INTRODUCTION
Food safety has become a critical concern among the general public in China, especially after the explosive coverage of the baby formula scandals in 2008. For instance, the number of media reports on food safety incidents during the mid-2010s was almost twice that of the previous decade.[Footnote 1 ] Since then, regulators have imposed tighter scrutiny over food manufacturing and retailing industries (Tung, Reference Tung2016). Meanwhile, consumers paid more attention to food safety and showed higher expectations for foreign brands. In this context, retailers, especially foreign retailers, have become targets of public attention. As a result, strategies for monitoring and managing public crises and for repairing legitimacy in the wake of negative incidents are becoming increasingly critical for MNC managers. Although the institutional and firm-level factors for such crises have been explicated (Zhao, Reference Zhao2013; Zhao, Park, & Zhou, Reference Zhao, Park and Zhou2014a; Zhao, Tan, & Park, Reference Zhao, Tan and Park2014b), how MNCs developed legitimacy repair approaches and their efficacies are not well examined.
Research in ethical decision-making and judgment highlights the constructivist nature of such legitimacy loss and repair issues (Sonenshein, Reference Sonenshein2007). Reflecting a central insight in the framing perspective, The Economist (2015: 66) suggests that ‘our response to questions very much depends on how the issue is framed’. Extending this insight into the China context, where the state has dominant control over the economy and society, we surmise that the framing work used by the government significantly influences how the MNCs react to legitimacy loss. Institutional theory highlights how the government exerts explicit, coercive isomorphic pressures on firms (DiMaggio & Powell, Reference DiMaggio and Powell1983) or direct control over their resources (Frynas, Mellahi, & Pigman, Reference Frynas, Mellahi and Pigman2006). However, little is known about how governments influence organizational responses through a subtle mechanism of framing, which in turn justifies a government's coercive actions toward the firms.
Moreover, the host country context for MNCs is not a homogeneous environment. The local contexts, through which we mean the political-economic aspect of a local environment, influence how local governments differentially categorize and judge issues, actors, and situations. We draw on the framing perspective to analyze an MNC's legitimacy repair approaches in response to varied framing by different local governments, the powerful frame creators within their administrative boundaries. Specifically, we seek to answer the following research questions: (1) how do local governments’ framing of an MNC and its misconduct vary in different local political-economic contexts? (2) What approaches do MNCs adopt to repair legitimacy in response to local governments’ framing work, and how effective are they?
To answer these questions, we conducted a case study of Walmart China's approaches in repairing its legitimacy loss over two eye-catching incidents. We selected these two incidents based on theoretical sampling. Their profile and framing reflect a contrast in local political-economic contexts. The first incident was a case of pork mislabeling that occurred in Chongqing in 2011. The role of Foreign Direct Investment (FDI) in the retailing industry had declined due to the increasing dominance by local state-owned retailers and the local government's leftist approach to socioeconomic development. The mislabeling incident was framed by the local government as criminal misconduct. More than ten Walmart outlets were closed down, with millions incurred in penalties and sales loss and several Walmart employees were sentenced. Walmart adopted a compliant repair approach, which was well received by the local government.
The second incident involved the company's sale of packed donkey meat mixed with fox meat (supplied by a local supplier) and occurred in Jinan, Shandong in 2013. In that context, the presence of Fortune 500 companies was limited and the local government highly valued FDI in all sectors. This incident was framed by the local government only as a negligent act. As such, Walmart developed a largely defiant response, which generated negative feedback from regulators, media, and consumers. The two highly publicized incidents triggered legitimacy crises for the giant retailer, showing a contrast in the feature of the incident, local contexts, government's framing work, repair approaches, and their outcomes. Our findings show that the feature of a negative incident and the degree of unfavorability of the local political-economic context toward MNCs prompted different framing strategies (legal and moral vs. administrative) from local governments, resulting in varying levels of the MNC's legitimacy loss. In response, the MNC enacted various legitimacy repair approaches (compliant vs. defiant), leading to different outcomes.
Our study makes several contributions. First, it contributes to research on MNCs under the conditions of institutional complexity (Kostova & Zaheer, Reference Kostova and Zaheer1999). It reveals not only how the national institutional environment (Ahmadjian, Reference Ahmadjian2016; Kostova & Zaheer, Reference Kostova and Zaheer1999) but also how local political-economic conditions shape MNCs’ legitimacy crises and their leeway for legitimacy repair. This analysis is particularly salient in the China context, given the complex institutional conditions and ubiquity of the government agencies across national and sub-national levels. Second, it demonstrates that the national and sub-national environments in China present a very different context for maintaining legitimacy as compared to other national contexts where there is a separation between social and regulatory legitimation processes. Unlike previous research on strategic framing, which focuses on the role of firms, media, and competitors (Bach & Blake, Reference Bach and Blake2016), our study reveals the role of local government, a powerful frame creator in China. Given the government's control of media, legitimacy with the general public (i.e., social legitimacy) is inseparable from that with government regulators (i.e., regulatory legitimacy), although such a separation might be expected in other national contexts. Lastly, the present study extends the literature on framing and legitimacy management by unfolding the linkage between local context, framing, legitimacy repair, and its outcome. The study shows that acts of defiance and counter-framing in a local context, in which the government has strong power over firms and the media, may backfire. Our case analysis offers insights to advance the institutional approach to legitimacy repair in China's national/subnational institutional environments, which contrasts with the argument presented in recent research that an integrative (both institutional and strategic) approach in legitimacy management could be more effective than a single approach (e.g., Scherer, Palazzo, & Seidl, Reference Scherer, Palazzo and Seidl2013). Essentially, our study shows that the effectiveness of the type of repair approach cannot be divorced from the local context in which it takes place.
In the following section, we first review the relevant streams of literature. Second, we outline our research design, data sources and collection, and analytic processes. Next, we present our case findings where we develop the linkages between local context, framing, legitimacy repair approach, and its outcome. In the last section, we discuss our theoretical contributions, outline future research directions, reflect upon doing qualitative research in emerging markets, and highlight implications for practitioners.
THEORETICAL BACKGROUND
Inseparability of Regulatory Legitimacy and Social Legitimacy in the China Context
MNCs’ misconduct causes losses in reputation, status, trust, and more fundamentally, legitimacy, all of which denote the right to exist in the market (Bitektine, Reference Bitektine2011). Legitimacy is defined as ‘a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions’ (Suchman, Reference Suchman1995: 574). It serves as an organizational resource to gain social and political support crucial for its survival (Oliver, Reference Oliver1991; Stevens, Xie, & Peng, Reference Stevens, Xie and Peng2016). A violation of normative standards of behavior often triggers organized consumer attacks (Schneider & Kozinets, Reference Schneider and Kozinets2011; Zhao et al., Reference Zhao, Park and Zhou2014a), leading to ‘legitimacy loss’ of the organization (Jonsson, Greve, & Fujiwara-Greve, Reference Jonsson, Greve and Fujiwara-Greve2009). As governments and the general public are two critical external stakeholders in the retailing industry, we focus on regulatory legitimacy and social legitimacy. Regulatory legitimacy refers to ‘legitimacy with the government regulators’ (Bitektine, Reference Bitektine2011: 154). It is called into question when the misconduct indicates a violation of rules and regulations. This may trigger regulatory sanctions, such as a suspension of the firm's license to operate and constraints on how it conducts its business. As regulators possess coercive power to ensure compliance (Zimmerman & Zeitz, Reference Zimmerman and Zeitz2002), their approval creates a significant impact on organizations (Deephouse & Carter, Reference Deephouse and Carter2005). Social legitimacy, defined as ‘legitimacy with the general public’ (Bitektine, Reference Bitektine2011: 154), is crucial for the survival of organizations and requires repairing after the occurence of negative incidents before it becomes an event-stigmatized organization (Hudson, Reference Hudson2008). We surmise that social legitimacy is inseparable from regulatory legitimacy in the China context because of the state power over firms (Tan & Tan, Reference Tan and Tan2012; Van Rooij, Reference Van Rooij, Bignami and Zaring2016) and its control over the media (Feng, Brewer, & Ley, Reference Feng, Brewer and Ley2012; Yang, Reference Yang2009), especially print media, which act as officialdom's mouthpiece (Feng et al., Reference Feng, Brewer and Ley2012; Marquis & Bird, Reference Marquis and Bird2018) and influence legitimacy evaluations.
The China context is often conceived at the national level. Few studies of MNCs have examined the influence of sub-national environment with the notable exceptions that examine the effect of local conditions on economic growth and performance of MNC subsidiaries (for a summary, see Meyer & Peng, Reference Meyer and Peng2016) and the impact of unfavorable local institutional and market conditions on MNC subsidiaries’ human resource management (Tsui-Auch & Möllering, Reference Tsui-Auch and Möllering2010). In fact, state intervention over firms and market activities in China has increasingly shifted to local governments (Luo, Reference Luo2002; Witt & Redding, Reference Witt and Redding2013), leaving provincial and municipal authorities with leeway to enact policies and enforce regulations to meet local political and economic agendas. Nevertheless, little is known about how local governments in varying local contexts shape MNCs’ legitimacy loss and responses. We attempt to address this gap in knowledge in Chinese management studies.
Framing: Frame Creator and Frame Adoption
Because people cannot access reality in its completeness, frames act as mental brackets, lenses, or schemata of interpretation that delimit what is in and out of the frame and thus construct order out of chaos (Goffman, Reference Goffman1974; Zerubavel, Reference Zerubavel1991). Recent literature on ethical decision-making (Sonenshein, Reference Sonenshein2007) and law enforcement (Hawkins, Reference Hawkins2003) points to the constructivist nature of misconducts. Corporate misconducts such as accidents, scandals, product safety, and health incidents (Marcus & Goodman, Reference Marcus and Goodman1991) trigger legitimacy concerns. In terms of deniability, product safety incidents lie in between accidents that could happen unexpectedly and scandals that are difficult to deny (Wang, Reference Wang2010), and are thus subjected to more diverse interpretations among various stakeholders.
Framing involves deliberate communicative action (Bach & Blake, Reference Bach and Blake2016) to promote, develop, and maintain a particular lens (Giorgi & Weber, Reference Giorgi and Weber2015). Framing is different from attribution. Attribution is about making causal interpretation, but framing is about allocating more weight to particular aspects over others to focus attention on only a part of reality. As such, framing is an intentional, strategic act whereby actors aim to advance specific interests such as garnering support (Bach & Blake, Reference Bach and Blake2016). Essentially, framing is not isolated from the political, economic, and social contexts; it involves a frame creator making sense of signals received from such contexts. Through framing, the frame creator promotes particular issue identification and moral evaluation, which eventually influences legitimacy evaluations (Anteby, Reference Anteby2008).
Whereas previous literature identifies firms, media agencies, social activists, religious groups, and NGOs as key frame creators (Bach & Blake, Reference Bach and Blake2016; Waller & Conaway, Reference Waller and Conaway2011), the role of regulators in framing the issues has received less attention. Hawkins (Reference Hawkins2003) points out that frame construction is central to the process of law enforcement for firms believed to have violated legal rules on occupational health and safety. An enforceable rule violation is socially constructed. A breach may not come to light if it is not attended to by regulators. If attended to, it can be dealt with in a range of ways from informal means of advice and education to more formal and punitive actions, depending on the frame used (Hawkins, Reference Hawkins2003). Given the substantial power of the government over firms and the media in China, we focus on frame construction by regulators. Frames differ in kind and salience within the legal institution (Hawkins, Reference Hawkins2003: 58). A legal frame is more coercive than moral or administrative frames (Hawkins, Reference Hawkins2003: 339, 343). Its salience (the extent to which the frame is discussed in the narratives) may influence public judgment. For example, regulators’ use of the term ‘noncompliant organization’ (based on the severity of rule-breaking and known past record of compliance) is likely to influence public judgment and prompt decisions about formal law enforcement. While regulators’ frame construction is a focus in the legal studies, organizational responses to regulators’ frame construction are beyond the scope of that investigation. To extend the existing literature on framing, we explore MNCs’ varied responses to local governments’ frame construction.
Legitimacy Repair Approaches
Several authors have identified factors at both institutional and firm levels for the emergence of public crises and the mitigating effect of social adaptation activities to lower the risks of encountering negative incidents (Zhao et al., Reference Zhao, Park and Zhou2014a; Zhao et al., Reference Zhao, Tan and Park2014b). However, once such incidents have occurred, how MNCs repair legitimacy and what contextual factors and mechanisms trigger such organizational responses remain to be examined.
The existing literature of MNCs’ legitimacy management focuses on how such enterprises gain and maintain legitimacy to overcome liabilities of foreignness in host nations. Among the various approaches to legitimacy management, both the institutional and strategic approaches are drawn upon the most. The institutional approach highlights that legitimacy can be gained through isomorphic adaptation to institutional pressures (Deephouse, Reference Deephouse1996; Mitchell, Agle, & Wood, Reference Mitchell, Agle and Wood1997). Legitimacy repair usually involves a reactive, compliant approach (Suchman, Reference Suchman1995): apologizing for wrongdoing, accepting an equitable punishment without resistance, disassociating the company from illegitimate business practices with strategies such as personnel replacement, and making internal and external rehabilitative changes (Pfarrer, Decelles, Smith, & Taylor, Reference Pfarrer, Decelles, Smith and Taylor2008; Suchman, Reference Suchman1995). In contrast, the strategic approach contends that an organization not only adapts to but also manipulates its environment, designing noncompliant actions to gain legitimacy by influencing the perceptions of important stakeholders or policy (Barley, Reference Barley2010; Elsbach & Sutton, Reference Elsbach and Sutton1992). Yet Scherer et al. (Reference Scherer, Palazzo and Seidl2013) suggest that an integrative approach with a simultaneous adoption of contrasting strategies is likely to be more successful in restoring an organization's legitimacy than adopting a single approach when facing conflicting demands amid complex and dynamic environments. However, the boundary conditions for the effectiveness of applying an integrative approach versus a single (compliant or defiant) approach in legitimacy repair remain unclear.
In summary, previous research on organizational responses to corporate misconducts has focused mainly on approaches to repairing losses in status and reputation and on consumer trust in individual organizations. We still know very little about how organizations respond to losses in, most fundamentally, legitimacy, particularly the contextual factors and mechanisms that shape MNCs’ legitimacy loss and repair approaches, and the boundary conditions for the effectiveness of such approaches especially in the China context. In this article, we attempt to address these gaps in the existing literature.
METHODS
Case Selection
The issue of MNCs’ legitimacy repair in response to framing is a highly delicate, complex, and sensitive phenomenon. Exploratory research and comparative analysis are argued to be more suitable than large-scale, cross-sectional studies to make sense of complex and dynamic management phenomena and thick description (Birkinshaw, Brannen, & Tung, Reference Birkinshaw, Brannen and Tung2011). We also maintain that case study methodology, with a focus on the contexts and a combination of primary and secondary data collection, offers an appropriate lens to unpack the contexts and mechanisms inductively for subsequent theory building (Tsang, Reference Tsang2014).
Among the foreign retailers in our empirical context, we select Walmart China to explore environment-level conditions for different repair approaches and outcomes for the following reasons. First, Walmart China represents MNCs in the retailing industry, given its foreign origin, large size, and high visibility as a subunit of one of the world's largest retailers. Second, since it began operating in China in 1996, the company has ranked among the top 10 retailers in China between 2011 and 2017 (China Chain Store & Franchise Association, 2018)[Footnote 2 ] and has ‘come across issues that most multinational companies have experienced in China, from the opening of the market to the more recent focus on food safety.…’ (China Law and Practice, 2015).
We first screened Walmart's misconducts in China and then selected two types of incidents: (a) those in which the retailer is generally perceived to be the single wrongdoer: the pork mislabeling incident in Chongqing in 2011; (b) incidents in which not only the retailer but also the suppliers were blamed: the donkey meat incident in Jinan, Shandong Province in 2013. While the two types of incidents give rise to different degrees of leeway for interpretation, the local political-economic contexts also provide a contrasting degree of unfavorability toward foreign retailers.
Data Collection
In this study, we used documentary research, interviews, and online surveys to examine the research questions from different angles and triangulate the data.
Qualitative data collection
Our data sources for qualitative analysis include (a) personal interviews and group interviews, (b) naturalistic observation of outlets of Walmart and its local competitors, (c) media reports, (d) social media posts, and (e) press releases; each serving unique and complementary functions (see Table 1). For first-hand data, we conducted multiple field visits to Chongqing and Jinan to interview various stakeholders.
Table 1. Data description
Given the sensitivity of negative incidents and the sociopolitical context of China, access to major players for interviews are extremely challenging for researchers (Sun, Mellahi, & Liu, Reference Sun, Mellahi and Liu2011). We enlarged the pool of interviewees by tapping on close contacts with retailers and external stakeholders in the local contexts. We conducted a total of 71 personal interviews with 57 interviewees in Mandarin. Some interviewees were interviewed twice for follow-up questions. Interviewees included managers and frontline employees of Walmart China, and external stakeholders such as government officials, competitors, suppliers, consumers, journalists, and academics in relevant disciplines (e.g., Marketing, International Business, Operation Management, and Economics) across various provinces between 2015 and 2017. Current and former employees of Walmart were selected from different functional departments relevant to our research context, such as Compliance, Food Quality Management, Promotion and Sales, Customer Service, Global Purchasing, and Human Resource Management. Their positions included departmental manager, manager of outlet, assistant to general manager, and frontline worker. In addition, we conducted three group interviews with 18 consumers to understand their perceptions of the negative incidents and evaluation of Walmart's legitimacy repair approaches. The length of interviews ranged from 10 minutes (e.g., for follow-up interviews) to 120 minutes. We used semi-structured interview guides to collect open-ended narrative data. To promote candor and protect respondent identities, we guaranteed the anonymity throughout the interview process. For documentary research, we collected English and Chinese news articles (using WiseSearch, Factiva, and CNKI databases) on major negative incidents involving Walmart China from 2006 (occurrence of its first major negative incident) to mid-2017. Media articles constitute an important source of data, as they are vetted by government departments and thus reflect governments’ framing after the occurrence of negative incidents, according to journalists we interviewed. Social media posts unfold the degree of legitimacy loss and responses to legitimacy repair approaches. Press releases reveal the company's official view and repair approaches. We focused on the pork mislabeling (and the duck expiry incident prior to it), and donkey meat incidents. They received extensive media coverage, constituting 503 articles out of a total of 1,460 reports on the company's major negative incidents and 1,258 posts from Sina Weibo (a popular social media platform in China). We label them as incidents rather than scandals, as do most articles in Chinese media and all journalists interviewed. Given that the two incidents received the most intensive media coverage, we focused on them in the survey data collection.
Survey data collection
To gauge the perception of negative incidents and examine the effectiveness of repair approaches, we launched anonymous online surveys among 25 experts and 154 consumers via Sojump, a widely used online survey platform in China. The experts included academics in business ethics, crisis management, and marketing, as well as practitioners in communication. Each expert was invited to read two summaries that we derived from news reports of both incidents and the legitimacy repair approaches. In contrast, each consumer was randomly assigned to read a scenario of one incident and the repair approach.
We measured the perceived intentionality of the MNC by ‘to what extent do you think Walmart did this on purpose’ on a six-point Likert scale (1 = strongly disagree, 6 = strongly agree), and whether the MNC was the sole party to blame. If a participant indicated Walmart was not the sole party to blame, we asked her/him to write the percentage of responsibility for which Walmart, supplier, and government should each bear. We measured the perception of legitimacy repair effectiveness by four items: to what extent the repair approach was satisfactory, sincere, reasonable, and efficient on a six-point Likert scale (1 = strongly disagree, 6 = strongly agree). As the reliability for each incident among both groups was higher than 0.86, the average score of the four items that was used as an indicator of legitimacy repair effectiveness.
Table 1 summarizes how each data source was used in the analysis. Overall, the media reports and interview data provided us the details of the local context, the incident, regulatory legitimacy, and framing and reactions from the regulators. Corporate statements provided us the information regarding the MNC's perception of the incident and its responses. Social media data were used as indicators of social legitimacy. The survey data, altogether with the group and individual interview data, provided us some evidence regarding social legitimacy and legitimacy repair outcomes. Together, the various data sources gave us insights into the institutional complexity, the interactive dynamics between the MNCs and external stakeholders, its legitimacy loss, varied repair approaches, as well as some outcomes.
Data Analysis
We created a coding scheme based on theory, prior research, and data to code the texts of all the transcripts, notes, press releases, news articles, and social media posts on the two incidents. Our coding was ‘guided by prototypes rather than exact terminology’ (Chong & Druckman, Reference Chong and Druckman2007: 108), allowing us to develop a contextual understanding of the data and uncover new themes that are not identified in the initial coding scheme. To enhance the reliability of our qualitative analysis, two research students independently coded each text. Three coauthors conducted regular meetings with them to discuss data interpretation and identify emerging themes. Second, we added codes when new facts, factors, or insights emerged, and recoded all the texts based on the final version of the coding scheme as shown in Table 2. We revisited our data constantly when insights emerged about the features of the incidents, environmental contexts, the framing acts, legitimacy repair approaches, and outcomes. We then compared the two incidents to identify similarities and differences (Eisenhardt, Reference Eisenhardt1989). The comparative analysis resulted in additional data collection, further interpretive results, and a development of key constructs as well as an identification of the relationships between the constructs.
Table 2. Coding scheme for qualitative data
RESULTS
The Changing National Environment in the China Market
The institutional and market environments for foreign retailers have been changing rapidly over the past few decades in China. Initially, foreign retailers were permitted to operate only joint ventures with local companies in first-tier cities and special economic zones, and the amount of capital invested and number of stores were strictly controlled by the Chinese government which intended to protect local players (Chuang, Donegan, Ganon, & Wei, Reference Chuang, Donegan, Ganon and Wei2011). The government, however, began to open the domestic retailing market to foreign competition after joining the World Trade Organization by abandoning its restrictions on location and entry mode in 2004. Yet, with the rise of local retailers and their catching up with MNCs in terms of performance and management, ‘foreign retailers’ historical mission in China has been completed’, according to a Beijing-based marketing scholar. Subsequently in 2008, the government removed the preferential treatment of foreign retailers.
In 2007, the central government launched a nationwide campaign to enforce the food safety regulations (Van Rooij, Reference Van Rooij, Bignami and Zaring2016). The baby formula scandals revealed after the Beijing Olympic Games in 2008 fueled additional anger toward wrongdoers and attention to food safety incidents. Reports of product-related safety concerns and corporate misconduct have increased substantially, reflecting higher expectations from consumers, scrutiny from full-time ‘anti-fake product fighters’, and the rapid rise of social media as watchdogs, according to interviews with journalists, academics, and retailers. In 2011, one deputy to the National People's Congress (NPC), Wang Lijun, then the chief officer of the Chongqing police, obtained the NPC's endorsement of his proposal to strengthen the punishment for offenses related to food and drug safety issues. Specifically, foreign retailers have been subjected to strict scrutiny. A journalist observed, ‘MNCs are leaders and representatives of the industry.… If a large MNC can't avoid food safety issues, the situation in the whole industry must be worse’. Table 3 summarizes the key findings for each incident in which Walmart was involved.
Table 3. Summary of results
The Solo Performance: Pork Mislabeling Incident in Chongqing
The local context
The national environmental effects were amplified by the government in Chongqing. The city, with a population over 30 million, was under the reign of the then-party secretary, Bo Xi-lai (from November 2007 to March 2012), who, prior to taking that post, had served as the minister of commerce, making him one of the key officials to steer the nationwide campaign to enforce food safety regulations (Economic Information Daily, August 29, 2007). Under Bo's leadership, the government, adopted populist nationalism (Xu, Reference Xu2001) and a leftist approach to socioeconomic development (Ming Bao, October 11, 2011; Oriental Daily News, October 15, 2011). The government had removed favorable policies toward MNCs, especially in the retail sector, which saw a surge in local retailing giants that include prominent state-owned enterprises. Moreover, it was determined to combat corporate misconducts, particularly regarding food safety incidents. As stated above, the proposal by the chief police officer to ‘sentence the wrongdoers to death without any forgiveness’ (sha wu she, 杀无赦) in food and drug safety incidents got endorsed by the NPC in 2011. A senior police officer recalled, ‘Chongqing took the lead in China to set up an anti-fake product team in the police force to combat counterfeit and shoddy products, raised the issues to the criminal level.… It is necessary to zhua dian xing (抓典型, to catch a typical actor or case, which is usually a representative in a sector)’. Which organization was picked as an example of bad actors for penalization depended on the political leaders’ assessment of the environment. A government official said, ‘In China, the will of political leaders is very important.… The government has to set a direction. It reflects with which card the government wants to play’.
One food safety incident of Walmart Chongqing was under the spotlight in early 2011 before the pork mislabeling incident. An outlet of Walmart was reported to sell deep-fried expired duck meat as cooked food. The act was caught on the spot, with almost 1800 kilograms of expired duck meat found in storage, and 200 kilograms of expired duck products identified to had been sold. This incident caused significant legitimacy loss indicated by Sina Weibo posts. In general, the public expressed disappointment: ‘Even large MNCs (or a Fortune 500 company) conducted such an evil deed; how can we trust any other retailer’? Some expressed a nationalistic resentment: ‘Walmart, get out of China’. Interestingly, one post remarked that ‘Although Brother Jun [Wang Li-jun] has just proposed the importance of food safety issues to the central government, the local Walmart publicly fried expired duck. [Walmart China's] act weakened the credibility of the Party and government… weakened the public's sense of safety, and threatened social stability.… Kill it without forgiveness’!
Framing by the local government
In September, 2011, a few months after the duck expiry incident, the local trade and industry bureau found that Walmart mislabeled some ordinary pork as the more expensive organic pork, obtaining an illegal benefit of 730,000 yuan. A staff member working in another foreign retailer regarded this mislabeling issue as ‘not uncommon for retailers.… In fact, it was hard to tell ordinary pork from organic pork without labels.… There could be human errors.…’.
In contrast, according to the then-party secretary, ‘This [pork mislabeling] incident is critical, as it is concerned with the daily life of the public…’ (Oriental Daily, Oct. 15, 2011). An official in the publicity department described how to deal with the wrongdoer: ‘Usually, you've got to arrest those in charge. You use the food safety laws. Once you get rid of the people, you can tone down the public opinion’. Several government officials called press conferences and emphasized the pork mislabeling as an act of deceptive pricing, putting profit over consumer rights and violating the consumer protection laws (e.g., Chongqing Daily, Oct. 10, 2011: 1). Accordingly, the local government publicized the incident as the top most representative cases of breaching consumer rights. An official emphasized, ‘Despite having been advised and educated and even fined 21 times over for past mishaps, the company refused to make corrections and improvements of its practices.… Therefore, the penalty this time is really severe’ (Chongqing Evening News, Oct. 10, 2011: 5).
The government's framing work helped justify harsh penalties on the company. The company was penalized with a 15-day store closure, which was unheard of in China. In addition, two employees were arrested, 25 were criminally detained, seven were placed under surveillance of residence, and three were released on bail. A fine of 3.65 million yuan was also imposed (Oriental Daily, Oct. 15, 2011). This was considered heavy in the context of Chinese retail industry. The local industrial and commerce bureau, according to Chongqing Morning Post (Oct. 26, 2011: 3), further imposed on the Chongqing outlets ‘comprehensive administrative guidance on employee training, amendments in procedures, and development of consumer rights protection’. This administrative guidance extended by the government was confirmed as ‘unprecedented and highly politicized’ by a senior police officer.
The government's framing work not only justified the penalties but set the tone for news reporting by eliminating varied interpretations. A journalist explained, ‘The government requires the media to help publicize its judgment.… It was willing to release statistics related to all its past and present incidents. We can then devote extra effort to publicize it. In the era of social media, the government's control over public opinion is tight.… We don't dare to do reporting based on our own view’. An academic expert in marketing remarked ‘Local newspapers are the mouthpiece of the local government. The reports of negative incidents have been vetted by the government's publicity department.… Once the government decides how to frame the incident, it will set the tone and the newspapers will not dare to reflect multiple interpretations. There will be a single voice on the incident.… Enterprises assess risks based on situations.… Walmart did not dare to place itself at the muzzle of a gun’.
Repair approach in response to legitimacy loss
Based on analysis of the Weibo posts, we ascertained that 77.3 percent of netizens blamed Walmart China for deliberately committing the misdeed and felt that the company should bear the full responsibility for it. The survey result also shows that the pork mislabeling incident was largely perceived as deliberate, and 70.1 percent of consumers and 68 percent of experts perceived Walmart as the sole party to blame (see Table 4).
Table 4. Descriptive statistics of consumers' and experts' perceptions of the two negative incidents
Notes:† Standard deviation is reported in bracket. For perceived deliberation and perceived effectiveness, a higher score indicates a higher perceived deliberation in conducting the incident, and a higher perceived effectiveness of legitimacy repair.
The top management of Walmart China issued apologies five times, acknowledging the infringement of consumer rights (Chongqing Morning Post, Oct. 26, 2011: 2). Just a week after the local authorities’ imposition of penalties, two top executives resigned for personal reasons. However, a Walmart manager remarked, ‘Because of this incident, the company has to find a scapegoat, at least to indicate an accountability to the government’. Another manager recalled, ‘The industrial and commerce bureau, police, and tax bureau sent officials to sit in our office daily. This reflects the leaders’ intention in Chongqing at that time; it was also [considered to be] a performance by the government’. Subsequently, Walmart China announced that it would overhaul its operations in the city, including the establishment of a compliance department in Chongqing.
The company not only fully complied with the government's demand for a temporary closure of a number of stores but also voluntarily shut down more stores. A current manager of the Chongqing branch recalled, ‘At least our store was voluntarily closed down by our company. It was the decision from Headquarters … to show our determination [to reform]’. An academic in Chongqing interpreted this voluntary store closure as a smart move, emphasizing, ‘It is certainly beneficial to gain support from the government’. Intriguingly, the MNC neither publicized its records of corrections and improvement of practices after the past incidents nor broadcast its voluntary store closure beyond the regulators’ requirement.
Some outcomes
In view of Walmart's low-profile, compliant repair approach, an academic commented, ‘There is a Chinese saying: “The gun shoots the outstanding bird”. If you show yourself as outstanding, consumers will pick on you, not to mention the regulatory authorities. It will give you more trouble’. Overall, the local newspapers reported that consumers were willing to ‘give it a chance to reform’ (Chongqing Morning Post, Oct. 26, 2011: 2).
Moreover, Walmart's repair approach gained endorsement from the government as well as an offer for support in the future. The mayor complimented the company: ‘The changes implemented by Walmart [China] are very reasonable and pragmatic.… indicating that the MNC has fulfilled its responsibility’ (Chongqing Evening Post, Oct. 25, 2011: 9). He added, ‘I believe Walmart China will have a better development after learning this lesson.… We have policy incentives for companies in the food industry, and Walmart China can also benefit from them’ (Chongqing Morning Post, Oct. 22, 2011: 2).
The Duet: The Donkey Meat Incident, Shandong
The local context
The incident occurred in Jinan, which, with a population of over seven million, is the capital city of Shandong Province. Despite the national policy on removing incentives to MNCs, the government of this sub-provincial city, which has had limited foreign direct investment as compared to that of the other 14 such cities, has maintained favorable policies toward MNCs under then-party secretary of Jinan, Min Wang (November 2011 to December 2014), and his successor, Wang Wen-tao (March 2015 to March 2018; The Paper, May 7, 2016). The government sought to attract large MNCs for economic development, employment generation, and enhancement of the ‘city image’ and ‘government image’, according to five interviewees. Walmart operated only two stores (versus 13 stores in Chongqing) as of 2013, and thus the government was eager to keep the company's presence in the local market. A Walmart manager in Jinan said, ‘In fact, the government has been pretty good to us’.
Framing by the local government
In December 2013, in response to a customer's complaint about the donkey meat products purchased from a Walmart outlet, the Shandong food and drug administration conducted tests on the products. It released test results eight days later to confirm that the donkey meat product contained fox DNA. The incident reflected a loophole in the regulatory inspection system as a manager from a local competitor explained, ‘In China, suppliers just need to provide a sample of a product for inspection. The quality of the product they actually supply to retailers can be different from that in the sample.… This is a critical loophole.… After all, the regulators and inspectors should improve the monitoring over the suppliers for quality management’.
While the government usually favors MNCs because of their significant contribution to the economy, ‘they cannot hit the red line [the warning line].… If they do, they will be dealt with [by the government]’, according to an official in the publicity department in the local government. An official from the food and drug administration emphasized, ‘If the case is serious and creates negative impact, the party leader and the city government have to work together to set the tone to guide the actions of the government departments’. Indeed, the press labeled it as the ‘reoccurrence of the pork mislabeling incident’ (Qi Lu Evening Post, Dec 25, 2013) and reported that Walmart China had sold fake products valued at more than 300,000 yuan. Typically, reports of negative incidents required ‘an approval from local governmental publicity departments’, said a local journalist interviewed.
In a press conference in early January 2014, the officials of the local industrial and commerce bureau blamed Walmart China for ‘not having fulfilled the necessary responsibilities’. First, the company did not check the quality control documents of the product. Second, it did not perform risk evaluation and quality control, because it expressed no doubt over the supply even when the price of the processed meat was lower than that of the raw meat. Third, it concealed the inventory data of problematic products upon investigation by the government. Last, its handling of consumer complaints caused much confusion (Qi Lu Evening Post, Jan. 3, 2014: N02). Nevertheless, the government imposed no fines on Walmart. Overall, the government used an administrative frame of ‘negligent act’ to depict the incident in terms of a sale of fake products, reflecting weak internal control. It blamed the company for not having fulfilled its responsibilities rather than accusing it for having violated consumer protection laws or failing to observe the food safety laws. This framing work helped play down the issue and distract attention from the loopholes in the regulatory inspection system.
Repair approach in response to legitimacy loss
Our analysis of the Weibo posts shows that 54.5 percent of netizens blamed Walmart for the incident, 18.2 percent blamed the supplier, 9.1 percent blamed the government, and 18.2 percent blamed others (e.g., the social context). The public considered that ‘In the donkey meat incident, Walmart could claim to be a victim…’. In line with our qualitative analysis, the survey result (see Table 4) shows that the MNC was moderately perceived as deliberate in its conduct, and that only 37.7 percent of the consumers and 16 percent of the expert panel considered Walmart as the sole party to blame.
As the administrative frame of ‘negligent act’ had a lower capacity to coerce than a legal frame, there was much room to engage in legitimacy repair. On top of observing standard practices (such as offering public apologies, transferring the compliance officer to another position, and compensating the customers and withdrawing all the products), the company undertook high-profile, defiant actions. First, the top management contested the administrative frame of ‘negligent act’ by delivering an apologia that was a de facto self-defense piece to counter the administrative frame by the government and to reframe itself as a victim of the errant supplier, and to distance itself from the charge of wrongdoing. The CEO pointed out, ‘The supplier intentionally committed the fraud. It deceived not only the government but also Walmart China and the consumers. The major responsibility should be borne by the supplier’ (Qi Lu Evening Post, Jan. 3, 2014: N02). A manager from Walmart, Jinan, added, ‘Surely, it is the responsibility of the other parties. [Our company] was the victim… Our purchasing department had all the documents about the product, which had been endorsed by the government’. Another Walmart employee said, ‘Walmart should take some responsibility, but the government should have borne more of it. The matter was about the government's monitoring’.
Furthermore, Walmart China's CEO put the blame on the Chinese authorities and urged them to ‘clean up its act’, pointing out, ‘In the U.S. and most other countries, it is usually the manufacturers, rather than the retailers, that have primary responsibility for the quality of the products they sell.… But in China, the manufacturers are not accountable. We are accountable… You resolve it by putting in place the right processes at the base. That is something the government should be held accountable for, as they are in most other countries’ (Wall Street Journal, Apr. 14, 2014). The company also held talks with China's food and drug administration, urging the officials to increase their inspection of all food purveyors. To counter the government's negative framing, the top management repeatedly announced that the company's goal was ‘not to be the biggest retailer in China … but the most trusted’ (Wall Street Journal, Apr. 30, 2015). It publicized its heavy investment in product quality assurance, including ‘food safety audits, third-party facility audits, and DNA tests’ (ENP Newswire, 2015), and its translation of an online training module on food fraud into Mandarin (ENP Newswire, 2014). The act of defiance was believed to address the legitimacy loss by convincing the public that other players should be held responsible for guaranteeing food safety, not just the retailer (Wall Street Journal, Apr. 14, 2014).
Some outcomes
Walmart's defiant approach through counter-framing and blaming the government appeared to be counterproductive. For example, a female consumer said, ‘We were disappointed that a retailer shifted the responsibility to others’. A government official in charge of food safety issues responded, ‘During this period, Walmart is under the spotlight. The public may consider Walmart as a scapegoat of the supplier and it is a victim itself. In fact, Walmart China is not treated unjustly at all.… In fact, Walmart committed “four sins,” including negligence of quality investigation and of internal control, inappropriate responses after the occurrence of this incident, and not being cooperative with legal enforcement officers…Overall, it can hardly absolve itself from the blame’. (Life Daily, Jan. 7, 2014: A03; Life Daily, Jan. 8, 2014: A02). In response to the act of defiance, a journalist reflected, ‘[Walmart China] asked the government to clean up its own act. This sounds like putting the blame on the government.… In China, this is definitely not the way to deal with the government.…’.
Consistent with our qualitative analysis, the survey result (see Table 4) shows that the MNC's legitimacy repair approach for the donkey meat incident was perceived as ineffective in the eyes of both consumers and experts. The defiant approach and claim as a victim were perceived negatively from the answers in the open-ended questions in the survey as ‘irresponsible’ and ‘insincere’.
DISCUSSION
A comparative analysis of two incidents provides us with rich data that reveal how negative incidents and contextual conditions lead to differences in governments’ framing, the degree of legitimacy loss, MNCs’ legitimacy repair approaches and outcomes. In Chongqing, where local political leaders elevated food safety violations to a criminal level, the pork mislabeling incidents led to severe legitimacy loss. Walmart adopted low-profile, reactive, and compliant actions for legitimacy repair in response to the Chongqing government's highly negative framing of the incident and the organization. The donkey meat incident in Jinan, unlike the Chongqing case, involved suppliers of the problematic product. Moreover, the local context in Jinan, Shandong, is favorable toward foreign investment in the retailing industry. Walmart, instead, adopted a defiant approach in response to the local government's slightly negative framing of the incident. Our study identifies an MNC's legitimacy repair approaches in response to the framing strategies deployed by local government under different local contexts. The findings unfold the linkage between local context, framing, legitimacy repair approach, and outcome.
Theoretical Contributions
China: Institutional complexity
Our study contributes to the literature on MNCs’ legitimacy management under conditions of institutional complexity (Kostova & Zaheer, Reference Kostova and Zaheer1999), especially in emerging markets. Our analysis shows that the host country environment in China has experienced dramatic changes over the years since mid-1990s. Foreign retailers, such as Walmart, were leading the retailing industry by introducing supermarkets into China. As local competitors developed over the years, the advantage of foreignness (Edman, Reference Edman2016; Shi & Hoskinsson, Reference Shi, Hoskisson, Tihanyi, Pedersen and Devinney2012) was gradually blunted. Meanwhile, consumers and regulators developed higher expectations for foreign brands and products. This fast-changing context poses increasing challenges for MNCs in retailing industry, which is subject to sub-national differences in sociopolitical and economic conditions.
In addition, our study reveals the role of not only the national environment but also the local governments in shaping MNCs’ legitimacy crises and leeway for legitimacy repair. Our analysis of the local contexts extends the focus on the institutional and market conditions (Meyer & Nguyen, Reference Meyer and Nguyen2005; Shi, Sun, & Peng, Reference Shi, Sun and Peng2012) to uncover the influence of the political conditions. As shown by our case data, the sub-national environments in China differ in terms of unfavorability from the local government toward FDI and local competitive dynamics. Specifically, the pork mislabeling incident occurred in Chongqing during a particular period, when the local government promoted ‘populist nationalism’ (Xu, Reference Xu2001) and a leftist notion of socioeconomic development, highly unfavorable to foreign retailers. Meanwhile, the presence of strong local retailers, especially SOEs, reduced the government's reliance on foreign retailers. Walmart was picked as a typical case (zhua dian xing, 抓典型) to show the determination of local government to fight offenders in food safety issues. In comparison, the Jinan government was relatively lenient toward Walmart, given the limited presence of large MNCs and the higher dependence on their contribution to economic development. Therefore, Walmart encountered a less severe legitimacy crisis and perceived a greater leeway for legitimacy repair.
Other than institutional complexity across levels, our case study reveals that China presents a different context for maintaining legitimacy as compared to other national contexts in which a separation exists between social and regulatory legitimation processes or social activism can be influential on large corporations (Rao, Yue, & Ingram, Reference Rao, Yue and Ingram2011). Given the government's control of media, social legitimacy is inseparable from regulatory legitimacy in China. The degree of social legitimacy loss appears to resonate with that of regulatory legitimacy loss in each incident. Therefore, Walmart needed to respond specifically to local regulatory agencies’ framing acts.
Framing: The local government as a frame creator
While prior studies on strategic framing focuses on the role of firms, media, and competitors (Bach & Blake, Reference Bach and Blake2016), our study reveals the role of the local government as a powerful frame creator in the China context. Our analysis thus contributes to the framing literature by unfolding the linkage between local political-economic context and the role of government in framing and its influence on MNCs. As legitimacy is essentially a perception that is socially constructed, the manner in which the negative incident and the organization are framed shape MNCs’ legitimacy loss and repair approaches. In the Walmart case, both incidents involved fake product sales and deceptive practices. However, the framing by local regulatory agencies substantially differs. As shown by our case data, negative framing is reflected in the kind (legal, moral, or administrative) of frames used and its salience (the extent to which the frame is discussed in the narratives).
In the pork mislabeling incident, the Chongqing government engaged in highly negative framing of Walmart and its role in the incident. It perceived the MNC to be deliberate in deceiving consumers, and thus to be the key party to blame. Thus, it used a legal frame, depicting the issue as a violation of food safety laws and consumer protection laws, which had a high capacity to coerce. It also adopted a moral frame, blaming Walmart for a deceptive act, exhibiting an intention to generate excessive profit. Furthermore, it characterized Walmart as a ‘noncompliant organization’ and a ‘repeat offender’ that refused to make corrections, extending a negative evaluation beyond the issue to the moral character of the organization. However, the Jinan government engaged in slightly negative framing to downplay the incident and to distract attention from its own regulatory loopholes. It could have used the legal frame to accuse Walmart of having failed to observe food safety laws, or having violated the consumer protection laws. Instead, it used an administrative frame, identifying the issue as a sale of fake products in relation to weak internal control, which had less capacity to coerce and appears to be contestable.
Legitimacy management: Repair approaches and outcomes
Last but not least, our study extends the legitimacy management literature by revealing the linkage of local political-economic context, framing, legitimacy repair, and its outcomes. Walmart, despite its reputation for a high level of standardization in operations, enacted varied organizational responses in response to different framing by the two local governments. In the pork mislabeling incident, the MNC perceived it to be risky to counter the Chongqing government's highly negative framing. Consequently, the company promptly adopted an institutional approach with reactive and compliant actions to focus on restoring regulatory legitimacy. This institutional approach appeared to be effective in repairing legitimacy, as shown by the archival data and cross-sectional survey data.
In light of the slightly negative framing in the donkey meat incident, the MNC perceived more room for charting its repair action, and adopted a defiant approach. Although the local political-economic context in Jinan was favorable to FDI and the negative incident can be attributed to the supplier's fault, Walmart's defiant approach was negatively received and caused highly negative responses from the local regulatory agencies, thus limiting the effectiveness of its legitimacy repair approaches. As shown in our study, acts of defiance in a political context, in which the government has strong power over firms and the media, may backfire. This argument was further supported by the cross-sectional survey data as the defiant approach was perceived among both consumers and experts as ineffective. Comments regarding the repair approach in the donkey meat incident also pointed to the perception of Walmart, through its act of defiance, as being irresponsible.
Our case analysis offers insight to support the institutional approach to legitimacy repair in the national/subnational institutional environments of China, which contrasts the argument that an integrative (both institutional and strategic) approach in legitimacy management could be more effective than a single approach (e.g., Scherer et al., Reference Scherer, Palazzo and Seidl2013). Ultimately, our analysis shows that the effectiveness of the repair approach cannot be divorced from the local political-economic context in which it is adopted.
Limitations and Future Research Directions
This study, like all other research works, has limitations. However, it can serve as a stepping stone to future research. While the generalizability of our findings based on a study of two incidents relating to a large MNC in two Chinese regions is limited, our arguments raised in this study could be applicable to many other MNCs facing unfavorable environments in China and other emerging markets. In the China context, there are more developed regions in which the government perception of foreign retailers’ contributions to local economies is low, and likely the degree of unfavorability of those environments is high (granted that it may not be as high as that of Chongqing under Bo's regime). Future studies of MNCs in these regions can help test and extend our arguments.
In other emerging markets, there are contrasting local contexts characterized by differences in political leaders’ ideology and policy inclination. For example, in India, Pepsi and Coca Cola have encountered more public crises arising from product safety and misuse of local resources in the state of Kerala (Firstpost, 2017; Jo, Reference Jo2006), with a government of a leftist ideology, than in other states with an openness to FDI. The MNCs’ strategies to repair legitimacy in different local contexts within emerging economies warrant comparative studies.
The ways in which local governments in various emerging markets frame public crises and the MNCs’ organizational responses to the frames merit further investigation. Future studies can look into MNCs’ legitimacy repair in countries such as India, Russia, and Brazil, where large foreign corporations have frequently encountered public crises in recent years. In such contexts, the extent to which the relationship between regulatory legitimacy and social legitimacy resembles or differs from China's emerging markets deserves further study. In particular, the challenges for MNCs’ legitimacy management can vary among emerging economies with a different degree of media control.
Our reflection of experience in doing qualitative research in China may benefit future researchers in several ways. First, the topic matters. The issue of negative incidents and legitimacy repair, while intriguing to academics, many consumers, and the public in general, is considered among the MNC personnel and government officials to be highly sensitive. Therefore, our extensive networks of local contacts across regions enabled us to get introduced to interviewees to encourage trust-based sharing. Second, although both the mainstream media and social media are under tight control by the governments, we realized that journalists, academics, and consumers were relatively open to unrecorded interviews. Overall, an ability to converse in local dialects without foreign accents could build faster and stronger rapport between interviewers and interviewees. Terms and jargons the interviewees used deepen our understanding of the socially constructivist nature of incidents with equivocality and the power of framing. Third, field visits and informal conversations with both internal and external stakeholders enabled us to immerse in the sociocultural context. Last, other sources of data such as corporate press releases, social media posts and media reports from both Chinese international media agencies are useful for the triangulation of interview data.
Practical Implications
Our research has practical implications for managers in international business. The first set of recommendations relates to legitimacy assessment. We highlight the vulnerability of MNCs when operating under institutional complexity, and the risks stemming from national and local environments on top of those related to their firm attributes. Second, we point to the susceptibility of MNCs to framing by host governments, which set the tone for the mass media in their reporting of negative incidents and shape the leeway for organizational responses. Our findings suggest that MNCs should assess the degree of negative framing that they face and the type of institutional environment when developing legitimacy repair approaches. Last, our findings, albeit tentative, caution against attempts to enact acts of counter-framing and defiance in a political context in which the government has substantial power over firms and the media. Essentially, organizations need to take the political context-legitimacy repair action fit into account.
