Tidigare studier har visat på att en finansiell kris leder till en minskad förekomst av earnings management som anses vara ett stort problemen i modern redovisning. Denna uppsats undersöker earnings managements förekomst bland svenska företag och hur den påverkas av en finansiell kris samt om det finns skillnader mellan olika branscher. Vi använder en modifierad variant av Jones modell för att mäta de diskretionära periodiseringarna, som likställs som earnings management, och jämför sedan förekomsten av earnings management före och under krisen samt mellan de fyra branscherna som studerats. Vi finner inte några skillnader i earnings managements förekomst innan och under finanskrisen men vi kan se att branschen Sällanköp i större utsträckning än övriga studerade branscher bedriver earnings management. Vår förklaring till detta är att övervakningen av företagens redovisning är generellt sett bra vilket minimerar möjligheten att bedriva earnings management men att det kan finnas brister för branschen Sällanköp.

Recent financial crises and accounting scandals resulting from perceived audit failures have given rise to increase criticism in the manner which accountants and auditors respect their professional code of ethics. As a consequence, the SEC rules in 2000 limits the services auditors are allowed to provide to their clients and also set limits on fees for internal audit services. In addition to this, they called upon all firms to disclose all fees paid to auditors for both audit and non-audit services. This thesis investigates whether the provision of Non-Audit Services is associated with auditor’s independence evidence from the earnings management perspective proxy by discretionary accruals. I began this study by examining the relevant textbooks and related articles of accounting and auditing. Furthermore, the hypotheses and research model are proposed based upon related accounting and auditing theories. The data for this thesis were manually collected from the NASDAQ OMX website from the annual financial reports of 107 Swedish public listed firms in the Stockholm Stock Exchange. The selection of these companies was based on their capitalization i.e., Large, Medium and Small size firms. In this study, quantitative research method has been used to explore the relationship between provision of NAS and auditor’s independence with the help of statistical soft wares (SPSS and Excel) for data analysis. My research outcome provides evidence that the amount of non-audit services rendered by a firm’s external auditors is not associated with earnings management; therefore, the auditor’s independence is not compromised for providing NAS to their clients. This no association is an indication that, the auditor’s independence is strengthened during the provision of non-audit services a result consistent with that of prior researches (e.g., DeFond et al. (2002); Crabtree et al. (2004); Chung & Kallapur (2003)). I also found out that, the provision of audit services is not associated with earnings management, an indication of no management discretionary accruals. Therefore, the auditor’s independence is not compromised when providing a joint audit and non-audit services. These results provide an insight concerning the perceptions of auditor independence and also afford empirical evidence regarding the role that non-audit and audit services fees plays in establishing confidence among the users of firms’ annual financial report.

This thesis first examines the determinants of earnings management in an international setting using the Limited Investor Attention Model of Hirshleifer and Teoh (2003). The model predicts that investor attention reduces earnings management. I have four key findings. First, I document that financial analysts curb adjusted absolute abnormal accruals and absolute performance-matched abnormal accruals in global firms. Second, I document that institutional block-holdings curb adjusted absolute abnormal accruals across the world. Third, I document that analyst following is related to more reduction in earnings management in common law countries than in code-law countries. Fourth, I find that institutional block-holders are more effective monitors in common law countries than in code law countries. This thesis also examines the relation between investor attention and stock mispricing of abnormal accruals in an international setting using the Limited Investor Attention Model of Hirshleifer and Teoh (2003). Consistent with the model’s hypothesis that investor attention reduces stock mispricing, I document three key findings. First, I find a significant and negative correlation between stock mispricing and analyst following in global firms. Second, stock mispricing is negatively correlated with institutional ownership in U.S. firms. Stock mispricing is not significantly correlated with institutional block-holdings in global firms. Third, stock mispricing per dollar of abnormal accrual is decreasing in analyst following for sufficiently large abnormal accruals in U.S. and global firms.

This thesis first examines the determinants of earnings management in an international setting using the Limited Investor Attention Model of Hirshleifer and Teoh (2003). The model predicts that investor attention reduces earnings management. I have four key findings. First, I document that financial analysts curb adjusted absolute abnormal accruals and absolute performance-matched abnormal accruals in global firms. Second, I document that institutional block-holdings curb adjusted absolute abnormal accruals across the world. Third, I document that analyst following is related to more reduction in earnings management in common law countries than in code-law countries. Fourth, I find that institutional block-holders are more effective monitors in common law countries than in code law countries. This thesis also examines the relation between investor attention and stock mispricing of abnormal accruals in an international setting using the Limited Investor Attention Model of Hirshleifer and Teoh (2003). Consistent with the model’s hypothesis that investor attention reduces stock mispricing, I document three key findings. First, I find a significant and negative correlation between stock mispricing and analyst following in global firms. Second, stock mispricing is negatively correlated with institutional ownership in U.S. firms. Stock mispricing is not significantly correlated with institutional block-holdings in global firms. Third, stock mispricing per dollar of abnormal accrual is decreasing in analyst following for sufficiently large abnormal accruals in U.S. and global firms.

In this dissertation, I examine the impact of earnings management and expectations management on the usefulness of earnings and analyst forecasts in firm valuation. Earnings and analyst forecasts are important inputs into accounting valuation models. Their ability to reflect current and predict future firm performance can help valuation models predict intrinsic value. However, increasing earnings management and expectations management activities in recent years may have adversely affected the usefulness of these information items in firm valuation. This study shows that intrinsic value metrics estimated using manipulated earnings or forecasts have less ability to track stock prices and predict future returns through V/P ratios, providing evidence for the joint hypothesis of (i) long-term market efficiency and (ii) the negative impact of earnings management and expectations management on the usefulness of earnings and analyst forecasts in firm valuation. It contributes to the accounting literature in several ways. First, it challenges the conventional view that more accurate and less biased forecasts are necessarily of better quality and proposes to assess the quality of analyst forecasts directly by examining their usefulness. It also introduces an improved measure for expectations management and presents new evidence on (i) the usefulness of earnings and analyst forecasts in firm valuation; (ii) the negative impacts of earnings management and expectations management on this usefulness; and (iii) the overall performance of accounting valuation models in firm valuation.

In this dissertation, I examine the impact of earnings management and expectations management on the usefulness of earnings and analyst forecasts in firm valuation. Earnings and analyst forecasts are important inputs into accounting valuation models. Their ability to reflect current and predict future firm performance can help valuation models predict intrinsic value. However, increasing earnings management and expectations management activities in recent years may have adversely affected the usefulness of these information items in firm valuation. This study shows that intrinsic value metrics estimated using manipulated earnings or forecasts have less ability to track stock prices and predict future returns through V/P ratios, providing evidence for the joint hypothesis of (i) long-term market efficiency and (ii) the negative impact of earnings management and expectations management on the usefulness of earnings and analyst forecasts in firm valuation. It contributes to the accounting literature in several ways. First, it challenges the conventional view that more accurate and less biased forecasts are necessarily of better quality and proposes to assess the quality of analyst forecasts directly by examining their usefulness. It also introduces an improved measure for expectations management and presents new evidence on (i) the usefulness of earnings and analyst forecasts in firm valuation; (ii) the negative impacts of earnings management and expectations management on this usefulness; and (iii) the overall performance of accounting valuation models in firm valuation.

No description available.

The main purpose of this study is to find out the relationship between the characteristics of board composition and earnings management. It is a case study of steel industry and telecommunications industry ,which have state-owned enterprises after privatization. From the perspective of earnings management, the Modified Jones Model is used for the detection of earnings management ,and the period of the study samples is from 2005 to 2007. There are six independent variables .They are the proportion of shareholding directors and supervisors, the pledged share ratio of directors and supervisors , the proportion of shareholding foreign investors, the proportion of shareholding government, the number of independent directors and the if the board has set the labor director or not. SPSS statistical software is used to do empirical analysis. The empirical results of this study shows that there is significant positively correlated between the proportion of shareholding directors and supervisors and earnings management ; there is significant negatively correlated between the proportion of shareholding foreign investors and earnings management ; there is significant negatively correlated between the proportion of shareholding government and earnings management. Above all , part of the variables of characteristics of board composition in this study are not significantly affected earnings management , resulting in some of the hypotheses do not hold .Thus, this study suggests that the concept of corporate governance in Taiwan is not already universal yet in the study samples, and Taiwan's independent directors just set up in the beginning stages of implementation. Besides, the labor director only interest in the labor rights issues, not in earnings management issues , and also they do not have the ability to judge them.

The purpose of the study is to use the data of the companies that are listed (including OTC) in Taiwan Stock Exchange to examine whether companies engage in earnings management in the years convertible bonds are issued and de-listed. We also want to know if there is any difference of earnings management whether convertible bonds are issued domestically or abroad. Discretionary current accruals are adopted as proxies for earnings management and the regression models are used to control the related variables. The empirical results indicate that companies conduct earnings management in the years convertible bonds are issued and there is no significant difference of earnings management whether convertible bonds are issued in Taiwan or abroad. However, the data after 2001 indicate that companies that issued convertible bonds abroad have less earnings management than those that issued convertible bonds domestically. The results also show no significant difference in the years convertible bonds are de-listed. The reasons may be due to the relatively small sample size or companies may not use discretionary accruals to conduct earnings management.

none