Information content of risk
appetite: an examination of risk
appetite reporting and
information asymmetry in
Dutch listed companies
Master thesis
Accountancy & Control
Jeannette de Bruijn (5979935)
23-6-2014
First supervisor: dhr. prof. dr. B.G.D. (Brendan) O'Dwyer Second supervisor: dr. G. (Georgios) Georgakopoulos
Table of contents
List of abbreviations ... 2
1. Introduction ... 5
2. Literature review and hypotheses ... 8
2.1 Risk (appetite) reporting ... 8
2.1.1 Regulations on risk disclosures in the Netherlands ... 8
2.1.2 Concept of risk appetite ...10
2.1.3 Relation between risk appetite and organizational strategy ...11
2.1.4 Compliance with regulations ...11
2.1.5 Relevance of risk appetite disclosure ...12
2.2 Economic consequences of reporting ...14
2.2.1 Information asymmetry ...14
2.2.2 Theoretical relation between disclosure and cost of capital ...15
2.2.3 Empirical evidence on relation between disclosure and cost of capital ...16
2.2.4 Influence of market competition ...18
2.3 Hypotheses ...19
3. Research methodology ...21
3.1 Empirical measurement ...21
3.2 Analysis of annual reports ...24
4. Preliminary analysis ...29
4.1 Sample selection ...29
4.2 Compliance with the Corporate Governance Code ...30
4.3 Testing of regression assumptions ...32
4.4 Descriptive statistics ...34 5. Empirical findings ...38 5.1 Bid-ask spread ...38 5.2 Share turnover ...40 6. Conclusion ...42 References ...45
Appendix 1: Risk appetite disclosures by Dutch listed entities ...48
List of abbreviations
AEX Amsterdam Exchange Index
AMX Amsterdam Midcap Index
AScX Amsterdam Smallcap Index
Code Dutch Corporate Governance Code
COSO Committee of Sponsoring Organizations of the Treadway
Commission
DASB Dutch Accounting Standards Board (RJ)
GAAP Generally Accepted Accounting Principles
Monitoring Committee Dutch Corporate Governance Code Monitoring Committee
NBA Netherlands Institute of Chartered Accountants
NIVRA Royal Netherlands Institute of Registered Accounts
RJ Raad voor de Jaarverslaggeving (DASB)
Abstract
Purpose – The purpose of this research is to empirically examine the relation between risk appetite reporting in annual reports and the information asymmetry component of a firms cost of capital.
Research design and methodology – The study is conducted by performing a cross-sectional regression whereby the information asymmetry component of a firm’s cost of capital is the dependent variable and the disclosure of risk appetite the independent variable. Information asymmetry is examined using the proxies bid-ask spread and share turnover. A sample is used of 52 companies listed on the AEX, AMX or AScX with a statutory seat in the Netherlands. The research is conducted on firms listed during 2009 to 2012. Control variables include market value, volatility, price, shares outstanding, beta, profitability, capital intensity and leverage.
Findings – The study shows that from 2009 to 2012 an increasing amount of Dutch listed firms reported on risk appetite. However, in 2012 only approximately a third of the companies complied with the Dutch Corporate Governance Code. The regression analysis provides support for the presence of a negative association between risk appetite reporting and the bid-ask spread. This indicates that firms which report on risk appetite are associated with lower information asymmetry. However, the results on the second proxy for information asymmetry do not substantiate this argument. The coefficient for risk appetite reporting was found not be significant in the share turnover regression.
Conclusion – The research provides support for the argument that the disclosure of risk appetite has information value. However, due to the limitations of this research and the insignificant result for the share turnover proxy the conclusion should be drawn with substantive precaution.
Contribution – Just before the introduction of the concept of risk appetite in the Dutch Accounting Standard, this study examine the information value of risk appetite. The study addressed the non-compliance of Dutch listed firms with the Corporate Governance Code.
Limitations – The empirical model may be subject to endogeneity, the non-normal distribution of residuals and the small sample affects the reliability of the research. Also, the examination of the annual reports for risk appetite is a subjective undertaking. The risk appetite disclosure has not been examined in isolation which could imply that the measure actually represents something else (e.g. the overall disclosure strategy).
Keywords: Corporate governance, risk paragraph, risk appetite, cost of capital, information asymmetry
Samenvatting
Doel – Het doel van deze studie is het empirisch onderzoeken van de relatie tussen de rapportage van risicobereidheid in het jaarverslag en de informatie asymmetrie component van de kapitaalkosten van ondernemingen.
Onderzoeksopzet en methode – Voor de studie is een regressieanalyse waarbij de informatie asymmetrie component van de kapitaalkosten de afhankelijke variabele vormt. De rapportage van risicobereidheid in het jaarverslag is de onafhankelijke variabele in het onderzoek. Informatie asymmetrie wordt benaderd door het gebruik van twee variabelen, namelijk de bied-vraag prijs (bid-ask spread) en aandelenomzet (share turnover). Voor het onderzoek zijn 52 bedrijven gebruikt die genoteerd zijn aan de AEX, AMX of AScX met een Nederlandse statutaire zetel. De studie is uitgevoerd op bedrijven die gedurende de jaren 2009 – 2012 beursgenoteerd waren. De gebruikte controlevariabelen in dit onderzoek zijn: marktwaarde, volatiliteit, prijs, aantal uitstaande aandelen, bèta, winstgevendheid, kapitaalintensiteit en de financiële hefboom (leverage).
Resultaten – De studie laat zien dat tussen 2009 en 2012 een toenemend aantal Nederlandse beursgenoteerde ondernemingen een beschrijving op heeft genomen van de risicobereidheid van de onderneming in het jaarverslag. In 2012 leefde echter slechts één derde van de ondernemingen de Nederlandse Corporate Governance Code na ten aanzien van het principe om over risicobereidheid te rapporteren. De resultaten van de regressieanalyse ondersteunen het bestaan van een negatieve associatie tussen de rapportage van risicobereidheid en de bied-vraag prijs. Dit impliceert dat ondernemingen die hun risicobereidheid rapporteren geassocieerd zijn met lagere informatie asymmetrie. De resultaten van de tweede regressie met aandelenomzet waren insignificant en ondersteunen deze associatie daarom niet. Conclusie – Het onderzoek ondersteunt het argument dat de rapportage van risicobereidheid waardevolle informatie geeft. Door de beperkingen van het onderzoek en de insignificante resultaten van de regressie met aandelenomzetten moet deze conclusie met wezenlijke voorzichtigheid getrokken worden.
Bijdrage – Vlak voor de introductie van het begrip risicobereidheid in de Richtlijnen voor de Jaarverslaggeving, onderzoekt deze studie de informatiewaarde van risicobereidheid. De studie laat zien dat een deel van de Nederlandse beursgenoteerde ondernemingen de Corporate Governance Code niet volgt ten aanzien van de rapportage van risicobereidheid.
Beperkingen – Het empirische model kan leiden onder endogeniteit en de niet normale verdeling van de residuen. Tevens zorgt de regressie op een beperkt aantal observaties voor een lagere betrouwbaarheid. Tevens is de analyse van de jaarverslagen op de rapportage van risicobereidheid een subjectieve onderneming. De rapportage van risicobereidheid is niet in isolatie onderzocht waardoor het mogelijk is dat eigenlijk iets anders is gemeten (bijvoorbeeld de wijze van verslaggeving als geheel).
Trefwoorden: Corporate governance, risicoparagraaf, risicobereidheid, kapitaalkosten, informatie asymmetrie
1.
Introduction
Although risks and uncertainties faced by corporations can be regarded to be problematic, they also provide opportunities for growth. Risks and uncertainties are thus inherently related to doing business. In 2007 the Corporate Governance Committee advised Dutch listed entities to report their attitude regarding the most predominant risks the entity faces. This attitude is also called the risk appetite (in Dutch: risicobereidheid) of a firm. The advice has been established in the Dutch Corporate Governance Code since its revision in 2009. Listed entities with a statutory seat in the Netherlands should therefore report on managements’ attitude regarding the most predominant risks of the firm.
Among others Mertens and Blij (2008), Eumedion (2014) and Van Daelen (2013) signal the importance of the disclosure of an entities’ risk appetite to stakeholders. Reporting the risk appetite of a firm could help investors gain insight in the risk attitude of a firm, evaluate how risky a firm operates and thereby determine how risky an investment is. However, prior research (Mertens and Blij (2008) and NIVRA (2010)) indicates that only a minority of the Dutch entities report on their risk appetite. According to Mertens and Blij (2008, p. 46), disclosing information on an entities’ risk appetite was still in its infancy in the Netherlands in 2008. This research examines the current status of risk appetite reporting in the Netherlands and the information value of risk appetite disclosures. Conducting research on this topic is especially interesting and topical as the Dutch Accounting Standards Board has recently proposed to expand the group of risk appetite reporting companies in the Netherlands. Not only will listed companies be encouraged to report on risk appetite, but also other large and medium-sized Dutch entities. The proposal is still under debate and this empirical study can provide valuable insights for the discussion on the future of risk appetite reporting in the Netherlands.
Economic theory suggests that an information gap between corporations and shareholders or between shareholders (also known as information asymmetry) increases the cost of raising capital due to the introduction of adverse selection (Leuz and Verrecchia (2000), Leuz (2003) and Botosan, Plumlee and Xie (2004)). Increasing the level of disclosure may diminish the information asymmetry and therefore lower the cost of capital (Leuz and Verrecchia, 2000). The theory on information asymmetry, cost of capital and disclosure is further outlined in the next paragraph. The production of information is however not without costs (direct and indirect) and therefore introducing new disclosure regulation should have a valid reason. Introducing new disclosures in annual reports should provide useful information (decision usefulness approach to accounting, Scott, 2012, p. 67). Although the economic consequences of risk disclosures has been studied by others (Kravet and Muslu, 2013 and Campbell et al., 2014), the disclosure of risk
appetite has not specifically been examined. This study is the first to focus specifically on the economic consequences of risk appetite reporting. The study examines whether there is a relationship between the disclosure of risk appetite and the information asymmetry component of a firms’ cost of capital. Hereby, I want to examine whether the disclosure of risk appetite provides valuable information to the users of annual reports. The research question is formulated in the following way:
What is the relationship between the disclosure of risk appetite and information asymmetry, measured by the proxies bid-ask spread and share turnover, in Dutch listed companies?
The study is conducted by performing a cross-sectional regression analysis using a sample of 52 firms listed on the Amsterdam Exchange Index (AEX), Amsterdam Midcap Exchange (AMX) or Amsterdam Smallcap Index (AScX) with a statutory seat in the Netherlands. The economic consequences of risk appetite reporting are examined for the years 2009 to 2012. The information asymmetry component of a firm’s cost of capital is proxied by the bid-ask spread and share turnover. To examine the risk appetite disclosure a dummy variable is used.
The study shows that in 2012 21 firms listed on the Amsterdam Stock Exchange reported on risk appetite, which amounts to a proportion of 31%. From 2009 to 2012 an increasing amount of Dutch listed firms reported on risk appetite. The output of the regression supports the presence of a negative association between risk appetite reporting and the bid-ask spread. This indicates that firms which report on risk appetite are associated with lower information asymmetry. However, the coefficient for risk appetite reporting was found not be significant in the share turnover regression. That result does not substantiate the argument that risk appetite reporting firms are associated with lower information asymmetry. The limitations of this research and the insignificant result for the share turnover proxy imply that a conclusion should be drawn with substantive precaution.
Verrecchia (2001) and Botosan, Plumlee and Xie (2004) have called for more research on the relation between public reporting and information asymmetry. This study responds to this call by examining the economic effect of risk appetite reporting. Although there is an increasing amount of literature on risk disclosures, I believe such a study has not yet been conducted in the Netherlands or elsewhere. Previous studies have primarily focused on how companies should report on risks and not on the effect of the disclosures. I consider conducting this research in the Netherlands to be relevant as the guidelines of the Dutch Accounting Standards on risk reporting are about to change. In March 2014 the Dutch Accounting Standards Board has published a
proposal which entails that large and medium sized entities should report on risk appetite. The amount of Dutch entities that will need to report on risk appetite will thus be enlarged. Performing this research may be helpful in assessing the relevance of disclosures on risk appetite of Dutch listed entities. It may support the claims of Mertens and Blij (2008), Eumedion (2014, 2013 and 2010) and Van Daelen (2012b and 2013) by providing empirical evidence on the relevance of providing information on risk appetites.
The societal relevance of the study is related to the relevance and the importance of risk disclosures. Increasing disclosure levels for entities may increase reporting costs, whereas it is questionable whether increased disclosure is beneficial to the entity and its stakeholders. By examining the economic consequences of the disclosure of risk appetites it can be determined whether increased disclosure lowers the cost of equity capital. The development of regulation on the disclosure of risk appetites might be helped by research indicating the economic consequences of the disclosure and further elaborating on the concept of risk appetite and the compliance. Also this study sheds a light on the compliance of Dutch firms with the Corporate Governance Code.
The remainder of this paper is structured in the following way. In paragraph two prior literature on risk (appetite) reporting and the economic consequences of reporting is examined. In this paragraph the hypotheses are also presented. In the following paragraph, the research methodology is outlined whereby a description is given of the empirical measurement and the manner in which the annual reports are analyzed is described. Subsequently, paragraph four comprises the preliminary analysis which explains how the sample is selected, what the tests on the regression assumptions have indicated and it presents the descriptive statistics. This paragraph also indicates the compliance of the sample with the risk appetite prescription of the Code. Paragraph five contains the empirical findings of the study. The final paragraph provides a conclusion and indicates the limitations of the study and directions for further research.
2.
Literature review and hypotheses
This paragraph provides an in-depth analysis of the literature on risk (appetite) reporting and the economic consequences of reporting. The first part of this paragraph examines literature on risk (appetite) reporting. It first outlines to what regulations regarding the disclosure of risk Dutch listed companies need to comply with, followed by an examination of literature on the concept of risk appetite, the relation between risk appetite and organizational strategy and on the level of compliance with regulations of Dutch listed companies. The third paragraph analyses the economic consequences of reporting. Specifically, the information asymmetry theory is considered, the theoretical and empirical relation between the information asymmetry component of a firms’ cost of capital and disclosure and the influence of market competition on the relationship.
2.1 Risk (appetite) reporting
2.1.1 Regulations on risk disclosures in the Netherlands
Dutch listed entities have to comply with different regulations which specify in what way entities should report on risk. Dinant (2010) explains what different regulations apply to the risk paragraph of these entities. First of all, article 391(1) of Book 2 of the Dutch Civil Code should be applied by Dutch listed corporations. It determines that the entity should give a description in its annual report of the most important risks and uncertainties it faces. The Civil Code doesn’t provide any further information of what information the risk paragraph should deliver, nor does it specify how entities should ensure control over the risks. This article of the Dutch Civil Code is not only applicable to listed companies but to all Dutch legal entities.
Secondly, the Dutch Corporate Governance Code (‘the Code’) is relevant for Dutch listed entities. As described by Van Daelen (2012a, p. 89) the Code, also known as the Frijns Code, originated from a prior version called the 2003 Code Tabaksblat. These codes are developed by the Dutch government as an implementation of European Commission Action Plan which had the intention to stimulate member states to write corporate governance codes for listed companies. The Code provides principles which are considered to be the generally accepted principles of good corporate governance in the Netherlands. Typical of the Dutch Code is that a ‘comply-or-explain’ basis is used. The principle means that entities must adhere to the principles or explain in their annual report how and why they don’t follow the principles outlined in the Code.
Principle II1.4 of the Code determines that in the annual report of the company, management should describe the most important risks the company faces in relation to its
strategy. An appendix to the Code provides an explanation of the terms used. Paragraph 14 of the Preamble specifies that the appendix with the explanation of and notes to terms used in the Code are an integral part of the Code. The appendix complements Principle II1.4 of the Code by stating that the management board should report on risks in the following way: “The description should also specify the company’s risk profile, in other words its risk appetite and, as far as possible, its risk sensitivity” (p. 39). The next section discusses the concept of risk appetite in more detail.
Apart from the Dutch Civil Code and the Corporate Governance Code, Dutch listed companies also have to pay attention to the guidelines of the Dutch Accounting Standards Board for the risk paragraph in their annual report. Paragraph 110a of guideline 400 of the Dutch Accounting Standards for medium sized and large legal entities states that entities should mention the most predominant risks and uncertainties in their annual report. The current paragraph states that risks related to strategy, operations, finance, financial reporting and legislation can be considered when determining the entities’ most prevalent risks and uncertainties. The list is not meant to be exhaustive. As Dinant (2010, p. 157) indicates guideline 400 has more to do with providing information on the risk profile of the company than on what control measures a company has to implement and whether it is in control. However, the Dutch Corporate Governance Code does provide regulations about control measures and how companies should disclose them. Although the Dutch Accounting Standards do not have a formal status, the standards can indicate what are regarded Dutch Generally Accepted Accounting Principles (Dinant, 2010, p. 157). A proposal to change the paragraphs on risk reporting in the guidelines was published at the end of March 2014. The proposed amendment explicitly mentions the requirement to report on risk appetite in paragraph 110c guideline 400 (RJ, 2014).
The three standards that are dealt with in the prior paragraphs deal with the regulation on the disclosure on risk in annual reports. In the context of risk management, the framework of the Committee of Sponsoring Organizations of the Treadway Commission (COSO) is regularly mentioned. The framework deals with internal control systems of corporations and how effective controls can be implemented. It is not legally binding for Dutch listed entities. The concept of risk appetite is also mentioned in this context and the COSO provides information on understanding its concept and how to communicate it (COSO, p.1). Furthermore, for Dutch banks and insurers special codes have been developed regarding their risk management. In the Dutch Banking Code and the Dutch Insurer’s Code the concept of risk appetite is mentioned in relation to how management should ensure proper risk management and remuneration. The codes do not stipulate how the banks and insurers should report on risk appetite.
This section provided an overview of what different regulations regarding the risk paragraph are applicable for Dutch listed entities. Three different standards are relevant for the risk paragraph of Dutch listed entities: article 2:391 of the Dutch Civil Code, the Dutch Corporate Governance Code and paragraph 110a of guideline 400 of the Dutch Accounting Standards for medium sized and large legal entities. However, only the Dutch Corporate Governance Code and the draft paragraph 110c of guideline 400 determine that listed companies should report on their risk appetite. The work of COSO can be helpful for corporations who need to report on risk appetite. The commission explains what the concept is about and how corporations should formulate their risk appetite. In the next section, an elaborate explanation of what the concept of risk appetite comprises is provided.
2.1.2 Concept of risk appetite
What the concept of risk appetite entails is outlined by several authors and institutions. According to Mertens and Blij (2008, p. 14) the risk appetite should be viewed as the preparedness of management to take risk. Van Daelen (2013) defines the concept in a similar way by stating that it is the level of risk the management of an entity is prepared to take. According to the COSO risk appetite is “the amount of risk, on a broad level, an organization is willing to accept in pursuit of value’ (2012, p. 1). Dinant (2010, p.155) and Eumedion (2010, p. 2) describe risk appetite in a comparable manner by stating that it is the attitude of the entity regarding the most predominant risks. Their definition is probably built on the one stated in the Dutch version of the Corporate Governance Code (p. 39). The code equates the concept of risk appetite with an entities’ risk profile. Whereas Mertens and Blij (2008) and Van Daelen (2013) do not limit the attitude to the most predominant risks, the others do. Nevertheless, all the mentioned definitions somehow imply that risk appetite has to do with the attitude of management regarding risk.
The proposed amendments of the Dutch Accounting Standards Board do however seem to take a somewhat different approach. According to the DASB entities should give a description of whether they are prepared to cover risks and uncertainties. This explanation of the concept gives the impression that risks and uncertainties are regarded to be negative and should be covered, whereas risks could also create opportunities. The DASB speaks about covering risks, whereas the definitions stated in the prior paragraph seem to consider risk appetite to be about the level of taking risk. As entities will currently need to comply with the Dutch Corporate Governance Code, the definition of risk appetite used by the Code will be used throughout this paper.
2.1.3 Relation between risk appetite and organizational strategy
The strategy of an organization is related to its risk appetite. NIVRA (2009, p. 8) points out that the strategy can determine the risk appetite and vice versa. Furthermore, Van Daelen (2013) clarifies that the information in the risk paragraph can be of value when the relation among risks and the overall approach to the risks is examined in combination with the entities’ strategy. Deloitte performs a yearly examination of the risk paragraphs of Dutch AEX-listed entities. In their study of the annual reports of 2012 (2013, p. 3) they point out that entities should pay more attention to report on the relation between risk management and the strategy of the organization. According to the NBA risk appetite should be part of the long term determination of goals (2013, p. 11). The Code pays attention to the relation of risk appetite and strategy on a higher level as it determines in Principle II1.4 that a description should be given of the most predominant risks in relation to the strategy of the company.
2.1.4 Compliance with regulations
Several researchers have studied to what degree companies comply with the Code regarding the risk paragraph. Mertens and Blij (2008, p.6) examined risk paragraphs in annual reports of 2007 of Dutch listed companies. They indicate that only 15% of the Dutch listed entities reported on their risk appetite in 2007. Furthermore they show that AEX (33%) companies are more likely to report on risk appetite than those listed on the AScX (5%). The NIVRA performed a similar study in 2009 which indicated that only a minor improvement took place. According to the research 19% (p. 9) of the Dutch listed entities report on their risk appetite. Kuijpers (2011, p. 9) provides an explanation for the non-compliance with the Code. He believes that not all companies disclose their risk appetite as the requirement is only stated in the explanatory appendix of the Code. However, as previously pointed out, the explanatory part of the Code is part of the Code and should thus be complied with or an explanation for non-compliance should be provided.
On a yearly basis, the Monitoring Commission Corporate Governance Code (‘Monitoring Commission’) evaluates the compliance of entities under its oversight. In the final report of the Monitoring Commission under the supervision of Jos Streppel it is stated that the disclosure on risk management is almost not worth monitoring any more in 2013 as the principles on risk are so well applied. Whether the Monitoring Commission believes the reporting of risk appetite is not worth monitoring (anymore) is unclear. The last time the Monitoring Commission mentions the risk appetite in its evaluation was in 2007 (p. 7). In the evaluation the Monitoring Commission recommends that entities give a description of their risk appetite.
Furthermore, a yearly price is awarded to the best Dutch annual report which is known as the Henri Sijthoff price. Although the report of 2013 states that risk paragraphs are regularly insufficient (p. 3), disclosure of risk appetite is not explicitly one of the criteria in the examination of the annual reports. However in the description of the criteria it is mentioned that the risk paragraph does need to provide users with good insight in the risk profile of the entity. Whether it includes risk appetite disclosures is uncertain. Surprisingly, both the Monitoring Commission and the Henri Sijthoff price do need seem to explicitly consider the compliance on the disclosure of risk appetite whereas the previous paragraph indicates that compliance with the Code regarding the disclosure of risk appetite is questionable.
2.1.5 Relevance of risk appetite disclosure
Several papers state that companies should better disclose their risk appetite. Mertens and Blij (2008, p.46) state that the examined disclosures weren’t concrete enough. One of their conclusions is that there is much room for improvement of the disclosure on risk appetite. The reporting on risk appetite is said to be infantile and inadequate. Mertens and Blij believe better risk appetite disclosures should be made to provide input for investors to enable them to make good and well-informed decisions. The subsequent study of the NIVRA (2010, p. 12) also concludes that the disclosure of risk appetite can be significantly improved. Dinant (2011, p. 155) states that institutional investors demand risk appetite disclosures in order to make deliberate investment decisions. Furthermore, Veenis and Dinant (2011, p. 13) believe that users of financial statements value the disclosure of risk appetite.
More recent claims of the relevance of risk appetite disclosures for Dutch entities is given by Van Daelen (2012b), Deloitte (2013), the reaction of Eumedion on the public management letter of the NBA (2013) and the recent reactions from Eumedion (2014) and EY (2014) on the proposal of the DASB (2014). Van Daelen (2012b) defines ten building blocks for the reporting on risks and uncertainties for Dutch entities. According to her, society calls for more transparency when it comes to risk. She believes companies should report on their risk appetite in order to prevent users of the report having to determine the firms’ preparedness to take risks themselves. In its report on the annual reports of 2012 Deloitte (2013, p. 4) states that companies pay too little attention to the reporting of risk appetite.
The NBA published a management letter on risk in 2013. In its reaction to the letter, Eumedion (p. 21) states that it has recommended the disclosure of risk appetite per risk category for a longer period. Eumedion (a corporate governance forum for institutional investors) publishes yearly a focus letter which is sent to the 75 largest Dutch listed entities to ask attention
for certain matters. In its focus letter of 2010 it stresses the importance of risk appetite whereas in subsequent years 2011 and 2012 only the risk paragraph as a whole is accentuated. In the letters regarding 2013 and 2014 little or no reference is made to the risk paragraph. This study may be helpful in explaining whether the disappearance of attention of Eumedion has to do with increased compliance with the risk appetite regulation of Dutch listed entities.
Eumedion also responded positively on the proposal of the DASB regarding the risk paragraph of large and medium sized entities. Eumedion states that it welcomes the introduction of risk appetite reporting in the Dutch Accounting Standards (Eumedion, 2014). EY (2014) is more critical on the proposal of the DASB to prescribe the disclosure of risk appetite for all large and medium-sized entities, whether they are listed or not. They support the idea of providing users of annual reports with information about how much risk an entity is prepared to take. However, they are critical on the formulation of the prescription which may lead to ‘boilerplate’ language and useless information.
Above, the literature on the risk (appetite) reporting has been examined. The section illustrates that the concept of risk appetite is not unambiguously defined. Furthermore it shows that the compliance with the Dutch Corporate Governance Code regarding the disclosure of risk appetite is questionable. However, the attention for the issue seems to be mixed. The previous section provides an argument that the disclosure of risk appetite is regarded to be useful. As these papers indicate, the information on risk appetite provided in the annual reports is believed to add value and therefore be relevant.
2.2 Economic consequences of reporting
In the previous section the literature on risk (appetite) reporting has been described. This section considers what the economic consequences of reporting are according to the literature. Reviewing the literature on the relation between reporting and the information asymmetry component of the cost of capital is relevant for this paper as it provides a framework for the research methodology. Also this study complements the current literature on the economic consequences of reporting. This section is structured in the following way. First, the information asymmetry theory is outlined. Subsequently, the theoretical and empirical literature on the relation between disclosure and a firms’ cost of capital is examined. At last the literature on the influence of market competition on the prior relationship is analyzed.
2.2.1 Information asymmetry
The need for financial reporting stems from the concept of information asymmetry. Before this reasoning is further developed, the theory on information asymmetry is outlined. Information asymmetry is built on the notion of information economics, which entails that in a business transaction one of the parties may have an information advantage over the other. When one party does in fact have an information advantage over the other they do not have equal access to information. In that case information asymmetry exists.
Information asymmetry can be divided into two different types, namely adverse selection and moral hazard. Scott (2012, p. 21) defines adverse selection as: “a type of information asymmetry whereby one or more parties to a business transaction, or potential transaction, have an information advantage over other parties”. The party with the most information may abuse his or her information advantage over the other party and thereby handle adverse to the interests of that party. Arrow (1984, p. 3) refers to this type of information asymmetry as ‘hidden information’. The party with the least information (the investors) will realize that the entity has an information advantage and therefore be less willing to invest leading to malfunctioning capital markets. According to Scott (2012, p. 21) adverse selection can be mitigated by financial reporting and accounting as the entity transfers inside information to its investors (the outsiders). The second type of information asymmetry is moral hazard. In business situations moral hazard may occur when ownership and control are not in one hand but separated between for example management and its investors. The principle (investor) will want to know whether the agent (management) acts in his or her best interest. However, it will be impossible for the investors to observe the work of management constantly as that would be too costly (Armstrong et al, 2010, p. 199). Moral hazard has thus to do with the (in)visibility of actions of the agent, also
referred to by Arrow (1984, p. 3) as ‘hidden action’. To ensure that management works in the best interest of the investors, the latter will come up with performance measures with which the performance of management can be determined (Armstrong et al., 2010, p. 199).
Now that the concept of information asymmetry has further been outlined, the need for financial reporting can be explained. The necessity of reporting stems from two roles it plays, namely the valuation and stewardship role. Beyer et al. (2010) explain that outsiders require financial information in order to assess whether a firm is interesting to invest in (ex-ante) and also to control the performance of management (ex-post). The contracts (both explicit and implicit) that are formulated due to the separation of ownership and control usually utilize accounting information to control the performance of management. When the firm provides (reliable) information whereby its management can be evaluated, the capital market will require a lower return (Beyer et al., 2010, p. 297).
2.2.2 Theoretical relation between disclosure and cost of capital
Hughes, Liu and Liu (2007) state that a firm’s cost of capital is equal to the risk-free return plus a risk premium. According to Easley and O’Hara (2004) the cost of capital has a wide impact on the organization as it can affect the capital structure of the entity, its operations, its investments and its profitability. Reducing the cost of capital can therefore be important for corporations. Theoretical studies imply that the cost of capital can be reduced by disclosures. The relation between the level of disclosure and cost of capital is explained by Campbell et al. (2014).
First of all, disclosure can positively affect the liquidity of securities as a higher level of disclosure reduces the information gap between current and potential shareholders. The study of Easley and O’Hara (2004) makes an interesting contribution to the literature as it shows that the amount and precision of information that is available to shareholders can affect an entity’s cost of capital. How a corporation decides to report can thus have an impact on its cost of capital. The study implies that investors demand a higher rate of return when information asymmetry exists. The uninformed investor is exposed to more risk as he knows that the other party has more information and therefor demands a higher return on his investment. The higher demand for returns stems from the risk the uninformed investor is exposed to as informed investors are better capable of adjusting their portfolio weights to include new information. A lower level of public information (thus more private information) introduces some kind of systematic risk. Investors want to be compensated for this risk which leads to a higher cost of capital.
The second explanation provided by Campbell et al. (2014) is built on the notion that due to a lower level of disclosure investors may have more difficulty with estimating the future rate of
return of the entity. The estimation errors are not diversifiable and therefor negatively impact the cost of capital. The final interpretation of the relation between disclosure and the cost of capital as presented by Campbell et al. (2014) originates from the study of Lambert et al. (2007). The latter show that the quality of information affects cost of capital in two ways. First, the quality of information has a direct effect on the expected distribution of future cash flows by market participants. This effect occurs as the covariances between cash flows of the firm and other firms are impacted by the disclosure quality. Second, the quality of information has an indirect effect on the cost of capital as it impacts real decisions made by firms which have an effect on the allocation of future cash flows. However, Lambert et al. (2007) note that their study is done on a firm level. The indirect impact of disclosure on the cost of capital may differ if examined across different companies.
2.2.3 Empirical evidence on relation between disclosure and cost of capital
In the previous paragraph the theoretical basis for the relation between the level of disclosure and a firm’s cost of capital has been outlined. In this paragraph empirical studies on this relation are examined. Leuz and Verrecchia (2000) for example empirically examined the economic consequences of German firms switching to IAS or U.S. GAAP. The later were considered to require more disclosures. Their research deals with the question whether increased disclosure, due to the switch from German to IAS or U.S. GAAP, decreases the cost of capital related to information asymmetries of the German firms. They use the following proxies to examine the information asymmetry aspect of a firms’ cost of capital: the bid-ask spread, the level of trading and the volatility of share prices. Leuz and Verrecchia (2000) conclude that their findings support the idea that companies who have a relatively higher disclosure level may experience economically and statistically significant advantages. According to the researchers the international reporting standards were associated with a lower level of the bid-ask spread and a higher turnover of shares. They controlled for several firm characteristics and self-selection. Leuz and Verrecchia (2000) did not find an association between share price volatility and the use of IAS or U.S. GAAP. They state that their findings should be cautiously interpreted as their sample was small. Nevertheless, the findings show that it didn’t matter whether the firms applied IAS or U.S. GAAP, indicating that the standard didn’t matter but only the increased disclosure level.
According to Leuz and Wysocki (2008) it is hard to draw conclusions on the relation between quality of information and a firms’ cost of capital as the literature is still developing. Their research explores the economic consequences of disclosure and provides an overview of
literature on the subject. Among other things, it deals with the relation between voluntary disclosure, the quality of accounting information and the cost of capital of a firm. Similar to the paper of Leuz and Verrecchia (2000), this paper also points out that self-selection is an issue when examining voluntary disclosure and financial reporting by corporations. They conclude that the empirical evidence that they have examined “appears to be sensitive to and can vary across different measures of cost of capital (i.e. realized returns versus ex ante cost of capital proxies), types of firms (i.e. different sizes), with the presence of other intermediaries (i.e. financial analysts), across types of disclosures or earnings attributes (i.e. annual reports versus timely disclosures versus conservative earnings), across types of investors (shareholders versus bondholders), and across different institutional environments (i.e. U.S. versus other markets).” (p. 34).
The paper of Leuz and Schrand (2009) acknowledges that the prior research on information quality, disclosure and the cost of capital is inconclusive. According to them the existence of information asymmetry can cause investors to increase the market risk premium and the asymmetry may increase the firm-specific cost of capital (p. 9). They examined the relationship from a totally different perspective than prior research. Their research used the “exogenous cost of capital shock” (p.2) that occurred after the Enron scandal as a starting point to study the effect of the shock on disclosures by corporations. By reversing the approach they tried to alleviate concerns that were put forward by prior research about the causality of the relation between information quality, disclosure and a firm’s cost of capital. Leuz and Schrand (2009) thereby tried to prevent that omitted variables together determined both the firms’ cost of capital and how the entity discloses information. They conclude that their experiment shows that entities increased their level of disclosure after the beta shock that occurred due to the Enron event. Entities that experienced positive shocks in their cost of capital, had larger needs of finance or had to deal with credibility concerns were more inclined to increase disclosure. Furthermore, the beta shocks stimulated firms to issue extra interim reports. Leuz and Schrand (2009) conclude that their research indicates that the disclosure responses of the entities “have significant subsequent market reactions (e.g. abnormal volume) and are effective in reducing the cost of capital” (p. 40).
Mohd (2005) examined the level of information asymmetry in software firms which expensed research and development costs compared to firms which capitalized such costs. He used the bid-ask spread and share turnover as proxies for information asymmetry. Capitalization is considered to lower information asymmetry as capitalizing research and development costs
provides valuable information to investors. Mohd (2005) examines whether such an association exists and finds evidence supporting this relation.
Kravet and Muslu (2013) do not study information asymmetry specifically but focus on the impact of risk disclosures on the risk perceptions of investors. They examine several measures around the 10-K filing to determine whether investors update their beliefs about the reporting company. Kravet and Muslu (2013) report that during the three days after the 10-K filing, a positive association has been found between a change in the risk disclosures and the amount of trading. A change occurs when there is a difference between the amount of sentences on risk disclosures compared to the previous year. According to Kravet and Muslu (2013) the risk disclosures cannot be called boilerplate.
2.2.4 Influence of market competition
Different researches indicate that market competition should be considered when examining the relationship between a firms’ cost of capital and information asymmetry. Armstrong et al. (2011) for example examined the influence of market competition on information asymmetry and the cost of capital. According to them in a perfectly competitive market, information asymmetry does not have a separate effect on a firms’ cost of capital. However, when the market is not considered to be competitive the cost of capital may be influenced by the level of information asymmetry. So, Armstrong et al. (2011) found out that the level of market competition should be considered a relevant conditioning variable when the relationship between cost of capital and information asymmetry is explored.
The conclusion of Armstrong et al. (2011) is confirmed by the research of Lambert et al. (2011) as they state that the relation between information asymmetry and a firms’ cost of capital is critically affected by the level of competition in the capital market. However, Lambert et al. (2011) limit their definition of information asymmetry to asymmetry between different investors. They explain that in a market which is not perfectly competitive, the level of market illiquidity is influenced by the amount of information asymmetry in a market. The illiquidity of the market in turn increases the cost of capital. Thus the cost of capital of an entity can be influenced by the interaction of asymmetric information and the competitiveness of the market. This implies that the results of this study should be interpreted in light of the Dutch level of market competition. The conclusions of this study should be cautiously generalized to other markets, thus affecting the external validity of the study.
2.3 Hypotheses
In the prior two paragraphs the literature on risk (appetite) reporting and the economic consequences of reporting has been outlined. This study aims to contribute to the literature on risk appetite reporting. The study of Mertens and Blij (2008, p.6) showed that only 15% of the Dutch listed entities described their risk appetite in the annual reports over 2007. The research of the NIVRA on the reports over the year 2009 show a small improvements to 19% (p. 9) of the Dutch listed entities reporting on risk appetite. The revised Dutch Corporate Governance Code became effective from the beginning of 2009 and prescribes the disclosure of risk appetite in principle II1.4. Due to the establishment of the principle in the Code and the fact that an increase of risk appetite reporting has been established by the NIVRA from 2007 to 2009, I hypothesize that during the period 2009-2012 more companies will report risk appetite:
H1: The proportion of Dutch listed companies reporting on risk appetite will increase during the period 2009 to 2012.
The amount of companies listed on the AEX, AMX and AScX is limited. Therefore, small changes in the index may have a significant impact on the proportion. As the denominator of the ratio may vary due to the number of companies with a foreign statutory seat and companies for which no data can be gathered, also the absolute amount of reporting companies will also be examined:
H2: The absolute amount of Dutch listed companies reporting on risk appetite will increase during the period 2009 to 2012.
This research also aims to add to the existing literature on the relation between disclosure and information asymmetry. Specifically, it examines the relation between risk appetite reporting and information asymmetry. The theoretical and empirical literature seems to illustrate a relation between the level of disclosure and the cost of capital. According to Easley and O’Hara (2004) information asymmetry makes investing more risky, resulting in a higher demand for return by investors and thus a higher cost of capital for the firm. Campbell et al. (2014) explain that more information production by companies may help investors to estimate the future return on their investment, resulting in a lower cost of capital. Furthermore, the quality of the provided information may affect the expectations of market participants on distribution of future cash flows and can also impact actual behavior of companies affecting cash flows. Leuz and
Verrecchia (2000) and Mohd (2005) studied information asymmetry in relation to disclosure level. Both came to the conclusion that more disclosure is related to lower information asymmetry. They found significant results indicating that a higher level of disclosure is related to a lower bid-ask spread and higher share turnover, which are both regarded to be proxies of information asymmetry. Leuz and Wysocki (2008) and Leuz and Schrand (2009) indicate that prior studies on quality of information and cost of capital are inconclusive. As the reporting of risk appetite is considered to be valuable information (Mertens and Blij (2008), Dinant (2011), Veenis and Dinant (2011), Van Daelen (2012b), Deloittte (2013), Eumedion (2013 and 2014) and EY (2014)), I hypothesize that its disclosure can lead to lower information asymmetry. Similar to the studies of Leuz and Verrecchia (2000) and Mohd (2005), this study will examine information asymmetry using the proxies bid-ask spread and share turnover. Consistent with these studies I expect that there is a relation between the disclosure of risk appetite and information asymmetry:
H3: The disclosure of risk appetite is negatively associated with the bid-ask spread.
3.
Research methodology
The aim of this study is to empirically examine whether there is an association between the disclosure of risk appetite in annual reports of Dutch listed entities and their information asymmetry component of the cost of capital. The research is conducted using a cross-sectional multiple linear regression analysis. The study examines the relationship between the independent variable risk appetite disclosure and a dependent variable, which is the information asymmetry component of cost of capital. To conduct the research the risk paragraphs in the annual reports of the Dutch listed companies are examined. The annual reports are retrieved from www.company.info or from the corporate websites. The other input for the regression analysis is gathered from the Datastream database.
3.1 Empirical measurement
The research methodology is primarily based on the research conducted by Leuz and Verrecchia (2000) and Mohd (2005). Leuz and Verrecchia (2000) looked at the effect on the information asymmetry component of a firms cost of capital of firms who switched from a German to an international reporting scheme, and thereby increased their level of disclosure. Their research used the following proxies to examine the information asymmetry component of the cost of capital: the bid-ask spread, trading volume and share price volatility. However, Leuz and Verrecchia (2000) are only able to find a significant effect on the bid-ask spreads and trading volumes of switched firms. Later research conducted by Mohd (2005) left out the share price volatility and only used the bid-ask spread and share turnover as proxies for information asymmetry. The study of Mohd (2005) examined the effect of accounting for software developments (capitalizing versus expensing) on information asymmetry. The bid-ask spread has been used by several other studies to examine information asymmetry (Campbell et al. (2014), Daske et al. (2008), Ecker et al. (2006), Leuz (2003)).
Mohd (2005) illustrates why the bid-ask spread and share turnover are proxies for information asymmetry. A high level of information asymmetry has two relevant effects. First, a high level of information asymmetry between an entity and its investors enables informed investors to trade at the expense of uninformed investors. Uninformed investors will realize that the informed investors have an information advantage. To protect themselves against the information disadvantage, uninformed investors will increase the bid-ask spread in order to compensate for their increased risk exposure. According to Leuz and Verrecchia (2000) the bid-ask spread addresses the adverse selection aspect of information asymmetry directly.
A second effect of high information asymmetry is that it will negatively impact share turnover. According to Petersen and Plenborg (2006, p. 133) the share turnover measure shows “the willingness of some investors to sell shares and others to buy”. The amount of trading is thus a liquidity measure. The willingness to trade is affected by the existence of information asymmetry. Mohd (2005) explains that uninformed investors are less likely to trade under high information asymmetry as they will realize that they have less information than others. The greater the amount of informed investors, the greater the demand for the stock (Easly and O’Hara, 2004). Informed investors will perceive the stock to be less risky, as due to the higher amount of information they are better capable of estimating the future rate of return and the distribution of future cash flows (Campbell, 2014).
Building on prior research, the proxies for information asymmetry used in this study are the bid-ask spread and share turnover. The relation between the information asymmetry component of the cost of capital and the disclosure of information about an entities’ risk appetite is examined using a dummy variable for risk appetite. The regression method used is the Ordinary Least Squares (OLS) regression (Leuz and Verrecchia, 2000). The analyses are performed using the following regression equations:
1. BID-ASKit = α0 + α1RISK_APPETITEeit + α2MVit + α3VOLATILITYit + α4PRICEit + α5SHARE_TURNOVERit+ α6BETAit + α7PROFITABILITYit +
α8CAPITAL_INTENSITYit +α9LEVERAGEit +v’t
2. SHARE_TURNOVERit = β0 + β1RISK_APPETITEit + β2MVit +
β3SHARES_OUTSTANDINGit + β4BID-ASKit + β5BETAit + β6PROFITABILITYit + β7CAPITAL_INTENSITYit + β8LEVERAGEit + w’t
Whereby:
Variable Definition Definition based on Datastream codes
BID-ASKit Logarithm of the yearly average
of bid-ask spreads. The bid-ask spread is the absolute difference between the bid and ask price at the end of each trading day divided by the average of the bid and ask price at the end of each trading day.
Leuz and Verrecchia, 2000, p. 105
Price - Bid: Euronext Amsterdam (AEX) (PB.AM) & Price - Ask: Euronext Amsterdam (AEX)
PA.AM)
SHARE_TURNOVERit Yearly average of the logarithm
of the daily turnover by volume
Mohd, 2005, p. 1219 Turnover By Volume: Euronext Amsterdam
RISK_APPETITEit Dummy variable whereby 0
means the company is not reporting on risk appetite and 1 means the company is reporting on risk appetite.
MVit Logarithm of the market
capitalization.
Mohd, 2005, p. 1222 Market Capitalization (MV)
VOLATILITYit Logarithm of the total return
index.
Total Return Index (RI)
PRICEit “Annual average of the
logarithm of daily stock price”.
Mohd, 2005, p. 1221 Price (Adjusted – Default) (P)
SHARES_OUTSTANDINGit “Annual average of the
logarithm of daily number of shares outstanding”.
Mohd, 2005, p. 1221 Number of Shares (NOSH)
BETAit Beta, (BETA)
PROFITABILITYit Five year average of return on
assets. Return on assets = Net Operating Income / Net Assets
Leuz and Verrecchia, 2000, p. 105
Net Operating Income (WC01250) & Net Assets (WC02999)
CAPITAL_INTENSITYit Total Intangible Other Assets
Net + Net Property, Plant and Equipment divided by Net Assets
Leuz and Verrecchia, 2000, p. 108
Total Intangible Other Assets Net (WC02649), Property, plant and equipment net (WC02501) & Net assets (WC02999)
LEVERAGEit Long Term Debt divided by Net
Assets
Long Term Debt (WC03251) & Net Assets (WC02999)
Control variables:
Several factors may influence the relation between the reporting on risk appetite and the information asymmetry component of a firm’s cost of capital. Apart from several of the control variables used by Mohd (2005), some other firm aspects as presented by Leuz and Verrecchia (2000) and Daske et al. (2008) which may influence the disclosure strategy of firms are included. I expect the disclosure of risk appetite to be influenced by the riskiness of the business; therefore the control variable beta is included in the regression. The control variables are: market value, volatility, price, shares outstanding, beta, profitability, capital intensity and leverage.
3.2 Analysis of annual reports
Mertens and Blij (2008) and the NIVRA (2010) did not define in their papers how they have inspected the annual reports of Dutch entities for risk appetite disclosures. One of the authors of the NIVRA study was however able to indicate which firms they believed reported on risk appetite in the year 2009. The analysis of the NIVRA is used as a reference for this research to ensure consistency in the studies. Realising that the examination of the annual reports on risk appetite is a subjective undertaking, I have used the expertise and professional judgement of the authors of the NIVRA report to guide my examination.
In paragraph 2.1.2 the concept of risk appetite as described in prior literature has been discussed. Considering those definitions, firms are regarded to report on risk appetite in this study if they in some form describe their attitude regarding the risks the entity is faced with. As a first step in the exploration of a risk appetite description the entire annual report is searched for the words ‘appetite’, ‘profile’, ‘take’, ‘tolerance’, ‘accept’, ‘prepared’ and ‘risk’. Prior literature did not stipulate which words are consistent with the notion of risk appetite. Therefore this list has been formulated after reading a subset of the annual reports in the sample, which is also done by Kravet and Muslu (2013). If the annual report contains one of the stated words that element is further examined by determining whether it describes the attitude of the company regarding the risks it is faced with. The list of words is thus used to identify parts in the entire annual report that could describe a firms’ risk appetite. If the search does not provide results, the risk paragraph of the annual report is studied. The risk appetite description is most probably found in this part of the annual report as the prescription on risk appetite is part of Principle II.4 of the Dutch Corporate Governance Code which deals with the risk paragraph. If the risk paragraph does also not provide indications of a risk appetite description, the entity is considered not to report on it. Companies that did not report on risk appetite in a certain year, where supposed not to have reported in earlier years.
This study uses a manual content analysis as the amount of annual reports examined is small and this method has several advantages. According to Li (2010, p. 145) a manual analysis of textual disclosures ‘can be more precise, detailed and tailored to the specific research settings’. A computer-based analysis would have required a very distinct framework for risk appetite, which has not been provided by prior literature. The data is manually collected as this probably provides more relevant outcomes in the case of a less strict available examination framework. An important drawback of manual content analysis is however that it is harder to replicate the study because of the subjective judgement (Li, 2010).
As stated previously, the examination performed by the NIVRA in 2009 has been taken as a reference for this study. The results of the NIVRA study were compared to the results of this examination to see whether the examinations were similar. One difference between the two studies can be identified. The study of the NIVRA considered Ballast Nedam to be a risk appetite reporting company in 2009 contrary to Nutreco. Nonetheless, this study has identified both to be risk appetite reporting companies. This is done as their annual reports are quite similar in that they both mention the words ‘prudent’, ‘attitude’, ‘accept’ and ‘risks’:
Ballast Nedam generally adopts a prudent attitude to accepting risks.
Annual report Ballast Nedam 2009 page 16
In general Nutreco adopts a prudent attitude with respect to the acceptance of significant business risks.
Annual report Nutreco 2009 page 52
To illustrate how the examination in this study has been made several examples are provided. Some companies mention the term ‘risk appetite’ in their annual report without explaining what their attitude is. Stating that a firm has a risk appetite is thus not enough to be qualified as reporting on risk appetite. The description should provide the reader with information of what that appetite is. Heineken for example is not considered to report on risk appetite although it does mention the term in its report:
Risk appetite
The Company is recognised for its drive for quality, consistency and financial discipline. Entrepreneurial spirit is encouraged across the Group in order to seek opportunities that support continuous growth, such as business development and brand building, while taking controlled risks. The international spread of the country portfolio geographically and between mature and emerging markets, the robust balance sheet and strong cash flow form the context of the risk appetite of the Company.
Annual report Heineken Group 2012 page 36
In its annual report of 2012 Accell clearly describes its risk attitude:
Accell Group has a relatively high risk appetite related to innovation, development and marketing. At the same time, the company has a low level of risk appetite when this relates to product safety.
However, other companies are less plain about how willing they are to accept risks. The following element from the annual report of Macintosh Retail Group in 2012 was for example considered not to be risk appetite reporting as it merely describes a financial policy rather than a risk attitude:
Macintosh Retail Group has always adopted a conservative financial policy…
Annual report Macintosh Retail Group 2012 page 73
Conducting the first academic study on risk appetite has its constraints as it complicates how the annual reports could be analyzed for risk appetite. Furthermore, it restricts the possibilities for in-depth analyses. Li (2010, p. 147) states that at least three aspects of corporate disclosure can be of interest to research: “the amount of disclosure, the tone and the transparency (or readability)”. I believe the lack of clear boundaries on risk appetite reporting would have made the analysis so dependent on personal judgment that this doesn’t outweigh the benefits of a more in-depth analysis. With the definition of risk appetite as a basis it was already challenging to distinguish reporting from non-risk appetite reporting companies. As the risk appetite disclosures only make up a small part of the annual reports, the in-depth analysis would have been very sensitive.
Kravet and Muslu (2013) and Campbell et al. (2014) have empirically studied the disclosures on risk by analysing the amount of disclosures. I believe such an examination is inappropriate for the current study on the disclosure of risk appetite. Due to the fact that prior literature has not explicitly defined how risk appetite reporting can be determined it is hard to decide which words or sentences describe a firms’ risk appetite. The following part from the annual report of 2012 of Akzo Nobel illustrates the grey area which hinders the determination of which parts describe Akzo Nobels’ risk appetite:
Risk appetite
Clarity on risk appetite and boundaries that determine the freedom of action or choice in terms of risk taking and risk acceptance is provided to all managers. Risk boundaries are set by our strategy, our Company Statement, Code of Conduct, company values, authority schedules, policies and corporate directives. Our risk appetite differs by objective area and type of risk:
• Strategic: In pursuing our strategic ambitions, we are prepared to take considerable risk related to achieve our growth, innovation and sustainability objectives. Returns on investment in the development of innovative products and sustainable solutions are never certain. Yet considerable funds and efforts are spent on research, development and innovation, even in less certain economic.
• Operational: With respect to operational risks, we continuously strive to minimize these risks. Our risk appetite is very limited. We are executing programs geared towards improving our operational and functional excellence. Our risk appetite is governed by our ambition to strive for top quartile safety performance, top quartile performance in diversity and talent development, top quartile eco-efficiency improvement rates and a top three position in the Chemicals supersector of the Dow Jones Sustainability Index.
• Financial: With respect to financial risks, we have a prudent financing strategy and a strict cash management policy and are committed to maintaining strong investment grade credit ratings. Our financial risk management and risk appetite for several financial risks are explained in more detail in Note 22 in the Financial statements section.
• Compliance: We do not permit our employees to take any compliance risk and have a zero tolerance policy in relation to breaches of our Code of Conduct. See the Corporate governance section for more details.
Annual report Akzo Nobel 2012 page 23
The sentences that are underlined clearly fall under the definition of risk appetite as it can be regarded to describe an attitude towards the different type of risks. However, it is questionable whether the other parts under the heading Risk appetite should be strictly considered to describe an attitude. The following part from the annual report of Mediq of 2012 illustrates that in other annual reports a strict approach is harder to apply:
RISK TOLERANCE
We stimulate the pursuit of new opportunities and accept the associated risks provided they contribute to the attainment of our strategic and operational objectives. The requirement we impose is that associated risks are identified and managed. We apply stringent financial criteria for acquisitions and investments: acquisitions must not only satisfy strategic criteria but also generate a return on average capital employed of at least 15% before tax within a limited number of years. We are prepared to accept the risk of several concurrent acquisition processes, as long as these satisfy our targets in terms of returns, management and other criteria. Our approach to risk is also influenced by a number of internal and external factors, such as our financial results and operating cash flows, our financing options, economic developments and statutory and regulatory requirements.
