Earnings Management And Financial Information Disclosure Of Listed Companies’ Empirical Study

In an environment of fast-growing domestic capital market, how to regulatefirms’ information disclosure behavior, and decrease their earning managementthrough enhancing the quality and transparency of information disclosure, become apressing problem for government and monitoring department. Based on the dataset oflisted firms in the security exchange of Shenzhen between2001and2011, the paperempirically investigates the impact of their information disclosure on earningmanagement behavior. The paper firstly reviews the literature and theory of earningmanagement and information disclosure, points out the potential innovation andcontribution, and then gives a brief introduction of background and logics to theimpact of information disclosure on earning management of listed companies. It thenintroduces the modified Jones’ model, KS model and fixed effect method, illustratesthe data source and variable choosing process, and shows the econometric regressionsresults. The conclusion is finally drawn on and caveats and policy suggestions arealso discussed. It turns out that, firms’ information disclosure indeed significantlydecreases the behavior of earning management. In addition, the firms’ size, equitydebt ratio and ownership concentration are positively related with earningmanagement, while the cash flow ratio and independent director are negativelyrelated. The policy suggestions contain, giving full play to the role of informationdisclosure, strengthening the capital structure, reshaping directors’role, improving thelaws and regulations of accounting, creating formal and harmonious environment.The innovation of the paper lies in two points. Firstly, it is creative in the aspectof research methods. Previous research, especially domestic earlier papers, mostlyeither use the qualitative analysis, or study the impact of earning management oninformation disclosure or the opposite relationship only. This paper instead employsthe modified Jones’ model, KS model and fixed effect methods, which can, to thegreatest extent, eliminate the causality between the earning management andinformation disclosure. Secondly, the dataset are renewed. Although some previous literatures use the method of quantitative analysis, they mostly do descriptivestatistical characteristics based on the macro datasets. Using the sample from theCSMAR dataset, this paper can not only proclaim the micro mechanism comparedwith the macro data, and also reveal the impact of information disclosure on theearning management these years, evaluate the effect of new accounting regulations in2007, and provide more and latest evidences for the negative influence of informationdisclosure on the earning management.