The Market Response to
Risk Disclosure
Abstract:
In this research I’ve examined the market response to the readability of risk disclosure, measured by share performance and corporate reputation, and the moderating effect of financial performance on the relationship between the readability of risk disclosure and corporate reputation. No significant evidence for these relationships was found. An important contribution was made, because this was the first study to examine the market response to the readability of risk disclosure.
Keywords
Market Response, Risk Reporting, Risk Disclosure, Readability, Share Performance, Corporate Reputation, Agency Theory, Stakeholder Theory, Impression Management.
Master Thesis Accountancy Name: R. S. Tuinsma Student number: 1872907
University: University of Groningen Faculty: Economics and Business Degree program: Master Accountancy First supervisor: J. E. Emanuels Second supervisor: W. G. de Munnik Date: 26-06-2013
2
Table of Contents
Introduction ... 3
Theoretical Framework ... 4
Theoretical Background ... 4
Risk and Risk Appetite ... 4
Risk Management ... 5 Risk Reporting ... 6 Readability ... 6 Share Performance ... 7 Corporate Reputation ... 9 Conceptual Model ... 11 Research Design... 11 Data Sample ... 11 Dependent variables ... 12 Independent variables ... 13 Control variables ... 13 Analysis ... 14 Results ... 14 Conclusion ... 18 Discussion ... 19 Further Limitations ... 21 Future Research ... 21 References ... 23
3 Introduction
The Institute of Chartered Accountants in England and Wales in 1998 and respondents to Solomon, Solomon, Norton, and Joseph (2000) already called for improved risk disclosure: directors need to provide more detailed risk information. These calls are from over a decade ago, but in 2010 NIVRA published the results of their investigation of risk disclosure in the annual reports of 2009 for listed firms. They concluded there was improvement in the quality of risk disclosure since the last investigation two years ago, but there was still a long road ahead. As Bloomfield and Fischer (2010) illustrate, disclosures are important for investors to make decisions, which will affect the cost of capital. The pervasiveness and detail of risk disclosure influence the perceived risk and the trust in management (Rose, Norman and Rose, 2010). Risk disclosures are used by investors to examine the risk and to make particular investment decisions, but they are, like other sections of the annual report, difficult or very difficult to read (Linsley and Lawrence, 2007). An interesting question that arises is: How does risk disclosure influence outcomes of investors decisions?
According to the UK Corporate Governance Code the board should, at least annually, conduct a review of the effectiveness of the company’s risk management and internal control systems and should report to shareholders that they have done so. The review should cover all material controls, including financial, operational and compliance controls (UK Corporate Governance Code, 2012). The company’s risk management system is commonly referred to as Enterprise Risk Management (ERM). Extensive research has been done about ERM, for example the determinants for implementation of ERM (Beasley, Clune and Hermanson, 2005), the challenges of risk management in financial companies (Cumming and Hirtle, 2001) and the value of ERM (Hoyt and Liebenberg, 2011). Although extensive research has been done about the readability of financial communication (e.g. Adelberg, 1979; Jones, 1988; Courtis, 1998; Clatworthy and Jones, 2001) and some research has been done about the use of disclosure information by investors (Bloomfield and Fischer, 2010; Rose, Norman and Rose, 2010), only one article examines the readability of risk disclosure (Linsley and Lawrence, 2007). They examine only the readability of risk disclosure and do not examine determinants or implications of the readability of risk disclosure. The proximity of the risk disclosure to the auditor’s report may add credibility to it (Neu, Warsame and Pedwell, 1998). By examining the market response to the readability of risk disclosure, I hope to provide management with implications how risk disclosure can be used to affect share performance and corporate reputation.
To the best of my knowledge, this study will be unique and innovative for three reasons. First, this will be the first study to examine the market response to the readability of risk disclosure, where only one study has examined the readability of risk disclosure in general (Linsley and Lawrence, 2007). Second, this study will use a more recent set of years, where Linsley and Lawrence (2007) use the annual reports of 2001, which is over a decade ago. Third, this study will be the first to examine the impression management role of risk disclosure, where other researchers have examined the impression management role of various other sections of the annual report (e.g. Adelberg, 1979; Jones, 1988; Courtis, 1998; Clatworthy and Jones, 2001).
4
The main research question addressed in this study is: What is the market response to the readability of risk disclosure?
The rest of this thesis is constructed as follows. First, I offer an overview of the theoretical framework with the consistent hypotheses. Second, I will discuss the research method with the research sample and the different variables. Third, I will present the results of my research. Finally, I will present the conclusion and discussion.
Theoretical Framework
Theoretical Background
Stakeholders and the management of a company behave in an agency relationship, a contract under which one or more persons (the principals(s)) engage another person (the agent) to perform some service on their behalf which involves delegating some decision making authority to the agent (Jensen and Mecklin, 1976). This relationship has three sources of costs, including information asymmetry, which occurs when the management has more and more accurate information than the shareholders (Saam, 2007). Because of their access to internal information resources, a manager is likely to hold more risk information than outsiders, who have no access to this information. Disclosure of risk information can reduce this information asymmetry (Linsley and Shrives, 2005a). Another motivation lies in the stakeholder theory, which can be used in a descriptive, normative and instrumental way (Donaldson and Preston, 1995). In this research the instrumental way is used. The instrumental way of the stakeholder theory is used to identify the connections, or lack of connections, between stakeholder management and the achievement of traditional corporate objectives. Freeman (1984) concludes that firms have stakeholders and should proactively pay attention to them. He noted that an enlightened but instrumental outlook would likely lead to better long-term performance by firms that attempted to explicitly manage their various stakeholders. Managers provide narrative information to reduce the cost of capital and increasing share performance (Baginski, Hassel and Kimbrough, 2000). Healy and Palepu (2001) showed that voluntary disclosure lowers capital costs, legal liability costs and competitor costs.
Risk and Risk Appetite
In September 2004, the Committee of Sponsoring Organizations of the Treadway Commission (COSO) issued Enterprise Risk Management - Integrated Framework. That framework defines risk as follows (COSO, 2004, p. 16): ‘Risk is the possibility that an event will occur and adversely affect the achievement of objectives.’ Events with adverse impact prevent value creation or erode existing value. Pagach and Warr (2010, p. 3) state: ‘Although risk is generally considered to be the possibility of outcomes that deviate from what is expected, it is primarily negative outcomes that are of most concern to firms.’ The COSO Framework (2004) has established four categories of entity objectives: Strategic, Operations, Reporting and Compliance. All risks can be divided in one of these categories. Risk appetite is the amount of risk, on a broad level, an entity is willing to accept in pursuit of value (COSO, 2004). The British
5
Standards Institution’s standard for risk management defines risk appetite as ‘the amount and type of risk that an organization is prepared to seek, accept or tolerate’ (BS 31100). The risk appetite of a company needs to be aligned with corporate strategy. ERM uses the firm’s risk appetite to determine which risks should be accepted and which should be mitigated or avoided (Pagach and Warr, 2010).
Risk Management
The COSO Framework (2004, p. 4) defines ERM as: ‘A process, effected by an entity’s board of directors, management and other personnel, applied in strategy setting and across the enterprise, designed to identify potential events that may affect the entity, and manage risk to be within its risk appetite, to provide reasonable assurance regarding the achievement of entity objectives.’ ERM consists of eight interrelated components; internal environment, objective setting, event identification, risk assessment, risk response, control activities, information and communication and monitoring.
Mikes (2009, p. 23) defines four types of enterprise-wide risk management; Risk silo management, Integrated risk management, Risk-based management and Holistic risk management. In traditional risk management individual risk categories are separately managed in risk ‘silos’, with a focus on risk quantification (Mikes, 2009; Hoyt and Liebenberg, 2011). ERM enables firms to manage a wide array of risks with a holistic approach (Hoyt and Liebenberg, 2011). Mikes (2009) sees ERM as holistic risk management, it encompasses, besides the measurable risk silos, risks that cannot be quantified or aggregated. These non-quantifiable risks include the risks of strategic failure, environmental risks and reputational risks.
Meulbroek (2002, p. 4) states: ‘If there were no value added, then direct expenses and distraction of management’s attention would make risk management a negative NPV proposition for the firm.’ So, there should be value added by risk management for a positive NPV proposition. Hoyt and Liebenberg (2011) found three main reasons why ERM can add value to the firm; natural hedging of aggregated risk, the potential interdependencies between risks across activities and the improvement of risk management disclosure. By using ERM and integrated decision making, firms can exploit natural hedging. By exploiting natural hedging, firms can avoid duplication in risk management expenditure. Individual risk management may reduce the probability of losses, but there are potential interdependencies between risks across activities. ERM provides a structure which combines all risk management activities into one integrated framework allowing the recognition of these interdependencies. ERM allows firms to provide improved information about the firm’s risk profile, which might enable opaque firms to better inform outsiders of their risk profile. This should serve as a signal of their commitment to risk management (Hoyt and Liebenberg, 2011).
Several researchers have examined the benefits of ERM for a firm. Gordon, Loeb and Tseng (2009) found that ERM and firm performance is contingent upon the appropriate match between ERM and the following five factors affecting a firm; environmental uncertainty, industry competition, firm size, firm complexity and board of directors’ monitoring. Pagach and Warr (2010) found limited evidence of ERM adopters for any significant change in various key
6
firm variables. They also found limited evidence of risk reduction in the firm’s earnings if they look at firm that might be expected to benefit more from ERM. Hoyt and Liebenberg (2011) found a statistically and economically significant positive relation between firm value and the use of ERM.
Risk Reporting
As mentioned in the introduction, the Institute of Chartered Accountants in England and Wales in 1998 and respondents to Solomon et al. (2000) already called for improved risk disclosure. Hermanson (2000) found in her analysis of the demand for reporting about internal controls that financial reporting did not have enough relevant information about the risks an organization has, but that the disclosure of management reports on internal control improved financial reporting. Recent corporate scandals have highlighted the importance of and drawn attention to financial reporting quality (Clarke and Dean, 2007; Donoher, Reed and Storrud-Barns, 2007). The corporate financial reporting scandals have led entity stakeholders to demand greater oversight of key risks facing the enterprise to ensure that stakeholder value is preserved and enhanced (Walker, Shenkir and Barton, 2002).
Mikes (2009, p. 25) states: ‘The reports from the Treadway Commission (COSO, 2004) and the Turnbull Committee (ICAEW, 1999), both considered as important milestones of Anglo-Saxon corporate governance, advocate ERM as a framework for capturing risks that are material from the point of view of the achievement of the strategic objectives of the enterprise.’ In general terms, risk reporting shall allow outsiders to assess the risks of an entity's future economic performance (Schrand and Elliott, 1998). The Turnbull report (1999) ‘Internal Control: Guidance for Directors on the Combined Code’ was the first report in the UK to provide guidance on the implementation of the internal control requirements of the Combined Code on Corporate Governance. Linsley and Shrives (2005a) found that risk information in annual reports is judged to lack coherence. The importance of providing coherent and readable risk information is that it facilitates assessments of the risk position of the firm and enables stakeholders to manage their own risk profiles (Linsley and Shrives, 2005b). To assure that management has implemented and works with ERM, Standard and Poor’s has introduced enterprise risk management analysis into its global corporate credit rating process starting in the third quarter of 2008 (Standard and Poor’s, May 2008). To assure that management provides information about ERM, the latest version of the UK Corporate Governance Code (2012, p. 18), section C.2.1 states: ‘The board should, at least annually, conduct a review of the effectiveness of the company’s risk management and internal control systems and should report to shareholders that they have done so. The review should cover all material controls, including financial, operational and compliance controls.’
Readability
The corporate annual report is a formal communication document comprising quantitative information, narratives, photographs and graphs. It seeks to inform shareholders, creditors and others about a company’s business history, its present financial status and its expected direction
7
(Courtis, 1995, p. 4). The readability of these narratives can be measured by readability formulas. Courtis (1986, p. 285) defines a readability formula as ‘a quantitative method of predicting whether prose passages are likely to be readable by a target audience.’ The success of a formula in providing meaningful predictive information depends on its ability to measure content, style, format and organization in the writing that are related to reader comprehension. Readability formulas produce single summary reading ease scores for measured passages of prose and thereby indicate whether the passages are likely to be read and understood by the intended readership (Courtis, 1995, p.6).
In their empirical analysis regarding impression management, Merkl-Davies and Brennan (2007) found that Flesch reading ease formula and Fog Index are the most used readability measures. The Flesch reading ease formula was developed by Rudolph Flesch and uses word length and average sentence length to calculate reading ease (Flesch, 1948). Robert Gunning developed the Fog Index as a check for the Flesch reading ease formula (Gunning, 1969). It uses the average number of words per sentence and the percentage of words with more than three syllables in the passage to calculate readability (Courtis, 1986).
Narrative annual report sections provide ‘almost twice the amount of information as do the basic financial statements’ (Smith and Taffler, 2000, p. 624). Risk disclosure provides the reader of the annual report with detailed information about the risks a company faces and the way a company will deal with these risks. Studies generally find annual reports to be difficult to read (Lewis, Parker, Pound and Sutcliffe, 1986; Courtis, 1986; 2004; Smith and Taffler, 1992b). Smith and Taffler (1992b, p. 94) state: ‘even users of the greatest sophistication have difficulty in fully comprehending financial narratives.’ Linsley and Lawrence (2007) examined the readability of risk disclosures for the 25 largest UK-listed companies published closest to 1 January 2001 and found that the mean Flesch ratings are all below 50, indicating that the level of readability of the risk disclosures is difficult to very difficult. As mentioned in the introduction, the Institute of Chartered Accountants in England and Wales in 1998 and respondents to Solomon et al. (2000) already called for improved risk disclosure. The NIVRA (2010) concluded that the quality of risk disclosure was improved, but there was still a long road ahead. This implicates the readability might be improved or is even improving during the years. Several stakeholders have asked for improved risk disclosure and companies might be responding to this development.
Share Performance
Risk management activities could be value increasing for shareholders when agency costs, market imperfections and information asymmetries interfere with the operation of perfect capital markets (Stulz, 1996; Nocco and Stulz, 2006). As stated before, Pagach and Warr (2010) found limited evidence of ERM adopters for any significant change in various key firm variables. However, Hoyt and Liebenberg (2011) found a statistically and economically significant positive relation between firm value and the use of ERM. Meulbroek (2002) found that ERM is likely to reduce the expected costs of regulatory scrutiny and expected capital if risk management disclosure is provided and that ERM can contribute to better performance evaluation by making
8
firm disclosures more informative, or at least more accessible. If risk disclosure is easier to read, it is easier and better understood by the users of risk disclosure. Therefore, readability can be associated with better performance evaluation, thereby lowering information asymmetry, lowering the cost of capital and increasing share performance.
Annual reports are the most read corporate documents and are effective vehicles of impression management because of their proximity to the auditor’s report, which may add credibility to the disclosures (Neu et al., 1998). Risk disclosure, as part of the annual report, can be used for impression management. Impression management is ‘how individuals present themselves to be perceived favorably by others’ (Hooghiemstra, 2000, p.60). In a corporate reporting context, impression management is an attempt ‘to control and manipulate the impression conveyed to users of accounting information’ (Clatworthy and Jones, 2001). Managers not only tend to emphasize successes, but also tend to obfuscate failures (Adelberg, 1979). Reading difficulty is a proxy for obfuscation. Courtis (2004, p. 292) defines obfuscation as ‘a narrative writing technique that obscures the intended message, or confuses, distracts or perplexes readers, leaving them bewildered or muddled’. Accounting narratives, such as risk disclosure, thus have a self-serving bias.
Several researchers have examined the relationship between readability of sections of the annual report and several financial performance measures (Merkl-Davies and Brennan, 2007). Courtis (1986) found that poor quality readability is not related to poor performance or high risk. According to Baker and Kare (1992), the association between readability and profitability is inconclusive. However, Adelberg (1979) found an inverse relationship between profitability and reading difficulty of non-standard footnotes of auditor’s reports and concluded that managers engage in impression management to influence the firm’s share price. Jones (1988) found that readability declines as turnover increases, over time and when the company becomes listed. Subramanian, Insley and Blackwell (1993) conclude that annual reports of good performers are more readable. Courtis (1995) found no significant difference between readability and firm performance. Variability of readability is not explained by firm performance (Clathworthy and Jones, 2001). Although Rutherford (2003) and Smith and Taffler (1992a) conclude that poorly performing firms do not obfuscate using textual complexity, Courtis (1998; 2004) concludes that low reading ease and high readability variability are associated with bad news and Li (2008) concludes that annual reports of firms with lower earnings are harder to read. Several researchers found that sections of the annual report are better to read when performance is better and the other way around, that sections of the annual report are harder to read when performance is worse, but other researchers found inconclusive results or found no relationship.
In short, by being easier to read, risk disclosure can reduce information asymmetry and thus increase share performance. Besides, risk disclosure can be used for impression management and thus can have a self-serving bias by emphasizing successes and obfuscating failures and can thereby increase share performance. Therefore I hypothesize:
9
Corporate Reputation
Previous research evidence suggests that discretionary narrative disclosures, such as the president’s or chairman’s letter to the shareholders, while not receiving much attention, contain important and useful information (Abrahamson and Amir, 1996; Smith and Taffler, 2000). Risk disclosure is also discretionary narrative disclosure. Management controls the content of these disclosures, so these disclosures are directly influenced by the image management wants to convey (Geppert and Lawrence, 2008). Therefore, they are perfect means for impression management.
Corporate reputation is based on the perception of multiple factors of those outside the company. Barnett, Jermier and Lafferty (2006) note that a precise definition of corporate reputation remains lacking. In an attempt to provide a common framework, they use a cluster analysis method to identify three distinct clusters of corporate reputation definitions. These are labeled awareness, assessment and asset value. Reputations can be built through observing a corporation’s, and thus management’s, response to items such as environmental impact, employee concerns, stockholder inquiries and other aspects of social responsibility impacting the firm (Geppert and Lawrence, 2008). Risks a company is facing and the management’s response to these risks might be an item of influence on corporate reputation as well. This process, called enterprise risk management, is disclosed in the risk disclosure.
Following the three clusters of Barnett et al. (2006), I expect that assessment and asset value are the reputation concepts most likely influenced by risk disclosure. The cluster awareness is of little value, since risk disclosure is distributed in the company’s annual report, which implies that the target audience is already familiar with the company. The cluster asset value is a result of a company’s visible actions and responses to external events (Barnett et al., 2006). Management can influence the asset value of a company’s reputation by highlighting the achievements with a discussion of actions and their results. This type of behavior is called ‘symbolic management’ (Westphal and Zajac, 1995; 1998). Risks a company is facing and the management’s response to these risks is a process called ERM. To make ERM visible, management provides risk disclosure in the annual report. Risk disclosure is an available mean for ‘symbolic management’ and can thus be used to highlight a management’s achievements. This is similar to the way risk disclosure can be used to emphasize successes by using impression management, as discussed in the previous section. If it is management focus to highlight achievements and emphasize successes, it generally uses easy to read narratives, instead of hard to read narratives, which are used to obfuscate failures. The readability of risk disclosure can influence the asset value of corporate reputation by highlighting management’s achievements and emphasizing successes or obfuscating failures.
The third cluster, assessment, is closely related to corporate image. According to Larkin (2003) the key feature of reputation is effective management of stakeholder relationships. In managing corporate image, it is important that the public perceive that the company’s communications are reliable (Greyser, 1999). He argues that central to the link between company behavior and reputation is firm credibility. Credibility requires stakeholders to perceive that the
10
message conveyed is transparent, and implicitly truthful. If risk disclosure is easy to read it is highly transparent, implicitly truthful and could therefore be perceived more credible. If risk disclosure is hard to read it will be seen as use for obfuscation and could lose its transparency and perceived credibility. The readability of risk disclosure has the potential to affect perceived credibility and transparency of the firm and can influence a company’s reputation.
In short, corporate reputation is build around three clusters, awareness, asset value and assessment. I expect that risk disclosure has no influence on awareness, since it is published in the annual report. However, the readability of risk disclosure can influence corporate reputation through the cluster asset value by highlighting management’s achievements and emphasizing successes or obfuscating failures and through the cluster assessment by affecting the perceived credibility and transparency. Therefore I hypothesize:
H2: The readability of risk disclosure and corporate reputation are associated.
As shown in the development of the first hypothesis, several researchers have examined the relationship between readability of sections of the annual report and several financial performance measures (Merkl-Davies and Brennan, 2007). Adelberg (1979) found an inverse relationship between profitability and reading difficulty, Subramanian et al. (1993) conclude that annual reports of good performers are more readable, Courtis (1998; 2004) found that low reading ease, and high readability variability, are associated with bad news and Li (2008) concludes that annual reports of firms with lower earnings are harder to read. Although other researchers (Baker and Kare, 1992; Smith and Taffler, 1992a; Clatworthy and Jones, 2001; Rutherford, 2003) found no or inconclusive evidence of this relationship, I posit a relationship exists between financial performance and the readability of risk disclosure.
Several researchers have examined the relationship between financial performance and corporate reputation (Sabate and Puente, 2003). They found in their empirical analysis of this relationship that the question was not only whether the relationship was positive or negative, but also which direction of causality there is. The empirical analysis shows that research has been done about the existence of the relationship, regardless of the direction, the relationship between financial performance and corporate reputation, the relationship between corporate reputation and financial performance and the double hypothesis circle, were both hypotheses are combined in a single study. Preston and Sapienza (1990) focused on the existence of this relationship and found a positive relationship between corporate reputation and financial performance. Several researchers have examined the relationship between corporate reputation and financial performance and found that this relationship is positive (Srivastava, McInish, Wood and Capraro, 1997; Deephouse, 1997; Roberts and Dowling, 1997; Cordeiro and Sambharya, 1997; Black, Carness and Richardson, 1999; Vergin and Qoronfleh, 1999; Jones, Jones and Little, 2000). The positive results have been validated, both for measures of corporate reputation and financial performance and by using several different lags (Sabate and Puente, 2003). Various studies examining the relationship between financial performance and corporate reputation found
11
a positive relationship (Fombrun and Shanley, 1990; Riahi-Belkaoui and Pavlik, 1991). Research by McGuire, Schneeweiss and Branch (1990) showed evidence supporting the double hypothesis circle. Although other researchers (Sobol and Farrell, 1988; Schultz, Mouritsen and Gabrielsen 2000; Hammond and Slocum, 1996; Chung, Schneeweiss and Eneroh, 1999; Dunbar and Schwalbach, 2000) found no or inconclusive evidence of this relationship, I posit a positive relationship exists between financial performance and corporate reputation.
As stated in the previous hypothesis development, I expect readability of risk disclosure and corporate reputation are associated. Risk disclosure can be used to emphasize successes and to obfuscate failures by being more, or less, readable. Through the use of these techniques, management can use risk disclosure to enhance the relationship between financial performance and corporate reputation when financial performance is good and weaken this relationship when financial performance is worse. So, I expect the readability of risk disclosure could have a moderating effect on the relationship between financial performance and corporate reputation. Therefore I hypothesize:
H3: Financial performance and corporate reputation are positively associated.
H4: The readability of risk disclosure moderates the relationship between financial
performance and corporate reputation. Conceptual Model
Research Design
Data Sample
I will perform a quantitative analysis where the readability of risk disclosure and financial performance will be the independent variables and share performance and corporate reputation will be the dependent variables. I will use data from the FTSE-100 companies, excluding financial companies. These companies are used since they are all publicly listed in the UK, which means they are subjected to the UK Corporate Governance Code and that their financial and reporting information is publicly available. According to Jones (1988), readability declines when a company becomes publicly listed. By only taking publicly listed companies, this risk in disturbance is avoided. These companies also apply to the Standard and Poor’s credit rating, which has introduced ERM analysis into its global corporate credit rating process since the third quarter of 2008. Therefore, I will use data from the fiscal years 2008-2011. Due to the absence of data for several firm-year combinations and the use of only complete firm-year combinations for
Readability of Risk Disclosure
Financial Performance Corporate Reputation
12
all four years, my final sample consisted of 39 companies and 117 observations. An overview of these companies is presented in appendix A.
This data will be collected from CompuStat global, Datastream and the magazine Management Today. CompuStat global consists of annual report data of listed companies, including EBIT, total assets and ROA, Datastream contains information about stocks, including information on equities, indices, futures and options, and the magazine Management Today publishes the Britain’s Most Admired Companies list.
Several researchers found that sections of the annual report are hard to read (e.g. Lewis, Parker, Pound and Sutcliffe, 1986; Courtis, 1986; 2004; Smith and Taffler, 1992b) and Linsley and Lawrence (2007) found that risk disclosures are difficult or very difficult to read. To be able to distinguish between the readability of risk disclosures in different years, I will measure the change in readability by dividing the readability of this year by the readability of the previous year. Consistent with Subramanian et al. (1993), who used the change in net profit compared with the previous year, I will measure the change in share performance and financial performance compared with the previous year. For the consistency of the whole model, I will use the change in corporate reputation compared with the previous year as well. To combine the data, I will use the readability of the risk disclosure and financial performance from the annual report published for the last fiscal year, the share performance during the first 5 days after the annual report has been published and corporate reputation for the year following on the fiscal year. By using the share performance during the first 5 days, I will examine the direct, short-term market response to the readability of risk disclosure and by using corporate reputation the indirect, long-term market response to the readability of risk disclosure.
Dependent variables
Share performance Cumulative Abnormal Return (CAR) is used in several studies to determine
whether investors receive new information which adjusts their expectations of stock performance over multiple days (Cowan, 1993). Although several researchers have examined the CAR for a relative long time window (e.g. Dodd and Warner, 1983; Linn and McConnell, 1983; Wruck, 1989), Bernard and Thomas (1989, p. 13) state: ‘a disproportionate large amount of the 60-day drift occurs within 5 days of the earnings announcement’. The earnings announcement is comparable with publishing the annual report in the case of risk disclosure. Since I want to measure the direct, short-term effect and a large amount of the drift in CAR occurs within 5 days of the publishing of new information, I will use a time frame of 5 days for CAR, starting at the day the annual report is published. Following Chaney and Philipich (2002), I will measure CAR as the difference between the actual return of a share and the expected return derived from the FTSE-100 index.
Corporate Reputation The best-known league tables for corporate reputation are those published
annually in the USA business magazine Fortune (Bromley, 2002). Although Fombrun, Gardberg and Sever (2000) question the validity of the Fortune list, Fortune is still the most used measure for corporate reputation (Bromley, 2002). Fortune, which is the America’s Most Admired
13
Companies (AMAC) list, has a British equivalent, namely the British Most Admired Companies (BMAC) list. Both lists are end of the year published peer review measures for corporate reputation. Therefore, I will use the BMAC score as measure for corporate reputation.
Independent variables
Readability of risk disclosure In readability studies regarding financial accounting, the Flesch
and Fog measure are the most used measures for readability (e.g. Jones, 1988; Smith and Taffler, 1992a; 1992b; Courtis, 1998; Clatworthy and Jones, 2001; Linsley and Lawrence, 2007). Although the Cloze and Lix measure have also been used by several researchers, they haven’t been used since 1995 (Smith and Taffler, 1992b; Courtis, 1995). I will measure readability using the Flesch measure, since it is the most used measure for readability in previous studies.
Financial Performance In tradition with corporate governance research, I will measure financial
performance using the return on Assets (ROA), measured by the earnings before interest and taxes (EBIT) divided by total assets (Cannella, Park, and Lee, 2008; Carpenter, 2002).
Control variables
Corporate reputation is influenced by several factors. In the analysis regarding corporate reputation, I will control for the variables size, market capitalization, industry and year. To be consistent with the research design, for size and market capitalization I will use the change compared to the previous year. Although CAR is significantly related to newly published information (e.g. Dodd and Warner, 1983; Linn and McConnell, 1983; Wruck, 1989), to be consistent in this research I will use the same control variables in the analysis regarding share performance.
Size For certain industries, corporate reputation is more related to size (Sobol and Farrell, 1988).
Therefore, I will control for size, which will be measured by the natural logarithm of total assets.
Market Capitalization Corporations with a higher market capitalization are expected to have a
better corporate reputation (Roberts and Dowling, 2002). Fombrun and Shanley (1990) found that the market valuation of a corporation could be related with corporate reputation. Therefore I will control for market capitalization, which will be measured by dividing the market value by the book value of a corporation.
Industry Sobol and Farrell (1988) showed that the relevance of financial performance in
affecting corporate reputation varied from industry to industry. The change in reputation could vary from industry to industry as well. Therefore I will control for industry. Following Fombrun and Shanley (1990) all variables are normalized with respect to the definitions of the FamaFrench industry qualifications. Sector 5, the sector other, will be the reference sector. Sector 1 is consumer, sector 2 manufacturing, sector 3 hi-tech and sector 4 health. Due to data limitations, there are no sector 4 companies in my dataset.
14
Year Consistent with Li (2008) this study will control for the year the annual report was
published. I will use the years as dummy variables. 2009 will be the reference year. Analysis
In this research I am using a trend analysis. To carry out this research I will use several different statistical analyses. First, I standardize the variables with significant differences to reduce the effect of possibly spurious outliers. Next, I will create a correlation matrix to see if multicollinearity exists between my dependent, independent and control variables. Third, I will proceed to carry out the linear regression analysis. In the regression analysis I will use dummies for year and industry and for the other variables I will use the change compared to the previous year. For the first, second and third hypotheses, I will use five models per hypothesis. In the first model, a regression analysis will be made with the control variables size and market capitalization only. In the next four models the hypothesis will be tested. In model 2 only the independent variable will be tested, in model 3 and 4 there will be controlled for year and industry, respectively, and in model 5 all variables will be combined. For the second and third hypothesis, the same procedure will be carried out. For the fourth hypothesis, I will use five models to test the hypothesis. In the first model only the independent variable will be tested. In the second and third model, there will be controlled for year and industry, respectively, in the third model there will be controlled for all control variables and in the last model the independent variables readability and financial performance will be included.
Results
Table 1 present descriptive statistics for my variables. In the first four columns, the normal descriptive statistics are presented and in the next four columns the statistics for the change is presented. On average, risk disclosure consists of 2319 words. The mean Flesch score for risk disclosure is 23,36. Linsley and Lawrence (2007), who used risk disclosures of the 25 largest companies of the FTSE-100 published in 2001, found an average of 2859 words with a mean Flesch score of 30,09. They used every sentence that could be considered as risk disclosure, where I have only used the sentences under the header risk management. Besides, in more recent annual reports, instead of long pieces of text, short sentences are presented in schemes. Risk disclosure has changed in the last years due to the attention paid to them and the introduction of
Variable Mean Min. Max. St. Dev. Mean Min. Max. St. Dev.
Number of Words 2319 531 5735 1249 Readability (Flesch) 23,36 11,96 42,69 5,98 0,997 0,520 2,080 0,184 Readability (Fog) 18,16 14,13 21,62 1,54 1,006 0,850 1,180 0,051 Financial Performance 9,9% -5,0% 25,0% 4,8% 1,032 -2,570 4,640 0,642 Share Performance 0,28 -13,90 15,17 4,48 -0,569 -98,450 63,920 11,691 Corporate Reputation 60,69 42,00 72,00 6,21 1,015 0,792 1,210 0,127 Market Capitalization 9,38 1,83 32,30 6,22 1,111 0,370 2,260 0,352 Size 18,27 16,07 21,80 1,29 1,003 0,990 1,060 0,009
15
various regulations. This can be the reason for a decline in average words and readability. The average financial performance, measured as the return on assets, is 9,9% with a standard deviation of 4,8%. Carpenter (2002) found an average financial performance of 10,0% with a standard deviation of 10,0%. The average corporate reputation is 60,69 with a standard eviation of 6,21. Brammer, Millington and Pavelin (2009), who used the 2002 BMAC list as measure for reputation, found an average reputation of 53,2 with a standard deviation of 7,99. The results of this study and the results of previous studies are in line with each other. The average Fog score is 18,16 with a standard deviation of 1,54, the average share performance, measured as cumulative abnormal return, is 0,28 with a standard deviation of 4,48, the average market capitalization is 9,38 with a standard deviation of 6,22 and the average size, measured as the natural logarithm of total assets, is 18,27 with a standard deviation of 1,29.
Table 2 presents the correlation matrix for all variables. The two measures for readability, Flesch and Fog, are negatively correlated. This is not an issue, since only one readability measure is used in this research, namely the Flesch score, since this score is the most used readability measure. Size is significantly related to the two readability measures and to market
Variable Read. (Flesch) Read. (Fog) Financial Perf. Share Perf. Corp. Rep. Market Cap. Size 1 -,623** 1 0,000 0,030 -0,099 1 0,748 0,290 0,016 -0,018 0,127 1 0,860 0,844 0,172 0,140 -0,095 0,007 0,053 1 0,131 0,307 0,941 0,569 0,134 -0,142 0,025 0,024 0,020 1 0,150 0,126 0,786 0,796 0,833 ,249** -,215* 0,018 0,045 -0,010 -,331** 1 0,007 0,020 0,849 0,630 0,911 0,000
* Correlation is significant at the 0.05 level (2-tailed) ** Correlation is significant at the 0.01 level (2-tailed)
Table 2. Correlation Matrix
Size Market Capitalization Corporate Reputation Share Performance Financial Performance Readability (Fog) Readability (Flesch)
16
capitalization. But, since the correlation coefficient is well below the critical value, multycollinearity is not an issue.
As discussed in the analysis section, I will use several models for the testing of the hypotheses. Table 3 contains the results of the regression analyses for the first two hypotheses. Models 1 through 5 contain the results for the first hypothesis and models 6 through 10 contain the results for the second hypothesis. Table 4 contains the results of the regression analyses for the third and fourth hypothesis. Models 1 through 5 contain the results for the third hypothesis and models 6 through 10 contain the results for the fourth hypothesis. In the next section I will discuss the results of the regression analyses and their implications for the hypotheses.
H1: The readability of risk disclosure and share performance are associated.
The results show that the readability of risk disclosure has no significant relationship with share performance (t = -0,048, p = 0,962). Even with the addition of the year and industry dummies there’s no significant relationship (p > 0,946). Therefore, hypothesis 1 can be rejected.
H2: The readability of risk disclosure and corporate reputation are associated.
The results show that the readability of risk disclosure has no significant relationship with corporate reputation (t = 1,589, p = 0,115). With the addition of the year and industry dummies there’s also no significant relationship (t > 1,572, p < 0,119). Therefore, hypothesis 2 can be rejected.
Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10
-79,122 -80,897 -159,277 -49,972 -126,587 1,070 1,704 1,101 1,291 1,291 129,449 135,220 140,238 136,306 141,702 1,406 1,453 1,519 1,548 1,548 1,458 1,497 3,194 0,896 2,523 0,007 -0,007 0,005 0,000 0,000 3,295 3,406 3,591 3,424 3,618 0,036 0,037 0,039 0,040 0,040 76,690 78,718 157,964 52,306 129,592 -0,061 -0,786 -0,176 -0,352 -0,352 127,782 135,175 140,227 135,962 141,376 1,388 1,452 1,519 1,544 1,544 -0,303 -0,427 0,000 -0,063 0,108 0,108 0,110 0,110 6,331 6,335 6,378 6,390 0,068 0,069 0,070 0,070 -5,680 ** -5,467 * -0,044 -0,043 2,813 2,817 0,030 0,031 -2,957 -2,950 -0,025 -0,025 2,701 2,702 0,029 0,030 -5,750 * -5,468 -0,023 -0,023 3,393 3,369 0,037 0,037 -5,003 -4,877 -0,017 -0,017 3,480 3,454 0,038 0,038 -3,190 -3,123 0,009 0,009 3,597 3,568 0,039 0,039 R-square 0,004 0,004 0,039 0,032 0,065 0,000 0,022 ,041 ,034 ,051 Adjusted R-square -0,014 -0,023 -0,004 0,032 -0,004 -0,017 -0,004 -,002 -,019 -,019 R-square change 0,000 0,035 0,028 0,026 0,022 ,019 ,012 ,011 F-Change 0,002 2,049 1,077 0,996 2,525 1,081 ,441 ,399 F-Value 0,214 0,142 0,906 0,609 0,940 0,023 0,857 ,947 ,643 ,732
** Correlation is significant at the 0.05 level (2-tailed) *** Correlation is significant at the 0.01 level (2-tailed)
Year 2011 Industry dummy 1 Industry dummy 2 Industry dummy 3
* Correlation is significant at the 0.10 level (2-tailed)
Constant
Market Capitalization Size
Readability (Flesch) Year 2010
17
H3: Financial performance and corporate reputation are positively associated.
The results show that financial performance has no significant relationship with corporate reputation (t = 0,070, p = 0,944). Even with the addition of the year and industry dummies there’s no significant relationship (p > 0,893). Therefore, hypothesis 3 can be rejected.
H4: Financial performance moderates the association between the readability of risk
disclosure and corporate reputation.
The results in table 4 show that financial performance has no moderating effect on the relationship between the readability of risk disclosure and corporate reputation (t = 0,562, p = 0,574). Even with the addition of the year and industry dummies there’s no significant relationship (p > 0,438). In the complete model with all variables the significance of the moderating effect is 0,213. Therefore, hypotheses 4 can be rejected.
In table 3, below model 1 I see that only 4‰ of share performance is explained by the control variables (R-square = 0,004). By adding the readability of risk disclosure, there is no change in R-square and no significant F-change (p = 0,962). The readability of risk disclosure combined with the year dummies explains around 4% of share performance (R-square = 0,039), where a significant relationship for the year 2010 and share performance is shown (p = 0,046).
Variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Model 10
1,070 1,071 0,440 1,257 0,634 1,032 0,337 1,212 0,522 1,256 1,406 1,412 1,476 1,439 1,507 1,412 1,478 1,440 1,512 1,551 0,007 0,007 0,021 0,002 0,016 0,010 0,027 0,005 0,022 0,006 0,036 0,036 0,038 0,036 0,039 0,036 0,039 0,037 0,039 0,040 -0,061 -0,064 0,569 -0,234 0,390 -0,027 0,669 -0,194 0,496 -0,356 1,388 1,395 1,459 1,419 1,487 1,394 1,460 1,418 1,490 1,548 0,001 0,003 0,001 0,002 0,005 0,019 0,019 0,019 0,019 0,019 0,137 * 0,073 0,008 0,012 0,007 0,011 0,020 0,015 0,015 0,015 0,015 0,016 -0,046 -0,044 -0,049 -0,047 -0,049 0,031 0,031 0,031 0,031 0,031 -0,021 -0,020 -0,020 -0,020 -0,027 0,030 0,030 0,029 0,030 0,030 -0,026 -0,024 -0,024 -0,020 -0,018 0,037 0,037 0,037 0,037 0,037 -0,013 -0,012 -0,011 -0,008 -0,012 0,038 0,038 0,038 0,038 0,038 0,008 0,009 0,010 0,012 0,013 0,040 0,040 -0,040 0,040 0,039 R-square 0,000 0,000 0,020 0,012 0,030 0,003 ,025 ,014 ,034 ,065 Adjusted R-square -0,017 -0,026 -0,024 0,012 -0,042 -0,023 -,019 -,040 -,037 -,023 R-square change 0,000 0,019 0,011 0,010 0,003 ,022 ,011 ,009 ,031 F-Change 0,005 1,092 0,421 0,372 0,318 1,235 ,400 ,344 1,774 F-Value 0,023 0,017 0,447 0,219 0,414 0,121 ,567 ,260 ,477 ,742 Constant Market Capitalization Size Financial Performance Year 2010
Table 4. Hypothesis 3 and 4
* Correlation is significant at the 0.10 level (2-tailed)
Readability (Flesch) x Financial Performance Year 2011 Industry dummy 1 Industry dummy 2 Industry dummy 3
** Correlation is significant at the 0.05 level (2-tailed) *** Correlation is significant at the 0.01 level (2-tailed)
18
The readability of risk disclosure combined with industry dummies around 3% of share performance (R-square = 0,032), where a significant relationship for industry 1, consumer, and share performance is shown (p = 0,093). By looking at model 5, I see that 6,5% of share performance is explained by all variables (R-square = 0,065), where a significant relationship for the year 2010 and share performance is shown (p = 0,055). Given this results, I can conclude that the readability of risk disclosure has little to no influence on share performance.
Below model 6, I see that market capitalization and size have no explanatory value for corporate reputation (R-square = 0,000). By adding the readability of risk disclosure, there is a relative large change in R-square (R-square change = 0,022) and an almost significant F-change (p = 0,115). The readability of risk disclosure is almost significant in this model (p = 0,115). The readability of risk disclosure combined with the year dummies explains around 4% of corporate reputation (R-square = 0,041), where the readability of risk disclosure is almost significant (p = 0,119). The readability of risk disclosure combined with industry dummies explains around 3% of corporate reputation (R-square = 0,034), where the readability of risk disclosure is almost significant (p = 0,115). By looking at model 10, I see that 5,1% of corporate reputation is explained by all variables (R-square = 0,051), where the readability of risk disclosure is almost significant (p = 0,118). Given this results, I can conclude that the readability of risk disclosure has an almost significant effect on corporate reputation.
In table 4, model 1 is the same as model 6 in table 3. By adding financial performance, there is no change in R-square and no significant F-change (p = 0,944). Financial performance combined with the year dummies explains around 2% of corporate reputation (R-square = 0,020) and combined with industry dummies around 1% of corporate reputation (R-square = 0,012). By looking at model 5, I see that 3% of corporate reputation is explained by all variables (R-square = 0,030). Given this results, I can conclude that financial performance has little to no influence on corporate reputation.
Below model 6, I see only a very small change in R-square when the moderator is added (R-square change = 0,003). The moderator combined with the year dummies explains around 2,5% of corporate reputation (R-square = 0,025) and combined with industry dummies around 1% of corporate reputation (R-square = 0,014). By looking at model 9, I see that 3,4% of corporate reputation is explained by the moderator and all control variables (R-square = 0,034). In model 10, where all the variables are combined, I see that only 6,5% of corporate reputation is explained by all the variables, where the readability of risk disclosure has a significant effect on corporate reputation (p = 0,064). Given this results, I can conclude that there is no moderating effect of financial performance on the relationship between the readability of risk disclosure and corporate reputation.
Conclusion
In this research I’ve examined the market response to the readability of risk disclosure, measured by share performance and corporate reputation, and the moderating effect of financial performance on the relationship between the readability of risk disclosure and corporate
19
reputation. No significant evidence for these relationships was found. This means the readability of risk disclosure has no direct effect on share performance and corporate reputation. However, there are a few recommendations which can be made. As mentioned in the introduction, directors can positively influence the perceived trust in management when risk disclosure is more detailed (Rose et al, 2010). So, risk disclosure needs to be more detailed to increase the perceived trust in management. Although the readability of risk disclosure has no effect on share performance or corporate reputation, I would personally recommend companies to improve the readability of risk disclosure. This might improve the perceived quality of risk disclosure, since it could otherwise be seen as use for obfuscating bad news. Thereby the perceived trust in management can be increased. Also, I recommend a table lay-out style for all the identified key risk factors with an overview of the description of the risk, its impact and the mitigating activities for this risk. Using an overview instead of large pieces of text provides the user with a better and more easy to understand risk disclosure. In short, by providing the risk disclosure in a table style lay-out with detailed and easy to read pieces of text, the perceived quality of risk disclosure and the perceived trust in management might increase.
Discussion
In the theoretical framework I suggested that risk disclosure can reduce information asymmetry and can be used for impression management, thereby increasing share performance. As discussed in the results, no significant evidence was found for this relationship. Although Stulz (1996), Nocco and Stulz (2006), Pagach and Warr (2010) and Hoyt and Liebenberg (2011) suggested or found that risk management activities could be value increasing for shareholders, these studies focused on the use of ERM and not on the readability of the risk disclosure. Meulbroek (2002) found that ERM can contribute to better performance evaluation by making firm disclosures more informative, or at least more accessible. However, by being easier to read, the risk disclosure is not automatically more informative or more accessible. As Adelberg (1979) stated, managers not only tend to emphasize successes, but also tend to obfuscate failures. He found that profitability was inversely related with the reading difficulty of non-standard footnotes of auditor’s reports and concluded that managers engage in impression management to influence the firm’s share price. However, in this study another section of the annual report is used and share performance is used as variable instead of share price. Several researchers have examined the relationship between readability of sections of the annual report and several financial performance measures (Merkl-Davies and Brennan, 2007). Several researchers found no significant evidence regarding this relationship, which could be an explanation why no significant evidence was found in this study (e.g. Courtis, 1986; Baker and Kare, 1992; Courtis, 1995; Clathworthy and Jones, 2001). Another reason could be that there are different findings regarding the obfuscating use of sections of the annual report (e.g. Rutherford, 2003; Smith and Taffler, 1992a, Courtis, 1998; 2004; Li, 2008). Besides the different findings of several researchers, these researchers have all used different sections of the annual report, like footnotes or the chairman’s report, and different financial performance measures, like the profitability or the market-to-book ratio. In short, risk disclosure might not be the best section of the annual
20
report to use in research and share performance might not be the best measure for financial performance.
The second relationship this research has examined is the relationship between the readability of risk disclosure and corporate reputation. The argument for this relationship was built on the fact that risk disclosure can be used for impression management and could thereby enhance corporate image and thus corporate reputation. No significant relationship was found regarding this relationship. An explanation that no significant evidence could be that in this research the readability of risk disclosure is used as measure for impression management, where reading difficulty is a proxy for obfuscation (Courtis, 2004). There might be better measures for impression management, like persuasive language or bias of themes (Merkl-Davies and Brennan, 2007). Besides, in managing corporate image, it is important that the public perceive the company’s communications as reliable, by being transparent, and implicitly truthful (Greyser, 1999). Reading difficulty might not be the best proxy to measure the transparency of a company’s communications. However further research regarding this relationship can be done using other measures for impression management or a very specific attribution of corporate reputation, for example the quality of management, since the perceived quality of management can be influenced by the management itself using impression management.
As discussed in the results, no significant evidence was found for the relationship between financial performance and corporate reputation. However, several researchers found evidence supporting the relationship between financial performance and corporate reputation (e.g. Preston and Sapienza, 2003; Deephouse, 1997; Roberts and Dowling, 1997; Fombrun and Shanley, 1990). Several researchers have used the Fortune score for corporate reputation (e.g. Preston and Sapienza, 2003; Fombrun and Shanley, 1990; Belkaoui and Pavlik, 1991; Roberts and Dowling, 1997) and several researchers have used ROA as measure for financial performance (e.g. Deephouse, 1997; Vergin and Qoronfleh, 1999), but only one research examined the relationship between ROA and the Fortune score (McGuire, Schneeweiss and Branch, 1990). They measured the average ROA of the period 1977-1982 and the reputation score for 1983 and found significant evidence for this relationship. They have taken a longer period to measure ROA, where in this research the change in ROA in one year was taken as measure for financial performance, which could be a too brief period to measure financial performance. Another explanation could be that ROA is not a good measure for financial performance, since other authors have found significant evidence regarding this relationship by using other financial performance measures, even without taking a longer period (e.g. Fombrun and Shanley, 1990; Srivastava, McInish, Wood and Capraro, 1997; Black, Carness and Richardson, 1999). This researchers have used financial performance measures like Tobin’s Q, Beta and Market value.
The moderating effect of financial performance on the relationship between the readability of risk disclosure and corporate reputation was the last relationship addressed by this study. This relationship was never examined and was built on logic reasoning following the expected relationship between readability of risk disclosure and corporate reputation, which was
21
already a combination of studies, and the already proven relationship between financial performance and corporate reputation. No significant evidence was found regarding this moderating effect of financial performance. Several explanations for this result have already been presented above. In short, reading difficulty might not be a good measure for impression management, risk disclosure might not be the best section of the annual report to use in research and ROA might not be a good financial performance measure.
Further Limitations
In the previous sections the specific limitations regarding the individual relationships were discussed. The research itself has some limitations as well. First, I’ve examined all the relationships using a trend analysis. Based on the literature used in this study, this is very uncommon. By using raw data, other results may be found. Second, the data I used was limited. My final sample consisted of just 117 observations, due to the use of multiple variables which lacked data for several company-year combinations. This has led to a large reduction in usable observations. Limiting the number of variables would have led to a larger data sample. By using a larger sample, other results may be found. Third, for several companies, the risk disclosure itself changed during the years. They have become longer or have become different in forms. Some companies first used a story in their risk management section and later used a table format. Fourth, reading difficulty might not be the best proxy for impression management. Other focuses of manipulation, like persuasive language or bias of themes, might be better measures for impression management (Merkl-Davies and Brennan, 2007). Besides, reading difficulty does not account for pictures or graphs, which also can be used for impression management (Courtis, 1995). Fifth, the two dependent variables used in this study, share performance and corporate reputation, are variables which are influenced by several factors. Share performance measured as CAR is dependent on the investor expectations and market movements. The expectations of investors vary in large degree and are influenced by all sort of announcements, for example earnings announcements (Bernard and Thomas, 2002). The publication of the annual report could be directly influencing the CAR, but in this research this direct effect was not examined. Corporate reputation is based on the perception of various stakeholders. Although there is controlled for size, market capitalization, year and industry, there are other factors which can influence corporate reputation, like media publications, downsizing and layoffs (Einwiller, Carrol and Korn, 2010; Zyglidopoulos, 2005; Flanagan and O'Shaughnessy, 2005).
Future Research
The first relationship I examined was the relationship between the readability of risk disclosure and share performance. No evidence for this relationship was found, but an interesting question is if the publication of the annual report itself has an impact on share performance. For the relationship between the readability of risk disclosure and corporate reputation no significant relationship was found, but future studies can examine this relationship more thoroughly. Regarding the future research of this relations I would recommend a larger data sample for better results. Further recommendations for future research are the use of different sections of the
22
annual report, the use of other measures for impression management, the use of other financial performance measures and the use of better or more specified dependent variables.
23 References
Abrahamson, E. and Amir, E. 1996. The information content of the president’s letter to shareholders. Journal of Business Finance and Accounting, 23 (8): 1157-1182.
Adelberg, A. H. 1979. Narrative disclosures contained in financial reports: Means of communication or manipulation. Accounting and Business Research, 10: 179-189.
Baginski, S. P., Hassell J. M. and Kimbrough, M. D. 2004. Why do managers explain their earnings forecasts? Journal of Accounting Research, 22 (1): 1-29.
Baker, H. E. and Kare, D. D. 1992. Relationship between annual report readability and corporate financial performance. Management Research News, 15 (1): 1-4.
Barnett, M., Jermier, J. and Lafferty, B. 2006. Corporate reputation: The definitional landscape. Corporate Reputation Review, 9 (1): 26-38.
Beasley, M. S., Clune, R. and Hermanson, D. R. 2005. Enterprise risk management: An empirical analysis of factors associated with the extent of implementation. Journal of Accounting and Public Policy, 24: 521-531.
Bernard, V. L. and Thomas, J. K. 2002. Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium? Journal of Accounting Research, 27: 1-48.
Black, E., Carnes, T. and Richardson, V. 2000. The market valuation of corporate reputation. Corporate Reputation Review, 3 (1): 21-31.
Bloomfield, R. and Fischer, P. E. 2010. Disagreement and the Cost of Capital. Journal of Accounting Research, 49 (1): 41-68.
Brammer, S., Millington, A. and Pavelin, S. 2009. Corporate Reputation and Women on the Board. British Journal of Management, 20 (1): 17-29.
Bromley, D. 2002. Comparing Corporate Reputations: League Tables, Quotients, Benchmarks or Case Studies? Corporate Reputation Review, 5 (1): 35-50.
BS 31100. 2011. Risk Management. Code of Practice, British Standard Institution.
Cannela, A. A., Park, J. and Lee, H. 2008. Top Management Team Functional Background Diversity and Firm Performance: Examining the Roles of Team Member Colocation and Environmental Uncertainty. Academy of Management Journal, 51 (4): 768-784.
Carpenter, M. A. 2002. The implications of strategy and social context for the relationship between top management team heterogeneity and firm performance. Strategic Management Journal, 23: 275-284. Chaney K. C. and Philipich K. L. 2002. Shredded Reputation: The Cost of Audit Failure. Journal of Accounting Research, 40: 1221-1245
Chung, S. Y., Schneeweis, T. and Eneroth, K. 1999. Corporate reputation and investment performance: The UK and US expertise, SSRN Electronic Paper Collection.
24
Clarke, F. and Dean, G. 2007. Indecent disclosure: Gilding the corporate lily, Cambridge University Press, Melbourne.
Clatworthy, M. and Jones, M. J. 2001. The effect of thematic structure on the variability of annual report readability. Accounting, Auditing and Accountability Journal, 14 (3): 311-326.
Cordeiro, J. J. and Sambharya, R.B. 1997. Do corporate reputations influence security analyst earnings forecast? An empirical study. Corporate Reputation Review, 1 (1): 94-97.
COSO. 2004. Enterprise Risk Management – Integrated Framework. Executive Summary & Framework. Courtis, J. K. 1986. An Investigation into Annual Report Readability and Corporate Risk-Return Relationships. Accounting and Business Research, 16 (64): 285-294.
Courtis, J. K. 1995. Readability of annual reports: Western versus Asian evidence. Accounting, Auditing and Accountability Journal, 8 (2): 4-17.
Courtis, J. K. 1998. Annual report readability variability: tests of the obfuscation hypothesis. Accounting, Auditing and Accountability Journal, 11 (4): 459-471.
Courtis, J. K. 2004. Corporate report obfuscation: Artefact or phenomenon? British Accounting Review, 36 (3): 291-312.
Cowan A. R. 1993. Tests for Cumulative Abnormal Returns over Long Periods: Simulation Evidence. International Review of Financial Analysis, 2 (1): 51-68.
Cumming, C. M. and Hirtle, B. J. 2001, The Challenges of Risk Management in Diversified Financial Companies, FRBNY Economic Policy Review, 7: 1-17.
Deephouse, D. L. 1997. The effect of financial and media reputations on performance. Corporate Reputation Review, 1 (1): 68-71.
Donaldson, T. and Preston, L. E. 1995. The Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications. Academy of Management Review, 20 (1): 65-91.
Donoher, W. J., Reed, R. and Storrud-Barnes, S. F. 2007. Incentive alignment, control, and the issue of misleading financial disclosures. Journal of Management, 33: 547-569.
Dodd P. and Warner J. B. 1983. On Corporate Governance: A Study of Proxy Contests. Journal of Financial Economics, 11: 401-438
Dunbar, R. L. M. and Schwalbach, J. 2000. Corporate reputation and performance in Germany. Corporate Reputation Review, 3 (2): 115-123.
Einwiller, S.A., Carrol, C. E. and Korn, K. 2010. Under What Conditions Do the News Media Influence Corporate Reputation? The Roles of Media Dependency and Need for Orientation. Corporate Reputation Review, 12 (4): 299-315.
