No abstract

No abstract

Corporate managers make several discretionary assumptions (i.e., the rate of salary growth and assumed return rate of pension plan assets) in calculating the fair value of pension assets and value of pension liabilities. In this study, we examine the factors that can induce managers to increase (decrease) fair value of pension assets (liabilities) through pension assumptions. We use the Taiwan setting as a natural experiment because Taiwan is planning to adopt IFRS from 2013 on. The estimation of pension asset (liability) value is similar between Taiwan accounting standards (i.e., TFAS 18) and international accounting standards (IAS 19); however, fair value is only disclosed in the financial statements under TFAS 18, but is required to be recognized in the balance sheet under IAS 19. The findings under TFAS 18 can provide important implications for future adoption of IAS 19. Using two key inputs for pension pricing model, this study finds that companies are inclined to increase (decrease) the value of pension assets (liabilities) by rising (lowering) the assumed expected rate of asset returns (expected salary growth). The manipulation is more pronounced for firms with high distress risk and complex ownership structure. Prior studies find that once firms are required to recognize the fair value (i.e., adopting IAS 19) as opposed to disclose the information (i.e., TFAS 18), firms have higher incentives to manipulate the model inputs. This suggests that the regulator needs to specify the authoritative enforcement rules regarding fair value estimation, and provides more insights on firms with high distress risk and complex ownership structure.

No abstract

No abstract

No abstract

No abstract

In this study, we examine whether overconfident CEOs strive to smooth dividends. Our findings show overconfident CEOs increase dividends more as earnings increase and decrease dividends less as earnings decline, resulting in downward dividend stickiness. This asymmetric dividend payout is consistent with the selective self-attribution bias. Furthermore, the effect of managerial overconfidence on dividend stickiness is more pronounced for firms without catering incentives. In addition, overconfident CEOs do not manage share repurchases as they do for dividends. Therefore, we do not find a positive effect of managerial overconfidence on total payout stickiness.