Managerial Overconfidence, Research and Development, and Earnings Management : Perspective from Capitalization and Expense

Managerial overconfidence, originally disputed in terms of finance and accounting, has been dated back to Malmendier and Tate (2005a,b). Galasso and Simcoe (2011) studied the relationship between CEO overconfidence and innovation, focusing on patents, in which they found that overconfident CEO underestimates risks in R&D activities, resulting in more R&D expenditure. However, little literature concentrates on expense and capitalization of R&D expenditure in China, where new accounting rule applied since 2007, and correspondingly the relationship between overconfidence and preference of expense and capitalization. By Tobit model and GMM model, we found that overconfidence managers have incentives to capitalize on R&D expenditure, because accrual earnings management motivated by capitalized R&D, different from real earning earnings management triggered by expense R&D, which cuts down R&D expense directly to beautify the earnings performance, adjust the proportion of research and development. Therefore, such a preference to capitalize R&D turns the corporation to deal with accrual earnings management instead of real earnings management