‘Justice for Janitors’ goes Dutch: the limits and possibilities of unions’ adoption of organizing in a context of regulated social partnership

Introduction

The decline in union membership has affected European countries in different ways, but even in countries where unions do not rely on membership rates for collective bargaining recognition, unions have had to recognize that their legitimacy can be called into question by employers and the state if there is a significant disparity between membership rates and coverage of collective agreements. In the Netherlands, unions hold a strong institutionalized position in the employment relations system, with high levels of collective bargaining coverage of 85 per cent (Visser, 2015). Yet, with increasing burdens on regulated social partnership models, particularly in the post-economic crisis period with government reductions in public spending and moves towards deregulation, unions are under pressure to adapt traditional responses (Gumbrell-McCormick and Hyman, 2013). With declining union membership density to 18 per cent (Visser, 2015) and recent challenges to their institutionalized role, Dutch unions have begun to ask questions about their purpose and have recognized the need to maintain representative legitimacy through membership recruitment and activism (De Beer, 2013; Roosblad, 2013). Dutch unions, and in particular the Confederation of Dutch Unions (FNV), the largest union confederation, have developed strategies inspired by the US ‘organizing model’, and have successfully implemented techniques and practices to recruit and mobilize workers, mainly from migrant and black and minority ethnic groups working in the cleaning sector (see Connolly et al., 2014; Marino, 2015; Roosblad, 2013).

The ‘Schoon Genoeg’ (‘Clean enough’) campaign was launched by the FNV in 2007 and is considered a unique campaign for the Netherlands, as the adoption of assertive organizing techniques, such as putting pressure on employers and mobilizing workers, represented a break with the consensus-based social dialogue tradition. The campaign to organize cleaners culminated in two successful prolonged strike actions: one in 2010 lasting nine weeks and one in 2012 lasting 15 weeks. These have been the longest strike actions in the Netherlands since 1933 and represent one of the most successful organizing drives. The strikes resulted in a national collective agreement with improved wages and working conditions, and increased membership and mobilizing capacity within the cleaning sector.

This article shows how organizing has been assimilated as a strategy for union renewal in a context of state-regulated social partnership, based on national and sectoral collective bargaining. The findings show how the outcomes of organizing approaches, particularly in the cleaning sector, have strengthened union legitimacy by extending representation and regulation to new groups of workers. At the same time, the narrative and tactics of organizing have stimulated a level of internal reflection and debate on union purpose and identity and have promoted a process of internal reform and democratization within parts of the union movement. It is still too early to assess the long-term outcomes of organizing in relation to sustained union membership increases and wide-scale organizational changes. However, our research shows that this innovative response might have potential long-term implications for internal union dynamics and relationships, and in turn the dominant source of union legitimacy and power.

This article is based on interviews with union officials and organizers, as well as non-participant observations of union organizing tactics mainly within the cleaning sector, where organizing was initially implemented. A subsequent round of interviews further explored how organizing has been framed within the union internal debate and the attempt to extend it to other sectors. In the first sections, the article discusses the significance of the turn to organizing and the implications of the adoption of organizing within regulated social partnership models. The employment relations context in the Netherlands is then discussed before moving on to present evidence on the Dutch union movement’s adoption of organizing approaches.

The turn to organizing

Organizing approaches have become a central feature of debates on union renewal in countries such as the United Kingdom, Australia and the US (Simms et al., 2013). The ‘organizing model’ advocates a shift away from providing services to existing members, to the recruitment, participation and empowerment of new members (see Bronfenbrenner et al., 1998). In the US, organizing has been linked to a new and dynamic attempt to organize more vulnerable and marginalized workers through a set of high-profile campaigns – the most famous of which is the ‘Justice for Janitors’ campaigns in California. The SEIU (Service Employees International Union) has been a driving force in configuring organizing approaches and linking mainly migrant workers from Central and South America into broader political and even international coordination strategies (Milkman, 2006).

Organizing – understood as an approach to recruit new workers, empower union members and encourage worker self-organization – is presented as a strategic opportunity for renewal and revitalization in the context of declining union density and institutional power bases (Frege and Kelly, 2004; Gumbrell-McCormick and Hyman, 2013). Yet, organizing is an ambiguous and contested concept (De Turberville, 2004, 2007; Simms and Holgate, 2010a) and over the years some argue that the ‘organizing model’ has become diluted and has ‘transmuted into a broad hook on which to hang ideas and practices’ (Gall, 2009). Organizing was originally introduced into the US union movement as ‘internal organizing’, used to refer to activity within unionized workplaces, with the focus on mobilizing current members for union action. The term ‘external organizing’ was used to refer to recruitment of new members and outreach work. The model is now indiscriminately used to refer to both internal organizing to mobilize members and external organizing to build support for union representation (Hurd, 2004). This distinction is important, as the way in which organizing is measured and understood in both academic and union debates tends to refer mainly to ‘external organizing’ with a focus on successes in recruiting new workers and mobilizing them to achieve a particular outcome.

Organizing has predominantly been researched in Anglo-Saxon countries where ‘external organizing’ and recruitment has been a necessity for union survival and renewal, and it has provided an important narrative for such attempts (Martinez Lucio and Stuart, 2009). However, in the US and the UK, apart from some symbolic victories, such as the ‘Justice for Janitors’ and ‘Justice for Cleaners’ campaigns – which have been used as reference points for other campaigns – organizing has not changed the overall trajectory of union decline (Milkman, 2006; Simms et al., 2013). This result does not necessarily reflect a failure of the ‘organizing model’ as such, but the move away from the original idea means that its success has mainly been measured in relation to ‘external organizing’, rather than analysing changes in the levels of participation and internal democratic dynamics.

In relation to internal dynamics, research shows that while the ‘organizing model’ is presented as a means of empowering grassroots workers to become the union, it can at the same time involve a consolidation of hierarchical authority at the centre (Savage, 2006). While it is generally recognized that organizing and mobilizing workers is key to any effort to revitalize, the question remains – how to achieve these goals and what must be sacrificed to do so? Accounts from union organizers in the US have criticized the way in which organizing has been managed from above and failures in self-sustaining organizing have been as a result of top-down interference in local unions’ attempts to organize (McAlevey, 2012). Yet, there evidently needs to be some form of structure within which organizing can take place and any form of democracy relies on and sustains some form of bureaucracy (De Turberville, 2007). Indeed, Milkman’s research on the SEIU’s successful organizing drives in Los Angeles shows that the keys to success are a combination of ‘top-down’ direction with ‘bottom-up’ mobilizing (Milkman, 2006). It is not the case of one substituting the other but of articulating and coordinating an approach that contains both drives (see Heery, 2005). This tension within organizing of where the locus of power should be reflects a wider, and more universal, tension within unions around reconciling strategy and democracy (Gumbrell-McCormick and Hyman, 2013).

The more radical potential of organizing approaches, with its encouragement of worker self-organization and empowerment, is for ‘internal organizing’ to overcome conservative unionism and to (re)create and support qualitatively different social relationships within existing unions (Carter, 2006). However, it also brings us to the more fundamental question posed by Simms and Holgate (2010b) of what unions are actually organizing for. The authors argue that many scholars writing on organizing in the UK have largely accepted organizing as a set of practices and tactics rather than as a wider political initiative. Also, the introduction of organizing in the US was essentially pragmatic, with the aim of gaining recognition from employers. Gallin (2014: 185–212), in discussing the SEIU, argues that, although organizing appears to be an approach based on a commitment to mobilization, in fact the end is to establish more effective and harmonious relations with employers based on mutual understanding. It does not necessarily follow that organizing means a shift in the way questions of partnership with management and joint working are conducted (McIlroy, 2013). Organizing is often viewed by unions as a first step to getting them into a stronger position for dialogue with employers or a way of building more membership. Heery (2002) argues that in the context of firm-based social partnership, as in certain spaces in the UK, there is no inherent or necessary incompatibility between organizing and partnership. The latter can be seen to be about preparing the ground and the organizational presence of the union within an employing organization for the purpose of partnership or other organizational activities.

A key point raised by the organizing literature is the loose interpretation of the ‘organizing model’ and the different meaning it assumes across different unions, organizational contexts and countries. Where unions have been inspired by the ‘organizing model’ as a strategy for renewal, there is no simple model to transfer, and no single rationale for its introduction – something the article reveals when looking at organizing in a context of state-regulated social partnership.

Organizing and social partnership

Much of the literature on the relationship between organizing and partnership is based on Anglo-Saxon understandings of partnership, which is not underpinned by social regulation as is the case in the Netherlands. A relatively unexplored area is how organizing approaches fit where the state is part of a more embedded partnership culture and set of relations (for an exception, see Berntsen and Lillie, 2016). The significance of organizing and how it is understood and implemented is obviously context-dependent. Within the revitalization approach ‘unions have moved toward organizing where their institutional position is weak, but where their institutional position is stronger or the political opportunity structure more open, unions have focused on building social partnership’ (Baccaro et al., 2003: 128). The degree of union institutional embeddedness to a great extent explains the adoption of specific revitalization strategies across national case studies (see Marino, 2012). The institutional embeddedness of unions results from the presence of formal and informal arrangements, and procedures, that provide unions with the ability to influence regulation at the national policy-making level. Hence, it constitutes a specific institutional resource that unions can draw on alongside other resources such as membership support and mobilization.

Heery and Adler (2004), in a five country study on organizing patterns, argue that in some contexts, focused workplace campaigns to activate rank and file workers to obtain a collective agreement simply make no sense, particularly where collective bargaining agreements are not dependent on membership strength. Implicitly, this view assumes that organizing is merely a way to maintain representative legitimacy, rather than a way to promote rank and file participation and internal democracy (Hyman, 1979). However, in most countries in western Europe, even where institutional embeddedness is high, unions have recognized the need to respond to falling union density and questions of representative legitimacy, and have to varying degrees accepted the ‘turn to organizing’ (Connolly et al., 2014; Gumbrell-McCormick and Hyman, 2013: 56). Therefore, while the responses of unions in western Europe in ‘hard times’ are to some extent path-dependent, reflecting traditional identities and ‘repertoires of action’ (Tarrow, 1998), there are signs of innovation and experimentation (see Gumbrell-McCormick and Hyman, 2013) aimed at finding a better balance between representation and participation.

The most explicit adoption of organizing approaches in European countries has been in Germany and the Netherlands (see Annesley, 2006; Berntsen and Lillie, 2016; Holtgrewe and Doellgast, 2012; Mundlak, 2007; Roosblad, 2013; Vandaele and Leschke, 2010). In Germany, organizing has had relatively limited success in terms of recruitment, and in reversing the decline in union density. To an extent this can be attributed to the ‘curse of institutional security’ (Hassel, 2007) and, as mentioned above, the fact that recruitment has traditionally been less important in highly institutionalized employment relations systems. This is not only in the case of the Ghent-type systems of Belgium, Denmark and Sweden, but in most European countries where unions have some level of institutional security underpinned by state regulation. In the Netherlands, where organizing was explicitly adopted as a strategy over a decade ago, social dialogue and collective bargaining is not dependent on membership density, but the declining membership rate has left unions open to challenge with regard to representative legitimacy (Vandaele and Leschke, 2010). In countries with regulated social partnership models, ‘external organizing’ offers the possibility to extend membership to ward off challenges by employers and the state. Outside these more conservative and pragmatic rationales, ‘internal organizing’ has the potential to encourage greater participation and mobilization of union members and promote more radical shifts in the dominant source of union legitimacy and power.

This article presents evidence on the development and implementation of organizing approaches in the Dutch union movement. Organizing approaches use assertive techniques that do not necessarily fit within a consensus-based repertoire of action, but the outcomes nonetheless help support and maintain representative legitimacy and institutional power bases through extending the representation and, fundamentally in the Dutch case, regulation of workers. Emerging research since the adoption of organizing in the Netherlands does not necessarily indicate a reversal of union fortunes in terms of union density (Gorodzeisky and Richards, 2013; Kranendonk and De Beer, 2016). Yet, the question posed in this article is how and to what extent the introduction of organizing has, in broad terms, promoted shifts in the internal workings of unions and greater sensitivity towards issues of representation and purpose. As other authors have noted (see De Beer, 2013), the current evolutions in Dutch unions could reflect a more fundamental shift from a dependence on a logic of influence to a logic of membership (Schmitter and Streeck, 1981), which has potentially important long-term implications for Dutch unionism.

Dutch unionism and employment relations

The Dutch system of employment relations, known as the ‘polder model’, has been considered an example of corporatism par excellence (Visser and Hemerijck, 1997), a ‘harmony model’ of political economy characterized by a high degree of consensus, cooperation and coordination among responsible ‘social partners’ of organized capital, organized labour and the democratic state (Hemerijck, 1995). Unions have long enjoyed strong institutional supports which have made them less dependent on membership and mobilization power sources (Gumbrell-McCormick and Hyman, 2013). There are two main union federations in the Netherlands: the Confederation of Dutch Unions (FNV), with over one million members, and the Christian-National Union Confederation (CNV), with around 300,000 members. The FNV is the most representative union in the Netherlands and is an umbrella organization representing 18 affiliated unions. Affiliated unions are generally recognized and they are directly or indirectly (through their confederations) represented in all national level advisory, consultation and policy-implementation bodies. FNV-Bondgenoten, representing mainly service sector workers, is the richest and biggest union in the Netherlands, with nearly half a million members, followed by the public services union FNV-Abvakabo, with 355,000 members.

The Dutch regulatory framework has proved to be relatively stable in the face of external challenges. The Netherlands has managed to minimize the effects of the 2008 economic and financial crisis if compared to other European countries (Hagima, 2013), thanks to low levels of public debt and low unemployment. Low unemployment is directly linked to the expansion of part-time work, which started during the 1990s and coincided with the rise of female labour market participation (Visser and Hemerijck, 1997). In the past two decades, however, the ‘success’ of this experiment has been challenged by a growing incidence of low pay and increasing use of flexible types of employment, especially in sectors such as agriculture, construction, domestic care and cleaning where union presence can be weak (Boonstra et al., 2010). Many unskilled and insecure jobs in these sectors are taken up by EU migrants through (bogus) self-employment or posting of workers (see Cremers et al., 2007), non-EU migrants and ethnic minority groups. Self-employment has increased from 12 per cent in 2007 to 16 per cent in 2014, which places the Netherlands above the European average (Institute for Public Policy Research (IPPR), 2015). The union role in promoting regulation in these segments of the labour market has diminished in recent decades (Berntsen, 2015) due to both declining union membership and collective bargaining, which has focused on managing flexibility rather than reducing it (Keune, 2013). Union density declined from 40 per cent in 1960 to 18 per cent in 2013 (Visser, 2015). Membership is relatively high in the public sector, especially in education and construction, while it is relatively low in the service industry, especially in the hotel and catering industry. Despite the membership decline, the highly centralized Dutch unions remain central actors with a relatively high collective bargaining coverage of 85 per cent. Although employers have never systematically tried to destabilize the unions or to create a union-free environment (Visser and Hemerijck, 1997), the decline in membership density undermines representative legitimacy, which the unions have recognized in the recent turn to organizing (Roosblad, 2013).

Research context and design

In 2003–2004, in the context of economic recession, social dialogue broke down between unions and the government regarding proposed changes to early retirement and pension reform. This breakdown exposed weaknesses in the ‘polder model’ and pushed the Dutch unions to stage the second largest demonstration in the post-war period, forcing the government back to the negotiating table. After nearly a month of negotiations, the union federations accepted a wage freeze in exchange for a softening of social security retrenchments. The FNV and affiliated unions attributed great importance to this mobilization, and promoting workers’ participation and representing under-represented groups became central issues in the 2005 FNV Congress. Topics such as representativeness, union democracy, workplace relations and participation were discussed, which also influenced the stances taken towards migrants and ethnic minority workers – who tended to be over-represented in sectors with low membership density and precarious working conditions. The union made ‘internationalist’ declarations on the need to recruit these workers and create structures of representation. In April 2005, the FNV published the results of a study on union innovations (Kloosterboer, 2007), with the explicit intent to provide new strategic references for Dutch unions, who were trying to ‘redefine themselves’. In order to build union membership and develop member engagement, Dutch union activists built up a series of links with both SEIU activists as well as with organizers working on the ‘Justice for Cleaners’ campaign in London. The organizing approach was embraced especially by FNV-Bondgenoten, the largest and arguably the most assertive of the affiliated unions, which made the first attempt at introducing organizing in the cleaning sector. The cleaning sector revealed itself to be the ideal terrain for an experiment and our data collection was mainly focused here, with the aim of understanding how the approach was initially introduced and its first stages of development.

The cleaning sector in the Netherlands is relatively small, but still an important part of the Dutch service economy, employing around 2 per cent of the workforce. The economic downturn from 2008 put pressure on labour conditions, already facing strong competition, resulting in a negative price spiral and lower wages. In 2010, there were an estimated 15,000 union members in FNV-Bondgenoten and 2500 in the CNV-affiliated union. Out of a total of 140,000 workers in the sector, this represents a membership density of around 12.5 per cent. The main employers are organized in one association with the 750 member companies representing 70 per cent of employees in the sector. There are around 3500 companies in the sector but collective agreements can be extended to cover the whole sector. The majority of the workforce is female and from migrant or ethnic minority backgrounds and 80 per cent of cleaning sector union membership in FNV-Bondgenoten is estimated to be female (EurWORK, 2012). The recent disputes in the cleaning sector, which are discussed in more detail below, have targeted contract companies in strategic segments such as the airports and the railways.

The research for this article draws on over 50 interviews with union officials and organizers from various levels and sectors within the Dutch union movement, as well as non-participant observation of union meetings and conferences. The broader question of how unions were responding to migrant workers was the main focus of the research, but in the Netherlands the introduction of organizing as a narrative for renewal and as a set of techniques dominated interviews and union activities during the field work. The research was carried out from 2008 to 2013 and the majority of interviews were conducted with officials and organizers in the service sector union, FNV-Bondgenoten. Where possible the interviews were recorded and transcribed and detailed field notes were taken during and after non-participant observations. A first set of interviews and non-participant observation was carried out from 2008 to 2011 in the context of a Leverhulme Trust-funded project entitled Social Inclusion, Unions and Migration. The non-participant observation involved attending union meetings with cleaning sector workers and FNV-Bondgenoten officials in the build up to strike action in Schiphol Airport and in the aftermath of the action. Additional interviews were carried out in 2013 within an ongoing Economic and Social Research Council (ESRC)-funded research entitled Migration and Trade Union Responses. The aim of this further data collection was to investigate more closely the way in which organizing was being framed within the internal debate and the attempt to promote its adoption in other sectors of the economy.

‘Justice for Janitors’ goes Dutch

The cleaners’ campaign was launched in 2007 during a meeting at Schiphol Airport attended by 500 cleaners. In the following months, organizing committees were created in Maastricht, The Hague, Utrecht and at Schiphol Airport. Migrants’ organizations, churches, mosques, social movement groups and others pledged their support. The campaign was framed around two issues: to fight for an increase to 10 Euros an hour and for greater respect for cleaning workers by employers. In the beginning, the union concentrated high levels of resources in the cleaning sector and also encouraged self-organization and the formation of leaders at the workplace level. The union activists used tactics and strategies of organizing common to campaigns used in other countries – not only ‘Justice for Janitors’ in the US but ‘Justice for Cleaners’ in the UK – which included mapping workplaces and targeting and ‘shaming’ the client companies of cleaning contractors. The union activists declared that they used ‘the classic organizing techniques, as developed by the SEIU, based on the blitz approach. This approach is very much based on mobilizing directly, raising profiles, getting the media in’ (Int.2: FNV-Bondgenoten organizer). The campaign was considered unique for the Netherlands and had public visibility through the media. According to one of the organizers, this visibility was a fundamental element of the campaign: ‘It had more to do with the vision … you know, we were in all the newspapers, we were on TV … many people in this country thought that all cleaners had joined the strike’ (Int.3: FNV-Bondgenoten organizer).

In Amsterdam, the campaign involved direct action against client companies, including the banks ING and ABNAmbro and Schiphol Airport. The cleaners and activists accompanied by a samba band and ‘rebel clowns’ stormed bank headquarters; they also went on ‘millionaires tours’, visiting the richest bosses of cleaning companies. A combination of grassroots organizing, direct action and broad coalitions applied pressure on employers and their contractors. The campaign showed results after a year when, in 2008, cleaners won higher wages as a result of the ‘10 Euro’ campaign. In early 2008, FNV-Bondgenoten negotiated an agreement on an increase in wages (from 8.90 to 10.00 Euros), vocational training, language courses and a more transparent collective agreement. These outcomes were celebrated and became an important reference in driving further union activity. The union also registered an increase in membership (about 2000 new members) and four new organizers were appointed (Int.9: FNV-Bondgenoten policy analyst).

In early 2009, FNV-Bondgenoten began a further campaign to organize cleaners in Schiphol Airport. The union recruited over half of the cleaning workers in the airport and the activists were able build on the success of the 2007/2008 campaign to mobilize workers to try and achieve better working conditions. The campaign was again successful and after four days of strike action, the cleaners won travel expenses, job security and a 50 Euro bonus. They also managed to negotiate a one-off bonus for all Dutch cleaners of 0.5 per cent of their yearly income. The campaign continued until 2010 and culminated in prolonged strike action concentrated in key areas of the economy, mainly Schiphol Airport, but the railways were also targeted. The cleaners won further concessions from employers and were able to negotiate sectoral level agreements in the cleaning sector. The campaign resulted in improved working conditions in the cleaning sector and led to the development of a core of union organizers in FNV-Bondgenoten, with some new representatives emerging from the cleaning workers.

On 2 January 2012, the cleaners went on a 105-day strike. They achieved a 4.85 per cent wage rise, better training, frequent workload assessments and more security for agency workers. They also achieved better terms for sickness absence. The cleaners held 10 ‘Marches of Respect’ and had support from major clients such as Dutch Railways, Schiphol Airport and the Erasmus Medical Centre. A major part of the campaign was to raise awareness among both employers and the public of the value of the work done by cleaners. The issue of recognizing value and respecting cleaning workers has been a key mobilizing frame used by workers to gain public support and pressure employers to accept better terms and conditions for these workers.

The success of the cleaners’ campaign was built on an ability to empathize and engage with the workforce in new and novel ways. In meetings during and after the campaign a very positive and supportive approach towards new activists from union officials and organizers was apparent. Key features of the ‘Schoon Genoeg’ (‘Clean enough’) campaign were the high levels of commitment of union organizers and high levels of resources concentrated on building up self-organization among the cleaners. The ideational and political investment was very strong, as was well described by one of the organizers:

This is not just a job, if I can come here and I see this as a job I will have quit it a long time ago. This is like what we are … I am part of a very historical moment in the Netherlands. I am part of something that has never happened before. (Int.10: FNV-Bondgenoten organizer)

Among some Dutch union organizers there was a highly enthusiastic almost cult-like status attached to organizing – which was reflected in our research by one organizer having ‘organize’ tattooed on his forearm.

Organizing began to represent a link into the new disorganized features of the Dutch economy and a new wave of migration that was less protected by the heritage of the ‘polder model’:

It’s like ‘Well we need to change’ […] For example, with undocumented work, that’s always been a situation of bad exploitation and people sort of close their eyes to [it]. Also, in sections like cleaning and also in catering, for example, there’s a lot of precarious work. (Int.8: FNV-Bondgenoten official)

Following the success achieved in the cleaning sector, other affiliated unions, such as the public sector union FNV-Abvakabo and the construction union FNV-Bouw, applied organizing techniques in other organizational contexts. It was recognized by some in FNV-Bondgenoten that organizing as a strategy would not work for all types of disputes or in all sectors of industry, but that it was gaining support within the union movement:

For example, there are also some older sectors, like manufacturing, with a strong position historically. They are not really all that interested in organizing. At some point they used to be a bit critical of it also. But actually you have to have some very good debates and I think criticism is going away and making room for us. They already have a very strong position, high density, so they don’t need those kinds of campaigns that you do with cleaning for example, and that makes sense, even though in some sectors we could do with a few more members still. But there is also a shared idea that as a union you have to be strong. You can’t just negotiate. (Int.8: FNV-Bondgenoten official)

Organizing campaigns were highly labour-intensive, which meant that other sectors had to subsidize organizing efforts in the cleaning sector. Some union officials argued that the approach did not fit Dutch approaches to dispute resolution and they stressed how strike action had to be a last resort when all other options (consultation) had failed. This shows an evident tension between a pragmatic and more principled support for organizing. As explained by an FNV-Bondgenoten official: ‘It’s a bit difficult for a Federation when they are trying to negotiate [with employers] to have these unions that are more and more … becoming activist [in nature] and thus more demanding’ (Int.8: FNV-Bondgenoten official). Organizing tactics therefore attracted some criticism from more moderate factions within the FNV and brought out tensions in the adoption of organizing within consensus-based sectors.

Organizing and social dialogue

The relationship between organizing and social partnership was complex, especially in those sectors of the economy where unions were still able to promote regulation through ‘traditional tools’. Despite the ideal emphasis placed on organizing, as underlined above, supporters of this strategy had not rejected social partnership or social dialogue per se. This reversion to type (in social partnership terms) is evident in the targets of organizing campaigns as well as in the scope of the outcomes achieved through them, as illustrated by the example below:

The strike ended with a collective agreement that established a committee that consists of three parties – the employers in the cleaning sector, the unions in the cleaning sector but also of the large banks, hospitals, Schiphol, the railways, which buy the cleaning services. And we, as a union, we are now trying to develop a system in which they will be paid a fair price for cleaning services, which will lead to less pressure on the workers. I think we were able to develop this strategy and commission because we started with organizing. […] And it is working rather well. If this commission will continue, it will get a monitoring task. So it will monitor how Schiphol will buy its cleaning services. Agreement will be reached within the tripartite commission. (Int.6: FNV-Bondgenoten policy analyst)

At the same time, however, the recognition of the value of promoting collective regulation through ‘partnership’ was supported by the importance of promoting a more democratic approach to employment relations to be achieved by ‘going back to basics’ (Int.2: FNV-Bondgenoten organizer). That, in practice, implied strengthening links with the union rank and file, in order to acquire a stronger bargaining power derived from a more active and militant membership, rather than relying on institutional ‘partners’ for recognition:

It has to be a deal [that] we sign from a position of power … Not from the institution, not from the fact that they say, ‘I accept that you’re sitting at the table’. No! They have to say, ‘Well, I’m not able to get rid of you at the table because you’re in power’. (Int.3: FNV-Bondgenoten organizer)

The shift towards organizing, therefore, began to involve a more systematic reflection of union purpose and identity and was linked to a concern for a return to ‘class politics’ in some form or other. This purpose required changes to be made to the organization itself and the way in which it worked internally, in relation to its members and counterparts. This movement, therefore, also involved the promotion of a new set of leaders – not externally recruited but embedded internally within the unions: ‘Now we recruit organizers, and at the end they will be officials, they will be the negotiators of the union. That’s a good way but it’ll take some time’ (Int.3: FNV-Bondgenoten organizer).

The idea therefore was not just to apply organizing techniques used by union experiences in other contexts, but to use organizing as a way to try to restart reform within the union. As one FNV-Bondgenoten official stated: ‘This could be the start of change within the union and more focus on decent work, more focus on a more democratic structure, more a movement type union’ (Int.8: FNV-Bondgenoten official).

The nature of the national context meant that organizing approaches appeared at first to be more at odds with specific cultures of consensus-based social dialogue. However, over time, organizing narratives and practices from within and beyond the union leadership were used to open up a broader reflection on the aims and purposes of organized labour. What is more, these new approaches acted as a reference point for new alternative agendas and internal union discussions about purpose and identity.

Emerging challenges of organizing

The formal adoption of the ‘organizing model’, as formulated in the 2005 FNV Congress, resulted in the implementation of organizing techniques at decentralized levels. The approach found supporters in union executive boards of the affiliated unions and sympathizers within the FNV and stimulated an internal process of organizational reflection. In the course of time, the approach acquired a greater measure of consent within the union official debate and several projects aimed at promoting union presence and activism at local levels emerged in affiliated unions not directly involved in organizing campaigns. Such projects, inspired by organizing principles, had the goal of promoting structural changes, attempting to move beyond the problem of ‘one-issue’ campaigns and to shift the dominant repertoire of action within the union movement.

Several challenges emerged as a result of the introduction of organizing tactics. First, organizing presented a departure from less confrontational strategies traditionally employed in the Dutch union movement. This presented difficulties for some unions in the context of a traditional repertoire of consensus-based social dialogue. For example, in the cleaning campaign, the cleaning activists targeted finance companies who were the client companies for cleaning contracts. In one campaign, inspired by tactics from US-style organizing campaigns, the cleaners offered a ‘golden shit’ award to the worst client company, which presented a dilemma for the finance sector employees and various union officers who had a good working relationship with their employer. The manner of such organizing tactics introduces a more conflictual element that can challenge consensual relations between unions and employers built up within a national social partnership model of employment relations. While the more pragmatic rationale of organizing, in terms of achieving greater collective representation and regulation, can complement social partnership models (Heery, 2002), the methods for achieving this outcome can create tensions within more consensus-based sectors and organizations.

Second, organizing campaigns have been concentrated around low-wage work and some union officials stated that it was difficult for unions to transfer organizing techniques into more traditional areas of the economy, such as nursing or ports and chemicals, for example. The cleaning campaign inspired follow-ups in other relatively less regulated sectors of the Dutch economy, such as domestic work, agriculture and the retail industry, where union presence was lower. However, the regulatory process in more highly unionized, traditional sectors still retained a form of institutionalism and social partnership approach towards employment relations (see Roosblad, 2013).

Third, there was the broader challenge around the sustainability of organizing, as the cleaning campaign was framed around achieving a particular result. Critics of organizing show that there is a tendency for ‘one-issue’ organizing campaigns to dissipate once the desired result is achieved. Linked to this is the extent to which members are really in control of the campaign – and even if organizing campaigns have led to securing better conditions of employment, there is a question as to what extent there has actually been more democratic unionism with more grassroots participation within the structures of the union and not just the organizational spaces of specific campaigns (McAlevey, 2012; Savage, 2006). There were signs that change would be more difficult given the prescribed way of doing union work, but the longer-term outcomes would require further research. In 2009, FNV-affiliated unions recorded increases in membership – with FNV-Bondgenoten growing by 2500 members in a three-month period. The assertive campaign in Schiphol Airport led to over half of cleaning workers becoming members of FNV-Bondgenoten. While these results are impressive, it is difficult to measure the extent of engagement of these workers in the post-campaign period. From observations of local union meetings in the post-strike phase, managing the expectations of the cleaners in cases such as the Schiphol campaign was difficult. The expectations of workers had risen, but the behaviour of management and supervision, and some of the basic conditions, in terms of breaks and monitoring for example, were still a focus of concern for workers. This post-event phase in organizing was clearly challenging as issues emerged on a constant basis, but union activity had to be measured and carefully planned due to the need to build more sustainable workplace institutions in order to be able to achieve better outcomes through negotiation. This observation reflects the difficult balance in the ‘top-down’ and ‘bottom-up’ nature of organizing discussed earlier (Heery, 2005; Milkman, 2006), and the tensions in reconciling strategy and democracy. The organizers effectively became ‘managers of discontent’ and sought to both contain and encourage worker self-organization and activism (see, also, Simms (2007) on ‘managed activism’ in organizing campaigns).

Despite these challenges, organizing in the Netherlands has been part of a process of internal reflection and a shift of emphasis in certain unions (De Beer, 2013; Roosblad, 2013). Organizers in the unions researched felt that union strength coming from a renewed membership might promote a shift in the dominant source of union legitimacy and power, but essentially without erasing the confidence in the dominant model based on regulated social partnership. The extent to which organizing principles will be able to promote organizational changes in terms of structure, culture and activity remain open to question, but its introduction already constitutes an important framework and reference point for union debates on change. As stated earlier, the impact of ‘internal organizing’ is something that is more indirect in many ways and difficult to measure.

Conclusion

This article has considered developments around organizing as a strategy for renewal in the Dutch context. The findings offer some broader lessons on the ways in which organizing approaches are understood and implemented in different regulatory contexts. In the Dutch case, the introduction of organizing constitutes, using Schmitter and Streeck (1981) categories, a move from a dependency on the logic of influence to one also based on the logic of membership, which has important potential long-term implications for the union movement and the way it responds to future challenges to its institutional power bases. In the Netherlands, where social regulation is relatively strong and consensus-based social dialogue is the dominant repertoire, compared to say the UK or USA, the goal of internal democracy and participation – and therefore membership recruitment and mobilization – has been considered less necessary as the union’s purpose is defined in terms of its institutional role and position (Baccaro et al., 2003; Marino, 2012). Unions are accepted as being embedded in a range of institutions in order to engage in the main task of collective bargaining. Considering the regulatory context, the institutional position of unions and traditions of union identity in the Netherlands, one would expect to find unions’ strategies for renewal to be based around the strengthening of social rights and regulation, and strengthening institutional power bases (Connolly et al., 2014). To a great extent this remains the case, with Dutch unions continuing to play a role in national level collective bargaining and continued recognition, formally at least, as a social partner by employers and the state.

However, while our research shows a level of path-dependency, it also demonstrates how unions have been experimenting with new ways of acting and engaging and how they have been inspired by organizing approaches (Gumbrell-McCormick and Hyman, 2013). In spite of the institutional and political supports, unions in the Netherlands are exposed to environmental uncertainty, and organizing has been an attempt to limit the union movement’s exposure, extend membership and maintain and reinforce representative legitimacy. The adoption of ‘external organizing’ techniques has led to successes in recruiting new workers and increasing mobilizing capacity, particularly in the cleaning sector, which has led to better regulation in the form of national collective agreements. Even though the main focus of organizing was on extending membership to under-represented groups, the more radical narrative and tactics of organizing (as opposed to social dialogue) has in turn promoted ‘internal organizing’, resulting in changes in internal dynamics and relationships. Therefore, there was a strategic ‘by-product’ of adopting organizing, which helped frame internal union reform and arguably encouraged a process of democratization within the union. The ‘vision’ of the ‘organizing model’ has acted as a space which provides new narratives and relationships, stimulating alternate ways of developing strategy. The overall repertoire of social dialogue is maintained but within this repertoire there have been changing and qualitatively different social relationships as a result of organizing, which point towards the potential of organizing to overcome a more conservative unionism and encourage greater democracy (Carter, 2006; De Turberville, 2007), although this is not an unambiguous inevitability.

The question is: how does organizing embed itself and maintain a momentum at the same time? This links in to fundamental questions about the underlying rationale of organizing and ‘what are we organizing for?’ (Simms and Holgate, 2010b). To an extent, this question has greater relevance in the UK (or USA) due to the lack of state regulation of employment relations. In the Dutch context, organizing is used to build state regulation through extending representation. If the central elements of organizing are worker self-organization, empowerment and mobilization, the evidence suggests that this occurs during certain stages of organizing ‘campaigns’, but the extent to which this is sustainable and relevant once the goal of greater collective regulation is achieved, is a matter of debate in a context of regulated social partnership.