- 學位論文
企業研發費用率變動對於條件式穩健衡量之影響
Impacts of Changes in Corporate R D Expense Ratio on Measurement of Conditional Conservatism
摘要
本研究旨在探討企業研發費用率變動是否會影響條件式穩健性之估計結果。本文認為，過去衡量穩健性的模型所採用之時效性差異可能部分來自於企業之研發費用行為。企業減少研發費用雖能使當期盈餘增加，然而因研發或有資產特性，透過減少研發來提高盈餘未必能增加企業價值進而提高股價報酬率；對於盈餘增加而股票報酬率未隨之上升的企業，條件式穩健模型之估計結果會顯示其盈餘更加反映好消息，而壞消息時盈餘反映消息的程度則下降，造成盈餘時效性的不對稱程度下降，導致該企業被認定的條件式穩健程度降低；反之，企業增加研發費用雖使當期帳列盈餘減少，然而增加研發費用導致的盈餘減少未必會沖擊企業價值而降低股價報酬率；在盈餘減少而報酬率卻不下降時，將使盈餘反映好消息的程度顯得更低，而反映壞消息的程度則變得更高，造成盈餘時效性的不對稱程度提高，導致企業被認定的條件式穩健程度增加。因此，研發費用率變動可能造成穩健程度被高估抑或低估。 本研究以2000年至2019年美國公司之研發相關變數來擴展 Basu (1997) 模型以進行實證分析。結果顯示，企業研發費用率之水準及其增減的確會影響穩健模型的衡量效度，且此現象可能造成個別樣本公司穩健係數因研發費用率上下波動而使得相關研究推論顯得不持續。因此，本研究建議後續使用Basu (1997) 模型進行條件式穩健相關研究時，能夠濾清樣本公司研發費用對穩健性估計之影響。若能控制研發費用這項變因，將能使得穩健性相關之研究推論更加強韌。
並列摘要
Prior studies on conditional conservatism document that earnings recognize bad news more timely than good news (Basu, 1997). However, this study argues that for certain companies such asymmetric timeliness of earnings in reflecting news may partly stem from the change of companies’ research and development (R D) expense to market value of equity ratio. Despite that reducing R D expenses will increase companies’ current earnings, the increase in earnings may not subsequently cause stock returns to rise because R D expenditures are likely to result in future economic benefits. Namely, the earnings increase but the stock returns do not rise. Consequently, earnings appear to reflect good news more but reflect bad news less. The extent of asymmetric timeliness degree decreases and such companies will then be more easily categorized by the Basu (1997) model as non-conservative firms. Conversely, although expanding R D expenses decrease companies’ current earnings, it may not lower stock returns because R D expenditures probably benefit future profitability. When earnings decrease but stock returns do not fall, earnings appear to reflect good news less but reflect bad news more. Thus, the asymmetric timeliness may be particularly pronounced and such companies will be more likely to be categorized as conservative firms. The empirical result shows that the change of R D expense ratio of U.S. firms contaminates the measurement of conditional conservatism. Moreover, the results of research on conditional conservatism using Basu (1997) model might be unstable due to the volatile characteristic of the R D expense ratio. Thus, this study suggests that future research examining conditional conservatism should take into account the impact of changes in R D expenses on the conservatism measure by including a control variable in the regression model to strengthen the inferences.