Book-tax conformity of earnings and the pricing of accruals

The purpose of this study is to investigate whether managerial incentives underlying book-only accruals and book-tax accruals affect stock prices, corporate leverage and dividend policy. The recent high-profile cases have generated considerable interest in understanding the firms' economic performance differences between book and tax reporting systems. Previous studies fail to examine how managerial incentives (e.g., tax incentives versus financial reporting incentives) underlying discretionary accruals affect prices, corporate leverage and dividend policy. To address this gap in the literature, the present study decomposes total accruals into discretionary book-tax accruals, nondiscretionary book-tax accruals, discretionary book-only accruals and nondiscretionary book-only accruals.; This study uses a large sample of U.S. firms over a period of 16 years (1986--2001) and applies the modified Jones model to compute discretionary book-tax accruals, nondiscretionary book-tax accruals, discretionary book-only accruals and nondiscretionary book-only accruals.; The Mishkin (1983) model reveals that the market's perceived persistence of accruals and cash flows differ significantly from the actual persistence for all earnings components. The market's inability to correctly assess the persistent of accruals is more severe for book-tax accruals than book-only accruals. In addition, a trading strategy of portfolios with high levels of discretionary book-tax accruals yields positive size-adjusted abnormal returns. This suggests that the market has difficulty interpreting book-tax accruals either because of lack of sufficient disclosure regarding these accruals or lack of understanding of their tax consequences. In addition, a system of simultaneous equations reveals that firms that use more income decreasing discretionary book-tax accruals issue less debt. Consequently, firms use income decreasing discretionary book-tax accruals as a tax shield substitute for interest expense. Finally, a simultaneous equations approach also shows that operating cash flows and book-only accruals are positively associated with dividend policy. On the other hand, book-tax accruals show no association with the corporations' dividend policy. These findings reveal that the positive association between dividends and earnings is mostly driven by the cash flow component of earnings and not the accrual components. Finally, as expected, more conservative firms pay less dividends.