IASB Standard Setting, Enforcement of Financial Reporting Standards, and Reliability of Sustainability Information

Transcription

1 IASB Standard Setting, Enforcement of Financial Reporting Standards, and Reliability of Sustainability Information INAUGURALDISSERTATION zur Erlangung der Würde eines Doktors der Wirtschaftswissenschaft der Fakultät für Wirtschaftswissenschaft der Ruhr-Universität Bochum Kumulative Dissertation, bestehend aus sechs Beiträgen vorgelegt von Diplom-Kaufmann Michael Wolfgang Stich aus Weiden in der Oberpfalz Juni 2011

2 Dekan: Prof. Dr. Helmut Karl Referent: Prof. Dr. Jürgen Ernstberger Korreferent: Prof. Dr. Bernhard Pellens Tag der mündlichen Prüfung: 27. Oktober 2011

3 Contents List of Tables. VI List of Figures.. IX 1 Introduction Motivation and scope Summary and publication details. 6 2 Do IASB s Standard Setting Pronouncements Meet Investors Demands around the World? Introduction Background, prior studies, and hypotheses Global importance of the IASB Due process of the IASB Standard setting strategy of the IASB over time Prior literature Hypotheses Research design Synthetic benchmarks Measurement variables Cross-sectional analyses Identification of events, data sources, and sample distribution Identification of standard setting pronouncement events Data sources Sample distribution Empirical findings IFRS adoption by jurisdiction Information content of (expected) changes in IFRS Expected net-benefits of (expected) changes in IFRS Comparison to a random event sample I

4 2.6 Summary and concluding remarks 58 Appendix 2.1 IASB standard setting pronouncements from 2001 to Appendix 2.2 Definition of propensity score matching variables.. 66 Appendix 2.3 Definitions of pronouncement content measures and control variables on the firm level Why do firms produce erroneous IFRS financial statements? Introduction Background: enforcement of IFRS in Germany Literature review and empirical predictions Literature review Financial reporting expertise Accounting resources Management incentives Governance quality Methodology Conditional logistic regression model Matching procedure Sample selection and data description Sample selection Data description Results Descriptive statistics and correlations Multivariate analyses Additional analyses Robustness checks Conclusion and limitations.. 98 II

5 4 Enforcement of Accounting Standards in Europe: Capital Market Based Evidence for the Two-tier Mechanism in Germany Introduction Background Enforcement of IFRS in the European Union The German enforcement model Literature review and hypotheses Related literature Investor reactions to error announcements Magnitude of market value discount Methodology Metrics of the short-term investor reactions to the error announcement Metrics of the long-term investor reactions to the error announcement Determinants model Empirical findings Sample selection Data sources, descriptive statistics, and correlations Short-term investor reactions to the error announcement Long-term investor reactions to the error announcement Determinants of short-term investor reactions to the error announcement Sensitivity tests Conclusions Economic Consequences of Accounting Enforcement Reforms: The Case of Germany Introduction Institutional background Prior research and hypotheses Enforcement reforms and the overall degree of enforcement. 148 III

6 5.3.2 Earnings quality effects of the enforcement reforms Capital market effects of the enforcement reforms Impact of enforcement through other mechanisms Research design, sample selection, and data Regression approaches Measurement variables Sample selection and data sources Empirical findings and sensitivity analyses Empirical findings Sensitivity analyses Conclusion Reforms to the Enforcement of IFRS and the Cost of Equity and Debt Capital Introduction Institutional background German enforcement prior to the reforms German enforcement reforms Prior research and hypotheses Overall degree of enforcement Enforcement and earnings quality Enforcement and trust in IFRS information Enforcement and the cost of equity and debt capital Research design and measurement variables Impact of the enforcement reforms on earnings quality and trust Direct impact analyses for the cost of equity and debt capital Path analyses for the cost of equity and debt capital Measurement variables Sample and data Sample definitions IV

7 6.5.2 Sources of data and sample distribution Empirical findings Descriptive statistics and correlations Impact of the enforcement reforms on earnings quality and trust Impact of the enforcement reforms on the cost of equity and debt capital Sensitivity analyses Concluding remarks Market Reactions to Increased Reliability of Sustainability Information Introduction Background Hypotheses development Research design Sample selection and data Event study methodology Multiple regression analysis Empirical findings and discussion Capital market reactions Determinants of stock market reactions Further discussion of competing hypotheses Interaction of company risks and economic uncertainties Effects of a deletion from the sustainability index Further sensitivity analyses Conclusions Bibliography Curriculum Vitae V

8 List of Tables Table 2.1 Sample distribution by jurisdiction Table 2.2 Sample distribution by type of pronouncement 40 Table 2.3 Sample distribution by year and type of pronouncement. 41 Table 2.4 Table 2.5 Table 2.6 Table 2.7 Table 2.8 Table 2.9 Table 2.10 Table 2.11 Table 2.12 Macroeconomic determinants of the IFRS adoption in different jurisdictions 43 Overall abnormal return volatility and cumulative abnormal returns to IASB standard setting pronouncements. 44 Abnormal return volatility and cumulative abnormal returns to IASB standard setting pronouncements by time period Abnormal return volatility and cumulative abnormal returns to IASB standard setting pronouncements by jurisdiction Abnormal return volatility and cumulative abnormal returns to IASB standard setting pronouncements by Leuz (2010) jurisdiction cluster Abnormal return volatility and cumulative abnormal returns to IASB standard setting pronouncements by industry.. 50 Abnormal return volatility and cumulative abnormal returns to IASB standard setting pronouncements by type of pronouncement Descriptive statistics for pronouncement content measures and control variables on the firm level Impact of the pronouncement content on cumulative abnormal returns.. 57 Table 3.1 Definition of variables 80 Table 3.2 Erroneous IFRS financial statements 83 Table 3.3 Impact of the IFRS financial statement errors on firm key numbers.. 85 Table 3.4 Distribution of observations by industry Table 3.5 Descriptive statistics of variables Table 3.6 Correlations of variables Table 3.7 Conditional logistic regression results.. 93 Table 4.1 Examinations completed and error findings by the DPR. 108 Table 4.2 Definition of variables 120 VI

9 Table 4.3 Descriptive statistics and correlations 125 Table 4.4 Short-term investor reactions to error announcements Table 4.5 Long-term investor reactions to error announcements Table 4.6 Determinants of short-term investor reactions to error announcements. 138 Table 5.1 Definition of variables 161 Table 5.2 Proportions of IFRS applying companies. 165 Table 5.3 Descriptive statistics and correlations 167 Table 5.4 Table 5.5 Table 5.6 Regression results for the earnings management, stock liquidity, and market valuation measures. 171 Impact of the level of enforcement through other mechanisms 175 Regression results for the earnings management, stock liquidity, and market valuation measures (matched sample approach) Table 6.1 Industry and country distribution of the sample 217 Table 6.2 Descriptive statistics Table 6.3 Correlations of regressor variables. 220 Table 6.4 Table 6.5 Table 6.6 Table 6.7 Regression results indicating the impact of the enforcement reforms on earnings quality and capital provider trust Regression results for the impact of enforcement reforms on the cost of equity capital and on the cost of debt capital 224 Path decomposition of the relationship between enforcement levels and the cost of equity capital. 227 Path decomposition of the relationship between enforcement levels and the cost of debt capital Table 7.1 Sample distribution by industry Table 7.2 Sample distribution by country Table 7.3 Definition of variables 249 Table 7.4 Investor reactions to an addition to the DJSI STOXX Table 7.5 Descriptive statistics of continuous control variables Table 7.6 Correlation matrix of cumulative abnormal returns and continuous control variables VII

10 Table 7.7 Table 7.8 Determinants of investor reactions to an addition to the DJSI STOXX (univariate results) 257 Determinants of investor reactions to an addition to the DJSI STOXX (regression results) VIII

11 List of Figures Figure 2.1 Yearly number of discussion papers, exposure drafts, and final standards pronouncements by the IASB 20 Figure 2.2 Distribution of abnormal return volatility. 44 Figure 2.3 Distribution of cumulative abnormal returns 53 Figure 4.1 Illustration of the long-term stock performance of censured and benchmark firms Figure 6.1 Illustration of the expected path analysis impacts. 207 Figure 6.2 Illustration of the decomposition IX

12 1. Introduction 1.1 Motivation and scope IASB standardsetting International Financial Reporting Standards (IFRS) are on the way to become the globally predominating accounting regime. Today, more than 120 jurisdictions around the world permit or require IFRS financial statements for domestic listed firms (AICPA, 2011). Most major economies that have not yet introduced IFRS have established time lines to converge with or adopt IFRS in the near future either on a voluntary or a mandatory basis. IFRS are expected to bring along several benefits to companies stakeholders and especially to their investors. The major argument for the IFRS adoption by the European Union (EU) in 2005 was the expectation that IFRS will help to eliminate cross-border financing barriers by ensuring that the financial statements of firms provide more reliable, transparent, and comparable information (European Commission, 2002). The IFRS Foundation states its objectives to develop, in the public interest, a single set of high quality, understandable, and enforceable global accounting standards that require high quality, transparent, and comparable information in financial statements and other financial reporting to help participants in the world s capital markets and other users make economic decisions. (IFRS Foundation, 2010, par. 2a) Recently, the chairman of the International Accounting Standards Board (IASB) stated his high expectations regarding the application of IFRS: 1

13 Global accounting standards will enhance the drive towards the free trade of capital internationally. By adopting a globally accepted set of standards, all companies large and small are able to attract capital from a larger pool of investors, driving down the cost of capital, and facilitating cross-border mergers and acquisitions activity and strategic investments. (Tweedie, 2011) The convergence between the IASB and the Financial Accounting Standards Board (FASB) is of particular importance for the global acceptance of the IFRS. Based on the Memorandum of Understanding (FASB and IASB, 2002), the two boards coordinated their efforts and conducted several joint standard setting projects. The elaboration of the financial reporting rules around the world is observed on highest diplomacy level. For instance, the leaders of the Group of 20 (G20) countries confirmed their willingness to get favored accounting standards in the foreseeable future: We call on our international accounting bodies to redouble their efforts to achieve a single set of high quality, global accounting standards within the context of their independent standard setting process, and complete their convergence project by June The IASB s institutional framework should further enhance the involvement of various stakeholders. (G20, 2009, par. 14) In June 2011, the US Securities and Exchange Commission (SEC) staff published a work plan for the consideration of IFRS incorporation into the financial reporting system for United States (US) firms. The proposed condorsement - procedure (SEC, 2011) is controversially debated. The first purpose of this Doctoral Thesis is to shed light on the economic consequences of the IASB standard setting process. The IFRS are broadly discussed in practice and theory, however, it is an open question whether the expected benefits of IFRS changes have actually materialized and whether the standard setting efforts of the IASB are beneficial from an investor perspective. 2

14 The first paper of this Doctoral Thesis ( Do IASB s Standard Setting Pronouncements Meet Investors Demands around the World? ) evaluates investor reactions to specific milestones in the IASB s due process. The empirical findings provide evidence that investors take notice of the standard setting activities of the IASB. However, there are no positive overall market reactions to the IASB standard setting pronouncement around the world. These results suggest that investors do not expect that the benefits of new standards outweigh the respective cost. Further, standard setting pronouncements which tend to increase options and choices as well as fair value accounting are associated with more negative market reactions than other pronouncements. Enforcement of financial reporting standards The second part of this Doctoral Thesis deals with different aspects of establishing enforcement mechanisms in Germany. While the development, implementation, and adoption of accounting principles and their consequences are well discussed in current accounting literature, considerably less is known about the impact of the application of these standards. Accounting researchers conclude that accounting standards by their own do not determine financial reporting outcomes. Besides other forces, the overall level of enforcement is identified as one driver of financial reporting quality. Several studies suggest that even nuances in how securities laws are applied and enforced can lead to considerable differences in observed outcomes. In the context of the rapidly occurring adoption of IFRS around the world, this is also an issue of great urgency. Even if there is a uniform set of financial reporting standards around the world, several expected benefits will not be fully realized unless the underlying institutional and economic factors evolve to become sufficiently similar. The chairman of the IASB notices: In Britain everything is permitted unless it is prohibited. In Germany it is the opposite, everything is prohibited unless it is permitted. In the Netherlands everything is prohibited even if it is permitted. And in France, of course, everything is permitted especially if it is prohibited. (Tweedie, 2009) 3

15 The following papers dealing with the impact of the enforcement level directly respond to a suggestion of Holthausen (2009) to analyze the impact of external enforcement reforms on reporting outcomes and capital market properties. The German government stated in 2004: Corporate scandals in the past caused by earnings manipulations have undermined the trust of investors in capital markets. It is an urgent target of the Federal Government to restore and to encourage the trust of investors in the accuracy of financial reports and thus in capital markets. (Deutscher Bundestag, 2004) The second paper ( Why do German Firms Produce Erroneous Accounting? ) discusses drivers of the IFRS accounting quality in Germany. The German Financial Reporting Enforcement Panel (DPR Deutsche Prüfstelle für Rechnungslegung) primarily attributes detected materially erroneous financial reports to the enormous scope and highly complex nature of IFRS, which overwhelms companies (DPR, 2009). The empirical findings indicate that opportunistic motives are conducive to erroneous accounting, while governance quality increases the probability that firms prepare compliant financial statements. However, the results suggest that material errors in mandated IFRS financial statements cannot be attributed to firms being unable to cope with IFRS. Especially the latter aspect is important for the IASB, other standard setters, regulators, and enforcers. The third paper ( Enforcement of Accounting Standards in Europe: Capital Market Based Evidence for the Two-tier Mechanism in Germany ) contributes to the understanding of capital-market based sanction-mechanisms in civil law countries such as Germany. A key assumption for the effectiveness of the currently introduced German enforcement procedures is that firms are deterred from publishing erroneous financial reports as they fear sufficiently negative market reactions if the firm has to announce this circumstance in the course of the enforcement mechanism ( name and shame ). Indeed, such negative market reactions occur. More in detail, the findings suggest that capital market 4

16 participants perceive the detailed description of the detected accounting errors published within the error announcements as well as the (un-)willingness of the firm to cooperate with the enforcement institutions. The fourth and fifth paper of this Doctoral Thesis ( Economic Consequences of Accounting Enforcement Reforms: The Case of Germany and Reforms to the Enforcement of IFRS and the Cost of Equity and Debt Capital ) investigate economic consequences of recent reforms to the enforcement in Germany. Results suggest a desirable impact of these reforms on the accounting policy as well as on important capital market properties of firms that fall under the new enforcement regime. Findings provide further evidence that these enforcement reforms leveled the playing field in the enforcement of financial reporting. Reliability of sustainability information Besides classical information provided in financial statements, sustainability information are published in a more or less standardized form commonly seen as important for the evaluation of current and future performance. In this context, sustainability is a term which is frequently used by companies to describe their economical, social, and environmental orientation. In the so-called Brundtland Report, the United Nations (UN) World Commission on Environment and Development describes sustainability as a development that meets the needs of the present without compromising the ability of future generations to meet their own needs. (UN, 1987) It seems to be worthwhile from an ethical perspective but also from the view of the homo oeconomicus to consider and support firms sustainability orientation. The sixth paper ( Market Reactions to Increased Reliability of Sustainability Information ) evaluates whether investors consider the reliability of companies sustainability information when determining companies market value. As theoretically predicted, findings support that the reliability of sustainability information is important from an investor s perspective. More specifically, the findings suggest that investors of firms carrying higher information and 5

17 investment risks particularly demand more reliable sustainability information. Finally, the results provide evidence that this demand is even higher in times of economic uncertainties. The remainder of this Doctoral Thesis proceeds as follows. Section 1.2 summarizes the individual papers as well as publication details, before the sections 2 to 7 present six research papers in all detail. 1.2 Summary and publication details The Doctoral Thesis at hand is a cumulative work that consists of six individual papers in the context of IASB standard setting, enforcement of financial reporting standards, and reliability of sustainability information. In this section, each of the individual papers is briefly summarized and co-authors are named (no first-author assignments). Moreover, publication details are given. Please note that there are later versions of all papers available in the respective journals as indicated in the text or on scientific online platforms. Thus, please cite the current versions of the papers. 6

18 Paper 1: Do IASB s Standard Setting Pronouncements Meet Investors Demands around the World? This paper investigates whether the IASB standard setting pronouncements, i.e., the publications of specific milestones (discussion papers, exposure drafts, and final standards) in the IASB s due process meet the demands of investors. Specifically, we analyze abnormal return volatilities and cumulative abnormal returns of IFRS firms to the pronouncements. The major challenge of the IASB in the standard setting process is to evaluate whether the benefits of new standards outweigh the costs in order to gauge their efficiency. Using a sample of all IASB pronouncements from 2001 through 2009 we find that IASB standard setting pronouncements are considerably perceived by investors. However, there are no positive overall market reactions to IASB standard setting pronouncements of investors around the world. Yet, the market reactions vary cross-sectionally by jurisdiction, time period, and type of pronouncement. Further, we provide some evidence that standard setting pronouncements which tend to increase options and choices as well as fair value accounting are associated with more negative cumulative abnormal returns than other pronouncements. Co-author: Jürgen Ernstberger. JEL Classification: G14, M48, and M41. Keywords: IFRS, IASB, standard setting, due process, event study, and market reaction. Publication details: Conference paper. 7

19 Paper 2: Why do German Firms Produce Erroneous Accounting? Given voiced concerns with the alleged complexity of IFRS, this paper investigates determinants of material errors in mandated IFRS financial statements. For a matched sample of firms censured by the German enforcement mechanism for producing erroneous IFRS financial statements, we investigate the role of complexity and thus sources of unintentional accounting errors, and control for intentional factors. In line with the previous literature, we find that, as predicted, the presence of opportunistic motives is conducive to erroneous accounting, while governance quality increases the probability that firms prepare compliant financial statements. In contrast, we find no evidence that unintentional factors pertaining to financial reporting expertise and accounting resources shape reporting quality. The results of this paper are important for regulators and standard setters, as they suggest that material errors in mandated IFRS financial statements cannot be attributed to firms being unable to cope with the new and complex standards, but are due to earnings management incentives and weak controls. Co-authors: Jürgen Ernstberger and Jörg-Markus Hitz. JEL Classification: M41. Keywords: accounting quality, enforcement, IFRS, regulation, restatements, and Germany. Publication details: Revise and resubmit at an international journal. 8

20 Paper 3: Enforcement of Accounting Standards in Europe: Capital Market Based Evidence for the Two-tier Mechanism in Germany On the background of regulatory initiatives that mandate development of comparable enforcement systems in EU jurisdictions to ascertain consistent and faithful application of IFRS, this paper provides capital market based evidence on investor reactions for one specific institutional setup: the two-tier enforcement system in Germany. In operation since 2005, the German enforcement mechanism consists of a private body, the DPR, which investigates compliance of financial statements published by publicly traded firms, and, upon error findings, involves the German securities regulator BaFin, which on a second level enforces disclosure of these findings to create adverse disclosure. For a sample of all error findings published until the end of 2009, we address the potential effectiveness of this name and shame mechanism by investigating short- and long-window capital market reactions upon error announcements. Results on abnormal returns, abnormal trading volumes and abnormal bid-ask spreads indicate that these announcements represent new, negative information and suggest that, despite an enforcement environment that is categorized as weak in the extant literature, the activities of the DPR/BaFin are effective in penalizing infringing firms and thus provide potential deterrence. In detailed analyses, we find that the magnitude of abnormal return reactions is positively associated with the severity of the error(s) and with the threat of follow up litigation, while indication of unintentional error(s) appears to reduce the valuation discounts imposed by investors. Co-authors: Jürgen Ernstberger and Jörg-Markus Hitz. JEL classification: M 41. Keywords: enforcement, IFRS, accounting quality, restatements, regulation, and Germany. Publication details: European Accounting Review (forthcoming). 9

21 Paper 4: Economic Consequences of Accounting Enforcement Reforms: The Case of Germany This study investigates recent reforms of financial reporting enforcement in Germany. The objective of these reforms was to promote a consistent and faithful application of accounting standards. Conducting multivariate analyses, we find a significant decrease in earnings management, a significant increase in stock liquidity, as well as limited evidence for an increase in market valuation for the companies that fall under the new enforcement regime. We document that companies that are characterized by an overall low level of enforcement through other internal and external mechanisms are particularly affected by these reforms. In our analyses, we carefully control for effects of the mandatory IFRS adoption in Europe as well as for various other effects arising from the application of different accounting principles. Our results hold for different research designs, incl., a difference-in-differences approach and a matched sample approach. We conclude that the enforcement reforms in Germany have leveled the playing field in the enforcement of financial reporting and have enhanced the trust in financial reports of the affected companies. Co-authors: Jürgen Ernstberger and Oliver Vogler. JEL classification: M 41 and M 48. Keywords: enforcement; regulation; IFRS; earnings management; stock liquidity; market valuation. Publication details: European Accounting Review (forthcoming). 10

22 Paper 5: Reforms to the Enforcement of IFRS and the Cost of Equity and Debt Capital This paper investigates the impact of recent reforms to the enforcement of IFRS financial reporting in Germany to the cost of equity and debt capital. The objective of these governmental reforms was to promote a more consistent and faithful application of financial reporting standards by publicly traded firms and for IFRS financial statements in particular. As a major regulatory chance a twotier external enforcement mechanism combining private and public financial reporting oversight was established. Beside this, the scope of the auditors oversight was restructured and new auditor independence rules were enacted. First, I evaluate if the reforms to the enforcement of IFRS have lowered the cost of equity and debt capital finding weak evidence for such an impact. Second, I perform a path-analysis using the increase in earnings quality and trust in information prepared by firms which are caused by the reforms to the enforcement of IFRS as a link to the cost of equity and debt capital. In this case, I find considerable evidence for the effectiveness of these reforms on the cost of equity and debt capital. Further, I document that the increase in earnings quality is a link which is equally important for equity and debt holders whereas the increase in trust in the information provided by IFRS reporting firms has a considerably higher impact for equity than for debt holders. JEL classification: M 41 and M 48. Keywords: enforcement, IFRS, governance reforms, cost of equity capital, cost of debt capital, and path analysis. Publication details: Conference paper. 11

23 Paper 6: Market Reactions to Increased Reliability of Sustainability Information This paper investigates whether investors consider the reliability of companies sustainability information when determining the companies market value. Specifically, we examine market reactions (in terms of abnormal returns) to events that increase the reliability of companies sustainability information but do not provide markets with additional sustainability information. Controlling for competing effects, we regard companies additions to an internationally important sustainability index as such events and consider possible determinants for market reactions. Our results suggest that first, investors take into account the reliability of sustainability information when determining the market value of a company and second, the benefits of increased reliability of sustainability information vary cross-sectionally. More specifically, companies that carry higher risks for investors (e.g., higher systematic investment risk, higher financial leverage, higher levels of opportunistic management behavior, and so on) react more strongly to an increase in the reliability of sustainability information. Finally, we show that the benefits of an increase in the reliability of sustainability information are higher in times of economic uncertainty (e.g., during economic downturns and generally high stock price volatilities). Co-authors: Julia Lackmann and Jürgen Ernstberger. JEL classification: M14, M41, and M48. Keywords: sustainability, corporate social responsibility, CSR, reliability, market reactions, and event study. Publication details: Journal of Business Ethics (forthcoming). 12

24 2. Do IASB s Standard Setting Pronouncements Meet Investors Demands around the World? Abstract This paper investigates whether the IASB standard setting pronouncements, i.e., the publications of specific milestones (discussion papers, exposure drafts, and final standards) in the IASB s due process meet the demands of investors. Specifically, we analyze abnormal return volatilities and cumulative abnormal returns of IFRS firms to the pronouncements. The major challenge of the IASB in the standard setting process is to evaluate whether the benefits of new standards outweigh the costs in order to gauge their efficiency. Using a sample of all IASB pronouncements from 2001 through 2009 we find that IASB standard setting pronouncements are considerably perceived by investors. However, there are no positive overall market reactions to IASB standard setting pronouncements of investors around the world. Yet, the market reactions vary cross-sectionally by jurisdiction, time period, and type of pronouncement. Further, we provide some evidence that standard setting pronouncements which tend to increase options and choices as well as fair value accounting are associated with more negative cumulative abnormal returns than other pronouncements. 2.1 Introduction During the last decade the International Financial Reporting Standards (IFRS) have become global accounting standards for the preparation of financial statements by publicly traded firms. The IFRS are issued by the International Accounting Standards Board (IASB) which began operations in 2001, when it succeeded the International Accounting Standards Committee (IASC). The major objective of the IASB is to develop a single set of high quality, understandable, and enforceable global accounting standards and to encourage convergence on these standards (IFRS Foundation, 2010, par. 2). IFRS are developed through a due process, i.e., an international consultation process that involves interested individuals and organizations from around the world. The primary challenge of the IASB is to evaluate whether the benefits of new standards and modifications 13

25 of existing standards outweigh the cost in order to gauge the efficiency of the IFRS application as a whole. This paper investigates whether the IASB standard setting pronouncements, i.e., the publications of specific milestone documents in the IASB s due process (discussion papers, exposure drafts, and final standards) meet the demand of investors. Specifically, we explore investors market reactions to all pronouncements between 2001 and As measures for the market reactions we use abnormal return volatility as well as cumulative abnormal return measures calculated for three-day event-windows around the publication of new pronouncements. We basically examine market reactions of all IFRS applying firms around the world from 2001 to 2009 at the day of the pronouncement of milestones in the IASB standard setting process. We find that the IASB activities are important for investors and substantially perceived when determining market values of the firms as we detect highly significant abnormal return volatilities around the days of the IASB standard setting pronouncements. However, these pronouncements are overall not associated with positive cumulative abnormal returns. In detail, we find an average market reaction of -0.13% over all pronouncements included in this study which is also economically significant. Basically, our results contradict the broadly articulated expectation that overall the IASB s standard setting pronouncements are perceived to entail net-benefits for investors. The reason for this result might be the worldwide scope of the IFRS which makes it even more difficult to understand the implications of standard setting pronouncements in different jurisdictions with different institutional settings. We discuss further reasons for these findings, incl., the possibility that the IASB pronouncements are net-beneficial for all stakeholders but not for shareholders of the firm. We also examine whether the market reactions vary by the type of standard setting pronouncements. We focus on discussion papers, exposure drafts, and final standards as the most important types of standard setting pronouncements as well as the most prominent and path-breaking steps of the IASB s due process. We do 14

26 not find economically important difference in the market reactions to these types of standard setting pronouncements. To explore whether market reactions change over time we conduct an in-depth analysis of the IASB s standard setting pronouncements and identify three time periods with different strategies, i.e., a period of creating a stable platform for the mandatory IFRS adoption in the European Union (EU) (years 2001 to 2004), a period of converging with the United States Generally Accepted Accounting Principles (US GAAP) and of improving as well as expanding the standards (years 2005 to 2007), and a period of reacting to the financial crisis (years 2008 to 2009). For these time periods we find significant differences in investors market reactions being most negative for the last time period. Further, findings suggest that the market reactions significantly differ between jurisdictions and between the three clusters of institutional environments identified by Leuz (2010). The market reactions seem to be more negative in environments which are not represented in the IASB by a member and which are not specifically in the focus of the IASB s standard setting strategy, i.e., in developing jurisdictions or in jurisdictions with weaker enforcement. Moreover, we document differences in the market reactions of different industries but without a clear tendency in the desirability of IASB standard setting pronouncements for specific industries. We also document that standard setting pronouncements which tend to increase options and choices for the preparers of financial statements are associated with more negative market reactions than other pronouncements. Finally, we provide some evidence that (proposed) standards which tend to increase fair value accounting in IFRS financial statements on average exhibit more negative abnormal returns than others. The paper contributes to the accounting literature in several ways. It is the first study which comprehensively documents market reactions to IASB standard setting pronouncement around the world and to assess the determinants of these 15

27 reactions. Our finding that additional options and choices as well as an extension of fair value accounting are associated with negative market reactions is relevant for future standard setting of the IASB. The purpose and findings of this study are related to several other streams in accounting literature. It enlarges the understanding of the desirability of an IFRS adoption as a whole. A broad number of studies convincingly provide evidence that investors experience net-benefits from a switch from local accounting standards (GAAP) to IFRS, i.e., in terms of cost of capital, earnings quality, stock liquidity, and market valuation (e.g., Daske et al., 2008; Leuz and Verrecchia, 2000). The results of this study indicate that the (expected) changes of the IFRS and/or the way they are procedurally modified (e.g., the due process and the way shareholder can participate) are not netbeneficial from an investor perspective. In addition, the findings of this study expand the understanding of the application of IFRS. Comprehensive reviews are provided for the consequences of the voluntary IFRS adoption by Soderstrom and Sun (2007) and for the consequences of the mandatory IFRS adoption by Brüggemann et al. (2010). Most related to this piece of research are several studies analyzing investors short-window responses to the likelihood of an adoption of IFRS, either on a voluntary or mandatory base, and its determinants (e.g., Armstrong et al., 2010; Christensen et al., 2007; Comprix et al., 2003). Furthermore, his study adds to the literature examining market reactions to the pronouncement of single new accounting standards (e.g., Beatty et al., 1996; Dechow et al., 1996a; D Souza, 2000; Espahbodi et al., 1991, 1995, 2002; Lys, 1984; Weber, 2004). Finally, our study is related to literature on the political economy of accounting standard setting. While most research in this area investigates constituents involvement in the due process by analyzing comment letters, only a few studies focus on the standard setting process itself. Allen and Ramanna (2010) examine the role of FASB members in shaping the properties of accounting standards. 16

28 The next section of the paper (section 2.2) provides the theoretical background of our study and develops our hypotheses. Section 2.3 presents the research methodology and our measurement variables. Section 2.4 describes our sample and the data used for the analyses. Section 2.5 provides the results and section 2.6 concludes. 2.2 Background, prior studies, and hypotheses Global importance of the IASB Today more than 120 reporting jurisdictions around the world permit or require IFRS financial statements for domestic listed firms (AICPA, 2011) and most major economies that have not yet introduced IFRS have established time lines to converge with or adopt IFRS in the near future. Also a large number of developing countries are on the way to adopt IFRS if they have not yet gone this step. For the United States (US), the Securities and Exchange Commission (SEC) issued (in December 2007) a rule that allows foreign issuers to submit financial statements to the commission using IFRS without an enclosed reconciliation of the IFRS data to US GAAP. Based on the Memorandum of Understanding (MoU) (FASB and IASB, 2002) the Financial Accounting Standards Board (FASB) and the IASB boards coordinated their standard setting efforts and conducted several joint standard setting projects. In November 2008, the SEC published a roadmap for a potential use of IFRS by US issuers, which sets four milestones that, if achieved, could lead to the required use of IFRS by US firms. In June 2011 the SEC staff published a work plan for the consideration of IFRS incorporation into the financial reporting system for US firms. However, the conduction of the proposed condorsement -procedure (SEC, 2011) is controversially discussed and there are still various uncertainties concerning the if, when, and how the US will adopt IFRS (for an analysis of the economical and political factors see, e.g., Hail et al. [2009]). 17

29 With expanding acceptance and growing importance of the IFRS, the IASB has become a global standard setter. The IASB is an independent standard setting body of the IFRS Foundation and based in London. 1 It consists of 15 full-time members 2 from ten jurisdictions and from different interest groups like national standard-setters, preparers, auditors, users, or academics. The IASB is funded by contributions from firms, private financial institutions, major accounting firms, and other international organizations throughout the world Due process of the IASB The standard setting of the IASB follows a certain due process which generally consists of six stages: first, adding a potential project to the agenda, second, planning the project, third, developing and issuing a discussion paper, fourth, developing and issuing an exposure draft, fifth, developing and issuing a new standard or changes to an existing standard, and sixth, holding meetings with interested parties, incl., other (e.g., national) standard setting bodies, to help understand issues related to the practical implementation and potential impact of the changes. In stage one the IASB consults the Standards Advisory Council (SAC), which consists of approximately 40 representatives from organizations that are relevant to standard setting, and discusses a possible project in public meetings. When adding a project to its active agenda, the IASB decides in stage two on whether to conduct the project jointly with the FASB and whether a working group is established. The IASB selects a project manager who draws up a working plan and appoints project team members. A discussion paper is published in stage three for major new projects to provide an overview of important issues and possible approaches in addressing these issues. It may also include preliminary views of the IASB on the possible approaches and seeks to receive public comments from interest groups. The IASB could publish a discussion paper drafted either by its 1 Effective in July 2010 the International Accounting Standards Committee Foundation (IASC Foundation) was renamed without changes in function and competences as International Financial Reporting Standards Foundation (IFRS Foundation). 2 At a meeting in January 2009 the trustees of the IFRS Foundation decided to expand the IASB to 16 members by

30 members or by supporting other standard setters. For instance, a recently published discussion paper about fair value measurement was developed by the FASB. Own discussion papers of the IASB are usually based on a research paper provided by its staff and considers comments by the SAC and other standardsetters. Stage four involves the publication of an exposure draft by the IASB, which is the main step of the due process in order to consult the public. Unlike a discussion paper, an exposure draft is mandatory in each due process and proposes specific rules for a new standard or amendments to an existing standard. Based on the public comments, the IASB revises its standard proposal and considers whether to issue a re-exposure for public comment. When the IASB decides that a final conclusion is reached about all major issues, the staff develops a pre-ballot draft in stage five. This draft is normally reviewed by the International Financial Reporting Interpretation Committee (IFRIC) and a near-final draft is published by the IASB on its internet platform eifrs. Finally, the new standard (amendment to an existing standard, respectively) is issued when the IASB members have balloted in favor of it. Stage six consists of educational activities to help interested parties in understanding and correctly applying the new standard (amendment) Standard setting strategy of the IASB over time In the following, we analyze the IASB s standard setting strategy (e.g., Allen and Ramanna, 2010) leading to several major pronouncements analyzed afterwards. Appendix 2.1 provides an overview of the standard setting pronouncement of the IASB for the years 2001 to Figure 2.1 illustrates the number of (projected) changes in standards pronounced by the IASB per year. We identified three time periods with different strategies: the first period from 2001 through 2004 was characterized by starting a convergence project with the FASB and creating a stable platform for the mandatory IFRS adoption in the EU. The IASB assumed accounting standard-setting responsibilities from its predecessor body, the IASC on April 1, This was the result of a fundamental restructuring of the IASC to realign its strategy and to foster its role as an independent and international standard setter. After consultations with national standard-setters and other interested parties, the IASB announced an 19

31 initial agenda of nine technical projects. In the following, the IASB amended several existing IAS and in 2003 issued its first new standard on first-time adoption of IFRS, named IFRS 1. In September 2002, the IASB had a joint meeting with the FASB and issued the so-called Norwalk Agreement in which they committed to a strategy of convergence between IFRS und US GAAP. To achieve this goal both standard setters revised or replaced existing standards or conducted joint projects in order to eliminate differences. By the end of 2003, the IASB completed its first Improvements Project which revised 14 standards and covered a variety of issues related to problems with implementing existing standards. In 2004, the IASB issued four IFRS, two revised International Accounting Standards (IAS) and an amendment to the financial instruments standard by the end of March and announced the completion of a stable platform of standards for use by firms adopting IFRS from January Figure 2.1 Yearly number of discussion papers, exposure drafts, and final standards pronouncements by the IASB Notes: This figure shows the yearly number of discussion papers (white column), exposure drafts (grey column), and final standards (dark grey column) pronounced by the IASB (no pronouncements in 2001). Annual improvement projects are included as one change in total. 20

32 The second period from 2005 through 2007 was characterized by the convergence project with the FASB and by an improvement and expansion of the standards. In two joint meetings, the IASB and the FASB reaffirmed the MoU in 2005 and issued a roadmap for completing the major MoU projects by The IASB started its second Improvements Project which was enacted in 2008 and expanded the scope of its standard setting, e.g., to management reports and to accounting rules for small and medium sized entities. The third period from 2008 to 2009 was characterized by reactions to the financial crisis. At the climax of the financial crisis in 2008 the IASB bypassed the regular due process and announced amendments to IAS 39 ( Financial Instruments: Recognition and Measurement ) and IFRS 7 ( Financial Instruments: Disclosures ) to relax reclassification rules for financial instruments measured at fair value due to political pressure (e.g., Bischof et al., 2009; Laux and Leuz, 2009). In the following, several financial crisis related projects were added to the agenda, e.g., the replacement of IAS 39, the amendment of consolidation rules, and a new standard about the measurement of fair values. The leaders of the G20 nations issued declarations after their meetings in Pittsburgh 2009 and in Toronto 2010 which requested the IASB and FASB to increase their efforts to converge their standards. The originally stated deadline of 2011 was not repeated in In addition, the Group of Twenty (G20) leaders encouraged the IASB to further improve the involvement of stakeholders, incl., outreach to emerging market economies. By the end of 2010, the IASB and the FASB published documents seeking views on their standard setting strategy. They acknowledged that several new standards were about to be finalized and they requested whether or how to sequence effective dates in order to reduce the burden to interested parties Prior literature Our study is related to the literature on market reactions to the pronouncement or adoption of single new accounting standards or other pieces of regulation. These studies document significant market reactions to the release of new accounting standards that affect reported income and/or shareholders equity. For example, Collins et al. (1981) and Lys (1984) investigate the determinants of the market 21

33 reactions to the proposed elimination of full cost accounting in the oil and gas industry which negatively affected reported earnings and shareholders equity. The studies find that debt covenants and firm size are important determinants of the negative abnormal market returns for firms in the oil and gas industry. Espahbodi et al. (1991) analyze the market reaction to nine pronouncements related to the proposed Statement of Financial Accounting Standards (SFAS) 106 ( Employers Accounting for Postretirement Benefits Other Than Pensions ) accounting for postretirement benefits other than pensions. They show that firms issuing such benefits exhibit significant negative market reactions around the issuance of the exposure draft. In line with the contracting hypothesis they find that the value-decreasing impact of the required disclosures are more pronounced for less mature, small, and highly leveraged firms. D Souza (2000) and Khurana and Loudder (1994) complement these results by suggesting a different market reaction for specific regulated firms. They provide evidence that investors in public utilities did not, on average, view the proposed standard as a valuedecreasing event. Moreover, the relation between financial leverage and the market impact of the proposed standard is contrary to that documented for the general population of firms. While the aforementioned studies explore the impact of eliminating income-increasing accounting method, Espahbodi et al. (1995) and Ricks and Hughes (1985) focus on the impact of introducing income-increasing standards. The former find no market reaction in weeks containing public pronouncements leading up to and including the Accounting Principles Board s adoption of a change from the cost to the equity method of accounting for longterm investments. The latter ones focus on accounting for income taxes, namely SFAS 96 ( Recognition of Assets and Liabilities ) and SFAS 109 ( Accounting for Income Taxes ). Around the exposure draft dates of these standards they find significant positive abnormal returns. The stock market reactions vary crosssectionally and are more pronounced for small firms with high debt ratios and high deferred tax liabilities or high loss carry-forward benefits. Beatty et al. (1996) analyze the impact of fair value accounting according to SFAS 115 ( Accounting for Certain Investments in Debt and Equity Securities ) on stock prices of bank holding firms and insurance firms. They find negative market returns of bank shares, but only little market reactions of insurance firm shares. 22

34 Analyzing the determinants of the market reactions, they provide evidence that banks with more frequently traded investments, with longer maturing investments and with more fully hedged portfolios were most negatively affected by the new standards. Dechow et al. (1996a) and Espahbodi et al. (2002) examine the market reactions to pronouncements related to accounting for stock-based compensation (in particular SFAS 123 [ Accounting for Stock-Based Compensation ]). While Dechow et al. (1996a) neither find any significant stock market reaction to these pronouncements nor any firm-specific variables to be related to the stock market reactions, Espahbodi et al. (2002) document significant abnormal returns around the pronouncements which are more pronounced for high-tech, high-growth, and start-up firms. Weber (2004) investigates market reactions to the adoption of Staff Accounting Bulletin (SAB) 96 which forced firms to choose between maintaining their share repurchase programs and using the pooling-of-interests method for business combinations. He documents a statistically significant negative market reaction of firms that had pending pooling-of-interests mergers on the date of the pronouncement. The reaction is less pronounced for firms that used pooling-ofinterests to avoid relatively larger reductions in future earnings. Both results imply that investors considered the pronouncement of the new restriction on the use of pooling-of-interests to be costly. Two studies (Lo, 2003; Marquardt and Wiedman, 2007) examine the impact of new regulations which do not affect reported income or shareholders equity. Lo (2003) investigates the 1992 revision of executive compensation disclosure rules and finds that firms lobbying against this regulation had, in comparison to control firms, abnormal stock returns of 6% over the 8-month period between the pronouncement of the proposed regulation and its adoption. This effect is more pronounced for firms which lobbied more vigorously against the new regulation. Both results are consistent with the notion that the proposed regulation induced corporate governance improvements to investors. In addition, Marquardt and Wiedman (2007) investigate the revision of financial reporting requirements for contingent convertibles (COCOs) to include them into diluted earnings per share. 23

35 The authors provide evidence of significantly negative stock returns around pronouncement dates associated with financial reporting changes. This is consistent with investor anticipating higher agency costs associated with the change of rules. Our study contributes to this prior literature by documenting the market reactions around the world on IASB standard setting pronouncements for the years 2001 through Hence, we provide comprehensive evidence of all pronouncements of a standard setter during a period of introducing and improving a relatively new accounting regime. In addition, this study adds to the literature which explores the economic consequences of applying IFRS. Comprehensive reviews are provided for the consequences of the voluntary IFRS adoption by Soderstrom and Sun (2007) and for the consequences of the mandatory IFRS adoption by Brüggemann et al. (2010). Extant literature provides evidence of positive reactions by capital markets and at the macroeconomic level upon the IFRS adoption, while they find mixed evidence for the impact on reporting practices. Our study is related to the capital market based consequences of IFRS adoption on, e.g., cost of capital, stock liquidity, stock price informativeness, information content of earnings pronouncements, analysts forecast properties or credit ratings. Most related to our research are studies analyzing investors responses to events that are expected to change the likelihood of an IFRS adoption by the EU. Comprix et al. (2003) explore short-window market reactions to events between 2000 and They provide evidence that the market reactions vary cross-sectionally with expected implementation costs, the quality of corporate governance mechanisms and the extent to which IFRS differs from domestic standards. Christensen et al. (2007) explore both short- and long-window market reactions for a sample of UK firms. They find a firm s willingness to adopt IFRS to be a determinant of the stock market reactions in the short and in the long-run. Pae et al. (2008) show that between 1999 and 2003 Tobin s q ratios of EU firms with high agency costs increased which they attribute to the IAS Regulation. Armstrong et al. (2010) investigate stock market reactions to 16 events between 2002 and 2005 that are 24

36 expected to change the likelihood and scope of mandatory IFRS introduction in Europe. The results are consistent with investors anticipating net benefits from an increase in earnings quality and comparability due to mandatory IFRS adoption. However, investors reactions are negative in code law countries that have weak enforcement environments. We complement the findings of these papers by exploring the capital market reactions to different standard setting pronouncements of the IASB and the determinants of cross-sectional variations of these reactions. Finally, our study is related to the literature on the political economy of accounting standard setting. While most research in this area investigates involvement of constituents in the due process by analyzing comment letters, only a few studies focus on the standard setting process itself. Allen and Ramanna (2010) examine the role of FASB members in shaping the properties of accounting standards. Their findings suggest that longer board tenure and a prior career in investment banking or investment management are associated with proposing standards perceived as decreasing accounting reliability. In contrast, FASB members contributions to the Democratic Party are associated with proposing standards perceived as increasing accounting reliability. We contribute to this literature by exploring the determinants of net benefits attributable to new accounting standards Hypotheses The objective of accounting information is to facilitate the exchange of resources and the enforcement of contracts among various parties related to the firm commonly denoted as the stakeholders of the firm, incl., its shareholders (Healy and Palepu, 2001; Jensen and Meckling, 1976; Kothari et al., 2011; Watts and Zimmerman, 1986). The characteristics of the available financial report information in an economy influence the efficiency of resource allocation as well 3 In line with Armstrong et al. (2010), we note that the of setting our study, which investigates investor reactions to the introduction of a great variety of financial reporting rules, shows some overlapping characteristics with prior research on the economic consequences of the establishment of broad pieces of legislations, e.g., of the Sarbanes-Oxley Act (SOX) in 2002 (e.g., Jain and Rezaee, 2006; Li et al., 2008; Zhang, 2007). 25

37 as the cost of capital (Kothari et al., 2011). Specifically from an investor perspective financial reports lower information and transaction costs between firms and capital providers, first, by enabling the latter to assess possible returns of investments in the firm and, second, by enabling them to monitor the use of capital by investors (e.g., Beyer et al., 2010). Changes in accounting standards can improve financial reporting quality for investors (e.g., Ball, 2006), e.g., by higher transparency and comparability, reduced information asymmetries, and less information risks (e.g., Barth et al., 1999, 2008; Leuz and Verrecchia, 2000) leading to desirable outcomes for investors (e.g., lower cost of capital and more efficient investment decisions). Furthermore, new standards could be beneficial for investors in terms of positive cash flow effects. For instance, Beatty et al. (1996) argue that improved standards have the potential to reduce contracting costs. For example, Hope et al. (2006) relate improved earnings quality to a reduction of the scope for managerial rent extraction. As the IFRS are conceptually focused on the information requirements of investors (IASB 2010, par. 10) new standards could affect firm values by influencing the information set of current and potential investors. Thus, the IASB standard setting pronouncements potentially lead to positive capital market reaction. In contrast, new standards come along with direct and indirect cost for firms and thus for their shareholders. Implementation and transition costs arise for firms adapting information systems and training the employees as well as for investors gaining expertise in the new rules. New standards also lead to costly changes in firms contracts, e.g., management compensation plans or debt contracts. Finally, the new standards might be a compromise between different views of the stakeholders in various jurisdictions which might not account for the needs of investors in a specific jurisdiction. Thus, IASB standard setting pronouncements contain important information about future investors benefits and cost. 4 However, it is ultimately an empirical question whether new pronouncements meet 4 To simplify the exposure of our hypotheses, we focus on aspects of cumulative abnormal returns (CAR it ) within our hypotheses development and do not explicitly refer to our information content measure AVOL it. However, hypothesis 2.1 to 2.5 are also statistically evaluated using AVOL it. In section 2.5, we describe our empirical findings for AVOL it and for CAR it in a row. 26

38 investors needs around the world. Because a priori arguments can be made for either positive or negative market reactions following the IASB s pronouncements, we state the following hypothesis: Hypothesis 2.1: IASB standard setting pronouncements contain information relevant for investors and overall lead to market reactions. Our second hypothesis examines whether the market reactions to standard setting pronouncements (in terms of the information content as well as concerning the desirability from an investor s perspective) have changed over time. We focus on the three time periods with different strategies of the IASB outlined in section 2.2 and conjecture that the market reactions are different. Because of higher uncertainties, increased political lobbying, and a general loss of trust in regulating authorities during the financial crisis the market reactions should be more positive for periods 1 and 2 in comparison to period 3. The difference between the market reactions between period 1 and 2 remains an empirical question. We hypothesize: Hypothesis 2.2: Market reactions to IASB standard setting pronouncements vary by different time periods. In the next analyses, we explore whether the sign and the magnitude of the market reactions to IASB standard setting pronouncements differ in different jurisdictions. Prior literature shows that legal and institutional differences of jurisdictions affect the properties of accounting outcomes (e.g., Ali and Hwang, 2000; Bushman and Piotroski, 2006; Leuz et al., 2003). This implies that uniform regulations for all jurisdictions allowing or requiring the use of IFRS might induce different market reactions. In contrast, new or proposed standards try to accommodate divergent national preferences, e.g., by providing choices and options to firms which might imply relatively comparable market reactions to their pronouncement. We hypothesize: Hypothesis 2.3: Market reactions to IASB standard setting pronouncements vary by different jurisdictions. 27

39 In addition, we evaluate whether the market reactions to IASB standard setting pronouncements vary cross-sectionally by industry. Theoretically, the IFRS are financial reporting rules which basically affect all industries, while there are some standards which actually are established to meet the activities of specific industries (e.g., IFRS 6 [ Exploration for and Evaluation of Mineral Resources ] and IAS 41 [ Agriculture ]). Further, there are IFRS applied by all firms but affect specific industries in particular (e.g., IAS 39 [ Financial Instruments: Recognition and Measurement ] for financials ). Moreover, some reactions of the IASB within the financial crisis are of higher importance for specific industries (e.g., the amendments to IAS 39 and IFRS 7 [ Financial Instruments: Disclosures ]; see, e.g., Bischof et al., 2009; Laux and Leuz, 2009). Thus it seems to be likely that the market reactions to IASB standard setting pronouncements vary crossindustrially leading to the following hypothesis: Hypothesis 2.4: Market reactions to IASB standard setting pronouncements vary by different industries. Further, we evaluate whether the sign and the magnitude of the impact of new IASB pronouncements are influenced by the type of the pronouncement. Our study focuses on discussion papers, exposure drafts, and enactments of standards by the IASB. A discussion paper is not published in each due process. On the one hand, discussion papers are less binding than an exposure draft or even a standard (amendment), as they only describe alternative solutions and provide preliminary views of the IASB on these solutions. Thus, a relatively small market reaction can be expected. On the other hand, discussion papers are often based on completely new theoretical bases. For instance, the discussion paper on leases requires the recognition of all lease contracts on the balance sheet which differs considerably from the current approach of recognizing only assets and liabilities from finance lease contracts. Further, participants of capital markets probably urge the IASB to improve the quality of the IFRS (e.g., to decrease existing uncertainties in standards or unbefitting accounting rules). Hence, there might be a rather large market reaction to discussion papers because it signals to market participants that the IASB is committed to improve the quality of a certain 28

40 regulation. An exposure draft is always published in a due process. It is more binding than a discussion paper because it states draft new regulations. However, the IASB could revise these regulation based on the comment letters received. Moreover, some of these regulations could be inferred from the discussion paper and from the public discussions of the IASB. In sum, the relative magnitude of the market reaction to an exposure draft in comparison to a discussion paper is unclear ex-ante. The final standard usually provides even less news to the market, but it will be mandatory beyond the effective date for all firms applying IFRS. Again, no ex-ante expectation about the difference in market reactions to a new standard (amendment) in comparison to discussion papers or exposure drafts can be stated. Thus, we test the following hypothesis: Hypothesis 2.5: The market reactions to discussion papers, exposure drafts, and final standards are different. Finally, we conjecture that market reactions to standard setting pronouncements by the IASB differ due to their nature. First, we investigate whether major changes, i.e., the introduction of a standard for a new issue or a major amendment of an existing standard by suggesting or requiring a new theoretical approach, trigger different market reactions. Second, we test whether (proposed) standards which provide additional choices or options to its users lead to different market reactions than others. Options and choices provide leeway for earnings management, but they could also be used to signal future performance. Third, the use of fair value accounting is assumed to provide more transparency to the market but also more volatility to earnings and equity. We investigate whether the market reactions to standard setting pronouncement differ between pronouncements increasing the use of fair value accounting and others. Fourth, we explore whether more disclosures in fact provides more information to users of financial statements or whether they cause increased cost for firms effectively bearded by their owners and could lead to an information overload. Based on the above discussion, our sixth hypothesis is as follows: 29

41 Hypothesis 2.6: The market reactions to IASB standard setting pronouncements are different for major vs. minor (proposed) changes, for (proposed) changes increasing vs. not increasing choices and options, for (proposed) increasing vs. not increasing fair value accounting changes, and for (proposed) changes increasing vs. not increasing disclosure requirement. 2.3 Research design Synthetic benchmarks In classical event studies (e.g., Binder, 1998; MacKinlay, 1997) researchers basically evaluate the effect of one firm-specific pronouncement on the market value of this firm relative to an index approximating the unobservable and not (substantially) affected market portfolio. For our study, the latter assumption does not hold as the worldwide market portfolio is substantially influenced by the high proportion of IFRS-applying firms. Accordingly, the first step is to define a benchmark for each jurisdiction under consideration ( IFRS-jurisdiction ) which is not (directly) influenced by the IASB pronouncements. For this purpose earlier studies use return series from other jurisdictions. For example, Armstrong et al. (2010) basically use the largest 1,200 Non-European firms out of the Dow Jones STOXX Global 1800 Index as a benchmark for EU firms. Zhang (2007) basically uses Asian, Canadian, and European firms as a benchmark for US firms (benchmarks on the continent level ). For our major analyses we expand these approaches and rely on jurisdiction-specific synthetic return benchmarks. To generate such a benchmark for each IFRS-jurisdiction, we identify the three most similar Non-IFRSjurisdictions by a propensity score matching procedure based on overall characteristics of these economies. Conceptually comparable to Hope et al. (2006), we estimate the multivariate probit regression model given in equation 2.1 over all jurisdictions I (worldwide) for each calendar year T separately. 30

42 P(IFRS IT =1) = Φ (α 0,T + α 1,T COMMON LAW ORIGIN IT + α 2,T ENFORCEMENT IT + α 3,T CAPITAL MARKET SIZE IT + α 4,T CAPITAL MARKET IMPORTANCE IT + a IT ), (2.1) where Greek letters stand for regression coefficients and a IT is the residual of the regression. IFRS IT is a binary variable taking a value of 1 if the firms of a jurisdiction I apply IFRS for their consolidated financial statements on a voluntary or mandatory base in calendar year T, 0 otherwise. COMMON LAW ORIGIN IT represents the legal origin of the jurisdiction and takes a value of 1 if jurisdiction I is classified as a common law jurisdiction, 0 otherwise (e.g., Djankov et al., 2008). 5 ENFORCEMENT IT is a proxy for the level of law enforcement in jurisdiction I in calendar year T and derived from the Rule of Law and Control of Corruption metric developed by Kaufmann et al. (2010a, 2010b). CAPITAL MARKET SIZE IT indicates the total size of the stock market and CAPITAL MARKET IMPORTANCE IT proxies for the relative importance and development of the stock market in jurisdiction I in the calendar year T. For technical details of the variables definitions see appendix 2.2. We define the three Non-IFRS - jurisdictions with the least absolute propensity score differences to each IFRS - jurisdiction under consideration as its benchmark jurisdictions. Next, we calculate daily returns (continuously compounded definition) for all listed firms around the world (R it ) and aggregate these daily returns for each jurisdiction I using equally weighted portfolios (R It ). To create a synthetic benchmark return series for each IFRS-jurisdiction I* for each calendar year T we regress the daily returns of the IFRS-jurisdiction I* (R IFRS I*,Tt) on the daily returns of the three Non-IFRS benchmark jurisdictions identified for IFRSjurisdiction I* by the above-explained procedure (R NIFRS1 I*,Tt, R NIFRS2 I*,Tt, and R NIFRS3 I*,Tt). For each IFRS-jurisdiction I* and for each calendar year T separately we perform the multivariate regression given in equation For a contemporary overview about the impacts of the legal origin see, e.g., Aguilera and Jackson (2003) and La Porta et al. (2008). 31

43 R IFRS I*,Tt = β 0,I*,T + β 1,I*,T R NIFRS1 I*,Tt + β 2,I*,T R NIFRS2 I*,Tt + β 3,I*,T R NIFRS3 I*,Tt + b I*,Tt, (2.2) where Greek letters are regression coefficients and b I*,Tt is the residual of the regression. We take the fitted values of this multivariate regression (FR IFRS I*,Tt) as the synthetic return benchmark series for IFRS-jurisdiction I* on day t in the calendar year T Measurement variables To evaluate our hypotheses, we use two measures for each event e on the firmlevel. The first measure (AVOL ei ) captures abnormal trading activities around the IASB pronouncements under examination in terms of abnormal return volatility. Abnormal trading activities indicate that the new information is relevant for investors and consequently reflected in stock prices. In line with prior research, we basically measure the abnormal volatility of stock returns at pronouncements of information as the ratio of the event window return volatility to the non-event window return volatility (Beaver, 1968; Landsman and Maydew, 2002; Landsman et al., 2011). We begin by computing daily abnormal returns using the marketmodel method (e.g., Binder, 1998). Thereby, firms abnormal returns are calculated as the difference of realized daily returns (R eit ) and expected daily returns (FR eit ). The latter are determined by a firm-specific market-model which is estimated for a benchmark period covering the 100 trading days prior to the event window. For each firm of the IFRS-jurisdiction I* we use the year-adequate synthetic benchmark return series (FR IFRS I*,Tt) as this is not influenced by the IASB pronouncements per definition. The modified market-model is given in equation 2.3: R eit = γ 0,ei + γ 1,ei FR IFRS I*,Tt + c eit, (2.3) 6 The average fit (adj. R²) of these regressions is 46.3%. 32

44 where γ 0,ei and γ 1,ei are coefficients and c eit is the error term of this model. Based on the residuals of this model, we calculate the stock return residual variance over the 3-day-event window, [-1;1], which approximately equals the average of the squared residuals over the event window (mean t [c 2 eit]) for each event e and firm i separately. Next, we calculate the stock return variance of the estimation window (σ 2 ei), [-101;-2] which equals the variance of the residuals over the 100-trading day estimation window. 7 As the quotient of mean t (c 2 eit) and σ 2 ei is highly right skewed, we follow previous studies (e.g., Landsman et al., 2011) and take its natural logarithm to calculate our abnormal return volatility ratio AVOL ei for event e and firm i. Thus, we get the AVOL ei definition given in equation AVOL ei = ln(mean t [c 2 eit] / σ 2 ei) (2.4) Our second measure for investors reactions to IASB standard setting pronouncements is a cumulative abnormal return metric. Also based on the previously described calculation of daily abnormal returns (AR eit = R eit FR eit ) we define cumulative abnormal returns as the sum of the daily abnormal returns over the three-day event window, containing the event day as well as the trading day before and after this day (CAR ei = AR ei,-1 + AR ei,0 + AR ei,1 ). Positive cumulative abnormal returns indicate that investors expect future net-benefits from the announced decisions of the IASB Cross-sectional analyses In line with our hypotheses, we evaluate whether specific kinds of accounting rule modifications announced in the IASB pronouncements are desirable from an investor-perspective. Thus, we conduct cross-sectional analyses of CAR ei on a set of variables of interest (separately and simultaneously) which indicate specific IFRS content changes. 7 We exclude event windows arising from other IASB standard setting pronouncement events from our estimation window. 8 In line with Landsman et al. (2011), we expect that the quotient of the average of the squared residuals over the event window (mean t [c 2 eit]) and of the abnormal return variance estimated of the estimation window (σ 2 ei) is one. Thus, we expect that the natural logarithm of this quotient is zero if there is no abnormal volatility. 33

45 First, we include FAIR VALUE ei which is a categorical variable taking a value of 1 (-1) if there is an increase (decrease) in fair value accounting, 0 otherwise. 13 (4) pronouncements are classified as fair value increasing (decreasing). We perform the regression model given in equation 2.5: CAR ei = δ 0 + δ 1 FAIR VALUE ei + δ 2 FIRST TIME IFRS ei + δ 3 INTERNATIONALITY ei + δ 4 LEVERAGE ei + δ 5 OWNERSHIP ei + δ 6 PROFITABILITY ei + δ 7 SIZE ei + δ 8 MOMENTUM ei + δ 9 STOCK VOLATILITY ei + firm fixed effects + type of announcement fixed effects + d ei, (2.5) where Greek letters are coefficients and d ei is the error term of this regression model. All variables as defined in appendix 2.3. We also include a binary variable for each firm as well as control variables for the type of IASB pronouncement in our equation. Second, we include OPTIONS ei which is a categorical variable taking a value of 1 (-1) if there is an increase (decrease) in the number of options or choices, 0 otherwise. 9 pronouncements are classified as number of options or choicesincreasing. We perform the regression model given in equation 2.6: CAR ei = ε 0 + ε 1 OPTIONS ei + ε 2 FIRST TIME IFRS ei + ε 3 INTERNATIONALITY ei + ε 4 LEVERAGE ei + ε 5 OWNERSHIP ei + ε 6 PROFITABILITY ei + ε 7 SIZE ei + ε 8 MOMENTUM ei + ε 9 STOCK VOLATILITY ei + firm fixed effects + type of announcement fixed effects + e ei, (2.6) where Greek letters are coefficients and e ei is the error term of this regression model. All variables as defined in appendix 2.3. We also include a binary variable for each firm as well as control variables for the type of IASB pronouncement in our equation. 34

46 Third, we include MAJOR CHANGE ei which is a binary variable taking a value of 1 if a (future) change in the IFRS is judged as fundamental, i.e., a new methodology is introduced, existing financial reporting rules are deeply modified, or if a new standard is announced. 36 pronouncements are classified as major change. We perform the regression model given in equation 2.7: CAR ei = ζ 0 + ζ 1 MAJOR CHANGE ei + ζ 2 FIRST TIME IFRS ei + ζ 2 INTERNATIONALITY ei + ζ 3 LEVERAGE ei + ζ 4 OWNERSHIP ei + ζ 5 PROFITABILITY ei + ζ 6 SIZE ei + ζ 7 MOMENTUM ei + ζ 8 STOCK VOLATILITY ei + firm fixed effects + type of announcement fixed effects + f ei, (2.7) where Greek letters are coefficients and f ei is the error term of this regression model. All variables as defined in appendix 2.3. We also include a binary variable for each firm as well as control variables for the type of IASB pronouncement in our equation. Fourth, we include DISCLOSURE ei which is a categorical variable taking a value of 1 (-1) if there is an increase (decrease) in the extent of disclosure requirements, 0 otherwise. 41 (2) pronouncements are classified as increasing (decreasing) the extent of disclosure requirements. We perform the regression model given in equation 2.8: CAR ei = η 0 + η 1 DISCLOSURE ei + η 2 FIRST TIME IFRS ei + η 2 INTERNATIONALITY ei + η 3 LEVERAGE ei + η 4 OWNERSHIP ei + η 5 PROFITABILITY ei + η 6 SIZE ei + η 7 MOMENTUM ei + η 8 STOCK VOLATILITY ei + firm fixed effects + type of announcement fixed effects + g ei, (2.8) 35

47 where Greek letters are coefficients and g ei is the error term of this regression model. All variables as defined in appendix 2.3. We also include a binary variable for each firm as well as control variables for the type of IASB pronouncement in our equation. 2.4 Identification of events, data sources, and sample distribution Identification of standard setting pronouncement events Our basic sample contains market reactions to all IASB discussion paper, exposure draft, and standard pronouncements from April 1, 2001 (the date on which the IASB assumed its duties) to December 31, We do not take amendments into account which are described as editorial changes by the IASB itself. Further, we do not rely on amendments, by way of consequence as these modifications are not caused by changes in content but by insurance of consistency between accounting standards. We take the information about the decision of publication of drafts from the staff summaries of the IASB as available on the IASB homepage and from additional explanations provided on this homepage ( One important issue of every event study is the correct identification of the event date. The releases of new standard setting pronouncements are likely to be reflected in stock prices if their content changes investors expectations about the amount and/or the uncertainty of future cash flows. Assuming sufficiently efficient stock markets we expect market reactions in short intervals around the pronouncements. To identify the event dates for our study we conduct interviews with a member of the IASB and with several firms financial accountants. According to these directly involved persons (projected) changes in financial reporting standards commonly are not noticed at the day of the IASB decision but when this information is published on wide-spread international publicly available platforms. Hence, we do not directly take the date of the IASB decision or publication of new drafts but the day this information is published on wide-spread international publicly available platforms. In detail, we use one of the most prominent homepages for news and for the development of IFRS which is provided by Deloitte ( It seems to be appropriate to take 36

48 such a lagged date (typically about one to three days after the IASB decision) as a proxy for the date information about the (expected) changes in IFRS are noticed and taken into account by capital market participants and investors in particular Data sources We collect daily stock market data (daily return indices and daily trading volumes) as well as the information about accounting principles used by the firms basically from the Datastream Worldscope database. Thereby, we follow the classification proposed by Daske et al. (2008) to identify IFRS adoptions. We assume an effective voluntary or mandatory adoption of IFRS in jurisdictions if more than 10% of the available observations are from IFRS financial statements. As previous studies found several unclear, contradicting or even erroneous entries in this database for the type of accounting principles used (e.g., Daske et al. 2008) we correct this data manually if the item contradicts the relevant commercial code. If the shares of a firm are traded on more than one stock exchange we use data from the main domestic stock exchange defined as the one with the highest absolute trading volume of the stock per year. Macroeconomic data are taken from the homepage of the World Bank. The basic sample of our study comprises all firms applying IFRS (maximum of 10,520 firms from 35 jurisdictions) with data available in the Datastream Worldscope database at any time in the investigated period (fiscal years 2001 to 2009). To construct our jurisdiction-specific synthetic benchmarks, we use all firms with available data from a maximum of 40 jurisdictions not applying IFRS in the respective calendar year (19,580 firms including 7,210 firms from the US). To calculate daily portfolio returns of several jurisdictions we require that there are at least 25% out of the maximal number of returns of a jurisdiction available. These criteria lead to a minimum of 48 jurisdictions for each year of the investigated period (with a maximum of 35 IFRS-jurisdictions in 2009). Further, we do not use abnormal daily returns which do not result from at least 25 daily observations in the estimation period (out of 100 days before the event window) of each event under consideration. We only take abnormal return volatilities 37

49 (AVOL ei ) and cumulative abnormal returns (CAR ei ) into account if there are daily abnormal returns available for all days of the event window. This requirements lead to 387,854 observations from 10,520 IFRS-firms from 35 IFRSjurisdictions. Our variables concerning the content of the IASB pronouncements are self-defined based on independent judgements of the authors. Only in the case if the authors rate the content of the pronouncement equally the information is used Sample distribution Table 2.1 to 2.3 provide an overview about the number of observations (abnormal return volatility [AVOL ei ] and cumulative abnormal returns [CAR ei ] show equal numbers of observations) resulting from all IASB pronouncements from 2001 to Table 2.1 shows the distribution of observations by jurisdiction. The table illustrates that most firms of our sample are from Australia (17.9%), from the UK (14.2%), and from Germany (9.9%). However, also smaller jurisdictions as well as developing countries are included. Table 2.2 gives an overview of the distribution of observations by industry following the 12-industries classification developed by Fama and French (2008) which bases on the SIC codes of the firms. The proportion of financials is 14.7% (56,916 observations), of business equipment firms is 10.7% (41,688 observations), and of manufacturing firms is 10.2% (39,540 observations), and The proportion of others is 7.1% (27,346 observations). For 24.5% of the observations no information about their industry classification is available in the Datastream Worldscope database. Jurisdictions considerably differ in the industry distribution, especially in the proportion of financials ( manufacturing firms ) with a maximal proportion spread of 32.0% (24.8%). This industry distribution of observations is broadly comparable to the underlying industry distribution of included firms. In total, our study includes 387,854 unique observations ,629 additional observations relate to pronouncement days at which two types of pronouncements we published. Excluding these observations from our analyses does not change the results at any stage of the study. 38

50 Table 2.1 Sample distribution by jurisdiction Total Number of obs. Prop. Austria 5, % Australia 69, % Belgium 9, % Bulgaria 2, % Chile % Cyprus 3, % Czech Republic % Denmark 7, % Estonia % Finland 7, % France 31, % Germany 38, % Greece 14, % Hungary 1, % Ireland 2, % Israel % Italy 16, % Kenia % Lithunia 1, % Luxembourg 1, % Morocco % New Zealand 3, % Norway 7, % Peru 2, % Philippines 8, % Poland 14, % Portugal 2, % Romania 1, % Russia 4, % Sweden 22, % South Africa 13, % Spain 7, % Switzerland 13, % The Netherlands 7, % United Kingdom 54, % Not jurisdiction-unique 3, % Total 387, % Notes: This table shows the distribution of observations by jurisdiction for our largest sample of 387,854 observations for our measures AVOL ei and CAR ei. 39

51 Table 2.2 Sample distribution by industry Total Number of obs. Prop. Business equipment 41, % Chemicals 8, % Consumer non-durables 25, % Consumer durables 5, % Manufacturing 39, % Energy 15, % Financials 56, % Healthcare 17, % Telecommunication 11, % Utilities 9, % Wholesale and retail 34, % Others 27, % Not classified 94, % Total 387, % Notes: This table shows the distribution of observations by industry for our largest sample of 387,854 observations for our measures AVOL ei and CAR ei. Table 2.3 shows the distribution of observations by year and type of pronouncement. 1.9% stem from the early period of the IASB (years 2001 to 2004). 31.7% are from the years 2005 to 2007, and 66.4% are from the years 2008 to ,308 observations (18.4%) relate to discussion papers, 196,057 (48.6%) relate to exposure drafts, and 133,118 (33.0%) relate to final standards. Except for 2009, the number of observations increases monotonically over the periods. Reasons are, first, the increasing data availability of the used database, second, the higher number of firms which apply IFRS (especially since the fiscal year 2005), and third, the increasing IASB activities resulting in a higher number of official standard setting pronouncements. 40

52 Table 2.3 Sample distribution by year and type of pronouncement Number of obs. Discussion paper Exposure draft Final standard Total Prop % , % ,017 1,023 2, % ,678 1,797 4, % 2001 to ,586 3,116 7, % ,408 6,679 16,643 26, % ,543 25,162 5,361 41, % ,892 42,135 11,986 60, % 2005 to ,843 73,976 33, , % ,575 47,761 47, , % ,890 69,734 48, , % 2008 to , ,495 96, , % Total 74, , , , % Prop. 18.4% 48.6% 33.0% 100.0% Notes: This table shows the yearly and total distribution of observations by type of pronouncement (discussion paper, exposure draft, and final standard) for our largest sample of 387,854 observations for our measures AVOL ei and CAR ei. Please note that there are days with multiple publications of difference types of pronouncements. Thus, the total numbers shown in this table (sum over all of 403,483 observations) do not necessarily fit to the results presented in the ongoing tables Empirical findings IFRS adoption by jurisdiction Table 2.4 shows the yearly results of the probit regression model described in section 2.3, which is the first step of our jurisdiction-level matching procedure to identify adequate benchmark jurisdictions ( Non-IFRS jurisdictions ) for those jurisdictions allowing or mandating the application of IFRS for publicly traded firms ( IFRS jurisdictions ). The proportion of IFRS jurisdictions in the yearly samples increases from 21.6% in 2001 to 70.8% in While the findings of these regressions are stable over time the explanatory power (McFadden R²s) of these probit models considerably increases from 9.5% in 2001 to 39.0% in We find limited evidence that the legal origin as well as the level of enforcement of the jurisdiction influences the likelihood of the IFRS adoption. Jurisdictions 41

53 characterized by a common law legal origin are less likely to adopt the IFRS while jurisdictions showing a higher level of financial reporting enforcement are more likely to adopt the IFRS. In line with Hope et al. (2006), we find evidence that smaller jurisdictions show a higher IFRS adoption probability. Last, jurisdictions characterized by a higher economical importance of the capital market are more likely to adopt the IFRS. 10 We use the propensity score of each jurisdiction in each calendar year to determine three benchmark jurisdictions not applying IFRS for each IFRS jurisdiction allowing or mandating the use of IFRS Information content of (expected) changes in IFRS In this study we use two different measures to evaluate market reactions to IASB standard setting pronouncements. The first measure, abnormal return volatility (AVOL ei ), indicates the information content of the pieces of information announced by the IASB (Beaver, 1968; Landsman and Maydew, 2002; Landsman et al., 2011). Figure 2.2 illustrates the distribution of AVOL ei for our largest sample. Referring to hypothesis 2.1, we find that IASB standard setting pronouncements show a highly significant information impact for investors (table 2.5). The mean AVOL ei is (t-stat. = ) and the median AVOL ei is (z-stat. = ). This means that investors take into account and react to the development and publication of financial reporting rules in general The variables CAPITAL MARKET SIZE i and CAPITAL MARKET IMPORTANCE i are positively correlated from to Thus, serious multi-collinearity concerns arise. However, these correlations do not seem to be critical at this stage of the analyses. 11 For an evaluation of the motives of an IFRS adoption see, e.g., Hope et al. (2006) and Ramanna and Sletten (2009). 12 We note that the test statistics reported in table 2.5 to table 2.12 are considerably driven by the large number of observations. Thus, we do not only take the statistical significance of the means and medians into account for our interpretations but also the absolute amounts. 42

54 Table 2.4 Macroeconomic determinants of the IFRS adoption in different jurisdictions P (IFRS = 1) Intercept (0.95) (0.86) (2.19)** (2.42)** (1.81)* (2.07)** (3.06)*** (2.87)*** (3.36)*** COMMON LAW ORIGIN (1.28) (1.50) (1.76)* (2.09)** (1.49) (1.41) (1.58) (1.67)* (2.42)** ENFORCEMENT (1.55) (1.83)* (2.37)** (2.50)** (3.40)*** (3.37)*** (3.42)*** (3.23)*** (3.87)*** CAPITAL MARKET SIZE (1.19) (1.07) (2.38)** (2.57)** (1.80)* (2.03)** (3.03)*** (2.79)*** (3.33)*** CAPITAL MARKET IMPORTANCE (1.53) (1.60) (2.67)*** (2.63)*** (2.57)** (2.70)*** (2.61)*** (2.78)*** (2.93)*** McFadden R² 9.5% 12.0% 21.1% 24.1% 27.2% 28.4% 30.2% 27.5% 39.0% Number of obs Prop. of IFRS jurisdictions 21.6% 21.6% 26.9% 30.8% 67.3% 67.3% 67.3% 69.2% 70.8% Notes: This table shows the first step of the propensity score matching procedure to identify the three most similar Non-IFRS jurisdictions for each IFRS jurisdiction for benchmark purposes as described in section All variables as defined in appendix 2.2. *, **, and *** indicate two-tailed significance at the 90%, 95%, and 99% levels, respectively. 43

55 Figure 2.2 Distribution of abnormal return volatility f(avol ei ) AVOL ei Notes: This figure shows the distribution of all abnormal return volatility (AVOL ei ) observations (387,854 observations). AVOL ei and CAR ei as defined in section 2.3, winsorized at the 1 st and 99 th percentile. Table 2.5 Overall abnormal return volatility and cumulative abnormal returns to IASB standard setting pronouncements AVOL CAR Mean Median Mean Median Number of obs. All announcements ,854 (119.95)*** (162.33)*** (12.71)*** (30.40)*** Notes: This table shows mean and median statistics for AVOL ei and CAR ei over all pronouncements. AVOL ei and CAR ei as defined in section 2.3, winsorized at the 1 st and 99 th percentile. *** indicates two-tailed significance at the 99% level. Referring to hypothesis 2.2, we evaluate whether the information content of IASB standard setting pronouncements has changed over time due to the changing role of the IASB (table 2.6). For the mean AVOL ei, we find that there is a remarkable increase of the information content of the IASB pronouncements in the periods of 44