1. Introduction

In today‟s society, many companies are focusing on increasing their brand assets Ambler et al. (2002). According to Swaminathan, Page and Gurhan-Canli (2007), the brand assets of an entity can be catastrophically damaged by negative press. This study investigated the impact of social initiatives on distribution marketplaces‟ relationships and the causal relationships among variables. Companies that receive positive press may receive favorable reviews while negative press can lead to controversy Ashley et al. (2011). For example, in a case involving Namyang Trading Co, angry consumers boycotted the company, which led the CEO to personally issue an apology Ashley and Tuten (2015). However, despite the apology, Namyang‟s brand image was not easily recovered. In another example involving the Korean Air Co., Korea‟s leading air carrier, the vice president of the company ordered a plane to return to the terminal due to inflight service issues, an issue that ultimately cost an estimated hundreds of billions of won Brennan (2014). Many studies have suggested strategies for responding to crises that pose fatal threats to a business Bris, Koskinen and Pons (2004). Social responsibility-related crises are increasingly observed, such as in the case of the Namyang Trading Co. The lower the degree of responsibility for an organization‟s crisis, the more defensive the strategy can be used, while when the organization‟s responsibility is greater, a more acceptable strategy is required.

Although companies cannot predict and avoid crises, they can manage them effectively. This study examines practical ways to manage such crises efficiently; we also specify ways through which a friendly corporate image or corporate reputation can be managed. Further, we assess the politicization of real-life reflection by studying the effects of crisis response strategies depending on the type of crisis.

2. Literature Review

2.1. Corporate Crisis Response Strategy

Cordes, Richerson and Schwesinger (2010) classified crisis types according to the nature of the causes and circumstances of the crisis. The most important aspect of the crisis response is the degree of responsibility of the organization. Deighton (1994) classified crisis types according to this aspect. The theory of situational crisis communication is based on the theory of attribution, which seeks to identify who is responsible for a crisis. Specifically, consumers want to know why a particular event occurred Dretsch and Kirmani (2014).

Dutta and Pullig (2011) defined a crisis as an important event that produces negative consequences for organizations, while Gill, Sridhar and Grewal (2017) defined it as a realistic event that threatens organizations‟ existence and/or future development. Harmeling, Moffett and Arnold (2017) argued that a crisis poses a threat to an enterprise that, if the consumer perceives the firm as being responsible for the crisis, creates a negative view of the business.

Some studies have suggested receptive responses have greater effects on crisis situations than defensive responses Heinze, Uhlmann and Diermeier (2014) found that the „apology‟ strategy has the highest level of acceptance among consumers. In other words, when a firm accepts responsibility for a crisis, consumers will show a more positive brand attitude.

2.2. Crisis Response

Because the causes of and consumer responses to crises are diverse, the ways by which companies respond to crises vary according to the type of crisis. Jin, Austin and Eaddy (2018) divided the responsibilities of organizations into victim, accident, and preventable clusters. Klein and Dawar (2004) categorized crises into disaster (e.g., earthquakes and fires) and business crises (e.g., mergers, acquisitions, bankruptcy, traffic accidents, and embezzlements).

According to Kniffin and Shimizu (2016), an apology response has the best effect on all crisis types, while a denial response is ineffective. This suggests that an organization facing a crisis should use strategies to accommodate public demands rather than use defensive strategies against the public. In this context, brand attitude is expected to be more positive if a receptive strategy is used. Kumar and Pansari (2016) first classified crisis types according to their rates of development (i.e., slow, fast) and duration. According to Ma, Yiu and Zhou (2014), common crises faced by companies are global crises, such as public awareness, market changes, product defects, and financial crises.

Macdonald (2000) classified crisis types into economic, information, material, natural disaster, human resource, and mental anomaly crises. Early studies on this topic focused on loss of brand assets because they mainly addressed the negative effects of crises related to product defects. However, recent research on the ethical and brand risks associated with corporate social responsibility have distinguished entities from risk type. Mitussis, O'Malley and Patterson (2006) argued that performance-related risks of company crises are not directly included in performance-related risks associated with a company, but rather social and ethical issues. Muller and Kräussl (2011) reported different responses to different crisis types (e.g., rumors, natural disasters, corruption, and unintentional and corporate-grade crises), and found that the most pertinent responses greatly restored brand credibility. Ozanne, Davis and Murray (2017) correlated consumers‟ corporate associations with corporate competence and professionalism as well as social responsibility. Saaksjarvi and Samiee (2011) identified capacity-level associations, such as innovation, manufacturing capabilities, customer service, and product quality.

2.3. Interaction Effects

Consumers process information that is relatively easy to recover from the negative information of an entity that conducts social contribution activities, thus moving away from the company‟s one-sided social contribution activities to pursue the overall benefits of corporate and marketing concepts Sandin (2009). The financial costs of social initiatives can be large for businesses, even if the activities increase the company‟s corporate image and ultimately profits.

Public service marketing attempts to market social initiatives to promote a positive reputation that can in effect buffer a firm from some of the effects of a crisis. This may require a long-term rather than short-term focus Schultz and Raupp (2010).

According to Sichtmann and Diamantopoulos (2013), if consumers perceive that an entity is working hard in the public interest, then that entity‟s intentions are perceived purely by consumers. The effectiveness and messages of public relations marketing have also been studied. Consumers prefer products from companies that are associated with corporate social responsibility messages, which elicit positive feelings and lead to positive buying behavior and moral satisfaction Viswanathan and Childers (1994).

Consumers can distinguish between specific and abstract arguments Wan and Yiu (2009). Wei, Ouyang, and Chen (2017) argued that favorable consumer attitudes are formed when an entity uses specific messages. Zhou and Ki (2018) offered a detailed interpretation of specific messages, along with detailed explanations of the goals and actions of the actors, the consequences of actions, and finally the expected company strategy.

This study examined specific alternatives for when an entity is obliged to carry out a defensive strategy in response to a crisis. When an enterprise uses a receptive strategy, consumers will not see any difference between abstract and specific messages because they will be more positive. However, when companies use defensive strategies, consumer reaction may differ by the company strategy employed; for example, consumers may be more likely to forgive a firm that has more specifically marketed its social responsibility than those that have engaged in only abstract terms. Indeed, Ashley and Tuten (2015) reported that specific messages were more effective than ambiguous ones. Therefore, based on these arguments, an entity‟s message prior to a crisis will vary depending on the type of activity it plans to carry out after the crisis. In other words, differences in attitudes to the types of public relations marketing messages will be greater than to defense strategies.

2.4. Public Interest Marketing

Intention is the perceived personal importance or level of involvement caused by stimulation in a particular situation Brennan (2014). This has been defined by many researchers Deighton (1994). Dretsch and Kirmani (2014) defined it in the context of attitude. Dutta and Pullig (2011) described it as the degree to which objects or ideas are centralized to an individual‟s value system. Klein and Dawar (2004) defined involvement as “an important level of perception or level of personal relevance to a particular subject” and argued that persuasion message processing depends on the degree of involvement. Kniffin and Shimizu (2016) classified it as a personal, physical, and contextual characteristic. Personal characteristics refer to different levels of involvement that are individually recognized. Physical characteristics are those resulting from differences in products within a particular product family. Contextual characteristics correspond to the decision to purchase, the purpose of the purchase, and the final purchase. Recent studies have focused on the application of such definitions to new targets Kumar and Pansari (2016).

According to Macdonald (2000), the more certain consumers are to express their opinions about a particular message, the more likely they are to express themselves and have more professional involvement. Prior studies have confirmed that consumer involvement affects attitude and the decision-making process. Therefore, this study examined the individual involvement of consumers in marketing the social initiatives of a company before it faces a crisis.

Many transportation companies have recently begun to address social issues due to crises Muller and Kräussl (2011). Distrust and criticism of transportation companies, such as local buses, passenger ships, and airlines, continue to increase. Accordingly, public transportation was chosen as the product line for our main experiment. The subway system operates as a public corporation. We selected a privately funded bus company to facilitate scenario manipulation in the experiment.

2.5. Brand Selection

We selected a fairly new and not-well-known company to avoid the effects of external variables on the use of well-known companies. To select the company, 65 undergraduates (men and women) were queried about their perceptions of seven companies and virtual businesses. The items of Sandin (2009) were modified with brand awareness to fit this study, using a seven-point scale where 1 = entirely. Women reported the lowest score.

2.6. Type of Crisis

Scenarios using the format of magazine articles were used to manipulate the type of crisis and response variables of the crisis in accordance with the researchers‟ intentions. Crisis types were divided into those related to corporate capabilities and those related to social responsibility. We offered a manipulated socially responsible scenario to 138 undergraduate and graduate students. We asked: “Do you think it is indirectly related to the quality?” and “How much of the social responsibility crisis do you think is related to the magazine articles and to women‟s social contributions?” The participants were asked to rate their responses using a seven-point scale (1 = not at all, 7 = very much so). The average result for capacity-level questions was higher than that for social responsibility-level questions.

2.7. Crisis Response Strategy

According to Bris et al. (2004), receptive strategies are those that recognize organizational responsibilities in times of crisis, and include apologizing to the public and making improvements. By contrast, a defensive strategy denies responsibility of the crisis. We tested both strategies by producing scenarios of each. In addition, the company measured the degree of responsibility for the crisis response strategy taken by the company after presenting a report-style scenario to 65 undergraduate and graduate students. The following questions were used to supplement the work of Ashley and Tuten (2015) “Do South Korean travelers acknowledge responsibility for the above incident?, ” “Are South Korean men trying to recover disappointment from the incident?, ” and “Are Korean women not trying to solve the problem at all?” The results showed a statistically significant difference between defensive and receptive responses.

3. Research Method

We used SAS ver. 9.4 software to analyze the results. We applied a 2 × 2 intergroup experimental design to our sample of subjects, who were undergraduate and graduate students at a university in Seoul, South Korea. The three experimental variables considered were the entity‟s risk response strategy, the crisis type, and public relations strategy. The experiments were conducted by presenting a hypothetical scenario to eight groups. Prior to this experiment, five preliminary surveys were conducted to determine the three variables just mentioned. Several selection criteria were considered. First, the available product should be low-cost to accommodate the lower purchasing power of the participants. Second, the participants should be interested in the experimental scenarios because they are recognizable. Third, the selected product family must clearly distinguish between a capacity-level crisis and a social-responsibility crisis. Preliminary surveys were conducted on the basis of these criteria.

Figure 1: Research Model

In the main experiment, the participants consisted of 321 undergraduate and graduate students from the university. Of these, 320 returned the questionnaires. All questionnaires were analyzed, except for 17 sections that contained invalid results. Analyses of the demographic characteristics of the sample group showed that 34% of the participants were male and 48% were female. The participants were randomly assigned to the eight groups noted above; each group contained 57–69 participants. For the main study, the participants consisted of 321 undergraduate and graduate students from a university in Seoul, South Korea. Of these participants, 320 participants returned questionnaires. All of these questionnaires were utilized for the analysis, except for 17 sections, which were filled with insincere results. An analysis of the demographic characteristics of the sample group showed that 34% of the participants were male, while 48% were female. In addition, 21% were used once or twice, 35% were used four to eight times, and 64% were used nine or more times. The participants were randomly assigned to the eight experimental situations mentioned above. At this time, each group of experiments was found to be from 57 to 69.

We provided somewhat negative information about a company to participants by presenting them with prepared magazine articles; other methods were also used. First, the researchers presented the participants with a different purpose than the actual purpose of the experiment. The subjects sought to avoid errors in understanding and responding to the intention of the experimenter, and the survey time was controlled to mitigate unfaithful responses. Before presenting the magazine articles for the study, we disclosed our involvement in public relations marketing by presenting some public interest marketing messages conducted by a hypothetical company. Next, we exposed the participants to 12 prepared YouTube videos and then conducted an investigation into a corporate crisis. This was done to control the consumers‟ favorable feelings about public relations marketing. Finally, the participants were asked to participate in preliminary surveys to expose the fabricated company‟s crisis and crisis response strategy. Then we measured brand attitude, relevance between magazine articles and companies, and crisis response strategies. After the experiments were completed, the original experimental purpose was explained to the participants, who were also advised that the companies and articles used in the experiment were fictitious. The participants were provided with a small token of appreciation.

4. Measure

The dataset, which comprised 320 samples, had mean squared and F values of 2.03 and 3.61, respectively (Table 1). Statistical analyses determined that the difference between the variables was 21.75 and that the difference was highly significant (0.0018). This indicated that the research results were highly consistent with the research direction. R-squared and adjusted R-squared values were 0.06 and 0.04, respectively, and between-variables t values were 82.36 and 83.27, highly significant at 0.0001. This suggests a great psychological influence of consumer attitude on public relations strategy. Specifically, the respondents were told to think about the brand as if they were people, and to think about a series of human features associated with the brand. Finally, respondents were asked to consider how they see themselves and to describe their own personality (the actual self).

Table 1: Demographic Characteristics of Enterprise Crisis and Public Relations Marketing Message Type (Unit: %)

Table 2: ANOVA Analysis

The respondents expressed their overall perceptions of the degree of consistency or discrepancy between the brand‟s personality and how they view themselves. The same procedure was used for ideal self-convergence. From a dependent variable perspective, we evaluated emotional brand attachment using 10 items previously used in consumer studies Brennan (2014). Deighton (1994) placed these items among seven second-order factors (i.e., excitement, joy, and happiness). In this study, we examined the negative impact of a company‟s strategy when responding to a crisis and its impact on consumer brand attitude. We used modified questions from Dretsch and Kirmani (2014). The response, “Yes, I think public relations marketing is valuable to me, ” was used to measure public relations marketing involvement and “Do you think it would be good to „receive‟ women who have been modified for this study” was used as a measure of brand attitude Dutta and Pullig (2011).

Table 3: Multiple Factor Analysis

4.1. Abstract and Specific Message

After showing five abstract messages to 160 people, we selected the most abstract messages according to Klein and Dawar (2004). Then we presented five specific abstract messages to 160 other students. The first eight messages were produced based on prior research that showed that consumers are more response to a domestic target‟s public interest activities than to an overseas target‟s activities. The most abstract message was „supporting the social outcasts, both of which are frequently used by businesses. The selected messages were presented to 320 undergraduate students. The questions were ranked using a seven-point scale from 1 (very abstract) to 7 (very abstract).

5. Result

Statistically significant differences were found between defensive and receptive responses and between abstract and specific messages. Concerning crisis type, the average response to capacity scenario-related questions was higher at a statistically significant level than the average response to social responsibility-level scenario statements.

Table 4: Multiple Factor Covariance and Correlation Analysis

However, within the social responsibility-level scenario, the average response to questions relevant to social responsibility was higher than the average response to capacity-level scenario statements. Engagement with 12 items significantly influenced brand attitude after a crisis and moderated the effectiveness of the response strategy

In residuals analyses, the fit value generally exceeded zero (Figure 2). Although the value was slightly lower in terms of density, it is of great significance in terms of the importance of the overall analyses.

Figure 2: Residuals Analysis

Hypothesis 1 predicted that consumers would have a more positive brand attitude toward receptive strategies than defensive strategies of a firm‟s crisis response. The results were statistically significant, and Hypothesis 1 was supported.

The impact of a company‟s strategy on differences in consumer attitudes toward a brand crisis, depending on the type of crisis, was not revealed by Hypothesis 2. The effects of involvement were significant. There was a significant interaction between the crisis response strategy and the crisis type. The effect of involvement was significant at the 5% confidence level.

Figure 3 shows that fit plot values were relatively orthogonal, with an overall squared fit of 0.99 (99% confidence level) indicating a large difference, which is desirable. Hypothesis 2 predicted that consumer attitudes would be much more positive toward a capacity-level crisis than toward a social responsibility crisis. Thus, Hypothesis 2 was supported. Hypothesis 3 predicted that the impact of crisis response would vary depending on the type of public relations marketing message. We found a significant interaction between the crisis response strategy and entity type. The effects of involvement were also significant. For defensive strategies, consumer attitudes were much more positive toward specific messages than abstract messages. On the other hand, for receptive strategies, no difference was found. Thus, Hypothesis 3 was supported.

Figure 3: Fit Plots

6. Discussion

We found that receptive responses have greater effects on crisis situations than defensive responses. In addition, there is a significant interaction between risk type and the type of public relations marketing message. Third, the degree of advanced public relations marketing was controlled by covariates as the sample comprised university students. Involvement significantly affected covariates, identifying an interaction between crisis response strategy, crisis type, and types of marketing message. In other words, our results indicate that an entity should be more receptive to a crisis at the corporate level, indicating that the issue needs to be actively dealt. Fourth, a company‟s crisis response strategy was influenced by the type of marketing messages presented by the company before the crisis. A company that presents a message in a specific way would generally be perceived to be more proactive to the consumer. Our results show that, for messages that are more specific than abstract, the effect varies depending on the type of crisis experienced. In absolute terms of consumer preference for effectiveness, it was found that, in the case of a corporate capacity-level crisis, it was most effective for the company to undertake a receptive response, such as an apology conveying accountability. However, our study further confirmed that if an entity uses a defensive strategy, the attitudes of consumers do not differ depending on whether the crisis is a social responsibility-level or capacity-level crisis. That is, consumer attitudes are very negative when companies use defensive strategies regardless of the type of crisis.

Regarding abstract vs. specific messages, if an entity‟s crisis response strategy is receptive, then no difference exists between abstract and specific messages. This result reveals some protection from the negative impact of brand attitudes on the firm due to its conventional receptive strategy. However, if the response strategy is defensive, then brand attitude will be higher for abstract messages if the public relations marketing message before the crisis was specific. This indicates that favorable marketing activities of companies prior to a crisis somewhat modify their responses. These results differ from previous studies related to crisis response in that they confirm public relations marketing as an effective tool by which companies can respond to crises.

6.1. Implications of this research

The focus of this study on different types of social initiatives (not covered extensively in previous research) has significant implications for research on brand assets and purchasing intention. Based on our findings, the following suggestions for branding strategies in implementing and implementing social initiatives are made. First, a company‟s social initiatives require authenticity; simply pursuing such initiatives does not guarantee improved brand assets. If a company‟s social initiatives are transparently inauthentic, then consumers will ignore the company‟s products and services in favor of other options. Second, social initiatives should be approached strategically. In this study, social initiatives were more synergistic when based on effective enterprise-wide strategies. Third, companies need experts to promote social initiatives. Corporate brand building is not accomplished in a short period of time, and by definition, requires long-term perspective and investment. Reflection on the real meaning of and need for social initiatives should be at the center of brand-building strategies.

6.2. Limitations and directions for future research

Some limitations of this study and proposed directions for future research should be mentioned. First, it is too restrictive to identify brand assets and purchasing intention according to type of social initiative based only on charitable donations or services. Further research should expand the model to include marketing communication, organizational culture, and human resources. Second, although companies are correct in their social initiatives, they should be categorized without any division of industry or categories. Finally, although this study confirms associations among main variables, the various subvariables that affect companies‟ social initiatives should be investigated.

7. Conclusions

We identified two crisis-based communication strategies (less-responsible defensive responses and highly responsible receptive responses, and applied them to crisis response situations to expand theoretical approaches. We found that, in the event of a crisis, receptive responses are generally more appreciated than defensive responses. It is desirable to take a receptive attitude, such as an active apology, at the corporate level, and to be honest and accept criticism for any complacent or irresponsible behavior. Finally, an entity‟s crisis response strategy can be affected by its favorable activities prior to the crisis. More in-depth research should present public relations messages and other crisis scenarios