ESSAYS ON SEGMENT REPORTING AND VALUATION

(1)

Peter Aleksziev ESSAYS ON SEGMENT REPORTING AND VALUATION

ISBN 978-91-7731-155-3

DOCTORAL DISSERTATION IN BUSINESS ADMINISTRATION STOCKHOLM SCHOOL OF ECONOMICS, SWEDEN 2019

ESSAYS ON SEGMENT REPORTING AND VALUATION

This thesis consists of three papers in the area of segment reporting and valuation.

Paper I: The role of segment reporting in corporate valuation

Paper II: On the usefulness of segment information for earnings forecasting Paper III: Heterogeneous investor beliefs and value creation through equity carve-outs

PÉTER ALEXANDER ALEKSZIEV holds a BSc in Finance and Accounting from the University of Szeged, Hungary, and an MSc in Accounting and Financial Management from the Stockholm School of Economics. Between 2014 and 2019 he was a PhD student at the Department of Accounting at SSE. He is currently teaching accounting- and finance-related courses at the Bachelor, Master, and Executive MBA levels at SSE and the University of Chicago Booth School of Business.

(2)

Peter Aleksziev ESSAYS ON SEGMENT REPORTING AND VALUATION

ISBN 978-91-7731-155-3

DOCTORAL DISSERTATION IN BUSINESS ADMINISTRATION STOCKHOLM SCHOOL OF ECONOMICS, SWEDEN 2019

Peter Aleksziev

ESSAYS ON SEGMENT REPORTING AND VALUATION

This thesis consists of three papers in the area of segment reporting and valuation.

Paper I: The role of segment reporting in corporate valuation

Paper II: On the usefulness of segment information for earnings forecasting Paper III: Heterogeneous investor beliefs and value creation through equity carve-outs

PÉTER ALEXANDER ALEKSZIEV holds a BSc in Finance and Accounting from the University of Szeged, Hungary, and an MSc in Accounting and Financial Management from the Stockholm School of Economics. Between 2014 and 2019 he was a PhD student at the Department of Accounting at SSE. He is currently teaching accounting- and finance-related courses at the Bachelor, Master, and Executive MBA levels at SSE and the University of Chicago Booth School of Business.

(3)

Essays on Segment Reporting and Valuation

Péter Alexander Aleksziev

Akademisk avhandling

som för avläggande av ekonomie doktorsexamen vid Handelshögskolan i Stockholm

framläggs för offentlig granskning fredagen den 29 november 2019, kl 13.15,

sal Ragnar, Handelshögskolan, Sveavägen 65, Stockholm

(4)

Essays on segment reporting and valuation

(5)

(6)

Essays on segment reporting and valuation

Peter Aleksziev

(7)

in Business Administration

Stockholm School of Economics, 2019

Essays on segment reporting and valuation SSE and Peter Alexander Aleksziev, 2019c ISBN 978-91-7731-155-3 (printed)

ISBN 978-91-7731-156-0 (pdf)

This book was typeset by the author using L^ATEX.

Front cover photo: c Peter Alexander Aleksziev Printed by:

BrandFactory, Gothenburg, 2019 Keywords:

segment reporting, disclosure, valuation, earnings forecasting, heterogeneous beliefs, diversification discount, equity carve-out.

(8)

To my family

(9)

(10)

Foreword

This volume is the result of a research project carried out at the Department of Accounting at the Stockholm School of Economics (SSE).

The volume is submitted as a doctoral thesis at SSE. In keeping with the policies of SSE, the author has been entirely free to conduct and present his research in the manner of his choosing as an expression of his own ideas.

SSE is grateful to Bankforskningsinstitutet, the Swedish Bank Research Foundation for financing this project for three years. SSE is also grateful for the financial support provided by Louis Fraenckels Stipendiefond, Friedlän- der Stipendiefond and C. F. Liljevalch J:rs Donationsfond which has made it possible to carry out the project.

We are also grateful for the support provided by Jan Wallanders och Tom Hedelius Stiftelse and Torsten Söderbergs Stiftelse to FIRE, the Research School in Accounting.

Göran Lindqvist Kalle Kraus

Director of Research Professor and Head of the Stockholm School of Economics Department of Accounting

Stockholm School of Economics

(11)

(12)

Acknowledgements

I would like to thank everyone who supported and encouraged me during these years. Without you, this dissertation would not exist. However, first I would like to thank Bankforskningsinstitutet, the Swedish Bank Research Founda- tion, for financing this project for three years.

I would like to thank my main supervisor, Kenth Skogsvik, for his tremen- dous help during these five years. You have taught me how to conduct research with integrity. You urged me to look at the big picture and understand it before starting to ask research questions. You were also kind to me during those years and went out of your way to help me, even corrected the grammatical and stylistic mistakes in the countless versions of this document. Thank you!

I would also like to thank my advisory committee members Niclas Hell- man, Rickard Sandberg and Vasiliki Athanasakou. Niclas, I could not have imagined a better advisor on financial reporting issues. You often answered questions I did not even know existed. I would like to thank Vasiliki for guid- ing me in the development of publishable research questions and helping me in improving the papers. Rickard, thank you for helping me with the econo- metrics issues in these papers and ensuring the appropriateness of my method- ological approaches.

Many other people contributed to the development of this thesis. I would like to thank Martin Walker for his guidance during the early years of this project. I also greatly benefited from the detailed discussions and on-the-spot suggestions by Henrik Nilsson and Sebastian Tideman. Earlier versions of the papers in this dissertation also benefited from the comments by Mattias Ham- berg, Erik Johannesson, Joachim Landström, Jan Marton and Emma Myl- lymäki.

Special thanks go to the people that challenged and encouraged me during the years. I am grateful for the always skeptical comments from Milda Tylait˙e;

your no-nonsense approach helped me make my arguments more convincing.

(13)

Ting Dong was always ready to help me, be it to offering comments on my papers or to sharing bits of wisdom about life. I really appreciate the support of Hanna Setterberg during my whole time at SSE. Tack!

Thank you Florian Eugster for always trying to find some time for my questions despite your always busy schedule. I had great evening discussions about financial analysis with my fellow night owl, Ingolf Kloppenburg. I have also appreciated the interesting discussions with the ‘quant’ Brown Bag seminar series members: Noor Alshamma, Liwei Zhu, Mariya Ivanova, and Antonio Vazquez.

I would also like to thank the Data Center at the Swedish House of Finance for providing access to the SDC Platinum database.

An important part of my PhD years was my development as a teacher.

Thanks go to Erik Alenius, who supported me in the beginning of this journey, and to Johnny Lind, who challenged me to start teaching in Swedish early on.

Tomas Hjelström, you had the most skin in this game. I really appreciate you taking me on as a co-teacher. I have learnt so much from you about how to be a better teacher. Per Strömberg was the first to trust me with executive teaching.

Thank you! It was a pleasure to teach with Gustav Johed, Catharina Pramhäll and Katarina Warg at different points in time.

Discussions with FIRE Workshop participants broadened my perspective in accounting. Thank you for all the insights, Ebba, Emilia, Kai, Kalle, Lukas, Martin, Patrik and Torkel. It was great to have the company of my fellow quant PhD students from other Swedish universities: Carl, Jason, Magnus and Savvas.

I had truly remarkable colleagues at the department. Per was always ready to give me career advice. I enjoyed discussing the PhD process and student representation with Anastasiya. Johan made sure we did not only work and arranged social activities for us. Thanks for all the fika, lunch and corridor discussions Anja, Christoph, Henning, Henrik A, Katerina, Lars L, Lars Ö, Malin, Marek, Peter, Stina, Tina, Walter and Zeping. Thank you, Anne and Christina for helping me with all kinds of problems at SSE and beyond!

When I doubted whether I could finish this project, Maja was always ready for some pep talk on Skype. Thank you! I am glad that Bence, Emese, Márk, Nóri and Patrik provided me with some non-academic free time activities during these years in Stockholm; Frici, Karina, Gréta, Bence and Robi in Hungary;

and Laci, my fellow traveler, all over the world.

(14)

ix I would like to thank my family for supporting me with this decision to do a PhD in Stockholm. I am particularly grateful to my partner, Judit, for being by my side during this journey.

Stockholm, September, 2019 Péter Aleksziev

(15)

(16)

Chapter 1 Introduction and overview of the dissertation

1.1 Introduction

This dissertation is about segment reporting and valuation. It consists of four parts, this introduction and summary and three separate research papers. All three papers address the valuation relevance of segment reporting, albeit from different angles and using different methodologies.

Segment reporting provides financial information, such as sales, earnings or assets, about different parts of a firm, incremental to the consolidated (firm- level) financial statements. It is essentially providing a breakdown of the consolidated numbers, attributing these numbers to different parts of the firm.

Previous research reports that users of financial statements use segment reporting for the purposes of valuing the whole entity or its shares (Previts et al., 1994; You, 2014). Given that there is consolidated information available to value the firm, it is an intriguing area to investigate how the segmental breakdown of firm-level information can be incrementally valuation relevant. This is the overarching topic of this dissertation.

The three research papers in this dissertation are the following:

• Paper I: The role of segment reporting in corporate valuation

• Paper II: On the usefulness of segment information for earnings forecasting

(21)

• Paper III: Heterogeneous investor beliefs and value creation through equity carve-outs

Paper I is a theoretical piece focusing on how segment reporting can be used in valuation. Previous research documented that financial statement users deem segmental information as valuation relevant, however, there is little analytical evidence in the literature as to how such information can be incrementally valuation relevant. The paper investigates the valuation relevance of segment disclosure from a theoretical and analytical angle. I benchmark the specific disclosure requirements of the different segment reporting standards, as well as the disclosure provided by the firms on the information needs of accounting information based valuation models, such as the discounted cash flow (DCF) model, the value added valuation (VAV) model and the abnormal operating income growth (AOIG) model.

I find that the standards do not require enough information on the segment level to enable the use of such models to value parts of the firm separately.

Furthermore, the disclosure of the reporting entities seem to fall short of the information needs of the valuation models. I investigate the valuation benefits of 1) providing sales and earnings on the segment level (partial disclosure) compared to no segment disclosure and 2) providing all inputs required by the valuation models (full disclosure) compared to partial disclosure. The valuation relevance of partial segment disclosure increases with the cross-segment differ- ence in growth rates and margins and decreases with segment concentration.

The analytical investigation comparing full disclosure and partial disclosure reveal that using earnings multiples for valuing segments separately can result in a value that is close to the value achieved through more sophisticated models. However, firms with high priced-in growth (i.e. for which a larger part of the market value cannot be explained by current earnings¹) can achieve higher valuation through signalling their superior quality, if the segment disclosure reveals that 1) they reinvest more of the operating income in the lower-risk segment, or 2) if their lower-risk segment is generating a larger share of the ab-

1In other words, where a larger part of the market value cannot be explained by the current earning power of the firm, as defined in Graham and Dodd (1951); the concept of earning power "combines a statement of actual earnings, shown over a period of years, with reasonable expectation that these will be approximated in the future, unless extraordinary conditions supervene." (p.418-419)

(22)

INTRODUCTION AND OVERVIEW OF THE DISSERTATION 3 normal operating income growth (AOIG)²and has more durable competitive advantage (higher economic goodwill).

Paper II is an empirical paper focusing on the earnings forecasting benefits of partial disclosure (disclosing sales and earnings on the segment level).

Previous research has documented that segment information can be useful for forecasting earnings (Collins, 1976). However, more recent evidence suggests that this result hinges on within-industry, cross-company comparability of segment performance (Fairfield et al., 2009; Schröder and Yim, 2018). This comparability has decreased after the introduction of management approach into segment reporting. Previously, under SFAS 14, companies were required to present their segments according to the business lines the firms operate in (industry approach). SFAS 131, effective after 1998, requires companies to present segments in line with the internal organization of the firm. This new standard resulted in decreased comparability of segment performance between firms.

Schröder and Yim (2018) documented that while industry information could improve earnings forecasts under the industry approach, this information does not improve earnings predictions under SFAS 131. Building on this research, my paper investigates whether segment information reported following the different standards can help to predict earnings, absent within-industry comparability. Segment information following the management approach reveals within-segment trends which might be useful for predicting earning. More- over, economy-wide comparability of the information presented might also improve earnings forecasts. I also investigate how segment reporting charac- teristics are associated with more accurate earnings forecasts.

The main findings of the paper is that segment information can improve earnings forecasts even without industry information, both under SFAS 14 and SFAS 131. However, my results suggest that the usefulness of segment information for forecasting earnings has not improved with the introduction of SFAS 131. I also find that disclosing more segments is associated with more accurate earnings forecasts but that proprietary information in the segment filing does not help forecasting.

Paper III is an empirical paper focusing on the value creation through equity carve-outs. In an equity carve-out, the firm sells shares in one of its sub-

2Abnormal operating income growth is an increase of operating income beyond the expected increase due to additional capital employed.

(23)

sidiaries (segments). Previous literature has documented that equity carve-out announcements are associated with positive abnormal returns. Two possible explanations have emerged in the literature to explain the value creation of equity carve-outs. The information asymmetry hypothesis in Nanda (1991) argued that when the firm decides to raise capital through selling shares in a subsidiary (instead of selling their own shares), they reveal private information that the firm is undervalued. The market receives this signal and corrects the misvaluation, leading to a higher share price. The divestitures gains hypothesis in Schipper and Smith (1986) on the other hand, suggests that there are benefits for listing the subsidiary separately on the stock market, such as better contracting on the group or subsidiary level or better monitoring from the market. The expectations of such benefits lead to the positive announcement returns.

Empirical studies have provided inconclusive evidence with respect to these hypotheses (Hulburt et al., 2002; Powers, 2003; Dasilas and Leventis, 2018;

Dereeper and Mashwani, 2018).

In Paper III, a third hypothesis built on the heterogeneous beliefs theory of Miller (1977) is presented. This theory argues that it is difficult to value firms, which requires considerable judgment and when investors are presented with the same information, they arrive at different value estimates. The investors with the highest value estimates ("optimistic" investors) invest in the share and the others forgo the investment opportunity. The hypothesis being put for- ward in the paper suggests that the abnormal return arises from the expectation that the firm will sell shares in the subsidiary to investors who have high value estimates of the subsidiary. They will pay a price higher than the value estimates of current investors, and the increase in the share price corresponds to this expected profit from selling subsidiary shares.

I document a significant decrease in the value of the consolidated firm from before the announcement to the completion of the carve-out (observed through a higher diversification discount). This is not in line with the asym- metric information and the divestitures gains hypotheses, the first predicting higher valuation levels as a result of revealing undervaluation and the second due to expected better performance in the future. The result is however in line with the heterogeneous beliefs hypothesis, as when investors are given the possibility to own shares in a segment separately, investors who are more optimistic about the segment will sell their shares in the group (decreasing its

(24)

INTRODUCTION AND OVERVIEW OF THE DISSERTATION 5 valuation) and buy shares in the subsidiary separately. This is the first paper to provide empirical evidence investigating the heterogeneous beliefs theory for corporate restructuring. It also contributes to the equity carve-out literature through offering an additional hypothesis for the value creation in equity carve-outs. Moreover, it provides additional evidence regarding sum-of-parts valuation and the diversification discount.

I continue this introduction by presenting segment reporting regulation over time under US GAAP and IFRS. Then, I review the relevant literature, starting with investigating the effects of regulation changes under the two regimes, mainly relating to evaluating the effect of the change from an industry approach to a management approach. This literature is also closely related to Paper I and II.

Disclosing segment information can be valuation relevant in two main ways. First, it might alter the expectation of future cash flows generated by the firm (the numerator effect). I present papers investigating the usefulness of segment information for forecast sales and earnings. This literature is closely linked to Paper II and the input improvement discussions in Paper I. Second, given the additional information, segment reporting can decrease information asymmetry between market participants which in turn can decrease the cost of capital and increase the value of the firm. Previous research has investigated the cost of capital effect of segment information extensively. In the dissertation I focus mainly on the numerator effect.

Providing segment information voluntarily is a choice and the management needs to evaluate the costs and benefits of providing this information.

Apart from the potential benefits, segment disclosures can bring about different costs, such as proprietary costs, agency costs and costs of producing the information. The third part of the literature review is about these costs and benefits of segment disclosures.

Finally, I review the literature on the diversification discount. An underly- ing assumption of the investigations in Paper I is that the value of the group can be calculated as the sum of the value of the parts. However, previous literature has documented the diversification discount, meaning that firms are valued less as a group compared to the value of the parts. Applying sum-of-parts valuation for investment decisions in practice requires an understanding of the diversification discount phenomenon. The investigation in Paper III is built on the clientele effect explanation of the diversification discount.

(25)

This discussion of regulation and prior literature sets the stage for the dissertation. In the next section I present an extended summary of the three research papers. The introduction and summary section of this dissertation fin- ishes with a discussion about how the papers fit into the literature on segment reporting and valuation.

1.2 Prior research on segment reporting

Segment reporting should supposedly provide useful information to market investors. Probably the first firm to provide segment information was Lever Brothers in the 1920s. While there was no explicit segment disclosure in its accounting reports, the chairman regularly addressed the performance of the different parts of the firm. Lever Brothers merged with two other oil- and fat- based product companies into Unilever in 1937, and Unilever kept providing segment information in the years following the merger (Camfferman and Zeff, 2003).

1.2.1 Segment reporting regulation

The importance of segment reporting became more evident in the 1960s. Con- glomerates grew by acquiring firms in different industries and conglomerate- level information appeared to be less useful for assessing the performance of the parts of the firm. SEC releases and APB opinions required disclosures about different parts of the business, such as different product lines and foreign operations. The first segment reporting standard by FASB was SFAS No. 14, effective after 1976.

SFAS 14 required companies to provide information about different business lines and geographic areas (FASB, 1976). Line-of-business (LOB) segments were defined following the industry approach: "Industry segment: A component of an enterprise engaged in providing a product or service or a group of related products and services primarily to unaffiliated customers (i.e., customers outside the enterprise) for a profit." (paragraph 10. a).

Effective after 1998, FASB introduced a new segment reporting standard, SFAS No. 131. This standard defined segments following the management approach, meaning that segments no longer were required to be engaged in providing similar products or services. Instead the management approach required

(26)

INTRODUCTION AND OVERVIEW OF THE DISSERTATION 7 reporting entities to define operating segments in line with the internal organization of the firm. In particular, SFAS 131 (FASB, 1997) defined operating segments as: "a component of an enterprise: a. That engages in business activities from which it may earn revenues and incur expenses (including revenues and expenses relating to transactions with other components of the same enterprise), b. Whose operating results are regularly reviewed by the enterprise’s chief operating decision maker to make decisions about resources to be allocated to the segment and assess its performance, and c. For which discrete financial information is available." (paragraph 10.). FASB argued that comparability was impossible to achieve and that the management approach provides more relevant and reliable information about the different parts of the firm.

Segment reporting standards under IFRS developed similarly. IAS 14, effective from 1983 required disclosure about the different parts of the firm following the industry approach. This standard was later revised. IAS 14R was effective for fiscal years beginning after 1 July 1998. It still followed the industry approach of defining a segment, but extended the segment reporting requirements of IAS 14 after reflecting on previous criticism (Prather-Kinsey and Meek, 2004).

IFRS 8, effective for fiscal years beginning on or after 1 January 2009, was a result of harmonizing US GAAP and IFRS. IFRS 8 is largely similar to SFAS 131 and also requires segments to be defined using the management approach.

Table (1.1) provides an overview of studies investigating the effects of changes in segment reporting standards. Nichols et al. (2013) provided an extensive review of the literature on the change from the industry approach to the management approach of segment reporting.

Ettredge et al. (2002a) investigated firms’ views on SFAS 131 by analyzing comment letters for the exposure draft (ED). They found that firms believed that the new standard would require more disclosures. Firms’ lobbying posi- tions could be explained by expected competitive harm as a result of the disclosures required by the new standard.

(27)

Table1.1:Priorempiricalresearchinvestigatingsegmentreportingregulation ArticleRegulationPurposeofstudySampleSample time- frame

Number ofobserva- tions

VariableofinterestMainfindings Herrmann andThomas (2000)

SFAS131Comparethesegment reportingpracticesun- derSFAS14andSFAS 131.

100ofthe250 largestUS companies

1997-1998 (adoption yearand theyear before)

100firmsDescriptivestatisticsMorefirmsprovidesegmentinformation,companies disclosesomewhatmoreitemspersegment, companiesreportcountry-basedGEOdisclosur steadofwidergeographicalareas. Streetetal. (2000)SFAS131Exploringthesegment reportingpracticesof firmsunderSFAS131as comparedtoSFAS14

UScompaniesin BusinessWeek Global1000,excl. energy&fin. firms,noM&A activities

1997-1998160firmsDescriptivestatisticsThenumberofsegmentsreportedandtheirconsis- tencyincreasedin1998,severalcompaniesstarted portingmoresegments.Fewcompaniesincreased itemsreportedandprofitfiguresreportedbecame comparable. Streetand Nichols (2002)

IAS14RExploringthesegment reportingpracticesof firmsunderIAS14Ras comparedtoIAS14.

Firmsreporting underIFRSwith Englishannual reports

1998-1999210firmsDescriptivestatisticsSignificantincreaseinthenumberofitemsdisclosed underIAS14R.Improvedconsistencybetween partsoftheAR.Averagenumberofprimar mentsdidnotincreasesignificantly,however,13 changedtodisclosingmoresegmentsunderIAS14R. Ettredge etal.(2002b)SFAS131Exploringthemotives behindsegmentaggrega- tionunderSFAS14.

USmultisegment firmsreporting segmentdatain 1996-1998

1996-19983435firmsHighestandlowest decileofthedifference betweenthenumberof 4-digitSICindustries andnumberof segmentsreported underSFAS14;CAR aroundeventdates.

Largefirmsinhighlyconcentratedindustries morelikelytoaggregateanumberofbusiness intoeachoftheirsegmentunderSFAS14.SFAS 131imposedunanticipatedcostsonfirmspreviousl reportinghighlyaggregatedinformation. Ettredge etal.(2002a)SFAS131Relatingfirmcharacter- isticsandcompetitive harmtofirms’position onSFAS131.

Respondentsfor theEDpreceding SFAS131

1996202response lettersfrom companies

Binaryor3-category scaleonthepositionof therespondent

Firmsbelievedthatthenewstandardwillrequire todisclosemoreinformation.Lobbyingpositions canbeexplainedbyexpectedcompetitiveharm changesindisclosure.SFAS131wasexpectedto substantialeconomicconsequences. (cont.)

(28)

Table1.1:(continued) ArticleRegulationPurposeofstudySampleSample time-frameNumberof observationsVariableofinterestMainfindings Bergerand Hann(2003)SFAS131Investigatingiffirmspro- videmoreinformationun- derSFAS131andifitaf- fectsinvestors’andanalysts’ abilitytopredictfirmperfor- mance.

USfirms,min.20 MUSDsales1997-19992999 observationsNumberofsegments,disaggregation,segmentconcen- tration,cross-subsidization, losssegments;mechanical earningsandsalesforecasts andtheirerrors;excess value.

SFAS131resultedinmoresegmentsbeing reported(andhigherdisaggregation).Sig- nificantimprovementinanalystforecast accuracy.Improvedinformationenviron- mentandmonitoring. Bar-Yosefand Venezia (2004)

SFAS131InvestigatingwhetherSFAS 131actuallyprovidescapi- talmarketparticipantswith morepredictiveabilitythan thepreviousregulation.

Twofirms,three differentexperiment groups(accounting studentsand professionalanalysts)

1998-199932pairsof accounting students,10 professional analysts

Forecastaccuracy;testing theconfidenceofpartici- pantswithaskingthemfor 95%conf.intervals

TheforecastsobtainedbasedonSFAS131 weremarginallymoreaccurateintwoof theexperiments,butlessaccurateinthe thirdone.Measuresofaccuracyshowed higherdispersionforSFAS131subjectsin alltheexperiments. Ettredgeetal. (2005)SFAS131EffectofSFAS131onthe stockmarket’seffectstopre- dictfutureearnings.

Compustatsample, availableobservations inbothperiods,no M&Aactivities

1995-20016812firms, 21698firm-year observations

MarketreturnFERCincreasedforpre-SFAS131mul- tisegmentfirmsirrespectiveofthemin- creasingthenumberofsegmentsornot. FERCalsoincreasedforlargefirmschang- ingfromreportingonesegmenttobeing multisegment. Ettredgeetal. (2006)SFAS131HowdidSFAS131affectseg- mentdisclosurequalityand whatdrivesfirms’segment disclosuredecisions.

USmultisegmentfirms underSFAS1311994-200034920 firm-years,of which8044are firmswho disclosedmore thanone segmentunder bothstandards

Segmentreportingcharac- teristics,cross-segmentdif- ferenceinprofitabilityto proxyforquality

SFASNo.131resultedinaslightincrease inthenumberofreportedsegmentsdis- closedandahighercross-segmentvariabil- ityofreportedprofits.Itincreasedthe transparencyofreportedsegmentprof- itability,butcontinuedtoallowmanagers ofhighproprietarycostfirmssomeabil- itytoconcealcompetitivelyharmfulin- formation. Hopeetal. (2008)SFAS131TheeffectofSFAS131on foreignearningsmultiples andinvestors’mispricingof foreignearnings.

USfirmswithboth domesticandforeign currentandlagged observationsofpretax incomeandincome taxes

1985-20042212firmsAnnualabnormalreturnForeignearningsresponsecoefficientin- creasessignificantlyafterSFAS131.For- eignearningsmispricingdisappearsfol- lowingtheintroductionofSFAS131. Park(2011)SFAS131TheeffectofSFAS131on thestockmarket’sability topredictindustry-wideand firm-specificcomponentsof futureearnings.

USfirms1995-20011962firms, 8558firm-year- observations

CARThemarket’senhancedabilitytopredict futureearningsismostlydrivenbythe improvedabilitytopredictindustry-wide, cross-industryperformances,ratherthen futurefirm-specificearningscomponents. (cont.)

(29)

Table1.1:(continued) ArticleRegulationPurposeofstudySampleSample time-frameNumberofob- servationsVariableofinterestMainfindings Nicholsetal. (2012)IFRS8Segmentdisclosurecharac- teristicsandqualityunder IFRS8andcomparedtoIAS 14R.

Companiesinthe top-tierindexof14 Europeanstock exchanges(exc.UK)

2008-2009 (orotherif adopted earlieror later)

361firmsDescriptivestatisticsUnderIFRS8companiesreportsignifi- cantlymoreoperatingsegmentson age,butmostfirmsdonotreport segments.Significantdecreasein amountofitemsdisclosed.Significant improvementsinthedisclosurefineness ofsegmentdata. Nicholsetal. (2013)IFRS/US GAAPResearchreviewontheef- fectsofthemanagementap- proach.

12IFRS8studiesand 25SFAS131studies1997-2013*37studiesLiteraturereview Faríasand Rodríguez (2015)

IFRS8Ifsegmentreportinghasim- provedduetochangefrom IAS14RtoIFRS8.

Spanishlistedfirms exceptforearly adoptersandsegment compositionchangers

2008-2010104firmsDescriptivestatisticsOnly18.3%offirmsincreasedtheir porting.Manyfirmscontinuetorepor segmentsthatareinconsistentwith operations. Bugejaetal. (2015)IAS14R, IFRS8Investigatingthenumberof segmentsreportedafterIAS 14RandIFRS8,relatingit tofirmindustrialdiversity, levelofindustryconcentra- tionandnumberofloss- makingsegments.Motives fordecreasingdisclosure.

Australianlisted companies2001-2002 and 2008-2009

1241firmsfor IAS14Radop- tionand1617 firmobserva- tionsforIFRS8 adoption

Changeupindicator variable,logit regressionon explanatoryvariables. Forecasterrorand dispersion

BothIAS14RandIFRS8resultinhigher numberofsegmentsreported.The ingsforIAS14Risconsistentwit agencyprinciplebutforneither dardswiththeproprietarycostpr ple.Analystforecastpropertiesdid improve,either. Cho(2015)SFAS131Capitalallocationefficiency ininternalcapitalmarkets followingSFAS131.

USmultisegmentfirms (non-financial, non-utility)

1996-1999 dec year-end; 1997-2000 non-dec year-end

1391firm-yearsCapitalallocation efficiency(signed capexdeviation variable,proxyingfor cross-subsidization)

Changefirmsexperiencedgreater provementincapitalallocation ciencythanthenon-changecontrol ple.Morepronouncedforcompanies withlessindependentboards,mor versifiedinternalcapitalmarkets greatertakeoverthreat. Notes:FERCisforwardearningsresponsecoefficient.GEOreferstogeographicsegmentdisclosure.MUSDismillionsofUSdollars.‘*’marksthatthestudiesreviewedwerepublished period.

(30)

INTRODUCTION AND OVERVIEW OF THE DISSERTATION 11 Studies investigating the change from SFAS 14 to SFAS 131 have documented that firms in fact disclosed more segments (Herrmann and Thomas, 2000; Street et al., 2000; Berger and Hann, 2003; Ettredge et al., 2006) and more items per segment (Herrmann and Thomas, 2000; Street et al., 2000) under the new standard. Also, fewer companies disclosed earnings on the geographical segment level (Herrmann and Thomas, 2000) as this no longer was required under SFAS 131 (FASB, 1997). Firms also changed their reported segment structure (Herrmann and Thomas, 2000; Street et al., 2000), i.e. it became more in line with the organizational structure presented elsewhere in the annual report. However, many firms still presented a different structure as compared to the MD&A section of the 10-K filings (Street et al., 2000).

Ettredge et al. (2002b) found that firms in highly concentrated industries were more likely to aggregate business areas into one segment under SFAS 14.

They found that SFAS 131 had a negative impact, suggesting that the new standard imposed additional proprietary costs on reporting entities. Ettredge et al.

(2006) also documented that the transparency of reported segment profitabil- ity increased, however, they argued that the standard still allowed managers of high proprietary cost firms to conceal information.

The higher disaggregation under SFAS 131 was also documented in Berger and Hann (2003). They found that the additional information helped analysts to improve earnings forecasts and resulted in a better information environment and potentially better monitoring. Botosan and Stanford (2005) documented that information asymmetry and analyst forecast errors increased marginally following the introduction of SFAS 131. Moreover, Ettredge et al. (2005) found increased future earnings response coefficients (FERC) after the introduction of SFAS 131 for firms disclosing more than one segment under SFAS 14, ir- respective of whether these firms increased the number of segments reported, suggesting that SFAS 131 brought about earnings numbers that were more in- formative compared to SFAS 14. Furthermore, Hope et al. (2008) documented increased foreign earnings response coefficients under SFAS 131, documenting that geographical segment disclosure under SFAS 131 helped to eliminate earnings mispricing. Park (2011) found that the market’s enhanced ability to predict earnings arised from the improved ability to predict industry-wide, cross- industry performance, and not firm-specific earnings components.

Bar-Yosef and Venezia (2004) observed in an experimental study that professional users of financial information (analysts and accounting students close

ESSAYS ON SEGMENT REPORTING AND VALUATION

ESSAYS ON SEGMENT REPORTING AND VALUATION

Peter Aleksziev

ESSAYS ON SEGMENT REPORTING AND VALUATION

Essays on Segment Reporting and Valuation

Péter Alexander Aleksziev

Akademisk avhandling

Essays on segment reporting and valuation

Essays on segment reporting and valuation

Peter Aleksziev

Foreword

Acknowledgements

Contents

Chapter 1

Introduction and overview of the dissertation

1.1 Introduction

1.2 Prior research on segment reporting