Abstract
This article reports on pre and post interviews with trade union leaders and senior Trades Union Congress (TUC) and union officials who held campaign responsibilities for the 2016 UK European Union (EU) Referendum. The article considers the development of union policy towards the EU, the determination of unions’ final positions, campaign resources and media, the arguments made and the drivers of and constraints upon active campaigning. Campaign intensity, resourcing and strategic decision-making varied widely across unions and was frustrated by resource constraints, fear of alienating members and in some cases lack of priority. The article concludes that unions must be better prepared to commit material resources and national officers’ time so that campaign issues are effectively framed in terms of member concerns.
Introduction
Despite the narrow margin of the Leave victory in 2016 and the critical role played by working class voters in the result (Clarke et al., 2017), little attention has been paid to the trade union role in the Referendum campaign in either academic research (Mckenzie, 2017) or popular accounts (Oliver, 2016). A similar inattentiveness marked the trade unions’ role in the 1975 Referendum on entry into the European Economic Community (EEC): no studies were published in leading industrial relations journals of the time with only one account subsequently identified (Whyman, 2008). Our research sought firstly to examine the key influences on both material and argumentative resources in trade unions on both sides of the Referendum campaign and the UK union federation, the Trades Union Congress (TUC). Secondly, we examined trade union post-Referendum evaluations of the effectiveness of their campaigning.
Whilst the pledge for a Referendum had been made in the 2015 Conservative manifesto with paving legislation passed later that year, the date of the Referendum remained a closely guarded secret until its announcement by Prime Minister Cameron on 20 February 2016 (BBC, 2016). Although most union policy positions favoured remaining in the European Union (EU), the TUC and many leading unions had reserved their position on the Referendum until the outcome of negotiations on revisions to British membership. In the event, these were concluded days before the announcement of the Referendum date. The absence of further dilutions of workers’ rights in the agreed package was sufficient to convince the TUC and the majority of affiliates to declare for Remain.
The narrow timescale thereafter gave trade unions and other interested parties two months to prepare campaign plans and materials before the formal start of campaigning on 15 April. The confirmation of the formal lead organisations (Britain Stronger in Europe and Vote Leave) by the Electoral Commission took place just two days before the start of campaigning. Over the next eight weeks, campaign expenditure reached some £32.6 million (Electoral Commission, 2019) and the result saw the highest vote ever recorded in the UK with 16,141,241 Remain voters defeated by 17,410,742 voting Leave. The 123 registered campaigners included just eight of the TUC’s 50 affiliated unions. This is somewhat misleading however as registered unions spent over £1 million, overwhelmingly supporting Remain. Though, perhaps expressing unions’ weight in the overall campaign, this was 5% of total registered Remain spending and only 0.3% of Leave spending (Electoral Commission, 2019). Of the Trades Unions’ 6,493,000 members (TUC, 2017), approximately 60% (3,916,000) belonged to unions that incurred spend in the Referendum.
This article provides findings from synchronous interviews conducted with 11 senior officers responsible for trade union and TUC campaigns during the Referendum and post-Referendum. These insider accounts examine the formation of trade union policy, campaign media, structure and themes, influences on decision-making and post-Referendum evaluation of campaign effectiveness. The next section outlines the history of UK trade unions’ policy towards EU membership and considers explanatory accounts; the third section presents methods and purposive sampling procedures; the fourth and fifth sections present the main findings from the interviews conducted during and following the Referendum and the final section discusses the results and draws conclusions for future union campaigning.
Trade union policy and the European Union
Trade union policy towards the EU underwent a significant shift from the late 1980s. The TUC and an overwhelming majority of unions had opposed Britain’s membership in the 1975 EEC Referendum. At that time, the developing European single market threatened sectoral and national bargaining structures, which then dominated wage setting (Brown, 1985; Whyman, 2008), supporting unions and employers associations alike (Gooberman et al., 2019). From the mid-1980s there has been a decline in union membership, density and collective bargaining, often precipitated by defeats in significant industrial disputes, a hostile state and a far less conducive legislative environment (Coderre-LaPalme and Greer, 2017; Freeman and Pelletier, 1990). Most recently the Trade Union Act 2016, the same year as the Referendum, continues the ‘authoritarian style of Conservative ideology and statecraft’ (Bogg, 2016: 299). Bogg’s legal argument about the repercussions of this Bill for trade unions suggests that British liberal democracy is in a fragile state (2016: 336). In this context, the EU has appeared to offer some protection and influence for workers and unions.
The 1988 address of the then President of the European Commission, Jacques Delors, to the TUC Congress is rightly regarded as a reference point for the growing significance of the EU to UK unions. Mitchell (2012: 42–43) argues that the TUC was ‘one of the most anti-EC union organizations in Europe, since 1988 the TUC has become one of the most vocal supporters’. Hyman (2008) discusses the importance of all types of EU social legislation noting ‘that British labour law today would be very different but for EU membership’ (2008: 27). However, support for ‘social Europe’ and the corporatism that influenced the extension of European Works Councils (EWCs) was never universal. If we accept the accuracy of Mathers et al.’s (2018) description of British unions as pragmatic Europeans (see also Hyman, 2017), then the positive relationship between trade union support for the EU and the protections afforded by it becomes clear. Overall, though, three main strands underpinned trade union policy towards the EU – policy and lobbying in Brussels, transnational campaigns and social dialogue.
Firstly, Mitchell (2012: 39–40) provides details of the TUC office established in Brussels in 1993; a Europe Monitoring Group who conveyed union European positions to this Brussels office; and a Network Europe Contact Point to inform unions and the TUC leadership about relevant European issues (2012: 39–40). She discusses the changing role of the TUC in Europe as it lobbied differing interests in Europe arguing that the:
. . . TUC’s European strategy, first presented in its 1988 policy paper ‘Maximising the Benefits Minimising the Costs’, centered on ensuring that the European social dimension was developed to the greatest extent possible and that unemployment resulting from the transition to (and increased competition from) the Single Market would be as insignificant as possible. (Mitchell, 2012: 42)
The multi-sector GMB and Graphical, Paper and Media Union (GPMU) also had Brussels offices (Hyman, 2017). However, many, especially since the increasingly neoliberal direction of EU policy, have questioned the effectiveness of union lobbying efforts compared to multinationals. For example, Waddington (2011: 208) reports that ‘Unionists grudgingly acknowledge[d] the efficacy of the employers lobbying of the Commission to dilute the terms of the Draft [recast European Works Council] Directive’. This reduction in union influence is also reflected at a national level. According to Culpepper and Regan (2014: 743), governments in the European periphery may not believe that ‘unions are worth the trouble of bringing in to privileged negotiation. They can be treated as just one interest group among many.’
A second mode of engagement has been through transnational European campaigns. Although Bieler and Erne (2014) note that a European trade union campaign with regard to EU austerity reforms had been difficult to organise, they highlight the success of factory workers in France, Greece and Italy. More specifically, Fox-Hodess (2017) discusses the European dockworkers union campaigns. Here Portuguese, Greek and English engaged in ‘an institutionalized form of rank-and-file internationalism’ (2017: 627). Interestingly, the UK (Liverpool dockworkers) was central to this from the start, with local workplace campaign coordination critical. Overall, though, problems aligning national union interests frustrate international cooperation, as Hyman (2017) notes when discussing the role of UK unions during the formation of the European Trade Union Confederation. Martínez Lucio (2010: 541) has also recently commented ‘labour internationalism(s) . . . is (are) born out of capitalism and its contours’. He notes a complex and multi-layered process based on four main dimensions: the sectoral and labour process; the ideological context; the character of networks, so their logic and form; and the national identities of participants, and ‘the regulatory politics of their contexts’.
A third mode of European engagement has been through social dialogue based on EU Directives (e.g. the EWC Directive) and framework agreements through European Social Dialogue (ESD) and European Sector Social Dialogue (ESSD) (see Begega and Aranea, 2018; Marginson and Keune, 2012). Collaboration here has extended beyond leadership and senior officers to involve junior officers and ordinary members. This has encouraged the development of formal and informal representative networks leading in some cases to successful solidarity and recruitment campaigns. For example, Bieler and Erne (2014: 165) highlight a successful transnational campaign that involved 40,000 European workers staging a one-day strike against the closure of a GM plant in Luton (UK). Fitzgerald and Stirling (2010: 321) recount a successful EWC influenced recruitment campaign at a call centre in Sunderland (North East of England), a Unison respondent here states ‘nationally we could do little, internationally we achieved everything’. Whittall et al. (2017) discuss Volkswagen’s transnational company agreements (TCAs) negotiated jointly by their EWC and Global Works Council. These TCAs related to both labour relations and temporary agency work and were highlighted as a positive development at the Bentley car plant in Crewe (North West of England). Interestingly, the UK locations noted here each had EU Referendum Leave votes returned (Luton 56%, Sunderland 61% and Crewe 56%). For members then perhaps the free movement of labour has overshadowed these potentially positive European engagements, influencing the vote to Leave the EU (see Goodwin and Milazzo, 2017 for a wider discussion of immigration and the vote). The May 2004 accession of Central and Eastern European (CEE) workers was significant as by June 2016 some 3.2 million CEE workers had been issued national insurance numbers (DWP, 2004–2016) and were able to work.
To present this in terms of a framing analysis (Goffman, 1974), austerity in working class communities, and workplaces and pressure on public services (Mckenzie, 2017) comprised the fundamental contextual issue. This was falsely attributed to migration by government, politicians and news media (Goodwin and Milazzo, 2017) and allied to concerns about funding paid to the EU. As Hyman (2010: 6) notes, ‘referendum campaigns themselves may be of critical importance in structuring perceptions and in framing the questions at issue’. The Referendum provided an opportunity for this framing to be embedded in working class communities by animating the Leave vote. For according to Leave, the solution was to take back control of borders and leave the EU (Goodwin and Milazzo, 2017).
By contrast, unions found it difficult to adequately frame the migration issue. They were inclusive of CEE workers (Fitzgerald and Hardy, 2010) including creating local union structures and support to encourage CEE union membership (see e.g. Aziz, 2015; Ciupijus et al., 2018). However, challenges were also evident, including perceived special treatment provided to particular groups (Mustchin, 2012). The most publicised example of localised trade union opposition to free movement came with the ‘British Jobs for British Workers’ dispute at the Lindsey Oil Refinery. Here seemingly strong xenophobic attitudes were expressed by trade union members in the GMB and multi-sector Unite. This dispute, though, was part of a longer running issue with the neoliberal turn of the Posting of Workers Directive (PWD) (see Fitzgerald, 2010) and significantly, as noted by Meardi (2012), was supported by some Italian trade unionists that understood ‘the British protesters’, as well as actually including ‘locally resident Polish workers’ (2012: 113). This PWD neoliberal turn chimed with the ongoing left critique of neoliberal Europe (see discussion of Culpepper and Regan, 2014). This echoes the market promoting features identified by unions opposing membership in 1975 (Whyman, 2008). Recent, powerful evidence here includes a limit to state subsidy and enforced competition, such as the introduction of the Fourth Railway Package two months prior to the UK EU Referendum (see De Francesco and Castro, 2018 for further discussion). In any event, as the EU expanded to take in low-wage and poorly organised economies from the 1990s, so the vulnerability of ‘social Europe’ to ‘social dumping’ increased and with it the potential support for Brexit. Waddington (2005) and Rubery (2011) discuss this vulnerability, identifying tensions between the European social model and the wider neoliberal policy framework. Rubery concludes that there is a ‘reconstruction of social models . . . and deconstruction to implement neoliberalism’ (2011: 671).
Overall, though, this does not mean members move to the right. Mosimann et al. (2019) highlight union membership can be a barrier to support for radical right-wing parties. They detail studies differentiated by country, class and data method such as the European Social Survey (ESS) and the Comparative Study of Electoral Systems. Their survey (involving some 1839 UK respondents) found union members less likely to support right-wing parties such as the UK Independence Party (UKIP). Béthoux et al. (2018) do suggest that both French and Irish unions and workers have become more Eurosceptic. Whereas Hyman’s (2010) analysis of European integration initiatives noted that surveys of ‘core’ union members (manual and to some extent white-collar workers) have often opposed EU integration, while union leaderships have become more pro-European. He concludes therefore that members are ‘clearly out of step with the policies of their unions’ (2010: 21). It is against this background that we set out to investigate the thinking of union leaders before and after the UK EU Referendum.
Method
Our research aims were to examine: (1) the key influences on material and argumentative resources in trade unions on both sides of the Referendum campaign including the UK main union federation, the TUC; and (2) post-Referendum evaluations of the effectiveness of trade union campaigning. This required us to generate synchronous data from respondents who were central to the trade unions’ Referendum campaigns. In line with similar small-scale, phenomenologically informed research, whilst we presupposed that our data might enable theoretical insight we neither aimed for nor claim generalisability to wider populations.
These aims required a purposive sample (Guest et al., 2006; Patton, 2015), comprising senior trade union officers with campaigning responsibility at national and regional levels. The common constraints of credibility, time and resource that attend studies with key informants (Patton, 2015) were particularly acute. Since accountability structures and responsibility for political campaigning vary across unions and research data needed to be captured during the Referendum campaign, the first named author exercised considerable flexibility to undertake interviews when respondents enjoyed limited availability. As is common with purposive sampling, brokers are necessary to facilitate high-level access and a senior TUC figure undertook this role. Although the numbers of respondents (11, see Table 1) is slightly lower than is common in work-based phenomenological studies (Saunders and Townsend, 2016), the seniority and credibility of respondents enabled the initial research objectives to be met. In line with the Ethical Approval granted by our institution, respondents were offered the opportunity of anonymity. All but three waived this and whilst some respondents are identifiable by role, those who asked for anonymity are not. Initial interviews were conducted in the immediate run-up to the Referendum in May and June 2016 and a ‘1’ signifies quotes from these in the findings. Follow-up interviews were undertaken following the Referendum in late 2016 and early 2017 and a ‘2’ signifies quotes from these in the findings.
Research respondents.
Following a merger with the Association of Teachers and Lecturers this became the National Education Union in 2017.
The respondent had become General Secretary when interviewed for the second time.
BECTU merged with the Prospect Trade Union in 2017.
Semi-structured interviews enabled the collection of descriptive information whilst also offering an opportunity for respondents to examine relevant rationales. Questions examined policy formation and influences, campaign scale, campaign media, structure and themes. The questions for the second set of semi-structured interviews were more limited, offering interviewees the opportunity to reflect on their campaigns and to evaluate their effectiveness. Whilst a common bank of interview questions was used, some variation enabled particular issues from respondents’ first interviews to be reconsidered. Analysis of data was conducted in a non-linear and iterative fashion with insights generated concurrently through the data collection process. Themes that emerged strongly from particular interviews (e.g. through repetition and emphasis) were checked against the results of others to examine areas of similarity and difference. The themes that emerged strongly as the key influences on material and argumentative resources in the first set of interviews then informed the generation of questions in the second set. These themes are:
Theme 1: Purpose – Pragmatic Europeanism.
Theme 2: Member – leader relationships.
Theme 3: Communications – Messaging and campaigning.
Findings: The Referendum campaign
Pragmatic Europeanism
Social Europe and two of the three main strands discussed (EU policy and social dialogue) were noted by a number of our respondents. However, Mathers et al.’s (2018) description of mainstream British unions as pragmatic Europeans for whom policy towards Europe depends upon a judgement about members’ interests can be confirmed in this research and echoes throughout the findings. As the likelihood of a Referendum strengthened from 2010, the TUC and affiliates developed and/or confirmed policy through successive conferences, with the RMT’s Pro-Leave motions to TUC Congress heavily defeated in 2012 and 2013. Following the election of a majority Tory government with a manifesto commitment to a referendum, two motions appeared at the 2015 Congress with USDAW supporting Remain but the GMB argued for a pause to await the outcome of Prime Minister Cameron’s renegotiation of Britain’s membership. Neither motion was pressed as Congress endorsed a General Council compromise but in the event, most unions followed the GMB’s ‘wait and see’ approach with a number postponing their decisions until their spring conferences just weeks before the Referendum. At that time, trade unions were also dealing with the draft Trade Union Act 2016. However, this was not highlighted as an issue that stopped or interfered with the launch of campaigning. In fact only the BFAWU respondent highlighted the Act, stating that it was evidence of the failure of social Europe. The mainstream unions and TUC policy of ‘wait and see’ kept open the option of Leave and was consistent with their longstanding pragmatism: So we have looked at the EU and asked if it is good for workers. So in 1975 it was not and in the 1980s it changed with the social side, so we were against the EU when it was just the Common Market. So for us it is about is it good for workers. (TUC National–1)
This emphasis chimed with reluctance to engage the arguments of the official Remain campaign. The pro-Remain unions were consistent in claiming that the aims of their Referendum campaigns were both to move the terms of the debate to those that mattered to workers and to make the argument for Remain on the basis of workers’ interests: We are not going into this discussion on their turf but on our turf. (Unite–1)
As noted, most trade union executives chose to refer the final decision on the Referendum. Although Unison was an exception, choosing to engage directly with members and branches before confirming decisions to both engage in and take a side in the Referendum. Whilst this was understood as involving risk: It was the best survey we have ever conducted; it had the greatest reach we have ever conducted on a policy issue and the BBC noted this was the largest survey of working people done during the Referendum. (Unison–1)
The Unison consultation does demonstrate that one large union endeavoured to frame their campaign to issues identified by members. The consultation involved approximately 60,000 members and concluded that members feared losing a number of employment rights and further cuts in public services if Brexit occurred. The survey also identified that 78% of surveyed members wanted them to have a position in the Referendum with an overwhelming majority of these stating Remain (95%). 1 The TUC decided against declaring its position until this consultation was completed and once campaigning began, TUC regions seemingly had little or no resources to coordinate campaigns. In Yorkshire and the Humber for example, a total spend of just £180 was incurred on ‘boards for street stalls and meetings’ (Y&HTUC–1).
Decisions on registering with the Electoral Commission depended on whether unions anticipated expenditure exceeding £10,000 (in the event both the Fire Brigades Union and construction union UCATT registered but did not exceed this). Union scale and the availability of political funds, following an expensive general election and against a dire financial background (Willman et al., 2016), were significant influences on spend and hence campaign strategy (Table 2). No simple relationship is apparent between union scale or policy and the intensity of their campaigning, as measured by expenditure per member; though some small unions including the BFAWU deliberately spent below the registration limit. The pro-Leave RMT spent most per head whilst the CWU spent a larger proportion of its income than other unions.
Total declared EU Referendum spend, union members and income.
Electoral Commission (2017); bCertification Office (2017); cPence per member spent on Referendum; dProportion of income spent on Referendum.
Member–leader relationships
As Gumbrell-McCormick and Hyman (2019) note, serious tensions around the coherence of leaders’ and members’ positions are an ongoing feature of union democracy. The anticipation of consequences for leader–member relations was a significant influence on unions whose leadership supported Remain. This manifested in different ways: so for some it justified neutrality and for others an even-handedness in the presentation of arguments. Despite its 2015 conference having rejected a Leave motion, the NUT’s Executive opted for neutrality and suggested that this should have been the position in the movement more widely ‘as their memberships would be so divided on this that taking a clear cut position either way might have an impact on membership’ (NUT–1). Whilst BECTU took a Remain position, their attempted even-handedness in communications was likewise predicated on a desire not to alienate members. This influenced the decision to limit their campaigning to a single ‘sensitive’ bulk email ‘as we do not wish to upset those in favour of leaving’ (BECTU–1). The TUC also attempted to maintain a degree of balance between Remain and Leave at both national and regional level: We try to be balanced by giving EU negatives and positives. Given the Electoral Commission regulations, we have been careful in getting involved in the workplace campaigns. So we have put out national and sometimes regional information and allowed others to campaign at local levels. (TUC National–1)
By contrast, ASLEF were confident that members’ and leaders’ opposition to the EU aligned with their wider political orientation, to fight neoliberalism as represented by the EU Fourth Railway Package. Confident in members’ understanding of the union’s position, they incurred no expenditure, with their only activity being the General Secretary’s speaking engagements and members’ support for the RMT’s day of action leafleting railway passengers. They contrasted their confidence in members’ support with that of other unions who undertook member consultation ‘we lead as a union and do not do focus groups’ (ASLEF–1).
Communications – Messaging and campaigning
The third theme was most prominent and this related to the effectiveness of different kinds of messaging and a variety of frustrations influencing campaigning; some of these were mutually implicative. Overall trade unions’ Referendum campaigns varied in policy, intensity, scope and priorities. Although, amongst our sample, Unite, Unison, USDAW and the RMT committed resources, as did the national TUC. In respect of campaign themes, a difference emerged between pro-Remain unions with Unite framing theirs differently, placing far more emphasis on migration than other unions whose focus was predominantly the protection of employment rights. But echoing Whyman’s (2008) findings, smaller, sectoral unions made specific arguments about the impact of the Referendum results on inter alia rail, ferries and the film industry whilst larger unions and the TUC focused on generic arguments (see Table 3). Some social media campaigning was undertaken by Unison and the TUC whilst traditional methods (leaflets and meetings) predominated elsewhere.
Main arguments referenced by respondents in campaigns/interviews.
Significantly, delay in taking a position until the spring conferences left little time to take decisions on registration, to hone messages and to respond to increasing hostility to migration at a local shop floor level. The issues this generated for Remain unions manifested in a variety of ways, including employers’ refusal to allow meetings on their premises (e.g. Tesco) and activists’ desire to make the argument in workplaces: . . . depends on the industries. Those industries that are heavily reliant on foreign investment and exports, particularly to Europe saw a vested interest in a continuing relationship with the European Union, there we had support in the main from the shop stewards . . . but as soon as you started to talk about how members would vote . . . it was very clear that huge numbers were going to vote Leave . . . Did not matter where you went, you could be in Cardiff or Bristol or Hull or Newcastle it did not matter. You would get the same questions; they were saying it with such conviction that you had to believe that they actually shared some of the views as our members. (Unite–2)
Surprise at the vehemence of members’ hostility was a regular feature of interviews during the campaign. The TUC Yorkshire and Humber reported that: People are now more willing to raise issues on a nationalistic basis. You know ‘I have lived here all my life and I can’t get this because . . .’ and trying to argue with someone from a deprived area that migrants might help them . . . (Y&HTUC–1)
Concern as to the direction that the campaign had taken was exacerbated by encounters between union officials and members; for example at a Unite workplace in Derby (a Leave voting area 57%) a pro-Remain battle-bus was ‘wrecked’ by pro-Leave members. Significantly, Unite have been highlighted as an inclusive union supporting CEE workers (Fitzgerald and Hardy, 2010). Further, the decision of ‘Britain Stronger in Europe’ to include Frances O’Grady (TUC General Secretary) as a speaker in the final televised debate
2
was both welcomed and seen as portentous: At the beginning of the campaign I don’t think there was a chance in hell that Frances would have been that prominent in the final line up but it became clear that actually that is where the battle needed to be fought. (TUC National–2)
Unison and Unite, Britain’s two largest unions, took different views of the need to combat arguments about migration during the campaign. Unison’s consultation results had shown migration to be insignificant as an issue and so whilst this was a ‘worry’ it was not a ‘game changer’. Unite, though, did consider this and austerity as issues but they were not adequately anchored (Goffman, 1974: 247), as recognised in the next section: . . . we have approached this head-on; you cannot avoid it. So we are out there arguing that it is an ideological austerity programme that is forcing the NHS into crisis so that they can privatise it. So there was research done that showed that where there is a 10% concentration of migrants in a community, NHS Waiting Lists are nine days lower than elsewhere [see Giutella et al., 2015]. Migrants tend to be younger, healthier, at work, not so reliant on the NHS and care services . . . It is not migrants who have brought down terms and conditions it is abusive, exploitative employers. (Unite–1)
Unite’s campaign, alone of the unions we spoke with, specifically targeted these messages at the East Midlands, Hull and other areas with migrant working communities. By contrast Unison, convinced that their members would predominantly support Remain, prioritised campaigns to encourage members to register to vote (as many as 300,000 were thought to be off the electoral register) targeting their 60,000 young members through social media. This also involved sharing membership data with electoral data pooled from the major political parties in the ‘Britain Stronger in Europe’ campaign. This data drop led to the identification of 140,000 Unison members in the key swing voter pool, to be targeted with ‘Britain Stronger in Europe’ material combating Leave claims on the NHS: We are not just saying to people vote Remain, we are trying to address a number of the issues that have been highlighted to us. So say to people these are the issues and when you vote, keep this in mind. It is a gentle approach and tested; in the last election we asked people which way they would vote and used this approach, and then went back and got a 12% swing to Labour. (Unison–1)
By contrast to this gentleness, arguments made by BFAWU on public expenditure chimed with the populist appeal of Leave campaign messages on increasing NHS spending: ‘it is to do with how money spent in the EU on membership could be spent in this country’ (BFAWU–1). As noted, pragmatism underpinned pro-Remain union campaigns, which overwhelmingly emphasised both the employment rights guaranteed by EU membership (maternity, health and safety, working time, holiday pay, discrimination and TUPE being mentioned most often) and the need for EU reform and a return to the social model: ‘our position is not anti-EU, it is anti-direction of EU right now, it is anti-neoliberalism right now, anti-deregulations right now’ (Unite–1). Indeed BECTU noted the failure to apply existing rights in teaching (working hours) and other sectors: ‘remember BECTU took Tony Blair to the ECJ because they didn’t want to fully implement the Working Time Directive’ (BECTU–1). Respondents on all sides identified that the argument that the EU could only be reformed from within gained little traction with members who wanted to Leave and the argument that the EU would impose a damaging Brexit deal fed into Leave’s narrative better than Remain’s. Whilst an EU sustaining post-war peace argument was significant to several respondents, its lack of impact amongst members was equally evident. Likewise arguments about EWCs had traction with campaigners but not most members.
Not all unions focused solely on the generic benefits of the EU however, and sector-specific arguments were evident in BECTU’s campaigning around protection for cultural programming, and for the advantages of free movement for members who were used to working in a transnational film industry; general unions however produced little sector-specific literature. Likewise, the RMT cited specific examples of the EU’s neoliberalism in representing the EU as ‘bosses’ Europe’ (see McIlroy, 2018 for a wider discussion of Bob Crow’s influence here), which ‘cannot be changed’ (RMT–1). The example of Caledonian Ferries being put out to tender against the wishes of the Scottish government proved a particularly useful example for unions committed to nationalisation of rail and maritime industries: What we are aiming for is one single public service that owns the entire rail service and its infrastructure, as well as, owns rail manufacture. So essentially back to the old British Rail [BR], for example much of the stock on the East Coast was built by BR. EU Directives do not allow that anymore and what the EU proposes and has proposed runs directly in opposition to our nationalisation strategy. (RMT–1)
Evaluating the campaign
To address our second research aim, we undertook interviews with respondents in late 2016 and 2017 to examine their evaluations of the effectiveness of their campaigns. This was influenced not only by the result but also by TUC polling in the immediate aftermath. This revealed that overall 62% of the joint Remain and Leave sample agreed or strongly agreed with the statement that ‘It is essential for the immigration system to reduce the number of migrants coming to Britain’ (Greenberg Quinlan Rosner, 2016: 9). Significantly this included 44% of Remain voters and 85% of Leave voters. Such were the results that union campaigning to increase voter registration and turnout was itself regarded as counter-productive inasmuch as people voting for the first time were regarded as overwhelmingly Leave voters (Unison).
The majority of respondents did not believe that the unions could have altered the result, itself a significant admission of the weakness of the influence of union leadership on members (Gumbrell-McCormick and Hyman, 2019). The Referendum provided a valve to release pressures that had been experienced by working people for years and perhaps even generations (see MacLeod and Jones, 2018; Telford and Wistow, 2019). Post-Referendum focus groups with which respondents had been involved had evidenced widespread confusion about both the European Union and the implications of the Leave vote. In the context of widespread ignorance, members and voters more widely were considered to have paid particular attention to the features of the campaign that appeared clear to them. In this context the promise that Leave would yield an additional investment of £350 million a week for the NHS was considered pivotal: We tried to supply people with simple facts about ‘don’t trust them with your NHS, there will be no NHS in 20 years’ time if we leave Europe’. But we just did not get anywhere because of the noise they were making with their £350 million. It was so effective and they knew it was effective from day one so they just kept on pushing it and pushing it. (Unison–2)
Despite their differences in Referendum policy and the intensity and focus of their campaigns, two themes dominated the post-Referendum evaluation: migration and campaigning.
Migration
Respondents on both sides of the Referendum agreed that the result was framed by alienation from political elites, austerity and competition from EU workers. Along with the actual extent of CEE migration, competition from EU workers also feeds into our earlier discussion of special union structures for CEE workers (Aziz, 2015; Ciupijus et al., 2018) as well as perceived special treatment (Mustchin, 2012). These factors provided an audience for anti-migration messages and made it difficult for members to appreciate the advantages of remaining: Most people wanted to stop large invasions of low-skilled Europeans descending on their towns doing casual work; I think this [a Remain vote] could have been achieved if we had not been such a deregulated employment market. (Y&HTUC–2)
Inasmuch as unions believed their own campaigns could have been more effective, their choice of campaign issues was a repeated concern. Union leaders universally underestimated the impact of the migration issue on the campaign and almost regardless of the focus they gave this (which, as we have seen varied), Remain respondents were particularly self-critical in respect of their failure to address migration directly enough or quickly enough: The message on immigration was weak and ambivalent and therefore would not attract those who called themselves Lexit if you like. So for me the key campaign on the Leave side was immigration and we did not address it, we just kept quiet because we did not want to upset anybody. (SERTUC–2)
Despite travelling the country to make the case that unscrupulous employers and lack of public investment were the causes of declining wages and failures in public services, and not migrants, Unite’s respondent argued that their engagement on immigration was inadequate both before the campaign and in a failure to react to the issue as its importance became clearer during the campaign: [The union] . . . took a conscious decision not to raise the immigration debate, so we weren’t going to get into their turf on immigration. So we wanted to run a positive campaign and we could have run a positive campaign about immigration of course but we decided not to run [an] immigration focused campaign . . . So we weren’t out there promoting the issue. I think early on it became clear that immigration was the one thing. (Unite–2)
In accounting for their failure to anticipate the importance and toxicity of the immigration debate, the USDAW respondent reflected on the failure of the assumption that the issues raised at conference would be a guide to the campaign. As it turned out, a quite different interpretation of delegates’ thinking was suggested by the contrast between the centrality of immigration to the Referendum debate and its absence during conference debate: . . . coming up to conference we were concerned that there would be a debate focused on foreigners taking British jobs but it did not come through. Nobody spoke against the statement to Conference about this; on reflection perhaps no one wanted to put their heads up above the parapet . . . I think this is one of the reasons I was lulled into as false sense of security. (USDAW–2)
On the Leave side, while there was ‘no regret’ (ASLEF) about the union’s position there was an acknowledgement that campaigns could have done more to counter the anti-migration narrative: . . . perhaps our decision was on slightly parochial grounds and the outcome would demonstrate that and that perhaps a much, much bigger picture to consider. That we did not take into account when we made our decision . . . I think if we could go back in time the decision should have been made on a much broader perspective rather than just the individual factors we considered. Given that our Executive is made up of very industrially minded individuals and not political it wasn’t surprising that we arrived at the decision we took. (ASLEF–2)
Similarly, BFAWU believed that failure to counter anti-migration narratives was important: . . . members of the government blamed migration because they were failing on education, the health service and I don’t think the trade union movement was good enough on refuting those types of things. Also in refuting the promises of Farage, Gove and Boris Johnson, the lies from them about the £350 million and those sorts of things. We should have been clearer that we are supporting Brexit to strengthen trade unions in . . . companies, so to push legislation through here rather than having to wait four or five years at a European level. We should have made it clear to people that they weren’t going to get the £350 million. This is not going to stop immigration, that is not going to happen in ten years. (BFAWU–2)
Union campaigns
When considering their own campaigns, the most commonly repeated frustration concerned the delays to the launch of union campaigning. The decision made by a number of unions to delay campaigning until the conclusions of Cameron’s renegotiation was compounded by taking recommendations to spring conference despite policy being long established and no further derogations having been agreed: I don’t think we really kicked off big campaigning until the beginning of April and we probably should have been campaigning for three to six months before then to really have an impact and change people’s minds. (TUC National–2)
Another concern was that despite unions on both sides wanting to run member-focused campaigns, there was confusion as to the alliance of traditional enemies on the Remain side: I think it was a confusing coalition and again in retrospect should Labour have gone alone, could they have pulled in more votes by going alone? Sharing platforms with David Cameron, Frances O’Grady with big business. It is not how we have traditionally run campaigns and I don’t know if we confused people and put them off. (Unison–2)
Amongst both Remain and Leave unions a key lesson was accepted in the need to counter hostility to migrants and to channel members’ anger towards employers and government: This is the real enemy within. What’s happening when they are taking away your overtime payments, when they are taking away your incentive bonus, if you have venom these are the people you should direct it to. So we have started that campaign and at the moment it has been successful, we are getting people out saying ‘you are right’. (BFAWU–2)
This theme chimed with the wider movement’s ‘workers must not pay the price for Brexit’ post-Referendum position. At the same time, the TUC has established a panel of key convenors in private sector companies to keep a watching brief on their post-Brexit strategies: As with the financial crisis, obviously some of the things that were done were because we had a financial crisis and some of it was a convenient excuse for doing stuff that the company wanted to do anyway. So we want to get a clearer view about what is actually a result of Brexit and what isn’t. (TUC National–2)
Discussion and conclusions
We would note that no two referendums ask the same question in the same set of socio-economic circumstances. Given this, it is unlikely that learning from one referendum will be readily transferrable to others, although the NUT’s decision to remain neutral was influenced by their neutrality in the Scottish Independence Referendum. However, what can be learnt is that campaigns need to be framed around specific issues, and timescales for identifying issues, developing strategies and seeking to convince members collectively need to be longer. Here the weakening of unions’ bargaining strength informed campaigns and thus a greater willingness to support European regulation that protected workers’ interests and hence European institutions (Hyman, 2008). During interviews Remain respondents noted the importance of a social Europe and the need to protect workers’ rights, although Leave respondents were critical of a neoliberal EU. This focus though meant that regardless of the side they took, unions maintained the pragmatic presuppositions that had informed them in 1975: namely that their position would be principally informed by the interests of members. This informed a determination, on both sides of the argument, to focus campaigning on what they believed to be members’ interests and those of workers more widely. Although this distinguished their arguments from the largely ‘blue on blue’ national campaigns in order to shift wider attention towards issues of employment. Pro-Remain unions’ attempts to frame the campaign in terms of employment rights were at odds with members who framed it in terms of migration in the context of austerity.
Interestingly, respondents reported that members’ understanding of, and receptivity to, arguments for free trade increased in export-focused workplaces or amongst workers whose routine experience involved working across borders (BECTU). Further, with the two Leave unions, ASLEF and RMT, ‘regaining control’ was counterposed to the increasing marketisation of the EU (EU Fourth Railway Package). In their view, a sovereign British Parliament provided a clear line of sight and the opportunity for the renationalisation of the railway system. For the NUT, though, the Referendum had little priority, as EU-driven employment protection had been ineffective, especially over working hours. Other disputes were of more immediate importance to members (see Stevenson, 2015, who discusses these neoliberal government reforms). As noted it seems that for many members the effects of austerity in working class communities, reduced terms and conditions, and EU free movement (Coderre-LaPalme and Greer, 2017; Telford and Wistow, 2019) negated any positive outcomes from trade union policy initiatives in the EU. Perhaps more importantly here it seems that the types of ‘localised’ transnational engagements identified (Bieler and Erne, 2014; Fitzgerald and Stirling, 2010; Whittall et al., 2017) either remain short in the memory or have involved too few members.
As significant was the lateness of union campaign engagements with members, which may have provided an opportunity to better frame issues and indeed counter any ‘lies’ (the £350 million) or migration myths. The pro-Remain union leaders and officials in our sample were dismayed at their inability to counter the argument that leaving the EU would provide extra resources for the NHS (the £350 million). Thus despite longstanding union positions towards the EU, the decision to wait for Cameron’s renegotiation and then for spring conferences delayed the start of campaigning. The subsequent intensity and focus of campaigning varied widely between unions. Whilst the RMT spent heavily per member and both Unison and Unite committed significant resources, other unions on both sides of the argument spent little, citing financial constraints or lack of priority. Data from the Electoral Commission confirm that this pattern was replicated in the wider movement: the vast majority of unions spent below the £10,000 reporting limit. This may corroborate Willman et al. (2016) in arguing that unions’ precarious financial position militates against participation in campaigns. But also underpinning this was a concern not to alienate members, to divide union leaderships from the membership – but this was seemingly starker in Remain unions. The fear of alienating members figured in the TUC’s attempts to provide ‘balance’, and in the cautious tone of member mailings. Members’ divisions were also understood to have frustrated campaigning at workplaces with the anger and hostility of Leave supporting members being reported back through branches and shop stewards throughout. Likewise a number of employers were reluctant to facilitate union pro-Remain campaigning, especially in retail, for fear of alienating employees and customers. These fears were informed by a vituperative campaign and the divisions this provoked and exposed. For both unions and employers the risks of arguing for Remain acted as a significant constraint on public pronouncements and support. The fear that general unions’ leadership would be disconnected from members (Gumbrell-McCormick and Hyman, 2019) found strong support from ASLEF and the RMT, who contrasted relationships between their own executives and members with those of general unions. ASLEF’s opinion of Unison’s use of focus groups was especially scathing.
Overall, though, as reported by respondents, free movement and a concern about migration were central during engagements with members. In explaining their failure to anticipate this they highlighted inadequacies in internal communications and culture that led to members not ‘putting their heads above the parapet’ to explain concerns. Likewise, local activists’ reports of the importance of migration during the campaign indicated that some activists shared members’ views but were unprepared to voice them as their own. Thus as well as ESD, ESSD and specifically EWCs not seemingly making an impression it seems that the recent national and local activities of many unions with regard to CEE workers (Aziz, 2015; Ciupijus et al., 2018; Fitzgerald and Hardy, 2010) had left little or a negative impression on members (Mustchin, 2012). Increasing awareness of the importance of migration for Leave-voting members provoked divergent responses. Whilst Unison’s extensive branch consultation before the Referendum had led them to conclude that migration would not be a critical issue, Unite decided to respond to ‘myths’ around migration but even in their case did not raise the issue. Such was the depth of shock and in some cases revulsion at the result and its aftermath that union leaders and senior officers, on both sides of the argument, have realised that a failure to engage more fully in the debate about migration aided the nationalist and sometimes racist narrative that was associated with it. It was recognised that this was a critical error, therefore we reach two final conclusions.
The first is that the issues that framed the Referendum campaigns did not address members concerned about migration and this has a wider application for future campaigns of all types. Here there is strong evidence that trade unions clearly did not have broad enough migration policies or initiatives. What we mean by this is more effective engagement with non-migrant members’ cultural values and beliefs. Trade unions have for a number of years administered anti-racist programmes, which have indeed increased following the EU Referendum. All are important but the types we seek to highlight and argue for are those that are widely applicable and engage with misunderstandings of race and racial terms and taking into account non-migrant cultural values and beliefs. For example, Unison and more particularly the TUC are currently providing Show Racism the Red Card programmes. 3 Anti-racist ambassadors’ programmes encourage both workplace representatives and members to express antagonistic opinions about migration or indeed immigrants. Then Show Racism the Red Card trainers facilitate open and honest discussions about these views, providing facts and other scenario information. Antagonistic opinions include racist views and terminology, which are allowed and expressed during an amnesty period. One of the key learning outcomes is that issues such as this have to be engaged with, discussed, understood and hopefully resolved, with discussion and communication the key.
Secondly, earlier engagement with non-migrant members’ cultural values and beliefs can contribute to issue framing and the mobilisation of members. This is applicable generally for unions. The reality with our study was, though, that both general unions and the TUC mainly concentrated their messaging at a national level, arguing broadly about employment rights and the importance of the EU in their protection. Whilst within sector-specific unions, arguments on both sides were far more likely to be illustrated by material examples that resonated with workers’ own experience (e.g. BECTU working on film sets in EU member states, or RMT having ferry services privatised). This suggests that the effectiveness of particular union campaigns is undermined by not seeking to engage either at an early stage or on an ongoing basis with the membership. Campaign mobilisations are enhanced with engagements and training such as the union Show Racism the Red Card programmes. But of course ongoing coordination and communication with those who have competed training need to either be established or maintained.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
