade unionists are reading this, the conclusion that the environment does not represent a "magic bullet" for unions' woes would, however, be an unremarkable one. But it would be even more remarkable if unions failed to engage with the greatest threat faced by this generation. The environmental agenda may not constitute a "magic bullet" for union growth. But it expands their field of competence, increases their relevance and confers legitimacy-helping, at least, create and sustain the conditions within which growth might eventually occur, even if this appears, for the time being, a long way off.