Understanding employee responses to COVID-19: a behavioral corporate social responsibility perspective

Purpose – The purpose of this study is to critically synthesize and integrate conceptual and empirical research on the behavioral perspective on corporate social responsibility (CSR) and explain why it is useful and necessary, especially in the wake of the COVID-19 pandemic. The authors explain why CSR can result in © Herman Aguinis, Isabel Villamor and Kelly P. Gabriel. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode The authors would like to thank Management Research Editor José Ernesto Amor os and two anonymous reviewers for their highly constructive feedback that allowed them to improve the manuscript. A previous version of this article was presented as a keynote address at the annual meetings of the Behavioral Science and Policy Association, May 2020. Employee responses to COVID-19

In the wake of the COVID-19 pandemic, Amazon decided that embracing corporate social responsibility (CSR) was the right thing to do and therefore would focus on the triple bottom line of financial, social and environmental performancenot just profits. Specifically, Amazon drastically expanded online grocery delivery to provide an essential service for individuals affected by the pandemic (Amazon, 2020). However, with the increased need for fulfillment centers to run smoothly and at full capacity, the company began facing a lack of organizational commitment from workers as they protested unsafe conditions such as not having adequate protective equipment and working among many that have tested positiveor even diedfrom COVID-19 (Weise and Conger, 2020). A distinguished Engineer and VP at Amazon Web Services even resigned because of the company's handling of worker concerns amid the pandemic (Dolsak and Prakash, 2020). From the perspective of employees, this supposedly well-designed socially responsible initiative may have been well-intended, but it asked too much of employees and had detrimental effects.
The real story described in this opening paragraph is a good example of organizations implementing CSR policies to try to do the right thing during the COVID-19 pandemic. CSR has been defined as "any 'responsible' activity that allows a firm to achieve a sustainable competitive advantage, regardless of motive" (McWilliams andSiegel, 2011, p. 1480). Thus, the actions that Amazon took during the COVID-19 pandemic were part of their CSR initiatives. But the question is, given the evidence of an overall positive relationship between CSR and firm performance (Cantrell et al., 2015;Jayachandran et al., 2013;Margolis et al., 2009;Orlitzky et al., 2003), what went wrong? Why did a well-intended CSR initiative lead to unintended negative consequences (Pierce and Aguinis, 2015)? What can organizations do to produce win-win results as they focus on the three "Ps" simultaneously (i.e. profits, people and planet; Aguinis, 2011)? What do we know based on the vast literature on CSR that can help us answer these questions, and what are the questions that future research should address?
The goal of our article is to offer insights regarding these issues by embracing a fairly novel behavioral perspective on CSR. Most CSR research conducted to date has adopted a firm or institutional perspective (Aguinis and Glavas, 2012). In other words, CSR research has focused mostly on why and when firms choose to engage in CSR by, for example, examining a firm's mission and the values of its top management teams (i.e. firm approach), as well as laws, industry standards and stakeholder pressures (i.e. institutional approach). However, as we illustrated with the Amazon story, although CSR policies are created by organizations, they are implemented by individual employees. Moreover, the way employees perceive and react to CSR policies and Employee responses to  actions are key determinants of how they are implemented and their success or lack thereof (Aguinis and Glavas, 2019;Closon et al., 2015;Duarte et al., 2014;Duarte et al., 2010;El Akremi et al., 2018;Rupp and Mallory, 2015). A behavioral perspective on CSR goes beyond the more conventional firm and institutional perspectives and facilitates, through an employee-centric conceptual framework (Rupp and Mallory, 2015), a deeper understanding of when and why CSR can lead to different types of positive as well as unintended negative outcomes. We explain why a behavioral perspective is useful and necessary, especially in the wake of the COVID-19 pandemic, by relying on theories and empirical research from the fields of organizational behavior (OB), human resource management (HRM) and others that focus on the individual and team levels of analysis (e.g. organizational psychology). Our article is organized as follows. First, we explain the usefulness of a behavioral perspective on CSR, particularly in the backdrop of the COVID-19 pandemic. Second, we describe two types of CSRembedded and peripheral (Aguinis and Glavas, 2013b)and their impact on employees. Third, we discuss reasons why CSR, which is intended to lead to positive outcomes, can actually result in both positive as well as negative outcomes. Fourth, based on our critical review and synthesis of the literature, we offer seven suggestions for future research on behavioral CSR and COVID-19 and describe how Latin America is a particularly suitable context for pursuing two of them. Finally, based on the existing evidence, we describe how a behavioral perspective on CSR leads to useful recommendations for practice and policymaking in the wake of the pandemic.
A behavioral corporate social responsibility perspective As we mentioned previously, research on CSR has traditionally focused on the firm and institutional levels of analysis and on assessing the relation between CSR and firm-level outcomes such as financial performance (Aguinis and Glavas, 2012;Duarte et al., 2010). In the past decade, however, conceptual and empirical research has expanded with a rise of a behavioral perspective on CSR (Closon et al., 2015;Duarte et al., 2014;Gond et al., 2017;Ng et al., 2019), what has often been labeled "micro-CSR" (Rupp and Mallory, 2015). This fairly new conceptual framework is person-centric and employee-centric in particular. Specifically, it focuses on how individuals experience and engage with CSR as they seek and find meaningfulness through work. Consequently, it also focuses on when and why employees experience CSR differentlyresulting in differing positive outcomes for themselves, their organizations and external stakeholders (Aguinis and Glavas, 2019).
This shift toward a behavioral perspective on CSR started less than a decade ago. Aguinis and Glavas' (2012) multilevel review of the state of CSR research showed that, at that time, only 4% of the CSR-related articles in 17 highly regarded management and organization journals addressed the topic at the individual level of analysis. Also, in their review of CSR, Aguinis and Glavas (2012) ascertained that the CSR literature was largely macro-focused and neglected CSR's multilevel nature. Further, little CSR research had explored underlying mechanisms (i.e. mediating effects) through which CSR might result in differing outcomes for different individuals, organizations and stakeholders (Aguinis and Glavas, 2012).
Since Aguinis and Glavas's (2012) review -exemplified by Rupp and Mallory's (2015) review, only three years later -CSR research adopting a behavioral perspective has flourished. This research stream resulted in new insights into not only the positive and negative effects of CSR on employees but also in how to measure CSR. For example, El Akremi et al. (2018) developed and validated a multidimensional stakeholder-based measure of employees' CSR perceptionsthe corporate stakeholder responsibility scale. By measuring CSR perceptions, we can now better understand why, how and when CSR perceptions are likely to lead to outcomes such as affective commitment, organizational identification and job satisfaction.
Research on behavioral CSR has demonstrated empirical support for the positive effect of CSR on individual performance, organizational attitudes and organizational attraction (Rupp and Mallory, 2015 for a review). In terms of performance, CSR is positively related to organizational citizenship behaviors (Hansen et al., 2011;Rupp et al., 2013b), employee engagement (Caligiuri et al., 2013;Rupp et al., 2013a), effort (Jones, 2010), involvement and improved employee relations (Glavas and Piderit, 2009).
In terms of organizational attitudes, CSR is positively related to employee identification with the organization (Kim et al., 2010) and job satisfaction (De Roeck et al., 2014;Dhanesh, 2014). Hofman and Newman (2014) documented the importance of employee perceptions of CSR by finding that employees' perceptions of CSR practices toward internal (compared to external) stakeholders were positively related to organizational commitment.
Finally, in terms of organizational attraction, employees find organizations that engage in CSR to be more attractive (Backhaus et al., 2002). Positive organizational attraction outcomes of CSR include job pursuit intentions (Gully et al., 2013;Rupp et al., 2013b), professional development (Caligiuri et al., 2013) and retention (Jones, 2010). For example, Caligiuri et al. (2013) ascertained that employees' volunteer assignments were most valuable for engagement when employees were able to find meaning through their work, believe they contributed meaningfully, use their professional skills and develop skills useful for their typical work role.
However, there is also a "dark side of CSR." A growing literature on the behavioral perspective on CSR has examined the negative consequences that sometimes occur when choosing or implementing CSR initiatives. For example, Ormiston and Wong (2013) found that firms that engaged in CSR were subsequently more likely to engage in corporate social irresponsibility (CSiR), and CEOs who presented themselves as moral individuals were more likely to be associated with this positive relation between CSR and subsequent CSiR.
In the context of the COVID-19 pandemic, research focused on how firms can do good and do well is more critical than ever (Bapuji et al., 2020). However, conducting this type of useful and impactful research can be difficult when trying to make sense of the vast and diverse body of CSR work. A first step toward understanding the effects of CSR using a behavioral perspective is to distinguish the types of CSR policies that firms may implement in response to COVID-19 and their respective impact on employees. As a preview, the key findings on behavioral CSR, the differences between embedded versus peripheral CSR, their effects on employees and why CSR often results in mixed effects on employees in the context of COVID-19 are listed below: Behavioral CSR: key findings (1) Definition and conceptualization Focuses on how individuals experience and engage with CSR as they seek and find meaningfulness through work. Focuses on when and why employees experience CSR differentlyresulting in differing positive outcomes for themselves, their organizations and external stakeholders.
Research on behavioral CSR has demonstrated empirical support for the positive effect of CSR on individual performance, organizational attitudes and organizational attraction. There is also a "dark side of CSR:" a growing literature on the behavioral perspective of CSR has examined the negative consequences that sometimes occur when choosing or implementing CSR initiatives.
(2) Embedded versus peripheral CSR Employee responses to  Embedded CSR relies on an organization's core competencies and integrates CSR within a firm's strategy, routines and operations. Peripheral CSR focuses on activities that are not integrated into an organization's strategy, routines and operations.
(3) Consequences of embedded versus peripheral CSR Embedded CSR is more likely to result in positive outcomes for employees when the CSR strategy is implemented taking them into consideration. When CSR is peripheral, employees do not engage with CSR every day, and they recognize that the CSR is not fully integrated within the organization. Employees involved in peripheral CSR might view it as symbolic or selfishly motivated by their firms. (4) Reasons for differential consequences of embedded and peripheral CSR: the role of sense making Employees are actively thinking about and making sense of their organization and their roles as employees, individuals and members of society. This psychological process of sense making is a fruitful way to understand the underlying reasons why CSR results in these positive and negative effects. Sense making can be an underlying and unifying mechanism through which individuals engage with CSR to search for and find meaningfulness through work. CSR is a particularly salient conduit for individuals to seek and find meaningfulness through work because CSR affects many stakeholders and outcomes and thus expands the notion of work to go beyond a task or job.
Embedded versus peripheral corporate social responsibility As Rupp and Mallory (2015) pointed out, CSR research faces a specificity dilemma. Several definitions and taxonomies of CSR have been proposed over time, most of them at the macro-level (Brammer et al., 2007;Matten and Moon, 2008;Waddock and Graves, 1997). Aguinis and Glavas (2013b) relied on the sustainability literature (Laszlo and Zhexembayeva, 2011) and suggested that CSR can be either embedded within or peripheral to the core functioning of a firm. This distinction between embedded CSR and peripheral CSR has become even more relevant in the recent pandemic. As firms quickly react to the new reality caused by COVID-19, they select the types of CSR initiatives to implement, often without considering the effects that these decisions may have on both their employees and their firms. Embedded CSR relies on an organization's core competencies and integrates CSR within a firm's strategy, routines and operations and, therefore, affects all employees. In contrast, peripheral CSR focuses on activities that are not integrated into an organization's strategy, routines and operations (Aguinis and Glavas, 2013b). Although peripheral CSR can fulfill noble purposes and create shared value for organizations and society, it is not integrated into the daily practices and routines of the firm. Further, although organizations rarely fully embed CSR, most do implement some types of CSR, making the embedded-peripheral continuum useful (Aguinis and Glavas, 2013b).
This embedded-peripheral conceptualization of CSR is built into the psychological foundations of CSR (i.e. the underlying mechanisms and processes). Examples of these psychological foundations are the fulfillment of internal and external stakeholder needs (Aguinis andGlavas, 2013a, 2013b), values congruence (Edwards and Cable, 2009), meaningfulness in and at work (Rosso et al., 2010), perceptions of organizational justice (Rupp, 2011) and the ability for individuals to present more of their whole selves at work (Kahn, 1990). It is through these psychological foundations that embedded and peripheral CSR have differing effects on employees.
An example of a firm practicing embedded CSR is Intel (Aguinis and Glavas, 2013b), which integrates CSR into its strategy and daily practices. The "enrichment of lives of every person on earth" is part of Intel's core vision, and in its 2019-2020 CSR report Intel highlighted its 2030 strategic goals of "responsible, inclusive, and sustainable" practices enabled through their "technology, innovation, expertise, and the passion of [their] employees" (Intel, 2020, p. 7). Intel's strategy and vision are operationalized through all facets of the organization (Aguinis and Glavas, 2013b). For example, performance management systems incorporate CSR and influence employee compensation decisions. Also, "design for the environment" is a collaborative approach between departments to ensure that CSR is built into every aspect of engineering, manufacturing and R&D (Aguinis and Glavas, 2013b). As a result of this embeddedness, employees across the organization become engaged and can find more meaningfulness in work, as there is a deeper purpose beyond short-term financial profit. In the context of COVID-19, Intel engaged in embedded CSR in committing $60m dollars to accelerate access to the technology needed to combat the pandemic and at the same time committing $100m to additional benefits, aid and compensation for its employees on the front lines (Intel, 2020). The company was able to leverage its pandemic leadership team created almost 20 years ago to ensure that its response was in line with its strategy, values and operations (Intel, 2020). With the resources and support from the company, employees have been engaging in creative CSR tactics, such as a group of Arizona employees who have been 3D printing face shields for first responders and vulnerable communities (We Are Intel, 2020). Now, let us return to the case of Amazon we described at the beginning of our article. Expanding its online grocery delivery seemed to be consistent with Amazon's strategy given that it "strives to offer [its] customers the lowest possible prices, the best available selection, and the utmost convenience" (Amazon, 2020). However, the initiative was not integrated effectively into the company's routines and operations related to employee health and safety. Thus, this initiative was more peripheral than embedded.
Finally, we make an important clarification regarding the Intel and Amazon examples. These illustrations are not meant to portray these companies as being perfectly or not at all socially responsible. Rather, these cases are meant to illustrate the approach used toward CSR and thus the usefulness of the embedded versus peripheral CSR distinction to understand when and why CSR leads to positive outcomes.
Consequences of embedded versus peripheral corporate social responsibility As Aguinis and Glavas (2019) explained, employees are affected positively when they perceive that CSR is embedded (Donia and Tetrault Sirsly, 2016). In contrast, when CSR is peripheral, employees do not engage with CSR every day, and they recognize that the CSR is not fully integrated within the organization. The reason for the stronger positive outcomes of embedded versus peripheral CSR is based on attribution theory. Specifically, employees involved in peripheral CSR might view it as symbolic or selfishly motivated by their firms (Aguinis and Glavas, 2019). Thus, it is not surprising that employees can have negative reactions to CSR when they are skeptical of the organization's claims (e.g. greenwashing) or when they view CSR as motivated by pure self-interest in profit (McShane and Cunningham, 2012). Jones et al. (2016) delved into the psychological processes behind these negative reactions and asserted that the way CSR is implemented sends signals to prospective employees; employees can draw inferences about a company's working conditions, thus influencing their attraction to the job. As a specific illustration of this general finding, during the COVID-19 pandemic, Jeff Bezos (Amazon's CEO) donated $100m to American food banks Employee responses to COVID-19 (Liao, 2020), which will certainly have a positive impact on the food security of people in need. However, he faced backlash from employees and the media who viewed this donation as an intentional diversion from scrutiny over the lack of protection for warehouse and logistics workers during the pandemic (The Economist, 2020).
Differential consequences of embedded and peripheral corporate social responsibility: the role of sense making What becomes apparent in our discussion of the positive and negative outcomes of embedded and peripheral CSR is that employees are actively thinking about and making sense of their organization and their roles as employees, individuals and members of society. This psychological process of sense making (Weick, 1995) is a fruitful way to understand the underlying reasons why CSR results in these positive or negative effects. Employees are proactive and intentional agents who make sense and give meaning to their worldsespecially in new experiences such as the COVID-19 pandemic where people, organizations and society are encountering heightened levels of ambiguity and complexity. Aguinis and Glavas (2019) explored sense making as an underlying and unifying mechanism through which individuals engage with CSR to search for and find meaningfulness through work. CSR is a particularly salient conduit for individuals to seek and find meaningfulness through work because CSR affects many stakeholders and outcomes and thus expands the notion of work to go beyond a task or job. Consequently, this mechanism can explain when and why employees experience CSR in a particular manner and the resulting outcomes of those experiences (Aguinis and Glavas, 2019).
Through CSR and sense making, individuals can find meaning at work and in work (Aguinis and Glavas, 2013b). Through peripheral CSR, employees might find meaning at work, as they can see their employer engaging in CSR, but the employee's own job does not directly contribute to CSR. For example, seeing that Amazon has donated $5m in Amazon devices globally to those in need (Amazon, 2020) may make an Amazon worker more proud of their employer even though their own job does not directly contribute to CSR (i.e. finding meaning at work). With embedded CSR, employees are more likely to find meaning at work and in work, as the CSR suffuses strategy and everyday routines and operations.

Implications and opportunities for future research
The organizational and societal context we are experiencing owing to COVID-19 brings a new need for research to improve our understanding of behavioral CSR. A behavioral perspective on CSR involves an expansion of conceptual and methodological frameworks. It also involves integrating varied fields (e.g. OB and HRM) and their theories that are impacted by CSR and sustainability issues (Aguinis and Glavas, 2013a). Because COVID-19 has impacted every country, business, industry and sector, there are numerous knowledge gaps and opportunities. In this section, we outline seven broad opportunities for future research. For the last two, we describe how conducting research specifically in Latin America would be particularly advantageous. As a preview, these future directions are listed below: Behavioral CSR and COVID-19: future research directions (1) HRM practices Is it appropriate for job analysis, recruitment, selection, performance management, reward and training processes to consider CSR? If so, how should they be designed? If these classic HRM activities have already been designed to consider CSR, how should they change in response to COVID-19?
What specific HRM theories can be used to improve our understanding of the relation between CSR and COVID-19 (and potential future pandemics and natural disasters)? (2) Bridging the micro-macro divide How are leadership characteristics related to firm-level CSR reactions to COVID-19? How do leaders' individual differences interact with firm-level factors (e.g. industry) in the selection of CSR initiatives? (3) Measuring CSR's impact on stakeholders outside the firm What organizational systems and processes prevent the successful implementation of CSR initiatives related to COVID-19? What are sound ways to capture the dimensionality of embedded and peripheral CSR? What are the specific systems and processes that might lead to better outcomes for extra organizational stakeholders? (4) Effects of CSR programs in reaction to  How do employees experience peripheral CSR (e.g. volunteering) while receiving compensation from their firm? (5) Indicators of employee resistance to CSR What are the boundary conditions in how CSR structure (embedded versus peripheral) impacts employee outcomes? What individual differences (e.g. work-family orientation, political orientation) that create variation in the way employees make sense of, give meaning to, and react to CSR are heightened by the pandemic? (6) Hybrid public-private collaborations What factors make these hybrid collaborations successful in responding to crises such as COVID-19? What are the consequences for both private and public enterprises that collaborate in these types of COVID-19 initiatives? (7) Social inclusiveness, inequity and vulnerable populations What is the impact that CSR initiatives from firms in Latin America can have in mitigating inequality, and how can this translate into more meaningfulness in and at work?
First, research can explore how HRM practices are impacted by and, in turn, impact CSR responses to COVID-19 (Morgeson et al., 2013). There is currently an opportunity to continue integrating conceptual streams and expand the research on HRM practices and CSR (Stahl et al., 2020). Despite the explosive research in the past years linking CSR and microresearch (Gond et al., 2020;Gond et al., 2017;Rupp and Mallory, 2015), most of this research has addressed theories and domains in OB compared to HRM. Thus, there remains a need to understand how HRM practices can include CSR. Future research could address whether it is appropriate for job analysis, recruitment, selection, performance management, reward and training processes to consider CSR and, if so, how these classic HRM activities should be designed. Or, if these processes have already been designed to consider CSR, research can consider how they should change in response to COVID-19. Moreover, the specific HRM theories that can be used to improve our understanding of the relation between CSR and COVID-19 (and potential future pandemics and natural disasters) are also still unknown.

Employee responses to COVID-19
Second, bridging the much-lamented micro-macro divide Aguinis and Glavas, 2012) and building on the multilevel nature of CSR (Andersson et al., 2013;Ren et al., 2017), future research can explore how leadership characteristics are related to firmlevel CSR reactions to COVID-19. For example, multilevel research can explore how leaders' individual differences might interact with firm-level factors, such as industry, in the selection of CSR initiatives (Aguinis and Molina-Azorín, 2015).
Third, related to the need to measure the impact of CSR on stakeholders outside the organization (e.g. community, environment and society; Barnett et al., 2020), understanding the organizational systems and processes that prevent the successful implementation of CSR initiatives related to COVID-19 could be a first step toward addressing the extra organizational impact of CSR. This research avenue provides the opportunity to develop ways to capture the dimensionality of embedded and peripheral CSR and the specific systems and processes that might lead to better outcomes for extra organizational stakeholders.
Fourth, future research might explore the effects of CSR programs in reaction to COVID-19 whereby employees engage in community service activities while receiving compensation from their firms. Organizations in the face of the pandemic are creating alternative ways to engage in CSR, with some attempting to deliver direct support to the community in the short term. Therefore, research that considers the way employees experience peripheral CSR, for example, by doing volunteering work but while receiving compensation, might lead to new insights on the nuances of how CSR structure (embedded versus peripheral) impacts employee outcomes.
Fifth, future research can explore the various indicators of employee resistance to CSR. Individual differences such as moral identity, communal values, work orientation and environmental values create variation in the way employees make sense of CSR (Aguinis and Glavas, 2019). The COVID-19 pandemic might be heightening some of these individual differences, such as work-family orientation or political orientation, that play a role in the way employees make sense of, give meaning to and react to CSR.
There are additional research opportunities related to the recent increase in research productivity in Latin America (Nicholls-Nixon et al., 2011;Ronda-Pupo, 2020;Ronda-Pupo and Katz, 2016) and Latin America's unique position to serve as a "natural laboratory" for building new theories and testing existing ones (Aguinis et al., 2020). Research on the behavioral perspective on CSR in Latin America would be uniquely valuable because it allows researchers to not rely on implicit assumptions of theories developed in the USA and other developed countries. Additionally, the region's relative homogeneity of institutional and economic conditions (Cuervo-Cazurra, 2016), together with distinctive conditions in some Latin American countries, make this a fruitful context. Specifically, two broad areas suggested by Aguinis et al. (2020) can be particularly generative for expanding our knowledge of behavioral CSR generally and in the wake of the COVID-19 pandemic, as further described.
Sixth, Latin America is well-suited to address future research on behavioral CSR regarding hybrid public-private collaborations that often occur in this context. Latin American countries sometimes lack sufficient investment in infrastructure and social services, so private companies often engage in designing and executing public activities, crafting public-private partnerships that bridge execution with government regulation (Engel et al., 2014). Cabral et al. (2013) ascertained that private companies reduced costs without neglecting critical services, which can be especially important during a pandemic when critical services are in high demand. The COVID-19 pandemic increases the potential for these types of collaborations in industries as diverse as health or fashion (e.g. creation of masks and medical equipment). Overall, this creates a unique context to understand employee reactions to embedded CSR, the factors that make these hybrid collaborations successful in responding to crises such as COVID-19, and the consequences for both private and public enterprises that collaborate in these types of COVID-19 initiatives.
A final area for which Latin America also offers an advantageous context for future research on behavioral CSR is social inclusiveness, inequity and vulnerable populations (Aguinis et al., 2020). As with most crises and pandemics, those with less suffer the hardest consequences, and "the COVID-19 pandemic has once again forcefully pushed societal inequalities into public consciousness" (Bapuji et al., 2020, p. 2). Latin America offers a unique context to test how CSR initiatives can mitigate COVID-19's impact on the vulnerable conditions under which millions live (e.g. the case of Rio de Janeiro's favelas; Briso and Phillips, 2020) and how these CSR initiatives translate into more meaningfulness in and at work for the employees of these firms.
As shown in this section, COVID-19 has opened numerous future research areas. We described seven possible research directions. But, clearly, opportunities for future research abound and additional areas could be added to our list. For example, many organizations fail to adopt the practices that management research suggests are most effective, and this discrepancy between research and practice has been labeled the research-practice gap (Cascio and Aguinis, 2008;Rynes et al., 2018). Future research on behavioral CSR could include partnerships between researchers and practitioners, who together are likely to generate knowledge that is both rigorous and relevant. Second, there is a rather vast literature on organizational responses to crises and adversities (Bundy et al., 2017;Williams et al., 2017). Future research could try to understand whether a behavioral CSR perspective is also useful for understanding employee responses to other types of crises and extreme situations.

Implications for organizational practices and policymaking
The research we synthesized and integrated in the previous sections has multiple useful implications for practitioners and policymakers. As a preview, these implications are listed below: Behavioral CSR and COVID-19: implications for organizational practices and policymaking When selecting a CSR strategy, leaders should keep in mind that embedded CSR (i.e. CSR that involves an organization's core competencies and integrates CSR within a firm's strategy, routines and operations) has been linked to positive outcomes for both the firm and employees when correctly implemented taking into account employees. Peripheral CSR (i.e. CSR that is not integrated with the firm's strategy, routines and operations but is based on initiatives such as donations or volunteering) results in mixed outcomes at best. "One size fits all" CSR does not always work. Employees involved in peripheral CSR might view it as symbolic or selfishly motivated by their firms and can have negative reactions to CSR when they are skeptical of the organization's claims (e.g. green washing) or when they view CSR as motivated by pure self-interest in profit. This provides an important caution to leaders seeking to implement CSR for image or public relations reasons. Employees quickly understand the type of CSR in which their organization is engaging. CSR initiatives can have negative effects on employees (i.e. the "dark side of CSR"). If a CSR initiative is peripheral and is also forced top-down on employees, it can backfire. HRM practices such as performance management and compensation can be helpful in making employees aware of CSR and show the firm's commitment to CSR. For example, employee performance evaluations and salary decisions can be based in Employee responses to COVID-19 part on employees' contributions to a firm's CSR initiatives. In other words, employees can get "performance points" and bonuses for being active participants in CSR. Because CSR is linked to sense making, making employees aware of and accountable for the firm's CSR involvement can lead to employee re-engagement through finding more meaningfulness in work.
First, our discussion of different types of CSR is useful regarding actions that governments and organizations of all types and in all industries can implement and their anticipated effectiveness. Embedded CSR involves an organization's core competencies and integrates CSR within a firm's strategy, routines and operations, while peripheral CSR does not.
Embedded CSR affects all employees as well as external stakeholders in multiple ways, including, among others, fulfilling internal and external stakeholder needs, enhancing consumer and employee pride and identification and providing a conduit to seek and find meaningfulness in and at work. Second, we explained why "one size fits all" CSR does not always work, which accounts for the documented variance in outcomes of CSR ranging from positive to neutral and even to negative. This provides an important caution to organizations seeking to implement CSR for image or public relations reasons. Employees quickly understand whether their organization is engaged in embedded or peripheral CSR. While embedded CSR is related to several positive outcomes when correctly implemented with employees in mind, this is not always the case for peripheral CSR.
Third, we described negative effects of CSR, what we refer to as the "dark side of CSR," as not all CSR initiatives and strategies have a positive impact on employees. Consider a situation where CSR is not only peripheral but is also forced top-down on employees. Leaders could be pushing for employees to engage in CSR while performance management systems encourage a focus on short-term financial results. In such situations where CSR is peripheral, symbolic and pushed top-down on employees, CSR can backfire. Leaders in Latin America should pay special attention to this point as a top-down leadership style is more common in this region. Both quantitative (Howell et al., 2007) and qualitative (D'Iribarne, 2002;Lindsley and Braithwaite, 1996;Rodriguez and Rios, 2008) cross-cultural research has found that authority-based and collective values in Latin America result in a unique leadership style characteristic of Latin America compared to the USA or Europe (Castaño et al., 2015). Specifically, this is typically referred to as the autocratic model of leadership and, more recently, paternalism leadership (Davila and Elvira, 2012), which combines high authority with close personal relationships and emphasizes the relationship orientation of respecting power and authority (Martinez, 2005).
Finally, we elaborated on how organizations and government agencies can use HRM practices such as performance management  and compensation to make employees aware of CSRwhich as a result, could increase employees' trust in their organization's commitment to CSR (as they are showing their commitment and not just talking about it). For example, employee performance evaluations and salary decisions can be based in part on employees' contributions to a firm's CSR initiatives. In other words, employees can get "performance points" and bonuses for being active participants in CSR. Because CSR is linked to sense making, making employees aware of and accountable for the firm's CSR involvement can lead to employee re-engagement by finding more meaningfulness in work.

Mini case study: Amazon
Given our previous summary of research findings and their implications for practice, let us now revisit the situation faced by Amazon as described in the opening vignette in our article. While Amazon's CSR initiative could have been an embedded CSR strategy because it involved one of the Amazon's core competencies (online delivery of goods) and was integrated within the firm's strategy, the firm seems to have failed to implement it in a way that involved employees. In other words, it likely was a "one size fits all" top-down CSR strategy that, as often happens, did not work as expected. Moreover, Bezos' donation, a peripheral CSR action, created further backlash (The Economist, 2020). Employees and the media quickly reacted against it as they considered it a hypocritical action to distract attention from the unsanitary conditions and high-risk situations that employees felt they were facing as a consequence of the CSR strategy to expand online grocery delivery. Amazon could have designed its expanded grocery delivery service with employee involvement and safety conditions in mind; used HRM practices such as performance management and compensation to make employees aware of CSR; and showed the firm's commitment to CSR (e.g. linking performance evaluations and salary decisions to employees' contributions to a firm's CSR initiatives). Had Amazon done that, employees would have been more likely to join the strategy and also become reenergized and find new meaningfulness through their work rather than resisting the CSR initiatives.

Concluding remarks
We described why a behavioral perspective on CSR is useful and necessary, especially in the context of the COVID-19 pandemic. Also, we described two types of CSRembedded and peripheraland their effects on employees, and we shed light on the reasons why CSR can result in both positive and negative outcomes for employees (refer to "Behavioral CSR: key findings"). Then, we offered suggestions for future research, such as how HRM practices are impacted by and, in turn, impact CSR responses to COVID-19 and pointed out the unique opportunity of research in Latin America to address these future research avenues (refer to "Behavioral CSR and COVID-19: future research directions"). Finally, based on the existing evidence, we offered recommendations on how a behavioral perspective on CSR is useful for practice and policymaking (refer to "Behavioral CSR and COVID-19: implications for organizational practices and policymaking"). Overall, a behavioral perspective on CSR offers novel theoretical insights on when and why CSR initiatives work as intended and when they do not, which is a critical information as organizations in all industries and of all types, as well as entire countries, implement actions and policies in response to COVID-19.