The limitations of the theory and practice of mobilization in trade union organizing

Introduction

That Kelly’s 1998 book Rethinking Industrial Relations has been well read is without doubt – it has been cited over 1400 times in scholarly writings ¹ since its publication. Kelly’s mobilization theory identifies under which conditions ‘individuals are transformed into collective actors willing and able to create and sustain collective organization and engage in collective action against their employer’ (1998: 38). The book then outlines a theory of how that collective organization and action take place, and the opportunities and constraints presented by the wider context. Given the political landscape at the time of Kelly’s writing, it was an important moment in industrial relations scholarship and in an in-depth review shortly after its publication, Martin (1999: 1208) argued that Kelly’s mobilization theory offered great strength to understanding collective action:

First, it generates researchable propositions about the structural conditions under which collective organizations might be expected to develop, while at the same time recognising the central importance of agency. Second, it focuses upon power, and responses to the differential distribution of power, as the central dynamic of industrial relations. Third, it recognises the importance of political, social, and economic factors in the development of collective action, relating them to each other in a systematic manner. Fourth, it provides an open-ended approach to the future of trade unionism.

Many of these issues have been picked up in the academic industrial relations literature using mobilization theory, although some to a greater degree than others: power, for example, and the importance of political, social and economic factors in the development of collective action have not received the same attention as structural conditions or agency. Martin (1999) made an important observation soon after Kelly’s book was published, that mobilization theory could provide a useful analytical approach to understanding renewal and the future of trade unionism. We agree that it still can, but it would be a much more useful systematic tool if there was a clearer conceptual understanding of what academics and practitioners meant by the term mobilization.

Our central argument here is that both in the original text, and in many subsequent applications of Kelly’s mobilization theory, there is a tendency to conflate two related, but distinct, concepts: mobilizing and organizing, and also that time and experience suggest that new definitions are now necessary. One distinction we make is that the starting point for mobilizing is the utilization of power resources already available, whereas organizing begins by asking where the power is that is needed to effect change and then works backwards to figure out a systematic strategy to develop the resources needed to win. Put another way, mobilizing is most often limited to activating an existing base of support (usually union members in the case of industrial relations), whereas deep organizing involves engaging and activating people who may not initially agree but who, through a process of collective organizing and the development of grassroots leaders, begin to self-identify as part of a community with a shared objective in seeking to challenge injustice. The distinction also extends to a difference in the use and understanding of power: while it is possible to win concessions by mobilizing a minority, a union’s power is much more clearly demonstrated when a large (‘super’) majority of workers is organized, actively showing support for the cause. The latter is achieved by a systematic power analysis of allies and enemies to understand how to build support and create leverage to weaken the opponent.

Han (2014) defines the difference between mobilizing and organizing in a more concise way by referring to the former as ‘transactional activism’ and the latter as ‘transformational activism’. In a mobilizing approach ‘participation is conceptualized as a transactional exchange between an activist and the association [union]’ where tasks requested are ‘strictly limited in their time commitment and require minimal effort on the part of the activist’ (p. 95). In comparison, in an organizing approach ‘the goal is not only to get work out of the activist in the short-term but also to invest in developing the activist’s capacity to act’ (p. 96) in the long-term and this is done through ongoing leadership identification and the teaching of power by working collectively with others in the union and beyond.

In making our argument, we are not criticizing the central pillar of mobilization theory and its contribution. Rather, we are urging industrial relations scholars to think more systematically about the distinction between ‘deep’ organizing and ‘shallow’ mobilizing, or Han’s (2014) ‘transactional activism’ and ‘transformative activism’, and how it affects the analyses undertaken, the conclusions reached, and the implications for practitioners. To do this, we reviewed 52 published research articles that apply Kelly’s mobilization theory in the context of trade union organizing activity. Our findings suggest that future scholarly analysis would benefit from a clearer distinction between these two terms, and thus we adopt an approach that sets mobilizing and organizing as two different and distinct activities within unions and other social movements (McAlevey, 2016).

Kelly’s mobilization theory in its historical context

In explaining why there has been a lack of consistent clarity between the two concepts of organizing and mobilization, it is important to note the publication date of Rethinking Industrial Relations. The book was published in year one of the first Labour government after 18 years of Conservative Party dominance of UK politics. Those 18 years had seen sweeping industrial relations reform, large-scale strike action in industries such as coal mining and printing, and violent confrontations between the state and unions, resulting in a political, economic and social context that was deeply hostile to trade unions and collective action. The election of a Labour government in 1997 brought an almost immediate shift in the approach of the state to regulating industrial relations. Of most significance to the ideas presented in Rethinking Industrial Relations was legislation introducing statutory processes allowing trade unions recognition for collective bargaining, and, at that time, that legislation was already being discussed, and the trade union movement was preparing for it. Those preparations included establishing training programmes for union staff at national level (see Heery et al., 2000 for a discussion of the establishing of the TUC’s Organising Academy) and developing campaigns at a local level (Gall, 2005; Heery et al., 2003; Holgate and Simms, 2008).

In this context, attention of both practitioners and academics was focused towards developing campaigns that attracted members and demonstrated union effectiveness in the workplace and beyond. Most of these were referred to as ‘organizing campaigns’, and indeed some unions did organize, but many were mobilizations of workers who were already union members in order to (re)gain union recognition rights for collective bargaining. Even though the legislation stated that only a simple majority of workers (51%) was needed to vote in favour of recognition, these simple majorities, however, leave a large part of the workplace unorganized.

Importantly, Rethinking Industrial Relations offered a ‘go to’ theoretical framework within which to analyse these developments. The empirical studies that followed the publication of the book have added important detail to our understanding of the micro-processes of union activities, and some have linked these to wider macro-developments, but without distinguishing, however, between the evolving conceptual differences of deep organizing and shallow mobilizing as set out by McAlevey in her book No Shortcuts (2016), and expanded upon in the section below.

Mobilization and organizing in perspective

Effective power, as McAlevey argues, is evidenced by US unions being able to demonstrate ‘super majorities’ of workers supporting action, not simply mobilizing the already supportive union members, who might indeed be a minority of the workforce (as in the UK in relation to strike ballots). ‘Mobilizing’ – or ‘moving workers into activity’ – is a necessary step in the process of organizing, but not all organizing work seeks to create a mass mobilization. While ‘mobilizing’ is an important tool and activity within organizing, it is not, on its own, organizing. ² To be more precise, a focus on just mobilization risks creating a movement that has little sustainability. At its most problematic, mobilization can create an illusion of strength (when activists are able to temporarily compensate for the inactivity of members or the wider workforce), but is unable to be sustained because of a lack of power that comes with mass supportive activity. Indeed, we argue that it is conceptually useful to understand deep organizing as a specific strategy to build power to win worker demands, often through strike action.

This is not to deny the important potential – and sometimes reality – of some collective mobilizations. Visible mobilizations are capable of creating a context that shifts the public discussion and can sometimes be enough to create wins. The Occupy movement and the more recent Black Lives Matter demonstrations in the US are examples of how protest can highlight inequality and discrimination. Strikes and union recognition mobilizations are examples in the arena of industrial relations. Kelly’s particular application of mobilization theory to industrial relations helpfully refocused attention on these visible worker actions, but focused only relatively briefly on the organizing activities that underpin decisions to act or that make mobilization possible.

In this article, deep organizing is a theory about how to win majority strikes, and is taken to refer to the long-term day-to-day grassroots activities, often undertaken by trade unions, but sometimes by other community groups, to effect change. Organizing draws on the concepts of collective interests (in the context here, mainly of labour and capital) and the agency of key actors (workers, unions) to challenge capital that are central to mobilization theory, but it is a much wider range of activity than end stage mobilization. Crucial in a deep organizing approach is the need to have a theory of change (strategy) about how workers might confront and rebalance the power of capital. Early on, this requires a power analysis of the strengths and weaknesses of both workers and the forces reigned against them. A decisive factor in deep organizing is the extent to which it is possible to broaden the base of activity (by increasing members or activists) and identifying new leaders and their capacity to act to increase the power resources of workers. It is about organic leaders who are trusted and respected by their co-workers, converting opponents or non-members into active supporters of the cause. This is something missing in mobilization which tends to rely on already-members or ‘the usual suspects’ to just turn up to protests or a one-off activity, but this, on its own, may not build the relationships that are necessary for developing capacity, power and grassroots leadership that is sustainable in the long-term. In mobilizing, action by members is often directed by staff and the measure of success is the numbers taking action, whereas in deep organizing support is provided by staff who train, coach and mentor, workplace leaders who can engage their fellow workers in activity, thus building a greater depth of support for the union. As a quick fix, mobilizing can be successful, but it is seldom, on its own, transformative or effective in changing the balance of power in the workplace.

Kelly’s use of mobilization and organizing in Rethinking Industrial Relations

Drawing on the work of social movement theorists such as Tilly (1978), McAdam (1988), Klandermans (1988) and Gamson (1992), Kelly puts mobilization at the heart of his theoretical proposals. In applying these theories to industrial relations and trade unionism he argues that in order for workers to define their interests as a collective, there needs to be a sense of dissatisfaction about breached values and norms. In the process of moving towards the development and articulation of collective interests, those dissatisfactions must be framed as illegitimate and attributed to an agentic other party – usually, in industrial relations, management – and then put forward as a set of demands. Leaders are essential in that process, although Kelly gives little detail about who he understands leaders to be. Mobilization can only then occur when a collective group (say, workers) perceive an injustice which is framed as such and attributed to another group, say managers. Whether or not mobilization occurs depends on the opportunity and judged effectiveness of acting. This cost-benefit analysis can be affected by many factors including the perceived likely risks, for example danger of counter-mobilization, and possible financial and personal costs – including a calculation of whether action will make the workplace a more or less attractive place to work.

Kelly (1998: 37–38) presents the forms of collective action that are the focus of his discussion and theorization: strikes, overtime bans, go-slows, work-to-rule, petitions, lobbies, collective appeals, and ‘the many other forms of non-cooperation and threats to employer legitimacy’. There is a centrally important point here: they are, in the main, mobilizations. They are all moments of action in the day-to-day grind of industrial relations and trade union representation. Kelly does not claim that mobilization theories should, or even could, explain the broader and long-term challenge of recruiting and organizing workers into unions. His focus is on worker collectivism, and specifically collective mobilization, as a mechanism for union members to use their collective power to pursue their interests and challenge injustice. The focus on mobilization was explicitly intended to move debate away from the descriptive and institutional focus of industrial relations into a realm that gave greater conceptual and theoretical clarity to agency (1998: 15). In doing so, Kelly had an explicit agenda to put power dynamics at the heart of his theorization, but this is something that has still not received much analysis in the industrial relations literature we reviewed, and specifically the literature on union organizing/renewal (1998: 9).

That said, broader issues of union organizing are not absent from Rethinking Industrial Relations. Within Kelly’s analysis the levels of membership and density of union representation are important factors for workers in judging the likely costs and benefits of mobilization. Drawing on Tilly (1978), Kelly argues that these are important variables in helping understand collective organization and action, but that ‘we need to look more deeply and try to gauge the extent to which members identify with the organization and the degree of interaction, or density of social networks, amongst members’ (Kelly, 1998: 37). In other words, he refers not just to structural aspects such as of union density, but to more qualitative aspects of union organization. Importantly, however, he pays little attention to how unions can achieve this. Rather it is taken to be one of the contextual factors that is required before workers engage in effective collective mobilization.

In sum, Rethinking Industrial Relations is mainly focused on understanding the dynamics of collective action, in particular on mobilizing people who are, in the main, already trade union members. In taking this approach, the broader context of union organizing is taken to be a contextual factor that contributes to judgements about the likely effectiveness (or not) of collective action.

McAlevey’s use of mobilizing and organizing

While McAlevey does not specifically address Kelly’s mobilization theory, she sets out why a distinction between the two concepts of mobilizing and organizing is important in understanding the way that workers develop their capacity to challenge injustice and win concessions. Her thesis is that each approach has a set of common traits that sets them clearly apart; they are not opposites, just very different methods of organizing workers. She begins, as others have done before (Simms and Holgate, 2010), by posing the important questions of ‘what are unions organizing for?’ and ‘what is the purpose of the union?’ This, and two other important factors – the way that power analysis defines the direction of the action, and the way the union is led and governed – distinguishes the two approaches.

McAlevey argues that mobilizing approaches risk having low day-to-day participation of members, with the exception of one-off mobilizations such as marches, strikes or rallies. Even when these activities do occur they tend not to have majority participation of the workforce. The focus of these mobilizations is usually to improve material conditions – such as winning wage growth – rather than broader reasons, such as to improve job quality or professional standards. Although the form of this approach to collective trade union representation varies in particular national and sectoral contexts, it is familiar. By focusing on occasional moments of collective member protest (mobilizations) and allowing professional union staff to undertake most of the ongoing day-to-day work of the union, the union asks relatively little from its members. Although a lot of effort is put into engaging members towards the occasional mobilizations, the main focus of activity is led by union staff and lawyers who direct and inform the workers. Workers themselves play a secondary role, if any, and this is particularly the case in terms of strategy. While union staff are able to mobilize members, most often these are the already committed ‘dedicated activists who show up over and over at every meeting and rally for all good causes, but without the full mass of their co-workers or community behind them’ (McAlevey, 2016: 10).

McAlevey contrasts this with a ‘deep organizing’ approach which is characterized by high levels of participation by union members who decide for themselves the issues around which they will organize: ‘organizing places the agency for success with a continually expanding base of ordinary people, a mass of people never previously involved’ (p. 10). While material concerns such as pay are often the key issue for workers, the labour process, or quality of service, or care, is also of considerable importance to workers who just want to be proud of their work and do their job well. The purpose of the union is therefore to reflect these concerns, which may relate to the workplace and beyond, with an understanding that union power needs to be utilized to increase both workplace and non-workplace standards. McAlevey describes this as ‘whole-worker organizing’ where workers, rather than staff, are the primary actors in an organizing union and are networked and embedded in their own communities from where they can draw further support and increase their power resources. As a result, they develop greater relationships of trust within the workforce. These ‘organic worker leaders’ are developed to be key organizers for the union and have a strong following at work and perhaps in their communities as well. Broad shop-floor issues are taken into open collective bargaining sessions run by workers themselves, who are able to work out what is needed to force managers to agree to their demands. In this approach, workers figure out through a comprehensive power analysis that super majority strikes can be an effective tactic in affecting change. No public action is taken until the union can demonstrate at least 65% support and the intention is to work towards increasing that figure to as close to 90% as possible. All this requires systematic, deep organizing workplace-by-workplace to ensure there is mass participation by members – only this way can a union be sure the issues have the full support of members who are prepared to commit to each other that they will act to achieve their goals.

In a practical example of the differences between the two approaches, McAlevey (2016: 58–59) cites the 1930s Congress of Industrial Organization (CIO) unions as those that understood deep organizing as distinct from mobilizing:

The CIO organizing methods incorporated an appreciation of power inside and outside the workplace. They used a systematic approach to recruiting support not only from the shop floor but also from the broader community in which the workers lived. Yet today, most good unions that organize inside the shop mobilize outside it: deep inside, shallow outside. It’s as if they can’t see the full extent of the battlefield or the vastness of their army. … To restore worker power to 1930s levels requires an organizing model inside and outside the shop, based on CIO practice in the 1930s and 1940s but adapted to today’s conditions.

In short, McAlevey’s contribution draws attention to the risks of focusing only on mobilization and therefore gives a framework through which to evaluate literature that has used Kelly’s mobilization theory. There are, undoubtedly, problematic aspects to the ‘vision’ of deep organizing developed by McAlevey. Most obviously, shop-by-shop organizing is not always a feasible strategy, particularly in very small workplaces, or in parts of the ‘gig’ economy. Similarly, mobilizations may be able to give wider meaning and solidarity to the struggles inside individual workplaces. An example here would be the Fight for $15 campaign in the USA, which has been successful in raising the pay of millions of low-wage workers through advocacy and protest, but on its own has done little to build the power of workers themselves to challenge corporations. Clearly any action to increase workers’ pay is to be applauded and this was a hugely successful large-scale mobilization, but it is a very different approach from deep workplace organizing and mass recruitment of union members as advocated by McAlevey.

What is important here is that McAlevey’s critique of mobilization, and clear articulation of an alternative that she labels deep organizing, provides a helpful framework through which to explore how these concepts have been deployed in empirical research to date. By systematically reviewing published empirical studies, we identify three areas that are central to both Kelly’s concept of mobilization and McAlevey’s concept of deep organizing: injustice, framing and leadership. By evaluating studies through the lens of contrast between organizing and mobilizing, this article extends existing analyses of union-building activities in two important ways. First, we make the case that researchers and practitioners need to be far more systematic in their thinking about the distinction between organizing and mobilization. Second, we argue that this lack of clarity helps explain why the resources and attention given to union renewal activity in recent decades (see, for example Simms, et al., 2013) have perhaps had such limited effect.

Methods

As researchers and writers in the area of union and community organizing we adopted a reflexive methodological approach in this article. We began with a systematic literature review. Using keyword searches of major academic databases, we identified four academic journals that have published the largest quantity of work in this area: Work, Employment and Society, British Journal of Industrial Relations, Industrial Relations Journal and Economic and Industrial Democracy. For each, we searched using the terms ‘mobilization theory trade union’, ‘organizing trade union’ and ‘Kelly mobilization union’. Taking the publication date of 1998 as the cut-off point, and using abstracts and keywords, we identified the 20 most relevant articles from each journal. In practice, not all the journals had 20 relevant articles using mobilization theory. For example, the journal Economic and Industrial Democracy generated 27 results in total but of those only 9 were considered pertinent to our study (e.g. some of the articles were about partnership and not union organizing and thus were not reviewed in-depth). The relevant articles were reviewed by one of the article authors and summarized using the following headings: (1) abstract, (2) summary purpose of article, (3) contribution to mobilization theory and (4) further comments. In total, 42 articles were reviewed in this way. This was then supplemented by a further search of academic databases using the search terms above. This yielded a further 10 relevant articles in other journals. ³ They were reviewed in the same way.

Reading through this vast literature, we categorized the articles by the main concepts of mobilization theory they drew upon. As a result, we focused on the three key areas of interest authors tend to use from Kelly’s theory when discussing union organizing campaigns: injustice, framing and leadership. In addition, we highlighted how many of these studies conflated ‘organizing’ and ‘mobilization’ as concepts using the distinction we make above. By addressing these themes, and by offering more conceptual clarity our aim is to show what a more nuanced approach to using mobilization theory might provide for theoretical and practical understandings of trade union organizing. Clearly, in an article of this length, it is not possible to review each of the 52 articles in any detail, therefore we have chosen to use examples from a selection of the papers to illustrate the ways that we think mobilizing and organizing have been conflated and where it would have been helpful to be more precise in describing the agency in the empirical examples outlined in the case studies.

How have these concepts been operationalized in studies of union organizing?

In our extensive review of the literature using Kelly’s 1998 work three areas demand attention as they highlight some of the problematic ways in which organizing and mobilizing have been conflated and the consequences of this conflation for our broader understanding of union activity. We focus on sense of injustice, framing and leadership, as these are the three areas that emerge most often in published empirical studies using Kelly’s book. That is not to say that we dismiss historical context and ideology as important factors when assessing the empirical examples (e.g. Healy and Kirton 2013). Rather, we focus here on these three areas because they offer the richest material to evaluate.

Discussion and conclusion

Rethinking Industrial Relations was written at a time of heightened political pressure on trade unions and the system of industrial relations more broadly. Kelly’s mobilization theory specifically analyses the conditions under which workers are likely to act as a collective against an employer-generated injustice. Importantly, however, the likelihood and success of the mobilization relies implicitly on the qualitative strength of the union organization. In other words, to a certain extent, for collective action to take place, there is an assumption that a qualitatively strong organization is already present. In many cases, however, this is not the case and the mobilization efforts may suffer as a result.

We therefore agree with scholars such as de Turberville (de Turberville, 2004, 2007; see also Carter, 2006) who argue that there is a risk of reading Kelly’s mobilization theory as being a largely ‘top-down’ theoretical framework. Although de Turberville uses mobilizing and organizing interchangeably he makes important points about the three areas of injustice, leadership and framing. Specifically, he draws upon data on employees’ sense of injustice and shows that, ‘those who attribute injustices to management has fallen from around 23 per cent in 1995 to around 15 per cent in 2002, while only 17 per cent of non-union employees thought that unionism is a solution to such injustices’ (de Turberville, 2007: 574). De Turberville’s argument is that while workers may feel aggrieved, they do not necessarily know who to blame. He uses the term mobilizing (as well as organizing) to critically refer to a top-down model of trade unionism. Putting aside the language used for a moment, we agree that mobilizing tends, in the case of trade unions, to be top-down and led by union staff. Further, if there is no qualitatively strong union present, the mobilization efforts will likely not be sustainable in the long-term. However, if the union engages in deep, whole-worker organizing, it requires a bottom-up process with members and often new members taking on active leadership roles and deciding the best strategic approach that meets their objectives. It also has the potential to increase the power base available to workers.

The review of empirical studies has illustrated that authors have a tendency to conflate the terms organizing and mobilization. This matters more than simply as a semantic discussion. It matters because a focus on mobilization risks taking a view of union renewal activity that is narrowly focused on the decision of a worker to take part in some kind of collective action. While that is undoubtedly an important part of union activity, it is – we argue – insufficient to create a wider context within which to challenge the power and interests of capital in a sustainable manner – in effect, to use Marx’s phrase, to transform workers into a ‘class for itself’. It is not the argument here that worker mobilization is unimportant. Indeed, there are many examples where collective actions and mobilizations have pressured employers and governments to shift position on important issues. What is argued here is that the wider context of union organizing – leading to membership growth and renewal – needs to be examined as often and as theoretically rigorously as mobilizations. McAlevey (2016) offers a useful way to differentiate between organizing and mobilization, and also a conceptual framework through which to understand those wider processes of deep organizing – of which mobilizing workers into collective action is a part. Kelly (1998) and McAlevey (2016) share many concepts: the importance of identifying injustices, issues and interests, framing those as collective interests and identifying leaders all feature strongly in both contributions. They also share a focus on the possibility of unions to disrupt power relations in the workplace and beyond. Where they differ is McAlevey’s focus on strengthening the wider power base of unions and on growth and sustainability and the power to turn out super majority strikes when necessary. Kelly’s focus, by contrast, is on moments of mobilization which can, of course, be extended in time but are nonetheless moments. It is clear that there is evidence of both forms of union activity in the empirical literature surveyed for this article. But by conflating the two, the evidence around what ‘works’ and what does not is less clear than it might be.

Again, this matters because unions have invested a great deal of time, energy and resources into renewal activity in recent decades and the results have been patchy at best. To develop sustained and sustainable approaches that facilitate long-term renewal, it is clear that focusing on moments of mobilization is not sufficient. Rather, unions need to engage in deeper organizing strategies to become a sustainable counterforce and realize that mobilization is often not enough to challenge the power structures reigned against workers. The ‘danger’ for unions in taking mobilization theory as a blueprint of how to develop a stronger union is that they risk focusing only on part of the challenge that they face.

Finally, it is important to note that this article should not be read as a rejection of mobilization theory. It was a hugely significant contribution to industrial relations literature at an important moment in time. It focused the attention of both scholars and practitioners on the opportunities for union renewal and offered a theoretically sound and structured way to think about the processes involved. It has proved remarkably resilient because it has found considerable empirical support. By offering this conceptual distinction between organizing and mobilization, we hope to extend the relevance of Kelly’s important contribution in ways that are useful to both academics and practitioners in the future.