The Structure And Information Content Of Accounting-tax Differences Among Chinese Listed Companie

The explosive growth of studies on book-tax differences can be traced back to the Enron event and the World Com Fraud that broke out in the early 2000 s.Regulators have questioned whether companies need to disclose more information about book-tax differences,and have even proposed eliminating the distinction between accounting profit and taxable profit.Therefore,research on the causes,structure and impact of book-tax differences has important social value,and regulators need theoretical evidence as the basis for future policy making.Although the development of corporate income tax is inseparable from the support of accounting(Chatfield,1977),due to the different objectives of accounting standards and tax law,there must be differences between book income and taxable income,which will provide space for corporate tax planning and earnings management.For example,a company can obtain benefits from reporting higher profits and lower taxable income.Previous studies believe that the indicators of taxable and book income can predict profit growth,and their differences are informative and helpful in predicting earnings persistence(Lev and Nissim,2004;Hanlon,2005;Blaylock et al.,2012).The difference between accounting profit and taxable profit cannot be eliminated.Given the inevitable existence of book-tax differences,the structure of book-tax differences and its information content are the main issues studied in this dissertation.This dissertation studies the impact of booktax differences components on information users by using the reconciliations of accounting profits and income tax expenses in the income tax footnotes,especially whether there are differences in the degree of attention that different report users pay to the components of book-tax differences.The tax base of corporate income tax is the taxable income,and the difference between it and the accounting profit in the financial statements is the book-tax differences,which includes permanent differences and temporary differences.Due to the confidentiality of tax returns,it is difficult for researchers to obtain taxable income.The traditional method is to estimate book-tax differences by using financial statement data.Specifically,the income tax expense is divided by the statutory tax rate to estimate the taxable income.But there will be certain errors and the components of book-tax differences cannot be observed.Accounting Standards No.18 Income Tax in 2007 requires that companies should disclose information related to income tax in the footnotes,including the main components of income tax expenses and the description of the relationship between income tax expenses and accounting profits.However,at the actual implementation level,the disclosure content and level of detail among companies are not uniform.Since 2014,companies are required to disclose relevant information on income tax expenses,including the reconciliations of accounting profits and income tax expenses.Therefore,since 2014,the disclosure of information related to income tax expenses,especially the disclosure of various items in the reconciliations of accounting profits and income tax expenses,has been more standardized.It provides opportunities for this dissertation to accurately calculate book-tax differences and distinguish internal compositions.Due to data limitations,existing studies on the components of book-tax differences are limited to the structure of deferred tax or to study a single item,such as valuation allowance(Bauman et al.,2001;Schrand and Wong,2003;Phillips et al.,2004).They did not delve into the components of permanent differences.Although permanent differences can be derived from book-tax differences minus deferred tax expenses,their internal structure remains a black box without disclosure.However,this dissertation constructs a database of permanent differences and their components through manual collection,and further examines the impact of the components and information content of book-tax differences.This dissertation adopts the reconciliations of income tax expenses disclosed in the footnotes as a measure of the book-tax differences,and divides it into three items related to multiple tax rates,operating and accounting judgments based on economic substance.The dissertation compares the actual disclosure of data with the estimated data,and then compare their earnings forecast ability.Further,the dissertation explores the impact of the book-tax difference on the different information users.The results show that there are errors in estimating book-tax differences,and errors vary with company characteristics.Compared with estimated book-tax differences,disclosed book-tax differences have a stronger predictive ability on future earnings.This shows that disclosure of total book-tax differences has an incremental informative effect.The components of book-tax differences also have an incremental effect on the forecast of earnings,and the persistence of different components in earnings vary.Tax rates and operating-related items are more persistent than deferred tax expenses and judgment-related items.This suggests that there are significant differences in the information content of the different components of the book-tax differences.This dissertation examines the impact of book-tax differences and their components from the perspectives of auditors,rating agencies,analysts and investors based on differences in the information users’ ability to access and process internal information related to book-tax differences.The results show that(1)Book-tax differences will increase audit fee and the probability of a modified opinion,among which tax rate-related items have the greatest impact on audit fees,but will not affect the final audit opinion;deferred tax and judgment-related items will have the greatest impact on the possibility of modified audit opinions.Operating-related items have an impact on the auditor’s decision-making,but the impact is lower than other items.(2)Accounting judgment-related items in book-tax differences are negatively correlated with corporate credit ratings.The greater the absolute value of judgmentrelated items,the higher the probability of obtaining a low rating.And the related negative increase in book-tax difference or decrease in positive book-tax difference will help to improve the rating(inverted "U" shape).The absolute value of operating-related items is not significantly related to the corporate credit rating,but sighed operating-related items will improve the credit rating,indicating that the cash flow effect of this item is dominant.(3)Book-tax differences will cause analysts’ forecast divergence and optimistic bias.In terms of different components,tax raterelated items,operating-related items,judgment items,and deferred tax will all cause analysts’ forecast divergence and optimistic.However,compared with other items,judgment-related items play a major role.(4)For ordinary investors,their response to unexpected earnings of companies with large book-tax differences is smaller,mainly for companies with large judgment-related items.Investors generally underestimate the persistence of earnings.The underestimation is most pronounced in companies with small book-tax differences.Book-tax differences provide investors with information about earnings.Companies that only have high judgmentrelated items or high rate-related items do not exhibit mispricing phenomenon,and investors can obtain earnings information from these items.For companies with complex book-tax differences,items related to high judgments can cause mispricing and it is difficult for investors to understand.The contribution of this dissertation is mainly in the following aspects: First,it reveals the necessity of disclosing book-tax differences.Book-tax differences have information content.It is helpful to predict the future earnings of companies(Hanlon,2005;Wu Lina et al.,2007;Blaylock et al.,2012;Xie Xiangbing,2015).But tax return is unobservable.Previous studies have used statutory tax rate to estimate taxable income.However,in the context of multiple tax rates in China,there may be errors in the selection of statutory tax rates.This dissertation uses the information disclosure of income tax expense footnotes to measure book-tax differences,and compares them with the estimated indicators.It is found that the estimated indicators have significant errors and the error part also has a predictive effect.The research of this dissertation supports the necessity of the disclosure of book-tax differences.In the case of unobservable tax adjustments,the book-tax differences inferred from financial statement may lack some information.This dissertation also proposes a new indicator of book-tax differences,which complements the relevant literature.Second,it provides empirical evidence for the necessity of detailed disclosure of book-tax differences.The information needs of different information users vary.Financial reports should take all needs into account and provide the decision-related information they need.This dissertation uses the components of book-tax differences disclosed in the footnotes to explore its impact on different users.The results find that there are differences in the impact of different components.It reveals the necessity of detailed disclosure of the internal composition of book-tax differences.On the other hand,it also shows that under the existing disclosure,users can obtain useful information from it.In addition,previous studies only focused on the total amount of book-tax differences or only one part of them due to data limitations.This dissertation enriches the research literature on the components of book-tax differences.Thirdly,this dissertation also has certain practical implications.For financial report users,book-tax differences and their components have earningsrelated information,and increasing attention to them can help users make decisions;for regulatory authorities,encouraging companies to detail tax-related information is conducive to the healthy development of the capital market.