The Effect Of Managerial Overconfidence On Accounting Conservatism

Accounting conservatism improves the contractual relationship of the enterpriseand the effectiveness of corporate governance. It is of great significance toshareholders and creditors. Especially the creditors, as the external of corporate, havestronger demand for accounting conservatism; they expect to maintain their interestsby using this. However, the choice of accounting conservatism depends largely on themanagers. The traditional view hypothesizes that managers are fully rational. Underthe constraint of many contracts, such as compensation contracts and debt contracts,they make the right decisions to consistent with the interests of the enterpriseshareholders and creditors. In fact, managers are bounded rational. Their judgmentsand decisions tend to be systematic cognitive biases, and these cognitive biases maydamage the interests of investors and creditors.In cognitive psychology, overconfidence behavior is a stable irrational behavior.Many scholars’ research found that overconfidence affects corporate investment,financing, dividend policies and some other financial decisions. Recent work inaccounting examines the implications of overconfidence for managerial forecasts ofearnings and misreporting or fraud. Ahmed and Duellman（2013）used the empiricaldata of public companies in the United States, for the first time, to research therelationship between overconfidence and conservatism, but there is no similarresearch in China. This study tries to take the characteristics of the Chinese capitalmarket into consideration, and to examine whether managerial overconfidenceinfluences the accounting conservatism of Chinese publics companies. We hope tofind effective supervision mechanism to restrain managerial overconfidence behavior,and then reduce the loss of shareholders and creditors.This study uses the empirical data of Chinese A-share public companies over2008to2012as samples, to test the influence that managerial overconfidence onaccounting conservatism. Models are based on the research of Ahmed and Duellman(2013). The measure of managerial overconfidence and the control variables ofregression models are both based on Chinese special system background andcharacteristics of the capital market. To avoid the defects of single measure, this paperuses three methods: earning prediction, executive compensation ratio and GAPEX, tomeasure managerial overconfidence respectively. The measure of conservatism isbased on the firm-specific asymmetric timeliness C-Score following Khan and Watts (2009). By using descriptive statistics, correlation analysis, regression model analysisand some other empirical analysis methods to test and analyze the sample, this studyobtains the following conclusion: there is a significant negative relation betweenaccounting conservatism and managerial overconfidence behavior; and strongmonitoring can not affect this relation; also, the magnitude of this negative effectwhich managerial overconfidence have on accounting conservatism of publiccompany, is lower in China than in the United States. The additional robustness test ofthe empirical results also support the above conclusions. The accounting conservatismis of great significance to shareholders, creditors and other stakeholder. So it isnecessary to take action to restrain the negative effect. According to empirical resultsin this study, to keep accounting conservatism in the appropriate level, the authorsuggest that companies should perfect independent director system, build appropriateincentive mechanism of executive compensation and establish effective learningmechanism.