Representation of Occupational Incidents in the Corporate Social Responsibility Reports of Chinese State-owned Enterprises: A Discourse Analysis

Abstract

Businesses today are subjected to intense public scrutiny. Organizations need to handle external communication carefully to maintain a positive image in the public eye. When an unexpected event such as a workplace incident occurs, organizations face an increased challenge to address information disclosure in a way that can minimise the impact of the bad news on organizational reputation and legitimacy.
In Mainland China, the rate of work-related fatalities is considerably higher than that in developed countries (CCID, 2014). Over the last decade, the government, the press, trade associations and social groups in China have actively encouraged state-owned enterprises (SOEs) to include accident disclosures within their corporate social responsibility (CSR) annual report, a document SOEs issue to the general public about their performance in terms of social, economic, and environmental responsibilities. China’s increased participation in the global economy has also driven large Chinese firms to adapt their practices to meet global standards regarding transparent corporate reporting and responsible business practices. Prior studies of CSR reporting mainly focus on large Western firms. Less is known about organizational practices in countries where there is a relatively short history of CSR engagement, such as China. An analysis of accident disclosures in the CSR reports of Chinese SOEs can thus add to our understanding of CSR and accident reporting in the Chinese context.
This thesis aims at examining the way in which workplace incidents are represented in the corporate social responsibility reports of Chinese state-owned enterprises. Based on a collection of 97 CSR reports (2.1 million tokens) published over the period 2005-2014 by 26 Chinese SOEs, the study conducts an in-depth investigation of accident disclosures within the sample texts through four major steps of analysis, each focusing on one aspect of the SOEs’ discourse. Theories and methods of content analysis (Krippendorff, 2004), genre analysis (Bhatia, 1993, 2004), legitimation in discourse (Benoit, 1997; Suchman, 1995; van Leeuwen, 2007) and critical discourse analysis (Fairclough, 1989, 1995) have been applied to the examination of the sample texts.
The first step of analysis seeks to find what information regarding workplace incidents has been disclosed in the sample reports. Methods of content analysis (Krippendorff, 2004), have been used to identify the presence, quantity, location, and categories of disclosures in the sample reports. It was found that the majority of the reports (85 out of 97) contain accident disclosures, although the information was not always disclosed in exact accordance with the reporting guidelines the firms claimed to have followed.
The second step of analysis explores the schematic structure of moves and steps for disclosure across the 85 sample reports that contain accident information. Adopting a genre-analytic approach to discourse (Bhatia, 1993, 2004), the analysis reveals that accident information as a form of “bad news” is typically disclosed within a schematic structure consisting of four basic moves: preparing the reader, conveying the bad news, mitigating the bad news, and reassuring the reader. By burying accident information strategically in positive content, firms attempt to soften the blow of the bad news they disclose.
The third step of analysis focuses on a smaller sample of 31 reports issued by firms with less satisfactory safety performance, namely, those had at least one fatal workplace incident during the reporting period. From a critical discourse perspective (Fairclough, 1989, 1995) and following frameworks of legitimation in discourse (Benoit, 1997; Suchman, 1995; van Leeuwen, 2007), this study investigates discursive strategies used by the poorly-performing firms for the purpose of self-legitimation. The findings suggest that firms tend to repair organizational image and legitimacy by directing reader attention to positive aspects of performance, and by constructing themselves as survivors fighting against external disasters.
The fourth and final step of analysis looks into disclosures made by Chinese SOEs from a comparative perspective. The study sets out to compare accident disclosures within the CSR reports of Chinese Fortune 500 SOEs (10 sample reports) and UK Fortune 500 firms (10 sample reports). The same methods that were used in the first, second, and third steps of study were applied to the analysis of the UK reports. Results show that accident disclosures from Chinese and UK firms are broadly similar in their content and form. Yet differences have been observed at a more detailed level, which reflects the influence of social, institutional, and cultural factors on the way in which the two groups of firms report CSR.
Overall, the study has mobilised a number of methods, theories and approaches to examine the content and presentation of accident information in the CSR reports of Chinese SOEs. Throughout the study, the unifying thread has been a focus on the nature and function of CSR discourse in a Chinese organizational context. By considering the way CSR reporting in Chinese SOEs is shaped by social, cultural, and institutional configurations at the local and global level, this study contributes to the understanding of the complexity and heterogeneity of corporate discourse in a transitional country where there is an increasing degree of globalization and marketization, and yet corporate behaviours are profoundly shaped by national culture and government signals.

Date of Award	9 Aug 2017
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	John FLOWERDEW (Supervisor) & Stephen BREMNER (Supervisor)