Understanding damage to and reparation of brand trust: a closer look at image congruity in the context of negative publicity

Purpose – Studies have shown that negative publicity adversely affects brand trust, but exactly how brand trust can be damaged remains poorly understood. This study aims to explore how negative publicity in ﬂ uences image congruity and, subsequently, brand trust. In addition, the study also examined the effectiveness of two corporate strategies to repair both congruity and trust. Design/methodology/approach – Based on a valid sample of 522 Chinese consumers between the ages 20 and 50, this study adopted a quasi-experimental design involving two types of negative publicity (performance-and value-related) and two initial corporate repair strategies (compensation and public apology) intended to repair brand trust. Findings – Negative publicity shaped brand trust through both functional congruity and self-congruity. Moreover, the type of negative publicity affected the role of image congruity in brand trust. The effectiveness of repair strategies further depended on the type of negative publicity. Research limitations/implications – Mobile phones were an appropriate focal product for this research, but examining only one product category may limit ﬁ ndings ’ generalizability. Negative emotions such as frustration or anger and their relationships with congruity can also be addressed in future work. Subsequent research can additionally consider more conditions to explore alternative routes of processing related to brand trust. Practical implications – Brand trust is a vulnerable brand asset on which negative publicity can have seriously negative consequences. Marketers and brand managers should assess the extent to which negative publicity can damage image congruity and brand trust and come up with different repair strategies subsequently. Originality/value – This study contributes to the limited and fragmented literature on consumers ’ evaluations of negative information. Findings offer fresh insight into the impacts of negative publicity on image congruity and brand trust. The implications extend beyond negative publicity to other forms of negative information, such as rumors, fake news and negative word of mouth. Results also highlight the importance of adopting appropriate repair strategies to restore consumers ’ trust in the event of negative publicity.


Introduction
Brands are often exposed to negative information in today's marketplace (Greyser, 2009).The Chinese Bureau of Consumer Protection (BCP) received more than 800,000 formal complaints in the year 2019 alone, most of which pertained to product and service brands (BCP, 2020).Other countries have reported similar trends.For example, between 2018 and 2020, the Australian Financial Complaint Authority (AFCA) fielded over 150,000 formal complaints; each involved one or more financial service brands (Australian Financial Complaint Authority [AFCA], 2021).Negative information, including fake news, has thus garnered growing attention in the marketing field (Yu et al., 2018).
The current issue and full text archive of this journal is available on Emerald Insight at: https://www.emerald.com/insight/1061-0421.htmMuch of the related research has shown that negative information adversely affects key brand assets or equities (Beldad et al., 2018), including brand trust (Hegner et al., 2016), brand image (Yu et al., 2019), brand attitude (Lerro et al., 2019) and brand loyalty (Lei and Lemmink, 2008).Brand trust is susceptible to the consequences of negative information (Bhandari and Rodgers, 2018).Despite this vulnerability, few studies (Hegner et al., 2016) have examined how negative information shapes brand trust.The mechanisms by which such information influences brand trust is consequently unclear.The present study attempts to bridge this gap by empirically investigating the role of negative information in brand trust through the mediation of image congruity.
Image congruity refers to the alignment (or lack thereof) between a consumer and an object (He and Mukherjee, 2007).Brand image congruity reflects the end state of a consumer's identification process when interacting with a brand (i.e.responding to an advertisement or making a purchase decision) (Liu et al., 2012).Numerous studies have explored brand image congruity (Hu et al., 2012;Mazodier and Merunka, 2012;Sop and Kozak, 2019).Functional congruity and self-congruity are likely the two most frequently considered concepts in this domain (Dean, 2004;Hegner et al., 2018).The majority of image congruity studies revolve around positive information (e.g.advertisements) that are conducive to positive image congruity.Far less research has assessed image congruity in a negative information context.
When a brand encounters negative information, an appropriate repair strategy can help the brand recover and rebuild brand trust (Dutta and Pullig, 2011).In other words, brands' mistakes can leave customers disappointed if mishandled.For example, Tesla owners have lamented brake failures in China (Tencent Net, 2021).Despite such safety incidents being possible, the company has yet to admit any wrongdoing and has refused to compromise, even going so far as to blame others for its missteps.Tesla's attitude has generated a spate of negative reactions online, leading to deep distrust of the brand among consumers with an accompanying decline in vehicle sales.
It is accordingly important to understand the roles of repair strategies in the relationships among negative information, image congruity and brand trust.Another contribution of the current study hence involves the effectiveness of repair strategies, which appears to have been sparsely addressed elsewhere.
Specifically, this study is guided by three main objectives: to investigate: 1 whether different types of negative publicity affect brand trust; 2 the potential mediating role of brand image congruity if negative publicity indeed influences brand trust; and 3 the effectiveness of repair strategies (i.e.compensation or a public apology), namely, how the potential interaction between negative publicity and repair strategies can shape brand trust.
Findings will clarify how negative publicity affects image congruity and, subsequently, brand trust.It is hoped that results from the current study also contextualize the effectiveness of repair strategies for different forms of negative publicity.

Negative publicity
Negative brand information can take several forms.Common types include rumors, negative word of mouth (WOM) and negative publicity (Andrews and Shimp, 2019).Negative publicity is defined as the "non-compensated dissemination" of negative information through a major medium (e.g.newspapers, radio broadcasts or television) that can potentially damage "a product, a service, business unit, or individual" (Sherrell et al., 1985).Compared with rumors and negative WOM, authorities typically verify the information about a negative publicity event.
Another key distinction is that, in a negative publicity event, information is shared through an expansive medium (e.g.television or a popular social media platform).The information is therefore widespread and reaches many consumers.Negative publicity is generally thought to be highly credible, although this supposition has not been empirically tested; thus, the impact of such publicity on consumers may exceed that of rumors or negative WOM (Bond and Kirshenbaum, 1998).Two primary types of negative publicity have been identified: performance-related and value-related (Beldad et al., 2018;Dean, 2004;Hegner et al., 2018).In performance-related negative publicity, a brand faces negative information related to functional brand aspects, such as quality.Value-related negative publicity features negative information about corporate values, business practices or social or ethical issues (Xie and Peng, 2010).A number of studies (Beldad et al., 2018;Liu and Sweeney, 2011;Pullig et al., 2006) have indicated that certain types of negative publicity can differentially affect brand assets, including brand trust.

Brand trust
Brand trust is a vital brand asset, as trust is one pillar of a sustainable consumer-brand relationship (Hegner and Jevons, 2016).Brand trust has been described in various ways but generally represents an average consumer's willingness to rely on a brand to fulfill its stated function (Chaudhuri and Holbrook, 2001).Hegner and Jevons (2016) suggested that trust can especially reflect perceptions of competence, predictability, benevolence and integrity.The effects of brand trust on other brand assets have been well established in the marketing literature: such trust can reduce consumers' risk perceptions (Song et al., 2012), enhance their purchase intentions (Doney and Cannon, 1997), cultivate brand loyalty (Mabkhot et al., 2017;Naggar and Bendary, 2019), and ultimately engender greater brand equity (Sharma and Jain, 2019).Despite research documenting what brand trust brings to a brand, most scholars have pondered brand trust in relation to positive information such as advertisements or sponsorships (Mazodier and Merunka, 2012).The underlying process through which negative information may affect brand trust remains unknown.
Negative information, such as negative publicity, may mold brand trust through a cognitive process as elucidated by expectancy disconfirmation theory (Oliver, 1980).Consumers hold certain expectations about how a brand should behave.Upon exposure to negative information, consumers may determine that a brand's behavior fails to match their expectations, thereby destroying trust in the brand (Hegner et al., 2016;Lerro et al., 2019).Beldad et al. (2018) pointed out that performance-related negative publicity seemed to significantly affect ability-based trust, whereas value-related incidents more strongly affected benevolence-based trust.The following hypothesis is thus put forth: H1. Brand trust will be negatively influenced by both performance-related (H1a) and value-related (H1b) negative publicity.

Brand image congruity
The relationship between brand image and trust has been scrutinized in the branding domain (Song, 2019).A number of studies have examined external information and brand trust (Mabkhot et al., 2017;Sharma and Jain, 2019) and found that positive information (e.g.advertising) might foster a positive brand image and, in turn, enhance brand trust.To expand this line of reasoning, the present study will explore the influence of brand image on trust from an image congruity perspective.Brand image congruity is often classified as either functional congruity or self-congruity.Functional congruity refers to the extent to which a brand's functional attributes (e.g.quality) meet consumers' anticipated product or brand performance (Kressmann et al., 2006).Functional congruity has been found to promote positive brand evaluations (e.g.brand trust or brand attitude) in multiple respects, such as in terms of products (Sirgy and Su, 2000) and tourism (Ahn et al., 2013).
Different from functional congruity, self-congruity embodies the compatibility between the image of a brand and that of a consumer (a notion also known as "consumer self-concept") (Sirgy et al., 1991).When a consumer selects a brand, they usually undergo an identification process in which they construct a self-image and display symbolic identities (Tasaki, 1992).Numerous studies have demonstrated that selfcongruity positively influences nearly all aspects of a brand, such as brand trust, brand attitude and brand loyalty (Kressmann et al., 2006;Sirgy, 1985;Sirgy et al., 1997aSirgy et al., , 1997b;;Wang et al., 2016;Yu et al., 2019).
Overall, some studies (Aaker, 1999;Kressmann et al., 2006;Sirgy et al., 1991) have examined the impacts of functional congruity and self-congruity on brand evaluations in a positive information context.If positive information influences selfcongruity or functional congruity, then it is reasonable to infer that negative information may also affect the congruity process but in a negative way.It is, therefore, novel and salient to investigate whether negative information influences either or both forms of congruity and, if so, to what extent.This discussion gives way to the following hypothesis: H2. Brand image congruity mediates the influence of negative publicity on brand trust.Specifically, the impact of performance-related negative publicity on brand trust will be mediated by functional congruity (H2a); the impact of value-related negative publicity on brand trust will be mediated by self-congruity (H2b).

Repair strategies
Proper handling of brand crises is critical to brands' survival due to spillover effects (Han et al., 2020;Robson and Farquhar, 2021).For instance, a brand crisis can spread to subbrands in the same product category, even if these trickle-down effects cover subbrands that are otherwise innocent.Businesses often adopt two approaches in crisis situations: diminishing and rebuilding (Claeys et al., 2010).Diminishing involves seeking to curb the severity of negative publicity and to lessen the affected brand's responsibility for the negative events.This tactic is rarely well received by consumers, as it can insinuate that the brand is attempting to evade responsibility.Rebuilding is often viewed as more proactive because it aims to restore consumers' trust in the affected brand (Dutta and Pullig, 2011).Compensation and a public apology have been identified as integral repair strategies in brand rebuilding (Dutta and Pullig, 2011).Based on situational crisis communication theory, Coombs and Holladay (2009) discerned that the effectiveness of rebuilding repair strategies' utility varied with the type of negative information.For instance, a few studies (Vlachos et al., 2009) have identified compensation (i.e.restitution for consumers' losses) as effective in mitigating negative publicity or a crisis.The act of public apology, in which a chief exective officer (CEO) or chief spokesperson makes an official apology on a major medium, is another standard repair tactic when a firm first encounters negative publicity.Hosmer (1995) claimed that public apologies should also be helpful for restoring brand trust because they demonstrate that a brand or business has owned up to its mistakes.Cremer (2010) observed that making a public apology may cause consumers to feel that the affected brands are conscientious and responsible-two attributes conducive to restoring a positive brand image.
Researchers have rarely compared the effectiveness of these two strategies.Compensation, which directly addresses consumers' functional value expectations, may be appropriate for repairing brand trust in cases of performance-related negative information (Snyder and Honig, 2016).Angie (2018) took a similar view and asserted that compensation can feasibly repair functional (or instrumental) consumer-brand connections.Public apologies represent pro-social behavior and might be more effective when dealing with value-related negative publicity (Hegner et al., 2016).This form of negative publicity may not be directly tied to product attributes (e.g.product quality); as such, customers might be inclined to positively respond to a crisis communication approach (i.e.apology) that explains the incident (Dutta and Pullig, 2011).Take, for example, an energy brand facing negative publicity due to greenwashing: the brand regained consumers' trust by apologizing publicly (Guo et al., 2018).
Drawing upon previous work (Jin and Lee, 2019), it has been argued that performance-related negative publicity will damage consumers' functional image congruity more than the valuerelated negative publicity.Compensation may, therefore, be more appropriate for repairing functional image congruity and will restore brand trust better than a public apology.In a similar vein, it is hypothesized that value-related negative publicity will influence self-congruity more significantly than performancerelated publicity, such that a public apology will be more suitable for repairing self-congruity and, subsequently, brand trust (Tsarenko and Tojib, 2015).More precisely, an apology can compensate for a psychological shortcoming, which may, in turn, repair self-congruity as postulated below: H3. Negative publicity (performance-related vs valuerelated) interacts with repair strategies (a public apology vs compensation) to influence brand trust: for performance-related negative publicity, compensation will be more effective than a public apology for repairing brand trust (H3a); for value-related negative publicity, a public apology will be more effective than compensation for repairing brand trust (H3b).
H4. Functional congruity mediates the impact of negative publicity on brand trust for the compensation strategy, whereas self-congruity mediates the effect of negative publicity on brand trust for the apology strategy.
Figure 1 depicts the conceptual model.

Sample selection
China is one of the most attractive markets for international brands but is highly competitive and complicated from a branding perspective.Tens of thousands of official complaints are lodged against domestic and international brands annually (China Economic Net, 2019).This country is likely one of only a few nations to host an annual Consumer Protection Day: on the 15th of March each year, China's Consumer Protection Bureau (CPB) publicizes brands that have received a large number of consumer complaints and that have been convicted of wrongdoing after official investigations.Consumers also post complaints daily through public media, including social media.
Negative publicity greatly affects brands given extreme pressure to attract and retain customers (Yu et al., 2019).
The current study focused on mobile phones, a popular product category among Chinese consumers.China is the world's largest consumer market for electronic products, yet these products (including local and foreign mobile phones) also constitute the category that amasses the highest number of consumer complaints (China Economic Net, 2019).In 2018 alone, the category received nearly 8,000 complaints via the National CPB, accounting for roughly 30% of all consumer complaints submitted that year (315Online, 2020).

Design and stimuli
This study developed stimuli based on actual negative events that had befallen mobile phone brands in China to improve the scenarios' realism.In the performance-related negative publicity scenario, a news article reported that problems with a technical spare part had severely diminished the phone's sound quality, resulting in a poor consumer experience.The valuerelated negative publicity scenario was also detailed through a news article; the piece stated that some workers working in the mobile phone factory had to work excessively long hours (e.g. 100 h per month) in peak seasons and were seriously underpaid (e.g.some received less than US$1 per day).Both articles explained that China's BCP had investigated and confirmed that each case was valid.A similar approach was adopted for the repairing scenarios: compensation scenario stated that the business offered a public apology as the initial response and the compensation scenario stated that it offered a monetary compensation (see Appendix for stimuli details) Manipulation checks were conducted for both negative publicity scenarios and repair strategies, using an approach similar to that of To and Patrick (2021).A pretest was performed with 50 third-year university students (27 women and 23 men; M age : 23.5 years).These participants were first asked if the negative publicity was performance-or valuerelated based on the definitions provided.Forty-nine participants chose the correct answer for each scenario.The two repair strategies (compensation and apology) were also checked, and all participants distinguished the strategies as expected (100%).The research designs were thus satisfactory.

Measures
The measures in this study were based on existing scales.Face validity was carefully verified by experts to ensure items' relevance.All items (originally written in English) were translated into Mandarin (Standard Chinese) before being back translated into English for meaning equivalence (Zhu et al., 2022).A pretest was carried out to ensure the questionnaire's clarity and flow.
Functional congruity was measured with four items adapted from Sop and Kozak (2019), such as "I think this brand has all of the functions I desire from a mobile phone."Self-congruity was evaluated using four items adapted from Liu et al. (2012), such as "I think this brand is consistent with my identity"; their scale was based on global self-congruity items developed by Sirgy et al. (1997aSirgy et al. ( , 1997b)).Brand trust was assessed with four items adapted from Chaudhuri and Holbrook (2001), such as "I rely on this brand."Perceived realism was measured using eight items (e.g."I think the event in the material portrayed possible real-life situations") adapted from Cho et al. (2012).All measures are summarized in Table 1.

Data collection and analyses
Collection and manipulation check This study included a total of six groups.Groups 1 and 2 (negative-publicity only) were conducted first.These participants were randomly assigned to either of the two negative publicity conditions (performance-or value-related).They were first asked to list their current mobile phone brand (only one), followed by questions related to their functional congruity, self-congruity and brand trust.Afterward, participants were asked to read the negative publicity news report about a fictional mobile phone brand.After reading the report, participants were asked to imagine a situation that the negative publicity had happened to their own mobile phone brand.Then the participants were asked to reevaluate their self-congruity, functional congruity and brand trust.Finally, the participants filled out the perceived realism and demographic questions.
The other four experimental groups (Groups 3, 4, 5 and 6) used a 2 (type of negative publicity: performance vs value) Â 2 (repair strategy: compensation vs apology) within-subject factorial design.Similar to the first two groups, these participants were asked to list their current mobile phone brand (only one), and then they were provided with a negative publicity case of a fictional brand.Then the respondents were asked to imagine that the case happened to their own mobile phone brands, followed by congruity-and trustrelated questions.Afterward, the respondents were asked to read another piece of the news report on the brand's offer of a public apology (or a compensation) as the initial response to negative publicity.The respondents were asked to reevaluate congruity and brand trust questions by imagining if their own mobile phone brands also offered the same repairing strategy.Realism and demographic questions were asked at the end.Table 2 provides details about the groups and procedures.
A total of 608 subjects were recruited through Wenjuanxing, the biggest online data collection platform in China (https://www.sojum.com/)that functions similarly to MTurk Prime.This online platform has been widely used for experimental research (Tong et al., 2019).Eight-six participants who had selected a specific number for all the scaled questions in the questionnaire were removed (Tezer and Bodur, 2020), leaving 522 valid responses.The sample showed a good gender balance (50.38% males and 49.62% females) with an average age of 28.76.The average income was about 4,500 yuan per month in Chinese currency, which indicated the sample were middle-income consumers in China.
More than 95% of participants identified the type of the negative publicity (performance or value) correctly after being exposed to the materials.Also, 98% of the participants correctly identified the repair strategies they were exposed to (apology or compensation).These results indicated that the manipulations of both negative publicity and repair strategy were successful.At the end of the survey, all respondents were asked if they could guess the intention of the study.None of the samples had been able to identify the true purpose.

Confirmatory factor analyses
Confirmatory factor analyses were conducted via AMOS for each construct included in the study.The AVEs for brand trust, selfcongruity and functional-congruity achieved satisfactory threshold values of more than 0.5 (e.g. from 0.59 to 0.76).The construct reliability of the three constructs also exceeded the recommended level of 0.70 (e.g. from 0.85 to 0.93).The Cronbach's alphas were considered good as well, ranging from 0.85 to 0.94.Table 3 provides detailed results.To examine the discriminant validity of the measurement model, the square roots of AVEs for the latent constructs were compared to the correlations of the constructs.None of the correlations surpassed the square roots of AVE, which indicated that the discriminant validity was also satisfactory (see Tables 4 and 5).Further tests were conducted to check if these six groups were significantly different in terms of gender, age and income.Results showed that gender was a significant covariant (p = 0.01), which was because the number of males was twice that of females in Group 3. In other words, the gender balance of Group 3 was not as satisfactory as the other five groups (which were equally balanced).As such, some cautions will need to be applied when interpretating the findings.No significant differences were detected in age (p = 0.106) or income (p = 0.161) in these six groups.

Hypotheses testing Negative publicity and brand trust
Results of Groups 1 and 2 showed both negative publicity types negatively influenced brand trust.After the exposure of performance-related negative publicity, the level of brand trust was significantly reduced [M performance before = 5.43, M performance after = 3.87, t(162) = 8.68, p < 0.05], suggesting that performance-related negative publicity can negatively Notes: All subjects have been evaluated on these constructs twice.1st refers to results when subjects were evaluated these constructs the first time; 2nd refers to results when subjects were evaluated these constructs the second time impact consumer-brand relationships.Likewise, after the exposure to value-related negative publicity, brand trust was also damaged [M value before = 5.59, M value after = 3.88, t(166) = 8.60, p < 0.05].These results support both H1a and H1b.

Mediation analyses of functional and self-congruity
To explore the underlying mechanisms of the effect of negative publicity types on brand trust, a bootstrapped mediation analysis by Hayes (2013) and Zhao et al. (2010) was conducted to examine the role of functional and self-congruity based on results of Groups 3 to 6. Results revealed an indirect but significant effect of performance-related negative publicity types on brand trust through functional congruity (b = À0.23,standard error (SE) = 0.10, 95% CI [À0.48,À0.07]), while self-congruity was insignificant (b = À0.11,SE = 0.08, 95% CI [À0.29, 0.02]).Results also showed that there was an indirect but significant effect of value-related negative publicity on brand trust through self-congruity (b = 0.24, SE = 0.10, 95% CI [0.05, 0.47]), while functional congruity was insignificant (b = 0.17, SE = 0.11, 95% CI [À0.02, 0.40]).Therefore, results supported H2a and H2b.In other words, when participants were exposed to performance-related negative publicity, their functional congruity perceptions were more negatively influenced than their self-congruity perceptions, which resulted in significantly lower brand trust.However, when exposed to value-related negative publicity, their self-congruity perceptions were more negatively impacted than their functional congruity, which resulted in significantly lower brand trust.These results imply that the mediation effect image congruity has on brand trust is contingent on the type of negative publicity (i.e.performance-or value-related).
The interaction effect between negative publicity and repair strategy A 2 (the type of negative publicity: performance-related vs value-related) Â 2 (the type of repair strategy: public apology vs compensation) analysis of variance test was conducted to further explore any interaction effect between the type of negative publicity and repair strategies on brand trust.Results showed a significant brand trust differences under the interaction of negative publicity and repairing strategy [F (1,352) = 87.60,p < 0.001, h 2 = 0.20].Brand trust was significantly better when exposed to a compensation than an apology repairing strategy under the exposure of performancerelated negative publicity [M compensation = 5.48 vs M apology = 4.11; F(1,177) = 51.53,p < 0.001].In contrary, when respondents were exposed to value-related negative publicity, brand trust were significantly better under the apology strategy than under the compensation strategy [M compensation = 4.03 vs M apology = 5.35; F(1,175) = 37.61, p < 0.001].The results confirmed an interaction effect between the type of negative publicity and the type of repairing strategy (see Figure 2).In other words, an apology strategy could more effectively reduce the negative impacts of value-related negative publicity as compared to what performance-related negative publicity could.Furthermore, a compensation strategy had a better mitigation effect on performance-related negative publicity than on value-related negative publicity.These results support H3a and H3b.Further analyses were conducted by including gender as a covariate; results suggested that the interaction effect on brand trust remained the same [F(1,351) = 86.47,p < 0.001, h 2 = 0.20].
A two-way analysis of variance was also conducted to further examine the interaction effect between negative publicity and repair strategies on functional congruity.Results showed a significant differences under the interaction of negative publicity and repairing strategy [F(1,352) = 7.67, p = 0.01, h 2 = 0.02; see A similar analysis was also conducted for self-congruity.Results showed a significant self-congruity differences under the interaction of negative publicity and repairing strategy [F(1,352) = 12.18, p < 0.01, h 2 = 0.03; see Figure 4].The inclusion of gender as a covariate did not change the interaction effect on self-congruity [F(1,351) = 11.73,p = 0.01, h 2 = 0.03].Self-congruity was significantly better when exposed to a compensation than an apology repairing strategy under the exposure of performance-related negative publicity [M compensation = 4.89 vs M apology = 4.23; F(1,177) = 11.12,p = 0.01].In contrary, when respondents were exposed to

Brand trust
Table 6 shows the mean differences in brand trust before and after adopting a repair strategy.First, under performancerelated negative publicity, brand trust increased from 3.87 after being exposed to publicity to 4.11 after being exposed to the apology strategy [M apology before = 3.87, M apology after = 4.11, t(170) = À1.06,p = 0.08].When subjects were presented with a compensation repairing strategy after being exposed to the performance-related negative publicity scenario, brand trust improved significantly from 3.87 to 5.48 [M compensation before = 3.87, M compensation after = 5.48, t(169) = À8.98,p <0.001].
Results of brand trust related to value-related negative publicity were also analyzed in a similar way.Brand trust increased, on average, from 3.88 after being exposed to negative publicity, to 5.35 [M apology before = 3.88, M apology after = 5.35, t(169) = À7.13,p < 0.001] after subjects were exposed to the apology scenario.Furthermore, brand trust on average increased from 3.88 after being exposed to negative publicity to 4.03 after being exposed to a compensation scenario [M compensation before = 3.88, M compensation after = 4.03, t(172) = À0.62,p = 0.39].These results suggested that brand trust was restored to different extents when the company used different repairing strategies to deal with value-related negative publicity.

Image congruity
Results showed that, under performance-related negative publicity, functional congruity was significantly repaired by compensation [M compensation before = 4.22, M compensation after = 5.51, t( 169

Mediation analyses of functional and self-congruity
The mediating role of functional congruity and self-congruity on brand trust moderated by repair strategy was examined by using PROCESS Model 8 with 5,000 bootstrapped samples (Hayes, 2017).In the analysis, negative publicity was the independent variable, repair strategy (apology vs compensation) was the moderator, brand trust was the dependent variable, and functional congruity and selfcongruity were treated as mediators.

Summary
A main objective of this study was to examine the impact of negative publicity on brand trust through the mediating role of image congruity.Another aim was to explore the interaction of negative publicity and repair strategy in restoring brand trust that has been damaged by negative publicity.Several notable results emerged.
Both performance-and value-related negative publicity were found to adversely influence brand trust.Image congruity, namely, functional congruity and self-congruity, played significant mediating roles in this association.In other words, the process of image congruity may partly explicate how various types of negative information can affect brand trust.
This study also provides empirical evidence regarding how brand trust can be repaired.Negative publicity can further interact with repair strategies to inform image congruity and brand trust: value-related negative publicity is better suited to the apology strategy, whereas performancerelated negative publicity aligns more with the compensation strategy.

Theoretical implications
This research contributes to theory in several ways.First, the current study provides a richer understanding of the effect of negative information on brand trust, which is a core brand asset.Negative publicity can threaten brand trust, albeit to different degrees depending on the type.This outcome echoes Yu et al.'s (2019) claim that negative publicity may not affect brand assets equally; rather, brands should carefully diagnose an incident prior to taking reparative action.
Second, image congruity was uncovered as a potential mechanism underlying the relationship between negative information and brand trust.This finding extends knowledge of image congruity in consumers' processing of negative information.In other words, negative publicity may reduce trust by lowering consumers' perceived congruity.Furthermore, results show that negative publicity can affect both functional and self-congruity, but the extent of related impacts appears contingent on the type of negative publicity.For instance, when negative publicity is performance-related, functional congruity is more seriously affected than self-congruity; the opposite is true when negative publicity is value-related.Results also suggest that the congruity process (i.e.evaluating if a brand fits one's functional or self-concept expectations) can possibly shape brand trust.Alternative routes of processing should be investigated in future studies.
Finally, repair strategies exhibit unique degrees of effectiveness in thwarting negative publicity.Compensation strategies (e.g.economically oriented approaches) were more effective in repairing brand trust among the sample if the negative publicity incident was performance-related.On the contrary, a public apology strategy was more effective in repairing brand trust when the negative publicity was valuerelated.These patterns lend support to crisis situation communication theory (Coombs and Holladay, 2009), which suggests that distinct measures should be taken to deal with negative publicity or crises.

Practical implications
Brands should respond promptly and effectively in the face of negative publicity (Ahluwalia et al., 2000).By clarifying the mechanism behind such publicity and brand trust, this study offers a few actionable suggestions to help businesses cope with negative publicity incidents.
First, although no brands are necessarily immune to negative incidents, businesses should strive to prevent these events if at all possible.Negative incidents, as indicated in this and other studies, have devastating impacts on brands; handling these incidents requires extensive human, material and financial resources (Robson and Farquhar, 2021).Businesses should, therefore, seek to maintain productive brand-stakeholder communication throughout daily operations.
Second, businesses should endeavor to repair the consumer-brand relationship when negative publicity occurs.Functional congruity and self-congruity were each found to mediate the impact of negative publicity on brand trust.Firms should thus gradually improve their brands' functional value while promoting their brands' symbolic value as part of routine business.Favorable functional congruity and self-congruity accumulated in advance may buffer negative publicity's adverse effects on consumerbrand relationships, such as in terms of brand trust.Special attention should be paid to self-congruity, which has been often adapted in many positive marketing public relations (PR) activities, such as corporate social responsibility advertising (Belch et al., 2020).Evidence from this study will call for a consideration of self-congruity in the reactive marketing PR activities, in particular, when dealing with value-related negative publicity.For instance, resonance strategies (e.g. based on learning from mistakes) could be adopted in marketing communications to restore selfcongruity.
Third, brands should tailor their repair strategies to negative publicity situations and take corresponding steps to restore brand trust: they should not only identify the causes of such publicity but also quickly devise response strategies.If the negative publicity is associated with product quality, businesses should initially consider choosing a compensation strategy.An apology strategy would be preferable if the negative publicity is value-related, as this approach can help businesses restore their moral standing.As an example, when Starbucks faced negative publicity related to racial discrimination, the company's CEO, Kevin Johnson, publicly apologized to victims immediately.Netizens appeared to unanimously endorse this response (NBC, 2019).
Finally, in an effort to avoid negative publicity, businesses should consistently strengthen their relationships with consumers and conduct timely reviews of brand dynamics.Doing so will better position a brand to address negative publicity as it occurs.

Limitations, future research and conclusion
This study has several limitations to be addressed in future work.First, the sample consisted of relatively young and educated participants.Their reactions to negative publicity could vary from those of other Chinese market segments.Although young and educated consumers constitute a key segment for many product categoriesincluding mobile phonesbecause of their lifetime customer value, scholars should turn their attention to other market segments to expand upon the results.
Another limitation centers around the selected product category.Mobile phones are an appropriate category to consider but investigating a single product type may temper the generalizability of the findings (Liu and Yu, 2013).For instance, responses to negative publicity regarding of fastmoving consumer products (such as food) may differ from consumer reactions to mobile phones, as the prices of mobile phones are much higher than those of fast moving consumer goods.Subsequent research can include multiple product categories to determine whether product involvement interacts with the congruity process.Negative brand publicity has also been found to influence consumers' beliefs about entire product categories (Roehm and Tybout, 2006).It is, therefore, important to understand whether the congruity process applies at the category level.
The experimental design has limitations as well.Although no participants ascertained this study's true purpose, asking the same questions before and after exposure to the research materials may not mitigate subjective bias.Future studies can use different sets of questions to control for this type of bias.Further, in the value-related case, the compensation strategy could be perceived as being geared toward affected workers.
Even though participants perceived both scenarios as realistic cases that could have happened to their own brands, the performance-related case may have resonated more with respondents (playing the part of a consumer) than the value case (playing the part of a worker).Researchers may assess psychological distance (Liberman et al., 2007) in follow-up work.Cultural and personal values could also be investigated in future studies (Yu et al., 2018).
Risk presents another interesting avenue for exploration (Lerro et al., 2019).Perceptions of trust and risk may be related but not necessarily interchangeable; even as some strategies can restore self-congruity and, in turn, trust, risk (e.g. as associated with purchases) may increase in the meantime.Negative emotions such as frustration or anger (Dalzotto et al., 2016) and their relationships with congruity can also be examined in future studies.
Finally, researchers could test combinations of repair strategies, such as blending an apology with economic compensation.Proactive strategies also merit scrutiny: corporate social responsibility initiatives or cause-related marketing (i.e.donations to charities) (Bowen et al., 2013) may be able to efficiently restore consumers' functional congruity and self-congruity.

Appendix. Research stimuli (translated from Chinese to English)
Performance-related negative publicity Please read the following news report from a national newspaper: "Recently, the Bureau of Consumer Protection (BPC) has received thousands of complaints about xxx, a Chinese mobile phone brand, of its poor sound quality which has resulted in the bad consumer experience.BCP has investigated these complaints and concluded that these complaints are valid: the sound quality issue is caused by a faulty technical part of the phone."After reading this news report, please imagine if this negative publicity incident happened to the mobile phone brand that you currently use, what would be your thoughts about [. ..]?
Value-related negative publicity Please read the following news report from a national newspaper: "Recently, the Bureau of Consumer Protection has received thousands of complaints about xxx, a Chinese mobile phone brand, of being a 'sweatshop'.Some workers at this mobile phone's factory had to work excessively long hours (e.g., 100 overtime hours a month) in peak seasons and were seriously underpaid.BCP has investigated these complaints and concluded that these complaints are valid."After reading this news report, please imagine if this negative publicity incident happened to the mobile phone brand that you currently use, what would be your thoughts about [. ..]?
Apology strategy under performance-related negative publicity Please read the following new report from a national newspaper: "In response to the poor sound quality issue, xxx Mobile Phone has held its initial press conference and offered a sincere public apology with a promise to pay extreme attention to product quality."After reading this news report, please imagine if the mobile phone brand you currently use adopted the same repair strategy, what would be your thoughts about [. ..]?
Compensation strategy under performance-related negative publicity Please read the following new report from a national newspaper: "In response to the poor sound quality issue, xxx Mobile Phone has held its initial press conference and promised that they would provide monetary compensations to consumers who have been affected by the sound quality problem."After reading this news report, please imagine if the mobile phone brand you currently use adopted the same repair strategy, what would be your thoughts about [. ..]?
Apology strategy under value-related negative publicity Please read the following new report from a national newspaper: "In response to the 'sweatshop' issue, xxx Mobile Phone has held its initial press conference and offered a sincere public apology with a promise that they would implement a more transparent overtime management system to safeguard the rights and wellbeing of its workers."After reading this news report, please imagine if the mobile phone brand you currently use adopted the same repair strategy, what would be your thoughts about [. ..]?
Compensation strategy under value-related negative publicity Please read the following new report from a national newspaper: "In response to the 'sweatshop' issue, xxx Mobile Phone has held its initial press conference and promised they would provide monetary compensations to workers who have not received correct overtime payments."After reading this news report, please imagine if the mobile phone brand you currently use adopted the same repair strategy, what would be your thoughts about [. ..]? Jingdan Feng is a PhD student in Marketing at the School of Management of Hainan University.Ms Feng's research is focused on branding, corporate social responsibility and green consumption.

About the authors
Fang Liu (PhD in Marketing) is an Associate Professor in Marketing and the Director of Undergraduate Programs at the Business School of the University of Western Australia (UWA).Dr Liu's research, teaching and consulting areas include branding and brand management, advertising, publicity and PR, consumer psychology and cross-cultural studies.Dr Liu has a wide range of publications in these areas and her articles appear in European Journal of Marketing, Journal of Product and Brand Management, Journal of Brand Management, International Marketing Review, just to name a few.Dr Liu's most recent interests include agri-food branding, tourism and destination branding, and comarketing and cobranding.Fang Liu is the corresponding author and can be contacted at: fang.liu@uwa.edu.auFor instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com

Figure 1
Figure 1 Conceptual model Figure 3].After gender was included as a covariate, the interaction effect on functional congruity did not change [F(1,351) = 7.49, p < 0.010, h 2 = 0.02].Results suggested there was no significant difference in functional congruity between the compensation and apology strategy under the exposure of performance-related negative publicity [M compensation = 5.51 vs M apology = 5.32; F(1,177) = 2.15, p = 0.144].In contrary, when respondents were exposed to valuerelated negative publicity, functional congruity was significantly better under the apology strategy than under the compensation strategy [M compensation = 4.75 vs M apology = 5.20; F(1,175) = 5.51, p = 0.020].

Figure 4
Figure 4 Interaction plot for self-congruity (estimated marginal means) Zelin Tong (PhD in Marketing) is a Professor in Marketing and the Director of MBA Programs at the School of Management of the University of Hainan, China.Dr Tong's research, teaching and consulting areas center on branding and brand management, corporate social responsibility, consumer behavior and cross-cultural studies.He has a wide range of publications in these areas.Dr Tong's most recent interests include destination branding, green consumption behavior and corporate social responsibility.

Table 1 Measures
This brand has all of the functions I desire from a mobile phone 2. This brand meets all my functional needs for using a mobile phone 3.The functional value provided by this brand is consistent with what I expect from a mobile phone brand 4.This brand performs well on the functional attributes I value the most for mobile phones The material showed something that could possibly happen in real life 2.The story in the material could actually happen in real life 3.Never in real life would what was shown in the material happen (negative item) 4. The material showed something that had really happened 5. What was shown in the material had actually happened 6.The material was based on facts 7. I believe what is described in the material 8.I think the incident described in the report is trustworthy

Table 2
Summary of groupsCongruities and brand trust were tested before and after exposure to NP Discriminate validity (2nd-time evaluation) Notes: Square roots of AVE estimates are on the diagonal; correlations of the constructs are below the diagonalTable 5

Table 6
Impacts of negative publicity and repairing strategies on brand trust and image congruityAnalyses were also conducted to understand the effectiveness of compensation or apology in repairing self-congruity.Under performance related negative publicity, self-congruity was not significantly repaired by compensation [M compensation before = 3.74, M compensation after = 4.89, t(169) = À5.54,p = 0.18].