新聞事件認列於盈餘之速度與非線性因素對穩健原則衡量之影響

Effects of the Speed of Recognition of News into Earnings and Nonlinearity on Conservatism Measurement

張明倫(Ming-Lun Chang)

指導教授 ： 劉啟群

國立臺灣大學/管理學院/會計學研究所/碩士(2013年)

https://doi.org/10.6342/NTU.2013.02641

摘要

Basu (1997)以盈餘對好（壞）新聞消息之不對稱反應衡量會計穩健原則，在價格領先盈餘與效率市場的假設下，以當期之正（負）股票報酬率作為好（壞）事件之代理變數，因好消息之認列門檻較壞消息高，因此壞消息會較多地反應於當期盈餘，此一盈餘不對稱反應被解釋為穩健原則存在之證明並廣為後續研究所用。林純央和劉啟群(2012)指出穩健原則存有非線性（非線性盈餘不對稱反應），即盈餘不對稱反應在新聞消息規模增大時會更加明顯，而非單純線性關係，因此衡量企業間穩健原則程度差異時，應以相同新聞事件規模作為比較基準。
　　本研究更進一步指出以股票報酬率作為新聞消息之代理變數，將忽略新聞事件本身特質對於盈餘不對稱反應之影響。本研究提出依新聞事件被認列於盈餘之速度不同而將新聞事件區分為不同類別之概念：在未考慮穩健原則之前提下，一新聞事件如被認列於當期盈餘之速度相對較慢，亦即盈餘需要當期及未來數期才能將新聞事件認列完畢，則該新聞可被歸類為未來新聞群組；相反地如該新聞被認列於當期盈餘之速度相對較快或僅需當期即可如數反應完畢，則可被分類為當期新聞群組，此為會計認列之自然情況。實證結果指出，在考慮穩健原則後此情況將會改變：當同樣面對未來群組之新聞事件時，相對於當期群組之新聞，好消息將更少地反應於當期盈餘，但壞消息卻相反地將更多地反應於當期盈餘。據此，在面對同樣未來群組之新聞時，好壞消息間之盈餘不對稱反應將被增大。過往研究未考慮新聞事件本身即存在不同之認列速度與其影響，其結果可能產生偏誤。
　　本研究亦將新聞事件認列速度差異之概念納入非線性穩健原則中，實證結果發現上述新聞認列於盈餘之速度對盈餘不對稱反應之效果在非線性模型下依然存在。此結果指出林純央和劉啟群(2012)以Basu (1997)模型建立之非線性模型只捕捉到部分穩健原則，在考量新聞事件本身認列於盈餘之速度後，非線性穩健原則之概念將趨完整。本研究建議在比較公司間之穩健原則時，不僅考量公司面對相同新聞規模，亦應考慮面對相同群組（相近認列速度）新聞，以作為比較基礎。

關鍵字

穩健原則 ； 盈餘不對稱反應 ； 非線性盈餘不對稱反應 ； 新聞事件認列於盈餘之速度

並列摘要

Basu (1997) interprets conservatism by measuring the asymmetric timeliness of earnings responses to bad news and to good news. He uses positive and negative returns as proxies of good news and bad news and refers that earnings reflect bad news in a timelier basis than good news due to the higher verification criteria for good news. Lin and Liu (2012) argue that nonlinearity would also have impacts on Basu (1997) methodology by indicating that earnings responses increase with absolute magnitudes of news and thus comparisons of conservatism among firms might be biased if not based on similar magnitudes of news.
　　In this study, I further argue that besides using returns as proxies of news when measuring conservatism, one should consider impacts of the asymmetric speed of recognition of news itself as well. I classify the news as either “Future Group of News” or “Current Group of News” according to the expectation of the relative speed of recognition of news into earnings without considering the possible impact of accounting conservatism on earnings recognition. Naturally, the recognition speed for Future Group is slower compared to Current Group without accounting conservatism. In other words, impacts of news in Future Groups would need more than one period to be reflected in earnings and thus the speed of recognition of news in Future Group into current earnings is slower. But after considering conservatism, the empirical results suggest that current earnings reflect much less good news when it is in the Future Group but contrarily reflect much more bad news if it is in Future Group in the same time. This opposite movement (decrease or increase) of earnings responses to good news and bad news when facing all future news also represents the magnification effect of the speed of recognition of news on asymmetric timeliness, compared to all current news. These effects are effective as well when considering nonlinearity of conservatism and indicate that it would be a more complete concept of nonlinearity after considering the speed of recognition of news. Comparisons of conservatism in prior studies might be more valid not only based on similar magnitudes of news but also facing similar Future or Current Group of News.

並列關鍵字

Conditional conservatism ； Asymmetric timeliness of earnings responses ； Nonlinearity in asymmetric timeliness ； Speed of recognition of news into earnings

參考文獻

Ball, R. and P. Brown, 1968. An empirical evaluation of accounting income numbers.

Journal of Accounting Research 6: 159-178.

Basu, S., 1997. The conservatism principle and the asymmetric timeliness of earnings.

Journal of Accounting and Economics 24: 3-37.

Beaver, W. H., R. A. Lambert, and S. G. Ryan, 1987. The information content of