The Confounding Effect Of Cost Stickness On Conservatism Estimates

Accounting conservatism is important concept of financial accounting.Recent years,the most common method of measuring accounting conservatism is the model of earnings and stock returns by Basu at 1997.Basu based his primary measure of asymmetric timeliness on a regression of net income on stock returns with separate slopes for positive and negative returns(proxies for good and bad news,respectively)and find that earnings incorporate bad news about future cash flows more quickly than good news.Then he interprets conditional conservatism as the higher degree of verification of good news as gains than bad news as losses.Cost is important concept of management accounting and cost stickiness refers to the asymmetric response of costs to sales increases versus decreases,which is a widely documented finding in the research on cost behavior.Throughout most of the existing literature at home and abroad,researches about accounting conservatism or cost stickiness is rich separately.But only a few studies combined accounting conservatism and cost stickiness together to find their relationship.Recently,some scholars argue that asymmetric behavior of costs can be an explanation of asymmetric timeliness.In fact,when cost components are "sticky",cost are more sensitive to sales increases than to sales decreases and because costs enter earnings with a negative sign,the direction of asymmetry is reversed for earnings.It means that costs stickiness has an asymmetric effect on earnings,which is stronger for sales decreases than for sales increases.And because sales changes correlate positi'vely with stock returns,the association between earnings and stock returns should be stronger for negative returns than for positive returns even if accounting conservatism is absent.Therefore,the estimates of the piecewise linear relation between stock returns and earnings in standard models reflect not only asymmetric loss recognition but also the confounding asymmetric effect of cost stickiness.So this thesis hypothesized that sticky costs could lead to an upward bias on average in the standard estimates of asymmetric timeliness.Meanwhile,cost stickiness likely varies with some of the standard proxies of the demand for conservatism,such as book-to-market ratio,size and financial leverage.So when cost stickiness exist,the confounding variation from cost stickiness can bias inferences in tests of theories about the causes or correlates of conservatism.Thus the past research may have wrong research findings for they do not include the confound effect of cost stickiness.The thesis suggest the second hypothesis:estimates of the variation in conservatism in standard models are distorted for these models do not control for the variation in cost stickiness.Finally,Cost stickiness may also confounds inferences about the correlates of conservatism,so in this thesis we exam the association between managerial overconfidence and accounting conservatism before and after control for cost stickiness to text if there is large bias.To test the hypothesis,this thesis based on analyzing and summarizing the documents about conservatism principle and cost stickiness both at home and abroad and then select listed IPO companies which don't have missing data during 2009-2015.Eliminate samples of finance and insurance companies,ST companies and companies whose earnings is negative.Based on the earnings and stock returns model by Basu and modify the model with cost stickiness as a missing variable to verify if there is large bias on estimation of average level of conservatism and extent of cross-sectional variation in conservatism.The results showed that sticky costs can distort standard asymmetric timeliness estimates.The standard models likely overestimate the level of accounting conservatism.After controlling the standard proxies of the demand for conservatism(book-to-market ratio,size,financial leverage),the variation of conservatism change a lot.Besides,cost stickiness also confounds inferences about the relationship between conservatism and managerial overconfidence.The relationship become uncorrelated after after control cost stickiness.At last,the thesis confirm that the asymmetry arising from cost behavior is distinct from conditional conservatism.Future empirical research about conditional conservatism should recognize the potential confounding effect of sticky costs to avoid overvalue of conservatism and change the variation of conservatism.