Accounting Reporting

lable at ScienceDirect

Journal of Cleaner Production 140 (2017) 1287e1297

Contents lists avai

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

The BP Gulf of Mexico oil spill: Exploring the link between social and environmental disclosures and reputation risk management

Mitali P. Arora, Sumit Lodhia*

Centre for Accounting, Governance and Sustainability, University of South Australia, Australia

a r t i c l e i n f o

Article history: Received 12 November 2015 Received in revised form 4 October 2016 Accepted 8 October 2016 Available online 11 October 2016

Keywords: Social and environmental disclosures Reputation risk management BP Gulf of Mexico Oil spill

* Corresponding author. E-mail address: sumit.lodhia@unisa.edu.au (S. Lod

http://dx.doi.org/10.1016/j.jclepro.2016.10.027 0959-6526/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

The aim of this study is to explore how social and environmental disclosures are used to manage reputation during a major incident, through the use of British Petroleum Plc's (BP) Gulf of Mexico crisis as a case study. It is argued that reputation is distinct from legitimacy and that the immediate actions undertaken by an organisation during a crisis are related to reputation management. Using a qualitative case study approach, a thematic analysis of BP's social and environmental disclosures on its website during the time period of this incident (April 21 e September 19, 2010) was undertaken to enable an understanding of the reputation risk management activities of BP. The key findings of this research revealed that BP used social and environmental disclosures on its website extensively to manage its reputation risk arising from the Gulf of Mexico crisis. The company was engrossed in providing accounts of its world class facilities and superior quality of management, diverting attention away from the severe environmental damage caused by the massive oil spill. It was observed that BP consistently used the corrective actions strategy by disclosing extensive information about its plans to solve the problem, and not its actual social and environmental actions during the disaster. The predominant reputation man- aging strategy used was the bolstering strategy, whereby BP stressed on its good traits when disclosing its social and environmental actions during the crisis.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Social and environmental disclosures can be used by manage- ment to extend their accountability from traditional economic performance reporting to providing accounts of a business's non- financial performance (Lodhia, 2013). Prior literature that has studied reporting practices during adverse events has highlighted that companies report negative information superficially (Douglas et al., 2004), portraying a one-sided view of the issues faced (Adams, 2004) and using disclosures to compensate for environ- mental degradation (Gray and Bebbington, 2000). The purpose of this paper is to explore how organisations disclose social and environmental information during a major environmental incident to manage their reputation. To address this objective, this study used the case of BP's Deepwater Horizon explosion on 20 April 2010 and the subsequent Gulf of Mexico oil spill (Steffy, 2011).

A significant number of studies have employed legitimacy the- ory for explaining the disclosure of social and environmental

hia).

information (Branco et al., 2008; Buhr, 1998; Cho, 2009; Islam and Deegan, 2008; Milne and Patten, 2002; O'Donovan, 2002; Pellegrino and Lodhia, 2012). This theory suggests that organisa- tions disclose social and environmental information to legitimise their existence to society (Dowling and Pfeffer, 1975; Lindblom, 1994). It also highlights that when faced with a crisis situation, companies will choose to highlight positive information and shift the attention away from negative issues (Milne and Patten, 2002; O'Donovan, 2002). However, it is observed that there is an under- lying need for a refined understanding of the social and environ- mental reporting phenomenon (Unerman, 2008), especially during a major incident. As researchers explore the disclosures of social and environmental issues, the most recent explanation is that of reputation Risk Management (RRM), a perspective offered by Bebbington et al. (2008). It has been argued that the RRM perspective has the potential to provide detailed insights into corporate disclosure during an ongoing event such as a major incident through a focus on the direct, immediate actions of an organisation (Bebbington et al., 2008). This approach is used to explore how BP used social and environmental disclosures during the Gulf of Mexico oil spill to manage its reputation.

Table 1 Differences between legitimacy and reputation.

Legitimacy Reputation

Long-term actions Immediate actions Longer term response Direct response Single organisation focus Comparisons with organisations Wider social status Bankable asset

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e12971288

The Gulf of Mexico oil spill incident (also known as the Deep- water Horizon incident) is one of the world's largest offshore oil spills (BBC, 2010). BP's Gulf of Mexico oil spill has severely affected the environment in the Gulf region and the surrounding area. This incident is defined in this paper as the actual explosion and the subsequent oil spill, which concluded when the oil spill was finally controlled. The incident has had an adverse impact on the repu- tation of oil companies around the world and questioned the ethics of offshore drilling. The case of BP's Gulf of Mexico oil spill is selected for this study with a view to understand how companies disclose social and environmental issues during a crisis. The moti- vation for selecting BP as a case study is as follows. Firstly, the Gulf of Mexico oil spill is one of the largest oil spills in history (BBC, 2010; Ramseur and Hagerty, 2013), which makes it an environ- mental crisis of significant ethical importance. Secondly, the com- pany responsible for this crisis, BP Oil, operates in the oil and gas industry, which is inherently an environmentally sensitive in- dustry. As a result, an incident of this magnitude necessitates the industry and the company involved to respond effectively (Patten, 1992; Summerhays and de Villiers, 2012). Thirdly, BP had sub- stantial social and environmental information available on its website about its response to the incident. These factors make the BP Gulf of Mexico incident an ideal case for a study into under- standing corporate disclosure practices as a response to an envi- ronmental disaster.

This research provides a new perspective to the literature on social and environmental accounting through its exploration of social and environmental disclosures on a corporate website during an ongoing incident. The findings illustrate that the application of the RRM perspective to BP's website disclosures during the Gulf of Mexico crisis provides an in-depth and detailed understanding of such disclosure. This study has implications for corporate reputa- tion management during a crisis and is also useful to stakeholders in comprehending corporate approaches to restore their image as a result of a crisis.

2. Theoretical perspective e reputation risk management

2.1. Corporate reputation and legitimacy

Corporate reputation has been defined in many ways. Prior literature suggests that corporate reputation is the perception of stakeholders of the organisation as ‘good’ or ‘bad’ (Tucker and Melewar, 2005). Fombrun (1996) highlights that corporate repu- tation produces tangible benefits to an organisation such as higher prices for products, lower costs for capital and labour, greater loy- alty from employees and a cushion of goodwill. Similarly, Fombrun and Van Riel (1997, p.10), drawing from multiple disciplines, state that “a corporate reputation is a collective representation of a firm's past actions and results that describes firm's ability to deliver valued outcomes to multiple stakeholders”. These definitions highlight that reputation provides a strategic advantage to an organisation and enables it to portray itself favourably to the various internal and external stakeholders.

Legitimacy focuses on the society's acceptance of an organisa- tion's activities. It has been described as a ‘condition or status which exists when an entity's value system is congruent with the value system of the larger social system of which the entity is a part’ (Lindblom,1994, p.2). Whenever a disparity exists between the two value systems, there is a threat to the entity's legitimacy (Dowling and Pfeffer, 1975).

Legitimacy has a broader and long-term focus, something that requires sustained evidence of performance (Lindblom, 1994). This can be contrasted to reputation which is immediate and can be managed directly by an organisation through specific activities at a

particular point in time (Deephouse and Carter, 2005; Deephouse and Suchman, 2008).

The distinction between reputation and legitimacy is a vexed issue for researchers with some using these terms interchangeably. However, Deephouse and Carter (2005) maintain that legitimacy emphasises the social acceptance resulting from adherence to so- cial norms and expectations whereas reputation emphasises comparisons among organisations. They also highlight that whilst legitimacy is often dependent on regulative, normative and cognitive dimensions, reputation transcends beyond these to encompass “virtually any attribute along which organisations may vary that can serve as a source of status comparisons” (Deephouse and Carter, 2005, p.332). Deephouse and Suchman (2008) take this further by stating that reputation differs from legitimacy and status by being a continuous measure of performance and being a fundament differentiator among organisations. They highlight that favourable reputation is often a strategic resource that firms can exploit for competitive advantage. Bitektine (2011) further em- phasises the differences between legitimacy, status and reputation from an evaluator's perspective, suggesting that judgements made on these concepts are categorically different. Drawing upon this preceding literature, Kuruppu and Milne (2014) conclude that reputation is a bankable asset managed through direct action, especially during negative incidents, while legitimacy is a wider social status that requires companies to conform to broader societal values and legal policies. Table 1 summarises the differences be- tween reputation and legitimacy.

Having distinguished between reputation and legitimacy, it is argued that it is the threat to an organisation's reputation which increases when there is a crisis such as the BP Oil spill, motivating the organisation to manage its reputation. Legitimacy on the other hand is a wider social condition which needs to be repaired in the longer term should the reputation management activities be un- successful. Prior literature has confined itself to annual report disclosure which enables an assessment of the legitimacy status of an organisation over a period of time since the occurrence of an incident. There is a need for studies that actually examine the im- mediate, direct actions of an organisation when a crisis occurs. This would require a focus on RRM, which enables an assessment of how organisations attempt to remedy the damage to their reputation. Advances in communication approaches more recently provide an understanding of how organisations manage their reputation through more immediate disclosure such as through the news media or their corporate websites (Lodhia, 2012, 2014).

According to surveys conducted by KPMG (2013) and Global Risk Management (2015), reputational consideration has recently been recognised as one of the major risks for companies. A recent study undertaken by Melo and Garrido-Morgado (2012) finds that strong corporate social reporting disclosure positively affects corporate reputation. The RRM perspective acknowledges that companies are motivated to disclose social and environmental in- formation as the means to manage their reputation. It is argued in the next section that RRM has the potential to provide a more detailed framework to explore a link between a company's disclo- sures and its attempts to manage reputation, especially during a major incident.

Table 2 Image restoration strategies.

Strategy Key Characteristic

Denial Simple denial Did not perform the act Shift the blame Act performed by someone else

Evasion of responsibility Provocation Responded to the act of another Defeasibility Lack of information or ability Accident Act was a mishap Good intentions Meant well in act

Reducing offensiveness of event Bolstering Stress good traits Minimisation Act not serious Differentiation Act less offensive Transcendence More important considerations Attack accuser Reduce credibility of accuser Compensation Reimburse victim

Corrective action Plan to solve or prevent problem Mortification Apologize for the act

Source: Benoit (1997, p. 179).

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e1297 1289

2.2. Reputation risk management

The RRM perspective is based on the notion that companies will use social and environmental disclosures in order to maintain their reputation, or manage any potential risk to their reputation (Bebbington et al., 2008). Unerman (2008) emphasised the signif- icance of using RRM in providing many valuable insights for un- derstanding underlying approaches to disclosing social and environmental information. Both Deegan et al. (2002) and Unerman (2008) opine that the current dominant theories used for explaining organisational approaches to reporting social and environmental information are broad theories in that they do not cover other possible detailed explanations. The proposed RRM theoretical model allows for a detailed analysis of social and envi- ronmental disclosures. It is particularly useful for exploring the immediate actions of an organisation during an incident, which as discussed in the preceding section, requires an emphasis on repu- tation rather than on legitimacy.

The RRM perspective offers a more “fine-grained analysis” (Bebbington et al., 2008, p.339) particularly for analysing content within social and environmental disclosures. According to Bebbington et al. (2008, p.338), the RRM perspective puts “flesh on the bones of legitimacy theory explanations”. The authors suggest that RRM provides one of the many possible positions for viewing corporate social reporting activities undertaken by companies.

To demonstrate the applicability of the RRM perspective, Bebbington et al. (2008) had analysed Shell's 2002 Corporate Social Responsibility report. They found considerable evidence from their analysis to support the RRM perspective for most of the disclosures reported. The authors suggest for further analysis to be carried out, especially in situations where an organisation has suffered a reputation damaging event. Hogan and Lodhia (2011) provided further empirical support to the RRM perspective through a study into the social and environmental disclosures of an Australian mining company pre and post a reputation damaging event of proposed carbon pricing regulation.

To facilitate an in-depth analysis of social and environmental disclosures, Bebbington et al. (2008) suggest examining these to understand how they relate to reputation management. To assess how disclosures relate to the reputation management perspective, they have proposed to study them in light of the elements of reputation. Their study identified five key elements of reputation, these are:

1 Financial performance, 2 Quality of management, 3 Social and environmental responsibility performance, 4 Employee quality, and 5 Quality of goods/services provided.

Therefore, according to Bebbington et al. (2008) if a company discloses social and environmental information predominantly around the five key elements of reputation during significant events such a crisis, it is highly likely that such disclosures are made with a view to manage reputation. These elements are also not mutually exclusive and it is possible for interrelationships between the various elements. An organisation could focus on a similar or identical approach in addressing several elements of reputation.

The RRM framework provides an understanding of the various reputation elements that could be disclosed extensively during an incident. However, there is a need to explore these reputation risk management disclosures further by analysing their role in attempts to repair the damaged reputation during an incident. As such, Bebbington et al. (2008) recommend the use of Benoit's image restoration strategies as the second component of this framework

for exploring social and environmental disclosures (see Table 2). They have observed that a company could use such image resto- ration strategies when responding to an incident.

The RRM perspective (Bebbington et al., 2008, p. 342) explains the significance of Benoit's image restoration strategies as providing a “conceptual link between RRM activities undertaken by organisations and discourses that one may observe as a result of the need for image restoration strategies”. Benoit (1997) suggests that when faced with a reputation damaging situation, companies choose different strategies to manage their reputation. Bebbington et al. (2008) suggested applying Benoit's (1997) image restoration strategies to social and environmental information to provide a detailed understanding of how companies disclose these issues in order to maintain their reputation.

To sum up, the RRM perspective as suggested by Bebbington et al. (2008) constitutes the five key elements of reputation, and the image restoration strategies developed by Benoit (1997). This perspective provides an understanding of how social and envi- ronmental disclosures can be used to manage reputation, especially during a major incident.

The prior discussions provide a basis for exploring the differ- ences between legitimacy theory and reputation risk management. The RRM perspective emphasises issues such as quality of man- agement and employees, and financial performance, which are barely discussed in studies using legitimacy theory. This could be due to the fact that reputation is a strategic resource that can differentiate an organisation from other organisations. Moreover, the use of the Benoit's image restoration strategies in RRM provides a more extensive exploration of the immediate, direct actions of an organisation in response to a crisis, compared to the legitimation tactics used by an organisation in response to repairing legitimacy, as suggested for example by O'Donovan (2002). Legitimacy stra- tegies have a long-term focus and require sustained evidence to ensure that organisational practices are congruent with societal explanations.

It is evident from prior literature that when companies are faced with a negative incident, they will use their social and environ- mental reporting as a tool to manage their legitimacy (Botes and Samkin, 2013; Cho, 2009; Deegan and Rankin, 1996; Patten, 1992; Deegan et al., 2000; Summerhays and de Villiers, 2012). Other studies have also linked impressions management to legitimacy strategies, suggesting that companies often portray a positive im- age even when faced with a reputation damaging situation (Michelon, 2012). The focus on periodic disclosures has limited

1 The incident occurred on the evening of April 20th and therefore, the focus of the analysis is from the first day after the incident as social and environmental disclosures on BP's website on April 20 are unlikely to relate to the spill. It was found that disclosures related to the spill emerged one day after the incident.

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e12971290

these studies to legitimacy explanations. However, studies on corporate communication have focused on reputation manage- ment, especially during a crisis. Examples include studies by Seeger and Ulmer (2003) on Enron and Waller and Conaway (2011) on Nike. More recent literature has focussed on Benoit's image resto- ration strategies (but not on reputation specifically), such as Harlow et al. (2011) specifically on BP and Lauzen (2015) on sexism in the Cannes Film festival selection process. Harlow et al. (2011) state that BP's strategies centred on describing what they were doing to correct the problem and compensate the victims, but did not include strategies such as shifting the blame to the other companies involved nor admitting their own blame. Similarly, Lauzen (2015) highlights image restoration strategies are not successful if there is no mortification or correction of action. Drawing upon such corporate communication literature, this study highlights that the RRM perspective has the potential to enhance social and environ- mental accounting literature.

Hopwood (2009, p. 437) commented that companies use their social and environmental reporting as a tool to “facilitate the con- struction of a new and different image of the company”. He stated that instead of using social and environmental disclosures to communicate the social and environmental performance of an organisation it can be used as a “corporate veil” (Hopwood, 2009, p. 437) to protect the negative information from the public at large. Similarly, Cho et al. (2015) highlight that contradictory societal and institutional pressures lead organisations to engage in hypocrisy and develop façades, leading to questions about whether social and environmental disclosures could ever lead to a less unsustainable planet. To examine these claims, there is an underlying need for a refined understanding of social and environmental disclosures (Unerman, 2008) than that undertaken in prior literature which has been limited to legitimacy theory explanations. As explained in this section, legitimacy and reputation differ and there is a need to move beyond legitimacy theory explanations used in the prior literature discussed here to comprehend the direct, ongoing actions undertaken immediately after a major incident. This requires an emphasis on reputation, rather than legitimacy. The RRM perspective provides a refined framework to examine social and environmental disclosures in order to obtain in-depth under- standing of managerial approaches to communicating social and environmental information during such an incident.

In summary, prior literature on social and environmental dis- closures during an incident has been limited to legitimacy theory explanations and a focus on periodic disclosure in annual or sus- tainability reports. The current study addresses these gaps in the literature. Using BP's Gulf of Mexico crisis as a context, this study explores the social and environmental disclosures of BP through a timely medium, the Web. The incident provides scope for exploring the ongoing disclosures on the corporate website, enabling an assessment of the direct actions undertaken by the organisation to manage its reputation during the crisis. This study utilises the RRM perspective which focuses on both the elements of reputation and Benoit's image restoration strategies.

3. Research methods

Adopting the case study approach, this study used qualitative research analysis employing the RRM framework. According to Yin (2013), case study research allows the examination of a single case, in depth, in order to analyse the case in a particular context. Case study research facilitates an in-depth understanding of a real-life phenomenon to shed some light on the decisions that were taken in the case, and explains why they were taken and how they were implemented.

Flyvbjerg (2006) explains that findings from a case study leads

to a substantial increase in knowledge which will further assist in the better understanding of a phenomenon. Yin (2013) suggests that a case study is the most appropriate method of research when a study is exploratory in nature, seeking to answer the ‘how’ or ‘why’ questions. As the aim of this study is to explain how orga- nisations use social and environmental disclosures to manage reputation, the case study approach is justifiable.

Within the context of the generalisability of research using a single case study, Flyvbjerg (2006) emphasises “that knowledge cannot be formally generalised does not mean that it cannot enter into the collective process of knowledge accumulation in a given field or in a society.” A single case study may not meet the criteria for generalising from a study using multiple cases; however, in- sights from the immense unaltered wealth of knowledge gained by studying a single case should not be discredited. Yin (2013) explains that a case study provides analytical generalisations rather than statistical generalisations.

Gerring (2007) highlights that choosing a case is the most important decision while attempting to use the case study approach. He has suggested that a case that is chosen for such research should contain the elements of the phenomenon that the study is seeking to explain. The appropriateness of the BP Gulf of Mexico Oil Spill case for the current study will be discussed in the next section.

Data was collected from the BP website. All the web disclosures from the start date of the incident, 21 April 2010, up till 19 September 20101 were collected. This specific time frame was selected because during this period, the oil spill was controlled and BP assumed its responsibilities regarding the incident. Using Internet Archive's Wayback Machine (www.archive.org/web/web. php), BP Oil's web pages during the time of incident were accessed. This tool allows access to archived website information that was captured each day as it appeared on the original website. There is a six month time lag in the information content available through Wayback machine but this was not a problem for the current research. Through the links on the front page of an archived web page, the ‘environment and society’ page on BP's website was accessed. Web disclosures encompassed standalone social and environmental information on the website as well as the avail- ability of information disclosed in other media on the website such as, reports or press releases in Portable Document Format (PDF).

The data collected was analysed using thematic analysis (Bowen, 2009; Green et al., 2007), which is often used for analysing narrative data. Given the depth of narrative data available on the Gulf of Mexico oil spill, this approach was the most appropriate for the current research. Thematic analysis is a rigorous process of sorting and classifying data based on dominant themes which assist in addressing the research question. This approach is consistent with the pattern matching approach suggested by Yin (2013) to ensure internal validity of case studies. The thematic analysis for this study focused on how BP Oil used its social and environmental disclosures to manage its reputation. Themes under this analytical approach are identified from the theoretical approach selected for the study and therefore, the application of the RRM framework assisted in examining the link between social and environmental disclosures and reputation management. To ensure inter-coder reliability, both the researchers independently analysed the data, then compared their findings and reached consensus on the data.

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e1297 1291

The thematic analysis was conducted in two parts. Firstly, the analysis identified which of the five elements that conceptualise reputation were used in order to build or maintain reputation. The data was then classified into these elements. Secondly, Benoit's (1997) image restoration strategies were applied to the data collected. This yielded information about the specific strategies BP used to disclose information relating to the crisis, in order to manage its reputation. The thematic analysis assisted in under- standing how companies use their social and environmental dis- closures to enhance reputation.

This study used secondary information collected from BP's website for addressing the research objective. As the scope and timing of this study was limited, it did not allow for direct engagement with BP to understand motivations for disclosing so- cial and environmental information during the crisis. Moreover, it can be difficult to gain access to engage with the managers of companies involved in a crisis situation.

4. The Gulf of Mexico oil spill

The BP Gulf of Mexico oil spill incident is one of the world's largest offshore oil spills (BBC, 2010). The incident occurred on the night of 20th April 2010, when a mobile offshore oil rig, Deepwater Horizon, exploded, caught fire, and consequently sank in the Gulf of Mexico. The oil rig was owned and operated by a company called Transocean and was leased by BP (King, 2013). Although a technical reason which has been provided for the oil spill is the failure of the blowout preventer (BOP), a number of studies have been conducted to understand causes that amounted to such an accident. This ac- cident claimed the lives of 11 men working on the rig, caused 17 injuries and millions of gallons of oil spill that lasted for approxi- mately 83 days (DHSG, 2011). As BP had the drilling rights for the Gulf of Mexico area, and had subcontracted the drilling work to Transocean, BP had to bear maximum responsibility for the acci- dent (King, 2013).

The BP Gulf of Mexico oil spill has caused considerable damage to the wildlife, the local coastal communities and a number of fishing and tourism businesses that depended on the Gulf for their day to day business (Cherry and Sneirson, 2011). This accident also questioned the business practices and ethics of oil companies around the world. In essence, even though this was an environ- mental incident, social and ethical impacts were associated with it. The loss of lives and livelihood questioned BP's social and ethical responsibilities to its stakeholders.

Studies have highlighted that the failure in BP's company culture and business practices (Hoffman and Jennings, 2011; McNicholas and Windsor, 2011) would have accumulated and ultimately led to the incident. Other studies have also pointed at lax regulation and safety culture within BP (Cherry and Sneirson, 2011; Cohen et al., 2011; McNicholas and Windsor, 2011) that could have caused the accident. The consequences of the environmental disaster questioned the operations of the company's ethics and its reputation.

5. Findings

5.1. Evidence of reputation risk management using the five elements of reputation

This section explains the extent to which BP used its social and environmental disclosures to manage its reputation during a crisis situation. A thematic analysis of BP's social and environmental disclosures has highlighted evidence for the five elements of reputation discussed in the literature.

5.1.1. Financial performance BP attempted to manage its reputation by highlighting its

financial performance, and diverting the attention of readers from its negative financial performance and reinforcing its commitment to regularly disclosing social and environmental issues faced during the Gulf of Mexico crisis.

Financial performance is characterised by information on corporate assets and financial soundness (Bebbington et al., 2008):

The Board believes that there are opportunities to divest assets which are strategically more valuable to other parties than they are to BP. (BP Press release, 20 July 2010c).

BP made the above statement upon selling the first of many assets it divested to raise cash to fund the relief work obligations. This was a bold statement giving an impression of BP's strong asset base, as the company claimed that such assets were ‘more valuable’ to other parties than to BP.

Amidst the critical incident with billions of dollars of future obligations arising due to the incident, BP published the following statement:

Under the current trading environment, we are generating sig- nificant additional cash flow. In addition, our gearing is currently below the bottom of our targeted range. Our asset base is strong and valuable, with more than 18bn barrels of proved reserves and 63bn barrels of resources as at the end of 2009. All of the above gives us significant capacity and flexibility in dealing with the cost of responding to the incident, the envi- ronmental remediation and the payment of legitimate claims. (BP Press release, 10 June 2010c)

This statement reinforces management's intention to boost the image of its financial performance. Even amongst such a turbulent financial situation, BP continued to portray a positive financial image, leading to the impression of being financially sound.

Furthermore, it was observed that BP posted a loss of $17,150 m for its second quarter in 2010 as opposed to a profit of $6,079 m reported during the first quarter of 2010 (BP, 2010c). This clearly shows that BP's financial position was straining as the crisis response activities were hampering its financial position. However, BP continued to disclose more financially positive information than they normally would.

BP's share price fell to a 13 year low in June 2010 owing to the environmental damage in the Gulf region and the accompanying extraordinary cash outflow for the response activities (Wearden, 2010). However, in its press release on the 9th of June 2010, BP stated that it was not aware of any reason for the fall in share price, but reiterated that its priority was to disclose information regarding the Gulf of Mexico oil spill incident. BP was clearly diverting the attention of readers from its falling share price to its commitment to disclosing social and environmental information. This can be interpreted as BP's attempts to manage its reputation by trying to distract attention from a negative issue, like the share price fall, to a positive issue, like their commitment to keeping people updated about the incident.

5.1.2. Quality of management According to the reputation ranking studies (as summarised by

Bebbington et al., 2008), a good reputation is created by good quality management. BP's explicit emphasis on disclosing infor- mation about it being a strong firm, its strong financial position, and its management's ability to tackle such situation was an attempt to project its management positively.

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e12971292

There were many instances where BP disclosed the superior nature of its management, showing its ability and capacity in dealing with a situation like the Gulf of Mexico crisis. The first of such evidence is captured in the initial disclosures that BP made regarding the incident. Even before deploying all resources for the response activity, BP made the following statement about its fa- cilities to deal with the crisis:

We have assembled and are now deploying world-class facil- ities, resources and expertise, and can call on more if needed. (BP Press release, 22 April 2010c)

This statement was made on the 22nd of April 2010, which was just hours after the spill began. It was an attempt by BP's man- agement to exude confidence in its ability to deal with a difficult situation with ease, even though one could question why the oil spill took place in the first instance if the company had such “world class facilities and expertise”. Drifting away from the consequences of the massive explosion at the rig, BP highlighted the assembling of its world-class facilities in its disclosure content just days after the explosion.

A week after the incident, BP's management confidently high- lighted that they would be judged by the success of dealing with this incident:

We will be judged by the success we have in dealing with this incident and we are determined to succeed. (BP Press release, 30 April 2010c)

Similarly, emphasising its ability to deal with difficult situations, the board stated on the 4th of June 2010, that:

The financial consequences of this incident will undoubtedly be severe, but BP is a strong company and we have weathered many storms before. (BP Press release, 4 June 2010c)

This statement was published in a press release, summarising the communication of BP's chairman and CEO with its share- holders. It was an attempt to show BP's confidence in management to deal with such incidents.

However, the oil leak took a fair bit of time to be fixed and BP's response to this incident did not seem to be adequate to justify their initial overconfidence in dealing with this issue:

This is a significant milestone in the response to the Deepwater Horizon tragedy and is the final step in a complex and unprec- edented subsea operation e finally confirming that this well no longer presents a threat to the Gulf of Mexico. (BP Press release, 19 September 2010c)

5.1.3. Social and environmental performance Companies can focus on social, community and environmental

responsibility to highlight their social, ethical and environmental performance in order to manage their reputation (Bebbington et al., 2008). A detailed analysis of BP's social and environmental disclo- sures during the crisis has shown evidence of reputation manage- ment. There were various examples of positive social and environmental information being disclosed, even though the actual actions during this incident did not match the corporate narrative.

In its communication with shareholders, BP's then CEO, Tony Hayward implied that their reputation was at stake in dealing with this crisis and that they had to maintain clear and transparent communication with the government and the American people.

This led to BP disclosing detailed social and environmental infor- mation like regular press releases, providing photos and other comprehensive information during the crisis.

BP disclosed positive social and environmental information from the start of the incident, regardless of the severe implications of the spill on the environment. An example is presented in the following quote:

The safety of the people working offshore is our top priority and the improved weather has created better conditions for our response, said BP Group Chief Executive Tony Hayward. This, combined with the light, thin oil we are dealing with has further increased our confidence that we can tackle this spill offshore. (BP Press release, 26 April 2010c)

Through the above quote, BP firstly described its priority to safety. The statement made by the CEO is an attempt to emphasise the attention that BP paid regarding the safety of its employees, highlighting positive social information. Secondly, BP described the oil from the spill as ‘thin’ and ‘light’. These words were carefully chosen to minimise the ostensible risks related to an oil spill. This shows BP's intention to present positive actions through the use of its social and environmental disclosures during the crisis.

Another example of BP's positive social and environmental disclosures is demonstrated through the following quote:

BP is doing everything we can to make this right. We continue to seal the Macondo well permanently, clean up the environment, and make sure that people are compensated for legitimate claims. (BP Press release, 30 April 2010c)

This statement highlights BP's attempt to portray its dedication to the oil spill. The statement draws attention towards BP's commitment to the restoration process, and provides a positive assertion of its social and environmental actions. It is a deliberate attempt to portray BP's willingness to extend its social and envi- ronmental responsibility to compensate those affected by the oil spill. As discussed earlier, in spite of these assurances, the oil spill took longer to control than originally anticipated.

BP provided detailed social and environmental information during the event such as disclosing its commitment to, and the steps taken to, restore the Gulf, and intentions to donate money. The CEO's note to the shareholders about providing adequate in- formation in response to the crisis was a reiteration of this. This is illustrated in the following quote:

In large part I believe that our reputation, rests on how we respond to the incident itself, and on this we are following four key principles: One, we will be transparent with the govern- ment and the American people in what we do and what we say. (BP Investor briefing, 4 June 2010b)

Clearly, positive social and environmental information domi- nated BP's disclosures in an attempt to deflect attention from the oil spill.

5.1.4. Quality of employees In addition to focusing on the quality of management, emphasis

on the quality of employees can be used to manage reputation (Bebbington et al., 2008). The quality of employees disclosures assisted in diverting the attention from the impact of the spill on locals to what the company was doing to manage the spill. This selective dissemination of information was used to manage BP's reputational risk arising from the Gulf of Mexico Oil Spill.

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e1297 1293

BP explicitly disclosed information relating to the quality of employees. On its website, BP published a number of videos showing people involved in working with the relief operations. Most of the videos showed people cleaning the beach, handling the spill off-shore and working in conjunction with other authorities. The videos highlighted people working with each other in groups, depicting a strong harmonious work culture. What stood out from these disclosures were specific reports that described the commitment of BP's employees.

BP's social and environmental disclosures on the website included a section on stories compiled by reporters. These stories claimed to capture the stories of people who are “most immediately affected by the oil spill” (BP, 2010a). One of the stories reported the commitment of an employee, titled ‘the tale of dedicated responder’. This story outlines the commitment of a BP employee who worked for 26 days without a break, day and night, to assist with the response work. According to the report, he took leave only when BP advised him to take a break. This was intended to be a heart-warming story to portray a strong positive impression about BP's employees' commitment to the oil spill response. The story is not only an attempt to shed light on the quality of workers, but also BP's employee management as the employees responded to man- agement's commitment to stop the oil spill. However, the motive for disclosing such a story can be questioned, given the damage caused by the oil spill in the first place. Moreover, this story ques- tions the company's human resource practices, highlighting that either the company was not aware of what the employee was doing or it enabled someone to work for such a long time period without concern for the employee's health or wellbeing.

What was disappointing was that these stories did not cover any impacts of the spill, or suffering faced by locals, which would have outnumbered the stories of such dedicated responders. Had the purpose of these stories been to highlight the conditions in the Gulf coast, there would have been many accounts of how the spill impacted local residents and businesses in their day to day life. In reality, these reports through their focus on employees only captured BP's side of the story. This questions the purpose of including such disclosures in the first place.

5.1.5. Quality of goods and services An examination of BP's social and environmental information on

its website during the incident found limited evidence of disclo- sures discussing the quality of goods and services. Even though the spill's entire focus related to crude oil (which not the final product as it needs to be further refined before being produced as petrol), there was emphasis on the worldwide needs for energy. It appeared that the oil spill could be justified as a by-product of BP's pursuit of a resource that is critical for success of modern economies.

5.2. Link between social and environmental disclosures and Benoit's (1997) image restoration strategies

This section discusses the results of applying Benoit's (1997) image restoration strategies (the second component of the RRM perspective) to BP's social and environmental disclosures during the incident. These image restoration strategies expand the dis- cussion of the elements of reputation discussed previously and further highlight the use of social and environmental disclosures to manage BP's reputational risk.

BP's social and environmental information disclosed on its website could be related to a number of image restoration strate- gies. These strategies are related to the disclosure during the actual incident and therefore, absence of any particular strategies does not imply that these were not used at other times or through different media.

According to the thematic analysis, it was evident that the strategy that was dominantly used was bolstering. Under the bolstering strategy, the company discloses information about its good traits in order to manage its reputation during a crisis situa- tion (Benoit, 1997).

BP used its website to shift the focus from the negative effects of the oil spill, to more positive issues like the company deploying its finest resources to manage the massive oil spill. Most of the infor- mation disclosed during the oil spill was centred on BP's superior capability of controlling the spill as discussed under the quality of management reputation element. An example of this bolstering strategy is presented in one of the disclosures published by BP during the initial days of the incident. BP disclosed the following statement on 30th April 2010, days after the incident took place:

We are doing absolutely everything in our power to eliminate the source of the leak and contain the environmental impact of the spill. We are determined to fight this spill on all fronts, in the deep waters of the Gulf, in the shallow waters and, should it be necessary, on the shore. (BP Press release, 30 April 2010c)

Such disclosures continued while the company attempted to control the leak:

We will halt this spill and put right the damage that has been done. We will rebuild the confidence of the American people and the world in BP. (BP Press release, 4 June 2010c).

These statements were an attempt to displays BP's confidence in its ability to handle the incident by stressing its commitment to stopping the oil spill. They focused on these good traits to bolster their reputation during the incident.

BP also attempted to deflect the attention from the oil spill to focus on the importance of energy. An example of such a bolstering strategy is as follows:

This is an inherently difficult business. The world needs energy and it trusted us to produce that in a safe and sustainable manner. We must regain that trust. (BP Investor briefing, 4 June 2010b)

This statement diverts the attention of the readers from the incident, to a more important issue, which is the worldwide need for energy. Here, BP attempted to emphasise on the world's energy needs and reiterate its role in providing this energy. BP highlighted the inherent difficulty of off-shore oil drilling, but at the same time, placed itself strategically to meet the energy needs of the world, hence justifying the incident. Using the above statement BP attempted to remind the stakeholders of its importance by bolstering about its position in meeting the energy needs of the world.

The second most evident strategy used by BP during the inci- dent was to disclose the corrective actions taken. Benoit (1997) classifies information describing plans to solve the problem as the corrective actions strategy. BP disclosed significant information that described the corrective action that it was taking in order to stop the oil spill. It gave continuous repetitive accounts of what corrective measures it was taking to respond to the incident. Ex- amples to demonstrate such behaviour are as follows:

We are determined to do everything in our power to contain this oil spill and resolve the situation as rapidly, safely and effectively as possible. (BP Press release, 22 April 2010c)

The Board of BP has been clear from the outset that all resources available to the company should be applied to meeting BP's

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e12971294

responsibilities in addressing these events. (BP Press release, 4 June 2010c)

These quotes highlight BP's intention to take corrective mea- sures for the incident. The then CEO, Tony Hayward, repeatedly underlined BP's place in repairing what had occurred. However, BP did not show the acceptance of responsibility of the incident through its disclosures, but emphasised taking sufficient corrective measures to control the impact arising from the incident. Such disclosures were made from time to time to highlight the actions taken.

The third dominant image restoration strategy employed by BP was that of transcendence. Transcendence involves strategically positioning the act in a more constructive position that shows positive outcomes as a result of the event (Benoit, 1997). In more than one instance, BP conveyed the message of learning greater lessons from the Gulf of Mexico incident. In BP's first investor briefing after the explosion, it explained the following to its stakeholders:

It should not have happened and we are bound and determined to learn every lesson to try and ensure it never happens again. (BP Investor briefing, 4 June 2010b)

The above statement persuades readers to overlook the past incident as a mere accident. Moreover this statement attempted to demonstrate BP's preparedness for such incidents in the future. This was intended to instil people's faith in the company and in line with the transcendence strategy, focuses on the positive outcomes of the Gulf of Mexico oil spill.

Another example that overwhelmingly indicates the transcen- dent nature of BP's disclosures is illustrated in the following quote from the same investor briefing on 4th June 2010 (BP, 2010b):

Overall, the paradigm in which BP, and the wider industry, operates needs to change. We need better safety technology. An example of this is the “fail safe” BOP,2 which this incident has clearly demonstrated is not failsafe. BP needs to, and will, take existing standards to a completely new level e we must.

In this statement, BP expressed an urgent need to overhaul its existing safety standards to a new level, in order to avoid such in- cidents in the future. However, BP's past social and environmental reports (BP, 2009) asserted its safety-first practices in all business operations. If this was the case, then the Gulf of Mexico incident should not have happened (Cherry and Sneirson, 2011). However, according to the above statement, BP foresees new and improved safety standards as a result of the current incident. This portrays the transcendent nature of BP's disclosures, by aspiring to be safer in the future.

BP extended its transcendence strategy by highlighting the benefit received by the hotel industry as a direct result of the oil spill. The following quote from BP's independent reporters in- dicates the transcendent nature of BP's disclosures:

Much of the region's other businesses e particularly the hotels e have been prospering because so many people have come here from BP and other oil emergency response teams. (BP Reports from the Gulf, 24 May 2010a)

2 Blowout Preventer as discussed in section 4.

BP's intention to highlight a positive consequence as a result of the oil spill is most certainly an attempt to use the transcendence strategy to divert the attention from those severely directly impacted by the oil spill and could be perceived as pushing the boundaries of acceptable corporate behaviour.

Another image restoration strategy used by BP during the inci- dent was minimisation. Under Benoit's (1997) image restoration strategies, minimisation is a strategy where a company facing crisis attempts to project that the crisis was not serious. To a certain extent, BP attempted to minimise the accident by disclosing se- lective information. The following quote from BP's press release just days after the incident illustrates this point clearly:

Given the current conditions and the massive size of our response, we are confident in our ability to tackle this spill offshore. (BP Press release, 25 April 2010c)

In the statement above, BP disclosed information that depicts the ease with which it will be able to tackle the incident and the following consequences. Only days into the incident, BP confidently expressed its ability to tackle the oil spill off-shore. This can be classified as BP's endeavour to minimise the consequences of the incident, even though it actually took a much longer time to control the spill.

By disclosing information such as “the oil response team had recovered more than 1000 barrels of oil-water mix of which the vast majority is water” (BP Press release, 25 April 2010c), BP most definitely attempted to minimise the impacts of the oil spill. This statement provides positive cues to stakeholders who are inter- ested in comprehending the impacts of the oil spill. By publishing such positive statements that minimised the seriousness of the incident, BP certainly attempted to protect its reputation.

Benoit's (1997) image restoration strategies include offering compensation to those affected, as a reputation managing strategy that reimburses the victim. While BP's Gulf of Mexico incident had severe consequences for the affected parties, not offering compensation was out of the question. BP's social and environ- mental information disclosed on its website included large quan- tities of information about BP's commitment to pay valid claims to affected parties. However, reiterating such information in multiple instances can be regarded as an attempt to discuss its commitment to compensation in order to maintain its reputation. An example of such disclosure is offered as follows:

As part of its commitment to restore the environment and habitats in the Gulf Coast region, BP today announced that it will donate the net revenue from oil recovered from the MC2523

spill to create a new wildlife fund to create, restore, improve and protect wildlife habitat along the coastline. (BP Press release, 8 June 2010c)

This quote was one of many that emphasised BP's commitment to repairing the damage caused to the environment and the wildlife as a result of the oil spill, thereby using compensation as strategy to restore its image.

At the inception of the incident, BP's initial disclosures were defeasible. Its first disclosure regarding the incident was merely a confirmation of the statement made by Transocean (the operator of the rig that exploded), as follows:

Transocean reported the fire earlier today on the rig, located approximately 41 miles offshore Louisiana on Mississippi

3 Mississippi Canyon Block Number 252 Rig.

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e1297 1295

Canyon Block 252, saying that a “substantial majority” of the 126 personnel on board were safe, but some crew members remained unaccounted for. A number of personnel were re- ported to be injured. (BP Press release, 21 April 2010c)

By confirming a formal statement made by Transocean, BP appeared to be unwilling to take responsibility for the accident. As this statement was made only hours after the incident, it is reasonable to assume that BP did not have sufficient information at that stage to accurately assess the extent to which it would have been responsible for the accident. Within the image restoration strategy, this act is classified as defeasibility (Benoit, 1997) where BP did not appear to be taking any responsibility on account of having insufficient information regarding the incident.

The most astounding observation made while analysing BP's social and environmental disclosures during the incident was that BP did not explicitly display any remorse for the incident. Apolo- gising for an act is one of the image restoration strategies described by Benoit (1997) as the mortification strategy. However, despite numerous trials and cases that held BP responsible for the incident, BP did not express any regret for its wrongdoings. The only apology offered by BP during the timeframe observed was during the in- vestors meeting on 4th June 2010, which was as follows:

The accident on April 20th cost 11 lives and led to an oil spill which has profoundly affected the people and the communities in the Gulf of Mexico. These events have shocked us all. We deeply regret and we are sorry for this tragic accident and its aftermath. (BP Investor briefing, 4 June 2010b)

According to Benoit's (1997) image restoration strategies, if BP apologised for its acts, the apology offered would be considered as a reputation managing activity. However, in this case, the apology was offered only at an investors meeting. This can be interpreted as offering an apology, but only to the investors and not the general public at large.

6. Discussion

BP's social and environmental information on its website was examined in light of the RRM perspective to explain how com- panies use their social and environmental disclosures as a tool to manage reputation during a crisis situation. The findings of this study support the usefulness of the RRM perspective in explaining organisational approaches to disclosing social and environmental information, consistent with previous studies by Bebbington et al. (2008) and Hogan and Lodhia (2011). These findings suggest that social and environmental information communicated by com- panies during a negative incident is disclosed with the primary intention to manage their reputational damage.

These findings expand prior literature that has sought to explain social and environmental disclosure practices of companies embroiled in negative events (Botes and Samkin, 2013; Cho, 2009; Deegan and Rankin, 1996; Patten, 1992; Michelon, 2012; Summerhays and de Villiers, 2012). These studies have confirmed that during a negative incident, where there has been significant damage to a company's reputation, it will disclose social and environmental information in order to legitimise its existence. The current study in line with corporate communication literature (see for example, Seeger and Ulmer, 2003; Waller and Conaway, 2011; Harlow et al., 2011; Lauzen, 2015) explains the strategies adopted by companies to manage their reputation and focuses on the at- tempts to manage reputational risk. This paper provides a more comprehensive framework for exploring reputation management

and in this process highlights that it is the reputation, rather than the legitimacy, that is of critical importance when companies un- dertake direct and immediate actions during an ongoing incident.

In comparison to legitimacy theory, the use of the reputation risk management perspective provides a more detailed analysis of BP's response to the Gulf of Mexico crisis through its social and environmental disclosures. The emphasis on reputation as a stra- tegic resource led to exploration of disclosures related to financial performance, management, employees and quality of goods and services. Moreover, a more “fine grained analysis” of the image restoration strategies was provided which is consistent with the prior work of Milne and Patten (2002), O'Donovan (2002) and Michelon (2012) that suggests that companies disclose positive information with the hope of restoring their image. This study provided evidence of the various instances of positive disclosures used by BP to manage its reputational risk during the Gulf of Mexico crisis.

Reputation is conceptualised by the five key elements discussed earlier. The current study's findings are consistent with Melo and Garrido-Morgado (2012) who highlight that social and environ- mental disclosures are aimed at managing corporate reputation. Bebbington et al. (2008) found that the quality of management sits at the core of reputation. An in-depth study of BP's social and environmental information disclosed during the Gulf of Mexico disaster shows evidence that projects superior quality of manage- ment. The social and environmental information disclosed throughout the period of the incident emphasised management's skills and capacity in handling the negative incident. Instead of focusing on the implications of what turned out to be one of the world's largest oil spills, the social and environmental disclosures focused on the ability of BP's management in dealing with the situation.

To report superior quality of management, BP's emphasis was on disclosing information about how it will sustain its financial per- formance during the crisis situation and provide information about its employee management. However, both the ability to sustain financial performance and the ability to manage employees portray a good quality of management. Consistent with the previous find- ings, it can be interpreted that projecting a good quality of man- agement “sits at the centre” (Bebbington et al., 2008, p.349) of managing reputation. Therefore, during a crisis situation, if there is an escalation in the amount of disclosures indicating good quality of management, it is quite likely that such information is deliber- ately included to manage reputation.

BP's social and environmental disclosures could be related to multiple strategies discussed in Benoit's (1997) image restoration strategies. These findings are in line with Harlow et al. (2011) who state that BP's strategies centred on describing what they were doing to correct the problem and compensate the victims, but did not include strategies such as admitting their own blame.

The most dominantly used strategy was observed to be the bolstering strategy. By focusing on information about its superior ability to tackle the oil spill and its strong financial and asset base available to depend on, BP stressed its good traits during the crisis situation. This can be explained as BP's dire need to disclose posi- tive information to counter the negative impact to its reputation as a direct result of the incident. This is consistent with the literature discussed in this paper that during a crisis, a company seeks to project a positive image to mitigate the impact of the negative events, and subsequently manage its reputation.

The second most dominantly used strategy was observed to be the corrective actions strategy. From BP's social and environmental disclosures it was evident that BP consistently disclosed its plans to stop the oil spill. Given the complicated nature of the oil spill, BP was justified in providing accounts of its strategies and plans to

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e12971296

tackle the oil spill. However, consistent repetition of such infor- mation can be interpreted as focusing on intangible corrective ac- tion plans rather than providing information on actual tangible impacts of the disaster. It can be deduced that during a disaster, companies attempt to manage their reputation by diverting the attention of stakeholders from actual negative effects of the event and focusing more on the positive plans and strategies to solve the problem.

The most unanticipated finding was the observation of tran- scendence and minimisation disclosures by BP during the incident. BP highlighted that the explosion and the consequent oil spill has provided it with many important lessons for the future. It also disclosed its aims to change its safety standards as a result of the lessons learnt from the disaster but downplayed the severity of the oil spill. BP pushed the boundaries of responsible behaviour when it discussed the prosperity of the Gulf region's hotel industry, directly as a result of the oil spill. Disclosing such transcendent information dilutes the purpose of providing social and environmental disclo- sures altogether.

Lastly, an expectation from the social and environmental dis- closures of a company responsible for an environmental disaster was to find evidence of communicating mortification. But BP did not offer a public apology for its actions during the course of the event. The only apology BP offered was at an investor's meeting, catered exclusively for its shareholders and investors. Although BP accepted its responsibility by taking appropriate corrective actions and bearing the cost of restoration, it chose not to publicly apolo- gize for its actions. In line with Lauzen (2015), image restoration strategies are usually not genuine if there is no mortification.

7. Conclusions

The aim of this study was to explore how social and environ- mental disclosures can be used to manage reputation during a reputation damaging event. The findings of this study have added to the existing literature that has found that the RRM perspective has the ability to provide detailed insights into the approaches for social and environmental reporting by facilitating an in-depth un- derstanding of such disclosures. If one of the primary motives for disclosing social and environmental information is the need to manage reputation (KPMG, 2013; Unerman, 2008) especially dur- ing an environmental incident, then the RRM perspective (Bebbington et al., 2008) contributes significantly in providing a useful framework to explore this phenomenon.

The analytical generalisations of this study suggest that when faced with a crisis situation, companies will choose to highlight positive information and shift the attention away from negative issues (Milne and Patten, 2002; O'Donovan, 2002; Michelon, 2012). As such, companies are motivated to use their social and environ- mental disclosures as a tool to manage the risk to their reputation arising from such incidents. Such a finding is consistent with the conclusions of other similar studies (Botes and Samkin, 2013; Cho, 2009; Deegan and Rankin, 1996; Patten, 1992; Michelon, 2012; Summerhays and de Villiers, 2012), which concluded that com- panies use social and environmental disclosures to manage their legitimacy during a crisis situation. The current study extends this literature by focusing on the immediate, direct actions during the crisis, thereby focusing on the various elements of reputation and providing a detailed analysis of image restoration strategies.

This study contributes to the academic literature by extending the applicability of the RRM perspective to explore social and environmental disclosures during a reputation damaging event. This perspective which focuses on corporate communication stra- tegies (see for example, Seeger and Ulmer, 2003; Waller and Conaway, 2011; Harlow et al., 2011; Lauzen, 2015) extends social

and environmental accounting literature, especially those that focus on disclosures. The RRM perspective provides an in-depth understanding of social and environmental disclosures by identi- fying the different strategies companies use to manage risks to their reputation. Comprehensive findings such as the focus on financial performance and quality of management and employees, the bolstering, corrective and transcendent approach to social and environmental disclosures, and the unwillingness to take re- sponsibility for actions resulting in a lack of an apology, may not have been evident through other established theoretical ap- proaches such as legitimacy theory. The focus on the ongoing incident through the website medium differentiates the current study from previous studies that have used the RRM perspective and emphasises the bankable asset, reputation, rather than the social condition, legitimacy. Moreover, this study provides a more extensive framework for exploring reputation risk management than in the corporate communication studies discussed in this paper by complementing Benoit's image restoration strategies with the five elements of reputation.

This research also adds to the limited literature on social and environmental disclosures during an incident and suggests that during such events, the corporate website is a critical source for social and environmental disclosure. The study has explored the timely disclosure medium, the Web and supports the richness of such a form of communication (Lodhia, 2004, 2010, 2012) for such disclosure, especially during ongoing events such as a social, environmental or ethical incident.

In relation to practice, the findings of this study serve as a guide for understanding organisational approaches to disclosing social and environmental information in the wake of an environmental crisis. Other companies operating in environmentally sensitive in- dustries can use the insights from this study as a demonstration of how social and environmental information disclosed during a crisis is interpreted. This will provide them with useful insights for formulating a strategy for presenting social and environmental disclosures when faced with a similar situation. On the other hand, these insights are useful to stakeholders to become aware of corporate attempts to manage reputation during a crisis.

The use of the Wayback machine in this study provides further avenues for research that seeks timely information as opposed to periodic disclosures. This tool has the potential to provide a wealth of information to researchers and can encourage studies of direct immediate responses to significant events as observed in the cur- rent study. Thus, social and environmental accounting research can be enriched through the use of this mechanism, enabling a focus on alternative media such as corporate websites.

Future research could extend the existing study to examine stakeholders' reaction to reputation risk management disclosures made by a company during a crisis. It would be interesting to un- derstand the success and failure of implementing reputation risk management strategies during a reputation damaging situation. As informed in the literature by Puncheva (2008), this will assist in understanding how management's attempts to manage reputation can affect stakeholder relationships.

References

Adams, C.A., 2004. The social, ethical and environmental reporting-performance portrayal gap. Account. Auditing Account. J. 17, 731e757.

BBC, 2010. Gulf of Mexico Oil Leak 'worst US Environment Disaster'. BBC. USA: BBC viewed 30 May 2010. http://www.bbc.co.uk/news/10194335.

Bebbington, J., Larrinaga, C., Moneva, J.M., 2008. Corporate social reporting and reputation risk management. Account. Auditing Account. J. 21, 337e361.

Benoit, W.L., 1997. Image repair discourse and crisis communication. Public Relat. Rev. 23, 177e186.

Bitektine, A., 2011. Toward a theory of social judgements: the case of legitimacy, reputation, and status. Acad. Manag. Rev. 36 (1), 151e179.

M.P. Arora, S. Lodhia / Journal of Cleaner Production 140 (2017) 1287e1297 1297

Botes, V., Samkin, G., 2013. BP's use of posture to respond to the deepwater horizon crisis. J. Econ. Financial Sci. 6, 359e382.

Bowen, G., 2009. Document analysis as a qualitative research method. Qual. Res. J. 9 (2), 27e40.

BP, 2009. Social and Enviornmental Review viewed 24 August 2013. http://www.bp. com/content/dam/bp/pdf/social and enviornmental/group-reports/bp_social and enviornmental _review_2009.pdf.

BP, 2010a. Reports from the Gulf viewed 6 September 2013. http://web.archive.org/ web/20100525235650/http://www.bp.com/sectiongenericarticle.do? categoryId¼9033611&contentId¼7061850.

BP, 2010b. Investor Briefing viewed 6 September 2013. http://web.archive.org/web/ 20100612140338/http://www.bp.com/liveassets/bp_internet/globalbp/ STAGING/global_assets/downloads/I/IC_BP_Investor_Briefing_webcast_4_June. pdf.

BP, 2010c. BP Press Release viewed 6 September 2013. http://web.archive.org/web/ 20100901023043/http://www.bp.com/PressReleaseDateSorterAction.do? categoryId¼2012968&contentId¼2006635&year¼2010&month¼0&taxono myId¼2021012.

Branco, M.C., Eug�enio, T., Ribeiro, J., 2008. Environmental disclosure in response to public perception of environmental threats: the case of co-incineration in Portugal. J. Commun. Manag. 12, 136e151.

Buhr, N., 1998. Environmental performance, legislation and annual report disclo- sure: the case of acid rain and Falconbridge. Accounting. Auditing Account. J. 11, 163e190.

Cherry, M., Sneirson, J., 2011. Beyond profit: rethinking corporate social re- sponsibility and greenwashing after the BP oil disaster. Tulane Law Rev. 85, 983e1038.

Cho, C.H., 2009. Legitimation strategies used in response to environmental disaster: a French case study of Total SA's Erika and AZF incidents. Eur. Account. Rev. 18, 33e62.

Cho, C.H., Laine, M., Roberts, R.W., Rodrigue, M., 2015. Organized hypocrisy, orga- nizational façades, and sustainability reporting. Account. Organ. Soc. 40, 78e94.

Cohen, M.A., Gottlieb, M., Linn, J., Richardson, N., 2011. Deepwater drilling: law, policy, and economics of firm organization and safety. Vanderbilt Law Rev. 64, 1853e1916.

Deegan, C., Rankin, M., 1996. Do Australian companies report environmental news objectively?: an analysis of environmental disclosures by firms prosecuted successfully by the Environmental Protection Authority. Account. Auditing Ac- count. J. 9, 50e67.

Deegan, C., Rankin, M., Voght, P., 2000. Firms disclosure reactions to major social incidents: Australian evidence. Account. Forum 2, 101e130.

Deegan, C., Rankin, M., Tobin, J., 2002. An examination of the corporate social and environmental disclosures of BHP from 1983-1997: a test of legitimacy theory. Account. Auditing Account. J. 15, 312e343.

Deephouse, D.L., Carter, S.M., 2005. An examination of differences between orga- nizational legitimacy and organizational reputation. J. Manag. Stud. 42 (2), 329e360.

Deephouse, D.L., Suchman, M., 2008. Legitimacy in organizational institutionalism. In: Greenwood, R. (Ed.), The Sage Handbook of Organizational Institutionalism. Sage Publications Ltd, London, pp. 49e77.

DHSG, 2011. Final Report on the Investigation of the Macondo Well Blowout viewed 6 August 2013. http://ccrm.berkeley.edu/pdfs_papers/bea_pdfs/ dhsgfinalreport-march2011-tag.pdf.

Douglas, A., Doris, J., Johnson, B., 2004. Corporate social reporting in Irish financial institutions. TQM Mag. 16, 387e395.

Dowling, J., Pfeffer, J., 1975. Organizational legitimacy: social values and organiza- tional behavior. Pac. Sociol. Rev. 18, 122e136.

Flyvbjerg, B., 2006. Five misunderstandings about case-study research. Qual. Inq. 12, 219e245.

Fombrun, C., 1996. Reputation: realizing Value from the Corporate Image. Harvard Business School Press, Boston.

Fombrun, C., Van Riel, C., 1997. The reputational landscape. Corp. Reput. Rev. 1 (1e2), 5e13.

Gerring, J., 2007. Case Study Research. Cambridge University Press, New York. Global Risk Management Survey, 2015. AON viewed 14 June 2016. http://www.aon.

com/forms/2015/2015-global-risk-management-survey.jsp. Gray, R., Bebbington, J., 2000. Environmental accounting, managerialism and social

and enviornmental: is the planet safe in the hands of business and accounting? Adv. Environ. Account. Manag. 1, 1e44.

Green, J., Willis, K., Hughes, E., Small, R., Welch, N., Gibbs, L., Daly, J., 2007. Gener- ating best evidence from qualitative research: the role of data analysis. Aust. N. Z. J. Public Health 31 (6), 545e550.

Harlow, W., Brantley, B., Harlow, R., 2011. BP initial image repair strategies after the Deepwater Horizon spill. Public Relat. Rev. 37, 80e83.

Hoffman, A.J., Jennings, P.D., 2011. The BP oil spill as a cultural anomaly?

Institutional context, conflict, and change. J. Manag. Inq. 20, 100e112. Hogan, J., Lodhia, S., 2011. Social and enviornmental disclosures and reputation risk

management: an Australian case study. Int. J. Account. Inf. Manag. 19, 267e287. Hopwood, A.G., 2009. Accounting and the environment. Account. Organ. Soc. 34,

433e439. Islam, M.A., Deegan, C., 2008. Motivations for an organisation within a developing

country to report social responsibility information: evidence from Bangladesh. Account. Auditing Account. J. 21, 850e874.

King, R., 2013. Deepwater Horizon Oil Spill Disaster: Risk, Recovery, and Insurance Implications. Congressional Research Service2010 viewed 5 May 2013. http:// www.fas.org/sgp/crs/misc/R42942.pdf.

KPMG, 2013. International Survey of Corporate Responsibility Reporting. http:// www.kpmg.com/Global/en/IssuesAndInsights/ArticlesPublications/corporate- responsibility/Pages/2011-survey.aspx.

Kuruppu, S., Milne, M., 2014. Managing Reputation and Maintaining Legitimacy: Understanding a Company's Responses to Social and Enviornmental Issues (Working Paper).

Lauzen, M., 2015. Image repair: a case study of Thierry Fr�emaux and the Cannes Film festival. Public Relat. Rev. 42, 170e175.

Lindblom, C., 1994. The implications of organizational legitimacy for corporate so- cial performance and disclosure. In: Critical Perspectives on Accounting Con- ference, New York, vol. 120.

Lodhia, S., 2004. Corporate environmental reporting media: a case for the World Wide Web. Electron. Green J. 1, 1e19.

Lodhia, S., 2010. Research methods for analysing World Wide Web social and enviornmental communication. Soc. Environ. Account. J. 30, 26e36.

Lodhia, S., 2012. Web Based Social and Environmental Communication in the Australian minerals industry: An application of media richness framework. J. Clean. Prod. 25, 73e85.

Lodhia, S., 2013. Social and Enviornmental Accounting and Reporting: an Overview, Contemporary Developments and Research Possibilities. Edward Elgar.

Lodhia, S., 2014. Factors influencing the use of the world wide web for sustainability communication: an Australian mining perspective. J. Clean. Prod. 84, 142e154.

McNicholas, P., and Windsor, C., 2011. The BP Gulf Oil Spill: Failed Regulatory and Corporate Governance Systems Analysed through a Regulatory Capitalist Lens. Paper Presented at the 2011 Accounting and Finance Association of Australia and New Zealand (AFAANZ) Conference, (Darwin, Australia).

Melo, T., Garrido-Morgado, A., 2012. Corporate reputation: a combination of social responsibility and industry. Corp. Soc. Responsib. Environ. Manag. 19, 11e31.

Michelon, G., 2012. Impression management and legitimacy strategies: the BP case. Financ. Report. 4, 340e363.

Milne, M.J., Patten, D.M., 2002. Securing organizational legitimacy: an experimental decision case examining the impact of environmental disclosures. Account. Auditing Account. J. 15, 372e405.

O'Donovan, G., 2002. Environmental disclosures in the annual report: extending the applicability and predictive power of legitimacy theory. Account. Auditing Ac- count. J. 15, 344e371.

Patten, D.M., 1992. Intra-industry environmental disclosures in response to the Alaskan oil spill: a note on legitimacy theory. Account. Organ. Soc. 17, 471e475.

Pellegrino, C., Lodhia, S., 2012. Climate change accounting and the Australian mining industry: exploring the links between corporate disclosure and the generation of legitimacy. J. Clean. Prod. 36, 68e82.

Puncheva, P., 2008. The role of corporate reputation in the stakeholder decision- making process. Bus. Soc. 47, 272e290.

Ramseur, J., Hagerty, C., 2013. Deepwater Horizon Oil Spill: Recent Activities and Ongoing Developments. Congressional Research Service. Congressional Research Service; p.15. http://www.fas.org/sgp/crs/misc/R42942.pdf. Viewed 7 August 2013.

Seeger, M., Ulmer, R., 2003. Explaining Enron: communication and responsible leadership. Manag. Commun. Q. 17 (1), 58e84.

Steffy, L.C., 2011. Drowning in Oil: BP and the Reckless Pursuit of Profit. McGraw- Hill, New York, US.

Summerhays, K., de Villiers, C., 2012. Oil company annual report disclosure re- sponses to the 2010 Gulf of Mexico oil spill. J. Asia-Pacific Centre Environ. Ac- count. 18, 103e130.

Tucker, L., Melewar, T., 2005. Corporate reputation and crisis management: the threat and manageability of anti-corporatism. Corp. Reput. Rev. 7, 377e387.

Unerman, J., 2008. Strategic reputation risk management and corporate social re- sponsibility reporting. Account. Auditing Account. J. 21, 362e364.

Waller, R., Conaway, R., 2011. Framing and counterframing the issue of corporate social responsibility. J. Bus. Commun. 48 (1), 83e106.

Wearden, G., 2010. David Cameron Admits Sympathy with Obama before Discussing BP Crisis. The Guardian. UK2010. http://www.theguardian.com/business/2010/ jun/10/bp-shares-plunge-gulf-mexico. viewed 10 June 2010.

Yin, R.K., 2013. Case Study Research: Design and Methods. Sage, USA.