The Effect of Integrated Reporting on the Cost of Equity Capital
An empirical study focused on companies operating in the Oil & Gas industry in
North America.
Name: Geert van de Rijdt Student number: 11419458
Thesis supervisor: Dhr. dr. A. Sikalidis Date: 23 June 2017
Word count: 11239
MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam
Statement of Originality
This document is written by student Geert van de Rijdt who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
The purpose of this research is to examine the potential effect of the quality of Integrated Reporting on a company’s cost of equity capital. Prior literature suggests that there is a negative association between voluntary disclosure and a company’s cost of equity capital. Where the disclosure of an integrated report is a form of voluntary disclosure, I predict that this negative relation holds between the quality of <IR> and a company’s cost of equity capital. In addition, I predict that this negative relation would be stronger for small companies compared to large companies. This paper focuses on companies operating in the Oil & Gas industry in the United States and Canada.
In testing the hypotheses, I rely on the research model of Dhaliwal et al. (2011). I find evidence of a negative relation between the quality of <IR> and a company’s cost of equity capital. However, this evidence is not statistically significant. I find evidence for the prediction that the effect would be stronger for small companies compared to large companies, however the relation between quality of <IR> and a company’s cost of equity capital remains not significant for both sub-groups. Potential explanations for not finding significant evidence are the relatively small sample and the use of a particular measure for the estimation of the cost of equity capital.
Table of contents 1 Introduction ... 6 1.1 Background ... 6 1.2 Research question ... 7 1.3 Contribution ... 8 1.3.1 Theoretical contribution ... 8 1.3.2 Practical contribution ... 8
1.4 The <IR> Framework ... 8
1.4.1 The fundamental concepts ... 9
1.4.2 Guiding principles ... 10
1.4.3 Content elements ... 10
1.5 Benefits of <IR> ... 11
1.6 The Oil & Gas industry ... 11
2 Literature review & hypotheses ... 13
2.1 Theories ... 13
2.1.1 Information asymmetry and agency theory ... 13
2.1.2 Signaling theory ... 14
2.1.3 Legitimacy theory in this context ... 15
2.2 The cost of equity capital (COC) ... 15
2.2.1 Definition of COC ... 15
2.2.2 An organization’s COC ... 16
2.3 Voluntary disclosure and COC ... 16
2.3.1 Information asymmetry ... 16
2.3.2 Information asymmetry and stock liquidity ... 16
2.3.3 Information asymmetry and estimation risk ... 17
2.4 Related literature & Hypotheses ... 18
3 Data and variables ... 19
3.1 Data construction ... 19 3.2 Research design ... 19 3.3 Regression variables ... 20 3.3.1 COC ... 20 3.3.2 Quality of <IR> ... 21 3.3.3 Control variables ... 21 4 Empirical results ... 22 4.1 Summary statistics ... 22 4.2 Pearson correlation ... 24 4.3 Regression analyses ... 25
5 Summary and conclusion ... 28
5.1.1 Summary ... 28 5.1.2 Conclusion ... 29 5.1.3 Limitations ... 30 5.1.4 Further research ... 30 References ... 31 Appendix A ... 35 Appendix B ... 36
1 Introduction
1.1 BackgroundEvery listed company is mandated to issue an annual financial performance report. These reports contain a company’s income statement, balance sheet, and notes regarding the financial statements. For stakeholders, such as investors, employers, and customers, it is important that the financial reports present high quality and accurate information about the company’s financial condition, in order to make respectively investment decisions, decisions whether or not to work for the company, and purchase decisions (Eccles & Saltzman, 2011).
Although financial reporting has received institutional legitimacy, it also has its critics. Jonas and Blanchet (2000) argue that financial reports are hard to understand for its users, due to the increasing complexity. Other criticism on financial reports is the time lag in issuing reports, the difficulty of finding the most relevant information, and, most importantly, the backward looking nature of the reports (Jonas and Blanchet, 2000). Since the financial reports do not contain information on non-financial performance that can determine a company’s long-term financial picture, questions are rising about whether a financial report presents a true and fair view of a company’s condition (Eccles & Saltzman, 2011).
These are some of the reasons that, in the last three decades, the bar raises for corporate disclosure standards. In response to this, companies began supplementing their financial reporting with the reporting on non-financial information (KPMG, 2011; Cohen et al., 2012). To report on non-financial information, different disclosure forms were used such as environment reports, stand-alone sustainability reports, corporate social responsibility (CSR) reports or within the annual report (Simnett et al., 2009; Cohen et al., 2012). Although Clarkson et al. (2004) has shown that this information is value relevant, the reports on non-financial information were overwhelming in quantity and up to 200 pages in length (KPMG and FERF, 2011), which results in stakeholders getting an information overload and do not know where to focus on anymore. Another weakness of these separate reports is that they are not integrated with the financial reports (Adams & Simnett, 2011). Moreover, O’ Dwyer et al. (2011) show that stand alone sustainability reports and CSR reports may actually reduce the visibility of corporate social and environmental impacts. Because of these weaknesses the information provided in the supplemental reports if often diminished (KPMG and FERF, 2011).
In order to provide a clear link between the reported non-financial information and the financial information in a manner allowing an assessment of the on-going future performance of the
company, the International Integrated Reporting Council (IIRC) was formed (IIRC, 2011, p. 2). The objective of the IIRC’s separate report was to integrate a companies’ financial and non-financial information. More specifically, Integrated Reporting (further referred to as <IR>) should:
• Improve the quality of information available to providers of financial capital to enable a more efficient and productive allocation of capital;
• Promote a more cohesive and efficient approach to corporate reporting;
• Enhance accountability and stewardship for the broad base of capitals and promote understanding of their interdependencies;
• Support integrated thinking, decision-making and actions that focus on the creation of value over the short, medium and long term (IIRC, 2013).
According to Paul Druckman, the IIRC CEO “Integrated reporting promotes a more cohesive and efficient approach to corporate reporting that draws on different reporting strands. Here, our innovators focus in on the sustainability perspective. They tell their journey, how they are connecting departments across their business, how they are developing their strategies for longer term value creation.” (IIRC, 2013, p. 3)
The potential benefits of <IR> are divided into two categories, the internal benefits and external benefits. Supporters of <IR> claim that <IR> enhances integrated thinking within organisations. Integrated thinking is the process of integrating intuition, reason and imagination in a human mind with a view to developing a holistic continuum of strategy, tactics, action and review and evaluation for assessing and solving a problem in any field (Douglas, 1986) and may lead to better decision-making (IIRC, 2014). The external benefits of <IR> include a better reputation and increased transparency, which could result in a lower cost of equity capital (IIRC, 2011). A PWC report shows that 63% of the investment professionals think that the quality of a company’s report, both, financial and non-financial, could have a direct impact on the cost of equity capital (PWC, 2014).
1.2 Research question
In the previous sections the background of <IR> is provided as an introduction to this thesis. In a study concerning the issues and future of the International Integrated Reporting Framework, Cheng et al. (2014) propose potential future research projects. They argue that “a driving force behind <IR> is the perceived inadequacy of financial information in informing the capital market about an organization’s “true” value creation potential” (Cheng et al., 2014, p. 13). Therefore an interesting research would be to examine whether and to what extent the quality of
<IR> affects the capital market. The relation between the quality of <IR> and cost of equity capital has not been examined before in the same context as this study (Zhou et al., 2015). The elaboration of the context will be provided in following sections. As Cheng et al. (2014) and the PWC (2014) report suggest, the quality of an integrated report could affect the cost of equity capital. This gives companies incentives to release a (better) integrated report. Hence the following research question:
Q: Does the quality of <IR> affect an organization’s cost of equity capital? 1.3 Contribution
This study is expected to make the following contributions:
1.3.1 Theoretical contribution
Where most studies apply the voluntary disclosure theory to the disclosure of financial information (Healy & Palepu, 2001), this study uses the voluntary disclosure theory to examine the effect of voluntary non-financial disclosure on the cost of capital. Non-financial disclosure research is mainly about CSR disclosure (Dhaliwal et al.,2011). This study focuses on the effect <IR> disclosure on the cost of equity capital. Therefore this study is expected to contribute to the existing literature concerning <IR> issues.
1.3.2 Practical contribution
As the number of voluntary adopters of <IR> is increasing rapidly and <IR> is mandatory for JSE listed companies (IIRC, 2012), <IR> might become a standard report and may ultimately replace the stand-alone financial reports. Therefore, research about <IR> from all perspectives can be relevant for all parties involved and affected. For example, this study to the effect of the quality of <IR> on the cost on capital can enhance better decision-making.
1.4 The <IR> Framework
To assess companies’ non-financial reports the IIRC built a framework, the International <IR> Framework. This is a principles-based document containing three main sets of requirements for the preparation of an integrated report, meeting the aims that are mentioned above. The first requirements are called fundamental concepts, the second are guiding principles, and the third are content elements (IIRC, 2012).
1.4.1 The fundamental concepts
The fundamental concepts underpin and reinforce the requirements and guidance in the Framework. They are:
• Value creation for the organization and for others • The capitals
• The value creation process (IIRC, 2013)
In the next paragraphs a short description of each concept is provided.
The value creation by an organization over time manifests itself in increases, decreases or transformations of the capitals caused by the organization’s business activities and outputs. That value can be created for the organization itself or for others, such as stakeholders and the society at large (IIRC, 2013).
In order to be successful an organization depends on various forms of capital. This Framework identifies the following six capitals, although organizations are not required to adopt this categorization:
• Financial capital – The pool of funds that is available to an organization for use in the production of goods or the provision of services and obtained through financing or generated through operations or investments.
• Manufactured capital – Manufactured physical objects that are available to an organization for use in the production of goods or the provision of services.
• Intellectual capital – Organizational, knowledge-based intangibles.
• Human capital – People’s competencies, capabilities and experience, and their motivation to innovate.
• Social and relationship capital – The institutions and the relationships within and between communities, groups of stakeholders and other network, and the ability to share information between them.
• Natural capital – All renewable and non-renewable environmental resources (IIRC, 2013). The value creation process reflects an organization’s business model, which draws on various capitals as inputs and, through its business activities, converts them to outputs. The organization’s activities and its outputs lead to outcomes in terms of effects on the capitals. The capacity of the business model to adapt to changes can affect the organization’s longer-term viability (IIRC, 2013). The figure below shows an organization’s value creation process.
Figure 1: The value creation process (IIRC, 2013, p. 13)
1.4.2 Guiding principles
The following seven guiding principles support the preparation and presentation of an integrated report:
• Strategic focus and future orientation • Connectivity of information
• Stakeholder relationships • Materiality
• Conciseness
• Reliability and completeness • Consistency and comparability
These guiding principles are applied individually and collectively for the purpose of preparing and presenting an integrated report (IIRC, 2013)
1.4.3 Content elements
An integrated report should give answers on the following eight questions that are linked to eight content elements:
• Organizational overview and external environment – What does the company do and what are the circumstances under which it operates?
• Governance – How does the company’s governance structure support its ability to create value in the short, medium and long run?
• Business model – What is the company’s business model and to what extent is it resilient? • Risks and opportunities – What are the specific opportunities and risks that affect the
company’s ability to create value over the short, medium and long term, and how is the company dealing with them?
• Strategy and resource allocation –Where does the company want to go and how does it intend to get there?
• Performance – To what extent has the company achieved its strategic objectives and what are its outcomes in terms of the capitals?
• Outlook – What challenges and uncertainties is the company likely to encounter in pursuing its strategy and what are the potential implications for its business model and future performance?
• Basis of preparation and presentation – How does the organization determine what matters to include in the integrated report and how are such matters quantified or evaluated? (IIRC, 2013)
The IIRC (2013) suggests that the consideration of these requirements will lead to the aims of <IR>.
1.5 Benefits of <IR>
Eccles & Saltzman (2011) identify three different classes of benefits arising from the use of <IR>. The first class is internal benefits, including better internal resource allocation decisions, greater engagement with shareholders and other stakeholder, and lower reputational risk. The external market benefits are the second class of benefits. These include meeting the needs of the mainstream investor, appearing on sustainability indices, and ensuring that data vendors report accurate non-financial information on the company. The last class is managing regulatory risk, including being prepared for a likely wave of global regulation, responding to request from stock exchanges, and having a seat at the table as frameworks and standards are developed (Eccles & Saltzman, 2011).
1.6 The Oil & Gas industry
This paper focuses on companies operating in the oil and gas industry in the United States and Canada. Oil and gas companies were among the first to regularly publish sustainability reports in
addition to the obligated financial reports (Patten, 1991). In addition, the companies have a large potential to have widespread of either very good or very bad reputations (Hughey and Sulkowski, 2012). Since companies tend the have either a very good or very bad reputation, the legitimacy theory could explain why these companies voluntary disclosure information. Companies hope to justify their operations by legitimizing its action via disclosure (Lehman, 1983). This theory will be discussed in detail in the next chapter. For these reasons, the oil and gas industry is considered to be a suitable sample for this study. Furthermore, the association between voluntary disclosure and a company’s cost of equity capital has not been studied before in this context.
2 Literature review & hypotheses
In the first chapter I explained how the rise of <IR> occurred, that the <IR> focuses on creating value in the short, mid and long term and what this paper will focus on. This chapter will include a literature review, where I explain the concepts of <IR> and the cost of equity capital further. I will also provide the reasoning behind the potential relationship between <IR> and the cost of capital followed by the hypotheses of this study.
2.1 Theories
The following paragraphs discuss the major theories that inform this thesis.
2.1.1 Information asymmetry and agency theory
The information or “lemon” problem arises under two conditions; the first condition is that an information asymmetry between two individuals, for example an investor and manager exist. The other condition is conflicting incentives between those two (Balakrishnan, 1993). The information problem can lead to malfunctioning of the capital market (Akerlof, 1970). For example, imagine a market where half of the companies are “bad” and the other half are “good”. Investors make investment decisions based on the information that is available to them, the public information. If investors cannot distinguish the “good” and “bad” companies, all the managers will claim that their company is “good”. Taking this possibility into account, investors will value all companies equal, at an average level. Therefore, if the information problem is not resolved, the market will undervalue some “good” companies and overvalue some “bad” companies. There are a few common solutions to the information problem. One of them is contracting. Optimal contracts between managers and investors mitigate the misevaluation problem, since the contracts provide incentives to the managers to disclosure private information (Kreps, 1990). Regulation is another solution since it could require managers to disclose private information. Due to the information problem, the demand for information intermediaries increases. Information intermediaries, such as rating agencies and financial analysts, engage in private information production to uncover managers’ private information (Healy & Palepu, 2001). Healy & Palepu (2001) distinguish two types of information asymmetry; moral hazard and adverse selection. Moral hazard occurs when, for example, an investor invests in an organization, and as result the manager of the organization takes more risks because the investor bears the cost of those risks (Bergemann & Hege, 1998). Adverse selection occurs when one party is better informed than the other party, and can therefore make better decisions (Healy & Palepu, 2001).
The agency problem, also known as the principal-agent problem, arises because separation of control occurs when a principal hires an agent. The principal (investor) hands the control over the company to the agent (manager). Consequently, once investors have invested their funds, the self-interested manager has an incentive to make decisions that expropriate the investors’ funds.
2.1.2 Signaling theory
Campbell et al. (2001) suggest that the signaling theory can contribute to the debate in this area. In order to explain information asymmetry in the labor markets, Spence (1973) suggested the signaling theory. Since then it has been developed in some other areas, including voluntary information disclosure (Ross, 1977). In the context of corporate reporting, the signaling theory suggests that, because of information asymmetry, companies that believe they are “better” than other companies, signal this to investors in order to attract investment and a more favorable reputation (Verrecchia, 1983).
Companies may signal by voluntarily disclosing information in excess of what is formally required by law and other regulation (Campbell et al., 2001). Thus by issuing an integrated report, companies could be signaling that they are a better investment than other companies. Also, not disclosing something can be a signal in itself. However, because of adverse selection, companies will signal that they are better than average by disclosing information. Therefore, in line with the signaling theory, companies tend to choose disclosing information over not-disclosing information (Campbell et al., 2001). Thus, theoretically, all companies will have an incentive to disclose all positive distinguishing information in order to maximize their own self-interest, so the equilibrium position will be full disclosure. Camfferman (1997) calls the mechanism that is leading to this the “relevation principle”. Watts (1986) argues that this diminishes against non-disclosure and may lead to over-non-disclosure. Over-non-disclosure is where the marginal costs of disclosure exceed the marginal benefits of disclosure. (Campbell et al., 2001).
2.1.2.1 Signaling and the agency theory
Morris (1987) suggests that the signaling theory and the agency theory overlap significantly. Signaling is wide ranging because it can be used to show that the company that signals is better than the others on many different aspects of the annual report (Campbell et al., 2001), for example, environmental or social disclosure, which also may be the case in the disclosure of <IR> information. The agency theory is useful, though the effects of motivation of disclosure are important. Signaling has been argued to be a useful addendum to agency theory (Chow & Wong-Boren, 1987). Camfferman (1997) also suggested that agency theory could be a valuable addition to the signaling theory, particularly in order to cover long-term disclosure issues.
2.1.3 Legitimacy theory in this context
The legitimacy theory suggests that corporate disclosure reacts to environmental factors, such as economic, social, political, and that disclosures legitimize actions (Lehman, 1983). Companies continuously seek to ensure that they operate within the bounds and norms of their respective societies. This requires the company to be responsive, since the bounds and norms change over time (Brown & Deegan, 2012). Guthrie and Paker (1989, p. 344) suggest that “this theory is based upon the notion that business operates in society via a social contract where it agrees to perform various socially desired actions in return for approval of its objectives, other rewards and its ultimate survival”. Companies therefore need to disclose sufficient environmental information for society to assess whether it is a good corporate citizen. Ultimately, the company hopes to justify its continued existence by legitimizing its actions via disclosure (Lehman, 1983). Therefore, voluntary disclosure is seen as a mechanism for companies to legitimize their operations (Brown & Deegan, 2012).
In this context, the legitimacy theory can be used in explaining why companies choose to disclose <IR> information in two ways. The first reason is that various events, such as frauds, financial scandals and corruption in the past years have damaged the reputation of many companies and questioned the credibility of corporate reporting. Furthermore, as discussed in the prior chapter, the Oil & Gas industry in particular is suffering from having a widespread of either very good or very bad reputations. Various incidents in the Oil & Gas industry have led to huge environmental damages. Therefore these companies are trying to, to some extend, recover their reputation by disclosing information on social and environmental issues. This could explain why companies, in particular companies in the Oil & Gas industry, choose to disclose <IR> information and so to legitimize their operations.
2.2 The cost of equity capital (COC)
2.2.1 Definition of COC
COC is the minimum rate of return that investors require in return for the provision of capital to organizations. A company’s cost of equity capital is the sum of the risk-free return and a risk premium (Botosan, 2006). Clinch (2013) describes COC as the expected rate of return on a organization’s stock, which is the risk-adjusted discount rate that investors apply to the organization’s expected cash flow when an organization valuate itself.
COC is a forward-looking concept that cannot be directly observed in the market place, therefore it is also referred to as the implied cost of equity capital (ICC) (Botosan, 2006). In most
of the empirical studies the ICC is estimated based on market prices, accounting numbers and forecasts of earnings and dividends (Easton, 2004).
2.2.2 An organization’s COC
According to prior research the role of COC is fundamental because it affects corporate decisions like investment and acquisition decisions (Easley & O’Hare, 2004). COC is also of interest to organizations in composing and adjusting the capital structure of the company (Easley & O’Hara, 2004). Based on the above it is arguably that organizations aim for a lower COC in order to improve their performance. Furthermore, lowering COC will also attract more investors and lead to greater investment, and will so benefit the entire economy (Shrestha & Mishra, 2012).
2.3 Voluntary disclosure and COC
Two main channels explain the relation between voluntary disclosure and cost of equity capital. The first channel is the information asymmetry in combination with stock liquidity; the second channel explains the relation with the concept of estimation risk (Botosan, 2006; Petrova et al., 2012). The basis of both channels is that more disclosure decreases the information asymmetry (Petrova et al., 2012). In the next paragraphs the concept of information asymmetry will be explained, as well as both channels that explain the relation between voluntary disclosure and the cost of equity capital.
2.3.1 Information asymmetry
As mentioned before, information asymmetry deals with the study of decisions in transactions where one party has more or better information than the order (Balakrishnan, 1993). Disclosure can diminish the information asymmetry (Healy & Palepu, 2001). Healy & Palepu (2001) distinguish two types of information asymmetry; moral hazard and adverse selection. Moral hazard occurs when, for example, an investor invests in an organization, and as result the manager of the organization takes more risks because the investor bears the cost of those risks (Bergemann & Hege, 1998). Adverse selection occurs when one party is better informed than the other party, and can therefore make better decisions (Healy & Palepu, 2001). This type of information asymmetry is, in the context of this paper where we look at the relation between <IR> and the cost of equity capital, the most important type.
2.3.2 Information asymmetry and stock liquidity
The first channel explains the relation between voluntary disclosure and the cost of capital with the concept of information asymmetry and the stock liquidity. Stock liquidity is the ease of
converting stock into cash, and vice versa (Amihud & Mendelson, 2000). Costs related to purchasing or selling stock are the drivers of liquidity. Costs of liquidity consist of adverse selection costs, opportunity costs, and direct costs (Amihud & Mendelson, 2000). More disclosure reduces the information asymmetry, which results in lower adverse selection costs, hence, in a higher stock liquidity (Amihud & Mendelson, 2000; Diamond & Verrechia, 1991). Additionally, the higher the stock liquidity, the lower the cost of equity capital (Amihud & Mendelson, 2000)
2.3.3 Information asymmetry and estimation risk
The link between accounting information and the cost of capital of companies is one of the most important topics in accounting (Lambert et al., 2006). More disclosure suggests less uncertainty, leading to better disclosure results, and subsequently to a lower cost of capital (Foster, 2003). Hughes et al. (2007) argue that greater information asymmetry leads to a higher uncertainty and higher factor risk premiums, and effectively in higher cost of capital. Vice versa, less information asymmetry leads to less uncertainty and lower cost of capital (Hughes et al., 2007). Moreover, Botosan (2006) argues that the estimation risk is an undiversified risk, which in case of poor information, results in a higher estimation risk, hence, higher cost of capital. Moreover, investors prefer stock with a low estimation risk, resulting in a higher demand for this type of stock. Disclosure lowers the estimation risk. For stock, of organizations that disclose more, there is more demand and higher prices, so the cost of equity for these organizations will be lower (Botosan, 2006).
2.3.4 <IR> as voluntary non-financial disclosure
Disclosure is the revelation of an item of generally unknown information on a financial statement or in the accompanying notes. <IR> aims to improve the quality of information, and to promote a more cohesive and efficient approach to corporate reporting. It integrates the financial and non-financial information (IIRC, 2013). Where disclosure is the revelation of an item of information, <IR> is a form of voluntary disclosure. Most prior research on the relation between disclosure and the cost of equity capital focuses on financial disclosure (Healy and Palepu, 2001; Armstrong et al., 2011). Margolis and Walsh (2001) argue that the mechanisms of the relation between disclosure and the cost of equity capital apply on both, financial and nonfinancial disclosure, as long as the disclosed information is value-relevant. Eccles and Saltzman (2011) find that <IR> disclosure is value-relevant, and therefore these mechanisms are expected to hold for <IR> disclosure.
2.4 Related literature & Hypotheses
Petrova et al. (2012) have examined the relation between voluntary disclosure and the cost of equity capital. They find a negative relation between voluntary disclosure and the cost of equity capital. More specifically, when the measure for disclosure increases with one unit, the cost of equity capital decreases with 0,1% (Petrova et al., 2012). Botosan & Plumlee (2002) have examined the relation between the release of annual, biannual, and quarterly reports and the cost of equity capital. They also find a negative relation (Botosan & Plumlee, 2002). Gietzmann & Ireland (2005) also find a negative relation, however this negative relation only holds for organizations with an aggressive accounting policy (Gietzmann & Ireland, 2005).
We have seen that disclosure can affect the cost of equity capital via two channels; information asymmetry and stock liquidity, and information asymmetry and estimation risk. More literature on the relation between disclosure and the cost of capital has been reviewed. This literature shows a negative relation between disclosure and the cost of equity capital (Botosan, 2006; Amihud & Mendelson, 2000; Diamond & Verrechia, 1991; Foster, 2003; Hughes et al., 2001; Petrova et al., 2012; Botosan & Plumlee, 2002; Gietzmann & Ireland, 2005). Since <IR> is a form of disclosure, I expect a negative relation between <IR> and the cost of equity capital. Therefore the hypotheses of this research is:
H1: Higher quality of <IR> lowers a company’s cost of equity capital.
The mechanism that affects the link between the quality of <IR> and a company’s cost of equity capital the most is information asymmetry. Armstrong et al. (2010) argue that companies with less information asymmetry tend to have a higher information environment quality, and vice versa. Hence why it is expected that the effect of the quality of <IR> on the cost of equity capital would be stronger for companies operating in a lower information quality environment. One of the most widely used proxies for the quality of information environment is the company size. Information asymmetry is higher in small companies compared to large companies (Chae, 2005; Drobetz et al., 2010). Therefore small companies have a lower information environment quality and thus the effect of the quality of <IR> on the cost of equity capital is expected to be stronger for small companies. The second hypothesis is as follows:
H2: The effect of higher quality of <IR> on a company’s cost of equity capital would be stronger for small companies compared to large companies.
3 Data and variables
In this chapter this study’s data construction will be provided, as well as a detailed description of the research design and each variable.
3.1 Data construction
To examine the association between the quality of <IR> and COC, this paper begins by merging four databases: Thomson Reuters ASSET4 (ASSET4), an environmental, social and governance database, Thompson Institutional Brokers Earnings Services (I/B/E/S), a database that provides analyst forecast data, Center for Research in Security Prices (CRSP), a stock and index database, and Compustat, which provides financial, statistical and market information. In this paper the focus is on the Oil & Gas industry. Using Datastream I am able to collect data on the quality of <IR> of 102 companies in the Oil & Gas industry in North America, including leading companies as Exxon, Paramount Resources and Apache Corporation. The list of companies included in the sample can be found in Appendix A.
In determining COC I follow Dhaliwal et al. (2006) and estimate the cost of equity in June of each year. To do so, I extract from the I/B/E/S summary file forecast data recorded in June for all organizations included in our sample that have positive 1- and 2-year-ahead consensus earnings forecasts and a positive long-term growth forecast. I then estimate COC using the Easton-model, which is discussed in the next paragraph.
Finally, I collect all necessary information for the control variables company size, beta, leverage, book-to-market ratio and the disclosure of a stand-alone CSR report. Later in this chapter a detailed description of these control variables will be provided. Due to the merge of four different databases, the observations for 57 companies included missing values and, therefore, are not included in the analysis. Furthermore, for every company one observation has been lost due to the creation of change variables. This procedure yields a final sample of 180 observations representing 45 unique companies. The companies included in the final sample can be found in Appendix A.
3.2 Research design
This paper’s hypothesis predicts that a higher quality of <IR> leads to a lower COC. In testing this hypothesis I follow Dhaliwal et al. (2011) by estimating the following regression model:
∆%𝐶𝑂𝐶!,!!!
= 𝛽!+ 𝛽!∆𝐼𝑅𝑄!,!+ 𝛽!∆𝑆𝐼𝑍𝐸!,!+ 𝛽!∆𝐵𝐸𝑇𝐴!,!+ 𝛽!∆𝐿𝐸𝑉!,!+ 𝛽!∆𝐵𝑀!,! + 𝛽!𝐶𝑆𝑅!,! + 𝜀!,!
Consistent with Dhaliwal et al. (2011), a lead-lag approach is used where the dependent variable is measured as of t+1 rather than on a simultaneous basis with the independent variables. To control for unobserved company characteristics that might be associated with the dependent variable, all variables are in change form. A negative coefficient on IRQ would support the hypothesis. In the next section a description of the variables will be provided.
3.3 Regression variables
3.3.1 COC
There are different models available to estimate COC, and a debate still exist about which model estimates the COC the most accurate (Easton and Mohanah, 2005; Botosan and Plumlee, 2005). According to Fama & French (1997) the single-factor model and the Fama and French (1993) three-factor model provide poor proxies for the cost of equity capital. Hail & Leuz (2009), and Chen et al. (2009) argue that the implied cost of equity capital approach is useful in this case because it makes an explicit attempt to isolate cost of capital effects from growth and cash flow effects. I use the Easton’s (2004) PEG ratio method to estimate the organization’s cost of equity capital. This is one of the most widespread and well-known estimates of the implied cost of equity capital (Botosan and Plumlee, 2005; Botosan et al., 2011). The estimation model is as follows: 𝐶𝑂𝐶!,! = (𝐸𝑃𝑆! !!! − 𝐸𝑃𝑆!(!!!) 𝑝!,! where: 𝐶𝑂𝐶!,! = 𝑡ℎ𝑒 𝑖𝑚𝑝𝑙𝑖𝑒𝑑 𝑐𝑜𝑠𝑡 𝑜𝑓 𝑒𝑞𝑢𝑖𝑡𝑦 𝑐𝑎𝑝𝑖𝑡𝑎𝑙 𝐸𝑃𝑆! !!! = 𝑡ℎ𝑒 𝑜𝑟𝑔𝑎𝑛𝑖𝑧𝑎𝑡𝑖𝑜𝑛!𝑠 𝑡𝑤𝑜 − 𝑦𝑒𝑎𝑟 𝑎ℎ𝑒𝑎𝑑 𝑚𝑒𝑎𝑛 𝑓𝑜𝑟𝑒𝑐𝑎𝑠𝑡 𝑜𝑓 𝑒𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝐸𝑃𝑆!(!!!)= 𝑡ℎ𝑒 𝑜𝑟𝑔𝑎𝑛𝑖𝑧𝑎𝑡𝑖𝑜𝑛!𝑠 𝑜𝑛𝑒 − 𝑦𝑒𝑎𝑟 𝑎ℎ𝑒𝑎𝑑 𝑚𝑒𝑎𝑛 𝑓𝑜𝑟𝑒𝑐𝑎𝑠𝑡 𝑜𝑓 𝑒𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝑝!,! = 𝑡ℎ𝑒 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑝𝑟𝑖𝑐𝑒
3.3.2 Quality of <IR>
To operationalize the variable quality of <IR>, this paper follows Sarafeim (2015) and employs the Integrated Rating score from ASSET4 as a measure. This score measures the company’s overall performance by including the individual sub-scores of the four main pillars: economic, environmental, social and governance performance. Since this proxy incorporates data items that are directly related to <IR>, this is considered to be a suitable measure for the quality of <IR> (Sarafeim, 2015). Based on the scoring methodology of ASSET4, scores for each measure range from 0 to 100, with 100 being the maximum possible score for the disclosure of highest quality. A negative coefficient is expected.
3.3.3 Control variables
In specifying controls shown to affect the cost of equity capital I follow prior research (Hail and Leuz, 2006; Dhaliwal et al., 2006). These control variables include company size, beta, leverage, book-to-market ratio and the disclosure of a stand-alone CSR report. Company size (SIZE) is calculated by the natural logarithm of the organization’s total assets at the end of each fiscal year. A negative coefficient is expected for company size. Beta (BETA) is estimated by using the market model and included to control for systematic risk. A positive coefficient is expected. Leverage (LEV) is estimated by total debt over total assets. The cost of equity capital is expected to increase as the level of leverage also increases, thus a positive coefficient is expected. The book-to-market ratio (BM) is the book value of the company divided by the market value of the company. For this variable a positive coefficient is expected. Finally, the disclosure of a stand-alone corporate social responsibility report (CSR) is measured as a dummy variable, coded 1 if the company issues a stand-alone CSR report and 0 if otherwise. A negative coefficient is expected (Dhaliwal et al., 2006).
All data necessary to calculate the control variables is obtained through DataStream and Wharton Research Data Services.
4 Empirical results
4.1 Summary statisticsTable 1 provides the summary statistics for the sample as well as for the sub-samples. The mean COC for companies in the Oil & Gas industry in North America is 0,1197. This is consistent with Easton and Monahan (2005), as they find that the COC using the PEG model is usually around 0,11 for US listed companies. The independent variable of main interest, the quality of <IR>, has an average of 59. At the same time, the standard deviation is 29, which suggests significant variation in the practice of <IR>. The lowest and highest rates on the quality of <IR> are 5 and 99, respectively.
Table 1
Summarize Statistics Full Sample
Variable Obs Mean Median Std. Dev. Min Max
COC 273 0,1197 0,1068 0,0674 0,0168 0,5889 IRQ 273 59,0761 59,71 28,5328 5,21 98,6 SIZE 273 9,6416 9,6107 1,3236 5,8345 12,7565 LEV 273 0,2243 0,2215 0,1085 0,002 0,6202 BETA 273 1,5002 1,5087 0,4865 0,4523 3,9545 BM 273 0,7029 0,5851 0,5471 0,1205 4,9955 CSR 273 0,4945 0 0,5009 0 1 Table 1.1
Small companies Large companies
Variable Obs Mean Obs Mean t-stat p-value
COC 136 0,1295249 137 0,10986 2,4316 0,0157 IRQ 136 37,32199 137 80,6715 -19,3123*** 0,0000 SIZE 136 8,578957 137 10,6967 -22,0638*** 0,0000 LEV 136 0,2597707 137 0,1890 5,6888*** 0,0000 BETA 136 1,629155 137 1,3721 4,5185** 0,0014 BM 136 0,7203843 137 0,6856 0,5253 0,5998 CSR 136 0,1911765 137 0,7956 -12,4914*** 0,0000 Significance level *** p<0,001, **p<0,01, *p<0,05
In table 1.1 the full sample is divided into the two sub-samples; small companies and large companies. The mean comparison of the variables between the two sub-samples is done by using independent samples T-test. The results indicate that there is a statistically significant difference between small and large companies for quality of <IR>, company size, leverage, beta and the issuance of a stand-alone CSR report. Larger companies issue higher quality reports and are more likely to issue a stand-alone CSR report, which is consistent with prior studies (Campbell et al., 2001). The sub-sample is a dummy of company size, and therefore a significant difference was expected, if not demanded. Furthermore larger companies have a lower leverage and higher beta.
As explained earlier in the methodology section, the change forms of all variables are used in the regression analysis to address the endogeneity concern. The change variables are used in the regression analysis and have therefore been winsorized. With winsorization, we set the extreme small and large observations equal to the values of less extreme small and large observations (Veenman, 2013). In this thesis, the data has been winsorized at the 1st and 99th percentiles of
their distributions. Because of the use of change forms, the first observation of every company is no longer included. In addition, there are 63 observations, which do not have previous data to calculate the changes in some variables. The sample size thus reduces from 273 to 180 observations for the regressions to be run. In table 2 the summary statistics of the change variables are presented. Stata computes the descriptive statistics on the level of each individual variable, therefore the number of observations differs. However, when running the regression analysis, only the observations without missing data (180) will be included.
Table 2
Summarize Statistics (change variables) Full Sample
Variable Winsorized at Obs Mean Median Std. Dev. Min Max
ΔCOC 99% 191 0,0088 -0,00062 0,0809 -0,1942 0,4780 ΔIRQ 99% 243 2,4159 1,1300 11,8930 -25,0400 36,5500 ΔSIZE 99% 243 0,0783 0,0770 0,2027 -0,5093 0,8368 ΔLEV 99% 263 0,0125 0,0076 0,0531 -0,1165 0,1956 ΔBETA 99% 261 0,0394 0,0279 0,4971 -1,2746 1,3519 ΔBM 99% 264 0,0835 0,0400 0,3470 -0,8980 1,4336
Table 2.1
Small companies Large companies
Variable Obs Mean Obs Mean t-stat p-value
ΔCOC 92 0,0110604 99 0,00663 0,3775 0,6469 ΔIRQ 112 3,320179 131 1,6427 1,0964 0,2740 ΔSIZE 112 0,1133572 131 0,0483 2,5203* 0,0124 ΔLEV 131 0,0167196 132 0,0084 1,2776 0,2025 ΔBETA 130 0,058735 131 0,0202 0,6252 0,5324 ΔBM 132 0,106048 132 0,0610 1,0541 0,2928 Significance level *** p<0,001, **p<0,01, *p<0,05
The descriptive statistics (table 2) indicate that the yearly average increase of the cost of equity capital is 0,0088 for the total sample, 0,01106 for small companies and 0,00663 for large companies. Furthermore, the mean change of the independent variables IRQ, SIZE, LEV, BETA and BM are 2,4159, 0,0783, 0,0125, 0,0394 and 0,0835, respectively. The differences of means in change variables between the two sub-samples are examined using the T-test and shown in table 2.1. There are no significant differences between the two groups, except for the variable company size (SIZE). The change in company size is higher for small companies than for large companies.
4.2 Pearson correlation
Table 3 provides Pearson correlations for all interval variables used in the study. The Pearson correlation matrix tests for correlation between variables and for possible multicollinearity issues. The quality of <IR> (IRQ) has a negative correlation with the dependent variable cost of equity capital (COC), which was expected. However the correlation is not significant. Beta is significantly correlated with cost of capital, and book-to-market ratio is significantly correlated with cost of equity capital and beta. Overall, the Pearson correlations are consistent with expectations and previous studies (Fama & French, 1992; Gebhardt et al., 2001; Dhaliwal et al., 2011) and provide some preliminary support for a negative association between the quality of <IR> and the cost of capital as predicted in this study. As shown in table 3, none of the unexpected correlations are sufficiently high as to raise collinearity concerns. Therefore, I conclude that there is no violation of the OLS assumptions.
Table 3
Pearson Correlation Analyses
ΔCOC ΔIRQ ΔSIZE ΔLEV ΔBETA ΔBM
ΔCOC 1,0000 ΔIRQ -0,0030 1,0000 ΔSIZE -0,0226 0,1795 1,0000 ΔLEV -0,0848 -0,1870 0,0819 1,0000 ΔBETA 0,2096* -0,0148 -0,0190 -0,0489 1,0000 ΔBM 0,308** 0,0708 0,0901 -0,0484 0,3796** 1,0000 Significance level *** p<0,001, **p<0,01, *p<0,05 4.3 Regression analyses
The final number of observations for running the OLS regression is 180. To resolve the problem of heteroskedasticity and unwanted correlation, Veenman (2013) suggests to make use of “clustered” standard errors. Therefore, in running the OLS regression, I have used the “clustered” standard errors.
Table 4 provides the results of the regression analyses for the full sample. I find that the model built explains approximately 18% of the variance in the response variable i.e. the cost of equity capital (adj. R-squared = 0,1836). This paper hypothesizes that the relationship between the cost of equity capital (ΔCOC) and the disclosure quality of <IR> (ΔIRQ) will be negative. As expected, ΔIRQ has a negative coefficient, which means that the increase in the quality of <IR> results in a lower cost of equity capital. However, this association is not statistically significant. This might have to do with the reduction of sample size from 273 to 180, because a relatively small sample size works against finding statistically significant results for hypothesis testing. As shown in the results, the coefficients for the other variables have the predicted signs, with two exceptions: leverage and the issue of a stand-alone CSR report. The coefficient for leverage is negative, while it was expected to be positive. A positive sign is found for the disclosure of a stand-alone CSR report, while a negative sign was expected. However, for both, leverage and disclosure of a stand-alone CSR report, the association is not significant, which means that is does not materially affect the cost of equity capital. The other variables have the expected coefficient signs. In addition, as shown in the table, a significant result is found for book-to-market ratio (BM). This implicates that a higher book-to-book-to-market ratio results in higher cost of equity capital, which was predicted.
Table 4
Regression Analysis: Testing H1
ΔCOC Expected sign Coefficient
Clustered SE t-stat. p-value ΔIRQ (-) -0,00548 0,004756 -1,15 0,251 ΔSIZE (-) -0,0269476 0,0450334 -0,82 0,411 ΔLEV (+) -0,0613735 0,0990993 -0,50 0,619 ΔBETA (+) 0,0113555 0,0131473 0,88 0,382 ΔBM (+) 0,1086607*** 0,0330788 5,65 0,000 CSR (-) 9,81E-06 0,0126867 0,00 0,999 Constant 0,0033716 0,0111413 0,36 0,722 Observations 180 R-squared 0,211 adj. R-squared 0,1836 F-statistic 7,71 p(F) 0,0000 Significance level *** p<0,001, **p<0,01, *p<0,05
H2 tests the effect of the quality of <IR> on the cost of equity capital, depending on the information environment. As mentioned in the hypotheses development, small companies are likely to have higher information asymmetry, and are thus expected to have a lower information environment quality. The effect of the quality of <IR> on the cost of equity capital is thus expected to be higher for small companies than for large companies. The small companies are all companies with total assets below median, where large companies are all companies with total assets greater than median. Table 4.1 provides the results of the OLS regression divided into samples. The total of observations in sample “small companies” yields 83, and in sub-sample “large companies” 97. The sub groups explain approximately 17% and 20%, respectively. The hypothesis predicts a stronger effect of the quality of <IR> on the cost of equity capital (i.e. a more significant coefficient) for small companies, which I also find. However, I only find weak evidence since the coefficient for both, small companies and large companies, remain not significant. For small companies, the coefficients for all variables have the predicted sign, except for disclosure of stand-alone CSR report. The coefficient of the disclosure of a stand-alone CSR report is positive, where it is expected to be negative. For large companies, the coefficients for all variables have the predicted sign, except for leverage. The coefficient of leverage is negative, where it is expected to be positive.
However, for both, leverage (large companies) and disclosure of a stand-alone CSR report (small companies), the association is not statistically significant, which means that is does not materially affect the cost of equity capital. The coefficient for book-to-market ratio (BM) remains significant for both sub groups, which implicates that a higher book-to-market ratio results in higher cost of equity capital.
Overall, the results in Table 4 provide support for H1 and show that a higher quality of <IR> results in lower cost of equity capital. However, the coefficient is not statistically significant and thus only provides weak evidence for this association. H2 predicts that the association would be more presence at small companies than at large companies. The results in Table 4.1 support this prediction. However, the coefficients remain not significant for both sub groups, and therefore only provide weak evidence for the association.
Table 4.1
Regression Analysis: Testing H2
Small companies Large companies
ΔCOC Coefficient SE Coefficient SE
ΔIRQ -0,008217 0,006972 -0,003663 0,0070060 ΔSIZE -0,0025764 0,0421286 -0,1003417 0,0581620 ΔLEV 0,1068679 0,1691406 -0,0416161 0,1903707 ΔBETA 0,0036428 0,0186733 0,0183085 0,0193143 ΔBM 0,1076512*** 0,0250246 0,1308368*** 0,0321257 CSR 0,0076283 0,0215051 -0,0120322 0,0182446 Constant -0,0013578 0,0120466 0,0163148 0,0171248 Observations 83 97 R-squared 0,1994 0,2381 adj. R-squared 0,1734 0,2046 F-statistic 3,99 5,22 p(F) 0,0016 0,0001 Significance level *** p<0,001, **p<0,01, *p<0,05
5 Summary and conclusion
In this section, I will provide a summary of this research followed by the conclusion, which I was able to draw from the results. Finally, I will discuss the limitations of this research along with recommendations for further research.
5.1.1 Summary
For all listed companies, the issue of a financial report at the end of every financial year is obligated. Financial reporting has received institutional legitimacy, however, it also has its critics. Financial reporting would be too complex, be backward looking, have a time lag in issuing reports, and make it difficult to find the most relevant information. Therefore the bar raises for corporate disclosure standards. As response to this, companies began additionally issuing reports with non-financial information such as environmental reports, stand-alone sustainability reports, CSR reports and, most recently, integrated reports. <IR> would promote a more cohesive and efficient approach to corporate reporting focusing on the sustainability perspective. This report reflects a company’s journey, how they connect departments and how they develop their strategy for longer-term value creation. Potential benefits of <IR> are better decision-making, a better reputation, increased transparency, greater engagement with all stakeholders, and a lower cost of equity capital. This study is expected to contribute to the existing literature concerning <IR> issues and can enhance, for example, better decision-making by examining the effects of the quality of <IR> on a company’s cost of equity capital.
Various studies have examined the relation between (voluntary) disclosure and the cost of equity capital. We have seen that disclosure can affect the cost of equity capital via two channels; information asymmetry and stock liquidity, and information asymmetry and estimation risk. Existing literature finds a negative relation between (voluntary) disclosure and the cost of equity capital (Botosan, 2006; Amihud & Mendelson, 2000; Diamond & Verrechia, 1991; Foster, 2003; Hughes et al., 2001; Petrova et al., 2012; Botosan & Plumlee, 2002; Gietzmann & Ireland, 2005). Since <IR> is a form of disclosure, I expect a negative relation between <IR> and the cost of equity capital. This study focuses on companies operating in the oil and gas industry in the United States and Canada. The first hypothesis is as follows:
H1: Higher quality of <IR> lowers a company’s cost of equity capital.
Small companies have a lower information environment quality (Chae, 2005; Drobetz, 2010), and thus the effect of the quality of <IR> on the cost of equity capital is expected to be stronger for small companies. The second hypothesis is as follows:
H2: The effect of higher quality of <IR> on a company’s cost of equity capital would be stronger for small companies compared to large companies.
To estimate a company’s cost of equity capital, I have used the Easton’s (2004) PEG ratio method. This is one of the most widespread and well-known estimates of the implied cost of equity capital (Botosan and Plumlee, 2005; Botosan, Plumlee and Wen, 2011). To operationalize the variable quality of <IR>, this paper follows Sarafeim (2015) and employs the Integrated Rating score from ASSET4 as a measure. Based on the scoring methodology of ASSET4, scores for each measure range from 0 to 100, with 100 being the maximum possible score for the disclosure of highest quality. A negative coefficient is expected. This paper follows Dhaliwal et al. (2006) in the research model, which includes the control variables company size, beta, leverage, book-to-market ratio and the disclosure of a stand-alone CSR report.
In order to prepare the data for running the analyses, I have created change variables and threated the extreme values by winsorization of data at the 1st and 99th percentiles of their
distributions. The mean comparison of the variables between the two sub-samples has been done by using independent samples T-test. To test for correlation between variables and for possible multicollinearity issues, the Pearson correlation matrix has been used. Finally, in order to solve the problem of hetoroskedasticity and unwanted correlation, I have used “clustered” standard errors in running the regression analyses.
5.1.2 Conclusion
This research does provide some evidence of a negative relation between the quality of <IR> and a company’s cost of equity capital. However, the relation I find is not statistically significant and therefore only provides a weak evidence for an existing relationship between the two variables. The lack of evidence for a statistically significant relation between the quality of <IR> and a company’s cost of equity capital might be due to the drop of observations. The coefficients for the other variables have the predicted signs, with two exceptions: leverage and the issue of a stand-alone CSR report. Additionally, I have found a statistically significant positive relation between the book-to-market ratio and a company’s cost of equity capital, which is in line with prior research. The second hypothesis predicts that the effect of the quality of <IR> on a company’s cost of equity capital is stronger for small companies compared to large companies. I find evidence for this hypothesis, however the relation between quality of <IR> and a company’s cost of equity capital remains insignificant for both sub-groups. For small companies, the coefficients for all variables have the predicted sign, except for disclosure of stand-alone CSR report. For large companies, the coefficients for all variables have the predicted sign, except for
leverage. As for the full sample, I also find a statistically significant relationship between the book-to-market ratio and a company’s cost of equity capital for the two sub-groups.
5.1.3 Limitations
In this thesis, I have found negative relation between the quality of <IR> and a company’s cost of equity capital. However, the findings are not statistically significant and thus too weak to confirm this relation. There exist some limitations in this study, which may cause that the relation is not significant. Firstly, the size of the final sample is relatively small. Due to the focus on Oil & Gas companies, the merging of four databases, the creation of change variables and the lead-lag approach the number of observation ultimately dropped to 180. Secondly, the measures chosen in this study also drive the results. To estimate the cost of equity capital the Easton (2004) PEG ratio method has been used. The results might have differed if a different method was used. In addition, the estimation of the quality of <IR> might not be direct or completely objective, due to the quantitative scores being based on a partly qualitative report.
5.1.4 Further research
As the number of voluntary adopters of <IR> is increasing rapidly, studies examining the potential effects of adopting <IR> become more interesting and relevant for all parties involved and affected. One of the limitations of this thesis is the small sample size. Therefore I would suggest further research to the effect of the quality of <IR> on a company’s cost of equity capital for all listed companies over the world. Additionally, sub-samples could be made to examine the potential existence of a stronger effect for companies from a certain country or operating in a certain industry. Furthermore, different measures for the estimation of the cost of equity capital could be used, the rule out the possibility that the choice of one particular measure drives the results.
References
Adams, S., & Simnett, R. (2011). Integrated Reporting: An opportunity for Australia's not‐for‐ profit sector. Australian Accounting Review, 21(3), 292-301.
Akerlof, G. A. (1970). The market for" lemons": Quality uncertainty and the market mechanism. The quarterly journal of economics, 488-500.
Amihud, Y., & Mendelson, H. (2000). The liquidity route to a lower cost of capital. Journal of Applied Corporate Finance, 12(4), 8-25.
Armstrong, C. S., Core, J. E., Taylor, D. J., & Verrecchia, R. E. (2011). When does information asymmetry affect the cost of capital?. Journal of Accounting Research, 49(1), 1-40.
Armstrong, C. S., Guay, W. R., & Weber, J. P. (2010). The role of information and financial reporting in corporate governance and debt contracting. Journal of Accounting and Economics, 50(2), 179-234.
Balakrishnan, S., & Koza, M. P. (1993). Information asymmetry, adverse selection and joint-ventures: Theory and evidence. Journal of economic behavior & organization, 20(1), 99-117. Bergemann, D., & Hege, U. (1998). Venture capital financing, moral hazard, and learning. Journal
of Banking & Finance, 22(6), 703-735.
Botosan, C. A. (2006). Disclosure and the cost of capital: what do we know?. Accounting and business research, 36(sup1), 31-40.
Botosan, C. A., & Plumlee, M. A. (2002). A re‐examination of disclosure level and the expected cost of equity capital. Journal of accounting research, 40(1), 21-40.
Botosan, C. A., Plumlee, M. A., & Wen, H. (2011). The relation between expected returns, realized returns, and firm risk characteristics. Contemporary Accounting Research, 28(4), 1085-1122.
Brown, N., & Deegan, C. (1998). The public disclosure of environmental performance information—a dual test of media agenda setting theory and legitimacy theory. Accounting and business research, 29(1), 21-41.
Camfferman, K. (1997). Voluntary annual report disclosure by listed Dutch companies, 1945-1983. Taylor & Francis.
Campbell, D., Shrives, P., & Bohmbach‐Saager, H. (2001). Voluntary disclosure of mission statements in corporate annual reports: signaling what and to whom?. Business and society review, 106(1), 65-87.
Chae, J. (2005). Trading volume, information asymmetry, and timing information. The journal of finance, 60(1), 413-442.
Chen, K. C., Chen, Z., & Wei, K. J. (2009). Legal protection of investors, corporate governance, and the cost of equity capital. Journal of Corporate Finance, 15(3), 273-289.
Cheng, M., Green, W., Conradie, P., Konishi, N., & Romi, A. (2014). The international integrated reporting framework: key issues and future research opportunities. Journal of International Financial Management & Accounting, 25(1), 90-119.
Chow, C. W., & Wong-Boren, A. (1987). Voluntary financial disclosure by Mexican corporations. Accounting review, 533-541.
Clarkson, P. M., Li, Y., & Richardson, G. D. (2004). The market valuation of environmental capital expenditures by pulp and paper companies. The accounting review, 79(2), 329-353. Clinch, G. (2013). Disclosure quality, diversification and the cost of capital. Australian Journal of
Management, 38(3), 475-489.
Cohen, J. R., Holder-Webb, L. L., Nath, L., & Wood, D. (2012). Corporate reporting of nonfinancial leading indicators of economic performance and sustainability. Accounting Horizons, 26(1), 65-90.
Dhaliwal, D. S., Li, O. Z., Tsang, A., & Yang, Y. G. (2011). Voluntary nonfinancial disclosure and the cost of equity capital: The initiation of corporate social responsibility reporting. The accounting review, 86(1), 59-100.
Dhaliwal, D., Heitzman, S., & ZHEN LI, O. L. I. V. E. R. (2006). Taxes, leverage, and the cost of equity capital. Journal of Accounting Research, 44(4), 691-723.
Diamond, D. W., & Verrecchia, R. E. (1991). Disclosure, liquidity, and the cost of capital. The journal of Finance, 46(4), 1325-1359.
Douglas, M. (1986). How institutions think. Syracuse University Press.
Drobetz, W., Grüninger, M. C., & Hirschvogl, S. (2010). Information asymmetry and the value of cash. Journal of Banking & Finance, 34(9), 2168-2184.
Easley, D., & O'hara, M. (2004). Information and the cost of capital. The journal of finance, 59(4), 1553-1583.
Easton, P. D. (2004). PE ratios, PEG ratios, and estimating the implied expected rate of return on equity capital. The accounting review, 79(1), 73-95.
Easton, P. D., & Monahan, S. J. (2005). An evaluation of accounting-based measures of expected returns. The Accounting Review, 80(2), 501-538.
Eccles, R. G., & Saltzman, D. (2011). Achieving sustainability through integrated reporting. Stanf Soc Innov Rev Summer, 59.
Fama, E. F., & French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of financial economics, 33(1), 3-56.
Fama, E. F., & French, K. R. (1997). Industry costs of equity. Journal of financial economics, 43(2), 153-193.
Foster, N. (2003). The FASB and the capital markets. The FASB Report (June 2003).
Gebhardt, W. R., Lee, C., & Swaminathan, B. (2001). Toward an implied cost of capital. Journal of accounting research, 39(1), 135-176.
