Essays on accounting conservatism

Abstract

This dissertation consists of three essays that investigate patterns of accounting conservatism in different institutional contexts. The first essay examines the cross-country determinants of accounting conservatism. Consistent with three primary explanations for conservatism (Watts, 2003), results show that accounting conservatism is positively related to suditor litigation risk, legal enforcement and tax-book conformity and findings provide direct evidence on the cross-country determinants of accounting conservatism. The second essay examines the cross-sectional determinants of accounting conservatism in Japanese insitutional environment. Japan provides a unique setting to revisit the contracting arguement for accounting conservatism. First, the joint equity and debt ownership enables Japanese financial institutions to resolve agency problem internally. Second, prior studies suggest that the choice between public bonds and bank loans depends on the agency cost of debt. Consistent with the contracting perspective, results provide strong evidence supporting my hypotheses that accounting conservatism decreases in the shareholdings by financial institutions, while it increases in the extent of public bond financing. The third essay documents the time-series variation in accounting conservatism and examines whether institutional characteristics can explain the patterns of accounting conservatism in the emerging Chinese market. The institutional setting in China provides an interesting background for investigating the time-series effect of litigation on accounting conservatism. Especially, auditors face an increasing litigation risks after a significant event, disaffiliation program, which had severed the ties between Chinese CPA firms and the sponsoring government bodies. Therefore, I hypothesize that accounting conservatism increases after the implementation of disaffiliation program and results provide strong supporting evidence. I also find that accounting conservatism decreases in the state ownership and increases in the bank financing. In sum, this dissertation provides systematic evidence on how different types of institutional factors are related to accounting conservatism. In particular, the primary findings suggest that various institutional characteristics lead to different demands for accounting conservatism. This study complements the extant literature by providing direct cross-country and cross-firm evidence on the relation between accounting conservatism and institutional factors based on those explanations proposed in recent work (Watts, 2003). The results of this study should be of interest to standard-setters, regulators and researchers.