Is reactive corporate social responsibility such a bad idea? It depends on your positioning

Patagonia, a brand known for its environmental activism, was reported to be sharing a certain Sri Lankan supplier with fast fashion brands. Apparently, neither the employees nor the manager of the factory in question noticed any difference in the working conditions offered by fast fashion brands and those granted by Patagonia (FTM, 2023). After the report’s publication, the brand’s founder, Yvon Chouinard, committed Patagonia to new corporate social responsibility (CSR) initiatives including funding a daycare center and a sanitation and health program in the affected community. This case is an example of reactive CSR, through which a brand responds to a crisis by investing in socially responsible initiatives. Using attribution theory (Weiner, 1985; Ginder et al., 2021; Jones and Nisbett, 1987), this study explores the effectiveness of reactive CSR in improving consumer responses toward a brand involved in a crisis.

Attribution theory (Weiner, 1985; Ginder et al., 2021) suggests that stakeholders make sense of CSR activities by attributing to them intrinsic or extrinsic motives (Maon et al., 2021). Intrinsic motives (e.g. a genuine commitment to environmental sustainability) are more likely to be perceived as authentic and altruistic. By contrast, extrinsic motives (e.g. enhancing brand image or responding to regulatory pressure) may be met with skepticism from stakeholders (Moreno and Kang, 2020). The attribution of motive significantly influences how stakeholders evaluate the company’s CSR efforts and judge its overall reputation as a benevolent, caring and warm company (Chen et al., 2020). Attribution theory emphasizes the importance of a company’s perceived motives in evaluating its CSR actions (Weiner, 1985; Ginder et al., 2021). The theory thus highlights the critical role of authenticity in shaping stakeholder responses and, ultimately, increasing corporate resilience (Carroll, 2021).

Reactive CSR is conceptualized as a potential response strategy to mitigate the backlash from a crisis (Coombs, 2007). It is a costlier response than the company verbally apologizing for the crisis and committing to avoiding a recurrence (Chaudhry and Loewenstein, 2019). Reactive CSR offers the company an opportunity to reassure consumers by demonstrating its willingness to contribute positively to society. Responses that demonstrate a strong commitment to redressing the harm caused are highly appreciated by consumers (Ohtsubo and Watanabe, 2009). When evaluating a crisis response, consumers scrutinize the potential underlying motives and whether the organization is well-intentioned in its efforts (Rim and Ferguson, 2020). Evidence of a costly commitment tends to reassure consumers of organizational benevolence as it signals the company’s genuine willingness to help (Ohtsubo and Watanabe, 2009). The fact that the company is willing to spend money to rectify its mistakes indicates that reactive CSR should be a more effective crisis response strategy than a less costly statement of apology.

However, existing studies question the extent to which reactive CSR can improve an organization’s reputation. These studies focus on reactive CSR as a company’s response to external pressure, such as potential new regulation, negative media coverage or evidence of potential wrongdoing, by implementing CSR initiatives (Ricks, 2005; Groza et al., 2011). This view stresses how CSR initiatives are driven by the need to mitigate reputational risks. This is contrasted with firms pursuing proactive CSR before negative reports emerge (Wagner et al., 2009). Consumers may see reactive CSR initiatives as opportunistic if they believe that the company is merely responding to public pressure (Groza et al., 2011) and is not genuinely interested in CSR (Rim and Ferguson, 2020).

Consistent with recent theorizing on CSR (Carroll, 2021; Siltaloppi et al., 2021; Elmassri et al., 2023; Tourky et al., 2020), we assume that the motivation for reactive CSR strategy would be perceived as authentic and benevolent only when the brand has built a stable CSR positioning in its core business strategy (Fatima and Elbanna, 2023). According to this perspective, reactive CSR as a crisis response can be seen as an extension of previous commitments by the same organization. In contrast, for companies that have built their core business positioning on performance or innovation, using reactive CSR as a crisis response strategy would be perceived as extrinsic and not sincere.

We conducted three experiments that considered both fictitious and existing brands to show that reactive CSR is persuasive for brands that can boast a preexisting commitment to CSR. For these organizations, reactive CSR has a better impact on consumer responses than an apology; this is not the case for brands positioned on performance. Our results make several contributions to existing research. First, we extend research on crisis communications (Bundy et al., 2017; Coombs, 2007) by offering the first conceptualization of reactive CSR as a potential crisis response strategy. We extend current research (Groza et al., 2011; Rim and Ferguson, 2020; Ricks, 2005) by demonstrating that brand positioning is fundamental in shaping consumers’ responses to reactive CSR following a crisis. Finally, we contribute to the literature on CSR by adding the ability to successfully implement a reactive CSR strategy to the list of advantages associated with CSR perceptions in previous research (Klein and Dawar, 2004).

Our study also contributes to the recent debate on the integration of CSR into business strategy (Carroll, 2021; Siltaloppi et al., 2021; Elmassri et al., 2023; Tourky et al., 2020) by showing that companies adopting a proactive and stable brand positioning focused on CSR can effectively use reactive CSR tools to restore their reputations as benevolent and caring in times of crisis. For these organizations, reactive CSR becomes an extension of their strategic, pre-crisis activities. This study offers a novel suggestion in the ongoing debate and supports the synergistic interplay between proactive and reactive CSR strategies during times of crisis.

Although we focus on CSR initiatives in response to a crisis, reactive CSR needs to be understood in the context of attempts to differentiate this from more proactive approaches to CSR. Reactive CSR refers to a company’s response to societal or environmental issues after they have emerged or become public concerns (Ricks, 2005; Carroll, 2021). Research suggests that companies often engage in reactive CSR in response to external pressures, such as negative media coverage, public outrage or regulatory action (Groza et al., 2011). According to attribution theory (Weiner, 1985; Ginder et al., 2021), reactive CSR can lead to extrinsic motives being attributed to the company since it is perceived as engaging with the problem defensively, only after receiving criticism from external stakeholders (Ricks, 2005).

By contrast, proactive CSR leads to inferences of intrinsic motives since the company has voluntarily addressed social and environmental issues before demands have been made to do so (Galbreath, 2009). Proactive CSR is seen as a strategic choice that aligns business goals with societal and environmental needs. It is generally considered effective in building and maintaining a positive corporate reputation (Ricks, 2005). The ongoing CSR debate (Carroll, 2021; Elmassri et al., 2023) suggests that companies should not only align their CSR practices with business goals but also integrate proactive CSR in its ethical, environmental and social practices, core business strategies and brand positioning (Kim et al., 2023; Tourky et al., 2020). Consistent with this, a brand positioning focused on CSR is an expression of the company’s DNA and its proactive engagement for the well-being of the community (Tourky et al., 2020). We build on this assumption, suggesting that reactive CSR can be effective and seen as genuine only when the company can rely on and leverage a solid and proactive CSR positioning that was established before the crisis.

To the best of our knowledge, there are only four studies in the literature that investigate the factors that can enhance reactive CSR effectiveness by reducing consumer perceptions of extrinsic motives. Groza et al. (2011) identified two factors that reduce consumer skepticism of reactive CSR and improve their attitude toward the company and their purchase intentions after a crisis. The first factor is the source of the reactive CSR information: externally published CSR information reduces consumer skepticism more than company communications. The second factor is where CSR initiatives are deployed. Consumers tend to be less skeptical when reactive CSR initiatives are local or very closely geographically. Rim and Ferguson (2020) investigated how the fit, as the degree of alignment, compatibility or mutual suitability between the crisis event and the reactive CSR actions, impacts the effectiveness of CSR strategy. They showed that in the case of a victim crisis, a reactive CSR strategy with a high fit to the crisis event has a positive influence on the company’s motives being perceived as altruistic, leading to the effective restoration of the company’s reputation after the negative event.

By contrast, in the case of a preventable crisis, reactive CSR is less effective in restoring a company’s reputation, especially if the fit with the crisis is very high. In the same way, Kim and Choi (2018) have suggested that consumers respond more favorably, in terms of attitude and purchase intentions, when a company launches a reactive CSR initiative that is congruent with the crisis and when the crisis is the result of an accident rather than an intentional transgression. Finally, Tao and Song (2020) demonstrated that to restore positive attitudes toward a company after a crisis, a response communication strategy that reaffirms or stresses previous CSR associations is more effective than one that only emphasizes associations with company’s ability. Extending this body of evidence, we show that the positioning of the brand contributes to the effectiveness of a reactive CSR strategy in response to a crisis.

In this study, we focus on reactive CSR as a potential response to mitigate the backlash from a crisis (Coombs, 2007). Carroll (2021) suggests that reactive CSR strategies can be perceived as extrinsically motivated if they are not aligned with a brand’s values or long-term CSR goals. We build on this assumption, suggesting that reactive CSR can only be effective and perceived as genuine when the company can rely on and leverage a solid and proactive CSR positioning established before the crisis.

The perspective of attribution theory (Weiner, 1985; Ginder et al., 2021; Jones and Nisbett, 1987) is that it is important to understand the relative hypocrisy or authenticity of a socially responsible initiative when consumers might consider reactive CSR as intrinsically motivated.

Perceived hypocrisy refers to the belief or impression that the company’s actions contradict its publicly stated values, commitments or promises (Wagner et al., 2009; Chen et al., 2020; Lee and Hur, 2024). This perception arises when consumers observe a disconnect between what a company says (e.g. ethical practices, social responsibility, sustainability) and what it does (e.g. unethical behavior, environmental harm, mistreatment of employees). There is some evidence that CSR can provide a buffer against the negative consequences of a crisis or wrongdoing (Kim and Park, 2020). At the same time, however, the literature on hypocrisy perception reveals that CSR can also make things worse when the crisis is in the same domain as the organization’s CSR engagements (Wagner et al., 2009; Ryoo and Kim, 2024).

In particular, CSR commitments can be counterproductive when subsequent events appear inconsistent with or directly contradict them (Wagner et al., 2009; Smith and Rhiney, 2020) due to consumer perceptions of company hypocrisy (Wagner et al., 2009). Wagner and colleagues (2009) further suggest that the perception of hypocrisy is linked to the sequence of events. CSR before a crisis, which is subsequently contradicted by the company’s failure to keep the promises made, leads to strong perceptions of hypocrisy. However, reactive CSR is not seen as hypocritical because it takes place after the crisis, and it is thus not seen as a contradiction or a failure to live up to promises made (Wagner et al., 2009).

While hypocrisy relates to a perceived inconsistency between words and actions, the concept of CSR inauthenticity is more about a general sense of or lack of genuineness, regardless of whether there’s a direct contradiction (Beckman et al., 2009; Muniz and Guzmán, 2023). Inauthenticity is the perception that CSR is not intrinsically motivated (Kim and Choi, 2018) and can lead to accusations of “woke washing” (Walter et al., 2024). Research emphasizes the need for attributions of sincerity, consistency and transparency to achieve authentic and intrinsic CSR perceptions (Beckman et al., 2009; Mazutis and Slawinski, 2014).

Considering reactive CSR as a strategy for crisis response, we reason that such an initiative is more likely to be perceived as intrinsically motivated when it is implemented by a brand already positioned on CSR. As such, reactive CSR should demonstrate the brand’s commitment to social responsibility by addressing the urgent and specific needs created by the crisis (Groza et al., 2011). Instead of viewing it as mere damage control, we suggest that reactive CSR can be seen as an opportunity for the company to reassess its values, positioning and practices, ultimately leading to a more authentic CSR framework (Rim and Ferguson, 2020; Walter et al., 2024). In other words, even though reactive CSR takes place after a negative event, it does not necessarily have to be perceived as extrinsically motivated. The perception of authenticity in CSR is often restored through concrete actions rather than empty promises (Mazutis and Slawinski, 2014; Walter et al., 2024; Carroll, 2021). Companies engaging in genuine reactive CSR should thus both rectify the immediate issues and demonstrate a long-term commitment to social responsibility and authentic remediation efforts (Mazutis and Slawinski, 2014).

Benevolence refers to a brand’s genuine and intrinsic concern for the welfare of stakeholders and society at large. Consistent with attribution theory (Ginder et al., 2021), perceived benevolence is the understanding that the brand’s CSR activities are driven by intrinsic motives and a sense of moral responsibility rather than financial or reputational gain (Xie and Peng, 2009; Kim and Park, 2020; Beveridge and Höllerer, 2023). Therefore, in the present study, we consider perceived brand benevolence to be a measure of the perceptions of an intrinsic CSR motive. A CSR crisis will negatively influence perceptions of brand benevolence, as it communicates a lack of concern for consumers and the consequences that corporate action might have for them (Bundy et al., 2017; Coombs, 2007). Reactive CSR can be deployed to limit the negative impact of a crisis and the reduction in brand benevolence that is likely to result by allowing the company to make authentic amends for its alleged wrongdoing (Ricks, 2005). Coombs (2007) has suggested that, depending on the nature of the crisis, companies may provide reparations or compensation to affected parties, such as financial compensation or CSR efforts to address the harm caused. Reactive CSR is a long-term strategy to repair damage and rebuild trust (Rim and Ferguson, 2020). It may begin soon after the crisis, but the efforts made, such as community involvement or implementing new sustainability practices, are designed to have a lasting impact. Reactive CSR aims to rebuild trust and improve the perceived benevolence of the brand over time, making it an ongoing process that continues well beyond immediate crisis resolution (Groza et al., 2011). For this reason, reactive CSR is costly, and the company must be confident in its effectiveness before relying on the strategy as a crisis response.

After a crisis, consumers scrutinize the company’s responses with the awareness that it might be trying to influence or manipulate them (Rim and Ferguson, 2020). There is some evidence to suggest that only authentic CSR initiatives communicated after an accidental crisis or congruent with the crisis domain can positively influence consumers’ perceptions of a company after a negative event (Kim and Choi, 2018; Rim and Ferguson, 2020). In different circumstances, while reactive CSR may address the crisis, it may also expose the company to scrutiny and skepticism if it is perceived as extrinsic or opportunistic.

An apology is an account used to acknowledge a mistake or failure, express regret and, ideally, restore trust and mend relationships (Coombs, 2007). An apology can be an effective crisis management tool for mitigating negative consumer responses (Rasoulian et al., 2017). However, exclusively verbal apologies might be dismissed by consumers as “cheap talk” (Rasoulian et al., 2017). In this respect, past research has shown that these verbal apologies might need to be paired with costly responses – such as financial compensation or investments in environmental restoration – to increase their effectiveness (Chaudhry and Loewenstein, 2019; Ohtsubo and Watanabe, 2009). Evidence of a financial commitment enhances the impression that the company means well and has authentic intentions. Given this evidence on the effectiveness of costly crisis response strategies, we expect consumers to respond more favorably to reactive CSR initiatives than to a less costly verbal response (i.e. an apology).

We considered several consumer responses to assess the impact of reactive CSR. The intention to forgive the brand is the foundation for relationship restoration after a crisis (Wei et al., 2020). Forgiveness intentions are closely related to subsequent behavioral responses such as purchase intentions and negative word of mouth (Antonetti and Baghi, 2023; Wei et al., 2020). We also measured the potential erosion of future willingness to buy the brand as a consequence of a crisis (Kim and Choi, 2018). Consumers often punish companies by spreading negative word of mouth about the brand (Baghi and Antonetti, 2021) and sharing negative information and content online (Muraro et al., 2023). Based on the preceding discussion, we hypothesize as follows:

Our predictions in H1 stand in contrast to evidence against reactive CSR. As discussed above, scholars have highlighted how reactive CSR can elicit skeptical responses from consumers who are unsure of the organization’s motivation in undertaking socially responsible initiatives (Groza et al., 2011; Rim and Ferguson, 2020). However, current debates about perceptions of reactive CSR neglect the potential of CSR brand positioning as an expression of proactive CSR efforts. As attribution theory suggests, the perception that CSR efforts are benevolent may be influenced by a variety of factors, including the company’s CSR track record (Vanhamme and Grobben, 2009). When a company demonstrates a long-standing commitment to CSR, it fosters a perception of stability, consistency and longevity in its efforts (Vanhamme and Grobben, 2009). These perceptions can enhance consumer trust, boost perceived brand benevolence and increase the company’s resilience in challenging times. Brands vary in the extent to which they rely on CSR activities to position themselves in consumers’ minds. While many brands may occasionally affiliate with societal or environmental causes, others have stronger, long-standing CSR commitments (e.g. The Body Shop or Patagonia). The latter become known as socially responsible brands, for whom CSR is a critical and long-lasting component of their brand identity and positioning (Guzman and Davis, 2017). Brands try to rely on the proactive CSR history and reputation to create a solid CSR brand positioning (Tourky et al., 2020).

There is a relevant literature on the role of brand identity in proactive CSR activities and its impact on consumer responses. Brown and Dacin (1997) defined the fit between brand identity and CSR activity as the alignment between the features of a company’s established core identity and the social or environmental initiatives it undertakes. This fit ensures that proactive CSR activities resonate with the brand’s core values, mission and personality, creating a cohesive narrative that enhances the overall perceptions of the company and its reputation (Brown and Dacin, 1997). A strong fit between brand identity and CSR initiatives improves responses to brand activities (Brown and Dacin, 1997), mitigates skepticism (Bigné-Alcañiz et al., 2012) and fosters positive perceptions of brand equity (Guzman and Davis, 2017).

In our context, we considered positive responses to reactive CSR to be independent of the specific domain of the proactive CSR activities in which the company is currently engaged. Irrespective of the exact fit between the domains of proactive CSR and the consequent CSR repair effort, proactive CSR offers reassurance about the benevolence of the brand (Beckman et al., 2009). In other words, we see reactive CSR as effective because it communicates to consumers that, despite the crisis, the company is coherent with its positioning and is well-intentioned and caring in its crisis response. This message is less convincing for a brand positioned on performance that does not benefit from the positive signaling associated with proactive CSR.

We propose that reactive CSR initiatives, developed in direct response to a crisis and with the explicit goal of addressing the harm caused, are likely to be well received when they are implemented by brands positioned on CSR but will probably be dismissed when they are delivered by brands that do not consistently engage in CSR. For CSR brands, reactive initiatives are less likely to be perceived as suspicious or motivated by instrumental, strategic concerns (Kim and Choi, 2018) because they are aligned with the pattern of previous engagements – this can be seen as an opportunity for the company to reassess its values and positioning, ultimately leading to perceptions that its CSR framework is authentic.

Furthermore, we know that consumers generally react more positively to responses by CSR brands following a crisis. Klein and Dawar (2004) demonstrated that in non-routine settings, such as crises, associations with CSR positioning are more likely to be activated and exert a positive influence on judgments about the company (Brown and Dacin, 1997). The perceived warmth of the brand achieved through CSR positioning can increase consumer benevolence toward the brand when it tries to repair the damage caused by a crisis. Consistent with this evidence, we hypothesize as follows:

Figure 1 presents the moderated mediation model examined in this research. We tested H1 and H2 through three between-subjects experiments in which we presented participants with a hypothetical crisis and evaluated responses to a reactive CSR strategy versus an apology ^[1]. In Study 1, we considered a fictitious brand, and in Study 2, we tested the model using real brands, comparing responses to a CSR brand positioned to focus on sustainability (Brand CSR) versus a performance-focused alternative (Brand Performance) involved in a fictitious crisis. In Study 3, we used fictitious brands to replicate our findings and confirm the validity of Studies 1 and 2, which demonstrated that the positive effect of reactive CSR is not due to an alignment between the domain of CSR brand positioning and the domain of reactive CSR action.

A sample of 300 Italian participants was recruited by graduate students through their personal networks for course credit. The sample included 45% men across several age groups: 37.5% were 18–24 years old, 49% were 25–34 years old and 13.5% were 35–44 years old. Participants were asked to take part in an online study. We used a 2 (brand response: apology vs reactive CSR) × 2 (brand positioning: CSR vs performance) between-subjects experimental design. Participants were presented with a description of a fictitious sportswear brand (BRAVE) and its core positioning. We then asked them to imagine reading an article in their favorite newspaper or another trusted source, identifying irresponsible behavior associated with BRAVE. BRAVE responded to the allegations either through reactive CSR or an apology. The details of the manipulations are available in Supplementary material. Finally, participants answered our questions, responded to the manipulations check and provided demographic information.

We used six items to check whether the participants’ perceptions of the brand positioning were correct. The results of the manipulation checks are presented in Table 1. We found that participants’ perceptions of whether the brand “cares about the environment,” “makes sustainability a mission for its business,” and “it is an environmentally friendly brand” (assessed via a seven-point Likert scale; we used the average of three items) were consistent with expectations; see results in Table 1. Their perceptions that the brand is “committed to the highest performance for its products,” “makes product innovation a mission for its business,” and “is a top-quality brand” (assessed via a seven-point Likert scale; we used the average of three items) were also consistent with expectations; see results in Table 1.

We used four items to check the manipulation of the brand response. Participants’ perceptions of whether the brand “implements more sustainable production processes” and “promotes new initiatives to reduce environmental pollution in many developing countries” (assessed via a seven-point Likert scale; we used the average of both items) were consistent with expectations; see results in Table 1. Participants’ perceptions of whether the brand “apologized for the damage caused” or “communicated its regret” (assessed via a seven-point Likert scale; we used the average of the two items) were consistent with expectations; see results in Table 1.

We borrowed scales from the literature to test our conceptual model (see Supplementary material for details). We measured perceived brand benevolence (Xie and Peng, 2009), purchase intentions (Groza et al., 2011), intention to spread negative word of mouth (Baghi and Antonetti, 2021) and forgivingness (Xie and Peng, 2009). We also collected demographic information, including participant gender and age, which we later used as covariates.

We conducted a 2×2 analysis of variance (ANOVA) ^[2] of mediators and dependent variables with the brand response (apology vs reactive CSR) and brand positioning (CSR vs performance) as between-subjects factors. Our analysis showed a significant brand response X brand positioning interaction (F (1, 299) = 20.07, p < 0.001, partial η² = 0.07) on perceived brand benevolence. When the brand is positioned on CSR, reactive CSR increases consumer perceptions of brand benevolence more than an apology (M_apology 2.68, M_{reactive CSR} = 4.76, t (148) = −8.57, p < 0.001, d = −1.52). However, when the brand has a performance positioning, reactive CSR does not increase the perceived benevolence of the brand compared to an apology (M_apology 3.83, M_{reactive CSR} = 3.69, t (148) = 0.51, p > 0.05, d = 0.08).

Our analysis shows a significant brand response X brand positioning interaction on consumer intention to forgive (F (1, 298) = 18.59, p < 0.001, partial η² = 0.07), intention to purchase the brand (F (1, 298) = 6.01, p < 0.001, partial η² = 0.07) and intention to spread negative word of mouth (F (1, 298) = 6.80, p < 0.05, partial η² = 0.08). Figure 2 shows all the interaction effects for the mediator and dependent variables.

We tested H1 by using a conditional process analysis using PROCESS (Model 4; Hayes, 2017) with 10,000 resamples to estimate bias-corrected and accelerated 95% confidence intervals. The independent variable was coded 0 (apology) and 1 (reactive CSR), and perceived brand benevolence served as a mediator. Age and gender were included as covariates (although our results do not meaningfully change if covariates are excluded). The positive effect of reactive CSR on consumers responses in mediated by brand benevolence (intention to forgive indirect effect = 0.07, CI 0.00–0.15, purchase intention indirect effect = 0.05, CI 0.01–0.09 and negative word of mouth indirect effect = −0.04, CI −0.02 to −0.08). H1 is confirmed. We tested our hypothesized moderated mediation (H1 and H2 together) by running a conditional process analysis using PROCESS (Model 7; Hayes, 2017) with the same approach as that taken in Model 4. We used 10,000 resamples to estimate bias-corrected and accelerated 95% confidence intervals. The moderator variable was coded 0 (performance positioning) and 1 (CSR positioning). Age and gender were included as covariates ^[3]. The full results are provided in Table 2. Reactive CSR increases benevolence when the brand has a positioning focused on CSR (effect = 1.12, CI 0.59–1.66) but not when the brand focuses on performance (effect = −0.12, CI −0.61–0.37). The mediation of perceived brand benevolence is consistently supported when the brand focuses on CSR, leading to an increase of intention to forgive the brand (indirect effect = 0.24, CI 0.08–0.43), intention to purchase the brand (indirect effect = 0.25, CI 0.10–0.46) and a reduction in negative word of mouth (indirect effect = −0.18, CI −0.37 to −0.02). The mediation is not supported for performance positioning (intention to forgive the brand indirect effect = −0.03, CI −0.15–0.09; intention to purchase the brand indirect effect = −0.03 CI −0.16–0.09; intention to spread negative word of mouth indirect effect = 0.10, CI −0.06–0.11). The index of moderated mediation ^[4] provides additional evidence in support of the mediation of benevolence only where the brand is positioned on CSR (intention to forgive the brand index = 0.26, CI 0.08–0.99; intention to purchase the brand index = 0.29 CI 0.10–0.53; intention to spread negative word of mouth index = −0.18, CI −0.42 to −0.02). H1 and H2 are confirmed.

Through a pre-test (n = 100), which is reported in Supplementary material, we identified two existing outdoor brands, Brand CSR and Brand Performance, as comparable brands that differed in their positioning (see results in Table 1) ^[5]. We conducted a 2 (brand response: apology vs reactive CSR) X 2 (brand: Brand CSR vs Brand Performance) between-subjects experimental design. The scenarios involved Brand CSR or Brand Performance placed in a fictitious crisis. We recruited a sample of 280 Italian consumers using the same procedures as in Study 1. Of the total sample, 43% were men and made up of various age groups: 41.5% were 18–24 years old, 39% were 25–34 years old and 19.5% were 35–44 years old. Participants were presented with the logo of the outdoor brand and a short description of their actual CSR/performance brand positioning ^[6]. The two brands had comparable ratings for brand awareness (M_{Brand CSR} = 4.51 vs M_{Brand Performance} = 4.14; t (278) = 1.25; p > 0.05), brand quality (M_{Brand CSR} = 5.00 vs M_{Brand Performance} = 4.83; t (278) = 0.73; p > 0.05) and brand attitude (M_{Brand CSR} = 5.12 vs M_{Brand Performance} = 5.08; t (278) = 0.52; p > 0.05).

The results of the manipulation checks (using the same measures as used in Study 1) are presented in Table 1. The two brands differed significantly in terms of the perception of their positioning. This was the case for the items measuring the positioning on environmental CSR and those measuring the positioning on performance (see results in Table 1). Participants were presented with the newspaper article used in Study 1, but applied to either Brand CSR or Brand Performance. The article ended with the response of the brand (reactive CSR vs apology). Participants were then asked to provide responses regarding our key variables (using the same scales as Study 1) and to answer the manipulation checks for the company response and demographic questions. We used the four items used in Study 1 to check whether the brand response was correctly perceived. Participants’ perceptions of whether the response of the brand constituted reactive CSR were consistent with expectations (see Table 1). Participants’ perceptions of whether the brand “apologized for the damage caused” or “communicated its regret” were consistent with expectations (see Table 1).

Further details on the brand descriptions, crises and organizational responses are presented in Supplementary material. At the end of the survey, participants were debriefed and informed that the crisis presented in the article was fictitious and developed for research purposes. Specifically, participants were informed that the brands mentioned were in no way involved in the crisis. We borrowed the scales used in Study 1 to measure our proposed mediator and dependent variables (see Supplementary material for detailed items). We also collected demographic information, including gender and age.

We conducted a 2×2 ANOVA of the mediators and dependent variables with the brand response after the crisis (reactive CSR vs apology) and brand positioning (Brand CSR vs Brand Performance) as between-subjects factors. The analysis shows a significant brand response X brand positioning interaction (F (1, 278) = 46.09, p < 0.001, partial η² = 0.07) on perceived brand benevolence. This indicates that when the brand with a positioning focused on CSR (i.e. Brand CSR) promoted reactive CSR in response to a crisis, consumers’ perception of brand benevolence is higher than with the apology response (M_apology 2.19, M_{reactive CSR} = 3.99, t (138) = −9.01, p < 0.001, d = −1.50). In contrast, when the brand focused on performance (i.e. Brand Performance) promoted reactive CSR, we do not observe any increase in the perception of brand benevolence compared to the case of the apology (M_apology = 3.17, M_{reactive CSR} = 3.04, t (138) = 0.74, p > 0.05, d = 0.13). The analysis also shows a significant brand response X brand positioning interaction on consumer intention to forgive (F (1, 278) = 18.76, p < 0.001, partial η² = 0.07), purchase the brand (F (1, 278) = 23.41, p < 0.001, partial η² = 0.08) and spread negative word of mouth F (1, 278) = 32.85, p < 0.001, partial η² = 0.12).

When Brand CSR engages in reactive CSR, there is an increase in consumers’ intention to forgive (M_apology 2.77, M_{reactive CSR} = 4.20, t (140) = −5.74, p < 0.001, d = −0.97) and purchase the brand (M_apology 2.93, M_{reactive CSR} = 4.10, t (140) = −8.88, p < 0.001, d = −1.23) and a reduction in negative word of mouth (M_apology 3.81, M_{reactive CSR} = 2.90, t (140) = 7.86, p < 0.001, d = 1.31). In contrast, reactive CSR do not influence the intention to forgive (M_apology 3.38, M_{reactive CSR} = 3.41, t (140) = −0.15, p > 0.05, d = 0.03) purchase the brand (M_apology 2.35, M_{reactive CSR} = 2.62, t (140) = −0.89, p > 0.05, d = 0.05), or share negative word of mouth (M_apology 3.88, M_{reactive CSR} = 3.81, t (140) = 0.35, p > 0.05, d = 0.06) for Brand Performance. Figure 3 shows the interaction effects.

We tested H1 through a conditional process analysis using PROCESS (Model 4; Hayes, 2017) following the approach used in Study 1. The independent variable was coded 0 (apology) and 1 (reactive CSR), and perceived brand benevolence served as a mediator. Age, gender and brand attitude were included as covariates ^[7]. We find evidence that the positive effect of reactive CSR on consumer responses is mediated by brand benevolence (intention to forgive indirect effect = 0.38, CI 0.22–0.58, purchase intention indirect effect = 0.39, CI 0.23–0.58 and negative word of mouth indirect effect = −0.39, CI −0.58 to −0.22). H1 is confirmed.

We tested the hypothesized moderated mediation (H1 and H2) by running a conditional process analysis using PROCESS (Model 7; Hayes, 2017) following the approach used in Study 1. The moderator variable was coded 0 (Brand Performance) and 1 (Brand CSR). Age, gender and brand attitude were included as covariates ^[8]. Detailed results are reported in Table 2. We find evidence that reactive CSR increases perceived brand benevolence when deployed by Brand CSR (effect = 1.83, CI 1.46–1.19) but not when it is deployed by Brand Performance (effect = −0.13, CI −0.51–0.25). The mediation of perceived brand benevolence is consistently supported for Brand CSR (forgivingness indirect effect = 0.45, CI 0.20–0.72; purchase intention indirect effect = 0.65, CI 0.36–0.99; negative word of mouth indirect effect = −0.58, CI −0.91 to −0.31) but not for Brand Performance (forgivingness indirect effect = −0.03, CI −0.13–0.06; purchase intention indirect effect = −0.05 CI −0.19–0.08; negative word of mouth indirect effect = 0.04, CI −0.07–0.16). The index of moderated mediation supports the mediating role of benevolence for a Brand CSR positioned on CSR (forgivingness index = 0.48, CI 0.21–0.78; purchase intentions index = 0.69, CI 0.37–1.10; negative word of mouth index = −0.61, CI −1.00 to −0.32). H1 and H2 are supported.

A sample of 200 Italian participants was recruited by graduate students through their personal networks for course credit. The sample was 42% male and included different age groups: 39% were 18–24 years old, 51% were 25–34 years old and 10% were 35–44 years old. Participants were asked to take part in an online study. We used a 2 (brand response: apology vs reactive CSR) × 2 (brand positioning: CSR vs performance) between-subjects experimental design. Participants were presented with a description of the fictitious apparel brand (KROMO) and its core positioning. As in the other studies, participants read an article identifying irresponsible behavior at KROMO. Study 3 differs from Studies 1 and 2 because, in this case, the reactive CSR implemented was not aligned with the sustainability positioning of the brand. The former focuses on worker rights while the latter concerns an environmental sustainability. For more details about the scenario and corporate responses see Supplementary material.

As in Studies 1 and 2, we assessed CSR brand positioning and found that the manipulation was successful. The results of the manipulation checks (measured as in Study 1) are presented in Table 1. Participants’ perceptions that the brand “committed to the highest performance for its products” were consistent with expectations (see Table 1). Participants’ perceptions of whether the brand “supported humanitarian associations that promote the protection of workers’ rights in Southeast Asia” and “launched humanitarian projects to ensure economic and social development based on fair and sustainable working conditions in developing countries” (assessed using a seven-point Likert scale and the average of both items) were consistent with expectations (see Table 1). Participants’ perceptions of whether the company “apologized for the damage caused” (same items as in previous studies) were consistent with expectations (see Table 1).

We borrowed the scales used in previous studies to measure our proposed mediator and dependent variables (see Supplementary material for detailed items). We also collected demographic data, including gender, age and education, which we later used as covariates.

We conducted a 2×2 ANOVA of mediators and dependent variables with brand response (apology vs reactive CSR) and brand positioning (CSR vs performance) as between-subjects factors. Our analysis shows a significant brand response X brand positioning interaction (F (1, 198) = 45.23, p < 0.001, partial η² = 0.06) on perceived brand benevolence. When the brand is positioned on CSR, reactive CSR increases consumer perceptions of brand benevolence more than the apology (M_apology 2.50, M_{reactive CSR} = 3.86, t (98) = −6.21, p < 0.001, d = −1.09). However, when the brand has a performance positioning, reactive CSR does not increase perceived brand benevolence compared to the apology (M_apology = 3.07, M_{reactive CSR} = 2.99, t (98) = 0.42, p > 0.05, d = 0.06). Our analysis shows a significant brand response X brand positioning interaction on consumer intentions to purchase the brand (F (1, 198) = 4.43, p < 0.05, partial η² = 0.02) and intention to spread negative word of mouth (F (1, 198) = 11.22, p < 0.001, partial η² = 0.06). Figure 4 plots the interaction effects and displays the descriptive statistics.

We tested H1 by running a conditional process analysis using PROCESS (Model 4; Hayes, 2017) with the same approach used in Studies 1 and 2. The independent variable was coded 0 (apology) and 1 (reactive CSR), and perceived brand benevolence served as a mediator. Age and gender were included as covariates ^[9]. We find evidence that the positive effect of reactive CSR on consumer responses is mediated by brand benevolence (intention to forgive indirect effect = 0.18, CI 0.21–0.38, purchase intention indirect effect = 0.29, CI 0.12–0.47 and negative word of mouth indirect effect = −0.29, CI −0.48 to −0.32. H1 is confirmed.

We tested the hypothesized moderated mediation (H1 and H2) by running a conditional process analysis using PROCESS (Model 7; Hayes, 2017) and the approach taken in Studies 1 and 2. Age and gender were included as covariates ^[10]. Detailed results are reported in Table 2. We find evidence that reactive CSR increases perceived brand benevolence when the brand has a positioning focused on CSR (effect = 0.89, CI 0.28–1.48) but not when the brand focuses on performance (effect = −0.54, CI −1.08–0.001). Reactive CSR increases forgivingness (indirect effect = 0.19, CI 0.03–0.41) and purchase intentions (indirect effect = 0.20, CI 0.03–0.44) and reduces negative word of mouth (indirect effect = −0.21, CI −0.46 to −0.04). The mediation is not supported for performance positioning (forgivingness indirect effect = −0.05, CI −0.14–0.07; purchase intention indirect effect = −0.09 CI −0.21–0.09) and the reactive CSR backfires leading to an increase in negative word of mouth (indirect effect = 0.12, CI 0.01–0.31). The index of moderated mediation supports the mediation of benevolence in the case of a brand with CSR positioning (intention to forgive the brand index = 0.32, CI 0.09–0.63; intention to purchase the brand index = 0.33 CI 0.09–0.64; intention to spread negative word of mouth index = −0.33, CI −0.71 to −0.08). H2 is confirmed.

We conducted three studies to demonstrate that the effectiveness of reactive CSR as a crisis response depends on the preexisting positioning of the brand when deploying this communications strategy. While the approach is effective for brands already positioned on CSR, companies focused on performance do not benefit from reactive CSR. Interestingly, the findings suggest that, for companies positioned on CSR that have been proactive in their CSR initiatives, stakeholders expect reactive CSR following a crisis. Results show that, for CSR brands, apologies might be even less effective than they are for performance-focused brands. Reactive CSR clearly emerged as the best possible response for an organization already positioned on CSR. For companies focused on performance, reactive CSR will not necessarily backfire but does not offer any specific advantage.

We have also demonstrated that it does not matter whether positioning on CSR and reactive CSR are aligned. Positioning on CSR activities alone would affect the relative persuasiveness of reactive CSR as a response strategy since proactive efforts reassure consumers about the motives underlying the company’s response to a crisis (Carroll, 2021). Our results show that a close fit in this respect is not necessary for reactive CSR to be effective. This is because CSR brand positioning reassures consumers about the benevolence of the company but does not create specific expectations in relation to the crisis and the response of the organization.

It should be noted that we focused on how having reactive CSR and apology as alternative responses influenced consumer perceptions. Our results do not suggest that performance-focused organizations should not engage in CSR initiatives following a crisis. On the contrary, there might be very good moral reasons to do so, as CSR initiatives can represent a meaningful effort to repair the damage caused by a crisis (Pfarrer et al., 2008). However, our evidence has illustrated that organizations focused on performance should not expect that reactive CSR will improve consumer responses.

Our study contributes to the recent debate on the integration of CSR into business strategies (Carroll, 2021; Siltaloppi et al., 2021; Elmassri et al., 2023; Tourky et al., 2020). There has been a growing emphasis on embedding CSR into core business strategies rather than treating it as an occasional or peripheral activity (Elmassri et al., 2023). This shift reflects a broader recognition that aligning social impact with business objectives can build resilience and drive overall success (Nguyen and Kanbach, 2024). Carroll’s (2021) contribution to this debate was to present a critical perspective on reactive CSR, highlighting the limitations of compliance-based CSR approaches. According to Carroll (2021), CSR strategies characterized by minimal effort, reactivity and mere accommodation to societal expectations are no longer sufficient in a rapidly evolving business landscape. Instead, companies are increasingly pressured to adopt proactive CSR leadership, integrating strong ethical, environmental and social practices into their core strategies and positioning (Carroll, 2021).

The results of this study support this claim by showing that companies adopting a proactive and stable positioning on CSR are also able to effectively use reactive CSR tools to restore their reputation as being benevolent and caring in times of crisis. The integration of CSR into their business strategy places additional tools at the company’s disposal that are more effective than a simple apology in managing a crisis. This study makes a novel contribution to the current debate, supporting the synergistic interplay between proactive and reactive CSR strategies during a crisis.

Our study contributes to extant research on reactive CSR (Ricks, 2005; Groza et al., 2011) in two ways. First, we conceptualized reactive CSR as a crisis response strategy and examined its effectiveness. Earlier research had considered this strategy in a more generic sense as a response to external pressures (Ricks, 2005; Groza et al., 2011): we extend the conceptualization of reactive CSR by articulating its potential role in responding to unexpected negative events that could harm external consumers (Bundy et al., 2017; Coombs, 2007). Second, we clarified that reactive CSR will lead to intrinsic attribution of brand benevolence when the strategy is deployed by organizations already engaged in CSR activities. The results suggest that evidence of proactive CSR, in form of CSR brand positioning, resolves potential concerns about brand motives that might otherwise be triggered by reactive CSR (Ricks, 2005; Groza et al., 2011). In this respect, we extend current debates by providing a further boundary condition for the influence of reactive CSR (Rim and Ferguson, 2020; Kim and Choi, 2018).

Our findings also have implications for research on brand hypocrisy (Wagner et al., 2009; Walter et al., 2024). Previous research has shown that brands can be seriously damaged by negative events occurring in the same area in which they have developed a strong reputation for proactive CSR (Baghi and Antonetti, 2021). The contradiction between the brand’s CSR discourse and evidence of a crisis triggers perceptions of hypocrisy, leading to a backlash against the company. In this context, reactive CSR initiatives can effectively insulate a brand from the negative hypocrisy effect. In our experiments, the brands positioned on CSR (who might have been especially punished for a crisis that raised the suspicion of hypocrisy) were significantly helped by the engagement in further reactive CSR initiatives. Our study, therefore, provides evidence of a suitable response for brands accused of hypocrisy regarding their existing CSR engagements.

More broadly, our study contributes to research on crisis communications (Bundy et al., 2017; Coombs, 2007) by suggesting that CSR can be a useful response to a crisis induced by a negative event. The approach tested in this study was similar to that taken in previous investigations focused on compensation or reparation as a quick way to resolve specific grievances or alleviate the harm caused by a crisis (Pfarrer et al., 2008). We have shown, however, that these strategies can be framed as part of a more general CSR effort and communicated to reinforce the brand’s existing and ongoing positioning on CSR. In summary, compensation is a quick, short-lived fix, while reactive CSR is a prolonged effort to restore reputation and foster positive stakeholder relationships. In this respect, this study also contributes to research on CSR as a resource for strengthening a brand’s reputation and social standing (Klein and Dawar, 2004). Our findings have demonstrated that the ability to leverage reactive CSR as an effective strategy in crisis communication depends on the brand’s prior engagement with CSR initiatives.

The study has two important implications for brand managers. First, our results have shown that managers of brands positioned on performance or innovation should avoid specifically linking their CSR initiatives to external pressures or (partial) crisis response. These organizations might benefit more from offering compensation or reparations not linked to CSR but explicitly presented as amends for the crisis (Bundy et al., 2017; Pfarrer et al., 2008). In contrast, brands positioned on CSR can present their post-crisis reparations as an unambiguous continuation of their preexisting CSR initiatives.

The second, broader managerial implication of this study relates to the importance of motive attribution when consumers evaluate CSR initiatives. Managers need to be proactive in managing the attribution of CSR motives, as these play a significant role in shaping whether and to what extent CSR will improve perceptions of the brand. Specifically, the study showed that, in relation to reactive CSR, consumers can infer negative motives based on the impression that reactive CSR is simply a ploy to deal with negative pressures (Ricks, 2005; Groza et al., 2011). Managers, however, can show that the company’s reactive CSR is not simply a set of new initiatives in response to a crisis. They can show that it is the outcome of a longer process in which the organization recognized that such initiatives should have been implemented earlier and is potentially apologizing for having failed to do so. This appears consistent with the example of Patagonia discussed above.

The study has a number of limitations that open interesting avenues for further research. In our three experiments, we have not controlled for the fact that the customer base of sustainability-focused brands might differ substantially from that of brands positioned on performance. There is some evidence that consumers interested in sustainability tend to have more altruistic values and may have different political beliefs (Doran, 2009). While this differentiation might not apply to all products, it might be salient in certain industries. As such, we cannot exclude that it might impact the results. Differences in values or political beliefs might influence the skepticism with which consumer groups evaluate reactive CSR. Further research should examine this point and consider whether certain personal differences across stakeholder groups might impact the results presented here. A further limitation of our study is that we only considered CSR positioning in terms of environmental sustainability. Future studies could test the robustness of these findings in other areas of CSR positioning (for example, human rights or social inclusion).

It is also important to consider that we have compared reactive CSR to an apology but not to a more costly strategy like compensation. Compensation and reactive CSR differ not only in their approach but also in terms of timing and duration (Coombs, 2007, 2020). Compensation is typically immediate and for the short term. It is a direct response to address the specific harm caused by offering those affected a financial settlement, refund or other form of restitution. The goal is to be reimbursed for the specific damages caused. Reactive CSR, by contrast, is often a longer-term strategy. It may begin soon after the crisis, but the efforts made, such as encouraging community involvement or implementing new sustainability practices, are designed to have a lasting impact. Reactive CSR is intended to rebuild trust and improve the company’s perceived benevolence over time, making it an ongoing process that continues well beyond the immediate resolution of the crisis (Groza et al., 2011). Because of these differences, compensation is likely to be less costly than reactive CSR. A relevant empirical question for future consideration is how compensation performs in relation to apology and reactive CSR. We speculate that compensation might be more persuasive than reactive CSR for brands positioned on performance.

In this research, we did not explicitly consider the motivation underpinning the crisis, although research has shown that there are significant differences between crises resulting from accidents and those apparently motivated by greed (Coombs, 2007). Crises that harm people directly and/or appear to consumers to be motivated by corporate greed might be perceived as more severe, and this could influence the relative effectiveness of reactive CSR. Future studies could also investigate the relative effectiveness of reactive CSR for brands that combine a CSR positioning with a unique positioning related to product performance. Most brands tend to have a mixed profile when it comes to CSR and performance, and it would be interesting to explore the relative role to such combinations when it comes to crisis response strategies such as reactive CSR. Such research could provide valuable insights for businesses in navigating challenges, building resilience and creating a positive and impactful brand identity.

The supplementary material for this article can be found online.