The Effects of IFRS 8 Geographical Disclosure Changes on the Valuation of Foreign Earnings

Transcription

1 The Effects of IFRS 8 Geographical Disclosure Changes on the Valuation of Foreign Earnings An Association Study on European Companies F.W. de Graaf 1 December 2011

2 The Effects of IFRS 8 Geographical Disclosure Changes on the Valuation of Foreign Earnings An Association Study on European Companies Master thesis Department Accountancy Faculty of Economics and Business Studies Tilburg University 1 December 2011 Author: F.W. de Graaf Organization: Deloitte Supervisor: Dr. A.J.M. Verriest Second reader: P.Y.E. Leung MSc

3 Acknowledgements I would like to express my gratitude to all those who gave me the possibility to complete this thesis. First of all I want to thank Dr. Arnt Verriest for his guidance through the thesis process and his support for the accelerated completion of this thesis. I also would like to thank Deloitte for the support and the use of their facilities. I am looking forward to the 1st of March when I start my career at their office in Eindhoven. Further I would like to thank my friends for their patience, support and their advice. Especially I want to mention the Biebers who were always there for me during the writing process. My gratitude goes also to my homies from Huize Loven who gave me the opportunity to relax after a day of hard work. Special thanks go to my mother, my sister and my grandparents. They always supported me during my entire study and showed their love and gave me advice when I needed it. At last, I would like to mention my father who is unfortunately not in the condition to support me but whose strength and ability to enjoy life inspired me to complete this thesis. Thank You! 1

4 Abstract This research investigates whether the market s valuation of foreign earnings is a function of the company s geographic segment disclosure. Specifically, it is examined what the effects are of an increase in the number of the disclosed geographic segments and the inclusion of earnings measures in geographic segments disclosures following the adoption of IFRS 8. IFRS 8 is comparable with the US standard on segment reporting, SFAS 131. Prior research following the adoption of SFAS 131 found strong evidence that these proxies for increased disclosure are positively associated wi th the valuation of foreign earnings. However, the results of this research do not find evidence for the existence of this relationship following the adoption of IFRS 8. The results are controlled for other variables that might affect geographic segment disclosure and for self-selection bias due to proprietary- and agency cost motives. Keywords: foreign earnings, IFRS 8, geographic segment disclosure, valuation, value relevance 2

5 Table of Contents Acknowledgements... 1 Abstract... 2 Table of Contents Introduction Topic and research question Theoretical overview Information Asymmetry and the Agency Theory Information Defined Information Asymmetry Agency theory Standard setting and regulation Evaluating Financial Standards IFRS Background Key Changes from IAS Geographical disclosure under IFRS Geographical disclosure Hypotheses development Research Methodology Research Design Variable definitions Cumulative abnormal stock returns Earnings measures Increase in the number of disclosed geographical segments Disclosure of geographical earnings items Sample selection Testing the Relation between Earnings and Return Descriptive statistics

6 3.5.1 Part 1: Samples description Part 2: IFRS 8 changes Number of segments Segment Items Main findings from the Descriptive statistics Tests and Findings Difference-in-differences tests Testing of Hypothesis Testing of Hypothesis Additional difference-in-difference test Controlling for Other Factors that might affect the Pricing of Foreign Earnings Self-selection tests Discussion of the findings Conclusion References Appendix

7 1 Introduction 1.1 Topic and research question This thesis is inspired by a paper written by Hope, Kang, Thomas, & Vasvari,(2009) about the effects of SFAS 131 on the valuation of foreign income of US multinational firms. They were the first to establish a link between cross-sectional differences in geographic segment disclosure practices and the valuation of foreign earnings. Foreign earnings are used by the users of financial statements as an important indicator for valuing the company s foreign operations. Because of increasing globalization and the growing importance of the international operations of companies, research to the valuation of foreign earnings is very important. Geographical segment reporting is an important element in the valuation of the international operations of a company. Hope et al. (2009) used a change in the segment reporting accounting standards to examine the effects of segment disclosure on the valuation of foreign earnings by the users of financial statements. In Europe, the international accounting standards board (IASB) also issued a new accounting standard on segment reporting, International Financial Reporting Standard (IFRS) 8 Operating Segments. IFRS 8 was published on 30 November 2006 and is effective for periods beginning on or after 1 January The standard replaced International Accounting Standard (IAS) 14 Segment Reporting. The main change introduced by IFRS 8 is the switch from a risk and reward approach to the management approach, which is comparable to the changes made by the adoption of Statement of Financial Accountant Standard (SFAS) 131 in the US. SFAS 131 was used as example for the formulation of IFRS 8. Therefore achieves the standard substantial convergence with the requirements of SFAS 131. There are only minor differences between IFRS 8 and SFAS 131, most necessary to make the terminology consistent with that in other IFRSs. Because of the high level of comparability between the IFRS 8 and SFAS 131, research findings on segment reporting in the US can be used to predict possible results of the adoption of IFRS 8. Research findings indicated that application of SFAS 131 resulted in more useful segment information than its predecessor, SFAS 14. According to the research, the management approach of SFAS 131 increased the number of reported segments and provided more information, it enabled users to see an entity through the eyes of management, it enabled an entity to provide timely segment information for external interim reporting with relatively low incremental cost, it enhanced consistency with the management discussion and analysis or other annual report disclosures, and provided various measures of segment performance. 5

8 Before the adoption of IFRS 8 the IASB discussed segment reporting at several meetings with users of financial statements. Most of the users supported the management approach under SFAS 131 for the reasons that were supported by the research findings. In particular, they supported an approach that would enable more segment information to be provided in annual reports. Consequently the IASB decided to adopt the US approach and issued IFRS 8. Because of the high level of comparability between the IFRS 8 and SFAS 131 and the reason for the IASB to issue IFRS 8 because of the expected increase in segment information, an increase in segment information is expected. Based on this expectation I formulate the following research question: DOES THE ADOPTION OF IFRS 8 INCREASE THE VALUE RELEVANCE OF FOREIGN EARNINGS? To answer this research question an association study is used to check the value relevance of foreign earnings. The effects of the adoption of IFRS 8 are measured by several conducting difference-in-differences tests. These tests showed no significant effect on the valuation of foreign earnings by shareholders after the adoption of IFRS 8 by a company. Thereafter tests were conducted to test for other variables that might affect the pricing of foreign earnings and if self selection-bias due to proprietary or agency cost motives was present. These tests did not show any of this. This thesis is relevant to the academic research because it is one of the first studies that examines the effects on the pricing of foreign earnings by geographical disclosure changes due to the adoption of IFRS 8. Further does this research extent the research on segment and specifically geographical disclosure which is very limited until now. This research is necessary to provide standard setters a good understanding of the effects of the accounting standards they issue. When accounting standards do better provide in the needs of users, the quality of the information will increase and information asymmetry cost will decrease, which benefits the entire society. The paper is structured as follows. The next section provides a theoretical background for testing the effect of the changes in the geographical disclosures on the valuation of foreign earnings by shareholders. In section 3 the research method is explained by the literature in section 2. The sample description and the descriptive statistics can also be found in this section. Section 4 describes the performed tests and their results. In the last section the conclusion is presented. 6

9 2 Theoretical overview With the increase in multinational operations in recent years, it is becoming increasingly important to gather information related to the foreign operations of a company. An important source of information is geographic segment disclosures in the annual report. The issue of IFRS 8 in 2009 changed the requirements for segment disclosure and therefore the requirements for geographic segment disclosure. This section provides a theoretical background for testing the effect of the changes in the geographical disclosures on the valuation of foreign earnings by shareholders. The section is divided into five subsections. The purpose of the first subsection is to get an understanding why accounting standards are needed and why they need revision over time. This is done by explaining information asymmetry and the agency theory. Thereafter in subsection 2, the process of standard setting is explained by taking IFRS as an example. The third subsection describes the association theory that is used by the academic research to measure the usefulness of disclosure regulation. The fourth subsection is about the globalization and the need for additional information of foreign operations of a company. The last paragraph will shortly describe IFRS 8, the standards that provide information about foreign operations and its differences with its predecessor, IAS Information Asymmetry and the Agency Theory In the first paragraph a definition of information is provided. Thereafter, in the second paragraph, the information asymmetric problems between the users of information are discussed. In the third paragraph the agency theory is explained and the role of accountancy and accounting standards in solving these information problems Information Defined The need for information is at the basis of the existence of the accounting profession. Therefore information has a central role in the profession and is related to almost all the academic accounting literature. Information has many definitions, therefore a definition is provided that is used in this thesis. In this thesis, the term information is limited to new information that has the potential to affect an individual s decision. Information that confirms already known knowledge of an individual or information that is new but does not affect an individual s decision, do not fall into this definition. The definition will be as follow: Information is knowledge that affects the existing optimal decisions of an individual. 7

10 2.1.2 Information Asymmetry In a market with ideal conditions does every participant receive perfect information at the same time and do they all interpret the information in the same way. Perfect information is information that does perfectly reveal the true state of nature and therefore it is highly reliable and relevant to the receiver. In reality it appears that markets with ideal conditions do not exist. Perfect information is very scarce and therefore information is mostly a tradeoff is between reliability and relevance of the information. Besides, information does not seem to be perfect, there exists in almost all markets information asymmetry between the different participants. This means that there is a difference in the information that participants have at the same time. Information that is only known by a selected group of participants and is not observable by others is called private information. When information is available to everyone at the same time it is defined as public information. Asymmetry can therefore be defined the difference between private information and public information Agency theory The agency theory, or principal-agent problem, is a problem that arises when management and ownership of a company are separated. For instance, the principal is a shareholder (owner) and the agent is the manager (management). The agent acts on behalf of the principal. The agency problem exists, because managers (agents) have different goals and motives compared to the shareholder (principal). Shareholders are focused on creating profit and value for the company, to maximize the return on their investment. On the other hand, managers are more interested in creating value and wealth for themselves. Managers will always try to maximize their utility, but this utility does not have to be the same as the utility wanted by the shareholders (Douma and Schreuder, 2002; Jensen and Meckling, 1976; Zimmerman, 2009). Preventing the agency problem is difficult, because information asymmetry exists. The agent knows precisely what is happening within the company, while the principal watches over from outside. So, the agent has better and more accurate information about the true economic situation a company is in, thus the agent is better able to make important decisions. A solution to reduce the information asymmetry is the introduction of financial standards. Investors need private information from a company to assess the profitability of the company s investment opportunities. Since they cannot observe every action by the manager, these investors need proof of the performance of the company. Therefore a need and demand for financial reporting exists, thus financial reporting is a tool to reduce the information asymmetry. 8

11 2.2 Standard setting and regulation As explained by the previous paragraph, are accounting standards necessary to solve the problem of the conflicting information needs by investors and management. This directly points out the difficulty of standard setting. Because there are different parties with different interests, conformity has to be reached. Therefore standard setting is a process that requires negotiation and comprise. Because this master thesis relates to European companies that report by International Financial Reporting Standards (IFRS), this process is used as an example. IFRS is developed by the International Accountings Standards Board (IASB). The IASB consists out of 14 individuals that possess technical skills and suitable international business and market experience. The IASB seeks to address a demand for better-quality information that is of value to all users of financial statements. The IASB follows the following procedure for designing standards. The first step is broad consultation with interested parties for topics that has to be added to the agenda IASB meeting. The IASB s approval to add agenda items, as well as its decisions on their priority, is by a simple majority vote at an IASB meeting. After a topic is added to the agenda a workgroup is formed that evaluates the issue and develops an exposure draft of the new standard. Normally this exposure draft is preceded by a discussion paper but this not permitted. A discussion paper explains the issue and solicits early comment from constituents which can contribute to the development of the exposure draft. When an exposure draft is completed it is first to be discussed and approved by the IASB. When the IASB gives its approval, the exposure draft is published for public comment. Comments are analyzed and a revised standard is prepared. Before the IASB vote for the standard all outstanding issues should be resolved. After a majority of nine of the 14 members approves the new standard the new IFRS is issued. The IASB does not have the legal jurisdiction to issue new accounting standards in the European Union (EU). A new issued IFRS should go through a process of endorsement before becoming law in the EU. In the process of endorsement the European Financial Reporting Advisory Group (EFRAG) consults interest groups. The EFRAG delivers an endorsement advice to the European Commission (EC). Thereafter the Standards Advice Review Group (SARG) issues its opinion whether EFRAG's endorsement advice is well-balanced and objective. Based on the advice of EFRAG and the opinion of SARG, the Commission prepares a draft endorsement regulation. The Accounting Regulatory Committee (ARC) votes on the EC s proposal. If the vote is favorable the European Parliament (EP) and the Council of the European Union (CEU) have 3 months to oppose the adoption of the draft Regulation by the EC. If the EP and the CEU give their favorable opinion on 9

12 the adoption or the 3 months elapsed without opposition from their side, the EC adopts the draft and the IFRS becomes law in the EU. The procedure is comprehensive and involves a lot of input by different parties, therefore when a IFRS is adopted it can be concluded that conformity between interest groups has been reached. Because the economic environment is dynamic and changes time over time therefore accounting standards should regularly being reviewed to ensure conformity between the interest groups. Standard setting is therefore a continuous process that constantly needs review of existing accounting standards and asks for adjustments if necessary. 2.3 Evaluating Financial Standards As follows from the previous paragraphs are the interests of different users of the financial statement protected by accounting standards. Although, many interest groups are consulted in the development process, it could happen that a standard favors one group to the disadvantage of the other. This imbalance between different users of the financial statements could also arise after changes in the economic environment. Imbalances in and/ or problems with financial statements can be detected by requests for review by the users itself or due to academic research. The field of accountancy that investigates how exogenous disclosure is associated with, or related to, the change or disruption in the activities of investors who compete in capital market settings as individual, welfare-maximizing agents is called association-based disclosure (Robert E, 2001). Research in this field uses value relevance studies to evaluate the effects of financial disclosure. Value relevance studies are based on the relation between earnings and stock return and are used to investigate the empirical relation between stock market returns and accounting earnings measurements for the purpose of assessing or providing a basis of assessing those numbers use or proposed use in an accounting standard. Value-relevance studies can be divided into three categories (Holthausen & Watts, 2001): relative association studies, incremental association studies, and marginal information content studies. Relative association studies compare the association between stock market values and alternative bottom-line measures. These studies usually test for differences in the R 2 of regressions using different bottom line accounting numbers. The accounting number with the greater R 2 is described as being more value-relevant. Incremental associations studies investigate whether the accounting number of interest is helpfully explaining value or returns (over long windows) given other specified variables. That 10

13 accounting number is typically deemed to be value relevant if its estimated regression coefficient is significantly different from zero. Marginal information content studies investigate whether a particular accounting number adds to the information set available to investors. They use typically event studies (short window return studies) to determine if the release of an accounting number is associated with value changes. Price reactions are considered evidence of value relevance. For the evaluation of accounting standards is mostly a relative or an incremental association study used (Bowman, 1983; Holthausen & Watts, 2001; Robert E, 2001; Kothari, 2001). A relative association study is for example used to examine whether the association of an earnings number, calculated under a proposed standard, is more highly associated with stock market values or returns than earnings calculated under an existing accounting standard. Incremental association studies are used to test whether a disclosure of additional information aids in explaining firm value of equity returns. Association studies use normally a long return window (For example, a quarter or a year). The marginal information content study uses a short return window (for example, two- to three days). This short time window is particular suitable for research to the first market reaction to an announcement or other event but not for evaluating the effects of an accounting standard for a longer period. The length of the return window period is concerned as one of the limitations of a valuerelevance study. Empirical research by (Lev, 1989) finds that value-relevance studies have low R 2 s. The literature distinguishes three possible explanations for this problem. The first deals with the possibility of measuring error in the earnings explanation model. Brown, Hagerman, Griffin, & Zmijewski (1987) tried to control for the measurement error by inserting earnings proxies without convincing result. A second possible explanation was provided by Beaver, Lambert, & Morse (1980), they state that accounting earnings are a measure of true earnings plus an error. The third explanation allows for imperfect earnings because not all value-relevant events observed by the market will be recognized as earnings during the return period, and conversely, earnings include the effects of events observed by the market prior to the return period. These effects can be reduced by introducing financial data other than earnings. These additional variables control for the value-relevant events not recognized in the earnings of the return period. Easton, Harris, & Ohlson (1992) use another approach to increase the correlation between earnings and market returns. This is explained by the theory that earnings aggregate over periods, so that earnings are more likely to reflect the value-relevant events or disclosure and their effects 11

14 over longer periods. They find that that the correlation between earnings and market returns improves with increases in the return interval. A return period of 10 years does almost explain all of the earnings. A related solution to the problem is used by Kothari & Sloan (1992), but instead of increasing the return period, they include a leading-period. Their reasoning is that the market return includes future earnings changes, but accounting measures do not. By including a leading return period, the coefficient should increase because the earnings that were taken in the leading period are now also included in the model. Besides the technical limitations of the value-relevance approach, state Holthausen & Watts (2001) that there also exists a problem with the relevance of the use of this approach for evaluating financial disclosure. They conclude that value added from value relevance research to the standards setters is very modest. They attribute this low usefulness of value relevance research to a number of factors. First, the underlying theory of the value relevance research is not a descriptive solid theory. The Efficient Market Hypothesis (EMH) is usually the underlying theory in value relevance research. The EMH states that the arrival of new information to the market is instantaneously reflected in the contemporaneous stock price. In general, the authors argue that there is a lack of connection between theory and accounting. Second, value relevance research lacks the explanation and logic underlying its assumptions. For example, research that test whether net income or comprehensive income is more associated with firm performance is not of relevance to the standard setters because there is no financial accounting standards showing interest in this comparison between net income versus comprehensive income. Third, even if the results form a value-relevance test effectively informs us about accounting s role in providing inputs to equity investor valuation, those tests still ignore the other roles of accounting and other forces that determine accounting standard. To the extent accounting standards are shaped by other roles and forces that are not perfectly correlated with the valuation role, the value-relevance literature misses key attributes for the valuation of the accounting standards. 12

15 2.4 IFRS Background In 1997 there was an international discussion about the future direction of segment reporting. The revision of the segment reporting requirements in the US resulted in the introduction of the management approach for identification of segments as well as measurement principles and the revised standard SFAS 131. The predecessor of the IASB, the International Accounting Standards Committee (IASC), and the US FASB at that time could not agree on a common way forward. As an outcome, the US Standard SFAS 131 and the IASC standard IAS 14 differed in their basic approach. It was clear that at a certain point in time the debate had to be re-opened with the aim to converge the standards based on experience gained so far. That moment came in 2005 when the IASB and the FASB started a short-term convergence project between IFRS and US GAAP called A Roadmap for Convergence between IFRSs and US GAAP. The objective of the roadmap was to reach a conclusion about whether major differences between the two standards should be eliminated through one or more short-term standard-setting projects and, if so, complete work in those areas by One of the topics for short-term convergence included segment reporting, which was to be examined by the IASB. Following the roadmap the IASB started to compare IAS 14 with the SFAS 131. The main difference between the two standards is that the requirements of SFAS 131 are based on a through the eyes of management approach, formally known as the management approach and IAS 14 requires the disaggregation of the entity s financial statements into segments based on related products and services and on geographical areas, the industry approach. The requirements of SFAS 14, the predecessor to SFAS 131, were similar to those of IAS 14. The approach to segment disclosure in SFAS 14 was criticized for not providing information about segments based on structure of an entity s internal organization that could enhance a users ability to predict actions or reactions of management that could significantly affect the entity s future cash flow prospects (Medina, 2007). In a study carried out by the American Institute of Certified Public Accountants (AICPA) in 1994 on financial reporting was the improvement of SFAS 14 listed as its first recommendation. The report listed the following important improvements needed in segment reporting: Greater number of segments for some enterprises More information about segments Segmentation that corresponds to internal management reports 13

16 Consistency of segment information with other parts of annual report Research findings on segment reporting after the adoption SFAS 131 indicated that these improvements were achieved (Medina, 2007). Thereafter the IASB discussed segment reporting at several meetings with users of the financial statements. Most of the users supported the management approach of SFAS 131 for reasons that were supported by the research findings. In particular, they supported an approach that would enable more segment information to be provided in interim financial reports. Consequently the IASB decided to adopt the management approach and published its proposal as an exposure draft in January After considering the responses in the public consultation period the IASB issued IFRS 8 on 30 November Key Changes from IAS 14 The main difference between IFRS 8 and IAS 14 is in the identification of the segments. IFRS 8 requires segments to be identified on the basis of the management approach. This requires identification of operating segments on the basis of internal reports that are regularly reviewed by the chief operating decision maker in order to allocate resources to the segment assess its performance (IFRS 8, IN11). In contrast, IAS 14 requires identification of two sets of segments, one based on related products and services, and the other on geographical areas. IAS 14 regarded one set as primary segments and the other as secondary segments. (IFRS 8, IN 11). Other important changes are in the measurement of segment information. IFRS 8 requires the amount reported for each operating segment item to be the measure reported to the chief operating decision maker for the purposes of allocating resources to the segments and assessing its performance. IAS 14 requires segment information to be prepared in conformity with the accounting policies adopted for the preparation and presentation of the consolidated financial statements. In contrast to IAS 14, IFRS 8 does not define segment revenue, segment expense, segment result, segment assets and segment liabilities, but requires an explanation of how segment profit or loss, segment assets and segment liabilities are measured for each operating segment. As a consequence, entities will have more discretion in determining what is included in segment profit or loss under IFRS 8, limited only by their internal reporting practices (Deloitte, 2006) Geographical disclosure under IFRS 8 Under IFRS 8, companies should at least disclose an analysis of revenues and non-current assets by the country of domicile and by all foreign countries in total from which the company 14

17 derives revenues. When an individual foreign country is material the revenues and non-current assets should be disclosed separately for that country. This represents a major difference compared to the geographic segment disclosure requirements under IAS 14. Under IAS 14, companies were required to disclose geographical segment information for each material geographic area but not for each country specific. This resulted in the disclosure of major geographic areas that provided limited information. IFRS 8 reduce this problem by inserting the requirement that revenues and non-current asset information is disclosed for each material country. Only if the information per country is not available and the costs for producing this information are extensive this information has not to be reported. This change in geographical disclosure standards is comparable with the change in geographical accounting standards as we saw by the adoption of SFAS 131 in the US. Research on the adoption of SFAS 131 on geographical segment disclosure showed an increase in the number of reported geographic segments (Herrmann & Thomas, 2000; Hope et al., 2009), with more countrylevel segments and fewer regional segments (Herrmann & Thomas, 2000). On the other hand, Herrmann & Thomas (2000) also note that many firms had no change in disclosure and some actually decreased the number of reported geographic segments. The decrease in the number of reported geographic segments occurs for the most part because IFRS 8 allows companies to report non-significant countries into one single foreign segment. IFRS 8 and SFAS 131 state further that if the necessary information is not available and the cost to develop it would be excessive, that disclosure of geographical segmentation is not required. This creates the opportunity for multinational companies to do not disclose geographic segments out of proprietary cost motives (Park, 2011) Geographical disclosure As the world continues to globalize, the foreign operations of companies are becoming more important as source of information. Despite the fact that foreign operations can experience different patterns of profitability, growth, and risk that differ from those of the consolidated operations, the disclosure requirements of foreign operations under IFRS and US GAAP are very limited and after the adoption of IFRS 8 and SFAS 131 even not mandatory anymore. Several studies examined the relation between geographical disclosure and the return on stock prices. One of the earliest studies on geographical disclosure was by Boatsman, Behn, & Patz (1993). They find no evidence that foreign income disclosures are value-relevant for US multinational firms return beyond their part in total income. However, a study in the same year by Ahadiat (1993) shows that geographical segmental earnings do appear to demonstrate a superior 15

18 predictive value over consolidated earnings. Seven years later Thomas (2000) finds some support for the value relevance of geographical data with stock returns. Bodnar, Hwang, & Weintrop(2003) and Bodnar & Weintrop (1997) conducted two studies in which they consider the value relevance of geographical disclosure by dividing total earnings in a domestic and a foreign component. Their studies provide evidence that for companies, located in the US, Australia, Canada and the United Kingdom, changes in the foreign income are more positively associated with annual returns than are domestic earnings. They conclude that their findings suggest that foreign income is perceived by the market to have different economic characteristics than domestic income. A recent study by Park (2011) about the impact of SFAS 131 on the extent to which stock prices incorporate industry-wide and firm-specific components of future earnings, supports that the changes in geographical disclosure can explain that public information does not improve analysts forecasts ability to predict firm-specific performance after the adoption of SFAS 131. Based on the findings of the studies it can be concluded that geographical disclosure practices have a significant influence on the return of stock prices. Hope et al. (2009) define three possible explanations for the relation between improved geographic disclosures and the higher pricing of foreign earnings. The first explanation they found is supported by a study of Holthausen & Verrecchia (1988) that suggest that the price reaction to the release of information is negatively related to the noisiness of the information signal. If geographic disclosures provide a noisy set of information about valuation-relevant future cash flows, then price changes associated with a given amount of unexpected foreign earnings will be smaller. To the extent that improved geographic disclosures can reduce the noise in foreign earnings, the price response to unexpected foreign earnings should increase. This is also supported by research of Collins & Salatka1(993). They test the model of Holthausen & Verrecchia (1988) following the adoption of SFAS 52 by multinational companies in the US. SFAS 52 was meant to improve foreign currency accounting compared with that under SFAS 8. They find that the response to unexpected earnings increases substantively after implementation of SFAS 52, suggesting that investors perceive earnings under the new standard to be a less noisy measure of future cash flows. The second explanation is based on a paper by Leuz & Verrecchia (2000) that found that higher-quality disclosures decrease the information asymmetry component of the cost of capital, reducing the discount applied by investors to stocks for which limited information is available. Information asymmetry arises either between the firm and investors or among investors. Regarding information asymmetry between the firm and investors, Lambert, Leuz, & Verrecchia 16

19 (2007) show how poor-quality disclosure creates information risk. Investors anticipate this and demand a higher risk premium (i.e., they charge a higher cost of capital). Regarding information asymmetry among investors. Easley & O'Hara (2004) show that, in a model with informed and uninformed investors, the information risk faced by the uninformed investors is not diversifiable and therefore will be priced. The information risk is reduced with the precision of firm disclosures. Regardless of its source, if information asymmetry is especially severe for foreign operations, then the risk adjusted discount rate for foreign earnings should decrease when the information environment improves. As last possible explanation provide Hope et al. (2009) that when the amount or quality of publicly available information about a firm is low, investors must undergo the cost of gathering and processing private information. This additional cost increases investors required return. As the firm s information environment improves, investors information acquisition cost is reduced because they can now free-ride on the information that the firm produces. The more the firm discloses, the more investors free-ride. Thus an improvement in geographic segment disclosures should reduce investors private information search costs related to foreign earnings, reducing the expected return. A second interesting aspect to examine the relation between geographic disclosure, foreign earnings, and firm value relates to the reported earnings items per segment. When companies define operating segments on any basis other than geographic area, IFRS 8 requires geographic items disclosure only of revenues from external customers (IFRS8, 33a) and long-lived assets for each reported segment (IFRS8, 33b). These companies do not have to report earnings items per segment. Findings of research on the adoption SFAS 131 show that most companies define their operating segments along lines of business after the adoption of SFAS 131(Herrmann & Thomas, 2000; Street, Nichols, & Gray, 2000). Also a limited research on the adoption of IFRS 8 by Backhuijs & Camfferman (2011) suggests that a majority of firms identifies segments by line of business. Therefore it is expected that most firms do not disclose geographic earnings. As proven by earlier researches are earnings a very important explanation for the companies stock returns over the long run and a significant determinant even in the short run. Therefore it is suggested that disclosure of geographic earnings in represents higher-quality disclosure compared with non-disclosure of geographic earnings and will therefore result in the higher pricing of foreign earnings. 17

20 Non-disclosure of geographic earnings potentially hampers the firm s information environment, resulting in an increase in information asymmetry, information acquisition costs, and noisiness of reported foreign earnings. Hope, Thomas, & Winterbotham (2009) find in a test of the model that firms that discontinue disclosing geographic earnings following implementation of SFAS 131 experience a significant decline in abnormal trading volume around subsequent quarterly earnings announcements. Their results are consistent with non-disclosure of geographic earnings reducing the ability of investors to utilize or generate private information in conjunction with the public announcement of quarterly earnings, which dampens trading. Consistent with the monitoring role of geographic segment disclosures, Hope and Thomas (2008) find that managers of firms that no longer disclose geographic earnings following SFAS 131 are more likely to engage in foreign empire building. Specifically, they find that non disclosing companies, relative to companies that continue to disclose geographic earnings, experience greater expansion of foreign sales and long-lived assets, produce lower foreign profit margins, and have lower firm value. 2.5 Hypotheses development Based on the found increase in geographical segments disclosure under SFAS 131 by Herrmann & Thomas (2000) it could be expected that the same effect hold for the companies that adopted IFRS 8. In combination with the literature from above, the following hypothesis can be formulated: Hypothesis 1: AN INCREASE IN THE NUMBER OF DISCLOSED GEOGRAPHICAL SEGMENTS INCREASES THE VALUE RELEVANCE OF FOREIGN EARNINGS. Based on the assumption that geographical earnings are not or less reported after the adoption of IFRS 8, the following hypothesis is formulated: Hypothesis 2: THE DISCLOSURE OF GEOGRAPHIC EARNINGS ITEMS DOES INCREASE THE VALUE RELEVANCE OF FOREIGN EARNINGS. 18

21 3 Research Methodology In this section the research method that is used in this thesis is explained. The first paragraph describes the used tests and the motivation for using them. In the second paragraph the variables that are used in the tests are described. The third paragraph explains the way the sample is constructed. In paragraph four the supposed association between earnings and the market return is tested. Thereafter, in the last paragraph a descriptive overview of the sample is presented. 3.1 Research Design As follows from the theoretical overview is an associations study a broadly used way to examine the adoption of accounting standards. The main issue in an association study is to examine the effects of an event. This is done by comparing the period before and the period after an event. The period before the event is used to estimate the expected values of the subject of research. For example, the stock returns before the event can be used to calculate the expected stock returns after the event. The difference between the actual return and the expected return, calculated by a capital asset model, is classified as the unexpected or the abnormal return. The abnormal return represents the impact on the expected return that was not expected at the beginning of the period. The abnormal returns could possible explain the effects of the event of interest. Although the capital asset pricing model still has some uncertainties regarding the expected returns, it is a general used method to examine an event (Bowman, 1983). In this research a version of the capital asset pricing model is used that is called the market model. The market model enables an ex post separation of the realized return into expected and abnormal return in a way that is easy to apply. This method is also used in the study by Hope et al. (2009). As appears from the literature overview it is expected that the adoption of IFRS 8 causes major changes in the reporting of geographical segments. These major changes are expected to be expressed primarily in changes in the geographical disclosure practices. Following Hope et al. (2009) a difference-in-difference test is used to measure whether the pricing of foreign earnings varies with geographical disclosure practices following the adoption of IFRS 8. The foreign earnings measure represents the change in earnings from the prior to the current year. Hypotheses 1 and 2 will be tested by inserting interaction variables between foreign earnings and the change in geographical disclosure measures on a measure from firm value. The earnings response coefficients of these interaction terms represent the influence of the geographical segment disclosure change on the valuation of foreign earnings by the shareholders. 19

22 Thereafter it will be considered whether these variables have differential foreign earnings response coefficients before the implementation of IFRS 8. If foreign earning response coefficients are as different in the pre-ifrs 8 period as they are in the post-ifrs 8 period, then it is hard to argue that cross-sectional disclosure practices under IFRS 8 relate to the pricing of foreign earnings. Further, a within-firm control will be included in the tests to control whether geographic segment disclosures affect the pricing of domestic earnings. Since the geographic disclosures relate mainly to foreign operations, it would not be expected that changes in geographical disclosure policies affect the pricing of domestic earnings in higher more significant way. If it does, the geographic disclosure changes likely reflect some other firm characteristics which affect the pricing of overall pricing. Next, companies decisions on whether to change disclosure policies are affected not only by the requirements of IFRS 8, but also by voluntary decisions. These decisions relate to the trade-off between the proprietary costs of these additional disclosures and the potential valuation benefits resulting from mitigating the information asymmetry between managers and investors and/ or between different users of the annual report. Hence, the conclusions may suffer from self-selection biases. That is, firms choices to increase the number of reported geographic segments and/ or include earnings measures may be caused by the existence of proprietary or agency cost motives. And it could be these other factors, rather than higher-quality geographic segment disclosure, that lead to differences in foreign earning response coefficients. A logit model with indicators for information asymmetry on the geographical disclosure measurements will be used as test for this possible threat. 3.2 Variable definitions In this section the main variables that will be used in the models in the next section are explained. For each variable the calculations, the origin and the definition are provided. Other variables that are used in this thesis will be explained in the text or in the footnotes Cumulative abnormal stock returns To compute the abnormal stock returns the following procedure is used. First the monthly Return Index for the companies and the market 1 is extracted from Datastream. Thereafter the 1 As benchmark for the market a DataStream level one market index of Europe is used. Europe can be seen as a standalone market that interacts with the rest of the world. Its own movements are not completely equal to the movements in other markets or the entire world market. Using this European index and considering its size, 2413 European listed companies, this is considered as a good benchmark for the European market. 20

23 percentage change between the months is calculated. The percentage change is the actual return for each month. Next, an estimation is made for the market model parameters for each company by running Equation 1. The parameters of the model are estimated using ordinary least squares regression and then used to calculate the residuals. The returns for the 36 months preceding the IFRS 8 adoption fiscal year are used for the estimation. Equation 1: Where: Return on the share of company i at t Return on the market at t Company specific parameters of the market model estimated over the preceding 36 months Residual Thereafter the abnormal returns, actual return by the expected returns., for the event period are calculated by subtracting the Equation 2: The expected returns are calculated by inserting the market returns for the event period in the company specific market model. The company specific parameters estimated with Equation 1 are used in this calculation. Equation 3: Where: Expected return on the share of company i at t Return on the market at t Company specific parameters of the market model 21

24 After that the abnormal returns are calculated and organized and grouped per company. Because for every company the same positive reaction is expected the companies are not divided into different portfolios. The usual way to study performance over longer intervals is by means of cumulative abnormal stock returns, where the abnormal stock returns are aggregated from the start of the event period, as follows: Equation 4: Where: is the annual cumulative abnormal stock return to firm i over the event period. is the monthly abnormal stock return to firm i Earnings measures The data items operating income 2,OI, and international operating income 3,FOI, are used to calculate earnings changes. First, the yearly operating income and international operating income for each company are extracted from the Datastream database. Thereafter the domestic operating income, DOI, is calculated by subtracting international operating income from operating income. Subsequently the yearly change for the three earnings measures is calculated by subtracting the current, t, minus the lagged year,t-1. To improve the comparability between companies, for each earnings measure the yearly change is divided by the number of shares 4 outstanding at the end of the current fiscal year 5, t. To facilitate cross-sectional and temporal comparisons, I normalize the earnings changes per share by the company s share price 6 at the end of the current fiscal year, t. 2 Operating income represents the difference between sales and total operating expenses. Extracted from WORLDSCOPE (Mnemonic: WC01250) 3 International operating income represents operating income generated from operations in foreign countries before adjustments and eliminations. Extracted from WORLDSCOPE (Mnemonic:WC07126) 4 Number of shares for each company is extracted from the Worldscope. (Mnemonic:WC Fiscal year ends are extracted from Compustat Global-Fundamentals via WRDS (Variable name: FYR) 6 The share price is extracted from Worldscope. (Mnemonic:WP) 22