Mobilizing precarious workers in Italy: two pathways of collective action intentions

ABSTRACT

Following increased flexibilization of labour market and related decline of traditional labour unionism over the last few decades, studying mobilization processes of precarious workers has become particularly timely. While localized forms of organization and unionization are gradually emerging, little is known about why workers intend to join these coordinated forms of collective action. Integrating social movement studies with social psychological literature on collective action, this study fills this gap by exploring collective action intentions in the current context of non-standard labour. To do so, we surveyed precarious workers enrolled by temporary hiring agencies in Italy (N = 379) and found two parallel psychological pathways explaining their collective action intentions. On the one hand, participants exhibited high collective action intentions when they were able to collectively identify with other precarious workers as part of the same social group. Collective identification with precarious workers increased group-based injustice that in turn predicted collective action intentions. On the other hand, participants also exhibited high collective action intentions when they were able to politically identify with unionized workers. Politicized identification with unionized workers increased collective efficacy that in turn predicted collective action intentions. By singling out the complementary role played by these two parallel pathways of collective action intentions among precarious workers, this study shed light on the socio-psychological determinants underlying the mobilization propensity of individuals still lacking any organizational affiliation, a topic that has been relatively ignored in scholarly literature. In doing so, we combine social movement studies and social psychological literature in innovative ways.

KEYWORDS:

• Precarious non-standard workers
• collective action intentions
• group-based injustice
• collective efficacy
• collective identification
• politicized identification

Introduction

Contingent and non-standard forms of contract are becoming the prevalent employment relation in contemporary capitalism, making labour increasingly ‘uncertain, unpredictable and risky from the point of view of the worker’ (Fudge, 2017; Kalleberg, 2009, p. 2). A growing number of workers lack guaranteed income, and they are characterized by remarkably low contractual power and limited or inexistent career perspectives (De Cuyper et al., 2008; Kirves et al., 2011; Nienhuser & Matiaske, 2006; Ross, 2009; Wagenaar et al., 2012). The heterogeneous set of all these fixed-term, self-employed, and non-guaranteed forms of work falls under the label of what scholarly literature defines today as ‘non-standard work’ (Cobble & Vosko, 2000). In lay thinking, this new workforce is most commonly known as ‘precariat’ (Standing, 2011).

If an abundant bulk of research has made sense of these labour market transformations over the last two decades (see, for instance, Kalleberg, 2011), only a limited number of studies have investigated the worker mobilization attempts to oppose such transformations (Grote & Wagemann, 2018). Non-standard labour is associated with specific processes of work fragmentation and precarization (Heiland, 2020), which are designed to isolate workers (Huws, 2014), and therefore jeopardize their capacity of collective action (Cini & Goldmann, 2020). Considering such circumstances, these workers have often remained voiceless and unable to organize collectively (Crouch, 2019; Giugni & Lorenzini, 2016). As a result, understanding the psychological processes why precarious workers may intend to move from a condition of non-mobilization to that of mobilization seems to be an especially relevant issue to address, especially for social movement studies (for a similar point, see van Stekelenburg & Klandermans, 2017). In line with classic attitude–behaviour models in psychology (e.g. Ajzen & Fishbein, 1977), indeed, intentions are good predictors of future behaviours, even in the absence of immediate concrete mobilization opportunities (De Weerd & Klandermans, 1999).

Integrating social movement studies with social psychological literature

Social movement research has traditionally focused on mobilized individuals (i.e. individuals actively participating in coordinated forms of collective action) by developing a useful kit of concepts and theories well adapted to understanding movement formation processes. Among such concepts (e.g. open political opportunities, resonant frames; see della Porta & Diani, 2020 for a review), mobilizing structures have been usually considered as pivotal (McCarthy & Zald, 1977). The assumption was that contentious politics needed dense networks of relations to support massive recruitment. However, the main implication of this assumption was to take for granted that people were prone to join such structures. In short, the analytical focus of these frameworks usually started from the moment of mobilization onwards.

What has been relatively missing from social movement studies was an in-depth analysis of the socio-psychological preconditions of collective action (for a notable exception, see Klandermans, 1997), namely, why certain individuals are more willing to engage in collective action than others and, more importantly, how and why the latter might be involved in such action. Accordingly, social movement scholars have mostly neglected the study of the motivations or attitudes of people not yet involved in mobilization processes. Only recently, few studies have started to explore the collective and political attitudes of individuals still lacking any organizational affiliation (Galais & Lorenzini, 2017; Portos et al., 2020).

On the other hand, research in social psychology has extensively investigated the collective action intentions of various subordinate and minority groups in society (e.g. Van Zomeren et al., 2004; Wright et al., 1990). Operationally, collective action was broadly defined as ‘the attitudinal support for protest as well as the protest intentions of members of a social group that are directed at removing the perceived underlying causes of the group’s disadvantage or problem (e.g. signing a petition and participating in a demonstration)’ (Van Zomeren et al., 2008, p. 512). Surprisingly, however, very few social psychological studies have explored collective action intentions among politically inactive workers (Manstead, 2018). The time seems ripe for social psychology to reverse this trend and study individuals’ collective action intentions in the work sphere.

In order to fill this gap, our study adopts an interdisciplinary approach able to integrate social movement and social psychology literature in innovative ways. Social movement theories provide social psychology frameworks with a greater capacity to understand the material dynamics accompanying individuals’ collective action intentions; conversely, social psychology literature lends to social movement studies a greater focus on the individuals’ interpretative processes underlying their motivation to join such mobilization attempts. In doing so, social movement and social psychological traditions can learn from each other and expand their explanatory power of complex mobilization (and demobilization) processes. This article represents one of the first attempts aiming at combining the two fields of study in order to explore the collective action intentions of individuals still lacking any organizational affiliation. Concretely, we merge insights retrieved from recent localized mobilization attempts among precarious workers in Italy and basic psychological tenets of collective action intentions, to provide first evidence about how these new forms of mobilization may catalyse the participation of politically inactive workers.

Mobilizing precarious workers in Italy

Over the last two decades, two parallel yet distinct forms of mobilization involving various categories of precarious workers have occurred in the Italian context: one more informal and based on the support of solidarity groups and social movements (Mattoni & Vogiatzoglou, 2014) and the other more traditional and based on the intervention of union confederations (Pulignano et al., 2016).

As for the first form of mobilization, since the early 2000s, various political collectives of activists have formed and spread in several Italian cities to organize precarious workers under the banner of ‘San Precario’, a fictitious patron saint invented and used as a political symbol to promote a shared collective identity (Hyman & Gumbrell-mccormick, 2017). Importantly, these spontaneous forms of organization have often been able to spur and support precarious workers’ self-organized mobilizations within different sectors and workplaces, such as call centres, publishing houses and airport services (Murgia & Selmi, 2012). These mobilizations have incentivized the growth of new labour actors – be they grassroot unions or self-organized labour collectives – able to fight for the dignity of all the categories of unprotected workers and capable of involving more recently even some platform workers (Tassinari & Maccarrone, 2020). In this sense, all these self-organized experiences of mobilization have not only advanced the labour rights of these workers but also, more importantly, pointed the possibility of the emergence of a new collective identity, ‘the precarious class’, to fight against collective injustices experienced by this subordinated workforce (Mattoni, 2012).

As for the second and parallel form of mobilization, Italian trade unions, and especially the General Confederation of Labour (CGIL), have founded specific branches for the representation of non-standard workers in the late 1990s. In 1998, the CGIL created a federation for precarious workers, Nuove Identità di Lavoro (NIdiL), aimed explicitly at increasing the protection afforded to self-employed or semi-autonomous work and temporary workers through collective bargaining and shop-floor representation. NIdiL, for example, described itself as a union that gives voice and representation to precarious workers without social protection: ‘It claims to pursue two main goals: raising awareness of the issue of precarious work and the problems and realities of precarious workers, and improving working conditions for precarious workers through re-regulation and organizing at multiple levels in industrial sectors, in society and at the workplace’ (Pulignano et al., 2016, p. 45). Albeit the representation of Italian precarious workers as a whole is still a far goal to achieve, some organizational initiatives have been successfully pursued, especially among some new service sector categories (Cella, 2012). In this sense, Italian trade unions have exhibited some efficacy.

Basic psychological tenets of collective action intentions

From a socio-psychological angle, the two mobilization attempts described hereinabove presuppose two basic psychological processes whose manifestation is needed for subordinate groups to mobilize (Kelly, 1998): a sense of injustice and a feeling of collective efficacy. First, individuals need to self-categorize as part of the same social group and to share feelings of injustice, namely, the conviction that an event, action or situation is ‘wrong’ or ‘illegitimate’ for the group as such. Second, subordinated groups must also perceive that their collective action may be effective and that when organized they are able to bring about social change. These two psychological processes are best known in social psychology as the ‘dual pathway model of collective action’ (van Stekelenburg & Klandermans, 2013; Van Zomeren et al., 2012).

Collective action intentions explained by perceived group-based injustice

Feelings of injustice about the way the subordinate group is treated are crucial to whether people react against collective disadvantage or not (Van Zomeren et al., 2008; Walker & Smith, 2002). When vulnerable people blame themselves for their economic hardship, they concurrently deny social injustice, thereby becoming politically acquiescent and anomic (Bourguignon et al., 2015). By blaming the social system instead of themselves, precarious workers stop attributing their own joblessness and precarious working conditions to personal responsibility and engage in political actions aimed at improving the situation (Breakwell, 1986; Herman, 1999). Group-based injustice is therefore critical for collective organization and action because they begin the process of detaching subordinate group members from loyalty to ruling groups or in Marx’s terms converting a class-in-itself into a class-for-itself (Jost, 1995). As similarly highlighted by studies exploring the labour context, it must be workers blaming ‘an agency for their problems, rather than attributing them to uncontrollable forces or events. That agency can then become the target for collective organization and action’ (Kelly, 1998, pp. 29–30).

Research has shown that the reappraisal of collective disadvantage in terms of perceived group-based injustice is closely tied up with the level of collective identification with the subordinated group (Van Zomeren et al., 2012). In this sense, participation in coordinated forms of collective action can be framed as ‘efforts by large numbers of people, who define themselves and are also often defined by others as a group to collectively solve problems they feel they have in common’ (Tajfel, 1981, p. 244). Subordinate group members gain social and emotional support from similar others, and they raise consciousness about injustice and willingness to strive for their rights (Herman et al., 2006; Walker & Mann, 1987). It follows that identification with precarious workers should be linked with external blame for the unfairness of collective disadvantage, which in turn should predict collective action intentions. We thus expect perceived group-based injustice to mediate the link between identification with the category of precarious workers and collective action intentions.

Collective action intentions explained by perceived collective efficacy

Perceiving social change as desirable is necessary but may not be sufficient to mobilize people. In order to participate actively in the claiming of social rights, individuals must also believe that collective action can be effective in achieving social change (Klandermans, 1997; Van Stekelenburg et al., 2009). Accordingly, collective efficacy can be framed as the belief that coordinated group action holds the potential to change current arrangements and improve group conditions. Research has shown that politicized collective identities, in particular, provide the necessary instrumental support and political cohesion to fuel appraisals of collective efficacy (Huddy, 2013; Meyer, 2017). By fulfilling psychological functions such as understanding and agency (e.g. Bandura, 1997), identification with organized political groups incentivizes people to develop a sense of collective efficacy (McAdam, 1982), which may lead them to engage in collective action (Simon & Klandermans, 2001). As similarly highlighted by studies exploring the labour context, the role of political groups has also been documented in contemporary studies of workplace mobilization (Cini & Goldmann, 2020). Bonds with unionized workers or union representatives increased workers’ sense of efficacy and, therefore, their propensity to act collectively. We thus expect perceived collective efficacy to mediate the link between identification with unionized workers and collective action intentions.

Our study

As a result of the 2008 collective labour agreement with the Italian Ministry of Labour and Social Policies, NIdiL promoted an educational project aimed at informing and warning precarious workers about the possible abuse of flexible and temporary contractual conditions. These educational projects targeted people whose field of activity varied greatly, ranging from members of the bar association, online market managers, web designers, nursing home assistants and warehouse workers. In most of the cases, participants were unemployed at the moment of the educational project or enrolled by temporary hiring agencies. Within this educational project, we designed a survey to investigate collective action intentions among such diverse categories of workers, who were all bonded by extremely precarious working conditions. Questionnaires were distributed before the class and administered by five external trainers in charge of thirty educational sessions in several temporary agencies across the province of Milano, Italy.

As part of this survey, we simultaneously focused here on precarious workers’ collective (i.e. identification with precarious workers as a social group) and politicized (i.e. identification with unionized workers) identities. In line with our theorizing, we organized these identity facets into two parallel pathways. We hypothesized that collective identification with precarious workers (Hypothesis 1a) and politicized identification with labour unions (Hypothesis 1b) both predict collective action intentions directly. Moreover, we hypothesized that collective and politicized identities both predict collective action intentions indirectly: On the one hand, collective identification should be positively related to perceived group-based injustice (Hypothesis 2a) that in turn should increase collective action intentions (Hypothesis 3a). On the other hand, politicized identification should be positively related to perceived collective efficacy (Hypothesis 2b) that in turn should increase collective action intentions (Hypothesis 3b). Thus, we expected two indirect effects through perceived group-based injustice (Hypothesis 4a), and through perceived collective efficacy (Hypothesis 4b).

Method

Participants

During class time, 379 participants (60.9% women, n = 231) filled out a pencil-and-paper questionnaire. Age ranged from 18 to 69 years (M = 31.00, S.D. = 10.73). Almost all of them were Italian mother tongue (> 90%). The instructor checked whether foreign workers had sufficient mastery of the Italian language before filling out the questionnaire. Of the total sample, 133 (35.1%) workers reported bachelor or higher levels of education; 204 (53.8%) reported high school degree; 35 (9.2%) reported secondary school degree or lower. A large majority of participants were unemployed at that moment (81.3%, n = 299); 227 (60%) respondents had long-term working experiences; 61 (16.1%) had short-term working experiences only; 72 (19.0%) had been hired solely as seasonal workers; 25 (6.6%) experienced only apprenticeship positions. Moreover, 122 (33.0%) participants reported major problems with their present or past employer. Importantly, only 11 (3.0%) participants were members of a trade union, and a solid majority of the sample was therefore lacking any organizational affiliation at the moment of the survey.

Measures

Wording, means and standard deviations for all indicators treated as exogenous factors are displayed in Table 1, whereas those treated as endogenous factors are displayed in Table 2. Higher scores always mean agreement with the statement. If not specified otherwise, all items ranged from 1 (Not at all in agreement) to 5 (Completely in agreement).

Table 1. Mean (standard deviation) and standardized loading of each observed indicator on exogenous factors.

	M (SD)	Collective Identification	Politicized Identification
I identify myself with precarious workers	3.86 (1.05)	0.80
I feel strong bonds with precarious workers	3.71 (1.04)	0.70
I define myself as precarious workers	3.66 (1.20)	0.92
I feel I belong to the category of precarious workers	3.76 (1.14)	0.91
I identify myself with trade unionists	2.84 (0.99)		0.64
I would like to be part of trade union’s activities	3.09 (1.08)		0.92
I see myself as trade unionist	2.63 (1.26)		0.82
I would be proud of being part of a trade union	3.00 (1.07)		0.83

Exact wording of each indicator is reported on the left-hand side of the table, followed by means and standard deviations, and standardized factor loadings extracted from a SEM performed using AMOS 18.

Table 2. Mean (standard deviation) and standardized loading of each observed indicator on endogenous factors.

	M (SD)	Perceived collective efficacy	Perceived group-based injustice	Collective action intentions
The way workers are treated is unfair	3.84 (0.88)		0.83
Working conditions are unacceptable	3.63 (0.96)		0.79
Many workers are exploited	4.26 (0.86)		0.66
Together workers could lay claim to better working conditions	3.86 (1.01)	0.77
Things better if workers become aware of their power	3.90 (0.99)	0.87
Favourable to minimum income	7.06 (2.36)			0.63
Willing to sign petitions for minimum income	6.94 (2.45)			0.66
Favourable to increase protections for workers who have lost their job	7.96 (1.87)			0.86
Willing to demonstrate supporting workers who have lost their job	7.11 (2.43)			0.79

Exact wording of each indicator is reported on the left-hand side of the table, followed by means and standard deviations, and standardized factor loadings extracted from a SEM performed using AMOS 18.

Collective identification with precarious workers was measured with four items adapted from Simon et colleagues (Simon et al., 1998), for example, ‘I identify myself with precarious workers’. Internal consistency was good (α = .90).

Politicized identification with unionized workers also comprised four items and was measured using the same formulation as before, for example, ‘I identify myself with unionized workers’. Internal consistency was good as well (α = .87).

Perceived group-based injustice comprised three items adapted from Van Zomeren and colleagues (Van Zomeren et al., 2008), for example, ‘The way workers are treated is unfair’. Internal consistency was good (α = .81).

Perceived collective efficacy was measured with two items inspired by Van Zomeren and colleagues (Van Zomeren et al., 2004), for example, ‘Together workers could lay claim to better working conditions’. The bivariate correlation between the two items was strong (r = .65).

Collective action intentions were measured with five items, for example, ‘Are you willing to demonstrate in order to support workers who have lost their job’; ‘Are you willing to sign a petition in favour of the minimum income?’.¹ All items were coded on an 11-point scale ranging from 0 (Not at all in agreement) to 10 (Completely in agreement). Exploratory Factor Analysis showed that all indicators saturated under the same factor except one (‘Are you favourable to strike to have less precarious working conditions?’). When the offending indicator was removed, the internal consistency improved from α = .78 to α = .85. Whereas the first four indicators concerned rather low-risk collective actions asking for more favourable working conditions, the latter indicator involved a blatant juxtaposition with an eventual employer. Thus, we excluded willingness to go on strike from the latent factor underlying collective action intentions and included it as robustness check in additional analyses.

Results

We tested our hypotheses with a Structural Equation Model (SEM) routine performed using AMOS 18. Working on the variance-covariance matrix of data (see Table S.1 in the Online Supplementary Material), SEM estimates a system of linear equations among unobserved (constructs) and observed (indicators) variables (Kline, 2015). This advanced statistical technique is particularly suitable for studying associations between latent (namely not directly measurable) variables. Before implementing the model, we evaluated the assumptions of multivariate normality and linearity. Using box plots and Mahalanobis distance, we observed no multivariate outliers. Also, missing values were negligible and random for any observed indicator (all < 5%); hence, Maximum Likelihood estimation procedure with listwise deletion was used (McKnight et al., 2007). After missing data deletion, the final sample included 343 cases.

Measurement reliability was assessed first, followed by the structural relation between latent factors. Two separate measurement models were estimated, one for the exogenous and one for the endogenous variables. Tables 1 and 2 show factor loadings of each indicator on the exogenous and the endogenous variables, respectively. Concerning the two exogenous variables (i.e. collective identification and politicized identification), the results were in line with our expectations. Collective and politicized identifications loaded on the four expected indicators, with the magnitude of loadings being generally high. Also, as compared to a single underlying dimension, the model fit improved substantially when the two latent factors were estimated separately, Δ χ²(1) = −753.33, p < .001; ΔCFI = .044; AIC = −751.33. The link between collective and politicized identifications (β = .05; SE = .03, p = .100) was weak and not statistically different from zero.

Concerning the three endogenous variables (i.e. perceived group-based injustice, perceived collective efficacy, and collective action intentions), the results were also in line with our expectations. Perceived group-based injustice and perceived collective efficacy loaded on the expected indicators. Also, the model fit improved substantially when the three latent factors were estimated separately, Δ χ²(1) = −334.14, p < .001; ΔCFI = .044; AIC = −751.33. Also, the link between perceived group-based injustice and perceived collective efficacy (β = .07; SE = .07, p = .267) was weak and not statistically different from zero.

Once the measurement models were defined, we estimated the structural model. Figure 1 provides unstandardized estimates and p statistics of each path of the model. Moreover, Table 3 shows total, direct and indirect effects of the exogenous variables on the two dependent variables. The model provided good fit to the data, χ² (107) = 223.92, p< .001; CFI = .965; RMSA = .057, 90% CI [.046; .067], p= .146; SRMR = .047. On the basis of the estimates obtained, we were able to test the two identity processes underlying collective action intentions.

Table 3. Total, direct and indirect effects of collective and politicized identifications on collective action intentions.

Total effect		Direct effect		Indirect effect
Collective identification	Politicized identification	Collective identification	Politicized identification	Collective identification	Politicized identification
0.56 (0.47; 0.65)	0.55 (0.42; 0.71)	0.35 (0.26; 0.44)	0.41 (0.29; 0.54)	0.20 (0.16; 0.27)	0.14 (0.08; 0.22)
p < .001	p < .001	p < .001	p = .001	p < .001	p = .001

Unstandardized estimates, confidence intervals, and p statistics are calculated performing Monte Carlo simulation with 4000 repetitions.

Figure 1. Dual pathway model of collective action intentions among precarious workers.

Estimates include standard errors and significance levels. The total variance explained is reported above each endogenous factor. For information about the measurement component and variables wording, see Tables 1 and 2. Model fit: χ² (107) = 223.92, p < .001; CFI = .965; RMSA = .057, 90% CI [.046; .067], p = .146; SRMR = .047. The results are consistent when gender, age, education and employment condition are introduced as controls (see Table S.4 for details).

Collective action explained by perceived group-based injustice

The first pathway shows that collective action intentions were predicted by collective identification as precarious workers and that the total effect of collective identification was mediated by perceived group-based injustice. In line with Hypothesis 1a, the total effect of collective identification on collective action intentions was significant. In line with Hypothesis 2a, collective identification was positively related to perceived group-based injustice (whereas it was not associated with perceived collective efficacy), so that the more the participants identified with precarious workers, the more they considered working conditions as unjust. In line with Hypothesis 3a, perceived group-based injustice was positively related to collective action intentions. In line with Hypothesis 4a, the indirect effect of collective identification on collective action intentions through perceived group-based injustice was significant (whereas the indirect effect through perceived collective efficacy was not significant).

Collective action intentions explained by perceived collective efficacy

The second pathway shows that collective action intentions were predicted by politicized identification with unionized workers and that the total effect of politicized identification was mediated by perceived collective efficacy. In line with Hypothesis 1b, the total effect of politicized identification on collective action intentions was significant. In line with Hypothesis 2b, politicized identification was positively related to perceived collective efficacy (whereas it was not associated with perceived group-based injustice), so that the more the participants identified with unionized workers, the more they conceived social change as possible. In line with Hypothesis 3b, perceived collective efficacy was positively related to collective action intentions. In line with Hypothesis 4b, the indirect effect of politicized identification on collective action intentions through perceived collective efficacy was significant (whereas the indirect effect through perceived group-based injustice was not significant).

Robustness checks

We performed a number of sensitivity analyses, to check the robustness of our results. First, we estimated the model only focusing on unemployed participants. Both coefficients and the fit indices were in line with our main analyses (see Table S.2 for details). Second, we estimated the model only on participants lacking organizational affiliation (i.e. not enrolled in any trade union according to their self-assessment). Alsoin this case, the findings were virtually the same (see Table S.3 for details). Finally, we estimated the model controlling for some potential covariates (worker’s gender, age, education and employment condition). None of the coefficients associated with these control variables were statistically significant, and the results were virtually the same (see Table S.4 for details).

Discussion

Our findings make two specific contributions to the study of collective action intentions of precarious workers, one empirical and the other theoretical. From an empirical point of view, our study has further contributed to disentangling two mechanisms explaining collective action intentions among a group of non-mobilized precarious workers in Italy. We used data from a unique sample collected during an educational project aimed at informing and warning precarious workers about the possible abuse of flexible and temporary contractual conditions. Our findings have pointed the presence of two psychological preconditions necessary for our participants to show mobilization propensity: a collective identification with precarious workers and a politicized identification with unionized workers. First, collective identification with precarious workers increased their collective action intentions via perceptions of group-based injustice. Although social stigma attached to exploitative or unemployed working conditions may be threatening for the social self (Furåker & Blomsterberg, 2003; Herman et al., 2006), the presence of a sense of collective identity helps precarious workers to feel they belonged to the same subordinate group, motivating them to undertake collective actions. Second, politicized identification with unionized workers further increased their collective action intentions via feelings of collective efficacy. Although old and new forms of organization and unionization are lagging behind structural transformations of labour market arrangement and power relations, these opportunities provide instrumental support and collective resources by also redirecting the struggles of precarious workers from within the workplaces toward the political realm (Grote & Wagemann, 2018).

From a theoretical point of view, our study has contributed to bridging social movement studies and socio-psychological literature on collective action. On the one hand, social movement studies have traditionally paid scarce attention to the socio-psychological preconditions of collective action, preferring to look at how material (and symbolic) resources can be secured and maximized during the mobilization process (della Porta & Diani, 2020). However, an analysis of the cognitive and motivational preconditions that trigger such processes seems to be particularly urgent, when exploring actors with apparently very scarce amounts of resources at their disposal, such as precarious workers. On the other hand, social psychological literature on collective action has traditionally overlooked the study of industrial relation actors, especially the workers, by preferring to look at ethno-cultural minority groups (Manstead, 2018). Such omissis seems particularly relevant in the current Italian context, as most workers experience various forms of social and relational fragmentation and face increasingly exploitative working conditions. Relatedly, we have shown that collective action intentions among non-mobilized individuals were grounded into social psychological processes of consciousness rising and identity politicization (Flesher Fominaya, 2010), whereby precarious workers interpreted their social environment as fundamentally unjust and changeable.

Concluding remarks: Limitations and future directions

Following theoretical insights from social psychological theorizing and current mobilization attempts among precarious workers, our study analysed the role of collective and politicized identification as parallel predictors of collective action. Yet, it is plausible that the two pathways are not independent processes: increased perceptions of group-based injustice may motivate people to take active part in political movements, thereby politicizing their social identity (McGarty et al., 2009). Possibly, socio-structural appraisals of a given social context, such as group-based injustice and collective efficacy, may loop back into identification dynamics (Van Zomeren et al., 2012). What is more, participation in successful collective actions is likely to fuel perceived collective efficacy and revive hope that social change shall be achieved (Drury & Reicher, 2005). In other words, the two pathways may not work in parallel, but rather reflect more complex causal loops that could not be captured in our cross-sectional design (Stürmer & Simon, 2009; Thomas et al., 2012). Indeed, the set-up of our analytical strategy prevents us from tackling causal mechanisms (Antonakis et al., 2010). Longitudinal studies should therefore be implemented in order to fully grasp the evolution of identification processes over time before and after that, the mobilization process has started (for a similar point, see Drury & Reicher, 2000).

Future investigations should also focus on alternative forms of collective action (Grote & Wagemann, 2018). It is plausible that more disruptive forms of politicization can also occur among precarious workers (Becker & Tausch, 2015), who have been largely excluded from labour union representation and conventional channels of political influence (Pulignano et al., 2016). Research should also move beyond classic ideas on collective identification and collective action to explore new forms of resistance and public expression of the self (Mcdonald, 2002).

Also, the effect of long-lasting periods of unemployment in terms of disengagement and increased fatalism should be better addressed (De Konig et al., 2015). Indeed, the perceived instability (or stability) of one’s precarious working condition may have implications in the way people (do not) react to material vulnerability and deprivation (Cohen-Chen & Van Zomeren, 2018; Van der Toorn et al., 2015). Finally, collective action intentions among precarious workers should be studied in countries that were less affected by the economic crisis. Whereas structural explanations of social inequalities may prevail in social contexts that were hit hardest by the financial crisis – such as the Italian labour market – internal attributions of responsibility may prevail in social contexts that were less impacted by the economic breakdown (Joffe & Staerklé, 2007; Staerklé et al., 2020).

Despite these limitations, our investigation provided convincing evidence that identification processes fuel collective action intentions among precarious workers. By doing so, we combined social movement theory and socio-psychological literature and articulated multiple identity facets into an integrated model. Importantly, we evidenced the key role played by social-psychological factors preceding formal and informal forms of collectivization among precarious workers, thereby empowering their struggle against inequality and ultimately promoting social change. More in general, our contribution seems to be particularly relevant for social movement research, whose main theoretical frameworks have only marginally incorporated cognitive and motivational determinants of collective action intention. By doing so, social movement research will be better equipped to pinpoint those social psychological drivers moving people from a condition of non-mobilization to that of mobilization. Concurrently, social psychology scholars will benefit from anchoring their analysis of basic psychological processes on existing mobilization structures and network opportunities, as exemplified in our contribution. By embedding micro-analysis of individual cognition into historical, cultural, and socio-structural factors, research on collective action will gain in ecological validity and better anticipate when social change is likely to occur.

Ethical statement

In line with the ethical procedures of the host institution (University of Lausanne, Faculty of Social and Political Sciences), correlational studies on societal issues employing adult samples were dispensed from an internal ethic committee. Nevertheless, the manuscript adheres to APA Code of Conduct ethical guidelines, as well as the Swiss National Science Foundation principles and procedures of integrity in scientific research. Moreover, this research was part of a project promoted by the Italian General Confederation of Labour (CGIL), department of New Working Identities (NIdiL). The research protocol was therefore discussed and validated by the NIdiL management office and integrated in their educational program.

Acknowledgments

The research presented in this article was possible through the collaboration with the Italian General Confederation of Labour (CGIL), department of New Working Identities (NIdiL). Special thanks goes to Jacopo Gamba, coordinator of the educational project, while this survey was conducted.

Data availability statement

Dataset, main analyses and Supplementary online material can be found on the Open Science Framework repository. Identifier: Open Science Framework

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Notes on contributors

E. Politi

E. Politi is currently a Postdoctoral Researcher funded by the Swiss National Foundation for Scientific Research (SNF) and visiting professor at Center for Social and Cultural Psychology of the Katholieke Universiteit of Leuven, Belgium. His main research interests are social psychological processes underlying political activism, social support and solidarity.

G. Piccitto

G. Piccitto is post-doctoral researcher at the Dondena Centre for Research on Social Dynamics and Public Policies of Bocconi University, Italy. His main research interests are structural characteristics and changing trends in labour markets, social stratification, inequality, and life course perspectives.

L. Cini

L. Cini is a post-doctoral researcher at the Faculty of Political and Social Sciences of the Scuola Normale Superiore of Pisa, Italy. He is member of COSMOS (Center on Social Movement Studies). His main research interests are social movements and conflicts in the current transformations of the labour market.

A. Béal

A. Béal is Mâitre de Conférence at the Ecole de Psychologues Praticiens of the Catholic University of Paris, France. He is member of the Capdroits coordination committee. His main research interests are the psychosocial determinants of take up and non-take-up of social rights and welfare benefits among people in vulnerable and precarious situations.

C. Staerklé

C. Staerklé is Associate Professor at the Social Psychology Laboratory and member of the Interdisciplinary Centre of Life Course Research of the University of Lausanne, Switzerland. His main research interests are public opinion about distributive justice principles and social psychological determinants of political ideologies.

Notes

1. We choose to ask participants about the minimum income, because a ballot initiative about minimum income was highly debated at the time of the survey as a measure to address the increased job insecurity in Italy.