Class politics,collective labor rights,and worker-management conflict in comparative perspective

Abstract

This article studies how perceptions of worker-management conflict are shaped by individual-level and macro-level variables. Drawing upon data from 33 countries from the 2015 International Social Survey Programme (ISSP), it uses multilevel models to examine how individual perceptions of worker-management conflict are affected by social class, union membership status, and the country-level protection of collective labor rights. The evidence supports the hypothesis that workers and union members perceive more conflict than employers and non-union members. The results also show that, as hypothesized, perceived workplace conflict is lower in countries with stronger protection of workers’ collective rights. Finally, contrary to an initial hypothesis, cross-level interactions suggest that in countries where collective rights are more strongly protected, union members perceive more worker-management conflict than non-union members. Contributions to the literature on class and power resources as well as to the recent debate on the “neoliberal convergence” of industrial relations (IR) systems are discussed.

Keywords

social class trade unions labor rights worker-management conflict power resources multilevel models

Introduction

Conflict is an integral part of industrial relations (IR) systems (Hyman, 1989). Therefore, the study of its causes and manifestations has occupied a central place in social sciences. IR scholars have studied conflict by focusing on various aspects of strike activity (Brandl and Traxler, 2010; Jansen, 2014), while scholars interested in class inequality have done so by analyzing how individuals’ class positions and membership in trade unions shape different (and sometimes opposing) attitudes toward work, employment relations, and worker-management conflicts (Ringqvist, 2021; Wright, 1997). In the last decades, scholars have paid special attention to how cross-national differences in labor conflict can be explained by institutional and political factors. For instance, the literature on neo-corporatism argues that corporatist IR institutions ameliorate industrial conflict through arrangements that facilitate centralized bargaining, the formation of powerful and centralized unions and employer associations, and the integration of unions into policymaking (Brandl and Traxler, 2010; Siaroff, 1999). Along the same lines, scholars espousing the power resources approach argue that democratic welfare states reduce class antagonisms by institutionalizing them in parliamentary politics through the action of labor parties. This, in turn, produces political outcomes (e.g. strong labor movements) and distributive outcomes (lower inequality) that ameliorate class disparities and reduce industrial conflict (Korpi, 1985; Korpi and Shalev, 1979).

Recent investigations on perceived class conflict reaffirm this argument. Through multilevel methods, they have shown that contextual factors such as trade union density, corporatism, or welfare state development reduce the perception worker-management conflict (see, e.g. Edlund and Lindh, 2015; Ringqvist, 2021). In doing so, they have argued that the national variations identified in prior investigations are crucial for understanding how people perceive class and employment relations.

In this article I contribute to this research by focusing on a particular aspect of national IR systems, namely, the laws and regulations governing collective labor rights, that is, the rights to unionization, collective bargaining, and striking. Scholars distinguish between individual employment laws and collective labor laws in order to differentiate between provisions on individual workers’ rights (e.g. working conditions, contract termination, severance pay, leave policies, etc.) and the regulations governing the collective action of labor organizations—that is, the organization of labor unions, the rules regarding collective bargaining and strikes, standards for dialogue between employers and workers, etc. (Botero et al., 2004; Cook, 2007). Building upon this distinction, the literature has shown that the legal protection of collective rights provides workers with institutional resources to organize and mobilize, thereby contributing to social equality and justice at work (Kerrissey, 2015; Schmalz et al., 2018). Kerrissey (2015), for example, demonstrates that when collective rights are more strongly protected, income inequality is lower. In line with the power resources approach, she explains this relationship by arguing that collective labor rights are fundamental because they enable worker organizations to press for redistribution (2015: 647).

Surprisingly, the existing research has paid little attention to how collective labor rights affect individual perceptions of workplace relations. In this article, I fill this gap by studying how perceptions of worker-management conflict are shaped by individual variables, such as social class and membership in trade unions, and by the level of protection of workers’ collective rights. Drawing upon data from 33 countries collected as part of the 2015 ISSP, I use multilevel models to examine how class position and union membership status affect individual perceptions of workplace conflict. Next, using the index of Freedom of Association and Collective Bargaining (FACB) rights proposed by Kucera and Sari (2019), I analyze how these perceptions are shaped by the country-level protection of collective rights. Finally, I study whether the relationship between class and perceived conflict and between union membership and perceived conflict differs across countries with varying levels of FACB rights. Consistent with my hypotheses, the results of the multilevel models show that workers and union members perceive more conflict than employers and non-union members. Similarly, in line with my hypothesis, the results indicate that FACB rights are negatively correlated with perceived worker-management conflict. Nevertheless, in contrast to the hypothesis derived from the power resources literature, cross-level interactions suggest that in countries where FACB rights are more strongly protected, union members perceive comparatively more workplace conflict than non-union members. At the end of the article, I discuss how these findings contribute to the literatures on class and power resources and to the recent debate on the “neoliberal convergence” of IR systems (Baccaro and Howell, 2017). I argue that, even if we recognize a move toward neoliberal economics, we should not assume that national differences in IR institutions are irrelevant to analyzing how people perceive workplace relations.

Class, unions, and workplace conflict

Scholars interested in class have shown that class position is a significant individual-level determinant of people’s perceptions and attitudes. These scholars have demonstrated, for example, that those who are in underprivileged positions—working-class or low-income people—are more likely to hold critical attitudes toward inequality (e.g. to perceive society as non-meritocratic) and more likely to endorse redistributive policies than those who are located in a privileged position in the social structure (Langsæther and Evans, 2020). These scholars have also shown that nonmanagerial workers perceive more workplace conflict and are more likely to oppose capitalist institutions and support progressive tax measures than employers or high-level managers (Edlund and Lindh, 2015; Svallfors, 2006; Wright, 1997).

Class analysts such as Svallfors (2006) explain these subjective effects of social class by arguing that classes differ in their attitudes because they have diverging “moral economies,” that is, different normative dispositions toward distributive justice. Erik O. Wright (1997), on the other hand, contends that classes represent people’s position in relations of exploitation and domination. Therefore, he argues, workers are more prone to hold “anti-capitalist” interests than employers (the exploiting class) and the members of the middle-class who are located in “contradictory class locations” (that is, positions where they simultaneously occupy the roles of exploiter and exploited). Wright emphasizes that distinguishing between a “capitalist” (property owner) class and non-property owner classes is crucial to understanding the basic contours of class conflict because such distinction gives rise to the class boundary between those who hire (and exploit) workers and those who sell their labor power (and are exploited). In this sense, Wright’s analysis departs from other approaches to class such as that proposed by Erikson and Goldthorpe (1992). Based on a conceptualization of employment relations rooted in the distinction between “service” and “labor” contracts, Erikson and Goldthorpe propose a class scheme that do not depend on the identification of a “capitalist” (employer) class, but rather on the identification of a privileged “service class” made of higher-grade professionals, managers, and large proprietors.

As for the relationship between union membership and perceived conflict, scholars have made similar arguments. Compared to their non-union counterparts, unionized workers not only perceive higher levels of worker-management conflict, but are also more likely to file grievances and less likely to be loyal to their employers (Lewin and Boroff, 1996; Lincoln and Boothe, 1993; Ringqvist, 2021). Similarly, union members tend to be less satisfied with their jobs and less committed to the organization’s goals and values than non-unionized workers (Hipp and Givan, 2015; Laroche, 2016; Lincoln and Boothe, 1993). Furthermore, union members are more “class conscious” and more politicized than their non-union counterparts (Dixon et al., 2004; Wright, 1997). Therefore, they are more willing to mobilize against employers and participate in politics than non-union workers (Dixon et al., 2004; Kerrissey and Schofer, 2018; Rosenfeld, 2014).

This evidence contradicts the argument that unions facilitate the communication between workers and managements and therefore help ameliorate labor conflict (see, e.g. Freeman and Medoff (1984)). Lewin (2005: 213) explains this apparent contradiction by arguing that unions are much more than a channel to communicate workers’ concerns to management. Unions, according to Lewin, represent the collective voice of their affiliates and, in the course of that process, build collective power to “win more” at the expenses of their employers—for example, to win more protective work rules or more money to settle grievances. Consistent with this, that is, other scholars explain the positive relationship between union membership and perceived conflict by contending that unions play a key role in the construction of workplace solidarity (Dixon et al., 2004; Wright, 1997).

In current neoliberal times, different categories of workers—for example, the informal self-employed and platform workers—work in precarious labor conditions in which they tend to be isolated from one another and in which their basic rights (most notably, the right to organize) are not recognized. In these conditions, unions’ capacity to build solidarity is limited at best (Vandaele, 2018; Tassinari and Maccarrone, 2020). Moreover, for those who have the right to organize, it is still unclear whether unions cause the “politicization effects” mentioned above or reinforce the attitudes of workers who are already more politicized prior to joining a union. However, in a recent longitudinal study Hadziabdic and Baccaro (2020) demonstrate that union membership produces “maturation effects”—that is, it strengthens workers’ politicization—which, once they are assimilated by workers, are unlikely to disappear after leaving the union.

Based on the evidence outlined above, the following hypotheses can be proposed:

H1.1

The perception of worker-management conflict is higher among people located in the working class (or in an adjacent class position) than among individuals located in the managerial or self-employed classes (employers or petite bourgeoisie).

H1.2

The perception of worker-management conflict is higher among union members than among non-union members.

Class politics, labor rights, and conflict

Comparative research has explained cross-national differences in the protection of workers’ rights by analyzing how labor policies are an outcome of class politics, that is, an outcome of the political struggle between class-based actors such as employers, unions, and parties (Cook, 2007; Korpi, 1985). Scholars espousing the power resource approach argue that when the linkages between unions and center-left parties are strong, unions have more opportunities to influence the policymaking process and protective labor legislation is more likely to pass (Cook, 2007; Korpi, 2006; Raess et al., 2018).

Without denying that labor policy is an outcome of class politics, scholarly research has also argued that labor policy is a mechanism in its own right. Once established, labor policy has “feedback effects” (Pierson, 1993) as it produces resources and incentives that shape the alternatives available for actors to pursue their interests (Cook, 2007). From this point of view, labor policy is a crucial “institutional” power resource: when workers’ collective labor rights are protected by law, unions’ bargaining power is greater, and workers have more resources to organize and mobilize (Schmalz et al., 2018). In this sense, collective labor rights represent “workers legal and practical ability to organize in unions, bargain collectively, and engage in protest” (Kerrissey, 2015: 627).

Emphasizing this latter aspect of labor policy, scholarly research has suggested that when collective labor rights are more strongly protected, industrial conflict is lower. Focusing largely on IRs in Europe, this research shows that the extension of collective labor rights was closely related to the consolidation of neo-corporatist institutions during the second half of the 20th century. Neo-corporatist institutions were designed to ameliorate class conflict by reducing workers’ need for open confrontation with employers (Korpi and Shalev, 1979). To do so, corporatist institutions facilitate the political inclusion of workers through centralized bargaining and state-sponsored pacts between employer associations and powerful and hierarchically organized unions (Brandl and Traxler, 2010; Kenworthy, 2003; Korpi and Shalev, 1979; Siaroff, 1999). This in turn enables unions to influence the policymaking process, and provides them with the power to exchange wage moderation for the promise of full employment, labor protections, redistributive policies, and welfare spending (Hicks and Kenworthy, 1998; Kerrissey, 2015; Korpi, 1985).

Corporatist institutions have similar effects on strike activity. In corporatist economies, centralized bargaining reduces strike activity by lessening the uncertainty during the bargaining process and by rising union leaders’ awareness of the economic effects of “excessive” wage increases—most notably, inflation and unemployment (Brandl and Traxler, 2010; Calmfors and Driffill, 1988). Additionally, corporatist IR institutions decrease individual unions’ propensity to strike through regulations that limit inter-union competition for members and bargaining outcomes (Akkerman, 2008; Jansen, 2014). In short, in corporatist systems where labor rights are more strongly protected, IRs are less conflictual and strike prone than in decentralized systems, that is, liberal market economies, where collective labor rights are less developed (Akkerman, 2008; Brandl and Traxler, 2010; Jansen, 2014; Kerrissey and Schofer, 2018).

To my knowledge, no research has analyzed how cross-national differences in the scope and extension of collective labor rights affect individual perceptions of workplace conflict. Nevertheless, consistent with the power resources argument, recent investigations show that in countries where collective labor rights are more strongly protected—most notably, countries with corporatist IR institutions, the aggregate perceptions of worker-management conflict are lower (Edlund and Lindh, 2015; Ringqvist, 2021). Ringqvist (2021), for example, demonstrates that the existence of corporatist regulations that protect unions’ right to be involved in policy formulation is negatively correlated with perceptions of worker-management conflict. Similarly, others have suggested that when IR institutions promote centralized bargaining, and by extension when unions’ bargaining power is stronger, workers are more likely to have higher levels of job autonomy and security, and are more satisfied with the material aspects of their work (e.g. income and advancement opportunities) (Edlund and Grönlund, 2010; Esser and Olsen, 2012; Hipp and Givan, 2015). In these contexts, working conditions are better for workers, and these may perceive IRs as less conflictual (Ringqvist, 2021).

Over the last decade, some scholars have argued that countries are converging around a common trajectory of liberalization. This has involved the “disorganization” of economies, the expansion of employers’ discretion in IRs and labor markets, and the decline of union membership rates (Baccaro and Howell, 2017; Streeck, 2009). Without denying the effects of neoliberalism on IRs, other scholars have insisted that national institutional differences continue to be important for explaining workplace relations. They have shown, for example, that in corporatist or coordinated market economies, unions have more power to defend protective labor policies and to resist employers’ attempts to increase work intensity (Adăscăliţei et al., 2021; Rathgeb, 2018).

Following these findings, it can then be posited that:

When collective rights are more strongly protected, the perceptions of worker-management conflict are lower.

The moderating effects of collective labor rights

No research has studied whether the country-level protection of collective rights moderates the relationship between class (or union membership) and perceptions of worker-management conflict. However, recent studies suggest that there are significant interaction effects between political-institutional contextual factors and individual-level variables such as class and union membership. Edlund and Lindh (2015) show that in more developed welfare states, class attitudes toward redistribution are more polarized than in countries where the welfare state is less developed. They explain this significant interaction effect by contending that welfare states politicize inequality and redistribution, thereby enhancing class differences in attitudes. While not using multilevel modeling, class analysts such as Wright (1997: 433–440) and Svallfors (2006: 74) reach similar conclusions. They show that in more developed welfare states, workers are more “class conscious” because left parties and unions are more powerful (Wright, 1997). This not only politicizes market inequality, but also makes workers more likely to interpret their experiences as “members of a collective” with certain interests rather than as “atomized market actors” (Svallfors, 2006: 74).

Nevertheless, recent investigations focusing specifically on worker-management conflict suggest that the relationship between these micro- and macro-level variables may be the opposite (see, e.g. Edlund and Lindh, 2015; Ringqvist, 2021). In a recent study of perceptions of worker-management conflict, Ringqvist (2021: 141) found a negative (but insignificant) interaction effect between trade union density and individual union membership. This suggests that stronger union power may reduce the positive effect of union membership on people’s perceptions of labor conflict. On the other hand, in their study of union membership and political participation, Kerrissey and Schofer (2018) found significant negative interaction effects between corporatism and individual union membership. This implies that although union members are more likely to participate in disruptive political actions (e.g. strikes) than non-union members, the effect of union membership is smaller in corporatist countries. According to Kerrissey and Schofer, this finding supports the argument that corporatist institutions favor the political inclusion of trade unions and, in doing so, reduce the need for confrontation between workers and employers (see also Cebolla-Boado and Ortiz, 2014).

Based on this evidence, two hypotheses can be formulated:

H3.1

The association between class location and perceptions of worker-management conflict is negatively moderated by collective labor rights. In countries where collective rights are more strongly protected, the positive effect of class location (e.g., of being a worker) is smaller than in countries where these rights are weakly protected.

H3.2

The association between union membership and perceptions of worker-management is negatively moderated by collective labor rights. In countries where collective rights are more strongly protected, the positive effect of union membership is smaller than where these rights are weakly protected.

Data and methods

For this article, I analyzed data from the ISSP 2015, module “Work orientations.” I restricted my analysis to 33 countries with data on the contextual variable of interest (collective labor rights). My sample of countries included liberal market economies (e.g. the UK, and U.S.), coordinated market economies (e.g. Austria, Germany, Iceland, and Sweden), other continental or southern European nations (e.g. France and Spain), and emerging economies from Latin American (Chile, Mexico), Central or Eastern Europe (e.g. Croatia and Poland), and Asia (Taiwan) (see Table A1, Appendix).

After listwise deletion, my sample included 19,538 individuals nested in 33 countries. Although the number of level-2 units (countries) is not exceptionally large, it is large enough to provide reliable estimates from multilevel linear models (Bryan and Jenkins, 2015: 19).

Method: multilevel analysis

The hypotheses that lead this study were tested using multilevel linear models (MLMs). MLMs are appropriate when we are interested in analyzing the relationship between a dependent variable (in this case, perceived worker-management conflict) and variables that measure attributes of both individuals (level-1 units) and countries (level-2 units). In fact, unlike fixed-effect regressions, MLMs allow for examining the relationship between variables controlling for appropriate country-level predictors (Bryan and Jenkins, 2015).

In this article I did so by estimating several random-intercept linear models with individual- and country-levels variables. I also fit models with random slopes for the two individual-level variables of interests, class and union membership. These models allowed me to analyze whether the relationship between the level-1 independent variables and the dependent variable varied across countries. The results of these analyses suggested that the models with random slopes for class were too complex to be supported by the data (i.e. to estimate the random coefficients precisely). They also suggested that the random slopes for class did not improve the models’ fit. Compared to the random-intercept models, these models had higher AIC and BIC values. By contrast, I found that the models with a random slope for union membership fit the data better (i.e. had lower AIC and BIC values) than the random intercept models (chi-square tests of deviance, p-value < .001). For this reason, in addition to presenting the MLMs with random intercepts, Table 2 only reports the models with a random slope for union membership (the models with random slopes for class are available by request).

Finally, I estimated models with cross-level interactions. This was key to fulfilling the goals of this article as two of the hypotheses described above (H3.1 and H3.2) refer to the existence of interaction effects between the class location/union membership and collective labor rights.

All the models were estimated using the lme4 package in R (Bates et al., 2015).

Dependent variable

The dependent variable is “Perception of worker-management conflict,” and was measured through the following five-point Likert-type question: In general, how would you describe relations at your workplace between management and employees? (1. Very good to 5. Very bad). This question does not directly measure worker-management conflict, nor does it distinguish between “individual” and “collective” conflicts—for example, between workers’ individual misbehavior and workers’ participation in collective actions. However, this variable seems appropriate to analyze perceived workplace conflict because it focuses on the respondents’ own workplace experiences (more specifically, on their perceptions about the quality of employee-management relations).

To facilitate interpretation, I followed prior research and transformed this scale into a 0 to 100 score (see, e.g. Edlund and Lindh, 2015: 317). Higher scores denote higher perceptions of conflict.

In robustness checks, I treated the dependent variable as a five-level ordinal scale and as a three-level ordinal scale (1. Very good or quite good relations, 2. Neither good nor bad, 3. Very bad or quite bad relations). Using the meologit command in Stata, I estimated multilevel ordered logistic models. These models, available upon request, produced the same results as the MLMs reported here.

The descriptive statistics for this variable and the independent variables are shown in Table 1.

Table 1.

Descriptive statistics.

	N	Mean	St. Dev.	Min	Max
Perceptions of worker-management conflict	19,538	27.45	22.75	0	100
1. Employers	19,538	0.01	0.09	0	1
2. Small employers	19,538	0.03	0.17	0	1
3. Petite bourgeoisie	19,538	0.08	0.28	0	1
4. Expert managers	19,538	0.04	0.19	0	1
5. Nonmanagerial experts	19,538	0.06	0.23	0	1
6. Skilled supervisors	19,538	0.12	0.33	0	1
7. Unskilled supervisors	19,538	0.07	0.25	0	1
8. Skilled workers	19,538	0.22	0.41	0	1
9. Unskilled workers	19,538	0.37	0.48	0	1
Union membership	19,538	0.23	0.42	0	1
Female	19,538	0.49	0.50	0	1
Age	19,538	42.85	12.72	15	95
Private sector	19,538	0.72	0.45	0	1
FACB rights	19,538	7.49	2.07	2.32	10.00
Gini	19,538	34.51	7.60	25.40	63.00

Individual-level independent variables: social class and union membership

Social class was measured through a modified version of the scheme proposed by Wright (1997). This scheme distinguishes class locations on the basis of three criteria: (1) the private ownership of means of production; (2) the level of skills; and (3) the possession of “organizational assets” (authority) within the production process.

Within the first criterion—ownership of means of production, I distinguished between: (1). Employers (owners of firms with 10 or more employees); (2). Small employers (with between 1 and nine employees); and (3). Petty bourgeoisie (self-employed).

Based on the second criterion (skill level) and third criterion (authority), I created six categories of salaried class position. In relation to skill levels, I distinguished wage earners according to their condition as experts, skilled laborers, or unskilled laborers. To do so, I used the ISCO-08 codes aggregated to two digits. Then, I corrected the “experts” category for education levels so experts were those employed in occupations belonging to ISCO groups from 10 to 26 who had also completed some post-secondary education. Finally, the “authority” dimension allowed me to differentiate between managers/supervisors and wage earners without supervisory authority. The ISSP dataset does not offer distinctions between the three authority levels proposed by Wright (1997: 74–90). However, it at least distinguishes between those who worked in a supervisory capacity and those who did not. Taking all these criteria into account, the six categories created to differentiate salaried class locations were (4) expert managers, (5) nonmanagerial experts, (6) skilled supervisors, (7) unskilled supervisors, (8) skilled workers, and (9) unskilled workers. Taken together with the three business owner categories described above, my class schema contains nine class positions total.

As noted above, this schema includes a “Petty bourgeoise” class formed by people who neither hire employees nor work for a direct employer. This implies that these respondents’ perceptions of employee-management relations are not based on their actual workplace experiences, but presumably on other subjective processes (e.g. desirability, past experiences, etc.). In spite of this, I decided to keep this class category because it allowed me to analyze whether the self-employed hold more “conservative” views than salaried workers, as suggested in recent investigations (Langsæther and Evans, 2020).

Union membership was measured with a dummy variable (1 = union member; 0 = non-union member).

Following recent research, my analyses also included the following sociodemographic controls: age (in years), gender (0 = male; 1 = female), and sector of employment (0 = public; 1 = private).

Country-level independent variable: collective labor rights

I used the index of Freedom of Association and Collective Bargaining (FACB) rights, developed by Kucera and Sari (2019). The FACB rights index is widely used in empirical research because it provides a comparable measurement of country-level trade union rights (Kerrissey, 2015; Vadlamannati, 2015). Based on the textual analysis of nine sources, including national labor legislations, ILO reports and other sources, Kucera and Sari’s index is constructed through the coding of 108 labor rights violations under five categories: (1) Fundamental civil liberties, (2) right of workers to establish and join organizations, (3) other union activities, (4) right to collective bargaining, and (5) right to strike. The coding process is done by experts who are asked to evaluate the 108 rights violations for each country. During the process, distinct criteria for violations of labor rights in law (de jure) and in practice (de facto) are applied. Additionally, the evaluation of each labor rights violation is weighted according to its severity (Kucera and Sari, 2019: 426–427). This results in a normalized 0 to 10 scale, with higher scores representing worse de jure or de facto labor rights. To facilitate interpretation, in this article I reversed the scale so that higher scores denote more protective labor laws and practices.

Following recent literature (Edlund and Lindh, 2015; Ringqvist, 2021), my models controlled for income inequality (Gini index, data obtained from the World Inequality Dataset [WID]). In alternative models, I also controlled for GDP per capita to assess whether the impact of FACB rights remains after controlling for differences in economic development/affluence. The results of these models did not differ from the results reported here.

To simplify comparison across models, the country-level variables were transformed into Z-scores before being entered in the models.

Empirical analysis

Table 2 presents the results of several MLMs. Following the standard practice, I first fit an empty model (M0), which suggested that approximately 5% of the variance in the dependent variable is explained by differences between countries (ICC = 0.044).

Table 2.

Multilevel linear models predicting perceptions of worker-management conflict in 33 countries (2015): main and interaction effects (standard errors in parentheses).

	M0: empty model	M1: random intercepts	M2: random slopes	M3: random intercepts	M4: random intercepts	M5: random slopes
Class location (ref.: Employers)
2. Small employers		1.486 (1.966)	1.244 (1.964)	1.188 (2.032)	1.412 (1.966)	1.226 (1.964)
3. Petite bourgeoisie		4.061* (1.827)	3.795* (1.825)	2.965 (1.889)	3.936* (1.828)	3.755* (1.826)
4. Expert managers		9.078*** (1.933)	8.811*** (1.931)	9.250*** (2.014)	8.953*** (1.933)	8.767*** (1.931)
5. Nonmanagerial experts		10.503*** (1.885)	10.266*** (1.885)	10.651*** (1.953)	10.357*** (1.886)	10.217*** (1.885)
6. Skilled supervisors		9.180*** (1.804)	8.943*** (1.803)	9.253*** (1.851)	9.064*** (1.805)	8.903*** (1.803)
7. Unskilled supervisors		10.319*** (1.848)	10.053*** (1.847)	10.399*** (1.893)	10.201*** (1.849)	10.015*** (1.847)
8. Skilled workers		10.998*** (1.784)	10.724*** (1.783)	11.117*** (1.831)	10.864*** (1.784)	10.680*** (1.783)
9. Unskilled workers		12.118*** (1.768)	11.767*** (1.767)	12.293*** (1.815)	11.963*** (1.768)	11.717*** (1.767)
Union membership
Yes		4.923*** (0.435)	4.686*** (0.808)	4.766*** (0.435)	4.722*** (0.441)	4.618*** (0.793)
Controls
Age		−0.026* (0.013)	−0.027* (0.013)	−0.027* (0.013)	−0.026* (0.013)	−0.027* (0.013)
Female		−0.498 (0.323)	−0.547† (0.323)	−0.575† (0.323)	−0.511 (0.323)	−0.549† (0.323)
Private sector		−0.603 (0.385)	−0.577 (0.386)	−0.581 (0.385)	−0.614 (0.385)	−0.581 (0.386)
Contextual variables
FACB rights (z-score)		−1.534† (0.915)	−1.739* (0.874)	−2.271 (1.797)	−1.752† (0.929)	−1.680† (0.879)
Gini (z-score)		−2.217* (0.912)	−2.154* (0.871)	−2.169* (0.925)	−2.171* (0.923)	−2.142* (0.875)
Cross-level interactions
Small employers* FACB rights				−0.797 (1.747)
Petite bourgeoisie* FACB rights				−1.646 (1.613)
Expert managers* FACB rights				0.774 (1.973)
Nonmanagerial experts* FACB rights				0.973 (1.775)
Skilled supervisors* FACB rights				1.106 (1.640)
Unskilled supervisors* FACB rights				1.739 (1.687)
Skilled workers* FACB rights				1.565 (1.594)
Unskilled workers* FACB rights				1.286 (1.574)
Union member* FACB rights					1.200** (0.466)	1.350† (0.817)
Constant	27.538*** (0.851)	18.096*** (2.052)	18.469*** (2.035)	17.984*** (2.095)	18.187*** (2.055)	18.496*** (2.037)
Random effects
Intercept	22.972	19.485	17.540	20.045	19.954	17.717
Residual	494.262	482.381	480.661	481.559	482.225	480.645
Union membership slope			14.111			13.257
Intercept – slope covariance			2.287			2.257
Observations	19,538	19,538	19,538	19,538	19,538	19,538
Number of countries	33	33	33	33	33	33
AIC	176,754.563	176,283.521	176,252.691	176,248.567	176,278.585	176,250.556
BIC	176,778.203	176,417.483	176,402.413	176,445.570	176,420.427	176,408.158
Log likelihood	−88,374.281	−88,124.761	−88,107.346	−88,099.283	−88,121.293	−88,105.278

^***p < .001; ^**p < .01; ^*p < .05; ^†p < .1.

The coefficients from models 1 (M1) through 5 (M5) show that class position and union membership are significant predictors of the level of perceived worker-management conflict. In the two cases, the effect is statistically significant (p-value < .001) and in the direction hypothesized: workers and union members score higher than employers or non-union members on the 0-100 scale of perceived conflict. This supports hypotheses 1.1 and 1.2.

To illustrate graphically the effect of class and union membership, I used the ggeffects package in R (Lüdecke, 2021) and plotted the marginal effects (predicted values) of these two variables. To do so, I used the coefficients from M1, holding the other variables constant at their mean (quantitative variables) or reference categories (categorical variables). The results are presented in Figure 1. Panel A of Figure 1 shows that, on average, skilled and unskilled workers score around 10 points higher than employers on the 0–100 scale of perceived conflict. Likewise, Panel B indicates that, on average, unionized workers score around five points higher on the scale of perceived conflict than non-union members. It must be noted that the differences between classes and between union and non-union members are not large. Furthermore, in the case of union membership the difference is even smaller. However, the coefficient for unionization is statistically significant. Thus, the alternative hypothesis should be accepted despite the overlap in the confidence intervals (see Schenker and Gentleman, 2001).

Figure 1.

Marginal effects of class and union membership.

Regarding collective labor rights, the coefficients indicate that there is a negative relationship between them and perceived conflict. In other words, in countries where collective labor rights are more strongly protected, the aggregate perceptions of worker-management conflict are lower. This relationship is statistically significant in most models (p-value < .1), which supports hypothesis 2.

As stated in the methodological section, the comparison between the random intercept and random slope models suggested that, unlike the models with random slopes for class, the models with a random slope for union membership fit the data better than the models without it. This suggests that, unlike the relationship between class and perceived conflict, the relationship between union membership and perceived conflict varies across countries. Consistent with this finding, the coefficients of M3 show that the cross-level interaction between FACB rights and class is insignificant. This implies that hypothesis 3.1 cannot be accepted. By contrast, M4 and M5 indicate that the interaction between collective rights and union membership status is statistically significant (p-value < .1). However, in contrast to hypothesis 3.2, this interaction is positive, meaning that the positive effect of being a union member on perceived conflict is larger in countries where labor rights are more protected. Thus, hypothesis 3.2 should be rejected.

Using the ggeffects package in R (Lüdecke, 2021), I plotted the marginal effects for the interaction between FACB rights and union membership presented in M5. The results are shown in Figure 2. The figure shows the predicted values in the 0–100 scale of perceived worker-management conflict for non-union and union members, in countries with “strong” and “weak” FACB rights (“strong” and “weak” FACB rights are represented by a score of 2 standard deviations above/below the mean). Figure 2 indicates that in countries with weak FACB rights, the difference between union and non-union members is smaller than in countries where FACB rights are more robust. In countries with weak FACB rights, the average scores of non-union and union members are, respectively, 20.7 and 22.6, while in countries with strong FACB rights they are 13.9 and 21.3. Like in Panel B of Figure 1, in Figure 2 the confidence intervals overlap. However, the fact that the coefficient for the interaction is statistically significant means that the alternative hypothesis should be accepted.

Figure 2.

Marginal effects of the interaction between collective labor rights and union membership.

Discussion

The results presented above are consistent with Wright’s (1997) claim about the “antagonistic” nature of class and employment relations. According to Wright (1997: 20), employers do not simply own the means of production and hire workers, they also dominate them to extract and appropriate the fruit of their effort (see also Thompson and Smith, 2010). This results in the formation of antagonistic interests between classes—that is, mutually opposed interests regarding how classes secure their economic welfare and enhance their economic power (Wright, 1997), which represents a crucial mechanism that links people’s objective class position with their subjective perceptions and attitudes. These antagonistic interests explain, in effect, why workers who are hired and dominated within production relations perceive more worker-management conflict than employers who hire and dominate them.

This finding has theoretical and empirical implications. Decades ago, analysts suggested that distinguishing a “capitalist” (property owner) class was no longer relevant to identifying the contours of the “privileged” classes (see, e.g. Erikson and Goldthorpe, 1992). As stated above, Wright departs from this assumption, noting that identifying a “capitalist” class continues to be important for analyzing exploitation, and thereby class antagonisms. In line with Wright’s argument, the results presented here suggest that distinguishing between property owner and non-owner classes is key to understanding variations in individuals’ perceptions of worker-management conflict. The similarities in the level of conflict perceived by large and small employers indicates, in effect, that rather than the size of the firm owned, what really shapes how property owners perceive labor relations is the very fact of being an employer, that is, the fact of owning a firm, hiring workers, and dominating them within production.

The importance of the property ownership/non-ownership distinction is also reflected in the comparatively low scores of the petite bourgeoisie. Recent research shows that the autonomy of the self-employed reinforces their individualistic attitudes and enhances their pro-business views (Langsæther and Evans, 2020). The evidence presented here is consistent with this argument, showing that the solo self-employed have perceptions of conflict that are more similar to those of the employers than to those of the salaried classes. That said, it is likely that this explanation holds only for the “traditional” segments of the self-employed class. In the current neoliberal context, platform and gig workers are legally defined as solo self-employed, although their work is highly controlled through platform algorithms. Moreover, despite the individualistic nature of their jobs and their precarious work situations, some of these workers have successfully organized, improving their labor conditions through collective action (Kougiannou and Mendonça, 2021; Tassinari and Maccarrone, 2020). Further research should clarify to what extent these two segments of the self-employed class differ in their perceptions of worker-management conflict.

As for union membership, the results concur with previous research showing that unions make workers more aware of the conflicting interests between them and employers (Ringqvist, 2021). The Marxist approach to IRs argues that unions arise as a collective response by workers to their subordinate position in the relations of production (Hyman 1989). Therefore, at the micro-level they are expected to increase the oppositional attitudes among union members (see also Wright, 1997). In line with this argument, my results indicate that unions are class-based organizations that build working-class solidarity and collective power, thereby strengthening the oppositional consciousness of workers. In doing so, my results provide new evidence contradicting the theory that unions facilitate the articulation of workplace grievances and thereby ameliorate workers’ perceptions of conflict (see also Lewin, 2005). As stated in the literature review section, the available evidence is inconclusive regarding whether unions reinforce or cause these types of subjective effects (see, e.g. Hadziabdic and Baccaro, 2020). The data and methods used here do not allow me to address this question, nor is the goal of this article to do so. Nevertheless, the results of the MLMs show that unions continue to be correlated with important subjective outcomes despite the fact that trade union membership rates have declined in most countries (Gumbrell-McCormick and Hyman, 2018).

Similarly, the significant and negative relationship between FACB rights and perceived conflict confirms previous research arguing that collective rights are an important “institutional” power resource for workers (Schmalz et al., 2018). Prior research suggests that collective labor rights enable unions to improve working conditions for workers, thereby creating less conflictual IRs (Esser and Olsen, 2012; Ringqvist, 2021). My findings add weight to this argument. Building upon the power resources approach, my findings suggest that collective labor rights can reduce perceived worker-management conflict by increasing workers’ power to press for redistribution. This can occur either directly through market mechanisms (e.g. by increasing unions’ bargaining power vis-à-vis employers) or indirectly through political processes (e.g. by augmenting unions’ power to push for state redistributive policies) (Kerrissey, 2015; Korpi and Shalev, 1979). In either case, the implication is that when unions are stronger, the aggregate perceptions worker-management conflict are lower.

This finding, along with the significant effect of union membership discussed above, help to understand why unions are sometimes described as “ambivalent” organizations. According to Hyman (1989: 78–79), such ambivalence is reflected in the fact that unions can increase class consciousness but, at the same time, serve as vehicles to institutionalize class compromise. The power resources approach explains this mixed function of unions arguing that unions’ growth and development is closely related to the institutionalization of class conflict through corporatist arrangements and welfare state expansion (Edlund and Lindh, 2015; Korpi and Shalev, 1979). My evidence demonstrates that this ambivalent nature of unions can be explained by distinguishing between individual and contextual levels of analyses. At the micro-level unions increase industrial conflict by reinforcing the oppositional attitudes of workers, whereas at the macro-level unions ameliorate it by reducing the aggregate perceptions of worker-management conflict.

The significant effects of FACB rights also add nuance to “convergence” thesis, which states that neoliberal globalization has essentially blurred the divide between national IR systems. The scholars espousing this view have rightly shown how in several countries neoliberalism has been associated with a general decline of union membership rates and strike activity, as well as with the decentralization of collective bargaining (Baccaro and Howell, 2017; Streeck, 2009). Without denying these effects of neoliberal policies, the results of the MLMs suggest that national differences in labor institutions—particularly in extent to which they protect collective rights—continue to be crucial for explaining variations in workplace conflict.

The results of the cross-level interactions provide more evidence for this latter argument. However, the interaction effect between FACB rights and union membership is contrary to what can be deduced from the literature on power resources and neo-corporatism. Instead, this effect is more consistent with recent research suggesting that industrial actions are more frequent when labor regulations protect the bargaining position of workers (see, e.g. Belloc, 2021). This interaction effect also concurs with the argument developed by class analysts such as Wright (1997) and Svallfors (2006). As stated above, these authors argue that union power “politicizes” class relations and foster partisan attitudes among workers, which reinforces their perceptions of class conflict. My evidence demonstrates, however, that FACB rights increase the level of perceived conflict only among unionized workers relative to their non-unionized counterparts, not among all workers relative to, say, employers.

Conclusion

Drawing upon data from 33 countries, in this article I analyzed how perceptions of worker-management conflict are shaped by individual-level and macro-level variables. I presented evidence that supports the argument that class and membership in unions are crucial for understanding workplace conflict in contemporary society (see also Edlund and Lindh, 2015; Ringqvist, 2021). Additionally, I showed that workplace conflict is also affected by national differences in labor institutions, as expressed in the extent to which they protect collective rights. Furthermore, I demonstrated that cross-national differences in the protection of workers’ rights moderate the relationship between union membership and perceived worker-management conflict.

These findings have important theoretical implications. They imply that we should adopt a more nuanced approach toward the “neoliberal convergence” debate. Even recognizing that countries have moved toward more “disorganized” political economies in which unions have a weakened position vis-à-vis employers (Baccaro and Howell, 2017; Streeck, 2009), we should not conclude that national differences in labor rights have faded away, let alone that these differences are not important for analyzing workplace conflict. Consistent with this, that is, the evidence presented here showed that there is a negative and statistically significant relationship between collective labor rights and the aggregate perceptions of worker-management conflict.

That said, an important caveat should be kept in mind. Contrary to my hypothesis, the results of the MLMs indicated that the interaction between FACB and union membership was positive. In other words, they suggested that the positive effect of being a union member on perceived conflict is larger in countries where labor rights are more protected. This implies that the function of the institutional arrangements established in the golden era of welfare capitalism to reduce industrial conflict may have been changing in the past few decades. Prior research has argued that these arrangements provided unions with institutional power resources that enabled them to push for redistribution without the need to resort to disruptive actions (e.g. strikes). Despite this, the results of the MLMs suggested that when these arrangements protect workers’ collective rights, unions members perceive more worker-management conflict than non-union members. As stated above, a possible explanation can lie in the ambivalent nature of unions (Hyman, 1989): at the macro-level unions can help reduce conflict even though, at the micro-level, they can increase the oppositional views of union members.

In this article, I contributed to the understanding of this “ambivalence” by clearly distinguishing between the micro- and macro-level effects of unions. Further studies with longitudinal data should investigate whether these effects have changed after the move toward neoliberalism. Over the past three decades, many countries have witnessed the emergence of a powerful business class eager to liberalize national IR systems (Baccaro and Howell, 2017). Neoliberal ideology has permeated center-left parties and union-party ties have weakened (Gumbrell-McCormick and Hyman, 2018). My findings suggest that these changes may have eroded corporatist institutions’ capacity to reduce industrial conflict. Longitudinal studies can help to clarify whether this is actually the case.

Finally, this article demonstrated that class position has significant effects on how people perceive labor relations. The increased inequality resulting from neoliberal policies goes far to explain why the main hypotheses derived from the work of Marxist class analysts such as Wright (1997) remain valid despite the weakness of organized labor. Future studies should interrogate how, in these contexts, the class differences in the attitudes toward workplace relations found in this article can translate into collective action (e.g. into the emergence of revitalized labor movements). My findings indicate that the degree of protection of workers’ collective rights is a central aspect of IRs. Therefore, implementing actions to strengthen institutional power resources should be an essential part of any strategy for trade union renewal.

Footnotes

Acknowledgments

The author would like to thank the anonymous reviewers for their helpful and constructive comments. He would also like to thank Valentina Andrade for her assistance with the database and to the participants of the VIII COES International Conference (November 15 - 17, 2021, Santiago, Chile) where a preliminary version of this article was presented. The research project that gave rise to this article was funded by FONDECYT Project 11190229 (“Institutional and political determinants of conflict between employers and workers: The cases of Argentina and Chile in comparative perspective,” PI: Pablo Pérez Ahumada) and by the Centre for Social Conflict and Cohesion Studies (COES; ANID/FONDAP/15130009).

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Fondo Nacional de Desarrollo Cientfico y Tecnolgico (grant no. 11190229) and Fondo de Financiamiento de Centros de Investigacin en reas Prioritarias (grant no. COES;ANID/FONDAP/15130009).

ORCID iD

Pablo Pérez Ahumada

Author Biography

Pablo Pérez Ahumada is Assistant Professor of Sociology at the University of Chile and Adjunct Researcher at the Centre for Social Conflict and Cohesion Studies (COES, Chile) He is also Director of the Observatory of Labor Strikes (OHL – COES/UAH). His research focuses on social class, industrial relations, labor movements, and politics in Latin America.

Appendix

Table A1.

Perceptions of worker-management conflict and FACB rights by country.

	Perceived worker-management conflict	FACB rights
AU-Australia	31.3	5.81
AT-Austria	20.5	9.62
BE-Belgium	33.5	8.72
CL-Chile	23.7	7.07
CN-China	31.4	3.94
TW-Taiwan	23.2	3.94
HR-Croatia	27.6	8.15
CZ-Czech Republic	30.5	7.66
EE-Estonia	28.8	8.91
FI-Finland	29.7	10.0
FR-France	37.9	8.56
DE-Germany	22.5	8.68
HU-Hungary	28.0	6.90
IS-Iceland	24.3	9.65
IN-India	24.6	3.39
IL-Israel	21.8	8.56
JP-Japan	38.3	8.52
LV-Latvia	23.5	8.94
LT-Lithuania	30.3	8.44
MX-Mexico	21.7	5.61
NZ-New Zealand	23.5	7.41
NO-Norway	30.3	9.83
PH-Philippines	25.3	2.32
PL-Poland	34.3	5.72
RU-Russia	31.3	5.33
SK-Slovak republic	26.7	10.0
SI-Slovenia	33.9	9.64
ZA-South Africa	21.5	7.21
ES-Spain	24.6	8.60
SE-Sweden	31.4	9.27
CH-Switzerland	20.7	8.73
GB-Great Britain/United Kingdom	26.2	8.35
US-United States	26.2	5.10

References

Adăscăliţei

Heyes

Mendonça

(2021) The intensification of work in Europe: a multilevel analysis. British Journal of Industrial Relations 60(2): 324–347.

Akkerman

(2008) Union competition and strikes: the need for analysis at the sector level. International Labour Review 61(4): 445–459.

Baccaro

Howell

(2017) Trajectories of Neoliberal Transformation: European Industrial Relations Since the 1970s. Cambridge: Cambridge University Press.

Bates

Mächler

Bolker

, et al. (2015) Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67(11): 1–48.

Belloc

(2021) Industrial actions and firing regimes: how deregulating worker “exit” reshapes worker “voice”. Structural Change and Economic Dynamics 56: 251–264.

Botero

Djankov

La Porta

, et al. (2004) The regulation of labor. The Quarterly Journal of Economics 119(4): 1339–1382.

Brandl

Traxler

(2010) Labour conflicts: a cross-national analysis of economic and institutional determinants, 1971-2002. European Sociological Review 26(5): 519–540.

Bryan

Jenkins

(2015) Multilevel modelling of country effects: a cautionary tale. European Sociological Review 32(1): 3–22.

Calmfors

Driffill

(1988) Bargaining structure, corporatism and macroeconomic performance. Economic Policy 3(6): 14–61.

10.

Cebolla-Boado

Ortiz

(2014) Extra-representational types of political participation and models of trade unionism: a cross-country comparison. Socio-Economic Review 12(4): 747–778.

11.

Cook

(2007) The Politics of Labor Reform in Latin America: Between Flexibility and Rights. PA: University ParkThe Pennsylvania State University Press.

12.

Dixon

Roscigno

Hodson

(2004) Unions, Solidarity, and Striking. Social Forces 83(1): 3–33.

13.

Edlund

Grönlund

(2010) Class and work autonomy in 21 countries: a question of production regime or power resources? Acta Sociologica 53(3): 213–228.

14.

Edlund

Lindh

(2015) The democratic class struggle revisited: the welfare state, social cohesion and political conflict. Acta Sociologica 58(4): 311–328.

15.

Erikson

Goldthorpe

(1992) The Constant Flux: A Study of Class Mobility in Industrial Societies. Oxford: Clarendon Press.

16.

Esser

Olsen

(2012) Perceived job quality: autonomy and job security within a multi-level framework. European Sociological Review 28(4): 443–454.

17.

Freeman

Medoff

(1984) What Do Unions Do? New York, NY: Basic Books.

18.

Gumbrell-McCormick

Hyman

(2018) Trade Unions in Western Europe: Hard Times, Hard Choices. Oxford: Oxford University Press.

19.

Hadziabdic

Baccaro

(2020) A switch or a process? Disentangling the effects of union membership on political attitudes in Switzerland and the UK. Industrial Relations: A Journal of Economy and Society 59(3): 466–499.

20.

Hicks

Kenworthy

(1998) Cooperation and political economic performance in affluent democratic capitalism. American Journal of Sociology The University of Chicago Press, 103(6). 1631–1672.

21.

Hipp

Givan

(2015) What do unions do? A cross-national reexamination of the relationship between unionization and job satisfaction. Social Forces 94(1): 349–377.

22.

Hyman

(1989) Strikes. 4th ed. London: Springer.

23.

Jansen

(2014) Effects of union organization on strike incidence in EU companies. International Labour Review 67(1): 60–85.

24.

Kougiannou

Mendonça

(2021) Breaking the Managerial silencing of worker voice in platform capitalism: the rise of a food courier network. British Journal of Management 32(3): 744–759.

25.

Kenworthy

(2003) Quantitative indicators of corporatism. International Journal of Sociology 33(3): 10–44.

26.

Kerrissey

(2015) Collective labor rights and income inequality. American Sociological Review 80(3): 626–653.

27.

Kerrissey

Schofer

(2018) Labor unions and political participation in comparative perspective. Social Forces 97(1): 427–464.

28.

Korpi

(1985) Power resources approach vs. Action and conflict: on causal and intentional explanations in the study of power. Sociological Theory 3(2): 31–45.

29.

Korpi

(2006) Power resources and employer-centered approaches in explanations of welfare states and varieties of capitalism: protagonists, consenters, and antagonists. World Politics 58(2): 167–206.

30.

Korpi

Shalev

(1979) Strikes, industrial relations and class conflict in capitalist societies. The British Journal of Sociology 30(2): 164.

31.

Kucera

Sari

(2019) New labour rights indicators: method and trends for 2000–15. International Labour Review 158(3): 419–446.

32.

Langsæther

Evans

(2020) More than self-interest: why different classes have different attitudes to income inequality. The British Journal of Sociology 71(4): 594–607.

33.

Laroche

(2016) A meta-analysis of the union–job satisfaction relationship. British Journal of Industrial Relations 54(4): 709–741.

34.

Lewin

(2005) Unionism and employment conflict resolution: rethinking collective voice and its consequences. Journal of Labor Research 26(2): 209–239.

35.

Lewin

Boroff

(1996) The role of loyalty in exit and voice: a conceptual and empirical analysis. Advances in Industrial and Labor Relations 7: 69–96. JAI PRESS INC.

36.

Lincoln

Boothe

(1993) Unions and work attitudes in the United States and Japan. Industrial Relations: A Journal of Economy and Society 32(2): 159–187.

37.

Lüdecke

(2021) Introduction to Adjusted Predictions and Marginal Effects in R. Available at: https://strengejacke.github.io/ggeffects/articles/introduction_marginal_effects.html (accessed 13 September 2021).

38.

Pierson

(1993) When effect becomes cause: policy feedback and political change. World Politics 45(4): 595–628.

39.

Raess

Dür

Sari

(2018) Protecting labor rights in preferential trade agreements: the role of trade unions, left governments, and skilled labor. The Review of International Organizations 13(2): 143–162.

40.

Rathgeb

(2018) Strong Governments, Precarious Workers. Ithaca, NY: Cornell University Press.

41.

Ringqvist

(2021) How do union membership, union density and institutionalization affect perceptions of conflict between management and workers? European Journal of Industrial Relations 27(2): 131–148.

42.

Rosenfeld

(2014) What Unions No Longer Do. Cambridge, MA: Harvard University Press.

43.

Schenker

Gentleman

(2001) On judging the significance of differences by examining the overlap between confidence intervals. The American Statistician 55(3): 182–186.

44.

Schmalz

Ludwig

Webster

(2018) The power resources approach: developments and challenges. Global Labour Journal 9(2): 113–134.

45.

Siaroff

(1999) Corporatism in 24 industrial democracies: meaning and measurement. European Journal of Political Research 36(2): 175–205.

46.

Streeck

(2009) Re-Forming Capitalism: Institutional Change in the German Political Economy. Oxford: Oxford University Press.

47.

Svallfors

(2006) The Moral Economy of Class: Class and Attitudes in Comparative Perspective. Stanford: Stanford University Press.

48.

Tassinari

Maccarrone

(2020) Riders on the storm: workplace solidarity among gig economy couriers in Italy and the UK. Work, Employment and Society 34(1): 35–54.

49.

Thompson

Smith

(2010) Debating labour process theory and the sociology of work. In: Thompson

Smith

(eds), Working Life: Renewing Labour Process Analysis. Basingstoke: Palgrave McMillan, pp. 11–28.

50.

Vadlamannati

(2015) Rewards of (Dis) integration: economic, social, and political globalization and freedom of association and collective bargaining rights of workers in developing countries. ILR Review 68(1): 3–27.

51.

Vandaele

(2018) Will trade unions survive in the platform economy? Emerging patterns of platform workers’ collective voice and representation in Europe. Working Paper 2018.05. Brussels: ETUI.

52.

Wright

(1997) Class Counts: Comparative Studies in Class Analysis. Cambridge: Cambridge University Press.