Is there any excuse for wrongdoing? The moderating role of perceived reasons for bank irresponsibility in blame appraisal and WOM recommendations

Abstract

Purpose

This study aims to outline the role of causal attributions in consumer responses to irresponsible corporate behaviour. Specifically, this paper presents a moderated mediation model that explains how four types of perceived motives behind an irresponsible action shape corporate blame and word-of-mouth recommendations.

Design/methodology/approach

To test the hypotheses, the study uses data from a large survey assessing consumer reactions to a real case of corporate socially irresponsible behaviour in the banking industry.

Findings

The findings show that market-, unethicality- and rogue employee-driven attributions increase corporate blame and subsequently make people more likely to spread negative comments regarding the culprit. The difficult situation of a bank, as a perceived reason for wrongdoing, does not reduce the blame attributed to the irresponsible organisation.

Originality/value

The literature offers little information on the attributions people make following egregious corporate behaviour; however, such cognitions can play an important role in stakeholders’ reactions to wrongdoing. This study therefore extends the understanding of how irresponsibility attributions affect consumers’ responses to misbehaviour. Given the empirical context, the findings might be particularly important for communication and bank managers.

Citation

Zasuwa, G. (2024), "Is there any excuse for wrongdoing? The moderating role of perceived reasons for bank irresponsibility in blame appraisal and WOM recommendations", Social Responsibility Journal, Vol. 20 No. 5, pp. 845-860. https://doi.org/10.1108/SRJ-04-2023-0245

Introduction

Corporate socially irresponsible actions produce negative effects not only on stakeholders but also on the focal firm. An egregious event undermines trust in the firm (Davies and Olmedo-Cifuentes, 2016) and endangers its good reputation (Nardella et al., 2020). The negative outcomes of corporate social irresponsibility (CSI) do not end with these unfavourable perceptions; they can begin a chain of outrageous responses. The literature indicates that news regarding corporate wrongdoing evokes moral emotions (Xie et al., 2015) that induce subsequent punitive responses, including negative word-of-mouth (WOM; Trautwein and Lindenmeier, 2019), reduced purchase intentions (Septianto et al., 2020) and increased boycotting willingness (Scheidler and Edinger-Schons, 2020).

The responses to a CSI incident are not always severe to the same degree (Davies and Olmedo-Cifuentes, 2016). The literature shows that an appraisal of a potentially irresponsible event is of key importance in consumers’ punitive reactions to corporate wrongdoing. A recent systematic literature review has demonstrated the role of four types of CSI appraisal, namely, severity, (un)ethicality, intentionality and blame (Valor et al., 2022). However, we still know little about the appraisal of the perceived reasons behind a CSI incident, although people have a natural predisposition to seek the motives behind any action they observe (Weiner, 2000). CSR research has clearly shown that the perceived organisational motives behind “doing good” are useful in explaining consumer responses (Story and Neves, 2015). However, little is known about the effects of these perceived reasons for corporate misbehaviour.

To help fill this research gap, this study aims to better understand how the perceived reasons behind the irresponsible actions of an organisation moderate the relationship among consumer awareness of such behaviour, blame and WOM recommendations. Specifically, this research examines the effects of four types of attributions consumers tend to make to corporate culprits, namely, market-, (un)ethicality-, rogue employee- and difficulty-driven attributions. Drawing on attribution theory (Weiner, 2000) and the literature on consumers’ reactions to corporate misbehaviour (Antonetti and Maklan, 2016; Guckian et al., 2018; Yuksel, 2013), this study develops a conceptual model. This proposed model is tested in the context of an instance of irresponsible lending that affected many consumers in Central and Eastern Europe (Olszewska, 2018).

Generally, this paper therefore improves the understanding of the mechanism of consumer responses to CSI incidents (Valor et al., 2022), particularly in the banking industry. Firstly, this study introduces the concept of causal attributions to this field, understood as consumers’ causal explanations for why certain corporate actions have happened (c.f., Weiner, 2000). The results show that the causes assigned to irresponsible behaviour function as boundary conditions that shape consumer responses to CSI news, including negative WOM. Specifically, market-, (un)ethicality- and rogue employee-driven attributions make people more likely to blame the focal corporation (i.e. bank); in turn, these internal attributions yield higher penalties for CSI (i.e. negative WOM).

Secondly, this paper contributes to the literature on crisis communication, as the understating of how people perceive a negative event is of key importance in developing an effective communication strategy (Coombs, 2022). Existing research has revealed that companies that face a scandal tend to shift blame to the other parties involved in a negative event (e.g. employees), although this strategy is considered an unethical way of communication (Kent and Boatwright, 2018). This research thus also corroborates studies showing how a “scapegoating” communication can backfire (Coombs, 2015). That is, the findings show that considering a few rogue employees the cause of corporate wrongdoing does not alleviates organisational blame but can aggravate the attributed responsibility of the focal organisation for this negative event.

Thirdly, by testing the proposed model in the context of the banking industry, this research contributes to an important but scant consumer literature on irresponsible lending (Cherednychenko and Meindertsma, 2019). Specifically, this study suggests that in the financial sector, where frontline employees have to strictly follow internal regulations and procedures, it might be difficult to distinguish between organisational and employees’ responsibility for misbehaviour. However, in other industries, e.g. automobile manufacturing, consumers can distinguish between “a few rogue employees” and a “rotten corporate culture” as the underlying factors for wrongdoing (Guckian et al., 2018).

The remainder of the paper is structured as follows: The next section introduces a conceptual model based on hypotheses. Then, the research method including the measures, the sample and analytical procedure is discussed. The subsequent section presents the findings. The paper concludes with a discussion of these findings’ theoretical implications, the relevant insights for practitioners and some directions for further research.

Conceptual model and research hypotheses

Figure 1 depicts the conceptual model of this research. A process underlining how news regarding irresponsible operations leads to punitive consumer responses is the core component in this model. This model also assumes that awareness of potentially irresponsible operations is necessary but not sufficient to evoke consumer responses to a CSI incident. Thus, this study introduces blame, the “degree to which customers perceive a firm to have directly or indirectly caused a CSI incident” (Grégoire et al., 2010, p. 742), as a mediator between awareness of misbehaviour and individual punitive response. The literature provides substantial support for this conceptualisation. Accordingly, blame predeceases moral emotions, such as anger (Antonetti and Valor, 2021), and thereby makes people more likely to boycott the products of a corporate culprit (Lim and Shim, 2019) or spread negative WOM (Antonetti and Baghi, 2019).

To better explain how CSI may evoke negative consumer responses, this study introduces perceived reasons for corporate wrongdoing (i.e. causal attributions) as boundary conditions that shape the relationship between news regarding irresponsible activities and blame. Following attribution theory (Weiner, 2000), this study assumes that a consumer who encounters information on irresponsible corporate behaviour tries to understand the causes of such actions, as the undermined trust in a firm (Zasuwa, 2018) and the negative valence of an event activate attributional processes (Folkes, 1988).

Overall, people can attribute an event, including wrongdoing, to internal and/or external factors (Kelley, 1967). Internal attributions refer to the specific features of an organisation, whereas external attributions include the potential causes outside the organisation. Given that different organisational factors can result in corporate wrongdoing (Zheng and Chun, 2017), this research distinguishes market ambition and unethical culture as two distinct internal reasons for irresponsibility (Zasuwa and Razmus, 2021). Market-driven attributions involve beliefs that a company engages in CSI to improve its performance and achieve a competitive advantage over other companies in its industry. Seeking profits at any price (i.e. greed) is considered a substantial antecedent to unethical behaviour in the organisational literature (Belle and Cantarelli, 2017). This type of perception, as a reason for CSI, received critical attention during and in the aftermath of the global financial crisis (Kenway and Fahey, 2010). Similarly, marketing research reveals that greed, as an underlying factor behind egregious events, increases the perceived unfairness of a firm’s behaviour (Antonetti and Maklan, 2016). Thus, market-driven attributions might increase blame, as a major tenet of blame is the undesirability of action (Malle et al., 2014). Formally, this study therefore proposes the following hypothesis:

H1.

Market-driven attributions strengthen the relationship among news regarding CSI, blame and negative WOM.

Unethical culture-driven attributions are beliefs that indicate an immoral organisational culture is a reason for corporate misbehaviour. A culture that has a favourable view of opportunistic behaviour can be the cause of CSI incidents, including earnings management, accounting frauds or option backdating (Liu, 2016). Research on VW scandals has revealed that when people attribute egregious behaviour to a rotten corporate culture rather than a few rogue employees, they are more likely to feel anger and tend to expect unethical actions by the focal firm in the future. Negative emotions and distrust reduce brand engagement (Guckian et al., 2018). These adverse responses indicate that when an unethical corporate culture is deemed a factor, the blame given to the perpetrator increases because of the higher intentionality and greater negativity of its corporate actions. Accordingly, taking these arguments together, one can reasonably expect that an unethical culture, perceived as a potential reason for wrongdoing, strengthens the effects of news regarding irresponsibility on corporate blame and the subsequent consumer responses:

H2.

Unethical culture-driven attributions strengthen the relationship among news regarding CSI, blame and negative WOM.

Although employees are members of an organisation, consumers distinguish between the responsibility of a firm and the responsibility of its employees concerning a negative event (Hess et al., 2007). The crisis communication literature reveals that firms tend to exploit this mechanism by shifting blame to their employees (Kent and Boatwright, 2018). Indeed, there might be several situations where a CSI incident is the fault of a person (or a few people) who seeks personal gain at the expense of other parties. Such misconduct could be the result of the actions of a few employees, as well as the consequence of the choices made by rogue executives (Pearce and Manz, 2011). This reasoning implies that regardless of the position rogue employees occupy in their firm, a CSI incident might be perceived as an outcome of the behaviour of a few people rather than an overall organisational problem. Indeed, when stakeholders believe that certain employees have caused a CSI incident, they are less likely to respond negatively to the focal firm (Guckian et al., 2018). Logically, when irresponsible behaviour is considered the fault of a few people, consumers are less likely to attribute blame to the focal firm and thus less likely to respond harshly to it (e.g. less likely to engage in negative WOM). Thus, the following hypothesis is proposed:

H3.

Rogue employee-driven attributions weaken the relationship among news regarding CSI, blame and negative WOM.

In addition to internal forces, external factors can result in irresponsible incidents and behaviours. External pressures, for example, pressure from headquarters to subsidiaries in developing countries (Surroca et al., 2013), financial stress due to financial loss and negative cash flow over the long term (Zheng and Chun, 2017), market cost pressure (Wu, 2014) or a higher-order backlog (Sun et al., 2019), positively correlate with CSI. Accordingly, preliminary research suggests that people can consider such external pressures and, consequently, the difficulties a firm faces as a potential reason for its misbehaviour.

Attribution theory indicates that when people consider external factors reasons for egregious behaviour, they are less likely to blame the focal actor(s) for an incident (Weiner, 1995). Consequently, the literature on ethical consumerism reveals that people are less likely to participate in a boycott when they believe external causes have resulted in a CSI incident (Yuksel, 2013). Thus, attributing a negative behaviour to external forces reduces the organisational responsibility for wrongdoing and makes consumers less likely to engage in punitive response. Thus, this study proposes the following hypothesis:

H4.

Difficulty-driven attributions weaken the relationship among news regarding CSI, blame and negative WOM.

Method

Procedure and sample

This study examined the proposed model via a field study that surveyed consumers’ perceptions and reactions to CSI in the banking industry. The specific empirical context is a Swiss franc mortgage crisis involving the largest retail banks in Poland. This situation originated in the 2000s, when banks, in addition to national currency mortgages, tended to offer cheaper CHF loans (Buszko, 2016). When the value of the Swiss franc increased as a result of the global financial crisis, the exchange rate risk materialised. As a consequence, thousands of borrowers saw their debts and instalments dramatically rise. This crisis intensified again in September 2022 when the Court of Justice of the European Union decided in favour of CHF borrowers (Businessinsider, 2022). The ongoing battle between customers and banks has received extensive public attention. The media has often condemned banks for using prohibited spread clauses and profiting at the expense of customers. Borrowers have felt cheated and complained of not being provided with adequate information on the exchange risk (GazetaPrawna, 2018). Thousands of borrowers have taken legal action against the banks. The vast majority of these cases have been ruled in favour of the banks’ clients (Rzeczpospolita, 2020). Thus, following the financial literature (Zunzunegui, 2018), this study considers the practice of offering CHF loans involving illegal clauses and unethical procedures a case of irresponsible corporate behaviour.

The data were collected in December 2022, when the problem of FCL loans in Poland was present in the public debate. This long-known issue was raised by the Court of Justice of the European Union, which judged in favour of a few consumers who accused the focal banks of using unfair terms in CHF mortgage loans [E.K., S.K. v D.B.P. (C-80/21), and B.S., W.S. v M. (C-81/21), and B.S., Ł.S. v M. (C-82/21), 2022].

Specifically, the research was conducted on 1,000 Polish bank customers. Technically, these respondents were recruited online from a nationwide consumer panel by a research agency (BIOSTAT). To select the sample, the quota sampling method was used with quotas reflecting the entire adult Polish population in terms of gender, age, education and a current CHF mortgage. Specifically, the sample included 54.2% females. The average age was 42.4 years, with a standard deviation of 14.1. Respondents with a secondary education accounted for 44.6% of the sample, followed by graduates and undergraduates (41.1%) and people with vocational schooling (11.7) or less (2.6). Two hundred participants had a mortgage, including a Swiss franc loan (100 respondents) or a domestic loan (100 respondents). There was an overrepresentation of people with a CHF mortgage in the sample compared to the general population (10% vs 8%).

To ensure ecological validity, we developed a brief description, drawing on actual court cases and financial penalties imposed on several banks by the President of the Office of Competition and Consumer Protection (UOKIK, 2021). At the beginning of the survey, respondents were asked to read a narrative of these. Then, they completed the scales that measured, inter alia, causal attributions, blame and negative WOM intentions.

Measures

Three items, derived from the scale of Klein and Dawar (2004) measured blame appraisal. Specifically, this scale assessed bank responsibility for the difficulties that CHF mortgage holders face. To check if a respondent had knowledge about the involvement of their primary bank in irresponsible lending, the survey asked whether the focal bank was mentioned in the excerpt among the financial institutions penalised for using unfair terms in Swiss franc loans (yes/no coding).

Causal attributions were measured by assessing banks’ perceived reasons for offering CHF mortgages. Specifically, these scales included four sets of items measuring market-, unethical culture-, difficulty- and rogue employee-driven attributions. The first three groups of items were drawn from studies involving exploratory research (Zasuwa, 2023; Zasuwa and Razmus, 2021). The adaptation of items involved changing the type of organisation from a company to a bank and mentioning the problem under investigation. Given that frontline employees can participate in mis-selling financial services (Brannan, 2017), a measure of employee-driven attributions was added. This measure was based on research showing that consumers can attribute a CSI incident to rogue employees (Guckian et al., 2018). For all the attribution items, a 5-point Likert scale from strongly disagree (1) to strongly agree (5) was used as a format for measurement. The scale for negative WOM was adapted from Grappi et al. (2013). In addition, the questionnaire measured the perceived irresponsibility of offering CHF mortgage loans in an analogous scale to that of Reidenbach and Robin (1990), similar to other CSI research (Trautwein and Lindenmeier, 2019).

Confirmatory factor analysis was conducted to verify the measures. The outcomes of this procedure revealed that all the items significantly loaded (Table 1). Commonly used fit criteria (Schumacker and Lomax, 2004), such as normed chi-square (CMIN/DF), comparative fit index (CFI) and the root-mean-square error of approximation (RMSEA), revealed a good fit of the measurement model. Specifically: χ² (254) = 840.52, p < 0.001, CMIN/DF = 3.303, CFI = 0.974, RMSEA = 0.048 with 90% confidence interval (CI) = 0.044–0.052 p > 0.05.

With regard to convergent validity, the average variance extracted (AVE) for every construct exceeded 0.50, and all composite reliability (CR) values were greater than 0.70. Concerning discriminant validity, Table 2 shows that the square roots of the AVEs of all the constructs were greater than the correlations among them, providing support for their independence based on the Fornell–Lacker criterion (Fornell and Larcker, 1981). Similarly, the highest heterotrait–monotrait (HTMT) ratio of correlations amounted to 0.581 (Table 3), showing there were no concerns regarding discriminant validity (Henseler et al., 2015).

Results

To test the mediating role of blame in the relationship among consumers’ knowledge of a bank’s use of unfair terms in mortgages and consumers’ negative WOM intention, this study applied Model 4 from Hayes’ Process macro in SPSS 4 (Hayes, 2018). Irresponsibility appraisal of a bank offering CHF loans and a consumer holding a CHF loan were two control variables introduced into the analysis, as the egregiousness of the act (Hartmann and Moeller, 2014) and the personal relevance of the action (Haberstroh et al., 2017) are significant antecedents of consumers’ responses to corporate wrongdoing. The findings show that knowledge about irresponsible lending affects negative WOM through the blame attributed to the bank (IE = 0.294; 95% CI: 0.221–0.377). Specifically, information on CSI has had a positive impact on blame (b = 1.151; p < 0.001), which subsequently affects WOM intentions (b = 0.255; p < 0.001). The direct effect of information on the involvement of a bank in irresponsible lending is close to zero and non-significant (b = 0.01; p > 0.05), demonstrating full mediation. Consistent with the hypotheses, irresponsibility appraisal and having a CHF mortgage are significant predictors of blame being attributed to a bank (respectively, b = 0.601; p < 0.01; b = 0.801; p < 0.001) and of negative WOM (respectively, b = 0.274; p < 0.001; b = 0.627; p < 0.001). Table 4 reports the detailed outcomes of this mediation analysis.

After testing the mediation mechanism, the study examined the salient moderation effects (Table 5). To perform this analysis, Model 7 of Hayes’ Process macro was used (Hayes, 2018). According to H1, market-driven attributions strengthen the effects of knowledge regarding the imposition of a financial penalty on the focal bank for using unfair terms in CHF mortgages on blame appraisal (b = 0.274; p < 0.05). Correspondingly, slope analysis [Figure 2(a)] shows that the line for a high level of perceived market-driven attributions is much steeper than a low level. In addition, the 95% CI for the index of moderated mediation (MMI) is significant (MMI = 0.070, 95% CI: 0.006–0.139), demonstrating that perceived market motivation, as a reason for irresponsibility, can indirectly affect negative WOM as well.

H2 posits that unethical culture-driven attributions strengthen the relationship among news regarding CSI, blame and negative WOM. In line with this hypothesis, the results reveal that unethical culture-driven attributions significantly interact with CSI awareness when shaping blame (b = 0.317; p < 0.01). Accordingly, Figure 2(b) displays a much steeper line for high than for low unethical culture-driven attributions. Further findings indicate that perceiving unethical culture to be an underlying factor for irresponsibility also indirectly affects WOM intentions (MMI = 0.081, 95% CI: 0.027–0.136).

However, contrary to H3, perceiving rogue employees as a potential cause of irresponsibility does not weaken but strengthens the relationship among news regarding CSI, blame and negative WOM. Specifically, regression analysis reveals that rogue employee-driven attributions directly (b = 0.216; p < 0.001), by interacting with knowledge regarding CSI, increase blame (b = 0.203; p < 0.05). Thus, the slope for a high level of rogue employee-driven attributions is steeper than the slope for a low level [Figure 2(c)]. Moreover, the 95% CI for MMI excludes zero, showing that the positive effects of rogue employees’ driven attributions can indirectly affect WOM intentions (MMI = 0.052, 95% CI: 0.024–0.101).

Regarding the moderating role of difficulty-driven attributions in the relationship among CSI knowledge, blame and negative WOM, the findings have shown that the interaction term between this type of attribution and CSI knowledge is negative, as predicted, but statistically non-significant (b = −0.025; p > 0.05). Accordingly, the MMI is also non-significant (MMI = −0.006, 95% CI: −0.054 to 0.043). Thus, the results of this field study provide no support for H4.

Discussion

General discussion

The aim of this study was to examine how causal attributions for CSI affect consumer responses to corporate misbehaviour, particularly blame and WOM intentions. The results of the above field research on the banking industry clearly show that the perceived motives behind corporate wrongdoing play an important role in shaping such reactions. In line with attribution theory (Weiner, 1995), the study shows that two internal types of causal attributions aggravate consumer outrage. The revealed moderating effects of market-driven attributions concur with the literature, showing that greed increases consumer unfairness (Antonetti and Maklan, 2016). The findings regarding the positive impact of unethical culture-driven attributions on blame and negative WOM extend the findings of Guckian et al. (2018), who show that perceiving a rotten corporate culture to be the reason for wrongdoing increases consumers’ anger and distrust in the focal firm.

Although this study has demonstrated the importance of the perceived motives behind CSI, not all the hypotheses have been supported. Contrary to what was expected and research on Dieselgate’s perceptions (Guckian et al., 2018), the findings suggest that consumers face difficulties in discerning between the focal organisation’s and its employees’ responsibility for bad behaviour. This effect may be attributed to the empirical context of the banking industry, given its relatively rigid protocols and the minimal decision-making freedom of its frontline employees. Thus, consumers might be less likely to believe that a group of rogue employees could bear sole responsibility for the egregious actions of a bank. Hence, rogue employee-driven attributions do not weaken but strengthen the blame towards a bank.

Regarding the effects of difficulty-driven attributions, the findings suggest that in the banking industry, perceived organisational distress, as an underlying factor for wrongdoing, does not reduce the blame attributed to the corporate culprit. A possible cause of this could be a lack of trust in banks and the institutions that oversee them (van der Cruijsen et al., 2016). At the time the survey was performed, the Polish banking industry was riddled with turmoil, including numerous unfavourable legal cases, which could have led the participants to doubt that banks’ irresponsible behaviour (i.e. using unfair terms in mortgage agreements) was caused by external pressures.

Theoretical implications

This paper offers a twofold contribution to the CSI and consumer literature. Firstly, it advances the understanding of the antecedents of blame attributed to a firm engaging in CSI when culpability is ambiguous. Studies have shown that perception of transgression severity (Laufer et al., 2005), wrongdoers’ intentionality (Pizzetti et al., 2021) and controllability over a CSI incident (Laufer et al., 2009), for example, impact the negative responses to a firm following egregious behaviour. This study thus extends the consumer literature by demonstrating how causal attributions of corporate misbehaviour affect blame attributed to the focal organisation. This knowledge is of high importance, as blame determines not only negative WOM recommendations but also other consumer reactions to CSI incidents, including moral emotions (Antonetti, 2020), boycotting intentions (Lim and Shim, 2019) and firm evaluations (Grappi et al., 2017).

A further contribution to the CSI literature is the finding that the blame attributed to a firm fully mediates the relationship between penalty awareness and negative WOM. CSI research, dominated by experimental approaches, has typically provided respondents with information on corporate wrongdoing, allowing them to easily identify the punitive responses to corporate culprits (Valor et al., 2022). However, in the real world, punitive stakeholder responses to companies are less evident (Nardella et al., 2023). Even a company found legally guilty can avoid social penalties (Nardella et al., 2020). The identified mediating effects of blame attributed to the focal firm thus improve the understanding of this paradox. Specifically, the findings suggest that mere knowledge that a particular firm has been legally punished for misbehaviour is not sufficient to evoke consumers’ negative responses to that firm. The significant mediating effect of blame reveals that consumers can punish a firm, provided they personally consider the firm guilty. Thus, subjective blame attribution might be the missing variable responsible for the lack of social penalty when a firm is legally guilty.

Practical implications

In addition to the above theoretical implications, the findings are useful for practitioners, particularly in the banking industry. The findings suggest that a bank that faces a responsibility crisis may have more difficulty defending its reputation than other organisations. The positive effects of rogue employee-driven attributions on organisational blame suggest that fostering public opinion regarding the egregious behaviour of certain employees as the cause of misbehaviour might even backfire and aggravate outrage responses. Clearly, the specific communication context and content might enhance the effectiveness of blame-giving campaigns (Antonetti and Baghi, 2019). However, this tactic may be highly risky in the banking industry. Similarly, the difficult situation of a bank cannot be considered an excuse for its wrongdoing. Although society requires firms to be profitable (Carroll, 1991) and people who have higher economic expectations of firms tend to less harshly assess corporate misbehaviour (Kim et al., 2019), the results of this study suggest that banks cannot justify their irresponsible operations by simply referring to their economic function in society.

Limitations and directions for future research

The study also has some limitations that should be considered. Firstly, this research only focuses on the effects of causal attributions on blame and negative WOM, as outcomes of CSI incidents. However, the findings evidence suggest that corporate misbehaviour can evoke a wide range of stakeholder responses. Thus, future research might address additional reactions to CSI, including moral emotions (Antonetti, 2020), consumer boycotts (Albrecht et al., 2013), negative media coverage (Fu, 2023) or damage to organisational reputation (Nardella et al., 2023).

In addition, the study examined the effects of perceived reasons for misbehaviour regarding a specific type of irresponsibility (i.e. irresponsible lending). This specific research context has allowed a better understanding of consumer responses to irresponsibility in an underexamined field (i.e. the banking industry); however, this context limits the generalisability of the findings. Thus, further research is needed to test how casual attributions affect consumer responses to wrongdoing in other industries, e.g. retail, hospitality or technology. Similarly, more studies are invited to test the proposed model using a wider catalogue of corporate transgressions, as different types of misconduct might exert different effects on stakeholders’ trust in firms (Davies and Olmedo-Cifuentes, 2016).

Conclusion

In conclusion, this paper provides valuable insights regarding attribution theory’s relevance for understanding consumers’ responses to CSI incidents. The results of the study suggest that companies, especially banks, should be aware of these various attributions and how they can affect consumer reactions. Overall, this paper suggests that the causal attributions people typically make following the occurrence of a bank’s irresponsible behaviour can increase consumers’ outrage. Thus, communication efforts aimed at justifying a bank’s irresponsible behaviour are likely to backfire.

Figures

Figure 1

Conceptual model

Figure 2

Moderating effects of causal attributions on blame

Table 1

Factor loadings, CR and AVE for constructs in a measurement model

Variable	Factor loading	CR	AVE
Negative WOM		0.936	0.786
I intend to say negative things about “my bank” to friends and relatives	0.844
I intend to recommend to my friends and relatives that they not use products of “my bank”	0.941
I intend to discredit “my bank” to my friends and relatives	0.825
I intend to suggest to my friends and relatives not to buy the products of “my bank”	0.930
Blame		0.934	0.824
“My bank” is to blame for the difficulties of Swiss franc borrowers	0.935
The difficult situation of Swiss franc borrowers is entirely the fault of “my bank”	0.925
To what extent is “your bank” responsible for the difficulties of Swiss franc borrowers?	0.862
Market-driven attributions (CHF mortgage loans were offered because…)		0.867	0.635
banks wanted to achieve better financial performance	0.416
banks wanted to become more competitive	0.898
banks wanted to gain an edge over competitors	0.894
banks wanted to improve their competitive position	0.872
Unethicality-driven attributions (CHF mortgage loans were offered because…)		0.904	0.760
ethical issues were treated as irrelevant in banks	0.829
banks permitted for unethical behaviour	0.941
banks had an unethical organisational culture	0.841
Difficulty-driven attributions (CHF mortgage loans were offered because…)		0.951	0.867
banks were in a difficult situation	0.942
banks wanted to defend themselves against difficulties	0.930
difficult situation must have prompted banks to do it	0.921
Rogue employee-driven attributions (CHF mortgage loans were offered because…)		0.963	0.868
some employees in banks were greedy	0.908
some employees in banks acted unethically	0.945
some employees in banks acted dishonestly	0.947
some employees in banks were immoral	0.926
Irresponsibility appraisal (Offering mortgage loans in CHF was…)		0.926	0.758
unfair	0.877
harmful	0.863
dishonest	0.890
disloyal	0.851

Notes:

CR = composite reliability; AVE = average variance extracted

Source: Own research

Table 2

Means, standard deviations, Cronbach’s alpha, correlations and the square roots for AVE

Variable	Mean	SD	Alpha	F1	F2	F3	F4	F5	F6	F7
Market-driven attributions (F1)	3.85	0.73	0.843	0.797
Unethicality-driven attributions (F2)	3.51	0.91	0.90	0.152***	0.872
Rogue employee-driven (F3)	3.37	1.01	0.963	0.138***	0.608***	0.932
Difficulty-driven attributions (F4)	2.89	1.06	0.951	0.068*	−0.042	0.051	0.931
Blame (F5)	3.87	1.59	0.933	0.079*	0.296***	0.279***	0.131***	0.908
Negative WOM (F6)	2.72	1.12	0.935	−0.064†	0.309***	0.259***	0.060†	0.454***	0.887
Irresponsibility appraisal (F7)	3.51	0.94	0.926	0.077*	0.423***	0.372***	0.113***	0.377***	0.234***	0.871

Notes:

*, *** and †denote statistical significance at p < 0.05, p < 0.001 and p < 0.10, respectively

Source: Own research

Table 3

HTMT analysis

Variable	F1	F2	F3	F4	F5	F6	F7
Market-driven attributions (F1)
Unethicality-driven attributions (F2)	0.222
Rogue employee-driven attributions (F3)	0.188	0.581
Difficulty-driven attributions (F4)	0.015	0.035	0.054
Blame (F5)	0.104	0.275	0.264	0.117
Negative WOM (F6)	0.025	0.294	0.268	0.053	0.422
Irresponsibility appraisal (F7)	0.121	0.405	0.351	0.105	0.342	0.229

Source: Own research

Table 4

Mediation analysis

Variable	Coefficient	SE	t	p	95% CI
					LL	UL
Dependent variable: Blame
Constant	−0.887	0.254	−3.49	0.001	−1.387	−0.388
CSI awareness	1.151	0.086	13.336	<0.001	0.982	1.321
Irresponsibility appraisal	0.601	0.045	13.261	<0.001	0.512	0.69
Having a CHF mortgage	0.81	0.144	5.643	<0.001	0.529	1.092
R2 = 0.284, F(3, 996) = 132.0, p < 0.001
Dependent variable: negative WOM
Constant	0.83	0.191	4.349	<0.001	0.455	1.204
CSI awareness	0.01	0.07	0.144	0.886	−0.127	0.147
Blame	0.255	0.024	10.813	<0.001	0.209	0.302
Irresponsibility appraisal	0.121	0.037	3.291	0.001	0.049	0.193
Having a CHF mortgage	0.42	0.109	3.859	<0.001	0.206	0.633
R2 = 0.198, F(4, 995) = 61.6, p < 0.001
Dependent variable: negative WOM
Constant	0.603	0.2	3.01	0.003	0.21	0.996
CSI awareness	0.304	0.068	4.472	<0.001	0.171	0.437
Irresponsibility appraisal	0.274	0.036	7.679	<0.001	0.204	0.344
Having a CHF mortgage	0.627	0.113	5.54	<0.001	0.405	0.849
R2 = 0.104, F(3, 996) = 38.6, p < 0.001
Total. direct and indirect effects
Total effect	0.304	0.068	4.472	<0.001	0.171	0.437
Direct effect	0.01	0.07	0.144	0.886	−0.127	0.147
Indirect effect	0.294	0.039			0.221	0.377

Note:

5,000 bootstrap samples for percentile bootstrap confidence intervals

Source: Own research

Table 5

Moderated mediation analyses

Variable	Coefficient	SE	t	p	95% CI
					LL	UL
Dependent variable: Blame, R2 = 0.305, F(5, 994) = 87.3, p < 0.001
Constant	1.258	0.234	5.375	<0.001	0.799	1.718
CSI awareness	1.148	0.085	13.477	<0.001	0.98	1.315
Rogue employee-driven attr.	0.216	0.045	4.812	<0.001	0.128	0.303
Interaction term	0.203	0.084	2.426	0.015	0.039	0.368
Irresponsibility appraisal	0.517	0.048	10.83	<0.001	0.423	0.611
Having a CHF mortgage	0.722	0.143	5.064	<0.001	0.442	1.002
Conditional effects of the CSI awareness at values of the rogue employee-driven attributions
−1.011	0.942	0.12	7.859	<0.001	0.707	1.177
0	1.148	0.085	13.477	<0.001	0.98	1.315
1.011	1.353	0.121	11.23	<0.001	1.117	1.59
Dependant variable: Blame, R2 = 0.311, F(5, 994) = 89.5, p < 0.001
Constant	1.291	0.236	5.461	<0.001	0.827	1.755
CSI awareness	1.156	0.085	13.626	<0.001	0.989	1.322
Unethical culture-d attr.	0.25	0.051	4.933	<0.001	0.15	0.349
Interaction term	0.317	0.092	3.437	0.001	0.136	0.499
Irresponsibility appraisal	0.507	0.049	10.401	<0.001	0.411	0.602
Having a CHF mortgage	0.728	0.142	5.133	<0.001	0.45	1.006
Conditional effects of the CSI awareness at values of the unethical culture-driven attributions
−0.913	0.866	0.12	7.245	<0.001	0.631	1.1
0	1.156	0.085	13.626	<0.001	0.989	1.322
0.913	1.446	0.12	12.074	<0.001	1.211	1.681
Dependant variable: Blame, R2 = 0.289, F(5, 994) = 80.9, p < 0.001
Constant	0.871	0.226	3.859	<0.001	0.428	1.314
CSI awareness	1.145	0.086	13.269	<0.001	0.976	1.315
Market-driven attributions	0.066	0.059	1.117	0.264	−0.05	0.182
Interaction term	0.274	0.118	2.326	0.02	0.043	0.505
Irresponsibility appraisal	0.597	0.046	13.099	<0.001	0.508	0.687
Having a CHF mortgage	0.813	0.143	5.672	<0.001	0.532	1.094
Conditional effects of the CSI awareness at values of the market-driven attributions
−0.728	0.946	0.121	7.802	<0.001	0.708	1.184
0	1.145	0.086	13.269	<0.001	0.976	1.315
0.728	1.345	0.122	11.017	<0.001	1.105	1.584
Dependent variable: Blame, R2 = 0.289, F(5, 994) = 80.9, p < 0.001
Constant	0.941	0.226	4.16	<0.001	0.497	1.385
CSI awareness	1.153	0.086	13.39	<0.001	0.984	1.322
Difficulty-driven attributions	0.105	0.041	2.579	0.01	0.025	0.185
Interaction term	−0.025	0.081	−0.309	0.757	−0.185	0.134
Irresponsibility appraisal	0.589	0.045	12.962	<0.001	0.5	0.678
Having a CHF mortgage	0.78	0.144	5.422	<0.001	0.498	1.063

Source: Own research