Comprehensive income : does the reporting location really matter?

Faculty of Economics and Business

Comprehensive income:

Does the reporting location really matter?

Student number: 10262555

Date: 21 June 2015

Course: MSc Accountancy & Control, variant Accounting

School: Amsterdam Business School

Faculty: Faculty of Economics and Business, University of Amsterdam First supervisor: Dhr. Wim Janssen

Statement of originality

This document is written by student Jordy Kuulkers who declares to take full responsibility for the content of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the content.

Abstract

On 15 December 2011 Accounting Standard Update 2011-05, Presentation of Comprehensive Income became effective. This update banned one of the reporting options for comprehensive income. From this date on firms can either report comprehensive income in a single statement together with net income or in a separate income statement-type format, both referred to as performance statements. In theory it does not matter where comprehensive income is reported because the same information is reported either way only the reporting location differs. However managers have concerns reporting comprehensive income in a performance

statement would cause their stock price to decrease. This paper contributes to prior literature by examining the effect of reporting location on stock price for companies listed on the S&P 100. Results show that the reporting location does not influence the stock price, thereby proving the concern of managers not to be valid.

Table of content

1. Introduction 1

2. Literature review & hypothesis development 3

2.1 (Other) comprehensive income 3

2.2 Background of SFAS No. 130, reporting comprehensive income 4

2.3 Value relevance of (other) comprehensive income 9

2.4 Motives behind reporting location choice 12

2.5 Relation between presentation format of comprehensive income 14 and stock price

3. Research Method 18 3.1 Methodology 18 3.2 Sample selection 21 4. Results 22 4.1 Descriptive statistics 22 4.2 Regression results 25 5. Conclusion 26 References 28

1. Introduction

1.1Background

The Financial Accounting Standard Board (FASB) issued Accounting Standards Update 2011-05, Presentation of Comprehensive Income on 16 June, 2011. This update is effective for fiscal periods that begin after 15 December 2011. This new update stated that all

companies are required to report comprehensive income in an income statement-type format. Therefore companies no longer have the option to report comprehensive income in the statement of changes in equity, which was possible before the Accounting Standards Update. With the new update two options remain to report other comprehensive income, which are both income statement-type formats. The first option is to report other comprehensive income at the bottom of a single statement of comprehensive income. The second option is to report comprehensive income in two separate statements split up in an income statement and a statement of comprehensive income.

As stated above, before the Accounting Standards Update companies were allowed to report other comprehensive income either in the statement of changes in equity or in an income statement-type format. It does not matter which way the firm chooses to report, because either way the same information is reported. Only the reporting location differs between the two methods. Moreover investors should not react to the reporting choice made, because rational investors should fully process all known information regardless of their location and as the same information is disclosed under the same name and no (future) cash flow effect is applicable, no different reaction of investors will be expected. However, before the Accounting Standard Update most companies acted if the reporting location did matter since they preferred to report other comprehensive income in the statement of changes in equity (Bhamornsiri & Wiggins, 2001; Pandit et al. 2006). Although the same information is reported standard setters had a preference for the income statement-type format since they viewed this method as the more transparent presentation (Bamber et al., 2010).

The FASB issued the Accounting Standards Update for two reasons. First, by reporting other comprehensive income in an income statement-type format the prominence and transparency other comprehensive income reported in the annual report will increase. Second, by limiting the presentation formats the FASB tries to facilitate the convergence between US Generally Accepted Accounting Principles (US GAAP) with International

Reporting Standards (IFRS), since the International Accounting Standards Board already made this change in 2007 (Henry, 2011).

The FASB claims reporting other comprehensive income in an income statement-type format provides comparability, consistency and transparency (Chambers, 2011). Most

managers however had concerns that reporting other comprehensive income in an income statement-type format would cause investors and other stakeholders to increase their

assessments of the volatility of the firm’s performance (Bamber et al., 2010). The researches of Hirshleifer & Teoh (2003) and Maines & McDaniel (2000) confirmed managers concerns that reporting comprehensive income in an income statement-type format would likely adversely affect investors and other stakeholders perceptions of the volatility of the firm’s performance. This could eventually hurt the stock prices of companies. To investigate whether these ex ante concerns of managers are valid, this paper will focus on the actual reaction of investors and examine the relationship between the other comprehensive income reporting location and the stock prices of companies. The research question developed to investigate this relationship is as follows: Does reporting other comprehensive income in an income statement-type format increases investor’s assessment of the firm’s risk and hurt the firm’s stock price?

To investigate this research question a linear regression is conducted on a sample which consists of 58 firms with 348 firm year observations. The results of this regression show that the interaction variable is positive, indicating that firms that report comprehensive income in a income statement-type format have a lower stock price as firms that report comprehensive income in the statement of changes in equity. However the interaction variable is not significant indicating the reporting location of comprehensive income has no influence on the stock price.

There have been a few researches on the reporting way of other comprehensive income and in the research of Bamber et al. (2010) there was a suggestion for future research to complement their study. This research will expand the research of Bamber et al. (2010) thereby contributing to the literature related to the relation between comprehensive income location and stock prices. From societal point of view this research will help managers in determining whether their ex ante concerns that reporting (other) comprehensive income in an income statement-type format will hurt their stock price are valid. Another societal

contribution is that it helps determining whether the Accounting Standards Update that abolishes the option to report comprehensive income in the statement of changes in equity is justified.

This research will proceed as follows. In section 2 background literature is discussed where more information is provided on comprehensive income, SFAS No. 130, ASU 2011-05, value relevance of comprehensive income, net income and other comprehensive income, and the relation between reporting location and stock price. Based on this literature a hypothesis is also formed in section 2. In section 3 the sample selection is provided as well as the

methodology that will be used to investigate the hypothesis stated in section 2. In section 4 the results will be presented. In chapter 5 the research will end with a conclusion.

2. Literature review & hypothesis development

This section will provide information on comprehensive income in section 2.1 as well as information regarding SFAS No. 130 and ASU 2011-5 in section 2.3. Furthermore research regarding the value relevance of comprehensive income, net income and other comprehensive will be discussed in section 2.3. In section 2.4 and section 2.5 the relation motivations

regarding the reporting location choice will be discussed as well as the relation between reporting location and stock price. Finally based on this background literature a hypothesis will be formed at the end of section 2.5.

2.1 (Other) comprehensive income

Since other comprehensive income is a part of comprehensive income it is very important to explain the concept of comprehensive income to understand the remainder of this paper. The FASB defines comprehensive income as “the change in equity of a business enterprise during a period from transactions and other events and circumstances from non-owner sources. It includes all changes in equity during a period except those resulting from investments by owners and distributions to owners” (FASB, 1997). After reading this definition it is still not clear what comprehensive income explicitly implies. However in most researches

comprehensive income is defined as the sum of net income and other comprehensive income (Brimble & Hodgson, 2005; Cahan et al., 2000; Chambers et al., 2007; Wang,

2003).

Comprehensive income is based on the ‘clean surplus relation’. This means that all changes in equity, except dividends and new equity issuances, are recognized in the statement

of comprehensive income (O’Hanlon & Pope, 1999). The following relation can be used to explain this (Ohlson, 1995).

Bt = Bt-1 + Xt - Dt

Bt = book value of equity at the end of the period Xt = earnings at the end of the period

Dt = dividends at the end of the period

Ii is required in the clean surplus relation that all non-owner changes in equity go though the income statement (Thinggaard et al., 2006). There are some items that do not go through the income statement but through the statement of changes in equity, these items are called dirty surplus items (O’Hanlon & Pope, 1999; Thinggaard et al., 2006). An example of a dirty surplus item is gains and losses on available-for-sale financial assets.

As stated above comprehensive income can be defined as the sum of net income and other comprehensive income. Net income can be calculated easily, other comprehensive income on the other hand is not a concept that can be calculated easily. Chambers et al. (2007) defined other comprehensive income using four classifications which are: (1) foreign currency translation adjustments, (2) available-for-sale marketable securities adjustments, (3) minimum required pension liability adjustments and (4) adjustments on derivative securities that qualify for cash flow or foreign currency hedge accounting treatment. With these concepts other comprehensive income can be calculated.

2.2 Background of SFAS No. 130, reporting comprehensive income

In June 1997 the FASB released accounting standard SFAS No. 130 reporting comprehensive income, which was effective for fiscal years that begin after 15 December 1997 (Pandit & Phillips, 2004). SFAS No. 130 required that all entities that were required to provide a full set of financial statements to report comprehensive income in a financial statement that was equally prominent as the other statements provided in the full set of financial statements. Organizations that are not-for-profit or have no comprehensive income besides net income are not required to report comprehensive income (Schmidt, 1999). The scope of SFAS No. 130 was limited to reporting comprehensive income and did not deal with the conceptual issues like how items of comprehensive income should be measured, when these items should be reported and what format should be used for that financial statement. SFAS No. 130 only

required that the entity displayed: (1) net income, (2) a classification of items of other

comprehensive income by their nature, (3) the total comprehensive income for the period and (4) the accumulated balance of other comprehensive income.

SFAS No. 130 allowed three alternative for reporting other comprehensive income and comprehensive income which were: (1) below the line for net income as part of a traditional income statement, this would result in a combined statement of net income and

comprehensive income. (2) In a separate statement of comprehensive income that begins with the amount of net income of the year. (3) In a statement of changes in equity. Initially the FASB proposed that comprehensive income would be reported in a performance statement as in option one and two. This showed that the FASB believed that comprehensive income was just as relevant as net income when measuring firm performance (FASB, 1997). The FASB also indicated that presenting comprehensive income in a performance statement would enhance its visibility which would benefit investors in using this information (FASB, 1997). However corporate managers were against presenting comprehensive income in a

performance statement. They disagreed with the implication that all comprehensive income items are performance related. They also claim that reporting comprehensive income in a performance statement would confuse and mislead financial statement users. Based on these arguments managers proposed to report comprehensive income in the statement of

stockholders’ equity (Maines & McDaniels, 2000). The FASB listened to the critiques of corporate managers and compromised between their original exposure draft and the opinion of corporate managers by allowing an alternative financial statement format.

The pressure to issue SFAS No. 130 came from internal as well as external sources. The external pressure came from financial statement users. Before SFAS No. 130 certain items, such as unrealized gains and losses from available-for-sale marketable securities, bypasses the income statement and were taken directly to the equity section of the financial statement. This way financial data which could be important for financial statement users to determine the value of the company was not transparent. Therefore the Association for Investment Management and Research (AIMR) rejected this practice and demanded a vehicle to address these accounting issues (Keating, 1999). The AIMR argued that it was necessary to report on comprehensive income to have useful financial reports. They also stated that all information regarding the economic activities of a company need to be reported in one place, this way financial statement users can sort out all the information and use whatever

information they need to make an informed decision about the company (Foster & Hall, 1996). This pressure from the AIMR pushed the FASB for the implementation of

comprehensive income. Besides the AIMR the Robert Morris Associates (RMA) also pushed the FASB to implement a change in the reporting of income (Keating, 1999). The objective of financial accounting is to provide information that is useful to potential investors and

creditors and other users of the financial statements to make an informed and rational

decisions about investing in the company or lending money to the company (Keating, 1999). Therefore it is important to draw attention to items of comprehensive income by reporting them according to one of the three options. This way investors and creditors can make

informed investment and credit decisions. It is also important to report comprehensive income because the unrealized gains or losses for example will most likely become realized gains and losses in the future, by disclosing them financial statement users can make predictions about possible future cash flows. Finally reporting comprehensive income is also important because financial statement users can judge management performance by using comprehensive

income. Financial statement users are able to judge management on their effectiveness of working with derivatives or use hedging (Keating, 1999). Although this information was already available before SFAS No. 130 it was very difficult to obtain this information. Before SFAS No 130 firms were not required to report on comprehensive income and its

components, however it was possible to estimate them by using balance sheet numbers. Although it was possible to obtain information about other comprehensive income it required a sophisticated understanding of accounting as well as several simplifying assumptions. These reasons stated above were important to financial statement users to pressure the FASB to issue an accounting standard that obligated companies to report comprehensive income.

An internal pressure to report on comprehensive income came from the fact that in the United Kingdom the United Kingdom Accounting Standards Board (ASB) released an

accounting standard that required companies to include a statement of total gain and losses, which is similar to comprehensive income, in their financial reports (Foster & Hall, 1996). Because of the emerging global markets and technological advancement the FASB felt the pressure to develop accounting standard that were similar to accounting standards around the world. This was necessary to increase the comparability between financial statements from different countries and give financial statement users advantages in comparing financial statements from different companies in different countries.

Another internal pressure to report on comprehensive income came from new

standards they developed themselves as a reaction to changes in technology and globalization. Examples of such standards were SFAS No.1 Disclosure of Foreign Currency Translation Information, SFAS No. 8 Accounting for the Translations of Foreign Currency Transactions.

SFAS No. 87 Employers’ Accounting for pensions and SFAS No. 115 Accounting for Certain investments in Debt and Equity Securities. These standards introduced new items such as unrealized gain and holdings from marketable securities, foreign currency translation gains and losses and minimum pension liability adjustments. However these new items were not recorded in the income statement and were therefore not transparent. As a reaction to this the FASB developed SFAS No. 130, where these unrealized gains and losses from marketable securities foreign currency translation gains and losses and minimum pension liability adjustments could be reported.

Besides motivations from financial statement users to implement SFAS No. 130, there were also arguments opposing the reporting of comprehensive income. These arguments came from corporate managers and financial statement preparers. One of the arguments they made was that transactions such as hedging and derivatives are part of an extremely volatile market thus making it difficult for managers to manage the unrealized gain and losses related to these transactions (Keating, 1999). SFAS No. 130 does not specify on how to recognize or measure comprehensive income, it only specifies location issues, making the arguments raised by corporate managers and financial statement prepares invalid. The arguments are invalid because the volatility of the market has nothing to do with the location issue. Another argument against SFAS No. 130 was related to the alternative option of reporting

comprehensive income in the statement of changes in equity. They argued that reporting comprehensive income in the statement of changes in equity did not enhance the visibility of comprehensive income. It is true that reporting comprehensive income in the statement of changes in equity is less transparent as reporting comprehensive income in an income statement-type format (Hirsh & Hopkins, 1998; Maines & McDaniels, 2000). However reporting comprehensive income in the statement of changes in equity is still better and more transparent as the previous treatment. Corporate manager and financial statement preparers also argued that reporting comprehensive income would confuse financial statement users (Keating, 1999). Financial statement users however testified during a public hearing this was not true (Keating, 1999). The final argument they made against reporting comprehensive income was that other comprehensive income items are extremely volatile and do not reflect the performance of the management and consequently should not be reported in an income statement (Keating, 1999). This argument however is invalid yet again because unrealized gains and losses on marketable securities can give information about management

performance. These items can show if management is effective in using investors’ recourses and are therefore performance related and should be included in the income statement.

When the FASB issued SFAS No. 130 they already mentioned their preference for reporting comprehensive income in either a single statement of financial performance including the components of net income and other comprehensive income or in a separate statement of comprehensive income starting with net income. However in practice corporate managers usually choose the other option that was allowed presenting comprehensive income in the statement of changes in equity, almost 80% of public companies presented comprehensive income in the statement of changes in equity (Chamber, 2011). On 16 June 2011 however the FASB issued accounting standard update ASU 2011-05. With this update companies were no longer allowed to report comprehensive income in the statement of changes in equity and only have two remaining options to present comprehensive income: (1) presentation within a single, continuous statement of comprehensive income or (2) presentation in two separate but consecutive income statements. Both of these options are referred to as performance

statements. Besides changing the location the update did not change much regarding reporting comprehensive income. The items that were included in other comprehensive income stayed the same (Henry, 2011). There were also no changes related to recognizing an item out of other comprehensive income and into net income (Henry, 2011).

By banning the option to report comprehensive income in the statement of changes in equity the FASB wanted to converge with the IASB who also only allowed reporting

comprehensive income in a performance statement (Henry, 2011). The FASB also believed that reporting comprehensive income in a performance statement would increase the

prominence of items of other comprehensive income and would make them more transparent to financial statement users (Chamber, 2011). Despite these believes the FASB never really supported these assumption with arguments. However several academic studies confirmed the believes of the FASB. For example Hirsh and Hopkins (1998) conducted an experiment with professional buy-side analysts and asked them to value a company where earnings were managed. The analysts that received comprehensive income information in a performance statement detected earnings management easier compared to analysts that received

comprehensive income information in the statement of changes in equity. This finding supported the argument that reporting comprehensive income in a performance statement would be more transparent. Another study that confirmed the believes of the FASB was the study of Maines and McDaniels (2000). They conducted an experiment using nonprofessional investors in the form of MBA students, providing them with financial reports where

comprehensive income was presented in different locations. Their findings suggest that investors give more weight to comprehensive income when it is presented in a performance

statement. These findings also support the argument of the FASB that reporting comprehensive income in a performance statement is more transparent.

Originally when proposing ASU 2011-05 the FASB only wanted to allow one option to report comprehensive income, which was the option to report comprehensive income in a single continuous statement. However just as with the issuance of SFAS No. 130 there were a lot of objection raised in comment letters responding to the exposure draft (Henry, 2011). One of the complains was that reporting comprehensive income in a single continuous statement would de-emphasize net income, this is because net income would no longer be a bottom-line item anymore but only a subtotal which could cause confusion in the capital market (Henry, 2011). People that commented on the exposure draft were also worried that a single-statement presentation would confuse investors as to what earnings numbers were used to calculate the earnings per share ratio (Henry, 2011). Finally there were concerns that items of other comprehensive income would be emphasized, however it was argued this is inappropriate because these items are most of the time noncore activities that can not be controlled by management. The objections made against this proposal were similar to the objections made against SFAS No. 130. Just as with SFAS No. 130 the FASB listened to the critiques and allowed the two options mentioned before.

2.3 Value relevance of (other) comprehensive income

Value relevance can be explained as the relation between accounting amounts/information and the market value of a company (Barth et al. 2001). Value relevance provides insight in the relevance of accounting numbers and shows whether shareholders are able to use this

information to determine the market price of a company.

Extensive research has already been conducted in relation to the value relevance of (other) comprehensive income, however the evidence is mixed. Biddle & Choi (2002)

investigated which measure of firm performance, comprehensive income, net income or clean surplus income, captures the underlying value of a company the best. Their results indicate the measure of firm performance that has the greatest value relevance for equity valuation is comprehensive income. Besides these findings the results also showed that reporting the components of other comprehensive income is more value relevant than net income. A few years later Biddle & Choi (2006) performed another research to test which measure has the most informational content and which measure has the best predictive value for net income, operating income, comprehensive income, clean surplus income and operating cash flows. For

this research the same data was used as in the previous research. Not surprisingly similar results were found as in the previous research. Their results suggest comprehensive income has the best informational value. The results related to the predictive value showed none of the income measures are the best in predicting every value.

These findings are supported by the research of Cahan et al. (2000). Cahan et al. examine whether comprehensive income is more value relevant than net income, and whether disaggregating other comprehensive income is more value relevant as reporting aggregated other comprehensive income. Their results suggest that comprehensive income is more value relevant than net income. Their findings also suggest disaggregating other comprehensive income is not more value relevant as aggregated other comprehensive income, which implies reporting aggregated other comprehensive is sufficient for investors.

Contrary to the findings discussed above, Dhaliwal et al. (1999) find evidence that net income is more value relevant as comprehensive income. Dhaliwal et al. (1999) conducted a research before SFAS No. 130 was issued to compare the value relevance of comprehensive income with net income and see which income measure is better measure for firm

performance. To test which measure is better Dhaliwal et al. (1999) conducted an association study to test the relation of comprehensive income and net income to stock returns and the market value of equity. Besides testing the value relevance of comprehensive income Dhaliwal et al. (1999) also tested the value relevance of the individual components of other comprehensive income. This test was performed by testing whether the association between the components of other comprehensive income improved the association with stock returns under SFAS No. 130. The results show that after the implementation of SFAS No. 130 net income is better in measuring firm performance compared to comprehensive income. The results of the second test Dhaliwal et al. (1999) performed, showed that only one component of other comprehensive income, which is marketable securities, increases the association with stock return.

Support for these findings was provided by O’Hanlon & Pope (1999). They examined value relevance of other comprehensive income in the United Kingdom. In the United

Kingdom reporting comprehensive income was also not required during the period investigated. To measure other comprehensive income O’Hanlon & Pope (1999) used the same method as Dhaliwal et al. (1999) to calculate other comprehensive income. Their findings provide weak evidence that other comprehensive income is not value relevant.

Similar results were found by Brimble & Hodgson (2005), who investigated Australian companies and tested whether including non-core operations in comprehensive

income was value relevant. They suggest that including non-core operations in comprehensive income confuses investors because their findings implicate that net income is more value relevant than comprehensive income. In addition their findings show that the components of other comprehensive income are not value relevant.

The papers discussed above were all conducted using data of the time period before SFAS No. 130 was issued. The following papers that will be discussed use either data after the implementation of SFAS No. 130 or both before and after the implementation.

The results of Chambers et al. (2007) however show contradictory evidence as

opposed to the research of Dhaliwal et al. (1999). Chambers et al. (2007) argue this is caused by measurement errors in the research of Dhaliwal et al. (1999). These measurement errors are caused because during the time frame investigated by Dhaliwal et al. (1999)

comprehensive income was not actually reported. Therefore Dhaliwal et al. (1999) used the annual change is the retained earnings of a firm plus dividends paid during the year to calculate comprehensive income. However Chambers et al. (2007) performed their research after the implementation of SFAS No. 130 and therefore they are able to use comprehensive income that is actually reported. To test whether other comprehensive income is value relevant Chambers et al. (2007) compare the value relevance of other comprehensive income before the implantation of SFAS No. 130 and after the implementation of SFAS No. 130. These results support the results of Dhaliwal et al. (1999) that other comprehensive income is not value relevant before the implementation of SFAS No. 130. However the results also show contradictory evidence and implicate that other comprehensive income is value relevant after the implementation of SFAS No. 130. This is caused due to the fact that after SFAS No. 130 companies were required to report comprehensive income which improved the

transparency and therefore other comprehensive income is valued by shareholders after the implementation of SFAS No. 130.

Similar to Chambers et al. (2007) Kanagaretnam et al. (2005) compare the value relevance of other comprehensive income before the implantation of SFAS No. 130 and after the implementation of SFAS No. 130. Kanagaretnam et al. (2005) also examine whether comprehensive income is more value relevant than net income using this method. Their findings indicate that net income has more value relevance and is better in predicting future income, comprehensive income and cash flows compared to comprehensive income. Besides this the findings also indicate that the components of other comprehensive income are all value relevant especially in the period after adopting SFAS No. 130. Several years later Kanagaretnam et al. (2009) performed the same test as in the previous research on Canadian

firms instead of firms in the United States. This research indicated only two components of other comprehensive income are value relevant instead of all the components. Contrary to the findings a few years ago, the findings provide evidence that comprehensive income is more value relevant as net income. These different findings indicate that the results can not be generalized to different countries.

This implication becomes clear by looking at the research of Goncharov & Hodgson (2011). In this research companies of 14 different European countries are investigated to test whether net income is more value relevant than comprehensive income. Their findings

indicate that net income is more value relevant the comprehensive income. This is contrary to the results of Kanagaretnam et al. (2009) indicating the suggestion that the results can not be generalized to different countries is correct.

2.4 Motives behind reporting location choice

As discussed above after SFAS No. 130 and before ASU 2011-05 managers had considerable discretion in how they presented comprehensive income. As a result managers had to choose one of the three options they had to present comprehensive income. Which method they choose to present comprehensive income is most likely affected by their motives. In prior literature two fundamental perspectives regarding managers’ motives are discussed, which are the opportunistic perspective and the information perspective (Riedl & Srinivasan, 2010). These perspectives give competing explanations why managers choose to present earnings information in a certain way over other possible presentation options.

According to the opportunistic perspective managers prefer to present information in a certain way. They have the intention to bias the perception of external parties regarding the firms performance in a direction management prefers. This perspective is based on the assumption that managers believe external parties face limitations when processing

accounting information and certain presentation forms have an impact on the extent to which these processing limitations shape external parties’ judgement. For example when

management reports comprehensive income in a statement of changes in equity they believe external parties pay less attention to these items compared to when comprehensive income is reported in a performance statement. So by reporting comprehensive income in the statement of changes in equity management wants to bias external parties and make use of their

limitations to influence the judgement of external parties regarding firm performance.

likely to select a presentation format that will portray firm performance in the best way possible. An example that is based on this principle is when managers use earnings

management to artificially increase firm performance, they will report comprehensive income in a statement of change in equity because this will decrease the likelihood their earnings management will be detected by investors. (Hirst & Hopkins, 1998; Hunton et al. 2006; Lee et al. 2006). Another example related to the opportunistic perspective is when managers report comprehensive income in the statement of changes in equity when they have a low job security or have equity incentives in their compensation contracts. Bamber et al. (2010) showed that manager believe that reporting comprehensive in a performance statement will increase the perceived volatility of firm performance by investors and therefore management is worried their stock price will drop. So when manager’s performance is being judged using stock prices or they have equity incentives in their performance contracts they are more likely to report comprehensive income is a statement of changes in equity. This is in accordance with the opportunistic perspective that states that management is more likely to select a presentation format that will portray firm performance in the best way possible when they pursue their own interests.

The information perspective on the other hand suggests that managers have two competing motivation when presenting information: (1) the motivation to report information in a transparent way and (2) the motivation not to disclose proprietary information because this gives companies competitive advantages. The assumption that lies underneath this perspective is the same as with the opportunistic perspective, the information perspective assumes external parties have processing limitations and accounting presentation can impact the extent to which processing limitations are able the shape the judgement of external parties. Contrary to the opportunistic perspective the informative perspective suggests managers have incentives to present information in a transparent way instead of using presentation formats to bias external parties’ judgement. Instead management tries to improve the user’ processing limitations instead of exploiting them, management believes this will lead to liquidity benefits associated with transparent disclosures (Libby & Emett, 2014). However the information perspective also states management balances incentives to report information in a transparent manner against incentives to protect proprietary information which gives them a competitive advantage. Because accounting reports can contain information about a company’s strategic decision management needs to weigh to incentives to use transparent reporting and the

incentives to protect proprietary information essential to maintain a competitive advantage. So according to the information perspective management will select a presentation format that

optimizes a trade-off between the benefits of reporting transparent and the costs related to transparent reporting.

2.5 Relation between presentation format of comprehensive income and stock price

This section will discuss the reasons why reporting comprehensive income in an income statement-type format will lead to a lower stock price. First of all general presentation attributes will be discussed and how these attributes can affect stock price. After that a connection will be made between reporting comprehensive income in an income statement-type format and an increase in volatility of firm’s performance perceived by investors and other stakeholder. Following the connection between presentation format and volatility a connection will be made between increase in volatile firm performance and more perceived risk by investors and other stakeholders. At the end a connection will be made between and increase in risk perceived by investors and other stakeholder and a decline in stock price.

Libby and Emett (2014) discuss three primary earnings presentation attributes in their research which are: (1) disaggregation, (2) location and (3) narrative attributes. Besides discussing these primary earnings attributes they suggested three mechanisms through which these presentation attributes can affect user behaviour. These mechanisms are: (1)

presentation attributes can directly affect information content, (2) Presentation attributes can indirectly affect information content through their effects on managers’ real or reporting actions and (3) Presentation attributes can affect ease or manner of processing. Their results show that disaggregation usually affects information content directly. Also shown in their research is that location is most likely to indirectly affect information content by affecting managers real of reporting actions and affect the ease of processing. Finally their results indicate that narrative attributes influence the easy of processing. Since this research focuses on the reporting location of comprehensive income and the stock price of companies the other attributes will no longer be discussed. An example of reporting location affecting managers real or reporting actions is when managers use earnings management to artificially increase their firm performance. Managers choose to report comprehensive income in the statement of changes in equity because previous research of Hirsh and Hopkins (1998) has shown that when comprehensive income is reported in the statement of changes in equity the earnings management is less likely to be detected.

In section 2.3 two perspectives were discussed that both had the same underlying assumption, managers assume they are able to influence the judgement of investor and/or

other stakeholders by presenting accounting information in a certain matter. As well as the assumption that investors and/or stakeholder have processing limitation. Managers can exploit these limitations by affecting the ease of processing by using certain reporting locations which may affect important judgements of investors regarding price creating in capital markets (Libby & Emett, 2014). Previous literature offers two fundamental perspectives with opposing views about price creation in capital markets, which are called the limited attention perspective and the market efficiency perspective (Libby & Emett, 2014).

The limited attention perspective assumes sophisticated investors are not able to set market prices equal to underlying firm values because they face arbitrage limitations (Libby & Emett, 2014). This results in the market price being a weighted average between the beliefs of

unsophisticated investors and sophisticated investors. (Libby & Emett, 2014). Consequently, when information in more difficult to process or extract this will be reflected in the stock price because unsophisticated investors are less capable of determining the underlying firm value (Bloomfield, 2002). It has been shown reporting comprehensive income in the statement of changes in equity can have the effect of increasing the difficulty the process.

Just as with the limited attention perspective the market efficiency perspective

assumes unsophisticated investors can lack the required capacity the fully process accounting information and determine the underlying firm values. However, unlike the limited attention perspective, the market efficiency perspective assumes market price will be equal to the underlying firm value because sophisticated investors compete with each other and drive market prices equal to underlying firm value (Libby & Emett, 2014). According to this perspective earnings attributes that affect information content can affect price, however earnings attributes that only affects ease of processing will not affect price (Libby & Emett, 2014). Consequently according to the market efficiency perspective reporting location will only affect market prices if it affects managers real or reporting action but not when it only affects easy of processing. This is because sophisticated investors will neutralize the judgement from unsophisticated investors that are not correct.

One of the main concerns of managers when presenting comprehensive income in an income statement-type format is that their firm’s performance will be viewed as more volatile by investors and other stakeholders. This is because other comprehensive income is more prominent when reported in an income statement-type format. The research of Bamber et al. (2010) shows that other comprehensive income is more volatile. This is because unrealized gains on derivative securities for instance, which are part of other comprehensive income, are influenced by uncontrollable and volatile market forces. Since other comprehensive income

components are influenced by market forces they are also less persistent. Comprehensive income, which consists of net income and other comprehensive income, is therefore also more volatile. When comprehensive income is reported in the statement of changes in equity it is less prominent, this is confirmed by Hirst & Hopkins (1998). They conducted an experiment with experienced and sophisticated financial analysts and show that when comprehensive income is reported in the statement of changes in equity comprehensive income is less prominent. They have proven this since half of the participants did not even see the term comprehensive income when reported income in the statement of changes in equity. When comprehensive income is reported in a more prominent manner, investors and other

stakeholder will perceive the firm’s performance as more volatile (Hirshleifer & Teoh, 2003). They argue that investors and other stakeholder perceive the firm’s performance as more volatile because they absorb more information when is it more visible for them. Their

argument suggests that when comprehensive income is reported in a more prominent manner investors put more weight on comprehensive income as a performance measure. Their argument also suggests that investors and other stakeholders are not capable to consider offsetting unrealized gains and losses on other assets and liabilities that are not recognized in the current accounting model. Both effects will result in investors and other stakeholders perceiving the firm’s performance as more volatile. A similar conclusion is reached by Maines & McDaniel (2000) who conducted a laboratory experiment. Their findings showed that nonprofessional investors only reflect the volatility of comprehensive income, when judging corporate and management performance, when it is reported in an income statement-type format.

Now I will hypothesize why an increase in perceived volatility by investors and other stakeholder results in believing there is a greater risk involved in trading these stocks. When proposing the mandatory reporting of comprehensive income in an income statement-type format, the FASB received over twice the average number of comment letters compared to earlier proposed accounting standards (Yen et al., 2007). Many of the comment letters stated concerns about the increase in perceived risk in firm performance by investors (Yen et al., 2007). Further, Farrelly et al. (1985) investigated whether financial reports had effects on the perceived risk by investors and other stakeholder. Their research showed that 79 per cent of the variation in risk where explained by the variables they researched, one of the main variables in this research, variability in earnings, showed to be one of the most correlated variables with the respondent’s risk rating. This implies that when comprehensive income is reported in the more prominent income statement-type format, investors and other stakeholder

perceive the firm’s performance to be more volatile which leads to an increase in perceived risk. These findings are supported by several other researches. Lipe (1998) for instance finds that (nonprofessional) investors perceive firms with higher volatility in earnings as firms with higher risks. Similar results are found by Koonce et al. (2005) who find that investors and other stakeholders perceive items that can not be controlled by the company as increasing risk. Lastly, Graham et al. (2005) also shows similar results. Graham et al. (2005) conducted a survey research, their findings suggest that managers believe that investors and other

stakeholders perceive more volatile performance measures as indicating higher company risk, even when the (future) cash flows are constant.

Finally the connection between an increase in perceived risk and/or more volatility and a lower stock price will be elaborated upon. As indicated before Graham et al (2005) suggest that managers believe that investors and other stakeholders perceive more volatile

performance measures as indicating higher company risk and indicate that this higher perceived risk can hurt the firm’s stock price. This is confirmed by Gebhardt et al. (2001), who discovered a relation between the earnings forecast by financial analysts and the firm’s implied cost of capital. These findings indicate that variability in earnings affects the cost of capital and therefore also the firm’s stock price. Another factor that can influence the stock price is information risk, which can be measured by earnings quality. When earnings have a lot of volatility they can be perceived to be of poor quality which can have negative stock price effects (Aboody et al., 2005). Francis et al. (2004) also confirm the concerns of managers that increased volatility indicate higher company risk which could hurt the stock price. Francis et al. (2004) investigate the relation between several attributes of earnings and the cost of equity capital. Earnings attributes are supposed to reduce information risk and as a result lower the cost of capital. Their findings suggest that firms generally have greater costs of equity capital when they have the least favourable values of each attribute compared to firms with the most favourable values. Two of the variables that have the strongest effect on the cost of equity capital are earnings persistence and earnings smoothness. This implies that firms with a smoother earnings path have lower costs of capital which results in higher return and therefore higher stock prices. Several other researches also find that investors demand higher risk premiums when they perceive earnings to be more volatile, because the perceived risk in expected returns increases due to the uncertainty of the return which in turn hurts the stock prices (Klein & Bawa, 1976; Jorion, 1985; Xia, 2001).

Based on the review discussed above it is expected that firms that report comprehensive income in a performance statement have a lower stock price. Hypothesis 1 summarizes this expectation.

H1: Firms that report comprehensive income in an income statement-type format have a lower stock price.

3. Research method

In this paragraph the methodology of the conducted research will be discussed. The first subsection the regression model will be explained as well as the variables of the regression model. In the second paragraph the sample selection will be discussed.

3.1 Methodology

For this research an empirical archival research method will be used. Otherwise known as a database research. The hypothesis will be answered using the following regression model:

RET = β0 + β1CHOICE + β2ASU + β3CHOICE * ASU + β4CI + β5NI + β6OCI + β7AFSSEC + β8PENSION +

β9FORCUR + β10CASHFL + β11VOLATILITY + β12LEVERAGE + β13 LOGSIZE

In table 1 the definitions of the variables used in this research are shown. The regression model is based on the model used by Bamber et al. (2010). However there are some

differences with the model used by Bamber et al. (2010). In this research reporting location choice is no longer the dependent variable but the independent variable instead. The

dependent variable that will be used is stock return. The dependent variable will be measured by using the traditional way as used in finance and also by Dechow (1994), which is the difference between the stock price relative to last year plus the dividend payout divided by last year’s stock price. Besides changing the dependent two extra independent variables are added into the regression, which is ASU and an interaction variable of CHOICE times ASU. This variable will indicate whether the fiscal year investigated is before or after ASU 2011-05 was issued. Besides these two independent variables, an interaction variable is incorporated in the model.

In addition to these primary variable used in the regression model to test the hypothesis, a number of control variables are used. First three extra control variables are discussed. Comprehensive income scaled by total assets (CI) is used as a control variable because previous research by Kanagaretnam et al. (2009) found evidence that comprehensive income is more value relevant that net income. Net income scaled by total assets (NI) is used as a control variable because Dhaliwal et al. (1999) and Goncharov & Hodgson (2011) found contradictory evidence which showed net income is more value relevant as comprehensive income. Finally other comprehensive income scaled by total assets is used as a control variable since Chambers et al. (2007) and Kanagaretnam et al. (2009) have shown that other comprehensive income is value relevant.

Besides the aggregated numbers of other comprehensive income, control variables for the magnitudes of each component of other comprehensive income are included in the

regression model. These variables are denoted as AFSSEC, PENSION, FORCUR and CASHFL. These control variables are added because in comment letters on SFAS No. 130 investors expressed concerns about the volatility of these items of other comprehensive income. An above- versus below-the-median method is used because managers are able to predict whether they are exposed to each item of other comprehensive income, but managers are not able to determine the ex-post effects of each items of other comprehensive income (Bamber et al. 2010). Since managers are able to predict whether their firm is exposed to the unrealized gains and losses of the items of other comprehensive income this may affect the reporting location decision (Bamber et al. 2010). However managers are not able to predict the magnitude of these unrealized gains and losses therefore an above- versus below-the-median method is used.

VOLATILITY is used as a control variable to control for the relative volatility of comprehensive income to income. A positive coefficient is expected, because when comprehensive income is relatively more volatile than net income firms are more likely to avoid reporting comprehensive income in a performance statement which according to the hypothesis should lead to a higher stock price. Aside from VOLATILITY Bamber et al. (2010) use LEVERAGE as control variable. Graham et al. (2005) provide evidence that managers of firms with a higher leverage prefer a smooth earnings path to minimize perceived risk. Consequently more leveraged firms are less likely to report comprehensive income in a performance statement which is expected to influence stock price. Finally a control variable is added to control for firm size which is also done by Bamber et al. (2010) and Lee et al.

To correct for extreme variables that can arise from database errors or scaling the variables RET, CI, NI, OCI, VOLATILITY, LEVERAGE and LOGSIZE have been winsorized at 1% and 95%. This action is performed to control for extreme observations that may contaminate the sample.

Table 1. Variable explanation

Variable Explanation

AFSSEC 1 if the absolute value of unrealized gains or losses from available-for-sales securities

scaled by total assets in the comprehensive income year exceeds the sample median and 0 otherwise, manually collected

ASU 1 if the fiscal year is after ASU 2011-05 and 0 otherwise

CASHFL 1 if the absolute value of unrealized gains or losses from cash flow hedges scaled by

total assets in the comprehensive income year exceeds the median sample and 0 otherwise, manually collected

CHOICE Reporting location choice, 1 if comprehensive income is reported in a statement of

changes in equity before the issuance of ASU 2011-5 and 0 for firms that report comprehensive income in a performance statement before the issuance of ASU 2011-5, manually collected

CI Total comprehensive income for the comprehensive income year scaled by total

assets, manually collected

FORCUR 1 if the absolute value of foreign currency translation scaled by total assets in the

comprehensive income year exceeds the sample median and 0 otherwise, manually collected

LEVERAGE Total non-current liabilities divided by total assets, manually collected

LOGSIZE Log of the common shares of the company, collected from Compustat

NI Net income scaled by the total assets of the company for the comprehensive income

year, manually collected

OCI Other comprehensive income scaled by total assets, manually collected

PENSION 1 if the absolute value of unrealized gains or losses of changes in the minimum

pension obligation scaled by total assets in the comprehensive income year exceeds the sample median and 0 otherwise, manually collected

RET Stock return of the company for the comprehensive income year, collected from

CRSP

VOLATILITY Volatility of other comprehensive income, calculated by dividing the standard

deviation of comprehensive income scaled by total assets with the standard deviation of net income scaled by total assets, measured over the comprehensive income year

3.2 Sample selection

The sample that will be used for this research consists of U.S. listed companies since they are bound to follow U.S. accounting regulations imposed by the FASB. The sample will consist of firms who are listed in the S&P 100 index. The S&P 100 covers approximately 45% of the market capitalization of U.S. equity markets. This provides a representative sample for U.S. companies as it represents a nearly half of the market capitalization of U.S. equity markets. The data that will be used consists of data from fiscal years 2009 until 2014. This time frame is explicitly chosen for two reasons. First of all, this time frame is chosen because ASU 2011-05 was issued in 2011 and became effective for fiscal periods beginning after 15 December 2011. Secondly, by using the time frame 2009-2014 there are an equal amount of fiscal years before ASU 2011-05 became affective as the amount of fiscal years after ASU 2011-05 became effective.

Accounting data and reporting location information is either collected from Compustat or hand collected from 10-K forms. Information about the market returns was retrieved from the CRSP database. An initial sample of 100 firms was obtained from the S&P 100. First of all, 11 firms were dropped because they did not provide financial statements covering the whole time frame. Secondly, 24 firms were dropped because they did not provide sufficient information in their financial statements concerning the components of other comprehensive income. Thirdly, 4 firms were dropped because they had more than one type of share. Lastly, 3 firms were dropped because information about their stock returns could not be extracted, because their fiscal year exceeded the time frame available in the CRSP database. Resulting in 58 firms included in the sample which consist of 348 firm-year observations. The sample composition is summarized in table 2.

Table 2. Sample selection

All firms listed on the S&P 100 100

Less: Companies that do not provide financial statements that covering the time frame 11 Less: Companies that do not provide sufficient information in their financial statements 24

Less: Companies with more than 1 type of share 4

Less: Companies with no information on stock return 3

4. Results

In this section the descriptive statistics will be reported as well as discussed. The chapter will also outline the results of the regression analysis to examine whether companies that report comprehensive income in a performance statement have a lower stock price. A conclusion is made on the hypothesis outlined in chapter 2, as well as relations/contradictions with previous literature that is discussed in chapter 2.

4.1 Descriptive statistics

Table 3 reports descriptive statistics for the independent and control variables. In panel A, the composition of the reporting location described. Panel A shows that 63.8 per cent of the companies reported comprehensive income in a statement of changes in equity before ASU 2011-05 was issued and consequently 36.2 per cent of the companies reported comprehensive income in a performance statement before ASU 2011-05 was issued. These results were unexpected, because Bamber et al. (2010) finds that only 19 per cent report comprehensive income in a performance statement. The mean of RET is 0.200 indicating the average stock return for the 58 companies is 20%. The mean of VOLATILITY of this sample is 1.576 which means that on average the variance in comprehensive income is one and a half times bigger than net income.

Table 4 shows the results of the Spearman correlation. Stock return (RET) is significantly related with comprehensive income and other comprehensive income. This result is expected because as indicated in section 2.3 Biddle & Choi (2002), Cahan et al. (2000) and

Kanagaretnam et al. (2009) showed in their researches that comprehensive income is value relevant. The same was done for other comprehensive income by Chambers et al. (2007) and Kanagaretnam et al. (2005). The significant negative correlation between stock return (RET) and company size (LOGSIZE) on the other hand was unexpected because in general bigger companies have higher returns. The correlation between comprehensive income (CI) and net income (NI) is high (0.922), this makes sense because comprehensive income consists of the sum of net income and other comprehensive income and net income determines a large part of comprehensive income. Therefore the small correlation between comprehensive income (CI) and other comprehensive income (OCI) is no surprise because other comprehensive income (OCI) determines a lesser part of comprehensive income (CI) compared to net income (NI).

Table 3. Descriptive statistics for independent variables

Panel A: Sample composition

Reporting choice n Percentage

Statement of changes in equity 37 63.8%

Performance statement 21 36.2%

58 100%

Panel B: Summary statistics

Variables Mean Standard deviation Lower quartile Median Upper quartile

RET 0.200 0.290 0.016 0.163 0.343 CHOICE 0.638 0.481 0 1 1 ASU 0.486 0.501 0 0 1 CHOICE * ASU 0.305 0.461 0 0 1 CI 0.074 0.058 0.030 0.067 0.107 NI 0.076 0.056 0.035 0.071 0.108 OCI -0.003 0.019 -0.006 -0.001 0.003 AFSSEC 0.353 0.479 0 0 1 PENSION 0.402 0.491 0 0 1 FORCUR 0.431 0.496 0 0 1 CASHFL 0.394 0.489 0 0 1 VOLATILITY 1.576 1.096 0.962 1.033 1.864 LEVERAGE 0.383 0.180 0.248 0.357 0.518 LOGSIZE 8.027 1.423 7 8.248 9.197

Table 4. Correlation among variables****

Variable RET CHOICE ASU CHOICE * ASU

CI NI OCI AFSSEC PENSION FORCUR CASHFL VOLATILITY LEVERAGE LOGSIZE

RET 1.000 CHOICE 0.039 (0.471) 1.000 ASU 0.097*** (0.071) -0.022 (0.687) 1.000 CHOICE * ASU 0.102*** (0.058) 0.499* (0.000) 0.681* (0.000) 1.000 CI 0.094*** (0.078) 0.152* (0.005) -0.122** (0.023) 0.003 (0.961) 1.000 NI 0.010 (0.849) 0.185* (0.001) -0.111** (0.038) 0.023 (0.663) 0.922* (0.000) 1.000 OCI 0.234* (0.000) -0.071 (0.181) -0.050 (0.354) -0.078 (0.148) 0.229* (0.000) -0.072 (0.183) 1.000 AFSSEC 0.053 (0.322) -0.081 (0.132) -0.081 (0.132) -0.111** (0.039) 0.075 (0.162) 0.055 (0.307) 0.112** (0.036) 1.000 PENSION -0.041 (0.446) 0.082 (0.129) 0.129** (0.016) 0.132** (0.014) 0.058 (0.283) 0.136** (0.011) -0.220* (0.000) -0.129 (0.017) 1.000 FORCUR -0.069 (0.202) 0.149* (0.006) -0.056 (0.295) 0.029 (0.588) 0.103*** (0.054) 0.162* (0.002) -0.130** (0.015) 0.024 (0.655) 0.268* (0.000) 1.000 CASHFL -0.046 (0.389) 0.203* (0.000) -0.041 (0.440) 0.080 (0.136) 0.231* (0.000) 0.287* (0.000) -0.078 (0.147) 0.081 (0.132) 0.191* (0.000) 0.261* (0.000) 1.000 VOLATILITY -0.061 (0.250) 0.338* (0.000) 0.001 (0.987) 0.177* (0.000) 0.092*** (0.085) 0.168* (0.002) -0.133** (0.013) -0.140 (0.009) 0.368* (0.000) 0.219* (0.000) 0.315* (0.000) 1.000 LEVERAGE -0.039 (0.468) -0.190* (0.000) 0.063 (0.238) -0.045 (0.402) -0.509* (0.000) -0.490* (0.000) -0.032 (0.549) -0.011 (0.836) 0.145* (0.007) -0.200* (0.000) -0.205* (0.000) -0.065 (0.224) 1.000 LOGSIZE -0.09*** (0.078) 0.002 (0.975) 0.012 (0.827) 0.004 (0.945) -0.046 (0.3930) -0.016 (0.763) -0.060 (0.263) -0.035 (0.512) 0.417* (0.000) -0.028 (0.611) -0.059 (0.611) 0.1458* (0.006) 0.181* (0.001) 1.000

4.2 Regression results

In table 5 the results of the regression are presented. The coefficient on comprehensive income is 4.185 which indicates that if comprehensive income increases by 1 the stock return will increase by 4.185. Comprehensive income is statistically significant (P-value 0.000) On the other hand the coefficient of net income is -4.175 which indicates that if net income increases by 1 the stock return will decrease by 4.175. This result is rather surprising because commonly if net income increases this is valued positively by investors and stock return should increase. Just as with comprehensive income net income is also statistically significant (P-value 0.000). The same observation can be made regarding other comprehensive income with a coefficient of -1.642. The expectation was that stock returns increase when other comprehensive income increases. This observation however can maybe be explained by the fact that other comprehensive income is relatively volatile and therefore stock return

decreases when other comprehensive income increases. Only other comprehensive income in not statistically significant compared to comprehensive income and net income. The

interaction variable has a positive coefficient indicating that firms that reported

comprehensive income in the statement of changes in equity have a higher stock return. However this interaction variable is not statistically significant (P-value of 0.554) and therefore it can be concluded that firms that report comprehensive income in a statement of changes in equity have no different stock returns than companies that report comprehensive income in a performance statement. Consequently the findings do not support hypothesis 1. These findings implicate the ex-ante concerns raised by managers are not valid and the FASB is justified in issuing ASU 2011-05. Most of the results are not significant which may

contribute to the fact hypothesis 1 is not supported. The fact that most results are not significant might also explain the negative coefficient of net income.

Table 5. Linear regression results (Dependent variable = RET)

Coefficient Standard deviation t-statistic P-value

CHOICE 0.013 0.046 0.28 0.778 ASU 0.001 0.041 0.23 0.815 CHOICE * ASU 0.038 0.064 0.6 0.554 CI 4.185 0.958 4.37 0.000* NI -4.175 0.983 -4.25 0.000* OCI -1.642 1.376 -1.19 0.238 AFSSEC 0.025 0.032 0.78 0.439 PENSION 0.020 0.034 0.58 0.563 FORCUR -0.022 0.032 -0.68 0.496 CASH -0.015 0.028 -0.53 0.597 VOLATILITY -0.005 0.010 -0.5 0.618 LEVERAGE -0.059 0.090 -0.66 0.514 LOGSIZE -0.023 0.010 -2.19 0.033

Notes: * significant at 1 per cent level; ** significant at 5 per cent level; *** significant at 10 per cent level

5. Conclusion

Since 2012 the FASB restricted the choices regarding reporting of comprehensive income. Instead of three choices namely, (1) reporting comprehensive in a single statement combined with net income, (2) reporting comprehensive income in a separate income-type statement format beginning with net income (3) reporting comprehensive income in the statement of equity, the FASB only allowed the first two options as of 2012. Although this choice should not make any difference because the same information is reported either way, managers were concerned that reporting comprehensive income in an income statement-type format would likely adversely affect investors and other stakeholder’s perceptions of the volatility of the firm’s performance and eventually hurt their firm’s stock price. They expressed these concerns in the comment letters regarding ASU 2011-05. Proponents of reporting comprehensive income on the other hand, which include the FASB, argue that reporting comprehensive income in a performance statement provides comparability, consistency and transparency.

This paper adds to the discussion about reporting comprehensive income in a performance statement by trying to clarify whether reporting comprehensive income in a performance statement leads to lower stock prices. The research sample used to answer this question consisted of firms of the S&P 100. After dropping several firms because of various reasons the final sample consisted of 58 firms which contained 348 firm year observations. One of the findings was that 63.8 per cent of the firms included in the sample reported

comprehensive income in the statement of changes in equity before ASU 2011-05 was issued. This result was surprising because previous research conducted by Bamber et al. (2010) showed a higher percentage of firms reporting comprehensive income in the statement of changes in equity. The results do not support the hypothesis I made. The regression does indicate that firms who reported comprehensive income in the statement of changes in equity have a higher stock price. However the results were not statistically significant. Resulting in an answer on the research question that reporting comprehensive income in an income statement-type format does not lead to lower stock prices.

The contribution made by this research is that is expands the research of Bamber et al. (2010) who indicated future research should investigate the concerns of managers regarding the effect of comprehensive income reporting location on stock price. The results showed these concerns are not supported and therefore the accounting standard update issued by the FASB is justified.

A limitation of this research was that the sample was relatively small. It consisted of only 58 firms which resulted in 348 firm year observations. This could have possibly affected the results of the research. Future research could replicate this research with a larger sample to investigate whether this will yield different results. Future research could also expand this research by examining whether reporting comprehensive income in a single statement will lead lower stock prices compared to firms reporting comprehensive income in a separate income-type statement. This research could be interesting because in their original proposal of ASU 2011-5 the FASB only allowed reporting comprehensive income in a single statement. However due to concerns expressed by managers in the comment letters that reporting comprehensive income in a single statement would lead to lower stock prices the FASB allowed both reporting income statement-type reporting options.