Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2017-02-10T12:19:50Z No. of bitstreams: 1 Adalene Olivia Silvestre_.pdf: 590609 bytes, checksum: 188661a0ac02964a45900cec5e4c0062 (MD5) / Made available in DSpace on 2017-02-10T12:19:50Z (GMT). No. of bitstreams: 1 Adalene Olivia Silvestre_.pdf: 590609 bytes, checksum: 188661a0ac02964a45900cec5e4c0062 (MD5) Previous issue date: 2016-12-20 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / PROSUP - Programa de Suporte à Pós-Gradução de Instituições de Ensino Particulares / A auditoria independente exerce um importante papel na relação entre a empresa e os usuários externos à entidade, devendo o auditor ser independente em relação à empresa auditada. No Brasil, o rodízio obrigatório de firma de auditoria é regulamentado pela Instrução CVM 308/99 na tentativa de contribuir com a manutenção da independência do auditor e, consequentemente, com a qualidade dos lucros divulgados pelas empresas. Diante disto, o presente estudo tem por objetivo analisar o efeito do rodízio da firma de auditoria sobre a qualidade dos lucros das empresas de capital aberto brasileiras listadas na BM&FBOVESPA no período de 2008 a 2015. Como medida de qualidade dos lucros foram utilizados os accruals residuais, que identificam a parcela discricionária dos accruals, medida inversa à qualidade dos lucros. Os accruals residuais foram abordados a partir de duas diferentes perspectivas: o gerenciamento de resultados, medido pelos modelos de Jones (1991) e Jones Modificado por Dechow, Sloan e Sweeney (1995) e os erros de estimativas, medidos pelos modelos de Dechow e Dichev (2002) e Dechow e Dichev modificado por McNichols (2002). Os resultados demonstram que o rodízio de firma de auditoria reduz o volume de accruals residuais e, assim, aumenta a qualidade dos lucros, quando esses são mensurados a partir da perspectiva do gerenciamento de resultados, através dos modelos de Jones e Jones modificado. Entretanto, o efeito do rodízio de firma de auditoria sobre a qualidade dos lucros não é observado quando os accruals residuais são mensurados a partir da perspectiva dos erros de estimativas contábeis, através dos modelos de Dechow e Dichev e McNichols. Por outro lado, os resultados demonstram que as empresas que realizam rodízio voluntário de firma de auditoria apresentam maiores accruals residuais e, consequentemente, menor qualidade dos lucros. / Independent audit plays an important role in the relationship between the company and external users, and the auditor must be independent of the audited company. In Brazil, the mandatory audit firm rotation is regulated by CVM Instruction 308/99, in an attempt to contribute to the maintenance of auditor independence and, consequently, with the quality of earnings disclosed by the companies. Therefore, the present study has the objective of analyzing the effect of the audit firm rotation on the earnings quality of Brazilian public companies listed on BM&FBOVESPA in the period from 2008 to 2015. Residuals accruals were used as a measure of earnings quality, which identify a discretionary portion of the accruals, inverse measure of earnings quality. The residuals accruals were approached from two different perspectives: earnings management, measured by Jones model (1991) and Jones modified by Dechow, Sloan and Sweeney model (1995) and the estimation errors, measured by the Dechow and Dichev model (2002) and Dechow and Dichev modified by McNichols model (2002). The results show that audit firm rotation reduced the volume of residuals accruals and, thus, increases the earnings quality, when these are measured from the perspective of earnings management through the Jones and Jones modified models. However, the effect of audit firm rotation on the earnings quality is not observed when the residuals accruals are measured from the perspective of accounting estimation errors, through the Dechow and Dichev and McNichols models. On the other hand, the results demonstrate that the companies that perform the voluntary audit firm rotation have greater residuals accruals and, consequently, lower earnings quality.

This study investigated earnings management in European firms. The private investors became victims of manipulated earnings where few laws offered regulatory oversight. The study forensically examined the attributes of earnings management identified using a discretionary accrual model published in Jones' work and Schippers' work. The firms' managers should fulfil agency theory when they made reporting decisions, and they should act in the investors' best interests to fulfil stewardship theory. The managers failed as they seemed to favor insiders when they reported manipulated earnings to outsiders like small investors even though the managers published financial reports conforming to the International Financial Reporting Standards. The investors depended on the decision usefulness of the reports. The study used the data of 432 listed firms in 11 code law nations. The paired t test identified significant differences between reported and economic earnings to find earnings management attributes and between economic and restated earnings to find earnings management cases. The research found that managers seemed to manipulate discretionary accruals to misstate earnings and reduce the decision usefulness of reporting. The data came from published financial reports and databases. The firms represented 11 nations and 9 industries that excluded banking and insurance. Almost 17% of nations and industry segments reflected earnings management attributes. About 29% of firms restated at least one annual earnings, and 84% of the restatements appeared to offset manipulation. The research results should prompt social change for small investors where regulators would redress the manipulation using stronger investor protection laws to improve the reported earnings quality and its decision usefulness.

Syfte: Studien avser att undersöka om det går att påvisa ett samband mellan goodwillnedskrivningar och företagets ekonomiska utveckling, företagsledningens incitament och företagets bolagsstyrning bland bolag i Sverige. Då den svenska bolagsstyrningsmodellen och ägarstrukturen skiljer sig från den anglosaxiska, syftar studien även till att utvärdera om utfallen i denna studie avviker från en tidigare genomförd studie i Storbritannien och om avvikelserna kan hänföras till just skillnaderna i ländernas ägarstruktur och bolagsstyrning.   Metod: Metoden är en replikering av en tidigare studie. Vi har använt oss av en kvantitativ metod, med deduktiv ansats. Studiens primära analysmodell är en Tobitregressionsanalys. Empirin är hämtad från databasen FactSet samt manuell granskning av årsredovisningar. Studien baserar sig på samtliga noterade bolag på Nasdaq Stockholm per den 1 oktober 2018 för åren 2015, 2016 och 2017. Detta resulterade i ett totalt urval om 330 bolag och 990 observationer. Efter bortfall baserat på olika förbehåll uppgick det slutliga urvalet till 408 observationer, varav 74 med och 334 utan rapporterade goodwillnedskrivningar. Slutsatser: Studien fann inget statistiskt säkerställt samband mellan företagets ekonomiska utveckling och goodwillnedskrivningar. Däremot tyder studien på att de bolag vars bokförda värde på eget kapital överstiger dess marknadsvärde redovisar goodwillnedskrivningar i större utsträckning. Vad gäller företagsledningens incitament till att manipulera resultatet så visade studien på ett signifikant samband mellan VD-byten och goodwillnedskrivningar. Detta indikerar att en nytillsatt VD använder sig av resultatmanipulering för egen vinning. Vi fann inget signifikant samband mellan företagens bolagsstyrningsmekanismer och nedskrivning av goodwill. Utfallen i den tidigare studien, som baserar sig på bolag i Storbritannien, skiljer sig markant från utfallen i vår studie. Detta skulle kunna förklaras av att ägarstrukturen och bolagsstyrningsprinciperna skiljer sig åt mellan länderna.

民國八十年後，政府積極推動金融自由化與國際化及開放民營銀行之設立等政策，致使我國銀行業面臨空前激烈之競爭，而政府為加強對銀行業之金融監理及金融檢查功能，必須仰賴銀行所提供之會計資訊作為定期檢查銀行業務時，分析銀行經營績效之參考；此外，一般投資人因資訊取得之困難與成本效益之考量，亦常傾向於以銀行公開之財務報表來評斷銀行經營績效。因此，深入探析我國銀行業之盈餘管理行為，將有助於政府及投資人評估銀行之經營風險及績效。 本研究以民國八十年至民國八十八年為研究期間，首先分析銀行業是否有延遲認列損益之行為，接著從股權結構面（包含公、民營銀行及董監事持股比率）探討我國銀行業之盈餘管理行為，藉以瞭解銀行在受政府法規高度管制之情況下，是否會因股權結構之不同而有不同之盈餘管理行為。其次，本研究亦從資本市場誘因之觀點，探析國內新上市銀行於上市前為符合上市條件，而上市後為避免更新已公開之財務預測，是否會造成盈餘管理行為有所差異。 實證結果顯示：（1）我國銀行業之損益操弄項目普遍存在季別差異之現象，除了有延後認列損益操弄項目至第四季外，同時也發現銀行第二季損益操弄項目有高於第一、三季之情形，似有半年報之效果呈現；（2）民營銀行之第四季處分資產損失、壞帳費用及提列其他損失準備佔全年比例顯著高於公營銀行，顯示民營銀行在第四季認列費用或損失之動機強於公營銀行；（3）銀行董監事持股比率會對第四季損益操弄項目佔全年比例有所影響之假說未獲得支持；（4）銀行上市前二年之營業利益及稅前純益佔實收資本比率之平均數，皆達10%獲利水準門檻，且皆顯著高於上市後第二年，據此間接證明銀行上市前之盈餘管理目標係在維持獲利水準達到10%之門檻；（5）達成財務預測之銀行會傾向於在第四季多認列壞帳費用，至於利益操弄項目，則並無第四季佔全年比例顯著高於前三季之情形。 / Banking industry in Taiwan is highly regulated and scrutinized by the banking law and government. Regulators of banking industry require that banks must satisfy certain capital adequacy ratio that is explicitly tied to accounting numbers. The regulating bodies have authority in inspecting banks' businesses and examining their financial reports periodically to evaluate their underlying performance. Based on a cost-benefit consideration and availability of information, the investors also analyze the performance of banks using their public available financial statements. The earnings behavior of firms in the banking industry is therefore essential for both regulators and investors in procuring the knowledge about the operating risk and performance of banking firms. Focused on the firms in the banking industry over the period of 1991 to 1999, this thesis first investigates whether the sample firm is subject to an earnings behavior of delayed recognition of their income and losses. The thesis also examines the influence of ownership structure on banks’earnings management behavior through the analysis of income manipulation ratios. Furthermore, in view of capital market incentives, this thesis studies whether going listed in Taiwan Stock Exchange would cast impacts on banks'earnings management behavior as banks are subject to different earnings management objectives (targets). The empirical results document that the fourth quarter ratios of the interim income manipulation items are larger and more variable than those of the other three quarters. This suggests that banks tend to delay their recognition of income manipulation items till the fourth quarter. The empirical findings reveal that the propensity for manipulating the fourth quarter expense/loss items is significantly greater for private banks than for state-owned banks. However, the empirical results cannot support the hypothesis that management ownership (the holdings of the directors and supervisors) will affect the magnitude of earnings management. With respect to capital market incentive issues, the empirical evidence indicates that for newly listed banks, both the two-year average operating income to capital ratio and income before tax to capital ratio before listing reach and not greatly exceed the 10% threshold. The evidence shows that these two ratios decline significantly in the second year after listing. The findings also suggest that listed banks which achieve their financial forecasts tend to recognize more bad debt expenses for the fourth quarter, while the pattern cannot manifest itself in the gain manipulation items.

本研究旨在探討我國一般產業之上市公司，其分析師盈餘預測與管理當局盈餘預測間之差異，並且檢視在資本市場中是否具有抑制公司經理人員任意發佈預測之機制。除此之外，亦檢測市場達成共識之程度是否為管理者盈餘管理之誘因，以及其盈餘管理的程度與方向。 研究樣本取自我國上市公司一般產業之財務預測年資料與實際經營結果財務年資料，研究期間涵蓋自民國85年至88年止。 實證結果發現管理當局不論是對市場傳遞好消息或壞消息，皆是其盈餘管理之結果。再者，當市場預期趨近於一致時，公司管理當局愈會對市場傳遞出中性消息，亦即分析師之盈餘預測與管理當局盈餘預測趨於無差異。同時公司亦可能基於市場壓力，而利用裁決性應計項目以從事盈餘管理。當操縱前盈餘高於分析師預期之平均盈餘時且市場達成共識時，經理人員會傾向不從事盈餘管理，而在市場預期一致性程度低時，從事盈餘管理；反之，當操縱前盈餘低於分析師預期之平均盈餘，且市場預期一致性程度高時才會進行盈餘管理以增加盈餘而達成市場的預期。 / I test a market expectations and market consensus hypothesis about earnings management in Taiwan's public firms from 1996 to 1999. First of all, no matter what a good news or bad news that managers spread in markets is the result of earnings management. Furthermore, when analysts have reached a consensus in their earnings forecasts, managers' earnings forecasts are close to market expectations as possible. At the same time, managers also have an incentive to manage earnings through discretionary accruals to achieve market expectations. Finally, all sample-corporations are split into two groups. Group 1 observations have nondiscretionary earnings below the mean analysts' forecast, and Group 2 observations have nondiscretionary earnings above the mean analysts' forecast. The results suggest the corporation managers in Group 1 make greater use of discretionary accruals to manipulate earnings to achieve market expectations when analysts have reached a consensus in their earnings predictions. Oppositely, the corporation managers in Group 2 make less use of discretionary accruals to save them in use of next period when analysts lacks consensus in their earnings forecasts.

基於資本市場先天資訊的不對稱，企業內部人本就享有私有資訊之絕對利益，再加上風險分攤的觀念，透過證券市場向投資人募集資金的方式多為企業的首選。而主管機關為防止投資人權益受損，對於企業現金增資審核之嚴謹門檻，卻導致企業盈餘管理的動機。國內研究對於上述因果及衍生問題已有相當多的研究，但未見針對現金增資目的與盈餘管理動機及工具之關聯性進行探討。 一般而言，公司可能為了補充營運資金、償還負債、擴充產能或購併他公司等目的申請現金增資。理論上，不同的增資目的其資金需求之迫切程度應不同，且不同的增資目的之資金需求也應透過財務報表中不同的會計科目來表達。本研究主要的實證議題有二：（1）現金增資的決策提供盈餘操縱的動機，而盈餘操縱的強度是否因增資目的不同而互異？及（2）不同的增資目的是否搭配不同的盈餘操縱工具，期使現金增資目的與財務報表所顯現出的現金需求相允合？ 實証結果顯示：（1）上市公司增資前確實傾向於採用增加盈餘之裁決性應計項目，同時配合操控業外收益來增加盈餘，但以資產或收益調整之裁決性應計項目之操縱行為並不顯著。（2）補充營運資金及償還負債二種增資目的盈餘操縱強度只有在增資前一季是大於其他增資目的，其他測試期間並無顯著不同。（3）補充營運資金之增資目的並不傾向採用裁決性應計項目作為盈餘操縱工具。（4）償還負債之增資目的傾向利用業外收益作為盈餘操縱工具。（5）補充營運資金及償還負債二增資目的之盈餘操縱動機之強度並無顯著差異。

地雷股係因本土性金融風暴而產生，眾多的股市投資人因此受到巨額損失，進而質疑財務報表的允當性及會計師、主管機關的監督不夠周嚴。因此，本研究主旨在探討地雷公司在遭遇本土性金融風暴時，財務報表是否具有資訊價值以及地雷公司會採用何種模式來管理盈餘，以穩定股價或隱藏財務風險。 本研究以股價行為的模式來定義地雷股，研究樣本包括有地雷股和相對同一產業的非地雷股樣本，研究方法主要以Modified Jones Model推估裁決性應計項目，並依常態性檢定之結果，以t檢定及Wilcoxon Sign Rank檢定，分析地雷公司和非地雷公司的可裁決應計項目和營業外項目變動情形，以檢驗地雷公司是否有盈餘管理的情形。 實證結果顯示：(1)本研究以地雷公司在亞洲金融風暴後地雷爆發前會發生盈餘管理假說，在實證上獲得支持，但是非地雷股也會有盈餘管理的情形產生。(2)地雷股盈餘管理的時點大多發生在民國八十七年第一季、第二季時。(3)地雷股盈餘管理的工具是傾向於使用營業外損益項目，相較之下營業項目是較不顯著的。

由於當公司管理機制設計出了問題時，盈餘管理動機會較容易通過篩檢，甚至管理機制本身可能都會產生盈餘管理動機，並使得管理當局可能容易操縱公司決策或是報表數字，又從國內報章雜誌的報導，我們可以發現一些台灣管理機制上的亂象，如董監一家親、董事不懂事、監事不監視或是一個家族、機構投資包辦董事會席次的情況。 在此環境之下，為了維護股東權益、降低盈餘管理行為與促進證券市場的穩定發展，本研究擬針對三構面—與股東有關、與董監事有關、與經理人有關之管理機制，如：機構投資者持股比例、外部大股東持股比例、董事會規模大小、最大家族成員董事持股、監察人是否具獨立地位、董事長是否兼任總經理、經理人持股比例、總經理薪資總額等管理機制設計，探討其與盈餘管理程度之影響。 本研究為了達到學術與實務結合，採用會計師與會計經理之實地訪談與實證研究兩種方法，以兼具深度之訪談與廣度之實證研究，探討公司管理機制設計對盈餘管理程度之影響。 一、會計師與會計經理之訪談 本訪談目的係為了瞭解國內上市公司管理機制設計與操縱之情形，因此挑選對於實務運作有深入瞭解的執業會計師與上市公司會計經理，透過依研究問題所設計的各項訪談問題，最後依照所獲結論彙總形成研究假說。而訪談結果大部分的會計師與會計經理皆認為，有些公司的確會在股東部分、董監事部分與經理人部分的管理機制作刻意的安排，以使得其盈餘管理目的較易達成。 二、實證研究 透過以上訪談形成研究假說之後，本研究以上市公司為研究對象，蒐集民國84年至87年間盈餘管理程度與各管理機制之相關資料，進行複迴歸分析，經由會計師與資深會計經理之訪談與實證研究，本研究得出以下綜合性結論： (一)股東部分之管理機制： 1. 機構投資者持股比例：從訪談中得知，機構投資者持股對於公司盈餘管理程度有相當程度之影響，但由於其身份真假難以區分，因此較難確定其所發揮之角色，及其對盈餘管理程度之正負影響；在實證分析結果部分，此管理機制則未與盈餘管理程度成顯著關係。 2. 外部大股東持股比例：從訪談結果中得知，由於國內企業普遍存在董事席次安排的問題，因此外部大股東對於公司決策有一定程度之影響；但由於較難確定此管理機制所發揮之角色，因此較難確定其對盈餘管理程度之正負影響；在實證分析結果部分，此管理機制則未與盈餘管理程度成顯著關係。。 (二)董監事部分管理機制： 1. 董事會規模大小：從訪談結果中發現，若具有共通性的成員越多，可能越容易從事盈餘管理行為，且在實證分析部分，民國八十七年度樣本分析中發現，董事會規模大小與盈餘管理程度呈顯著正相關，即表示董事會規模越大，公司之盈餘管理程度越高。此實證分析結果與專家訪談所獲結論相符。 2. 最大家族成員董事持股比例：從訪談結果中發現，若最大家族在董事會中勢力越大，則對公司影響程度越大，而在實證分析部分，於民國八十七年度樣本分析中發現，最大家族成員董事持股比例與盈餘管理程度呈顯著正相關，表示最大家族成員董事持股比例越高，公司盈餘管理程度越高。此實證分析結果與專家訪談所獲結論相符。 3. 監察人身份是否獨立：從訪談中發現，若是機構監察人可能由於較容易操縱，不易維持獨立地位，因此較自然人監察人不易發揮監督功能，而在實證分析部分，則未與盈餘管理程度呈顯著關係。 (三)經理人部分管理機制： 1. 職務雙重性：從訪談中得知，若具有職務雙重性較不易執行相互報告與負責之制度，可能使得盈餘管理行為較易發生，但可能由於仍有其他董事會成員兼任經理人之情況未納入考慮，使得實證分析部分，未與盈餘管理程度呈顯著關係。 2. 經理人持股比例：從訪談中發現，有些產業偏好以股票招攬或鼓勵員工，因此對於經理人的激勵有相當大的誘因，但也可能因此產生盈餘管理的動機，但可能由於未能考慮不同持股來源之不同動機程度，使得在實證分析部分未能與盈餘管理程度呈顯著關係。 3. 總經理薪資總額部分： 從訪談結果中發現，總經理的薪資通常包括本薪與依照盈餘計算之紅利，因此可能由於未能詳細區分薪資內容之盈餘管理動機，使得在實證分析部分，未與盈餘管理程度成顯著關係。 目 錄 第一章 緒 論………………………………….............1 第一節 研究動機與目的…………………………………… 1 第二節 研究問題…………………………………………… 3 第三節 研究方法…………………………………………… 4 第四節 研究架構…………………………………………… 6 第二章 文獻探討…………………………………............8 第一節 盈餘管理相關文獻………………………………… 8 第二節 公司管理機制之相關文獻…………………..…….. 10 第三節 公司管理機制與盈餘管理相關文獻……………… 21 第四節 文獻彙總…………………………………………… 30 第五節 研究延伸…………………………………………… 38 第三章 資深會計經理與會計師之訪談………........ 43 第一節 訪談目的……………………………….…….……... 43 第二節 訪談問題之形成………………………….………… 43 第三節 訪談結果記錄…………………………….……….... 48 第四節 訪談內容比較與分析…………………….………… 58 第四章 研究方法論……………………………........ 61 第一節 觀念性架構………………………………...……….. 61 第二節 研究方法………………………………….…..…….. 62 第三節 研究假說…………………………………….…..….. 62 第四節 變數衡量………………………………………...….. 74 第五節 實證模式………………………………….……….... 81 第六節 研究期間與抽樣限制條件……………….………… 82 第七節 資料來源………………………………….………… 83 第八節 資料分析方法……………………………...……….. 84 第五章 實證研究結果…………………………... .........85 第一節 敘述性統計…………………………………….…… 85 第二節 複迴歸分析結果…………………………….…….... 86 第六章 結論與建議……………………….…….. .........93 第一節 研究結論………………………………………...….. 93 第二節 研究限制……………………………………………....96 第三節 建議及未來研究方向……………………….…….... 96 參考文獻………………………………............…………….99 表 次 表一、國外盈餘管理文獻彙總表………………………………. 31 表二、國內盈餘管理文獻彙總表………………………………. 31 表三、國外管理機制文獻彙總表—屬理論探討者……………. 32 表四、國外管理機制文獻彙總表—屬實證分析者……………. 32 表五、國內管理機制文獻彙總表………………………………. 34 表六、國外管理機制與盈餘管理相關性之文獻彙總表………. 35 表七、國內管理機制與盈餘管理相關性之文獻彙總表………. 36 表八、研究延伸彙總表…………………………………………. 42 表九、研究問題與訪談問題對照表……………………………. 44 表十、訪談問題形成之彙總表…………………………………. 45 表十一、問題一結論彙總表……………………………………. 49 表十二、問題二結論彙總表……………………………………. 50 表十三、問題三結論彙總表……………………………………. 52 表十四、問題四結論彙總表…………………………………… 53 表十五、問題五結論彙總表……………………………………. 54 表十六、問題六結論彙總表……………………………………. 55 表十七、問題七結論彙總表……………………………………. 56 表十八、問題八結論彙總表…………………………………… 56 表十九、問題九至十六結論彙總表……………………………. 57 表二十、會計師與會計經理訪談結果比較表…………………. 58 表二十一、變數衡量彙總表……………………………………. 80 表二十二、所有樣本之產業分布表……………………………. 82 表二十三、敘述統計分析表……………………………………. 85 表二十四、總樣本分析VIF值表……………………………… 86 表二十五、總樣本迴歸分析表…………………………………. 87 表二十六、各年度樣本獨立性假設分析表……………………. 90 表二十七、各年度樣本迴歸分析表……………………………. 90 表二十八、管理機制與盈餘管理程度之研究結果彙總表……. 94 圖 次 圖一、研究方法概念流程圖…………………………….……... 5 圖二、研究架構圖……………………………………….……... 7 圖三、訪談問題形成流程圖…………………………………… 44 圖四、股東部分管理機制與盈餘管理之觀念架構圖………... 61 圖五、董監事部分管理機制與盈餘管理之觀念架構圖……… 61 圖六、經理人部分管理機制與盈餘管理之觀念架構圖……… 62 附 錄 附錄一、國內外管理機制整理表…………………………….. 105 附錄二、統計結果彙總表……………………………….….… 110 附錄三、依照產業別估計裁決性應計項目之迴歸分析….…. 114 / If there are some deficiencies on corporate governance structure, the motivations of earnings management will pass through the corporate governance easier. The corporate governance will even create some motivations, and managers (might be owners also) will tend to manage the decisions or reported earnings. Hence, whether the corporate governance of public companies be designed and works well is the hottest topic in Taiwan, in order to assert stockholders’ rights, decrease the behaviors of earnings management, and improve the stock market develop steadily. This thesis combines interviews (with CPAs and accounting managers) and empirical study to investigate the relation between corporate governance and earnings management. 4. Interviews with CPAs and accounting managers: The main purpose of interviews is to understand how the corporate governance of public companies be designed and manipulated, so the interviews are made with CPAs and accounting managers who realize the true meanings of CG more profoundly. By the conclusions of the interviews, I can further ascertain the foundations of my assumptions and give the outcomes of empirical tests stringent explanations. In the conclusions of the interviews, most CPAs and accounting managers consider that the corporate governance in stockholders, members of board, and managers’ dimensions are indeed manipulated to achieve some purposes of earnings management in many public companies in Taiwan. 5. Empirical study: My study objects are the public companies in Taiwan Stock Trade. I collect financial and corporate governance data and run regression analysis. The comprehensive conclusions of interviews and empirical tests are as followed: 3. Stockholders’ dimension: (1) Institutional stockholders’ holdings: From the contents of interviews, I realize that institutional shareholders’ holdings have some impacts on earnings management, but the real roles of institutional shareholders are hard to clarify. As the result, it’s hard to predict the relation between institutional holdings and earnings management as positive or negative. In the empirical test, there is no significant relation between institutional holdings and the level of earnings management. (2) Outside big stockholders’ holdings: From the contents of interviews, I realize that most of big shareholders of some family company have been in the board of directors, and in the empirical test, there is no significant relation between outside big stockholders’ holdings and the level of earnings management. 4. Members of board’s dimension: (1) Size of board of directors: From the contents of interviews, I realize that the more the number of members of board with similar characteristics, the easier will they manage reported earnings. In the empirical test of 1998 samples, there is significant positive relation between size of board of directors and earnings management. It means that the bigger the board of directors, the higher the level of earnings management. (2) The holdings of the biggest family members of directors: From the contents of interviews, if the biggest family has more power in the board of directors, it influences the board of directors more. In the empirical test of 1998 samples, there is significant positive relation between the holdings of the biggest family members of directors and earnings management. It means that the higher the holdings of the biggest family members of board, the higher the level of earnings management. (3) Whether the supervisors have independent roles: From the contents of interviews, I realize that because of some regulations of supervisors, the institutional supervisors are manipulated easier than natural supervisors and hard to supervise the behavior of earnings management. But in the empirical test, there is no significant relation between supervisors’ independent role and the level of earnings management. 5. Managers’ dimension: (1) Duality (whether the chairman of board and CEO are the same): From the contents of interviews, I realize that if the chairman of board and CEO are the same, there might be deficiencies of reporting and mutual responsibility and higher level of earnings management. But in the empirical test, there is no significant relation between duality and earnings management. (2) Managers’ holding: From the contents of interviews, I realize that some companies of particular industries are preferred to give stocks to recruit or encourage employees, so the holdings of managers are very incentive. But in the empirical test, there is no significant relation between managers’ holdings and the level of earnings management. (3) CEO’ compensation: From the contents of interviews, the CEO’s compensation is often included fixed salary, bonus based on the earnings, so it might has some relation with the level of earnings management. But in the empirical test, there is no significant relation between CEO’s compensation and the level of earnings management.

本研究旨在觀察國內上市公司因經營不善、財務困難而遭管制機構處以降類、變更交易方式或終止上市之處分前，是否進行盈餘管理。由於遭降類等處分後，不僅其公司經理人有遭撤換的可能，更使公司形象、信譽受損，故本研究由實是性會計理論的觀念推衍，並參考相關的法令規定，推論經營不善之上市公司，其管理當局會有強烈的動機進行盈餘操縱，以避免遭降類等之噩運。 　　本研究利用可裁決性應計項目(discretionary accruals)在降類等事件日前後的尺度是否具顯著性差異，來偵測財務困難公司是否在降類等處分前利用可裁決性應計項目進行盈餘操縱。本研究總樣本共64家，涵蓋14種產業，其中降類公司樣本25家，變更交易方式公司樣本26家，下市公司樣本13家。樣本期間為民國66年至83年，每個樣本財務資料涵蓋7個年度。經過實證分析後獲致下列結果： 　　1. 股票遭降類公司，降類前的可裁決性應計項目有大於降類後之現象，其中尤以降類前一年度最為顯著。 　　2. 股票遭變更為全額交割公司，變更前的可裁決性應計項目有大於變更後的現象，其中尤以變更前一年度最為顯著。 　　3. 股票遭下市公司，其下市前的可裁決性應計項目與下市後的差異並不顯著。 　　綜合上述實驗結果，本研究可獲以下結論： 　　1. 第一類股票上市公司因經營不佳而遭變更為第二類股前有利用可裁決性應計項目進行盈餘管理之情形，尤以降類前一年度最為明顯。 　　2. 上市公司因財務困難而遭變更為全額交割股前，有利用可裁決性應計項目進行盈餘管理之情形，尤以變更前一年度最為明顯。 　　3. 公司下市前利用可裁決性應計項目進行盈餘管理的情形並不顯著。究其原因，可能係由全額交割股至下市這一過程乃一不算短的時間，由於應計項目具有迴轉的特性，或因公司財務結構、獲利能力的日趨惡化，使盈餘操縱的空間愈小，使公司已無能力於下市前進行盈餘操縱。

整體而言，會計盈餘提供財務報表使用者有關於企業獲利能力之相關資訊。然而，由於會計盈餘同時包含了雜訊 (noises) 與偏差 (biases)，因而影響到會計盈餘對公司獲利能力評價的正確性。因此，過去的會計文獻發展出不同的盈餘管理估計模型 (如: Healy 1985; DeAngelo 1986; Friedlan 1994; DeFond and Jiambalvo 1994 以及Dechow, Sloan and Sweeney 1995)，以嘗試去除這些雜訊與偏差。然而，究竟哪一個估計模型能夠提供最為純淨之非裁量性淨利 (nondiscretionary income) 的衡量指標，則並無定論。在效率市場 (market efficiency) 的假說下，本研究透過本益比 (P/E ratio) 的平均數復歸 (mean-reverting) 現象來評比五種盈餘管理估計模型。由於過去的文獻同時發現盈餘成長率與風險係數均會影響本益比的高低，因此，本研究同時將這兩個變數納入考量。 　　實證結果發現，依照上述五種盈餘管理估計模型所估計之本益比皆有平均數復歸的現象。其中Friedlan (1994) 模型在全體樣本與控制盈餘成長率之後，其本益比平均數復歸現象均較其他模型為快；次佳之盈餘管理估計模型為DeFond and Jiambalvo (1994) 與Dechow et al. (1995) 兩模型；最差的則為 Healy (1985) 模型。 / Overall speaking, accounting earnings provide financial statement users with useful information about a firm's profitability. However, because of the biases and noises included in the accounting earnings, the accuracy and reliability of accounting earnings to the evaluation of a firm's profitability may be adversely influenced. In light of this, prior earnings management studies have developed various estimation models of nondiscretionary income (e.g., Healy 1985; DeAngelo 1986; Friedlan 1994; DeFond and Jiambalvo 1994; Dechow, Sloan & Sweeney 1995) with an attempt to remove the biases and noises embedded in the accounting earnings. Nonetheless, there is no consistent empirical evidence about the relative performance of these estimation models. Assuming market efficiency, the main purpose of this study is to utilize the mean-reverting phenomenon of P/E ratios to evaluate the relative performance of these models. Since prior studies have found that earnings growth rate and risk coefficient may affect the magnitude of P/E ratios, we also control for these two variables in our analyses. 　　The empirical results reveal several findings. First, P/E ratios calculated using different earnings management estimation models exhibit the mean-reverting phenomenon. Second, the Friedlan (1994) model has the best performance among all models when we use the overall sample and three subsamples grouped based on the earnings growth rate. In addition, the DeFond and Jiambalvo (1994) and Dechow, Sloan & Sweeney (1995) models perform moderately. Finally, the Healy (1985) model shows the worst performance.