using operating cash flow data to predict financial distress some extensions.pdf

1、 Wiley and Accounting Research Center,Booth School of Business,University of Chicagoare collaborating with JSTOR to digitize,preserve and extend access toJournal of Accounting Research.http:/www.jstor.orgAccounting Research Center,Booth School of Business,University of ChicagoAccounting Research Cen

2、ter,Booth School of Business,University of ChicagoUsing Operating Cash Flow Data to Predict Financial Distress:Some Extensions Author(s):Cornelius Casey and Norman Bartczak Source:Journal of Accounting Research,Vol.23,No.1(Spring,1985),pp.384-401Published by:on behalf of WileyAccounting Research Cen

3、ter,Booth School of Business,University of ChicagoStable URL:http:/www.jstor.org/stable/2490926Accessed:12-04-2015 22:28 UTCYour use of the JSTOR archive indicates your acceptance of the Terms&Conditions of Use,available at http:/www.jstor.org/page/info/about/policies/terms.jspJSTOR is a not-for-pro

4、fit service that helps scholars,researchers,and students discover,use,and build upon a wide range of contentin a trusted digital archive.We use information technology and tools to increase productivity and facilitate new forms of scholarship.For more information about JSTOR,please contact supportjst

5、or.org.This content downloaded from 130.194.20.173 on Sun,12 Apr 2015 22:28:43 UTCAll use subject to JSTOR Terms and ConditionsJournal of Accounting Research Vol.23 No.1 Spring 1985 Printed in U.S.A.Using Operating Cash Flow Data to Predict Financial Distress:Some Extensions CORNELIUS CASEY*AND NORM

6、AN BARTCZAKt 1.Introduction Recently,financial statement users and regulators of publicly reported financial accounting data have argued in favor of the disclosure of detailed information on firms current operating cash flows(Harris et al.1980,FASB 1981,Smith 1982,and Thomas 1982).The FASB suggests

7、that such disclosures will allow users to assess better the amount,timing,and uncertainty of future cash flows.It states that the greater the amount of future net cash inflows from operations,the greater the ability of the enterprise to withstand adverse changes in operating conditions 1981,p.11.The

8、 presumption that historical operating cash flows enable better assessments of future cash flows,however,is based on intuition rather than on empirical evidence(Griffin 1982).Our study was conducted to assess whether operating cash flow data and related measures lead to more accurate predictions of

9、bankrupt and*Associate Professor,Dartmouth College-the Editors of the Journal of Accounting Research were saddened to learn of the death of Professor Casey in October 1984;t Lecturer,Harvard University.The authors gratefully acknowledge the helpful comments of partici-pants in accounting workshops a

10、t several universities.Suggestions made by E.Altman,P.Griffin,J.Horrigan,W.Mikkelson,T.Selling,J.Shank,C.Stickney,and P.Williamson are appreciated,as is the computing assistance rendered by D.Bower,S.Hare,D.Roberts,and G.Peterson.J.Ohlson kindly provided many of the financial statements for the bank

11、rupt firms.Funding for this project was provided by the Tuck Associates,Dartmouth College,and the Division of Research,Harvard Business School.Accepted for publication September 1984.384 Copyright C),Institute of Professional Accounting 1985 This content downloaded from 130.194.20.173 on Sun,12 Apr

12、2015 22:28:43 UTCAll use subject to JSTOR Terms and ConditionsUSING OPERATING CASH FLOW DATA 385 nonbankrupt firms.Recent studies(Gombola and Ketz 1983 and Gombola et al.1983)found that ratios based on operating cash flow load on a separate statistical factor,suggesting that operating cash flow vari

13、ables may be useful in descriptive and predictive studies involving financial ratios.Our results suggest otherwise,at least with respect to bankruptcy prediction.In another study(Casey and Bartczak 1984)we reported that accrual-based multivariate discriminant models forecasted corporate bankruptcy m

14、ore accurately than any single operating cash flow ratio.The focus of the present study is on the marginal predictive content of the operating cash flow ratios,in contrast to our previous study which examined their univariate predictive value.The poor performance of the operating cash flow ratios in

15、 the previous study does not preclude their ability to enhance predictive power when used in combination with accrual-based ratios.Previous accounting studies have been unable to document the existence of incremental information content in operating cash flow data(Bowen,Burgstahler,and Daley 1984).U

16、nlike the present study,these studies have used restrictive surrogates for measuring cash flow from operations,and thus the issue of whether operating cash flow data have incremental information value is unresolved.In section 2 we review prior bankruptcy studies that have examined the potential usef

17、ulness of operating cash flow data.Section 3 describes the methodology employed in the present study,including selection of sample firms and analyses applied to the firms accounting data.Results are presented in section 4,followed by some limitations of the study and possible directions for future r

18、elated research in section 5.2.Previous Research Most accounting and finance studies of corporate financial distress define cash flow as net income plus nonworking capital expenses,so they omit items such as changes in current assets and current liabilities which may have a significant impact on a c

19、ompanys actual cash flow from operations.This has led to suggestions for broader measures of cash flows(e.g.,Largay and Stickney 1980 and Gombola and Ketz 1983).Our definition of cash flow from operations(CFO)is essentially the same Bankruptcy was selected as the specific form of financial distress

20、and as the criterion event for three primary reasons:(1)the direct and indirect costs of bankruptcy are significant in relation to the value of the firm(Altman 1983a):(2)results from this study can be compared with previous studies of bankruptcy prediction;(3)the costs of data gathering and the prob

21、lem of interpreting the economic significance of other events(e.g.,loan default)were viewed as greater than the benefits of using events whose occurrences are arguably less subject to noneconomic factors than is bankruptcy.Thus,while bankruptcy prediction per se is not the main focus of this study,b

22、ankruptcy is nonetheless a convenient event for assessing one aspect of the information value of operating cash flow data.This content downloaded from 130.194.20.173 on Sun,12 Apr 2015 22:28:43 UTCAll use subject to JSTOR Terms and Conditions386 JOURNAL OF ACCOUNTING RESEARCH,SPRING 1985 as theirs:w

23、orking capital provided by operations,plus or minus changes in the noncash working capital accounts except for short-term indebt-edness(seasonal bank loans,nontrade notes payable,and the current portion of long-term debt).This concept of operating cash flow is also consistent with the recent FASB Ex

24、posure Draft 1981.Empirical research on the relationship between CFO and financial distress has provided only limited evidence that such data are useful in discriminating between troubled and healthy firms(table 1).None of the studies listed there was validated using separate holdout samples of bank

25、rupt companies.Moreover,the proportion of failed firms in all of the studies except that of Largay and Stickney 19801,which involved only one company,was close to or equal to 50%.This is likely to lead to artificially low overall error rates(Deakin 1977).2 Some basic differences between our study an

26、d those of previous studies are:(1)a focus on the potential marginal improvement in classification accuracy using CFO;(2)a larger number of sample firms and a smaller proportion of failed firms;(3)a validation of the results based on a holdout sample;(4)a preliminary canonical correlation analysis o

27、f the amount of variance in the operating cash flow ratios already accounted for by accrual-based ratios;and(5)the inclusion of first-order interac-tions between the operating cash flow and accrual-based ratios.The expectation of the FASB(and others)that the level of operating cash flow will serve a

28、s a useful indicator of the likelihood of financial distress does not derive from any formal theory.It is merely consistent with an apparent shift in preference by many analysts to use cash flow data for assessing a firms financial performance(Hawkins 1977).The present study neither proposes nor tes

29、ts any theory of the probability of financial distress.While such theories do exist(see Scott 1981 for a survey;also Emery and Cogger 1982),no one theory is generally accepted(Ball and Foster 1982)and none gives specific attention to our measure of operating cash flow.3.Method 3.1 SAMPLE FIRMS The s

30、ample firms are the same companies used in our previous study on operating cash flows(Casey and Bartczak 1984).Sixty firms were selected that had petitioned for bankruptcy during the period 1971-82.The bankrupt firms were a subset of the 105 failed firms used in a previous study(Ohlson 1980)and incl

31、uded firms listed by Dun and Bradstreet and the Wall Street Journal Index.Criteria for inclusion in 2 Type II error is the classification of a nonfailed firm as failed;Type I error is the classification of a failed firm as nonfailed;overall error rate is a weighted combination of Type I and Type II

32、error rates.This content downloaded from 130.194.20.173 on Sun,12 Apr 2015 22:28:43 UTCAll use subject to JSTOR Terms and ConditionsUSING OPERATING CASH FLOW DATA 387 the present study were:(1)the stock of the company was publicly traded,and(2)at least five,and in some cases six,years of published f

33、inancial statements were available prior to the failure date.The first criterion was imposed in order to undertake a related study involving market value data.The second allowed for the inclusion of a trend variable in the analyses.We also made sure that the data for the first year prior to failure were actually publicly available prior to the bankruptcy petition date(Ohlson 1980)by requiring all