In South Korea, non-regular employment has proliferated since the 1998 financial crisis. Many companies carried out unprecedented downsizing of regular employees en masse and began to adopt flexible employment relations by expanding the use of non-regular labour; this trend was encouraged by the government’s neoliberal labour market reforms. The diffusion of a non-regular workforce[1] has resulted in the growing polarization of labour markets, because non-regular workers’ employment conditions are quite inferior to those of regular workers in terms of wages, fringe benefits, and institutional protection.

During the post-1998 economic crisis period, some non-regular workers organized their own labour unions and engaged in protest actions against discriminatory treatment and precarious employment conditions. Their protest actions were largely triggered by employers’ despotic behaviours, including forced wage cuts, unilateral termination of employment contracts, discriminatory and illegal employment practices, and violent suppression of unions. Non-regular workers’ struggles are viewed as a new wave of labour insurgency in South Korea, symbolizing a revolt of the precariat class, a term coined by Standing (2011), against the neoliberal labour regime; these workers and their unions have shown militant activism to resist the labour market flexibilization that has victimized them.

Many of the precarious workers’ struggles failed to achieve their intended outcomes, due to a lack of resources and employers’ oppressive acts. Some of them, however, succeeded in achieving their demands and saw union membership increase through their protest actions. What leads to the difference between successful and unsuccessful precarious workers’ struggles? This is our research question, which we aim to address by examining major cases of non-regular workers’ struggles in South Korea. We examine the factors that help these precarious workers overcome tyrannical employers against the backdrop of the neoliberal capitalist state. Our study contributes to the theoretical elaboration of labour movement revitalization for the precariat class, which is likewise victimized under the neoliberal globalization regime, by shedding light on the activism of precarious workers and the conditions under which they achieve desired outcomes from their struggles. Literature in the English-speaking world has paid little attention to the active role of such atypical workers in staging various protests against employers’ inhumane and discriminatory treatments against the background of a market-driven labour regime, although there is an emergent research interest in this subject in recent labour movement literature.

This study explores causal conditions of successful non-regular workers’ struggles via abductive reasoning, which is a form of logical inference that moves from observations to a theory to account for the observations. In adopting the abductive approach, we employ the research method of fuzzy-set qualitative comparative analysis (fs/QCA) to examine middle-N cases. The fs/QCA is appropriate because of its emphasis on identifying conditions rather than variables. Hence, we apply this abductive approach toward theorizing key conditions for enabling precarious workers to win against employers, by drawing upon 30 cases of non-regular workers’ struggles pertaining to the period between 1998 and 2012. Our focus is on three dimensions of precarious workers’ struggles to provide a “grounded” explanation concerning the success of those struggles: 1- internal solidarity; 2- external solidarity; and 3- struggle repertoire.

As South Korean firms began expanding their use of non-regular labour to replace regular employees during the 1998 economic crisis, the presence of precarious workers increased sharply (beginning in the late 1990s). In fact, the share of temporary and daily workers, as estimated by the National Survey of Economically Active Population, rose sharply from 41.8% in 1995 to 52.1% in 2000. According to the government’s official estimates, based on the Economically Active Population Supplementary Survey, which has been conducted since 2001 to capture the size of the non-regular workforce and the related employment conditions, such atypical employment rose from 26.8% in 2001 to 37.0% in 2004, and has since declined to 32.5% as of 2015 (Korea Labor Institute, 2015). Non-regular workers, comprising nearly a third of the entire workforce, have faced discriminatory treatment and institutional exclusion—phenomena that are closely related to their precarious employment status. A substantial wage gap exists between regular and non-regular workers, and has been widening over the past 15 years. The average hourly wages of non-regular workers have fallen from 80.5% in 2001 to 65.0% in 2015, compared with that of regular workers (=100). Moreover, a majority of non-regular workers have been excluded from the statutory protection of social welfare (i.e. employment insurance, medical insurance, and national pensions) and labour standards (i.e. severance pay, overtime work pay, and paid vacations), as demonstrated in Table 1.

Given the growing presence of the non-regular workforce and their discriminatory employment conditions, these precarious workers have organized unions and launched a variety of collective actions to protest against employers’ unfair treatments since the 1998 crisis. During the post-1997 period, union organizing and protest actions by non-regular workers have steadily increased in various industrial sectors, such as construction, manufacturing, transportation, and service in both private and public sectors. Consequently, the relative share of their struggles in the total number of labour disputes has soared from 0.7% to over 20% over the same period, as shown in Figure 1.[2]

Non-regular workers’ unions, however, are severely constrained by enterprise unionism, which is rooted in the traditional setting of South Korean enterprise-based industrial relations. Non-regular workers are often denied membership in existing unions, which represent only regular workers within the boundary of the enterprise. As a result, non-regular workers attempt to unionize themselves by organizing their independent unions. In the segmented workplace, the former group is primarily viewed as a buffer to guarantee job security and lower work burdens for the latter group (Lee and Frenkel, 2004). Problematically, non-regular workers are commonly faced with limited power resources, resulting from their precarious employment status and vulnerable working conditions, in contrast to regular workers, whose unions have sufficient organizational resources and entrenched bargaining power in labour-management relations at the enterprise level. Accordingly, when defending their union from the employer’s harsh oppression, or mobilizing protest actions to press their claims, non-regular workers tend to engage in protracted struggles—often lasting longer than a hundred days—and resort to an unconventional and high-risk struggle repertoire, which goes beyond the workplace, due to the constraint of their limited organizational resources (Lee, 2016). Under the disadvantageous context of enterprise-based industrial relations, it is difficult for non-regular workers’ unions to maintain organizational sustainability, which is evidenced by the decline in union membership for these workers from 5.2% in 2004 to 2.8% in 2015, in contrast to regular workers’ union membership, which has remained steady at a level of 16.7-16.9%.

In response to the growing presence of atypical employment patterns in both Western advanced countries and non-Western developing countries, the English-language labour movement literature has recently paid closer attention to how labour unions react to the changing landscape of labour markets. Much research in the labour movement literature casts light on how labour unions formulate their policy stance toward precarious labour and launch organizing campaigns to recruit atypical workers. This research discusses the nature of challenges created by atypical labour for labour unions and highlights how those unions deal with such challenges by re-establishing their strategic approach toward precarious workforces in terms of labour market regulations, membership recruitment, and collective bargaining or organizational representation, whether from a macro perspective of cross-country comparison (Pulignano et al., 2016; Kornelakis and Voskeritsian, 2016; Gumbrell-McCormick, 2011; Pernicka, 2005), or from a perspective of micro-level case studies (Benassi and Dorigatti, 2015; Simms and Dean, 2015; Conley and Stewart, 2008). Most of this literature focuses on how existing unions respond to atypical labour by treating this worker group as an object of union representation, but hardly pays attention to the new activism of precarious workers, who act as key agents to unionize themselves and mobilize their own protests at the margins of labour markets. Only a fraction of the recent labour movement literature delves into the active roles that precarious workers assume in their struggles: among them are Chun (2009), who highlights the symbolic leverage of precarious workers having weak structural power to strengthen their associational power by classification struggles and public drama in USA and South Korea, and Atzeni (2016), and Mattoni and Vogiatzoglou (2014), who commonly underscore collective interests or identification to mobilize precarious workers’ self-activity in a bottom-up approach. In light of this, our study will address a research gap by exploring the conditions under which precarious workers, as key agents of new labour movements, succeed in accomplishing demands and organizing unions through their collective action.

As Rodgers and Rodgers (1989) indicate, precarious workers, equated with non-regular workers in South Korea, are typically characterized as having insecure low-income jobs, as experiencing a lack of control over working conditions, and being excluded from legal and organizational protection. Given such vulnerable employment conditions, it is very difficult for those precarious workers to organize their own unions and mobilize collective action against employers, in contrast to regular workers who are guaranteed job security and an institutionalized voice in the workplace. Precarious workers tend to be hesitant to join unions and participate in union-led action, because of their insecure employment status and fear of job loss. Additionally, the unionization of precarious workers is often confronted with employers’ determined opposition, which is closely linked with their primary motivation to use non-regular labour to avoid the organizing and expansion of labour unions in the workplace. Because of their disposable employment status and fragile collectivity, precarious workers are likely to lack both associational and structural power[3] at the workplace level, and suffer disadvantageous conditions with which to cope with employers’ determined repressions, even when they succeed in organizing a union (Lee, 2016).

Precarious workers who are involved in a contentious situation with their employers require particular conditions that enable their protest actions to succeed in overcoming their precariousness and fulfilling their demands, thereby producing “the paradox of power.”[4] Since precarious workers lack power resources from within their unions to deal with employers’ repressive reaction, it is important for them to build solidarity networks externally and mobilize effective protest tactics with their limited action resources, in order to pressure employers to recognize their unions and accept their demands. As Frege and Kelly (2003) underscore, coalition building with other social movements becomes a key union revitalization strategy for overcoming the crisis of union movements; solidarity networking is a necessary condition for precarious workers to organize a union and mobilize protest actions against their employer’s repression, while lacking sufficient power resources. In addition, the protest repertoires devised and employed in struggles of precarious workers with limited action resources are significant for their success.

Solidarity networking is conceptualized as building a community of common interests, common feelings, and joint actions (D’Art and Turner, 2002). Indeed, solidarity between regular and non-regular workers, who have different interests and feelings in the segmented workplace, could be a crucial condition in helping the latter group make gains in conflicts with an employer. The regular workers’ union, which has an established position within the workplace and possesses sufficient power resources, can offer material and emotional support to the precarious workers’ struggle, which is restrained by poor action resources, and mobilize joint action to pressure the employer to address the latter’s protest. Solidarity networking with community and civil society movements outside the workplace could be another significant source of union power for precarious workers. Precarious workers’ claims are largely aimed at employers’ arbitrary dismissals, inhumane working conditions, and illegal treatments, including union suppression. Thus, the precarious workers’ struggle to defend their basic well-being and labour rights is often viewed as a public drama (Chun, 2009), with a societal effect that resonates to create public concern over employers’ injustice imposed on those marginal workers. Accordingly, when precarious workers get involved in a contentious situation with an employer, solidarity support from labour and civil society organizations plays a key part in making up for the lack of power resources, and public opinion that sympathizes with their protest becomes a powerful ally to push the employer into correcting unjust treatments against those workers.

The choice of struggle repertoire or protest tactics also affects the outcome of the precarious workers’ struggle. A new repertoire of collective actions, extending beyond the boundary of the workplace and appealing to the public, has become increasingly important to the precarious workforce.[5] McAdam and his colleagues (2001) differentiate between the contained and transgressive forms of protest repertoire; the former refers to collective action in which protesting actors employ well-established means of claim making, while the latter refers to protest action in which newly self-identified actors adopt an innovative repertoire. McAdam (1986) also distinguishes protest action between low-risk/cost and high-risk/cost types.[6] In this light of theoretical distinction concerning protest repertoires, precarious workers tend to devise and mobilize a variety of transgressive and, sometimes, high-risk protest repertoires to overcome employers’ repressive actions and attract public attention under a disadvantageous power imbalance. This is in contrast with the regular workers’ unions, which can achieve their claims by resorting to conventional and contained struggle repertoires in an institutionalized setting of labour-management relations (Lee, 2016; Lee, 2015).

In summary, internal solidarity (with regular workers), external solidarity (with community and civil society movements), and the strategic choice of protest repertoires are presumed as key conditions in producing successful outcomes for precarious workers’ struggles, which are constrained by meagre power resources, against employers’ repressive domination in the context of the neoliberal labour market regime. These three conditions—internal solidarity, external networking, and protest repertoires—are highlighted as the key elements of unions’ strategic capacity in the union power model, elaborated by Lévesque and Murray (2010), and Ganz (2000).

For fuzzy-set analysis, the measured score of the conditions and outcomes for each case is converted into a fuzzy membership score. Four of the dimensions have a range of scores from 0 to 2, and the “struggle repertoire” dimension has a range from 2 to 7. Three points are required to calibrate for fuzzy-set analysis: a cut-off point for “fully in” the set, a cut-off point for “fully out” of the set, and the crossover point, as displayed in Table 3. The cut-off point for “fully in” is the minimum value in the range for each condition, and “fully out” is the maximum value. The crossover point, which means “neither in the set nor out of the set,” is the median value within the range.

Based on the calibration rule noted above, fuzzy-set scores for the five dimensions of the 30 cases are presented in Table 4.

In this study, we carry out the standard fuzzy-set analysis by using fs/QCA software (version 2.5), developed by Ragin and Davey (2014), and the intermediate solution is adopted for producing result configurations.[11] Conditional configurations, having no relevant case, are excluded from the analysis. If the conditional configuration has a value of consistency higher than 0.75, the outcome is coded as “1”; if not, the outcome is coded as “0.” The result configurations that have coverage higher than 0.3 and consistency over 0.75 are accepted as significant causal conditions.

In Korea, many struggles of non-regular workers, who attempted to organize their own unions and engaged in militant action to protest against employers’ inhumane discrimination and illegal exclusion, have failed to achieve the desired outcomes due to the inadequacy of their action resources, which is largely associated with their vulnerable employment status. In this light, our study examined the conditions that led to victory in precarious workers’ struggles, by focusing on three attributes (i.e. internal solidarity from regular workers, external solidarity from labour and civil society groups outside the workplace, and protest repertoires mobilized), among various factors affecting these struggles. Specifically, this study identified the configurations of these three conditions that produced successful outcomes in precarious workers’ struggles, in terms of bargaining gains and organizational sustainability, by employing fs/QCA modelling to examine 30 major cases of non-regular workers’ struggles, occurring during the 16-year period from 1998 to 2013.

A key finding is that the conditional configuration of strong external solidarity, strong internal solidarity, and fewer struggle repertoires constitutes a significant causal path to successful outcomes, as identified in the fs/QCA models concerning bargaining gains, as well as the combined outcome of bargaining and organizational gains. Moreover, the configuration of strong external solidarity and fewer struggle repertoires is found to be another significant causal path to success in terms of organizational gains. These findings reaffirm the idea that strong solidarity bridging, whether with regular workers that have a different employment status in the segmented workplace, or with labour and civil society groups outside the workplace, is the crucial causal condition for precarious workers to achieve desired outcomes from struggles, as highlighted in the existing literature of union revitalization and social movement unionism. In this vein, it cannot be overemphasized that the building of “organic solidarity” (Zoll, 2000) or “protest alliances” (Wills, 2009) with workers of differing statuses and social actors in the civil community is key to achieving desired outcomes from struggles of precarious workers, who are trapped in poor circumstances for protest action. This is true from a theoretical perspective of social movement, addressing resource mobilization as a sine qua non factor that influences the course and outcomes of labour activism (McCarthy and Zald, 1977). Thus, it should be underscored that precarious workers, lacking mobilizable action resources, must consciously go beyond the differentiated status and interest structures of the segmented workplace and forge working-class bonds to mobilize the internal solidarity resources provided by regular workers. They must also publicly dramatize their issues and protest actions as a means of generating broad resonance with labour organizations and civil society groups outside the workplace, thereby inducing their active involvement and provision of external solidarity resources.

At the same time, our analysis presents the interesting, but unexpected, finding that when precarious workers’ struggles mobilize fewer struggle repertoires, they are likely to achieve successful outcomes of bargaining and organizational gains. Thus, mobilized protest repertoires negatively affect their success, even if they receive internal and external support While the existing literature on social contention and activism (i.e. McAdam, 1986; McAdam et al., 2001) addresses the forms (i.e. contained or conventional vs. transgressive) and nature (i.e. low risk/cost vs. high risk/cost) of protest repertoires, our analysis finds that the quantity of action repertoires has a significant effect on the outcome of workers’ struggles. Specifically, the finding that employing fewer protest repertoires is the key condition for precarious workers’ struggles to achieve better outcomes implies that excessive mobilization of repertoires might lead protest actions to become extreme and isolated from actors who could provide internal and external solidarity, and who become offended by those workers’ over-militancy. As indicated by Chun (2013), the mobilization of more protest repertoires tends to lock workers into protracted struggles, which involve declines in union members’ participation and solidarity actors’ support, and escalate protests into desperate, deadlock battles. Linking with the typology of protest repertoires addressed by McAdam (1986), this finding implies that precarious workers who fail to obtain their desired outcome with protest repertoires of low risk/cost are likely to move on to struggle tactics of high risk/cost, but unlikely to achieve what they demand, thereby leading to the paradox of an infertile struggle situation where the more and riskier protest repertoires mobilized by those workers, the less desirable outcomes they gain. As such, the relationship between the quantity of protest repertoires and struggle outcomes can be interpreted as presenting a polarization of moderation (fewer struggle repertoires and better outcomes) and extremization (more struggle repertoires and worse outcomes) (Lee, 2016).

It should be noted that our research has several limitations. First, the context of our analysis is specific: the South Korean institutional framework of industrial relations, as exemplified in enterprise unionism, differs from Western advanced countries, including those in North America. Second, our fs/QCA models do not examine the effect of non-union organizational factors, such as employers’ counter-responses, which might significantly affect the outcomes of non-regular workers’ struggles. This is associated with limited secondary source data, from which we tried to extract relevant information concerning the events of non-regular workers’ struggles, but which did not contain enough information about other organizational information. Third, the sample of our fs/QCA cannot be representative of all the struggles of precarious or non-regular workers, since we cannot avoid a non-random sampling bias by mainly selecting salient cases (largely from large firms) for which sufficient information on non-regular workers’ struggles was available. The next step would be to apply the hypothetical findings, drawn from this fs/QCA, to both qualitative case study and quantitative analysis to further delve into the causal mechanism concerning the process and outcomes of precarious workers’ struggles.