Sustainable HRM and class-based inequality and stable effect on occupational careers, above and beyond other HRM, class and inequality

Purpose – This paper points out that common human resource manageement (HRM) research and practice have overlooked employee ’ s class of origin. Workers ’ class of origin can be seen as “ the elephant in the room ” in current HRM, being that it significantly affects organizational decision-making with negative social (increased class-based inequality) and organizational (inefficient allocation of human re-sources) effects. Design/methodology/approach – The paper summarizes the partial, fragmented and multi-disciplinary literature on HRM and employees ’ social class of origin. Findings – Thepapershowshowrecruiting,selection,traininganddevelopmentpracticessystematicallyreinforce class-based inequality by providing high-class employees with more resources and opportunities compared to low-class employees. Practical implications – The paper provides sustainable HR practitioners, educators and researchers with recommendations on how to address employees ’ social class of origin, improving organizational competitive advantage and reducing class-based inequality at the societal level. Originality/value – The paper focuses on a topic which, in diversity management, is an elephant in the room (i.e. workers social class of origin).


Introduction
In modern societies, individuals' position in a social hierarchy is supposed to depend only on their merit and not on ascribed characteristics (Bell, 1973). Common human resource manageement (HRM) research and practice have supported this shift from ascription to merit; diversity management, in particular, aims at avoiding bias and discrimination against employees based on visible and invisible ascribed characteristics like gender, age, race, physical ability/disability, religion, sexual orientation or gender identity (e.g. K€ ollen, 2021).
Social stratification research provides extensive empirical evidence that individuals' class of origin has a substantial and stable effect on occupational careers, above and beyond other ascribed characteristics, as well as merit-related variables such as education level (Goldthorpe, 2003). However, several scholars have recognized that common HRM practice tends to ignore an employee's class of origin (e.g. Bidwell et al., 2013;Cobb, 2016;Dundon and Rafferty, 2018;Grothe-Hammer and Kohl, 2020;Jonsen et al., 2011), even when considering diversity management practices (Jonsen et al., 2021;Ricc o and Guerci, 2014). For example, a recent analysis of DiversityInc's 2019 list of 50 companies shows that "not a single company statement referred to social class" (Ingram and Oh, 2022, p. 6). Likewise, employees' class of origin is rarely considered in common HRM research (Loignon and Woehr, 2018). As such, HRM practice contributes to the reproduction of class-based inequality in the workplace (van Dijk et al., 2020;Avent-Holt and Tomaskovic-Devey, 2019), e.g. in terms of the "class ceiling" (Ingram and Oh, 2022).
The neglect of social class is problematic for two reasons. First, equal treatment is a fundamental ethical value in modern societies (Barry, 2005;Holvino, 2010) and informs the Sustainable Development Goals (Doyle and Stiglitz, 2014;van Bergeijk and van der Hoeven, 2017). This made the issue of inequality a central construct in the academic literature as well, especially in the fields of business and society (e.g. Marens, 2018) and business ethics (e.g. Beal and Astakhova, 2017). The reproduction of class-based inequalities may thus have negative effects on the social legitimacy of organizations, the HRM profession and academic research (Farndale and Paauwe, 2018). Given that employees from low social class of origin "are most likely to be targets of objectification, prejudice, discrimination, and subject to a pattern of blaming, shaming, and dispositional attribution" (Markus, 2017, p. 217), the perceived lack of equality can undermine the overall legitimacy of the recent efforts made by the HRM community regarding diversity management and social stratification (e.g. Romani et al., 2019;Pullen et al., 2017).
Second, the reproduction of class-based inequality may have detrimental effects on organizational performance. With class-based inequality, organizations create inefficient systems of staff deployment, exchanging the contributions of competent and motivated employees from lower classes with sub-optimal contributions of employees from higher classes (Ashley and Empson, 2013). Moreover, organizations do not take full advantage of diversity-related benefits (Martin and Côt e, 2019), and are exposed to negative effects on both their employees, who may perceive a differentiated distribution of resources and opportunities as unfair or unjust (e.g. Schmidt et al., 2018), and on their external stakeholders (Kalev et al., 2006).
Against this background, the present paper has two aims. First, based on a literature review, we want to provide insights on which common HRM practices reproduce class-based inequalities, and how they do so. Second, we want to provide the sustainable HRM community (i.e. practitioners, educators and researchers) with recommendations on how to minimize this reproduction. Specifically, our recommendations are aimed at developing sustainable HRM practices which balance the social, environmental and economic performance in line with the Sustainable Development Goals (De Stefano et al., 2018;Kramar, 2014). In the past, other papers have triggered changes in the HRM community by showing that HRM practices were not neutral, but inequality-reproducing (see for example the seminal papers on gender inequality and HRM by Dickens, 1998). We hope that this paper will help to trigger such a change process in regard to class-based inequality, supporting the sustainable HRM community in overcoming the growing gap between sustainable HRM research and practice (Podgorodnichenko et al., 2021).

Class-based inequality defined
The notion of "social stratification" refers to the existence of a hierarchy of social classes (Grusky, 2019), which have differentiated access to economic (e.g. wealth or income), cultural (e.g. lifestyles or habits) and political (e.g. political support and/or political power) resources (Weber, 2019). Every society is stratified, since the amount of socially necessary work is divided into a set of occupations, to which individuals devote much of their lives. The social division of labor is also associated with a differentiated distribution of resources, since some occupations are considered more technically complex or more impactful on organizational performance and are therefore rewarded with higher resources, such as income, prestige or power.
While social stratification is a structural feature of societies, it needs to be justified and legitimized by a series of values and discourses which may vary over time and across social groups (Lenski, 1966). The view currently dominant within contemporary Western societies associates social stratification with "merit" or "meritocracy", so that the social positioni.e. social classof individuals is based on merit-related factors, and depends particularly on their education and productivity (Barry, 2005).
Research on social stratification studies the structure of the "social ladder", but also how people move across different classesi.e. social mobility. Sociological analyses, in particular, focus on the link between individuals' class of origin and their "class of destination", i.e. the social position they achieve in their lives. Class of origin is usually measured in terms of the class of the family of origin, while class of destination is measured in terms of the class to which individuals belong at a specific point in their career. The analysis of social mobility allows analyzing whether social stratification is reproduced across generations or whether it can be dissolved through merit. The closer the association between the class of origin and the class of destination, the smaller the influence of merit (Breen, 2004) Empirical research has shown that class of origin influences class of destination in two ways (Blau and Duncan, 1967). Firstly, a large part of its effect occurs indirectly via inequality of educational opportunities, i.e. through the link between individuals' class of origin and their educational attainments (Breen and Jonsson, 2005). Students from higher classes (1) have better school performances than students from lower classes, (2) are more inclined to study for a longer period and/or (3) attend more selective and prestigious educational institutions (Boudon, 1974). Hence, belonging to a higher class provides individuals with greater opportunities to develop their education, skills and experience. Secondly, this effect occurs through the direct impact of class of origin on class of destination. This direct impact occurs when two individuals with the same human capital (i.e. education, skills and experience) reach different classes of destinations because of their different classes of origin. Typically, an individual from a higher class of origin is more likely to reach a higher class of destination. When this happens, social stratification combines with class-based inequality. While the mechanisms of the indirect effect of class-based inequality can be identified within the functioning of the educational system, the direct effects are embedded in the process of allocating individuals among the various positions of the occupational structure, i.e. in the specific HRM practices that organizations employ.
Which common HRM practices reproduce class-based inequality, and how? Organizational practices can normalize inequality, through "actions that conveywhether implicitly or explicitly, whether deliberately or unknowinglythat existing societal economic inequalities are acceptable, thus adding to the risk that such inequalities go unnoticed, unquestioned, or unaddressed"; reinforce inequality, through "actions that have an effect of maintaining, strengthening and reproducing current societal economic inequalities"; or reduce inequality, through "actions that have an effect of reducing the existing inequalities in society, such as equal opportunities and diversity initiatives aimed at improving employment access to individuals from marginalized demographic groups" (Bapuji et al., 2020, p. 5). In this section, we summarize extant researchwhich is partial, fragmented and informed by several disciplines (Rivera, 2020) showing which common HRM practices reinforce classbased inequality, and how they do so.
Before we proceed, three preliminary considerations are necessary. First, we suggest that common HRM practices do not seek to reinforce class-based inequality explicitly and intentionally. These practices are often framed by companies which intend to reduce inequality and use merit-based discourses comprising words such as "meritocratic", "neutral" and "value-free" (Amis et al., 2020, p. 202). Amis et al. (2020), however, noted three "institutional myths" (i.e. meritocracy, efficiency and positive globalization) which legitimate common HRM practices, but also reinforce inequalities. Similarly, Bal and D oci (2018) present the "fantasmatic logics" of freedom, meritocracy and growth and progress as sources of legitimation of common HRM practices which otherwise reinforce class inequality. In this regard, Berry and Bell (2012) argued that, whereas inequality on the basis of gender or race is today hard to legitimize with such discourses, myths and logics, "class-based differentials are still widely viewed as legitimate" (p. 244) because dominant views tend to consider them inevitable (Acker, 2006). Second, given our focus on the reproduction of class-based inequalities, we do not analyze all common HRM practices related to the concept of social class, but only those operating on social mobility (i.e. on movements among classes) because they may inhibit upward movements of employees from lower classes of origin. Accordingly, common HRM practices which determine how material and immaterial resource allocation is dispersed among social classes (such as compensation or organization design) are not considered in this paper. Third, our focus is on single common HRM practices even if these do not work "in isolation" (Jackson et al., 2014), and it is highly likely that they have cumulative effects on inequality reproduction (Amis et al., 2020).

Internal and external recruiting and selection practices
Internal and external recruiting and selection practices typically provide employees from higher classes of origin with employment and career opportunities, and block entry or career development to employees from lower classes (Rivera, 2012). This happens for five reasons.
First, the processes used by companies for hiring and selection choices are often biased (Pitesa and Pillutla, 2019). Even sophisticated hiring processes, which increase information about candidates' and employees' human capital, are not as neutral as they appear (Friedman and Laurison, 2019;DiTomaso et al., 2007). They are instead shaped by existing patterns of disadvantage, and they combine to penalize and discriminate against employees from lower classes, even when they appear bias-free to medium to high class employees -(Kang et al., 2016). For instance, when HR managers use information about credit accounts to identify which job candidates are likely to engage in misconduct, they might equate bad credit with an untrustworthy, irresponsible disposition. In doing so, they act as an instrument of class-based discrimination since lower classes may have worse credit because of fewer financial resources (Kiviat, 2019). In addition, employers have partial information about candidates' and employees' human capital. Consequently, they tend to compensate for the lacking information with assumptions about employees' social position (Baron, 1984;Dacin et al., 2010;Jackson, 2009). Employees from higher classes of origin also usually possess noncognitive skills acquired in the family of origin and which, being well rewarded in the labor market, make those individuals overconfident (Bourdieu, 1979). This overconfidence is often mistaken for intelligence in selection processes (Belmi et al., 2020).
Second, recruiting and selection processes may be biased by principles of "cultural fit", i.e. the idea that candidates and employees with individual characteristics that fit the culture of the organization are more likely to be hired or promoted (Rivera, 2012;Rivera and Tilcsik, 2016). Although organizational cultures may work as strong coordination mechanisms, the selection and promotion of employees according to their cultural fit may contribute to segregating individuals from lower classes in organizational areas and jobs associated with limited resources (Acker, 2006;Wasserman and Frenkel, 2015). Recognizing that an individual can belong to multiple cultures based on different dimensions (for example, based on sexual orientation or nationality), current research has shown that belonging to classbased cultures radically affects companies' decision-making processes. Dale and Burrell (2008), for example, showed how traditional corporate cultures and artefacts are framed around the values of medium and high social positions which, clashing with the values into which employees from lower classes have been socialized might result in the marginalization of low class employees.
Third, recruiting and selection processes may rely on candidates' and employees' social networks, which give them access to information on vacancies and career paths, and which also influence decisions concerning hiring and promotion (Amis et al., 2020). Research has shown that people from higher classes are likely to have more and better access to contacts that can assist them during hiring and promotion processes (Lin, 1999). Ryan and Mooney (2020), for example, showed that the opportunities for employees in luxury hotels to network within their organization related to their class of origin. In such a context, individuals from lower classes might be less willing to "play politics", which makes them less promotable (Belmi and Laurin, 2016;Stern and Westphal, 2010). In addition, when the available knowledge about employees' human capital is not perfect, organizational decision-makers promote those personally closed to them, who typically belong to a higher class of origin (Ruiz Castro and Holvino, 2016).
Fourth, past performance is a key criterion used to make decisions about who should be promoted, especially in the case of internal selections. Empirical evidence shows, however, that performance management systems are often characterized by a double standard whereby employees from lower classes have to outperform others to attain the same levels of promotability (e.g. Laurison and Friedman, 2016). This process becomes stronger when performance management systems are competence-based, since this makes them value the soft competencies typically developed through informal processes by employees from higher classes of origin (Stephens et al., 2017).
Finally, career aspirations may differ among employees from different classes. People from higher classes tend to have a longer-term perspective, and are more likely to take risky career decisions that pay off later in their lives, especially to avoid social demotion (Breen and Goldthorpe, 1997;Erikson and Jonsson, 1996). Leaders and HR managers often reinforce these diverse aspirations because employees from lower classes are constantly informally and implicitly advised not to pursue careers via continuous feedback regarding their inadequacy (Collien, 2018). Employees from lower classes may therefore engage in selflimiting behaviorsfor example, by refusing a challenging job assignment or by declining an opportunity for career advancementwhich reinforces class-based inequality (Ilgen and Youtz, 1986).

Training and development practices
Training and development practices reproduce class-based inequality since employees from higher classes are usually provided with more opportunities to develop new skills which can support their further upward social mobility compared to employees with lower origins which have fewer opportunities for upward social mobility (Rivera and Tilcsik, 2016). This happens for three main reasons.
First, the allocation of employees from lower classes at low level-jobs reduces their learning opportunities (Chan and Anteby, 2016). High-level jobs are usually designed following post-Taylorist principles, i.e. they are characterized by high enlargement (job variety) and high enrichment (job autonomy) (Hauff et al., 2014). This provides employees in high-level jobs with ample opportunities for learning on the job. Differently, low-level jobs are usually designed following Taylorist principles, i.e. they are characterized by low enlargement and low enrichment. Employees in these low-level jobs have thus fewer opportunities for skills development and fewer opportunities to master the skills needed for social mobility (Duggan et al., 2020).
Second, organizations make choices about training-related investments for employees based on their current jobs, which reinforces class-based inequality (Riaz, 2015). Indeed, organizations tend to invest more in jobs that are of high strategic value (Huselid and Becker, 2011). Since hiring and selection practices make strategic jobs less accessible to employees from lower classes, disproportionate investments in jobs depending on their strategic value reinforce class-based inequality.
Third, training and development practices may reinforce social inequality through talent management, i.e. when training decisions are differentiated according to the attributed potential and performance of the employees (Iles et al., 2010). A guiding idea here is that "an unequal allocation of resources in talent management relates to positive organizational outcomes, such as a higher return on development investments, higher productivity, and retention" (Gelens et al., 2013, p. 344). Two issues emerge in this case. On the one hand, the concept of "employee potential" entails the risk of (intentionally or otherwise) incorporating biases concerning employees' classes: employees from lower classes may be systematically considered to have less potential than employees from higher classes (Thunnissen et al., 2013;Meyers et al., 2013). On the other hand, when decisions about training investments in employees are taken by considering their performance, the fact that, as said above, performance management systems systematically undervalue the performance of employees from lower classes makes them less likely to receive training and development opportunities.

Recommendations for the sustainable HRM community
Premise: what is sustainable HRM, and why it relates with (class-based) inequality HRM plays a central role in facilitating and supporting a transition toward sustainability (Lopez-Cabrales and Valle-Cabrera, 2020). This has led to the development of the idea of sustainable HRM as "the adoption of HRM strategies and practices that enable the achievement of financial, social and ecological goals, with an impact inside and outside of the organization and over a long-term time horizon while controlling for unintended side effects and negative feedback" (Ehnert et al., 2016, p. 90). Thus, sustainable HRM is seen as an extension of common HRM, being aimed at simultaneously preserving, regenerating and developing the economic, environmental, social and human resources of an organization (e.g. Kramar, 2014).
In the perspective of sustainable HRM, social inclusion and equality are considered key components of the social performance of the firm (J€ arlstr€ om et al., 2018). This claim is supported by two theories, namely resource dependence theory and stakeholder theory. Resource dependence theory states that organizations depend on resources which ultimately originate from an organization's social and physical environment (Davis and Cobb, 2010;Pfeffer and Salancik, 2009). Therefore, employers try to be perceived as equal and inclusive since it provides them with social legitimacy within their social context, which is a key requirement for companies to have access the contextual resources. Stakeholder theory assumes that a stakeholder is "any individual or group who can affect or is affected by actions, decisions, policies, practices or goals of an organization" (Freeman, 1984, p. 25). Since several internal and external stakeholders are interested in inclusion and equality, companies should take their claims seriously. Following those theories, Bapuji and Neville (2015) have recently shown that when companies increase inequality and social exclusion they lose their social legitimacy. This implies three typical risks: first, the risk of protests, which may constrain and coerce companies' decision-making processes; second, as a result of a lack of legitimacy, the risk that alternative and competing organizational forms may emerge; third, the risk that the labor income inequality generated by dispersed pay structures deteriorates institutions. Overall, sustainable HRM argues that HRM practices which reinforce inequality (and, relevantly here, also class-based inequality) deprive a company of external resources important for its success (following resource dependence theory) and/or cause negative reactions in its stakeholders (following stakeholder theory) (Stahl et al., 2020). Therefore, starting from the idea that sustainable HRM has equality as one if its central values (Diaz-Carrion et al., 2018;Richards, 2022), we develop recommendations for the members of the sustainable HRM community in regard to the objective of mitigating class-based inequality.

Recommendations for sustainable HRM practice
Firstly, we suggest that organizational decision-makers should include the social class of origin in the set of dimensions that their organizational diversity initiatives address. This inclusion would require to adopt a sustainable HRM paradigm to design three types of measures: (1) non-discrimination practices, so that HRM choices (e.g. related to hirings, promotions or training investments) are based on qualifications and abilities, and not influenced by class of origin, (2) resource practices, so that employees from lower classes are provided with additional support and opportunities (e.g. mentoring, sponsorship programs, targeted and recruitment) and (3) accountability practices, so that responsibility for and monitoring of progress in the inclusion of low class employees are increased (e.g. by including class diversity as a criterion in managers' performance evaluations) (Dobbin et al., 2015). In this effort, it is important to avoid unintended negative effects. The critical diversity management literature has shown that diversity management initiatives may have undesired effects, such as increased discrimination against diverse employees, or the disengagement of non-diverse employees (for a review of these criticisms, see Gotsis and Kortezi, 2015). These negative outcomes become more likely when the initiatives are built around paternalistic signals like "low class employees need help", but they are minimized when initiatives are built around moral signals like "class diversity is a moral duty our organization is not ignoring" (Leslie, 2019). Importantly, high-class individuals tend to infer that individuals from low classes lack competence, and that their inferior employment outcomes are deserved (Dover et al., 2020). Therefore, when designing actions targeted on low-class employees, it is important to confront high-class individuals with reliable data on the unequal opportunities of low-class individuals, so that they can start framing this class-based inequality as morally wrong and unacceptable.
Secondly, we suggest that sustainable HRM practitioners should develop training interventions targeting the unconscious class-related biases of organizational decisionmakers. This is similar to what several HRM practitioners are doing in relation to other employees' ascribed characteristics (K€ ollen, 2021). For example, in the field of sexual orientation, the aim of such interventions is not to institutionalize a de-sexualized view of the organization, but rather to make managers and employees aware of the powerful role played by sexual orientation(s) in organizational dynamics (K€ ollen, 2016). Similarly, in the field of social inequalities, training interventions should make organizational decision-makers aware of the risk that employees' social class of origin may affect their decisions, making the whole organization an engine for social inequality reproduction. A relevant concept is that of class work, i.e. "the identity-protecting and -restoring behaviors that individuals invoke during cross-class encounters" (Gray and Kish-Gephart, 2013, p. 677). Making organizational members (and the HRM professional community itself) aware of the several facets of class work could be a first step toward dismantling individual decisions and behaviors which may perpetuate class-based inequality.

Recommendations for sustainable HRM education
Management education has been under public scrutiny for many years, and concerns about its outcomes (e.g. Pfeffer and Fong, 2002) and its normative assumptions (e.g. Khurana, 2010) have been repeatedly presented, with some scholars recently calling for its dismantlement (Parker, 2018). Within this debate, Fotaki and Prasad (2015) have deplored the fact that class-based inequality is largely neglected in management education. Nonetheless, we suggest that teachers can rely on an already-existing resource to address class-based inequality, i.e. "critical management education" (Grey, 2004). A series of papers has discussed how management education can be more responsive to societal challenges, and several journals have paid growing attention to class-based inequality (e.g. Perriton and Reynolds, 2018). Few HRM scholars have drawn on the research tradition of "critical HRM" (Watson, 2010;Delbridge and Keenoy, 2010;Godard, 2020) to import these considerations into the field of HRM education. To help institutions and sustainable HRM educators to move in this direction, Bratton and Gold (2015) developed the idea of "critical HRM education". This pedagogy emphasizes the need to help learners adopt the process of "reflective critique" (p. 498). Critically reflecting on HRM could support learners in "avoiding a sole reliance on sanitized representations of corporate HR", in "looking outwards to connect personal and workplace problems to larger macro and global social structures" (p. 498), and in "exploring [. . .] the interconnections between the power relations and the socio-economic context in which worker and employer representatives interact" (p. 503).
This (self-)reflective approach to HRM suggests that sustainable HRM education should encourage students to appraise HRM outcomes critically. Rather than designing the syllabus solely around "performance"-oriented priorities, a wider set of outcomes should be included, so that learners can critically explore which actors benefit most from the implementation of HRM practices. This approach supports learners in adopting a wider perspective on HRM. It does so by showing that certain HRM practices reproduce class-based inequality, and penalize workers from low classes. Furthermore, sustainable HRM education should be designed to support learners in "a critique of prevailing theories, dominant assumptions and existing ways of meaning and propositions" (Bratton and Gold, 2015, p. 503). This approach could help students de-naturalize those taken-for-granted HRM practices which reproduce class-based inequality by questioning the discourses, such as those of meritocracy or employee freedom (Amis et al., 2020;D oci and Bal, 2018), by which those practices are legitimized (Seeck et al., 2020). These pedagogical features seem very promising, as evidenced by scholars who have reflectively critiqued their own applications (Taipale and Lindstr€ om, 2018).

Recommendations for sustainable HRM research
Overlooking class-based inequality in HRM research carries two serious risks (Côt e, 2011). First, HRM research is not building the knowledge of workers from low classes. Indeed, common HRM research is often developed on the accounts of workers occupying important positions in the organization (typically individuals from high classes), at the expense of employees who occupy low level positions or are precariously employed. For example, HRM strategies like a high-commitment HRM system focus on employees who work in strategic jobs or who are of high value for the organization (Lepak and Snell, 1999;Hauff et al., 2014). Another example is provided by HRM studies within the healthcare sector which usually focus on the perspective of nurses or doctors, rather than that of people performing low-level jobs. Furthermore, taken-for-granted assumptions in HRM often derive from studies developed in sectors, such as IT, or companies such as professional service organizations, difficult to access by employees from low classes.
The second risk concerns the building of knowledge by researchers from high classes. HRM researchers are often unaware of how their own class of origin influences their views on organizational dynamics, and their choice of research issues may unconsciously induce them to develop a science-informed by assumptions related to their own social class of origin. For example, the extensive employment by HRM researchers of individual personality traits can be interpreted as the result of dispositional explanations of behaviors and careers that characterize individuals from high classes, whereas individuals from low classes tend to explain behaviors and careers by adopting contextual explanations (Grossmann and Varnum, 2011).
Mitigation of the two above-reported risks would require sustainable HRM researchers to make a radical change in current HRM research, which could encounter several forms of resistance. First, the fact that "research on social class is bound to be provocative, controversial, and infused by ideology" (Côt e, 2011, p. 64) might be problematic for the HRM field, which is considered to be still dominated by a positivist and scientist posture which depicts research as objective and value-free (Harley, 2015). Second, the incorporation of the class of origin into current HRM research might raise ethical concerns, especially considering the managerialist attitude of common HRM research, which is often focused on predicting performance (Marchington, 2015). What if, for example, findings show that employees from low classes perform better under hard HRM, and employees from high classes perform better under soft HRM? Is it ethical in this case to recommend HRM practitioners only interested in improving organizational performance to "segregate" employees from low class of origin in positions managed through hard HRM? We recognize that exploring class-based inequality can highlight the ethical dilemmas that HRM practitioners confront with, and of which they are often unaware. Several studies (typically developed outside the HRM field) might help HRM researchers to deal with those resistances. For example, HRM researchers can rely on methodological contributions that explain how to design and perform organizational research ethically while being sensitive to class of origin (e.g. Mao and Feldman, 2019), or how to conduct reflexive research with appropriate awareness of the influence(s) of their own social class of origin (e.g. Gray, 2019).
In this regard, we highlight here four broad areas in which sustainable HRM research could be enriched by incorporating class-based inequality. The first area for future sustainable HRM research concerns the turn to employees' perspectives on HRM practices. van Beurden et al. (2021, p. 22) acknowledge that "predictors of employee perceptions of HR practice are largely unknown" and call for future research to examine them. Similarly, in a recent editorial of a special issue on HRM attributions the authors concluded "that there is scope to develop a better understanding of the determinants of HR attributions" (Sanders et al., 2021, p. 4). We argue here that, in sustainable HRM research, employees' class of origin can be considered a powerful predictor of their perceptions of HRM practices and on how they make sense of those practices. For example, since individuals from low classes have been shown to consider respect for authority a dimension of morality (e.g. Kusserow, 1999), they may tend to recognize and interpret voice practices differently from individuals from high classes. Côt e (2011, p. 57), for example, theorized that employees from a low social class of origin may "hesitate to express concerns and ideas because these expressions typically challenge the status quo and the leaders who have developed and endorsed the current procedures. Thus, if two members of the same organization are identical in abilities, personalities and demographic characteristicsbut one is from higher class and the other from lower classthe one from higher class may be more likely to voice".
The second area for future sustainable HRM research is related to the HRM architecture model (Lepak and Snell, 1999). This model suggests that, to improve its organizational performance, an organization should implement different HRM systems, and that such HRM systems should be differentiated according to the strategic value and the uniqueness of employees' human capital. These HRM systems would be then characterized by diverse levels of investment, so the organization should invest more in those employees with high strategic value and unique human capital, and less in employees with low strategic value and not-unique human capital. A recent review shows that the association between this model and organizational performance has been supported by 18 empirical studies (Luo et al., 2021). Even if some critical papers on HRM differentiation exist (Marescaux et al., 2021), none of them has incorporated class-based inequality. This represents a significant research gap since class-based inequality might be important to understand why and when individuals are likely to fall in certain quadrants. For example, it is possible that the low strategic value/notunique human capital quadrant systematically includes individuals from low social class of origin. This could reinforce class-based inequality because those individuals experience less opportunity for skill and career development and fewer chances of upward social mobility.
The third area relates to social stratification research, which has empirically shown how employees' class of origin is a powerful predictor of their attitudes and behavior (Manstead, 2018), such as organizational commitment and job satisfaction (Loignon and Woehr, 2018) or propensity for teamwork and group thinking (e.g. Kraus et al., 2010). Since common HRM research has extensively explored the effects of HRM practices on employees' attitudes and behaviors, we argue that a contextual factor which may significantly affect those associations is class of origin. Loignon and Woehr (2018), in this regard, noted that: "one of the primary avenues of future research for organizational scientists is exploring the extent to which social class functions as a boundary condition for many of the established effects in our field [. . .]. There is already a sizeable, albeit fragmented, body of literature suggesting that one's social class can moderate many of the effects that organizational scientists regularly study [. . .]. Thus, much like age and gender have been found to moderate basic phenomena within the workplace, social class too may create distinct contexts" (p. 83). A relevant (and emerging) construct which can push sustainable HRM research in this direction is "HR salience", i.e. "the prominence of an HR practice in an employees' cognitive field relative to other practices" (Garg et al., 2021, p. 12). Testing the relevance of HR salience, Garg and colleagues found that it moderates the relation between specific HRM practices and employee job pursuit intentions, job satisfaction and turnover intentions. Since having a certain class of origin may change the cognitive field of individuals (Kraus et al., 2012), sustainable HRM research could explore if social class affects whether certain HRM practices become more prominent than others, with differentiated effects on attitudes and behavior.
The last area concerns a reflective exploration, in qualitative or theoretical ways, of the reasons why social class has been neglected by common HRM scholarship (Kaufman, 2014). For example, it has been recognized that the unitarist framework, which conceives the organization as characterized by shared interests among employers and employees, has continuously evolved and persistently shaped common thinking about HRM (Greenwood and van Buren, 2017). The prevalence of this framework could be related to the strong influence of neoliberalism on common HRM research (Bal and D oci, 2018). It might be possible that the unitarist framework of the research field, together with the influence of neoliberalism on its development, have progressively moved HRM scholars away from the employment of critical analytical frameworks (such as the critical view of society as framed by social class dynamics). Future research might explore why the abandonment of critical frameworks based on social stratification has occurred, in parallel with the growing employment of the idea of corporate social responsibility, which has been recognized as a class-free construct (Fuchs, 2017).

Conclusions
This paper is a first attempt to render sustainable HRM practice, education and research sensitive to the concept of class-based inequality. Neglecting class-based inequalities does not only make organizations inequality reproducers, with negative effects on their social legitimacy; it also jeopardizes organizational performance, in particular through inefficient staff deployment and/or negative attitudes and behaviors of low-class employees due to discrimination and unequal treatment. We have shown how common HRM practices like recruitment and selection, as well as training and development, directly influence the social mobility of employees and thereby contribute to social inequality. Building on these analyses, we have provided the sustainable HRM community with some recommendations on how to avoid class-based inequality. We have also made some suggestions as to how future sustainable HRM studies can support such efforts. Note 1. Research on social mobility also considers so-called intra-generational (or career) mobility, i.e. the mobility among social positions that the same individual has occupied during her/his career. For the sake of brevity, in this article e will not deal with this type of mobility.