Data Envelopment Analysis of Corporate Failure for Non- Manufacturing Firms using a Slacks-Based Model

Data Envelopment Analysis of Corporate Failure for Non Manufacturing Firms using a SlacksBased Model By D Andre Wilson Supervised by Dr. Joseph C. Paradi A thesis submitted in conformity with the requirements for the degree of Masters of Applied Science The Centre for Management of Technology and Entrepreneurship Graduate Department of Chemical Engineering and Applied Chemistry University of Toronto Copyright 2012

Data Envelopment Analysis of NonManufacturing Firms Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms using a SlacksBased Model Abstract D Andre Wilson Masters of Applied Science Department of Chemical Engineering and Applied Chemistry University of Toronto 2012 The purpose of this work was to study the ability of the SlacksBased Model of Data Envelopment Analysis in the prediction of corporate failure of nonmanufacturing companies as compared to Altman s Z score model. This research looks at nonmanufacturing firms specifically and attempts to classify companies without looking at the asset size of the firm. A DEA model based on the Altman s Z score financial ratios was created as well as a revised DEA model. The overall accuracy of the models showed the revised DEA model to be more accurate than the original DEA model as well as the Altman Z score. This indicated that bankruptcy could be predicted without the use of total assets or liabilities as variables. This also showed the ability of an SBM DEA model to predict bankruptcy. ii

Abstract The purpose of this work was to study the ability of the SlacksBased Model of Data Envelopment Analysis in the prediction of corporate failure of nonmanufacturing companies as compared to Altman s Z score model. DEA had been tested for corporate failure before, however the DEA model used was a BCC model and this was tested against Altman s original Z score model, which is an assetdominated model. This research looks at nonmanufacturing firms specifically and attempts to classify companies without looking at the asset size of the firm. Data was collected on nonmanufacturing companies that filed for bankruptcy between 2000 and 2006 for up to five years before bankruptcy. Nonbankrupt companies were found to match these, based on the SIC codes of the companies. The following data was collected for each company from the financial statements: Current Assets, Current Liabilities, Working Capital, Retained Earnings, Operating Income, Book Value of Equity, Total Assets, Total Liabilities, Number of Employees, and Number of Shareholders. The means of these variables were compared and this showed that the nonbankrupt companies had larger averages for all of these variables, except for the number of employees. The data was split into two groups, one to create the DEA model and determine a cutoff point and the other to test that cutoff point. The Altman Z model used the following formula. iii

The EBIT was replaced with Operating Income for this study due to lack of data. The first DEA model utilised the ratios from the Z score and made all numerators outputs and all denominators inputs. A second DEA model was created to exclude total assets and liabilities, to show that bankruptcy can be determined without the total asset size of the firm. The working capital was split into current assets and current liabilities for this second model and two new variables were also added, the total number of employees as an input and the total number of shareholders as an output. Altman s model classified more companies as bankrupt than DEA, whereas DEA classified more as nonbankrupt. Altman had a lower type I error in the first and second years before bankruptcy but DEA had lower type I error in the preceding years. DEA had a lower type II error overall. The overall accuracy of the models showed the revised DEA model to be more accurate than the original DEA model as well as the Altman Z score. This indicated that bankruptcy could be predicted without the use of total assets or liabilities as variables. This also showed the ability of an SBM DEA model to predict bankruptcy. All models had an increase in accuracy as the time before bankruptcy increased, though the revised DEA model had the highest accuracy in the th year before bankruptcy, showing it as a better model for predicting bankruptcy further before the date of bankruptcy. iv

Acknowledgements I would first and foremost like to thank Dr. Joseph Paradi, for all of his guidance and advice, as well as for doing everything possible to help me succeed. I truly could not have completed this without him and I am very grateful to him for sticking with me through the ups and downs of this past year and a half. I would like to thank all of the members of CMTE who helped me along the way and for all being so supportive when I needed them, in particular Angela Tran for making herself available when I needed her, as well as Haiyan Zhu for her advice. I would like to thank my family at the National Society of Black Engineers, the Engineering Outreach Office, and the Leaders of Tomorrow for helping to enrich my time at the University of Toronto. I would especially like to thank Nnaziri Ihejirika, for his constant support throughout my journey, for putting up with all of my mood swings and for always being there for me when I needed him. I would like to thank my parents, Larry Wilson and Fran DilletWilson, as well as my siblings Chauntez, Chadeau and Jazz for always supporting me no matter what. Most of all, I would like to thank God, because through Him all things are possible. v

Table of Contents Abstract... iii Acknowledgements... v Table of Contents... vi Table of Figures... ix List of Tables... x List of Symbols... xi Chapter 1: Introduction... 1 Thesis Structure... 2 Chapter 2: Literature Review... 4 Beaver s Univariate Study... 4 Altman s Multivariate Model... 6 Subsequent models... 8 Data Envelopment Analysis in Bankruptcy prediction... 13 Summary of Literature Review... 17 Chapter 3: Data Envelopment Analysis... 19 Charnes Cooper Rhodes model... 22 BankerCharnesCooper model... 23 SlacksBased Model... 2 Advantages/Disadvantages of DEA... 27 Chapter 4: Model Development... 29 Dealing with Negative Values... 30 Model Revision... 31 Chapter : Data Acquisition... 33 Chapter 6: Results and Discussion... 38 Univariate Analysis... 38 Second dataset... 41 Altman Z results... 43 Second group... 47 vi

DEA Model... 48 Revised DEA Model... Comparison of models... 8 Comments on DEA scores... 62 Chapter 7: Conclusion and Recommendations... 64 Conclusions... 64 Recommendations for Future Work... 6 References... 68 Appendix A: List of companies... 72 Group 1... 72 Group 2... 79 Appendix B: Financial Data... 86 Group 1... 86 1 year before bankruptcy... 86 2 years before bankruptcy... 89 3 years before bankruptcy... 92 4 years before bankruptcy... 9 years before bankruptcy... 98 Group 2...101 1 year before bankruptcy...101 2 years before bankruptcy...10 3 years before bankruptcy...109 4 years before bankruptcy...113 years before bankruptcy...117 Appendix C: List of paired companies...121 Appendix D: List of Altman Z Scores...123 Group 1...123 Group 2...12 Appendix E: List of DEA Scores for Original Model...128 vii

Group 1...128 Group 2...130 Appendix F: List of DEA Scores for Revised Model...133 Group 1...133 Group 2...13 Appendix G: Tstatistics for comparison of means...138 Group 1...138 Group 2...138 viii

Table of Figures Figure 1: Simak Model 1... 14 Figure 2: Simak Model 2... 14 Figure 3: Simak Model 3... 1 Figure 4: DEA Example... 20 Figure : DEA BCC vs CCR efficient frontier... 23 Figure 6: Slacks Based Model... 26 Figure 7: Profile Analysis of Variables in first DEA model... 40 Figure 8: Profile Analysis of remaining variables for revised DEA model... 41 Figure 9: Percent error of Altman Z'' Score... 44 Figure 10: Classification accuracy of Altman Z'' Model... 4 Figure 11: Classification accuracy of Altman Z'' model including grey area... 46 Figure 12: Total classification within each zone of Altman Z'' model... 47 Figure 13: Cutoff point determination on year 1 results... 49 Figure 14: Closer look at cutoff point for year 1... 0 Figure 1: Cutoff point on year 2... 1 Figure 16: Cut off point for up to years before bankruptcy... 2 Figure 17: Average DEA scores for group 1 first DEA model... 3 Figure 18: Cutoff points for up to years before bankruptcy for revised model... 6 Figure 19: Average DEA scores for group 1 in model 2... 7 Figure 20: Comparison of overall accuracies... 9 Figure 21: Comparison of bankrupt and nonbankrupt classification accuracies... 9 Figure 22: Comparison of type I and type II error... 60 Figure 23: Comparison of percent classifications... 61 ix

List of Tables Table 1: DEA Example... 19 Table 2: Number of companies in group 1... 36 Table 3: Number of companies in group 2... 36 Table 4: Number of companies in group 1 used in revised model... 36 Table : Number of companies in group 2 used in revised model... 37 Table 6: Profile analysis of group 1 bankrupt companies... 38 Table 7: Profile Analysis of group 1 nonbankrupt companies... 39 Table 8: Profile Analysis of bankrupt companies in group 2... 42 Table 9: Profile Analysis of Nonbankrupt companies in group 2... 42 Table 10: Results of Altman Z'' Model on group 1... 43 Table 11: Results of Altman Z'' model on group 2... 48 Table 12: Cutoff points for first DEA model... 4 Table 13: Classification accuracies of group 1 used to determine cutoff... 4 Table 14: Classification accuracies of cutoff point for model 1 tested on group 2... Table 1: Intersection points for each year of model 2... 6 Table 16: Cutoff points for revised DEA model... 7 Table 17: Classification accuracies used to determine cutoff points for revised model... 7 Table 18: Classification accuracies of cutoff points for revised model on group 2... 8 x

List of Symbols Z Altman's Z Score Z" Altman's Z" Score for nonmanufacturing firms EBIT Earnings before Income and Taxes WC Working Capital RE Retained Earnings OI Operating Income BVE Book Value of Equity TA Total Assets TL Total Liabilities CA Current Assets CL Current Liabilities SH Number of Shareholders EM Number of Employees Efficiency CCR DEA Score Input Oriented CCR DEA Score Output Oriented Coefficient for DEA output Coefficient for DEA input DEA output DEA input BCC DEA Score input oriented DEA Score dual multiplier form Vector of DEA inputs Vector of DEA outputs Row vector with all elements unity BCC DEA Score output oriented Vector of coefficients for DEA inputs and outputs Negative DEA slacks Positive DEA slacks DEA Score for Slacks Based Model xi

Chapter 1: Introduction When an individual or group is managing a company or investing in a company, something of great importance is the corporate health of that company. A very valuable piece of information would be if that company is headed for corporate financial stress or failure. In the past, many have attempted to predict corporate failure before it occurs. One of the most prevalent methods is to use financial ratios to determine the health of a company. A number of studies have been done to use the information from financial statements, particularly financial ratios to predict failure [BEAV67]. A prominent method of predicting bankruptcy is the Altman Z score [ALTM68]. Edward Altman used Multiple Discriminant Analysis to create a model that uses basic financial ratios in a linear formula to give a score. This score is used to determine whether a company is at risk of corporate stress or failure, whether they are healthy or whether they are classified in an undetermined zone or grey area. Later, in 1997, Paul Simak and Joseph Paradi attempted to compare this method to another analysis type, DEA or Data Envelopment Analysis [SIMA97]. This method used the BCC version of the DEA model and showed that this model was preferable to Altman ZScore for predicting bankruptcy of companies up to 3 years before the bankruptcy date. The problem with these methods is that they were generalized and in particular, did not look at firms that were nonmanufacturing, i.e. there was a large focus on the assetsize of the firms involved [GRIC01B] [HILL04]. In more recent times, more companies are nonmanufacturing Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 1

and are more service oriented and thus have less focus on the overall assetsize of the company [WORL09]. In response to the focus on manufacturing companies in his original model, Altman created another model, the Altman Z score [ALTM02]. With this score he tested nonmanufacturing firms and redeveloped the coefficients and cutoff points of the original model to suit. This model however, is still substantially based on assets, though more companies are becoming focused on service, are people driven, and do not have a large asset base [WORL09]. Thus an investigation of the Altman Z score for nonmanufacturing firms was done here. This research tests the following: The feasibility/accuracy of DEA as a predictor of bankruptcy in nonmanufacturing firms as compared to the Altman Z model The effectiveness of a SlacksBased DEA model as opposed to a BCC DEA model for bankruptcy prediction The accuracy of bankruptcy prediction without regard to the total asset size of a company and to add a 'human capital' value that focuses on the serviceoriented side of nonmanufacturing firms. The effectiveness of both Altman and DEA in the prediction of bankruptcy for up to five years before the bankruptcy date. Thesis Structure This thesis is structured as follows: Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 2

o Chapter 2 outlines key papers that have contributed to bankruptcy prediction literature with the use of financial statements. o Chapter 3 outlines Data Envelopment Analysis as an efficiency measurement technique, giving an overview of different DEA models as well as mathematical formulas, key terminology and appropriate examples. o Chapter 4 outlines the method for developing the model and the reasoning behind the innovations that were made to the model. o Chapter outlines the method for data acquisition and summarizes the quality of the data used. o Chapter 6 presents the results of the analysis and comments on the findings. o Chapter 7 summarizes the key findings and suggests recommendations and potential areas for further research. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 3

Chapter 2: Literature Review Having an understanding of the studies related to bankruptcy prediction and analysis techniques are imperative to create a model that is not redundant in literature and that can contribute to the existing knowledge of analysis of corporate failure. This section outlines some of the most relevant findings with bankruptcy prediction and summarizes the implications of this research based on those findings. Beaver s Univariate Study One of the first attempts to predict insolvency or bankruptcy was done by William Beaver in 1967 [BEAV67]. Beaver defined failure as the inability of a firm to pay its financial obligations as they mature and a financial ratio as a quotient of two numbers, where both numbers consist of financial statement items. He also introduced a third term predictive ability which is essentially the usefulness of a data item in identifying an event before it occurs [BEAV67]. Beaver collected data from Moody s industrial manual between 194 and 1964, inclusive. Each failed firm from Moody s was compared to a nonfailed firm in the same industry of the same asset size. At the time there was statistical reason to believe that a larger of two firms will have less probability of failure even if they have identical financial ratios. Therefore he believed that firms of different assetsizes could not be accurately compared [ALEX49]. Beaver compiled 30 ratios and showed 14 to be the most effective: Cash flow/total debt Net income/total assets Total debt/total assets Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 4

Current assets/total assets Quick assets/total assets Working capital/total assets Cash/total assets Current assets/current liabilities Quick assets/current liabilities Cash/current liabilities Current assets/sales Quick assets/sales Working capital/sales Cash/sales Beaver s results ultimately showed cashflow to total debt ratio as the best predictor, with total debt to total assets as second best. He noted that the most crucial factor was the net liquid asset flow supplied to the reservoir while the size of the reservoir was the least important factor. Beaver also visited the concept of likelihood ratios. The likelihood ratio (LR) is the ratio of these to values. ( ) ( ) (1.1) Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

( ) ( ) A likelihood ratio could be found for every interval for each of the financial ratios in each year before bankruptcy. However, Beaver was inconclusive in his analysis of the likelihood ratios. He stated that in the year before failure the likelihood ratio mirrored the financial ratio, however in years before that the results varied greatly [BEAV67]. He also stated that though his work was univariate, it would be valuable to consider a multivariate approach. This is where Altman stepped in [ALTM68]. Altman s Multivariate Model In 1968, Edward Altman attempted the first multivariate approach to bankruptcy prediction. The analysis technique that he adopted was MDA, multiple discriminant analysis. In Altman s time, MDA was not as popular as regression analysis and was used mainly in biological and behavioural sciences [ALTM68]. MDA is a statistical technique used to classify an observation into one of several a priori groupings dependent upon the observation s individual characteristics. [ALTM68]. It is usually used to classify the variable into a qualitative group e.g. male or female, bankrupt or nonbankrupt [ALTM68]. The process used for MDA was first to establish groups, which could be more than two, and then collect data for objects within each of those groups. Then a linear combination is created from the data collected that will best discriminate between the groups. This is done by assigning coefficients to each piece of data. For the case of bankruptcy, a coefficient is assigned Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 6

to each financial ratio chosen and the output of the linear combination is a number that can assign a firm into bankrupt or nonbankrupt. MDA allowed for the entire profile of variables to be analyzed simultaneously rather than individually [ALTM02]. To develop the model Altman took a sample of 66 corporations with 33 firms in the bankrupt group and 33 in the nonbankrupt group. All bankrupt firms were manufacturers that filed a bankruptcy petition under Chapter 11 of the National Bankruptcy Act between 1946 and 196. The nonbankrupt firms were selected by a paired sample method (similar to Beaver) [BEAV67]. A list of 22 potential ratios was compiled which were split into five standard ratio categories: liquidity, profitability, leverage, solvency and activity ratios. From the list of 22, five ratios were selected to be able to do the best overall job at collectively predicting bankruptcy. These were selected based on: (1) statistical significance of various potential functions while determining the relative contribution of each individual variable, (2) the intercorrelation between the variables, (3) the predictive accuracy of various profiles and (4) judgement of the analysis. [ALTM68] The final model was: (1.2) Where: (1.2a) (1.2b) Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 7

(1.2d) (1.2e) With cutoff zones: Altman found a classification accuracy of 83.% for his model and he showed that his model could predict bankruptcy up to three years before the bankruptcy date. Subsequent models Later in 1972, Edward Deakin revisited Beaver s analysis. [DEAK72] He used the 14 ratios that Beaver found to be most effective and attempted to use a discriminant analysis similar to Altman s. Deakin also attempted to look at data up to years before the date of bankruptcy. In his analysis he found that the significance for each ratio changed across the years. And he found that he was only able to get significant prediction result for up to 3 years before the date of bankruptcy. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 8

In 1980, James Ohlson attempted an alternative method of bankruptcy prediction using a probabilistic approach [OHLS80]. Ohlson looked at data between 1970 and 1976. Essentially for this method he looked at the vector of financial ratios and determined a vector of parameters for those ratios and looked at the probability of bankruptcy for those ratios and parameters. He then attempted to find a cutoff probability point between zero and one for bankruptcy and nonbankruptcy. The ratios that Ohlson employed were: Size = log(total assets/gnp pricelevel index) TLTA = Total liabilities divided by total assets WCTA = Working capital divided by total assets CLCA = Current liabilities divided by current assets ONENEG = One if total liabilities exceeds total assets, zero otherwise NITA = Net income divided by total assets FUTL = Funds provided by operations divided by total liabilities INTWO = One if net income was negative for the last two years, zero otherwise CHIN (NIt NIti)/(INItI + INItil), where NIt is net income for the most recent period. The denominator acts as a level indicator. The variable is thus intended to measure change in net income. Ohlson, however did not find promising results with this model as compared to Altman s model and thus is not commonly used today. In 1984, Zmijewski explored the potential methodological drawbacks of the previous bankruptcy prediction techniques [ZMIJ84]. His main issue was that previous studies that had used nonrandom samples, i.e. bankrupt and nonbankrupt groups had predelineated before modeling. Zmijewski attempted to use random sampling and incorporated a probit model to Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 9

test bankruptcy. The firms chosen for this study were from the American and New York Stock Exchanges with SIC codes of less than 6000 and were obtained between 1972 through 1978. What Zmijewski did was to create a variable B where if B>0 then the company is at risk of bankruptcy. His model is below: ( ) ( ) (1.3b) Where : o ROA = Net Income to Total Assets (Return on Assets) o FINL = Total Debt to Total Assets (Financial Leverage) o LIQ = Current Assets to Current Liabilities (Liquidity) o u = Normally Distributed Error term. However, Zmijewski found that his results were qualitatively similar to those that use nonrandom sampling and that there was no apparent improvement on the overall classification rates [ZMIJ84]. In the 90s there were many critiques of bankruptcy prediction. In 1993, SuJane Hsieh, criticized methods for determining the cutoff point of bankruptcy [HSIE93]. Some issues that were pointed out were the fact that the cutoff point is determined by trial and error not by statistics and that the cutoff point is determined without considering the relative loss for Type I and Type II errors. Hsieh derived a modified Bayesian decision model to estimate an optimal cutoff point for bankruptcy prediction models. A function was added in this model to account for the error costs of Type I and Type II errors and attempted to minimize these costs and not simply the probability of the error. However though Hsieh came up with this method for determining Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 10

the cutoff point, it has never actually been applied to previous bankruptcy models to determine its effectiveness vs. the common trial and error approach. In 2001, John Grice and Michael Dugan noted another drawback that models may not be as effective outside of the time period in which that model was created [GRIC01A]. That same year Tyler Shumway attempted to create a bankruptcy prediction method using a hazard model to account for changes over time [SHUM01]. He collected data for over 31 years and used the same ratios that Altman had used in his ZScore model. Shumway model, though it showed results better than Altman s in the first year before bankruptcy, had a significant decline in accuracy before the second year before bankruptcy. It can be seen that many bankruptcy models have used Altman s model as a benchmark for bankruptcy prediction. In 2001 another study was done by John Grice, along with Robert Ingram to look at the generalizability of the Altman Z score model [GRIC01B]. Grice looked at data between 1988 and 1991 and again showed that Altman s model was not as accurate during that time as it was on the time that it was developed. It was also shown that Altman s model was significantly more effective at predicting bankruptcy at a sample of specifically manufacturing firms than for a general dataset of companies. In 2004, a study was then done by Sudheer Chava and Robert Jarrow to look at the industry effects in bankruptcy prediction [CHAV04]. Data was collected from 1962 to 1999 and firms were taken from the AMEX, NYSE and NASDAQ listings. This study looked at both yearly and monthly intervals and showed that monthly intervals had the potential of being better predictors of failure if the data can be collected. A hazard model was run on the variables from Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 11

Altman s model [ALTM68], Zmijewski s model [ZMIJ84] and Shumway s model [SHUM01] and showed that industry groupings had a significant effect on the slope and intercept coefficients in these models. In 2004, Stephen Hillegeist, Elizabeth Keating, Donald Cram and Kyle Lundstedt attempted to use an options pricing model to look at the probability of bankruptcy [HILL04]. However again this model looked only at manufacturing firms to compare to Altman s Z score [ALTM68] and Ohlson s model [OHLS80] and it was suggested by the authors that the coefficients should be updated for industry adjustments. From the literature review it can be seen that industry is a factor in bankruptcy prediction. One of the most wellknown authors of bankruptcy prediction, Edward Altman also made another model based from his Altman Z score, specifically for nonmanufacturing firms [ALTM02]. This model is shown below. (1.4) Where: (1.4a) (1.4b) (1.4c) (1.4d) With cutoff zones: Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 12

As shown in studies such as the 2001 study by Grice [GRIC01B], Altman s original model is very much focused on manufacturing firms as are other models. Altman model for nonmanufacturing firms, called the attempts to look at an alternate industry set, but has not been studied much in subsequent papers and should be explored further. Data Envelopment Analysis in Bankruptcy prediction Comparing Altman Z score model to results from Data Envelopment Analysis (DEA) was done in 1997 by Simak and Paradi. The idea was to show that DEA would also be a valid model in the prediction of corporate failure [SIMA97]. Simak stated that when planning to invest in a company, most would like to know if the risk is acceptable for the return [SIMA97]. This is one of the main reasons why early warning indicators for corporate distress are widely researched. One of the more common methods is regression based multivariate ratio analysis, however it has many shortcomings associated with it so there are needs to explore other methods. Simak chose to use Data Envelopment Analysis (DEA) for several reasons. The advantages of DEA are outlined in the following chapter. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 13

In Simak s model, the focus was on the Z score model. He split this model into a DEA model by making all of the denominators into DEA inputs and all of the numerators into outputs. He had the following model: Figure 1: Simak Model 1 Simak made two revisions to this model. First he changed the market value of equity to the book value. This allowed for his DEA model to be applied to private companies whose shares did not have market values, and also that the book value was more readily available on the balance sheet of a company [SIMA97]. This gave the model shown in Figure 2. Figure 2: Simak Model 2 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 14

Simak also changed the EBIT, Earnings before Interest and Tax, with Operating Income [SIMA97]. The EBIT is defined as follows: Whereas the operating Income is defined as: The NonOperating Income is mostly miscellaneous income such as interests on notes receivable, capital gains/losses etc. Replacing EBIT with Operating Income, the model focuses more on the management s competence at running the normal operations of the firm. This model is shown in Figure 3. Figure 3: Simak's Model 3 Simak collected data from New Generation Research Inc. [NEWG11] for companies that applied for bankruptcy between 1991 and 199, which included 426 companies with assets between $1 million and $1 billion. From these a random sample of 43 companies was selected. These 43 companies were matched with nonbankrupt counterparts based on their SIC numbers. The companies had to have the first 3 digits of the SIC number match. Data was classified into January 1993, January 1994 and January 199, and each company s data was allocated to these groups based on the closest proximity to their fiscal year end. Three groups were created from Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 1

that dataset, one year before bankruptcy, two years before bankruptcy and 3+ years before bankruptcy. The Jan 1993 data was used to determine a cutoff point for the models and this was tested on the other years. Trial and Error determined 0.66 as a cutoff point and Simak found 7% accuracy at classifying bankrupt firms, 6% accuracy at classifying nonbankrupt firms and an overall accuracy of 63%. Simak commented on the low classification accuracy of nonbankrupt firms and that the walking wounded should be considered, i.e. companies that were near bankruptcy but still survived [SIMA97]. He also noted that narrowing down to more specific industries could likely increase the accuracy of the results. Simak used the BCC Inputoriented DEA model for all of his analyses. He noted that he tried some studies using the outputoriented BCC and obtained similar results, however he stated that perhaps using different DEA models for analysis could be of interest [SIMA97]. Another study was done in 2004 by Anja Cielen, Ludo Peeters, and Koen Vanhoof on bankruptcy prediction using DEA [CIEL04]. This study compared a linear programming approach, a customized data envelopment analysis model and a rule induction/decision tree model. This study looked specifically at companies in Belgium that declared bankruptcy between 1994 and 1996, inclusive. These authors selected 11 ratios that they had found from previous literature. These ratios were: Equity ratio Equity/total assets Retained earnings/total assets Retained earnings/total assets Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 16

Expired taxes Tax and social security charges/short term debt Cash ratio Cash/restricted current assets Inventories Work in progress, finished goods and contracts in progress/ current assets Financial debtratio Amounts payable within one year at credit institutions/short term debt Gross return Operating cash flow before taxes/total assets Coverage of debt Cash flow before dividends/total debts Net return Operating profit/loss after depreciation before financial charges and taxes/total assets Current ratio Current assets/short term debts Quick ratio Amounts receivable within one year + Investments + cash/ amounts payable within one year Leverage or debt ratio External liabilities/total assets Ratios that were found to have a positive correlation with bankruptcy were inputs to the model and ratios that with a positive correlation were made to be outputs to the model. This study also showed DEA to be a better predictor of bankruptcy than the LP model or the decision tree model. DEA has not been studied extensively as a predictor of bankruptcy but from the studies that have been done there seems to be potential for this analysis technique. Summary of Literature Review There is a rich literature on bankruptcy prediction using financial data and we included here only the most relevant to our work. Some have included many different models, from univariate analysis [BEAV67] [MERW42] to MDA [ALTM68][DEAK72], to DEA [SIMA97] [CIEL04] Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 17

as well as with other forms such as neural networks [BACK94], and hazard models [SHUM01], yet Altman seems to be referenced the most frequently [SIMA97][DEAK72][CIEL04][SHUM01]. Though Altman s original model has been modified to apply to different areas of industry, i.e. the Altman Z model [ALTM77] [ALTM02] this model has not been studied in much detail. Similarly with DEA, the few papers using this technique for bankruptcy prediction has shown it to be a promising method [SIMA97] [CIEL04] but not much has been done in this aspect particularly with respect to nonmanufacturing firms. This paper attempts to look at that niche of bankruptcy prediction for nonmanufacturing firms using data envelopment analysis. It also attempts to focus less on asset size, as do many other models [ALEX49] and attempt to incorporate the human capital [WEAT03]. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 18

Chapter 3: Data Envelopment Analysis The purpose of this section is to give brief overview of DEA and to provide a description of the main DEA models used as well as the model that will be used in this research. Data envelopment Analysis, DEA, is a fractional linearprogramming model used to measure the relative efficiency of different DecisionMaking Units (DMUs) [COOP78]. Efficiency in the most simplistic sense is just the measure of the output of a certain thing (DMU) relative to the input to that DMU. It is calculated with equation (3.1). (3.1) This calculation can be used to measure different DMUs against each other. These DMUs can be anything that has certain characteristics that can be used as inputs and outputs to determine the efficiency of said DMU. For example, we can look at the efficiency of a hospital, it might measure its efficiency by the number of patients it treats per number of doctors. The hospital is the DMU, the number of patients treated is the output and the number of doctors is the input. If we had data for a number of different hospitals we could compare them using theirs efficiency. For example: Table 1: DEA Example Hospital 1 2 3 4 6 7 Doctors 2 3 3 4 6 7 8 Patients 1 1 3 3 4 6 Efficiency (Patients/Doctors) 0. 0.33 1 0.7 0.67 0.714286 0.7 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 19

Patients By comparing these institutions we can see that Hospital or DMU 3 is the most efficient. In DEA, this DMU would be the most efficient and apparently would have nothing to improve. The other DMUs would be compared to this one in terms of efficiency i.e. all other DMUs would want to reach an efficiency of 1 to be similar to DMU 3. If we plot this on a graph we can visualize this [COOP07]. 8 7 6 4 3 2 1 0 0 2 4 6 8 10 Doctors Figure 4: DEA Example The line passing through the point (3,3 and 0,0) corresponds to the efficiency of 1 which is what DMU 3 has. This line is known as the efficient frontier. All other DMUs are enveloped by this frontier. DEA states that all DMUs that fall on the efficiency frontier are said to be efficient whereas all others below it are inefficient. But DEA takes this a step further. Suppose you have more than one input or more than one output. You would have to come up with another way to measure efficiency. DEA does this by making a virtual input and virtual output, keeping the equation for efficiency similar [COOP07]. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 20

(3.2) We can take the virtual input as a sum of the inputs and the virtual output would be defined similarly. Going back to the hospital example, suppose another input was the number of nurses. The efficiency would then be: However there is a flaw in this. One could argue that a doctor would be a more important input than a nurse. Perhaps for this hospital a nurse can only treat 1 in every 4 patients that comes in, and a doctor must be present to treat the other 3. Thus the doctor should be treated as a more important input. We could say that the importance or weight of the doctor is three times that of the nurse. Therefore the efficiency becomes: The virtual input now depends upon the weighted sum of the inputs. This is the case in DEA. The virtual input is defined as a weighted sum of the inputs and likewise the virtual output is defined as a weighted sum of the inputs. (3.3) (3.4) Where v and u are the weights for the inputs and outputs respectively. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 21

It is not always straightforward to determine the weight of a particular input or output, especially when there are a large number of inputs/outputs. Linear programming is used to solve this. DEA has two main models using a fractional LP to calculate the frontiers, the CCR and the BCC model. Charnes Cooper Rhodes model The CCR model or Charnes Cooper Rhodes model, used a constant returnstoscale (CRS) approach [COOP07]. It is designed to find the optimal weights for each input and output for each DMU. If we take the equation for the efficiency that was shown above we can solve with the following formula. Subject to: ( ) (3.) We can change this into a linear programming model by simply splitting the efficiency ratio into the following: Subject to: ( ) (3.6) Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 22

Patients This would be an outputoriented model. This can also be done in another way, called an input oriented model [COOP07]. As shown in Equation (3.7). Subject to: ( ) (3.7) BankerCharnesCooper model The BCC or BankerCharnesCooper model uses a variable returnstoscale (VRS) approach [BANK84]. To visualize the difference between CRS and VRS let us go back to the hospital example. The efficient frontier outlined in Figure 4 would be a CRS efficient frontier. However for a VRS model, the efficient frontier would be as shown in Figure. 8 7 6 4 3 2 1 0 0 2 4 6 8 10 Doctors Figure : DEA BCC vs. CCR efficient frontier Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 23

Notice now there are three points on the efficient frontier rather than DMU3 only. Portions of the VRS frontier shows different returnstoscale behaviour, which is variable depending of the size of the DMU's scale, and allows for more efficient DMUs and also that the inefficient DMUS are now closer to the frontier and are thus more efficient. In CRS there is a constant returns to scale and the frontier is simply a straight line this means that for a unit of input the output production is always the same while the VRS frontier is piecewise linear and has different returns to scale depending on the DMU's scale. The BCC model also has output and input oriented models. The input model is shown below: Subject to: (3.8) Where is a scalar. and are the vectors of the inputs and outputs respectively for all DMUs, and and are the vectors for the DMU being optimized.. is a vector of the coefficients for the inputs and outputs with all elements nonnegative and is a row vector with all elements unity [COOP07]. The dual multiplier form is: Subject to: (3.9) Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 24

, free in sign Again and are the vectors of the inputs and outputs respectively, but here rather than, and are the vectors of variables for the inputs and outputs, similar to the example of virtual efficiency above. The output version is: Subject to: (3.10) The dual form is: Subject to: (3.11), free in sign SlacksBased Model Another model is the SlackBased Model, or SBM, which is an Additive DEA model. Additive DEA models, rather than establishing the radial projection onto the efficient frontier (input or output directions), it optimises the positive and negative slacks and with the projection is not radial that is, the suggestion for improvements show both input reductions and output Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 2

Output augmentations thus the nonradial projections. The radial approaches look at the changes that occur proportionally, i.e. a projection from an origin, whereas SBM looks at the nonradial projection, or inputs that change nonproportionally, i.e. looks at each point as its own origin. This is shown in Figure 6. 7 6 4 3 2 1 s s+ 0 0 2 4 6 8 10 Input Figure 6: Slacks Based Model The is the negative slack and the is the positive slack. SBM uses the values of these slacks to calculate the efficiency of the units, unlike the BCC or CCR models which would use a ratio of the distance from the point to the efficient frontier with the distance to the origin. SBM models are units invariant but not translation invariant [COOP07]. The model for SBM is shown below. Input oriented: Subject to: Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms (3.12) 26

Output Oriented model: Subject to: (3.13) Advantages/Disadvantages of DEA DEA has several advantages over other analysis techniques, such as regression. DEA does not take an average it looks at an efficient frontier and attempts to project DMU performance onto that. It does not have fixed coefficients for the variables, i.e. it is nonparametric; it varies the coefficients of each DMU to maximize their efficiency. In this way it also focuses on individual inputs/outputs by analyzing the coefficients produced for each one. One of the biggest advantages of DEA is the ability to handle multiple inputs and outputs simultaneously, Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 27

unlike regression, and these inputs and outputs can have different units and still be analyzed together. A disadvantage of most DEA models is that they do not account for random error and assume that all deviations from the frontier are due to lack of efficiency of the DMU [TRAN07]. While there are stochastic and fuzzy DEA models, they are seldom used because the error distribution in the variables for the former and the fuzzy membership functions for the latter are hard to ascertain for real data sets. DEA has decreased accuracy with small sample sizes and must typically have a minimum number of DMUs. The rough rule of thumb is: { ( )} (3.14) Where n, m and s are the number of DMUs, inputs and outputs, respectively [COOP07]. Some DEA models cannot process negative numbers and allowances must be made for that. Also most DEA models, because they assign coefficients based on maximizing efficiency, can end up giving some coefficients zero or near zero values to make the DMU look the best it can be and often this is not acceptable, but there are techniques in DEA that can address this aspect satisfactorily. Lastly, DEA only looks at relative efficiency, comparing DMUs to other DMUs in their set, and does not look at the actual optimal efficiency. Some work has been done on this through the creation of a theoretical frontier [CHAR94]. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 28

Chapter 4: Model Development The section outlines the method used to develop the two DEA models used in this study, the first is based on Altman s Z model and the second is a revised model attempting to test the dependence on asset size and the relevance of human capital. The first phase of this model is to compare it with the most popular bankruptcy prediction model, the Altman Zscore. Edward Altman had three main Z score models; the original general Zscore, the Z model and the Z model, the latter of which is the model of interest here [ALTM78]. The Z score was designed for nonmanufacturing firms [ALTM02] and is defined as follows: (4.1) Where: (4.1a) (4.1b) (4.1c) (4.1d) With cutoff zones: Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 29

This Z score was used as the basis to create the first DEA model. The ratios were split and recombined into virtual inputs and outputs for the DEA model. All of the numerators were made to be outputs and the denominators were inputs in the model. The ratios were not inputted directly as it has been shown that ratios used as inputs or outputs in DEA can affect the accuracy of the model [SIGA09]. Due to data availability, EBIT was substituted for operating income. It has been shown that Operating Income is also a valuable indicator of corporate health in DEA [SIMA97]. Therefore the model consisted of: Outputs: Working Capital (WC) Retained Earnings (RE) EBIT or Operating Income (OI) Book Value of Equity (BVE) Inputs: Total Assets (TA) Total Liabilities (TL) And a virtual efficiency of: (4.2) Dealing with Negative Values The DEA model chosen was a SBM, or slacksbasedmodel, due to lack of previous work on corporate failure being done with this DEA model type. However, there is the issue with negative values. Many bankrupt companies will have negative values for their working capital, Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 30

retained earnings, operating income and even equity. However, negative values cannot be computed with this model. Not much research has been done to accommodate negative values in certain DEA models [KERS09], thus some sort of solution would have to be found to accommodate them. Hence the model was revised. Each output was split into a positive and negative measure. For example, Working capital (WC) was split into WC+ and WC, WC+ being the positive values for Working Capital and WC being the negative. The absolute value of the negative variable would be taken but it would then be placed in the model as an input rather than an output, i.e. WC would be an input, whereas WC+ would be an output. This method is essentially saying that Working Capital is an output, and therefore should be made as large as possible to optimize the efficiency. Working Capital is an inflow to the company. However a negative Working Capital can be view as an outflow from the company and therefore should be minimized. Thus it would be defined as an input, as the inputs to the model are minimized. Therefore we get a new visualization of the virtual efficiency. (4.3) Model Revision The model was then revised to accomplish one of its main purposes, to see how accurately bankruptcy can be measured regardless of assetsize. Hence the model was run again without total assets in the model. The total liabilities input variable was also removed and working capital was split into current assets and current liabilities. In an attempt to test the relevance of human capital, which is important to smaller, nonmanufacturing firms to the model, the number of employees and the number of shareholders was added to the model. Number of Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 31

employees was added to introduce the measure of individuals as a contributor to the efficiency of a company. The number of shareholders was added because for many smaller nonmanufacturing firms the shareholders have decisionmaking power and invest both time and money that contribute to the success of a firm. The number of shareholders investing in a business can also be seen as a reflection of the financial wellbeing of a company as viewed by the public. The revised model had the following virtual efficiency. (4.4) Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 32

Chapter : Data Acquisition One of the most critical components of this research is the collection of viable data. This was not an easy task as there are few databases for bankrupt companies. This section will discuss how the data collection was done, how companies were selected and which data was collected on each company. One of the first tasks was to find a list of bankrupt companies. This is difficult as once a company has reached failure it is usually removed from most exchange listings and other databases. However a list of bankrupt companies was found at New Generation Research Inc. [NEWG11], which was also used by Simak [SIMA97]. This site has filings for companies that date back to 1986 listed in alphabetical order of the company name. This list was narrowed down to those that could be classified as nonmanufacturing or servicebased firms. These companies must also have filed for bankruptcy between the years of 2000 and 2006. The reason for these dates was that more recent filings would be more easily obtained, and more easily compared to current companies. Bankruptcy filings from 2007 to present were not selected due to the economic recession taking place. During that time, significantly more bankruptcy filings took place than in previous years and it was considered that data from this time period could skew the results. The companies filing bankruptcy during that period would be more so for external reasons, which is not what is being studied in this project. This list consists primarily of companies filing for bankruptcy in the United States. No other extensive list of bankrupt companies was found and thus the focus of this work will be on the United States and Canada. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 33

Once the list from bankruptcydata.com was examined the names of companies were crossreferenced with the Mergent Online database [MERG08]. From that database the SIC, NAICS and incorporation date of the companies were extracted. The companies were also confirmed to be resident firms in the US or Canada and that they were publicly traded. Data was also collected on the exchange that the company was listed on before its bankruptcy. However, most of these companies were primarily traded overthecounter or through National Bulletin Board trading. For each of these bankrupt companies, financial data was compiled for up to years before the date of bankruptcy being filed, as it was shown that there is potential to predict bankruptcy up to years in advance [BEAV67] [MERW42]. Some companies did not have a full years of data and thus only had the number of years before bankruptcy collected. Whenever it was possible to identify them, the companies that had filed for bankruptcy but did not fail were excluded from the study. Many of these companies filed for bankruptcy for reasons other than complete insolvency, some liquidated due to legal issues, and others though they were suffering financial distress, filed in an attempt to reorganized and restructure their corporate strategy and alleviate some debt. These companies, similarly to the walking wounded described by Simak [SIMA97] are difficult to properly classify. Data from the full Balance Sheets, Income Statements, Cash Flow Statements and Retained Earnings were collected. From the Balance Sheet, current assets, total assets, current liabilities, total liabilities, retained earnings and shareholders equity values were extracted. From the Income Statement, the operating profit was calculated using the formula Net Sales Cost of goods Expenses. The number of employees and number of shareholders were also collected for the second model. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 34

Once the data was collected for the bankrupt companies, healthy companies, or nonbankrupt companies, were then found. A healthy company was chosen for every bankrupt company based on SIC number and on the years of health. Healthy companies had to be in existence at least years after the bankruptcy of their bankrupt counterpart. Healthy companies also must not have filed for bankruptcy during the time that they are being compared to the bankrupt counterpart. The same financial data was collected for the healthy company as the bankrupt counterpart within the same years. For example, if a bankrupt company filed bankruptcy in 2002, financial data was collected for 19972001. The healthy company would have to have been in existence and not to have filed for bankruptcy between the years of 1996 to 2006. In some cases a suitable healthy match could not be found and thus the number of bankrupt companies exceeds the number of nonbankrupt ones. Data was organized by year before bankruptcy, up to years before bankruptcy. The companies collected was split into two groups, one which would be used to create the model and determine the cutoff point for the DEA model, and the second which would be used to test the model. The number of companies used in the Altman test and in the first model is the same, however in the second/revised model; some companies were omitted due to lack of data on the number of employees or shareholders. The numbers are shown Table 2. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 3

Table 2: Number of companies in group 1 Year before Bankruptcy Number of Bankrupt Companies Number of Healthy Companies 1 44 31 2 44 31 3 43 31 4 42 29 40 28 The number of companies analyzed decreased each year due to the lack of financial data available. Table 3 shows the numbers for the second group, used to test the model. Table 3: Number of companies in group 2 Year before Bankruptcy Number of Bankrupt Companies Number of Nonbankrupt Companies 1 46 41 2 4 41 3 46 41 4 43 40 38 37 Though the same companies were run through both the first and second models, the number of companies was not the same due to lack of data, particularly in the number of employees and shareholders. The total number of bankrupt companies are analyzed is shown in Table 4. Table 4: Number of companies in group 1 used in revised model Year before Bankruptcy Number of Bankrupt Companies Number of Healthy Companies 1 40 29 2 34 28 3 31 26 4 32 24 26 23 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 36

The companies used in this second data set were the same used in the second data set for the first model. However, again due to the lack of information, the number of companies varied. The number of companies from the second data set that were analyzed using the second model is shown in Table. Table : Number of companies in group 2 used in revised model Year before Bankruptcy Number of Bankrupt Companies Number of Nonbankrupt Companies 1 42 3 2 38 34 3 39 34 4 32 30 26 27 Of both of the groups a total of 49 different SIC numbers were included, all from industries that were nonmanufacturing/ service based. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 37

Chapter 6: Results and Discussion This chapter shows the results of the data collected in the Altman Z model, the first DEA model based on the Altman Z model, as well as the revised DEA model. First a univariate analysis was done on each of the inputs and outputs. Univariate Analysis Over the five years before bankruptcy there were a number of inputs and outputs collected for each company, both bankrupt and healthy. First, these values were analyzed on their own. Tables 6 and 7 show the averages of each of these values over the five years, with measurements in thousands of dollars except shareholders and employees which are in number of shareholders and employees, respectively. Table 6: Profile analysis of group 1 bankrupt companies Bankrupt Year 1 Year 2 Year 3 Year 4 Year Current Assets 131,879 141,449 131,68 113,96 119,184 Current Liabilities 206,117 118,072 9,76 69,104 72,676 Working Capital 74,238 23,378 36,109 44,861 46,08 Retained Earnings 83,01 19,862 3,184 1,649 9,676 Operating Income 38,36 72,711 4,7 17,319 19,48 Book Value of Equity 37,317 94,934 9,18 64,370 47,830 Total Assets 360,30 414,048 399,76 271,344 276,06 Total Liabilities 324,039 318,606 307,366 209,941 22,207 Employees 17,740 18,263 18,823 19,60 23,938 Shareholders 1,272 1,12 1,334 1,444 1,89 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 38

Table 7: Profile Analysis of group 1 nonbankrupt companies Nonbankrupt Year 1 Year 2 Year 3 Year 4 Year Current Assets 387,839 33,12 274,79 24,666 221,683 Current Liabilities 260,623 232,649 181,219 17,219 127,942 Working Capital 127,216 102,862 93,360 97,447 93,741 Retained Earnings 339,646 267,14 198,312 162,0 122,330 Operating Income 1,76 139,617 103,228 84,331 70,180 Book Value of Equity 31,714 49,193 36,249 314,060 279,823 Total Assets 1,030,768 879,143 692,34 614,072 30,739 Total Liabilities 498,960 419,911 336,176 299,494 249,428 Employees 11,777 11,193 9,42 8,29 7,707 Shareholders 8,793 8,89 6,829 6,222,426 If we plot these over the five years we can compare the averages of bankrupt and nonbankrupt companies. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 39

Figure 7: Profile Analysis of Variables in first DEA model In Figure 8 we plot the comparisons for the other inputs and outputs for the second model. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 40

Figure 8: Profile Analysis of remaining variables for revised DEA model It can be seen that nonbankrupt companies have higher averages than the bankrupt ones for all variables, except in this case the number of employees. The bankrupt companies have a higher number of employees for all years before bankruptcy, which is something to note. This observation affirms the decision to put the number of employees as an input variable for the DEA model, i.e. that it should be minimized to increase efficiency. Second dataset The Averages were also compared for the second set of data. Summarized in Tables 8 and 9: Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 41

Table 8: Profile Analysis of bankrupt companies in group 2 Bankrupt Year 1 Year 2 Year 3 Year 4 Year Current Assets 92,823 111,122 97,307 83,433 84,33 Current Liabilities 133,827 77,99 64,410 1,610 0,09 Working Capital 41,004 33,163 32,897 31,822 33,826 Retained Earnings 1,149 11,260 8,186 3,092 8,104 Operating Income 3,607 4,124 3,728 4,686,031 Book Value of Equity 3,771 2,19 3,93 3,426,929 Total Assets 19,319 222,48 191,003 12,314 167,362 Total Liabilities 173,888 168,427 139,0 100,739 113,429 Employees 1,18 1,40 1,842 2,12 1,760 Shareholders 846 1,276 1,213 1,468 1,62 Table 9: Profile Analysis of Nonbankrupt companies in group 2 Nonbankrupt Year 1 Year 2 Year 3 Year 4 Year Current Assets 431,773 419,662 349,639 306,840 320,141 Current Liabilities 30,398 308,711 2,070 208,62 217,643 Working Capital 126,37 110,91 94,69 98,188 102,498 Retained Earnings 220,471 194,13 198,78 340,841 406,824 Operating Income 63,67 0,148 32,082 62,309 64,823 Book Value of Equity 440,22 437,928 429,4 419,02 08,77 Total Assets 1,094,197 1,070,311 974,314 863,790 899,132 Total Liabilities 64,430 630,864 44,692 444,916 391,781 Employees 13,798 13,712 13,182 13,14 10,18 Shareholders 3,474 4,27 3,438 3,2 3,688 The trends were very similar to those of the first group, except that for this group the nonbankrupt companies also have higher averages of employees. A simple ttest was done, the results of which are shown in appendix G. For the first group, the ttest concluded that all variables showed a difference in the means of the bankrupt and nonbankrupt companies with a significance of 0.0. However for the second group, the ttest showed that the bankrupt and nonbankrupt companies were different to a significance of 0.01 except for the means of the Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 42

number of employees. The discrepancy in results shows a bit of variability in the accuracy of the number of employees as a predictor of corporate wellbeing. Altman Z results The data collected was then processed through the Altman model to test the accuracy of his Z model. The Altman Z scores for each company can be found in Appendix D. When each of the companies was processed through this model the classification results were as follows. Table 10: Results of Altman Z'' Model on group 1 Year 1 2 3 4 Type I error 18.2% 27.3% 37.2% 3.7% 41.% Type II error 4.2% 48.4% 38.7% 41.4% 42.9% Bankrupt accuracy 77.3% 9.1% 44.2% 47.6% 1.2% Nonbankrupt accuracy 41.9% 3.% 48.4% 44.8% 46.4% Total accuracy 62.7% 49.3% 4.9% 46.% 49.3% Bankrupt accuracy including grey area 81.8% 72.7% 62.8% 64.3% 8.% Nonbankrupt accuracy including grey area 4.8% 1.6% 61.3% 8.6% 7.1% Total accuracy including grey area 70.7% 64.0% 62.2% 61.9% 7.9% Total bankruptcy 64.0% 4.7% 41.9% 4.1% 47.8% Total nonbankrupt 28.0% 30.7% 41.9% 39.4% 43.% Total within grey area 8.0% 14.7% 16.2% 1.% 8.7% Type I error is a bankrupt company classified as nonbankrupt. The type II error is a nonbankrupt company being classified as bankrupt. Figure 9 shows the trend of the errors over the years before bankruptcy. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 43

60% Percent Error 0% 40% 30% 20% Type I error Type II error 10% 0% 0 2 4 6 Year before bankruptcy Figure 9: Percent error of Altman Z'' Score The first year before bankruptcy, the Altman Z score does a relatively good job of classifying bankrupt companies, it has a type I error of 18%. As expected, the percent error increases as the time before bankruptcy increases, where the error more than doubles to 38%. The opposite effect is true for the nonbankrupt companies. The type II error is as high as 4% in the year before bankruptcy and decreases as the time before bankruptcy increases to a percent error of 38% five years before bankruptcy. The Type II error stays relatively constant compared to that of the Type I error, due to the fact that the bankrupt companies have more defined indicators closer to the bankruptcy making it easier to predict bankruptcy and less bankrupt companies being classified falsely. However for nonbankrupt companies they are not approaching bankruptcy and thus their statements are similar regardless of year, therefore the nonbankrupt companies that are doing poorly, i.e. the walking wounded are constantly classified as bankrupt regardless of year. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 44

The accuracy is plotted in Figure 10. 90.00% 80.00% 70.00% 60.00% 0.00% 40.00% 30.00% 20.00% 10.00% 0.00% Percent classification accuracy 0 2 4 6 Year before bankruptcy Nonbankrupt Overall Accuracy Bankrupt Figure 10: Classification accuracy of Altman Z'' Model Here in Figure 10, the nonbankrupt companies have no clear trend and the classification accuracy fluctuates between 3% and 48% over the range of the years before bankruptcy. The bankrupt companies however have a clear trend, the classification accuracy decreasing the further from the year of bankruptcy, with a high classification accuracy of 77% in the year before bankruptcy but declining drastically to 43% in just the third year before bankruptcy. The overall accuracy also shows a downward trend of 62% in the year before bankruptcy and 4% five years prior to bankruptcy. These results do not currently include the grey area of the model. This grey or undetermined area is where Altman classified companies as unsure whether they were in danger of bankruptcy or not. Figure 11 shows the accuracy, including the grey companies in the classification. I.e. Bankrupt companies classified in the grey area are deemed bankrupt and nonbankrupt companies classified in the grey area are deemed nonbankrupt. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 4

90.00% 80.00% 70.00% 60.00% 0.00% 40.00% 30.00% 20.00% 10.00% 0.00% Percent accuracy including grey zone 0 2 4 6 Year before bankruptcy Bankrupt Nonbankrupt Overall Figure 11: Classification accuracy of Altman Z'' model including grey area If the grey area is included, the accuracy increases, with a bankruptcy accuracy of 81%, a nonbankrupt accuracy of 4%, and overall accuracy of 70% in the first year before bankruptcy. However, the bankrupt and nonbankrupt accuracy converge to approx. 61% at year before bankruptcy, with the overall and bankruptcy accuracies decreasing and the nonbankrupt increasing as seen before. The total number of companies classified is shown in Figure 12, regardless of the accuracy of the classification. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 46

70.00% Total classification amounts 60.00% 0.00% 40.00% 30.00% 20.00% Bankrupt Nonbankrupt Grey area 10.00% 0.00% 0 2 4 6 Year before bankruptcy Figure 12: Total classification within each zone of Altman Z'' model Similar trends are seen for bankrupt and nonbankrupt, though it could be noted the large difference in the percentage for bankrupt and nonbankrupt in the year before bankruptcy, with a percent bankrupt classification of 64% and a nonbankrupt of 28%. For the grey area, the classification actually seems to decrease the closer the date is to the bankruptcy date with the exception of year five, with fewer classifications as unsure as the bankruptcy approaches. It is not clear why this trend occurs, though it could be suggested that as the date of bankruptcy approaches the indicators between the bankrupt and nonbankrupt companies are more distinct and thus more companies are classified as either bankrupt or nonbankrupt with less falling into the ambiguous area. Second group As mentioned previously, there were two data sets analyzed. The first was to be used to create the DEA model cutoff and the second to test those established cutoff points. The data above was for the first data set. Table 11 summarises the results for the second data set. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 47

Table 11: Results of Altman Z'' model on group 2 Year 1 2 3 4 Type I error 13.3% 18.2% 34.1% 41.% 40.% Type II error 42.% 30.0% 27.% 27.% 2.0% Bankrupt accuracy 77.8% 9.1% 0.0% 41.% 3.1% Nonbankrupt accuracy 47.% 2.%.0% 2.% 63.9% Total accuracy 63.%.9% 2.4% 46.9% 49.3% Bankrupt accuracy including grey area 88.9% 86.4% 70.% 70.7% 83.8% Nonbankrupt accuracy including grey area 60.0% 72.% 7.0% 7.0% 88.9% Total accuracy including grey area 72.9% 69.1% 9.% 60.% 67.1% Total bankruptcy 61.2% 4.2% 39.3% 34.6% 30.1% Total nonbankrupt 29.4% 34.% 44.1% 46.9% 2.1% Total within grey area 11.8% 23.8% 20.2% 2.9% 36.9% DEA Model Each year was then analyzed with Data Envelopment Analysis, SlackBasedMeasure model. The model was input oriented, designed to minimize inputs. Through this analysis a score was given to each of the companies. This is listed in Appendix E. Once each company was given a score, a measure of bankruptcy had to be determined. For each year every possible cutoff point was tested at intervals of 0.0 from 0 to 1 to determine the classification accuracy of both bankrupt and nonbankrupt firms at each of those potential cutoff points. Figure 13 shows the percentages vs. the cutoffs. For example for a cutoff point of zero, no bankrupt companies are classified as bankrupt and all nonbankrupt companies would be classified as nonbankrupt. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 48

Percentage 120 Year 1 100 80 60 40 NonBankrupt 1 Bankrupt 1 20 0 0 0.2 0. 0.7 1 Cutoff Point Figure 13: Cutoff point determination on year 1 results Figure 13 shows the cutoff for year 1. We want to have the highest percentage of both bankrupt and nonbankrupt companies being classified accurately. The dilemma is that as the cutoff point changes the accuracy of one increases while the other decreases, as seen in Figure 13. If only one cutoff point was chosen, then the best place would be where the graphs intersected, giving the highest possible accuracy for both simultaneously. Here that point would be approx. 0.. Because we are choosing not one but two cut off points, an upper, above which would be the nonbankrupt zone, and a lower, below which would be the bankrupt zone, we need to focus on the points just above and below the 0.. We zoom in around that point in Figure 14. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 49

90 80 70 60 0 40 30 20 10 0 0.3 0.4 0. 0.6 0.7 Bankrupt Nonbankrupt Overall Accuracy Figure 14: Closer look at cutoff point for year 1 Here we also plot the overall accruracy of the Cutoff points. As we can see we get a high point in the overall accuracy at a cutoff point of 0., with 64% overall accuracy. Here the bankrupt companies have a classification accuracy of 61.36% and the nonbankrupt companies have a classification accuracy of 67.74%. For one year before bankruptcy these would be the chosen bottom cutoff. There is another maximum point at a cutoff of 0.6. Here the overall accuracy is 6.33%. The classification accuracy for bankrupt companies at this point is 72.73%, whereas for nonbankrupt companies the classification accuracy is 4.8% This would be the top cutoff point. However, this is only for one year before bankruptcy. If we look at the following years the graph changes quite a bit. Figure 1 shows the second year before bankruptcy. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 0

Percentage 120 Year 2 100 80 60 40 NonBankrupt 2 Bankrupt 2 20 0 0 0.2 0.4 0.6 0.8 1 1.2 Cutoff Point Figure 1: Cutoff point on year 2 It can be seen that the crossover point moves up to 0.6 in the second year which is quite different from the first. Because this model attempts to predict bankruptcy for up to years before the bankruptcy date, cutoff points for all years up to years before bankruptcy are examined in Figure 16. Here Bankrupt 1 is the plot of bankruptcy classification for 1 year before bankruptcy, and Nonbankrupt 1 is the nonbankrupt classification for 1 year before bankruptcy. Bankrupt 2 is the bankrupt classification for 2 years before bankruptcy, etc. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 1

Percentage 120 Cut off points for up to years before bankruptcy 100 80 60 40 20 NonBankrupt Bankrupt NonBankrupt 1 Bankrupt 1 NonBankrupt 2 Bankrupt 2 NonBankrupt 3 Bankrupt 3 NonBankrupt 4 Bankrupt 4 0 0 0.2 0.4 0.6 0.8 1 Cutoff Point Figure 16: Cut off point for up to years before bankruptcy Figure 16 clearly shows that for one year before bankruptcy the numbers are significantly lower with the numbers trending up as would be expected. In Figure 17 we compare the averages of the DEA scores for the years. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 2

DEA score 0.8 Average DEA scores 0.7 0.6 0. 0.4 0.3 0.2 Bankrupt Nonbankrupt 0.1 0 0 2 4 6 Year before bankruptcy Figure 17: Average DEA scores for group 1 first DEA model The DEA scores for the nonbankrupt companies are higher than that of the bankrupt companies for all years, but with a much more significant difference in the 1 st year before bankruptcy and with that difference decreasing as the time before bankruptcy increases. At one year before bankruptcy the average bankrupt score is 0.446, using that and the fact that we have a maximum efficiency in bankruptcy classification at 0., the bottom cutoff is chosen to be 0.. We can see that for one year before bankruptcy the average for the nonbankrupt companies is 0.647 we can choose the top cutoff point at 0.64. Because we are more concerned about the accuraccy of classification for bankrupt companies than nonbankrupt we will shift these points up. By comparing the values over the years as well as the averages, the finalized cutoff points are as follows: Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 3

Table 12: Cutoff points for first DEA model θ>0.7 0.<θ<0.7 θ<0. Nonbankrupt Grey area Bankrupt This corresponds to the following clasiffication accuraccies: Table 13: Classification accuracies of group 1 used to determine cutoff Year 1 2 3 4 Type I error 18.2% 29.6% 37.2% 40.% 36.6% Type II error 38.7% 32.3% 29.0% 24.1% 32.1% Bankrupt accuracy 63.6% 4.% 39.% 3.7% 34.2% Nonbankrupt accuracy 4.2% 4.2% 4.2% 1.7% 0.0% Total accuracy 6.0% 4.3% 41.9% 42.3% 40.6% Bankrupt accuracy including grey area 81.8% 70.% 62.8% 9.% 63.4% Nonbankrupt accuracy including grey area 61.3% 67.7% 70.9% 7.9% 67.9% Total accuracy including grey area 73.3% 69.3% 66.2% 66.2% 6.2% Total bankruptcy 3.3% 40.0% 3.1% 30.9% 33.3% Total nonbankrupt 29.3% 36.0% 40.% 4.1% 42.0% Total within grey area 17.3% 24.0% 24.3% 23.9% 24.6% These cutoff points are then tested on a second set of data that had not been influenced by the creation of the cutoff. This second set of data had the same inputs and outputs from the same genre of companies as the first group, with the same cutoff points determined by the first set, <0. is bankrupt, >0.7 is nonbankrupt and in between is a grey area, the see Table 14. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 4

Table 14: Classification accuracies of cutoff point for model 1 tested on group 2 Year 1 2 3 4 Type I error 13.0% 28.9% 28.3% 30.2% 39.% Type II error 21.9% 14.6% 19.% 20.0% 29.7% Bankrupt accuracy 71.7% 1.1% 4.7% 41.9% 36.8% Nonbankrupt accuracy 6.9% 68.3% 6.9% 60.0% 62.2% Total accuracy 68.9% 9.3%.2% 0.6% 49.3% Bankrupt accuracy including grey area 86.9% 71.1% 71.7% 69.8% 60.% Nonbankrupt accuracy including grey 78.1% 8.4% 80.% 80.0% 70.3% area Total accuracy including grey area 82.8% 77.9% 7.9% 74.7% 6.3% Total bankruptcy 48.3% 33.7% 33.3% 31.3% 33.3% Total nonbankrupt 37.9% 47.7% 4.9% 44.6% 0.7% Total within grey area 13.8% 18.6% 20.7% 24.1% 16.0% Revised DEA Model The original DEA model was a comparison with the Atman Z score. It took the ratios that Altman used, split them into their numerators and denominators and made them outputs and inputs respectively in the DEA model. A second model was created and analyzed from the same data set. This used a similar set of inputs and outputs but was modified to not include total assets/liabilities and also to incorporate human capital by incorporating number of shareholders and number of employees. The DEA scores for this model can be found in Appendix F A similar process was done to determine the cutoff point for this model. The cutoff point determination shown summarized in Figure 18. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

Percentage 120 Cut off points for up to years before bankruptcy second model 100 80 60 40 20 0 0 0.2 0.4 0.6 0.8 1 Cutoff Point NonBankrupt Bankrupt NonBankrupt 1 Bankrupt 1 NonBankrupt 2 Bankrupt 2 NonBankrupt 3 Bankrupt 3 NonBankrupt 4 Bankrupt 4 Figure 18: Cutoff points for up to years before bankruptcy for revised model The intersection points for each year were considered. These are shown in table 1. Table 1: Intersection points for each year of model 2 Year before Bankruptcy Intersection Point 1 0. 2 0.6 3 0.6 4 0.7 0.7 The average scores for the years are shown in Figure 19. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 6

DEA Score 0.9 0.8 0.7 0.6 0. 0.4 0.3 0.2 0.1 0 Average DEA Scores revised model 0 2 4 6 Year before bankruptcy Bankrupt Nonbankrupt Figure 19: Average DEA scores for group 1 in model 2 The cutoff points determined were the following: Table 16: Cutoff points for revised DEA model θ>0.8 Nonbankrupt 0.6<θ<0.8 Grey area θ<0.6 Bankrupt These cutoff points corresponded to the classification accuracies shown in Table 17. Table 17: Classification accuracies used to determine cutoff points for revised model Year 1 2 3 4 Type I error 20.0% 29.4% 41.9% 31.3% 42.3% Type II error 41.4% 3.7% 38.% 2.0% 26.1% Bankrupt accuracy 7.0%.9% 4.2% 43.8% 46.2% Nonbankrupt accuracy 48.3% 46.4% 3.9% 4.2% 73.9% Total accuracy 63.8% 1.6% 49.1% 48.2% 9.2% Bankrupt accuracy including grey area 80.0% 70.6% 8.1% 68.8% 7.7% Nonbankrupt accuracy including grey area 8.6% 64.3% 61.% 7.0% 73.9% Total accuracy including grey area 71.0% 67.7% 9.7% 71.4% 6.3% Total bankruptcy 60.9% 46.8% 42.1% 3.7% 36.7% Total nonbankrupt 31.9% 37.1% 47.4% 41.1% 7.1% Total within grey area 7.3% 16.1% 10.% 23.2% 6.1% Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 7

As with the first model, these cutoff points were then tested on the second data set. Using the cutoff points above on this data, the results are shown in Table 18. Table 18: Classification accuracies of cutoff points for revised model on group 2 Year 1 2 3 4 Type I error 14.3% 31.6% 30.8% 21.9% 42.3% Type II error 22.9% 11.8% 11.8% 6.7% 18.% Bankrupt accuracy 78.6% 7.9% 46.2% 3.1% 38.% Nonbankrupt accuracy 62.9% 61.8% 73.% 66.7% 70.4% Total accuracy 71.4% 9.7% 8.9% 9.7% 4.7% Bankrupt accuracy including grey area 8.7% 68.4% 69.2% 78.1% 7.7% Nonbankrupt accuracy including grey area 77.1% 88.2% 88.2% 93.3% 81.% Total accuracy including grey area 81.8% 77.8% 78.1% 8.% 69.8% Total bankruptcy 3.3% 36.1% 30.1% 30.7% 28.3% Total nonbankrupt 36.4% 4.8% 0.7% 43.6% 6.6% Total within grey area 10.4% 18.1% 19.2% 2.8% 1.1% Comparison of models Here the results of the Altman Z with the two DEA models are compared. The results are all for the companies in group 2. Figure 20 shows the classification accuracies of each company. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 8

Percentage 80.00% 70.00% 60.00% 0.00% 40.00% 30.00% 20.00% 10.00% 0.00% Overall Accuracy 0 2 4 6 Year before bankruptcy Altman First DEA Model Second DEA Model Figure 20: Comparison of overall accuracies Not surprisingly, it can be seen that there is a general trend with all of the models such that the accuracy decreases as the time before bankruptcy increases. The second DEA model actually has a slightly higher overall classification accuracy than the first DEA model, which is slightly better than the Altman Z score. This accuracy can be split into bankrupt and nonbankrupt accuracy, as shown in Figure 21. Figure 21: Comparison of bankrupt and nonbankrupt classification accuracies Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 9

For bankruptcy, the trend is similar to the overall accuracy, though with bankrupt classification the Altman Z model fares better against the other models. With nonbankrupt accuracy there is no clear trend over time, however it can be seen that DEA has higher classification accuracy than the Altman Z score in this regard. Figure 22 shows the type I and type II errors. Figure 22: Comparison of type I and type II error Type I error has an increasing trend, with the exception of an outlier on the second DEA model in the fourth year before bankruptcy. Other than that outlier, the second model appears to have a higher type I error than the first DEA model. This mirrors the trend in bankruptcy classification accuracy, though the Altman Z model has lower error in the first 2 years before bankruptcy and after which the DEA model shows less error. Similarly the type II error mirrors the trend in the nonbankrupt classification accuracy, with no defined trend over time, however with the DEA model having much lower error than the Altman model. Something to also consider is the total percentage of companies classified as bankrupt, nonbankrupt and in the grey area, shown in in Figure 23. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 60

Figure 23: Comparison of percent classifications Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 61

Altman classifies more companies as bankrupt than the DEA models. However for nonbankrupt classification the DEA scores classify more as nonbankrupt in the first 3 years and then in the fourth and fifth years before bankruptcy the three models have similar nonbankrupt classifications. The Altman model seems to have more companies classified in the grey area as the time before bankruptcy increases. The DEA models have an increasing trend until the th year before bankruptcy where it dips again. Comments on DEA scores Upon a further analysis of the DEA scores, which can be found in Appendices E and F, it could be noted that a particular number of firms were classified as being efficient, i.e. having a score of 1. These companies were both bankrupt and nonbankrupt companies. Some of the nonbankrupt companies were the following: o Home Depot o American Consumers o Eat at Joes o Family Room Entertainment o Children s Place o Arden Group o ELXSI o Rocky Brands This is understandable for the nonbankrupt companies. In this study, companies that do not end up bankrupt are efficient. However, there were a number of bankrupt companies constantly classified as efficient. Those were as follows: o Image Innovations Holdings o CD Warehouse o etoys o Letchers Inc. o Quokka Sports o WHSU Inc. (aka Micro Warehouse Inc.') Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 62

It is difficult to say why these were classified as efficient even though they went bankrupt. The values that they have for the input and output variables are comparable to those of nonbankrupt companies. Upon further investigation, companies like etoys and CD Warehouse were actually repurchased by other companies upon bankruptcy. In these instances the company could still have positive financial statements, but simply dissolved the company as a business strategy. Some other companies, such as Image Innovations Holdings, filed for bankruptcy but still exist today and could have similarly had alternate intentions for filing bankruptcy, such as a Chapter 11 reorganization. In future studies these companies should be accounted for in the model or excluded from the data set. Other companies however, such as Quokka Sports, were reported to be doing poorly but still have a classification of efficient. This shows that there is still error in the DEA model that must be further investigated. Also a few nonbankrupt companies were constantly classified as bankrupt. Some of which are: o AMC Entertainment o Carrols Corp o ACG Holdings o Jennifer Convertibles o Amazon.com o All American SportPark Inc. o TIX Corp o United Artists Theatre Circuit These would be what Simak referred to as the walking wounded. Though the second DEA model did classify more companies as nonbankrupt, it still did not sufficiently take into account those that performed poorly but still survived. More tests need to be done to look into this. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 63

Chapter 7: Conclusion and Recommendations This section presents the conclusions and suggests potential areas for future research. Conclusions In conclusion, the DEA models more accurately predicted bankruptcy when compared to the Altman Z model. The overall accuracy of the original DEA model was higher than that of the Altman Z and the revised DEA model, slightly more than that of the original. The higher accuracy of the revised model showed that the total assets or liabilities of a company are not necessary in determining bankruptcy. This is important for companies like nonmanufacturing/retail companies which may not have a large investment in hard assets. The Altman Z model did a fair job in classifying bankrupt companies and the results of the DEA models were essentially the same in terms of bankrupt accuracy. However, DEA did a better job in classifying the nonbankrupt companies correctly, which factored into the overall higher accuracy. The classification errors varied. For type I error, i.e. bankrupt companies classified as nonbankrupt, Altman had lower errors in the first 2 years, then DEA had lower errors in the third and fourth year and in the fifth year the overall errors for all three models were high. For type II error, i.e. nonbankrupt companies classified as bankrupt, the Altman model definitely had a higher error than DEA. The Altman Z model would classify more companies as bankrupt than the DEA models in all years and the DEA models classified more companies as nonbankrupt in the first 3 years. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 64

Comparison of the average values of the inputs and outputs was also done. These input and output variables were: Current Assets Current Liabilities Working Capital Retained Earnings Operating Income Book Value of Equity Total Assets Total Liabilities Number of Employees Number of Shareholders Comparison of means showed that for all variables, nonbankrupt companies had higher means than bankrupt ones, except in the case of the number of employees, where the bankrupt companies had a higher average. Overall, this research showed that a SBM DEA model can be used for predicting bankruptcy or financial distress, that total asset size is not necessary for bankruptcy prediction in services oriented firms and also showed that the method for dealing with negative values, by splitting them into positive and negative values could be a viable option when needed in DEA analyses. This research has many useful conclusions but it also has some areas that should be examined further. Recommendations for Future Work There are many areas that could be improved on in this model. The following is a list of possible recommendations for future work. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 6

This work looked at an inputoriented SlacksBasedModel. Alternate DEA models could be studied to attempt to gain further accuracy. In particular, the possibility of using the Assurance Region model to restrict the weights and if possible, reconfiguring DEA to allow weight restrictions between inputs and outputs to account for the positive and negative values that had to be split in this model. Creating a different model with new inputs and outputs perhaps from sources other than financial statements could be investigated. The use of employees and shareholders, two nonmonetary variables, were an attempt to do this and their impact could be explored in further detail. Financial statements only look into one aspect of a company s health and other inputs could possibly make the model more accurate. A different method of determining the cutoff point could be explored. The trial and error approach is a simple and intuitive way, however a more statistically sound method could be developed. Decision trees were a discussed method and could be looked into for future research. Other analytical methods could be explored, rather than or in conjunction with DEA. Altman used MDA, however there could be more studies done with other models, such as logistic regression. Looking at a more homogeneous subset of companies could be helpful. DEA looks at classifying DMUs which are similar in culture, the more similar they are, the more accurately the culture can be defined. This study looked broadly at nonmanufacturing/retail companies. This could be more specific; however the problem could be finding a large enough sample size of data. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 66

The walking wounded was not sufficiently addressed as well as those companies that are doing relatively well but file for bankruptcy. In this case, more than simply the fact that a company has filed bankruptcy should be looked at when classifying a company as failed. The curiosity of the bankrupt companies having a larger number of employees than nonbankrupt firms should be explored in the context of the human capital being a significant factor in these firms. Future models could do more to look into the efficiency of a company based on the number of employees and the efficiency of the employees. Incorporating cash flows into a future model should be considered, as a few models have used this in the past and has shown that it could possibly be a good indicator of bankruptcy prediction [AZIZ88]. Testing the model on companies in a different corporate environment, different countries, such as the UK and Europe could be considered. Applying this lack of assets approach to a different analysis type, such as neural networks [BACK94] could also be used to test the legitimacy of this approach. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 67

References [ALEX49] [ALTM02] [ALTM06] [ALTM68] [ALTM68] [ALTM77] [ALTM93] [AZIZ88] [BACK94] [BACK9] [BALL82] [BANK78] [BEAV67] Sidney Alexander, "The Effect of Size of Manufacturing Corporation on the Distribution of the Rate of Return," Review of Economics and Statistics, August, 1949, pp. 2293 Altman, Edward. Bankruptcy, Credit Risk and High Yield Junk Bonds. Blackwell Publishers Inc. Malden, Massachusetts. 2002. Altman, Edward. Hotchkiss, Edith. Corporate Financial Distress and Bankruptcy. Third Edition. John Wiley and Sons Inc. 2006. Hoboken, New Jersey Altman, Edward. Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy, Journal of Finance (September 1968) pp. 89610. AItman, E., "Financial Ratios, Discriminant Analysis and the prediction of Corporate Bankruptcy". Journal of Finance. Sept.1968. pp. 89609 Altman, E.. Haldeman, R., Narayan, P.. ZETA Analysis A New Mode1 to Identify Bankruptcy Risk of Corporations". Journal of Banking and Finance. June 1977, pp. 293. Altman. Edward. "Corporate Financial Distress and Bankruptcy", Second Edition, John Wiley & Sons. 1993. Aziz, A., Emanuel, D.C., Lawson, G.H., Bankruptcy Prediction An Investigation of Cash Flow Based Models, Journal of Management Studies 2:, Sept 1988, pp. 419436. Back, B., Oosterom, G., Sere, K., and van Wezel, M. (1994), A Comparative Study of Neural Networks in Bankruptcy Prediction, Proceedings of the 10th Conference on Artificial Intelligence Research in Finland, Turku, Finland, p. 140148. Back, Barbro., Sere, Kaisa., van Wezel, Michiel C. Choosing the Best Set of Bankruptcy Predictors, proceedings of the 1NWGA, Vaasa, Finland, Jan 199, pp. 28300 Ball R., Foster G. Corporate Financial Reporting: A Methodical View of Empirical Research Journal of Accounting Research, 20 (supp.), 1982, pp. 161234 Banker, R.D., Charnes, A. and Cooper, W.W. Some Models for Estimating Technical and Scale Inefficiencies in Data Envelopment Analysis Management Science vol. 30. pp. 10781092 (1978) Beaver, William H. Financial Ratios as Predictor of Failure, Empirical Research in Accounting: Selected Studies, 1967, supplement to Vol., Journal of Accounting Research pp. 71111 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 68

[CASE80] [CASE84] [CASE8] [CHAR78] [CHAR94] [CHAV04] [CHUV03] [CIEL04] [COOP07] [CYBI03] [DEAK72] [GRIC01A] [GRIC01B] [HILL04] [HSIE93] Casey, C.M. Jr., The Usefulness of Accounting Ratios for Subjects renditions of Corporate Failure: Replication and Extension, Journal of Accounting Research, Autumn 1980. pp. 603613 Casey, C. Bartczak, N., Cash Flow Its Not the Bottom Line. Harvard Business Review, 62, July/Aug. 1984, pp. 6066 Casey, C. Bartczak, N., Using Operating Cash Flow Data to Predict Financial Distress: Some Extensions, Journal of Accounting Research, Spring 198, pp. 38 341 Charnes, A., W. Cooper, and E. Rhodes (1978): Measuring the Efficiency of Decision Making Units," European Journal of Operational Research, 2(6), 429444 Charnes,A., Cooper, W. W.. and Seiford.L.M., "Basic DEA Model", in Data Envelopment Analysis: Theory, Methodology and Application, Charnes et al. Editors, Kluwer Academic Publishers, 1994 Chava, S., Jarrow, R., 2004. "Bankruptcy Prediction with Industry Effects". Review of Finance 8: 37 69, Kluwer Academic Publishers. Chuvakhin, Nikolai., Gertmenian, L. Wayne., Bankruptcy Prediction in the WorldCom Age. Graziadio Business Review. Vol. 6 Issue 1. 2003 Cielen, A., Peeters, L., Vanhoof K., 2004. "Bankruptcy prediction using a data envelopment analysis." European Journal of Operational Research 14 (2004) 26 32 Cooper, W. W., Seiford, L.M. and Tone, K. Data Envelopment Analysis: A comprehensive Text with Models, Applications, References and DEASolver Software, 2nd Edition. Springer Science+Business Media, LLC. 2007 Cybinski, P.J. Doomed Firms: An Econometric Analysis of the Path to Failure. Ashgate Publishing Limited. Burlington, VT. 2003 ISBN 07463007 Deakin, E.B., A Discriminant Analysis of predictors of Business Failure, Journal of Accounting Research, Spring 1972, pp. 167179 Grice, J., Dugan, M. 2001. "The Limitations of Bankruptcy Prediction Models: Some Cautions for the Researcher." Review of Quantitative Finance and Accounting, 17: 11 166, Kluwer Academic Publishers. 2001 Grice, J., Ingram R., "Tests of the generalizability of Altman s bankruptcy prediction model." Journal of Business Research 4 (2001) 3 61 Hillegeist, Stephen A., Keating, Elizabeth., Cram, Donald P., Lunstedt, Kyle G.2004. Assessing the probability of bankruptcy. Review of Accounting Studies 9, 34 Hsieh, SuJane. 1993. "A Note on the Optimal Cutoff Point in Bankruptcy Prediction Models." Journal of Business, Finance and Accounting 20(3), 0306686X. Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 69

[KERS09] [LIBB7] Kristiaan Kerstens*, Ignace Van de Woestyne** Negative Data in DEA: A Simple Proportional Distance Function Approach. *IÉSEG School of Management, CNRS LEM (UMR 8179) **Hogeschool Universiteit Brussel, Belgium April 2009 Libby, R. Accounting Ratios and the Prediction of Failure, Journal of Accounting Research, 197. [MERG08] Mergent Online http://www.mergentonline.com/ Date Visited: September 2 th, 2011.Copyright 2008 by Mergent, Inc. [MERW42] [NEWG11] [OHLS80] [PATE02] [SHUM01] [SIGA09] [SIMA97] [TAFF82] [TRAN07] [WEAT72] [WORL09] Charles L. Merwin, "Financing Small Corporations in Five Manufacturing Industries ",.192636. National Bureau of Economic Research,. 1942. New Generation Research Experts in Bankruptcy Research http://www.bankruptcydata.com/ Date Visited: September 2 th, 2011. New Generation Research. inc. Suite 801, 22 Friend Street. Boston MA 021 14. Ohlson, James, 1980. "Financial ratios and the probabilistic prediction of bankruptcy". Journal of Accounting Research 18, 109 131. Pate, Carter (2002) Business Chapter 11 Bankruptcies: Recent Levels and Implicationsfor 2002, PricewaterhouseCoopers, March 2002 Shumway, Tyler, 2001. "Forecasting bankruptcy more accurately: a simple hazard model." Journal of Business 74, 101 124. Sigaroudi, Sanaz, Paradi, J. Incorporating Ratios in DEA: Applications to Real Data, 2009, Masters Thesis. Centre for Management of Technology and Entrepreneurship, University of Toronto, Toronto, ON Simak, Paul, Paradi, J. DEA Analysis of Corporate Failure 1997, Masters Thesis. Centre for Management of Technology and Entrepreneurship, University of Toronto. Toronto, ON. Taffler, R.J., Forecasting Corporate failure In the UK using Discriminant Analysis and Financial ratio Data, Journal of R. Statist. Soc. A 14, 1982, pp. 34238 Tran, Angela Paradi, J Two Stage Financial Risk Tolerance Assessment using Data Envelopment Analysis ProQuest Dissertations and Theses; 2007; ProQuest. Masters Thesis, Centre for Management of Technology and Entrepreneurship, University of Toronto, Toronto, ON. Weatherly, Leslie A. Human Capital The Elusive Asset: Measuring and Managing Human Capital: A Strategic Imperative for HR SHRM Research, Society for Human Resource Management, 2003 Growth of the Service Sector. "http://www.worldbank.org/depweb/beyond/beyondco/beg_09.pdf" Date Visited: December 2th, 2011. World Bank 2009 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 70

[ZMIJ84] [ZOPO98] Zmijewski, Mark E., 1984. "Methodological issues related to the estimation of financial distress prediction models." Journal of Accounting Research 22 (Suppl.), 9 82. Zopounidis, Constantin. Dimitris, Augustinos I. Muticriteria Decision Aid Methods for the Prediction of Business Failure. Kluwer Academic Publishers, P.O.Box 17, 3300 AA Dordrecht, The Netherlands. 1998 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 71

Appendix A: List of companies *Note that for the Year of bankruptcy descriptor, for nonbanktupt companies, the year refers to the year of bankruptcy of their bankrupt counterpart. Group 1 DMU Company Bankrupt/Non Bankrupt 1 1800 Flowers.com Inc 2 A.C. Moore Arts & Crafts Inc Year of bankruptcy Nonbankrupt 2001 Retail Specialty Nonbankrupt 2001 Retail Specialty 3 AccuHealth Bankrupt 2001 Diagnostic & Health Related Services Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms Sector Listed Country Exchange SIC SIC NAICS NAICS Incorpo Description Description rated US NMS 992 Florists 43110 Florists 1976 US NMS 94 Hobby, toy, and game shops US NBB 8082 Home Health Care Services 4 ACG Holdings Nonbankrupt 200 Printing US 274 Commercial printing, gravure AHT Corp Bankrupt 2000 Diagnostic & Health Related Services 6 All Star Gas Corp 7 AMC Entertainment 8 American Banknote 9 American Consumers Bankrupt 2003 Retail Specialty US NBB 8099 Health and allied Services US NL 984 Liquefied Petroleum Gas Dealers Nonbankrupt 2001 Entertainment US ASE 7832 Motion Picture Theaters ex drive in Bankrupt 200 Printing US NBB 274 Commercial printing, Nonbankrupt 2000 Retail Food & Beverage, Drug & Tobacco gravure US NBB 411 Grocery Stores 41120 Hobby, Toy, and Game Stores 621610 Home Health Care Services 323111 Commercial Gravure printing 621999 All Other Miscellaneous Ambulatory Health Care Services 44312 Liquefied Petroleum Gas (Bottled Gas) Dealers 12131 Motion Picture Theaters (except Drive Ins) 323111 Commercial Gravure Printing 44110 Supermarkets and Other Grocery (except 1997 1983 1989 1993 1988 1920 192 1968 72

10 American Eco Corp 11 Ames Department Stores Bankrupt 2000 Miscellaneous Consumer Services Bankrupt 2002 Retail General Merchandise/ Department Stores 12 Arden Group Nonbankrupt 2001 Retail Food & Beverage, Drug & Tobacco 13 Ascena Retail Group Nonbankrupt 2004 Retail Apparel and Accessories 14 Avado Brands Bankrupt 2004 Hotels, Restaurants & Travel 1 Big Buck Brewery & Steakhouse 16 Big V Supermarkets Bankrupt 2004 Hotels, Restaurants & Travel Bankrupt 2000 Retail Food & Beverage, Drug & Tobacco 17 BioScrip Inc Nonbankrupt 2001 Diagnostic & Health Related Services 18 BonTon Stores Nonbankrupt 2002 Retail General Merchandise/ Department Stores Canada NBB 7699 Repair Services US NBB 331 Variety Stores US NMS 411 Grocery Stores US NMS 621 Women's clothing stores US NBB 812 Eating Places US NBB 812 Eating Places US 411 Grocery Stores US NMS 912 Drug Stores and Proprietary Stores US NMS 311 Department Stores Convenience) Stores 33419 Other Measuring and Controlling Devices 42990 All Other General Merchandise Stores 44110 Supermarkets and Other Grocery (except Convenience) Stores 448120 Women's Clothing Stores 722110 Full Service Restaurants 722110 Full Service Restaurants 44110 Supermarkets and Other Grocery (except Convenience) Stores 446110 Pharmacies and Drug Stores 42111 Department Stores (except Discount Department Stores) 19 Borders Nonbankrupt 2001 Retail US NBB 942 Book stores 41211 Book Stores 1994 Group Inc Specialty 20 Briazz Inc Bankrupt 2004 Hotels, US NBB 812 Eating 722211 Limited 199 1962 1977 1966 1986 1993 1990 1996 1929 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 73

21 Caliber Learning Network 22 Carmike Cinemas Inc Restaurants & Travel Bankrupt 2001 Educational Services Places US OTC 8299 Schools & Educational Services Nonbankrupt 2001 Entertainment US NMS 7833 Motion Picture Theaters ex drive in 23 Carrols Corp Nonbankrupt 2004 Hotels, Restaurants & Travel 24 Casual Male Corp Bankrupt 2001 Retail Apparel and Accessories 2 CD Warehouse Bankrupt 2002 Retail Specialty 26 Children's Place Retail Stores Inc 27 Cinemaster Luxury Theaters Inc 28 Commodore Applied Technologies 29 Computer Learning Centers Nonbankrupt 2000 Retail Apparel and Accessories US NBB 812 Eating Places US NBB 611 Men's & Boys Clothing US NBB 73 Record & Prerecorded tape stores US NMS 61 Family Clothing Stroes Bankrupt 2001 Entertainment US NAS 7834 Motion Picture Theaters ex drive in Nonbankrupt 2002 Sanitation Services Bankrupt 2001 Educational Services 30 Converse Bankrupt 2002 Apparel, Footwear & Accessories 31 Cooker Restaurant Corp Bankrupt 2001 Hotels, Restaurants & Travel US NBB 499 Sanitary services US OTC 8299 Schools & Educational Services US OTC 3021 Rubber and plastics Footwear US OTC 812 Eating Places Service Restaurants 611430 Professional and Management Development Training 12131 Motion Picture Theaters (except Drive Ins) 722110 Full Service Restaurants 448110 Men's Clothing Stores 41220 Prerecorded Tape, Compact Disc, and Record Stores 448140 Family Clothing Stores 12131 Motion Picture Theaters (except Drive Ins) 62998 All Other Miscellaneous Waste 611430 Professional and Management Development Training 316211 rubber and plastics footwear 722110 Full Service Restaurants 32 Crown Books Bankrupt 2001(2000) Retail US OTC 942 Book stores 41211 Book stores 1981 1996 1982 1968 198 1996 1988 1989 1996 1987 1908 1986 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 74

Corp Specialty 33 Dairy Mart Bankrupt 2001 Retail Food & Beverage, Drug & Tobacco 34 Drug Emporium Inc 3 Eagle Food Centers 36 Eat At Joes Ltd Bankrupt 2001 Retail Food & Beverage, Drug & Tobacco Bankrupt 2000 Retail Food & Beverage, Drug & Tobacco Nonbankrupt 2004 Hotels, Restaurants & Travel 37 ELXSI Corp Nonbankrupt 2001 Hotels, Restaurants & Travel 38 etoys Inc Bankrupt 2001 Retail Specialty 39 Express Scripts Inc 40 Family Room Entertainment Corp Nonbankrupt 2001 Diagnostic & Health Related Services Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms US NBB 499 Miscellaneo us Food Stores US OTC 912 Drug stores and propietary stores US OTC 411 Grocery Stores US NBB 812 Eating Places US NBB 812 Eating Places US OTC 94 Hobby, toy and game shops US NMS 912 Drug Stores and Proprietary Stores Nonbankrupt 2001 Entertainment US NBB 7812 Motion picture & video production 41 Ferrellgas L.P. Nonbankrupt 2003 Retail Specialty 42 Florsheim Group Inc 43 Fresh Choice Inc Bankrupt 2002 Apparel, Footwear & Accessories Bankrupt 2004 Hotels, Restaurants & Travel US 984 Liquified Petroleum Gas Dealers US NBB 3143 Men's footwear, except athletic US NBB 812 Eating Places 44299 All Other Specialty Food Stores 446110 Pharmacies and Drug Stores 44110 Supermarkets and Other Grocery (except Convenience) Stores 722110 Full Service Restaurants 722110 Full Service Restaurants 41120 Hobby, Toy and Game Stores 446110 Pharmacies and Drug Stores 12110 Motion Picture and Video Production 44312 Liquified Petroleum Gas (Bottled Gas) Dealers 316213 Men's Footwear (except Athletic) 722110 Full Service Restaurants 1972 1977 1987 1988 1987 1996 1986 1969 1994 1892 1986 7

44 Furr's Restaurant Group Inc Bankrupt 2003 Hotels, Restaurants & Travel 4 Gadzooks Inc Bankrupt 2004 Retail Apparel and Accessories 46 Gerald Stevens Inc 47 Grand Union Company Inc 48 Hartcourt Companies Inc 49 Hastings Entertainment Inc Bankrupt 2001 Retail Specialty Bankrupt 2000 Retail Food & Beverage, Drug & Tobacco Nonbankrupt 2001 Educational Services Nonbankrupt 2002 Retail Specialty US NYS 812 Eating Places US NBB 621 Womens Clothing Stores 722212 Cafeterias 1990 448120 Women's Clothing Stores 1983 US NBB 992 Florists 43110 Florists 1970 US NBB 411 Grocery Stores Us NBB 8299 Schools & educational services US NAS 73 Record & prerecorded tape stores 0 HCI Direct Inc Bankrupt 2002 Textiles US NBB 221 Women's hosiery, 1 Healthcare Integrated Services 2 HeiligMeyers Company Bankrupt 2002 Diagnostic & Health Related Services Bankrupt 2000 Retail Furniture & Home Furnishings 3 Home Depot Nonbankrupt 2001 Retail Hardware & Home Improvement 4 Homeland Holding Corp Bankrupt 2001 Retail Food & Beverage, Drug & Tobacco except socks US NBB 8071 Medical Laboratories US NBB 712 Furniture Stores US NYS 211 Lumber and other building materials US NBB 411 Grocery Stores 44110 Supermarkets and Other Grocery (except Convenience) Stores 611699 All Other Miscellaneous Schools and Instruction 41220 Prerecorded Tape, Compact Disc, and Record Stores 31111 Sheer Hosiery 62111 Medical Laboratories 442110 Furniture Stores 1928 1983 1968 1991 1972 444110 Home Centers 1978 44110 Supermarkets and Other Grocery (except Convenience) Stores Horizon Bankrupt 2001 Retail Food & US NBB 912 Drug stores 446110 Pharmacies 1992 1987 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 76

Pharmacies Inc 6 House2Home inc 7 Image Innovations Holding Inc Beverage, Drug & Tobacco Bankrupt 2001 Retail Hardware & Home Improvement Bankrupt 2006 Leisure Equipment 8 Integra Inc Bankrupt 2002 Hospitals & Health Care Facilities 9 Integrated Health Services Inc 60 Jacobson Stores Inc 61 Jennifer Convertibles Inc 62 Jos. A Bank Clothiers Inc 63 Kasper ASL Ltd 64 KushnerLocke International Inc 6 LaCrosse Footwear 66 Lamonts Apparel Inc Bankrupt 2000 Hospitals & Health Care Facilities Bankrupt 2002 Retail General Merchandise/ Department Stores Nonbankrupt 2000 Retail Furniture and Home Furnishings Nonbankrupt 2001 Retail Apparel and Accessories Bankrupt 2002 Apparel, Footwear & Accessories and propietary stores US NBB 211 Lumber and other building materials US NBB 279 Commercial printing US NBB 8093 Specialty outpatient clinics, nec US NBB 8399 Social Services US NBB 311 Department Stores US NBB 712 Furniture Stores US NMS 611 Men's & boy's clothing stores US NBB 2337 Women's and misses' suits and coats Bankrupt 2001 Entertainment US NBB 7812 Motion picture and video production Nonbankrupt 2002 Apparel, Footwear & Accessories Bankrupt 2000 Retail Apparel and Accessories US NMS 3021 Rubber and Plastics Footwear US NL 61 Family Clothing and Drug Stores 444110 Home Centers 1989 323119 Other Commercial Printing 621498 All Other Outpatient Care Centers 623110 Nursing Care Facilities 42111 Department Stores (except Discount Department Stores) 442110 Furniture Stores 448110 Men's Clothing Stores 31234 Women's and Girls' Cut and Sew Suit, Coat, Tailored Jacket and Skirt 12110 Motion Picture and Video Production 316211 Rubber and Plastics Footwear 448140 Family Clothing Stores 1998 1991 1986 1939 1986 1982 1997 1986 1897 198 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 77

67 Lechters Inc Bankrupt 2001 Retail Furniture & Home Furnishings 68 Loews Cineplex Entertainment Corp Stores US NBB 719 Misc. homefurnish ings stores Bankrupt 2001 Entertainment US TSX 783 Motion Picture Theaters ex drive in 69 Med/Waste Inc Bankrupt 2002 Sanitation Services 70 Meritage Hospitality Group Inc 71 Mexican Restaurants Inc 72 New York Health Care Inc Nonbankrupt 2004 Hotels, Restaurants & Travel Nonbankrupt 2003 Hotels, Restaurants & Travel Nonbankrupt 2001 Diagnostic & Health Related Services 73 RadNet Inc Nonbankrupt 2002 Diagnostic & Health Related Services 74 Rocky Brands Inc 7 Sagemark Companies Ltd Nonbankrupt 2002 Apparel, Footwear & Accessories Nonbankrupt 2002 Hospitals & Health Care Facilities US NBB 499 Sanitary Services, nec US NBB 812 Eating Places US NBB 812 Eating Places US NBB 8082 Home health care services US NMS 8071 Medical Laboratories US NMS 3143 Men's Footwear, except athletic US NBB 8093 Specialty Outpatient Clinics 442299 All Other Home Furnishings Stores 12131 Motion Picture Theaters (except Drive Ins) 62998 All Other Miscellaneous Waste Management Services 722211 Limited Service Restaurants 722110 Full Service Restaurants 621610 Home Health Care Services 62111 Medical Laboratories 316213 Men's Footwear (except Athletic) 621498 All Other Outpatient Care Centers 197 1977 1991 1986 1996 1983 198 1932 1961 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 78

Group 2 1 AFC Enterprises Inc Incorporated Non Bankrupt 2 AllAmerican SportPark Inc Non Bankrupt 3 Amazon.com Non Bankrupt 4 Amsurg Corp Non Bankrupt Blockbuster Inc Non Bankrupt 6 Cache Inc Non Bankrupt 7 Cala Corp Non Bankrupt 8 CareGuide Inc Non Bankrupt 9 Charming Shoppes Inc 10 Circuit City Stores Inc Non Bankrupt Non Bankrupt 11 DGSE Companies Non Bankrupt 12 Finlay Enterprises Non Inc Bankrupt 13 FragranceNet.com Non Inc Bankrupt Year of bankruptcy Sector Listed Country Exchange SIC SIC Description NAICS NAICS Description 2004 Hotels, Restaurants & Travel 2004 Sporting & Recreational 2002 Retail Specialty 2001 Hospitals & Health Care Facilities 2001 Retail Specialty 2004 Retail Apparel and Accessories 200 Property, Real Estate & Development 2004 Diagnostic & Health Related Services 2000 Retail Apparel and Accessories 200 Retail Appliances and Electronics 2003 Retail Specialty 2003 Retail Specialty 2000 Retail Specialty US NMS 812 Eating Places 722211 Limited Service Restaurants US NBB 7992 Public golf courses US NMS 961 Catalog and mail order houses US NMS 8011 Offices & clinics of medical doctors US NBB 7841 Video tape rental US NMS 621 Women's clothing stores US NBB 7011 Hotels and Motels US NBB 8099 Health and allied services nec US NMS 621 Women's 713910 Golf Courses and Country Clubs 44111 Electronic Shopping DMU Company Bankrupt/ Non Bankrupt 1992 1984 1994 621493 Freestanding Ambulatory Surgical and Emergency Centers 1992 32230 Video Tape and 1989 Disc Rental 448120 Women's 197 Clothing Stores 721120 Casino Hotels 198 18111 Internet Service Providers 448120 Women's clothing stores Clothing Stores US NBB 731 Radio, TV & 443112 Radio, Electronic Stores Television, and Other US ASE 944 Jewelry Stores 448310 Jewelry Stores 199 1969 1949 US NBB 944 Jewelry Stores 448310 Jewelry Stores 1988 US NBB 999 Miscellaneous retail stores 424210 Drug and Druggidtd Sundries Merchant 1987 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 79

14 Great Atlantic & Pacific Tea Company Inc 1 Image Entertainment Inc Non Bankrupt Non Bankrupt 16 Internet Infinity Inc Non Bankrupt 17 Kirkland's Inc Non Bankrupt 18 Lowe's Companies Inc 19 Michaels Stores Inc 20 Million Dollar Saloon Inc Non Bankrupt Non Bankrupt Non Bankrupt 21 Movie Gallery Inc Non Bankrupt 22 National Record Mart 23 Natural Wonders Inc 2001 Retail Food & Beverage, Drug & Tobacco Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms Wholesalers US NBB 411 Grocery Stores 44110 Supermarkets and Other Grocery (except Convenience) Stores 2000 Entertainment US NBB 7822 Motion Picture and tape distribution 2000 Internet & Software 2003 Retail Furniture and Home Furnishings 2001 Retail Hardware & Home Improvement 2001 Retail Specialty 2001 Hotels, Restaurants & Travel 2000 Retail Specialty Bankrupt 2001 Retail Appliances and Electronics Bankrupt 2000 Retail Specialty US OTC 099 Durable Goods, nec US NMS 947 Gift, novelty and souvenir stores US NYS 211 Lumber and other building materials 12120 Motion Picture and Video Distribution 423990 Other Miscellaneous Durable Goods Merchant Wholesalers 43220 Gift, Novelty and Souvenir Stores 192 197 199 1966 444110 Home Centers 192 Us NYS 94 Hobby, toy, and game shops 41120 Hobby, Toy, and Game Stores US NBB 812 Eating Places 722211 Limited Service Restaurants US NBB 7841 Video tape rental US NBB 73 Record & pre recorded tape stories US 999 Miscellaneous retail stores, nec 24 Netflix Non Bankrupt 2004 Entertainment US NMS 7841 Video tape rental 2 Netter Digital Bankrupt 2000 Entertainment US NBB 7812 Motion Picture & Video Production 32230 Video Tape and Disc Rental 41220 Prerecorded Tape, Compact Disc, and Record Stores 43998 All Other Miscellaneous Store Retailers (except Tobacco Stores) 32230 Video Tape and Disc Rental 12110 Motion Picture and Video Production 1983 1987 198 1937 1986 1997 1979 80

26 New Horizons Worldwide Inc 27 New York Bagel Enterprises Inc 28 Noble Roman's Inc 29 Omnicare Inc Non Bankrupt Non Bankrupt 2006 Business Services Bankrupt 2000 Hotels, Restaurants & Travel Non 2002 Hotels, Bankrupt Restaurants & Travel 2001 Diagnostic & Health Related Services 30 One Price Clothing Inc Bankrupt 2004 Retail Apparel and Accessories 31 Orbit Brands Bankrupt 2004 Sporting and Corporation Recreational 32 Overstock.com Non 2003 Retail Bankrupt Specialty 33 Paper Warehouse Inc 34 Park Pharmacy Corp 3 Paul Harris Stores Inc 36 Payless Cashways Inc 37 Pdg Environmental Bankrupt 2003 Retail Specialty Bankrupt 2002 Retail Food & Beverage, Drug & Tobacco Bankrupt 2000 Retail Apparel and Accessories Bankrupt 2001 Retail Hardware & Home Improvement Non Bankrupt 2000 Sanitation Services 38 PharMor Inc. Bankrupt 2001 Retail Food & Beverage, Drug & Tobacco 39 PHC Inc Non Bankrupt 40 Piccadilly Cafeterias Inc 2002 Hospitals & Health Care Facilities Bankrupt 2003 Hotels, Restaurants & Travel US NBB 8243 Data processing schools 611420 Computer Training US NBB 812 Eating Places 722110 Full Service Restaurants US OTC 812 Eating Places 722110 Full Service Restaurants US NYS 912 Drug Stores and Proprietary Stores US NBB 621 Women's clothing stores US NBB 7992 Public golf courses US NMS 961 Catalog and mail order houses US NBB 947 Gift, Novelty and Souvenir shops US NBB 912 Drug Stores and proprietary stores US OTC 621 Women's Clothing Stores US NL 211 Lumber and other building materials 446110 Pharmacies and Drug Stores 448120 Women's Clothing Stores 713910 Golf Courses and Country Clubs 44113 Mail Order houses 43220 Gift, Novelty and Souvenir Shops 446110 Pharmacies and Drug Stores 1988 199 1972 1981 1987 1991 1998 1987 1986 448120 Women's 192 Clothing Stores 444110 Home Centers 1988 US NBB 493 Refuse Systems 62211 Hazardous Waste Treatment and Disposal US NBB 912 Drug Stores and proprietary stores US ASE 8082 Home health care services 446110 Pharmacies and Drug Stores 621610 Home Health Care Services 1987 1982 1976 US NBB 812 Eating places 722212 Cafeterias 196 41 Pier 1 Imports Inc Non 2000 Retail US NYS 719 Misc. home 442299 All Other Home 1986 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 81

42 Planet Entertainment Corp Bankrupt Furniture and Home Furnishings furnishings stores Bankrupt 2001 Advertising US NBB 7313 Radio, TV, publisher representatives 43 Play Co. Toys and Entertainment Corp Bankrupt 2001 Retail Specialty 44 Prandium Inc Bankrupt 2002 Hotels, Restaurants & Travel 4 Premier Concepts Inc Bankrupt 2003 Retail Specialty 46 Prosoft Learning Bankrupt 2006 Educational Corporation Services 47 Provell Inc Bankrupt 2002 Retail Specialty 48 Questar Assessment Inc Non Bankrupt 2000 Educational Services 49 Quokka Sports Inc Bankrupt 2001 Sporting & Recreational 0 Redline Bankrupt 2004 Retail Performance Automotive Products Inc Furnishings Stores 41840 Media Representatives US OTC 94 Hobby, toy and game shops 41120 Hobby, Toy and Game Stores US NBB 812 Eating Places 722110 Full Service Restaurants 1996 1974 1986 US NBB 944 Jewelry Stores 448310 Jewelry Stores 1988 US NBB 8243 Data processing schools US NBB 961 Catalog and mail order houses US NBB 8299 Schools & educational services US NBB 7999 Amusement and recreation nec US NBB 61 Recreational Vehicle Dealers 1 Regal Cinemas Inc Bankrupt 2001 Entertainment US NMS 7832 Motion picture theatres ex drive in 2 Rx for Africa Inc Non Bankrupt 2001 Leisure Equipment 3 Salex Holding Corp Bankrupt 2000 Miscellaneous Consumer Services 4 Samuels Jewelers Bankrupt 2003 Retail Inc Specialty Schlotzskys Inc Bankrupt 2004 Hotels, Restaurants & Travel 6 Showscan Entertainment Inc US NBB 7822 Motion Picture and tape distribution US NBB 749 Automotive Services 611423 Computer Training 44111 Electronic Shopping 611699 All Other Miscellaneous Schools and Instruction 711219 Other Spectator Sports 441210 Recreational Vehicle Dealers 12131 Motion Pictures Theatres 12120 Motion Picture and Video Distribution 811111 General Automotive Repair 198 1986 1976 1996 1989 1986 1990 US NBB 944 Jewelry Stores 448310 Jewelry Stores 1982 US NBB 812 Eating places 722211 Limited Service Restaurants Bankrupt 2000 Entertainment US NBB 7832 Motion picture theaters ex drive 12199 Other Motion Picture and Video 1993 1984 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 82

7 Sight Resource Corp 8 Southern Investors Service Company Inc Bankrupt 2004 Diagnostic & Health Related Services Bankrupt 200 Hotels, Restaurants & Travel 9 Spiegel Inc Bankrupt 2003 Retail Specialty 60 Standard Management Corp Non Bankrupt 61 Star Buffet Inc Non Bankrupt 62 Steakhouse Partners Inc Non Bankrupt 2002 Retail Food & Beverage, Drug & Tobacco 2000 Hotels, Restaurants & Travel 2001 Hotels, Restaurants & Travel 63 Strouds Inc Bankrupt 2000 Retail Furniture & Home Furnishings 64 Stylesite Marketing Inc 6 Tender Loving Care Health Care Services Bankrupt 2000 Apparel, Footwear & Accessories Bankrupt 2002 Diagnostic & Health Related Services 66 Tesseract Group Inc Bankrupt 2000 Educational Services 67 Tilden Associates Non Bankrupt 68 TIX Corp Non Bankrupt 69 Tops Appliance City 2000 Miscellaneous Consumer Services Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms US NBB 8099 Health and allied services US NBB 7011 Hotels and motels US OTC 961 Catalog and mail order houses US NBB 912 Drug Stores and Proprietary in Industries 423460 Ophthalmic Goods Merchant Wholesalers 721199 All Other Traveler Accommodation 44111 Electronic Shopping 446110 Pharmacies and Drug Stores Stores US NBB 812 Eating Places 722110 Full Service Restaurants US NBB 812 Eating Places 722110 Full Service Restaurants US OTC 719 Misc. home furnishings stores US NBB 2399 Fabricated textile products nec US OTC 8082 Home health care services US NBB 8299 Schools & Educational Services US NBB 749 Automotive Services 2001 Entertainment US NBB 7999 Amusement and recreation nec Bankrupt 2000 Retail Appliances and US NBB 722 Household appliance stores 442291 Window Treatment Stores 1992 1972 196 1989 1997 1996 1987 812331 Linen Supply 1993 621610 Home Health Care Services 611430 Professional and Management Development Training 811198 All Other Automotive Repair and Maintenance 713990 All Other Amusement and Recreation Industries 443111 Household Appliance Stores 1999 1986 199 1993 1992 83

70 Toys R Us Inc Non Bankrupt 71 U.S. Physical Non Therapy Bankrupt 72 U.S.A Floral Products Inc 73 Ultimate electronics Inc 74 Unapix Entertainment 7 United Artists Theatre Circuit Inc 76 United Petroleum Corporation Electronics 2001 Retail Specialty 2001 Hospitals & Health Care Facilities Bankrupt 2001 Retail Specialty Bankrupt 200 Retail Appliances and Electronics Bankrupt 2000 Entertainment US NBB 7822 Motion Picture and tape distribution Non Bankrupt 2001 Entertainment US 7832 Motion Picture Theaters ex drive in Bankrupt 2001 Retail Food & Beverage, Drug & Tobacco 77 USA Biomass Corp Bankrupt 2000 Sanitation Services Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms US NYS 94 Hobby, toy, and game shops 41120 Hobby, Toy, and Game Stores US NMS 8093 Specialty 621498 All Other Outpatient Outpatient Care Clinics Centers US OTC 193 Flowers & 424930 Flower, Nursery Florists supplies Stock and Florists Supplies Merchant Wholesalers US NBB 731 Radio, TV & 443112 Radio, Electronic Stores Television and Other Electronics Stores 12120 Motion Picture and Video Distribution 12131 Motion Picture Theaters (except Drive Ins) US OTC 411 Grocery Stores 44120 Convenience Stores US NBB 493 Refuse Systems 62211 Hazardous Waste Treatment and Disposal 78 Valley Media Inc Bankrupt 2001 Entertainment US NBB 7822 Motion picture and tape distribution 79 Video City Inc Bankrupt 2004 Entertainment US NBB 7841 Video Tape Rental 80 Video Update Inc Bankrupt 2000 Entertainment US NMS 7841 Video Tape Rental 81 Vision America Inc Bankrupt 2001 Hospitals & US NBB 8011 Offices & Health Care clinics of Facilities medical doctors 82 Wall Street Deli Bankrupt 2001 Hotels, Restaurants & Travel 83 West Coast Entertainment Inc 12120 Motion Picture and Video Distribution 32230 Video Tape and Disc Rental 32230 Video Tape and Disc Rental 621111 Offices of Physicians (except mental health specialists) US NBB 812 Eating Places 722211 Limited Service Restaurants Bankrupt 2001 Entertainment US OTC 7841 Video Tape and Disc Rental 32230 Video Tape and Disc Rental 1928 1992 1997 1993 1986 1926 1970 1988 1998 1984 1983 1984 1966 199 84

84 Western Sizzlin Corp 8 Whole Foods Market Inc 86 WHSU Inc (fka Micro Warehouse Inc) Non Bankrupt Non Bankrupt 2003 Hotels, Restaurants & Travel 2000 Retail Food & Beverage, Drug & Tobacco Bankrupt 2003 Retail Specialty 87 Zany Brainy Inc Bankrupt 2001 Retail Specialty US NBB 812 Eating Places 722110 Full Service Restaurants US NMS 411 Grocery Stores 44110 Supermarkets and Other Grocery (except Convenience) Stores US NMS 961 Catalog and mail order houses US NBB 94 Hobby, toy and game shops 44111 Electronic Shopping 41120 Hobby, Toy and Game Stores 1992 1980 1987 1991 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 8

Appendix B: Financial Data ** Dollar values are recorded in thousands of dollars Group 1 1 year before bankruptcy DM U Current Assets 1 Current Liabiliti es 1 Workin g Capital 1 Retaine d Earning s 1 Operati ng Income 1 Book Value of Equity 1 Total Assets 1 Total Liabiliti es 1 Employe es 1 Sharehold ers 1 1 88821 61412 27409 118678 4473 117816 1927 77441 2400 86 2 98018 4196 6422 29920 16133 73727 123811 0084 4091 87 3 12417 16214 3797 19061 262 12004 1430 2634 128 69 4 6068 76430 1772 16916 20907 18877 267913 46688 220 100 301 2822 679 99081 13621 1064 13890 3326 10 6 463 89137 8474 1641 1071 9740 3811 9821 298 8 118880 113013 7 194003 22360 29602 0161 18763 8669 6 12800 497 8 66798 138719 71921 39806 44 624 184281 178027 220 309 3141.1 1122.8 2018.3 1691.96 2472.36 3673.3 1201.1 9 93 63 3 83.449 6 8 72 187 880 10 0 6729 6729 46262 3110 4924 236924 191000 19972 19914 11 870197 76672 10344 1344 70610 39682 7 32700 6200 12 864 31221 27424 76186 19838 77267 118160 40893 2313 128 13 31289 166893 146002 197438 4736 2278 46183 23420 800 2100 14 240 131677 107622 8118 3848 26 30129 29781 1200 6400 237.2 18777. 16419. 124.77 24129.9 221. 1 42 02 8 13389. 0 6 8 7 376 134 16 9903 107038 8003 38771 4366 30836 31244 343290 700 17 62391 737 11184 6398 179 390 116402 76897 288 127 18 228829 111671 11718 98401 19600 203261 38944 186283 8700 341 13310 111790 204710 120060 19 0 0 217200 170300 13100 84600 0 0 30000 4163 20 1377 437 4060 8182 8274 417 6963 11138 326 108 1266. 13319. 662.67 29874.4 23620. 21 67 3 2 100474 3363.7 624.21 24 346 44 22 94428 67319 27109 30102 9964 129121 777033 647912 9097 740 86 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

23 6390 8300 27110 2007 26740 112384 493072 6046 16300 0 24 2334.7 110362.2 124983. 38174.1 30119.0 8 89726.2 9 376627. 2 286900.9 6977 700 2 9198.7 72 180.8 64 7347.9 08 17331.2 8336.9 434.1 1114.0 6 6619.9 07 293 70 26 7384 46244 27340 29127 7837 120066 17099 0893 773 4200 27 3184.1 88 6402.4 2 3218.2 6 24480.9 4494.39 273.91 4 1793.2 12679. 29 411 108 28 2461 22032 1971 6243 3981 171 286 26462 6 21 29 7397 72397 160 3912 13699 419 124308 79149 2128 121 30 80171 213312 133141 19182 13943 14019 97183 237378 110 190 31 3733 10401 66680 1876 12617 168 12112 10467 30 2900 32 103291 842 17749 2609 30748 3800 126926 91126 4000 230 33 016 69877 19712 3973 9890 22272 190717 212989 3900 2200 34 211970 186138 2832 6262 21179 2727 2622 237968 03193 618 3 108007 17806 70049 2928 736 21978 260416 238438 6087 1463.4 3076.9 97 440.4 49 2997.36 7 328.3 6282.9 02 12 8 36 13681.6 191.218 37 22247 1384 6863 146924 624 7179 10222 27343 2609 298 38 214286 49301 16498 22043 193027 21386 4218 209772 940 942 11194 227664 17140 39 998169 4 11777 287414 200212 70244 4 0 29 403 1917.8 47.38 1370.4 4343.24 4890.63 47.38 40 2 8 32 686.47 277.648 9 7 8 8 1119 41 144264 136472 7792 13023 11119 2 1069 1 216764 3 42 69898 112631 42733 102463 114 46972 8374 13046 1100 1300 43 2632 8366 734 32179 84 1710 3084 1144 198 313 44 92 28620 1936 48888 788 1331 76649 7318 4700 4200 4 8418. 49 3464. 66 0772. 83 4140.3 2007.04 8637. 1 12127. 2 38769. 67 4387 94 46 32938 36086 3148 60647 42196 132669 20830 73161 6200 123 47 21920 178302 73618 77483 71811 307617 10892 0 781633 13000 2100 48 4867.3 2 877.0 99 3889.7 8 4803.4 7779.34 480.93 3 16980.1 4 9218.9 94 136 11344 49 163292 113380 49912 43368 8 77344 22981 1207 6264 40 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 87

0 72193 7316 1323 1273 1098 46267 142262 18829 88 36 1 1192. 44 1041. 61 109.8 27 2112.7 92.18 388. 19673.0 3 2331. 8 122 1 2 113027 4 749941 380333 237806 18044 60102 19477 2 13426 0 2200 3300 3 777700 0 43800 0 339200 0 10110 00 419100 0 10040 00 21380 00 63700 0 227000 212010 4 8094 48838 3176 3438 4033 21097 17978 18661 4381 843 3124 42227 10703 2438 10333 1832 3908 2076 903 108 6 283683 124784 18899 498 107829 324638 84711 260073 800 310 7 7090.3 4 2219. 4 4870.8 1621.24 2368.02 3 281.96 6 701.0 6 2219. 4 1 412 8 2760 1130 8770 9824 14646 71 806 11676 92 900 9 730397 37882 6 3042 9 226241 0 396687 93707 337908 0 43161 73200 1600 60 1260 882 9382 8879 4237 7166 267274 19618 4300 1140 61 2274 29190 644 3079 4937 326 30992 34248 429 229 62 62003 3333 2860 840 9073 4708 8894 43246 1241 136 63 88663 240649 11986 101490 42619 18041 28690 240649 1170 110 64 124822 106099 18723 48888 44226 2303 11200 1202 7 67 6 6387 2834 2783 1987 308 414 79316 37771 390 330 66 4762 64224 16662 4461 67 14602 9389 79293 100 1100 67 183242 3018 148224 63129 20143 144161 24921 10360 6119 79 16778 10148 68 70049 80147 780098 48971 284060 176097 6 9 69 891.2 2 31773. 67 2882. 4 16248.2 3290.8 13821.8 8 4607.7 1 3223. 83 320 200 70 121.3 4 4496.6 1 324.2 7 2736.3 8 31.28 1829.64 8373.71 8 0030.4 6 4166. 7 1700 630 71 2860.9 8 97.3 63 877.72 27.0 16947.7 28982.7 8 1203. 08 2300 64 6811.4 4823.8 1987. 3841.93 8699.90 487.9 72 37 84 3 869.84 908.43 8 3 6 129 104 73 394 6281 26987 14432 19102 44340 128429 172769 9 4097 387.1 44266. 1672.3 37.99 1043.3 7469.8 23616. 74 4964 03 9 4 7 3 1 48 936 147 7 100 788 712 3731 1686 4237 6667 2430 8 8273 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 88

2 years before bankruptcy DM U Current Assets 2 Current Liabiliti es 2 Workin g Capital 2 Retaine d Earning s 2 Operati ng Income 2 Book Value of Equity 2 Total Liabiliti es 2 Total Assets 2 Employe es 2 Sharehold ers 2 1 13490 2776 82129 7737 781 18918 224641 6723 2200 117 2 8148 37980 47168 20413 10763 63681 107392 43711 3339 96 3 18876 18327 49 10846.262 384 21970 2824 180 6 1406 31297 107699 278441 386140 246 100 4 73473 103293 29820 21,68 13789 7779 78277 20190 29890 44634 14744 16 20 6 634 3866 32302 241 3966 684 496 101649 289 8 7 10288 18906 86468 026 837 1146 97730 86026 12300 490 8 66764 10824 41760 86013 3934 47477 20167 249044 290 3049 9 3382.7 13 1280. 7 2102.1 6 1744.7 98 108.19 27.64 1 4033.32 8 147.6 87 179 901 10 128,286 37048 91238 9 28881 90238 20383 16014 11 92,927 662071 29086 10143 29488 63266 19729 4 134002 8 3300 6200 12 77 31098 24677 6848 16830 69276 111279 42003 2170 1321 13 270908 163049 10789 167297 8621 191914 422963 231049 9100 2000 14 37,460 111482 74022 311 47441 2746 396221 330408 14000 300 3026.2 9 9777.9 1 1,10 4176.7 43 4 0 13.03 3 24880.6 1 19261. 43 419 338 16 70,02 80167 966 360 41077 27199 247664 274863 200 17 8382 76387 899 47 4797 3187 11683 80496 243 112 18 241487 99176 142311 9217 23076 198862 40038 206176 9000 337 19 119820 0 102790 0 170300 126700 166200 802600 191480 0 111220 0 30000 421 20 2,133 7042 4909 70187 1444 3429 1070 7276 34 121 21 31,76 1270. 3 1919. 03 66471. 8 2210.2 1311 3284.1 8 22820. 37 297 46 22 3094 10946 732 44974 1628 204197 807494 603297 11068 721 23 21493 6230 4087 31 36498 20970 486874 46904 1900 0 24 188,616 6826. 93 120088.8 46202. 8 17709.1 1 78183.3 2 324034. 7 2481.3 614 728 2 9,70 2070.4 38 7634.1 6 860. 3 8192.62 13214.8 3 19707.1 2 6492.2 9 30 6 26 6220 26989 331 922 344 80607 110761 3014 3700 1600 27 2,78 2476.2 8 281.8 8 19249. 1393.6 700.16 6 1139.2 2 8134.0 3 439 140 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 89

7 28 7141 24017 16876 8689 10201 78 37473 29199 69 181 29 77,972 671 10421 13727 4073 436 130736 76200 2184 9 30 122,096 16400 42404 131737 7402 1124 12363 264908 2024 2017 31 4,898 22239 17341 31007 11876 44789 149298 10409 60 2100 32 141,83 841 7068 46041 467 844 176897 92442 0 260 33 67,27 69148 1891 10302 7923 6869 209799 202930 3900 2400 34 186,324 121692 64632 21748 1204 222 230689 22392 46081 0 3 114,4 97193 1732 22663 794 28386 28331 24929 649 2168.4 100.18 6 2268.6 1 3486.78 4 378.742 386. 26 12 3374 36 2 1342 101.148 37 31280 10943 20337 19993 96 63877 8981 2974 2864 334 38 26,817 4996 21821 30826 29099 24098 30666 668 306 127 10663 11003 248731 178782 39 8 34003 29640 136784 699482 1 9 3283 29 1924.2 8 660.8 2 1.09 40 1862. 09 3786.7 84 4371.49 9 818.28 3 3786.7 84 0 1101 179463 41 10341 141242 9099 118917 912401 882233 4 42 88,666 60846 27820 4214 28034 13472 171690 18218 1400 1400 43 3,98 8089 4104 26173 1331 21632 36329 14697 2046 317 44 14,749 270 1096 397 10074 2434 89431 9186 400 2000 4 82,141 3101. 7 47039. 67 42707. 91 9447.72 3 8682.7 4 12680. 6 38827. 82 4677 88 46 2,783 32291 608 1807 9420 13973 173023 3700 377 0 12443 47 209,280 992 786272 98743 40349 46663 892231 0 0 48 919.29 3146.9 77 2227.6 9 43174 748.24 246.12 2 9760.44 1 376.8 44 77 1200 49 14813 10186 4667 3931 9278 7791 213484 137693 624 32 0 7,070 7337 1713 122722 39890 46799 1612 202411 1077 41 401.1 1 1 14,729 617.7 27 9111. 8 6246.60 4 16738.9 9 39129.1 2 21911. 33 11 74 20971 14883 1,293,4 2 8 70211 91397 26410 190071 60914 3 9 23100 3300 639000 36600 273400 791400 37900 123410 170810 473300 3 0 0 0 0 0 00 00 0 201000 196126 4 72,328 42834 29494 28649 71 2764 16784 140200 42 922 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 90

38,217 23646 1471 9449 848 120 9831 4481 96 107 6 42,449 18471 267698 20819 20627 394686 727742 33306 9400 3229 7 421 198.0 4 222.61 1 638.69 26.94 239.928 437.982 198.0 4 16 107 8 3,122 8108 4986 81189 948 642 14693 8268 12 83 9 821,090 479890 341200 22483 73249 133196 39312 8 406116 3 84000 1600 60 131,73 63742 67993 61666 1220 74443 233664 19221 0 1190 61 1832 28933 1081 37 170 8036 2614 34181 443 229 62 732 29240 26492 10864 38 43786 8471 4096 1174 143 63 149,700 239680 89,980 2820 1916 93601 337117 24316 1600 114 64 160,74 94262 66312 9634 23921 6074 1811 112790 0 96 6 70247 42487 27760 27806 2126 49494 9798 48104 790 300 66 44,664 0289 62 0 062 1996 97468 1024 1600 1371 67 193,396 28922 164474 7687 9798 19134 267644 10810 6246 798 68 60,228 21381 13623 332 23000 618802 190738 9 12888 7 27499. 19961. 10080. 19797.4 016.0 3038. 92 3 9 8724.46 7 4 7 30 192 69 7,39 70 71 106.2 6 2648.2 96 6860.6 79 4683.8 27 802.3 37 100.6 7 3177. 6 314.0 4 6029.1 7 267.19 2 6873.7 1286.62 97 2 32.89 7 79.839 7609.34 9 1717.2 1 037.41 9 467.6 4 30066.6 1 10333.1 39146. 3 1700 630 14349. 4 2191 66 29.7 33 1311 104 1760.0 72 09 2397. 6818. 4487. 10879 73 9 71 8 19932 1496.2 6024.1 9062.1.2 68 2730 6997. 6796.3 0200. 1041. 318.13 032.6 8601.1 372. 74 31 46 97 1 7 7 7 1194 172 7 2910 272 2638 1899 1741 607 6329 272 3 18193 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 91

3 years before bankruptcy Workin g Capital Retaine d Earning Operati ng Income Book Value of Equity DM U Current Assets 3 Current Liabiliti es 3 3 s 3 3 3 Total Assets 3 Total Liabiliti es 3 Employe es 3 Sharehold ers 3 1 12120 386 8619 1027 8171 109003 1823 7332 2100 101 2 7274 28649 4662 1386 897 6972 90617 3364 3119 93 3 8281.7 23 964.7 0 1282.9 8 422.1 12193. 9 1039. 09 96 6 390.797 164. 140340 26092 96020 28013 37633 2430 98 4 79032 79897 86 62902 128 61644 3084 6416 93100 9438 128 700 20 6 706 8778 8013 12180 37417 4919 4949 9414 293 9 7 10764 173836 66191 21042 10214 1394 79780 6632 12700 480 8 7870 43906 13964 39431 44199 8781 21771 22632 0 3629 9 339.7 6 1268.4 84 2091.2 81 1803.0 98 116.80 2634.8 46 427.22 2 1622.3 76 190 94 10 108642 4873 9907 30188 20971 107099 211786 104687 11 70372 71274 11702 88016 76080 324008 148339 3 11938 36400 6487 12 1286 239 2747 6831 16010 838 93126 34768 1916 1364 13 3493 123994 23099 279672 382 33423 48247 123994 8900 2000 14 34831 141333 10602 10071 9606 8792 609681 448864 20000 200 1 1103.0 12 498.4 37 349.4 3 77.0 0 6786. 24 26228.3 7 18978. 03 68 423 16 70161 86844 16683 3717 37198 27186 24946 282132 100 17 82980 6317 19823 0790 274 3904 110106 7102 27 117 18 2637 11487 141788 84627 24831 190691 417492 226801 9700 327 19 113330 0 988800 14400 36400 167300 71100 176660 0 1010 0 27200 406 20 7768 3702 4066 08 697 17916 2193 0 40 121 21 37380. 84 9909.7 34 27471. 11 4340. 2 28998.7 37614. 81 6218.9 6 2444. 1 270 39 22 28831 8466 834 6687 2183 226309 69743 471234 10234 744 23 26400 938 3298 2174 4804 2669 47377 446918 16100 0 121367 47394. 19638.9 7262. 33067. 29804 24.6 7 4 1 4.9 6630 809 2 190691.9 12031. 4 69324. 3 239.8 91 9491. 20.121 1674.36 21814. 88 28877.7 7062.8 68 43 68 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 92

26 38390 1812 20238 2684 1446 8467 7933 20886 2730 1800 27 3960.3 79 307. 8 42.82 1 17663. 3 6894.7 891. 38 17146.3 84.7 8 487 143 28 6001 19 806 39430 388 991 16047 6096 10 137 29 78210 9349 18861 1908 14846 07 11271 6416 1291 4 30 14134 134296 708 88129 1826 69310 19006 264316 268 2200 31 230 16699 11469 3489 18381 012 13267 10311 616 2100 32 1074 77606 72968 46901 381 87461 18082 93391 2800 0 33 49769 8832 9063 7806 10981 927 181331 172074 3600 948 34 187177 108064 79113 17949 10797 1390 219784 22460 444344 0 3 112717 99769 12948 1876 16443 32237 261624 229387 6337 36 89.02 2313.0 86 2223. 8 13041. 9 40.69 3077.3 38 698.281 377.6 19 36 3374 37 20970 974 11216 180343 8441 460 66636 21076 272 112 38 1811 3 146 2268 2270 134 2927 182 0 0 39 66907 39296 117611 146322 89234 249694 692.8 10946 1 84767 3283 210 40 283.82 7 17.19 266.63 2 88.27 416.74 8 434.73 17.82 10 0 41 124939 12906 467 126691 6874 6674 72490 42 117884 48609 6927 11720 484 44648 19732 12704 1889 100 43 6198 6914 716 2444 1394 2329 34840 1181 1763 316 44 133 28712 1177 63918 7796 9162 89463 9862 600 2000 4 81307. 91 36322. 31 4498. 6 36699. 22 194.8 79387. 94 118793. 7 3940. 73 442 84 46 6702 29 1443 1732 2178 6948 14727 7779 194 1198 47 197829 172640 2189 293210 70634 13210 106083 9 114904 9 1000 4000 48 961.91 7412.1 8 640.2 8 36383. 1 7192.2 8.66 9 937.67 2 8020.3 69 49 170987 103692 6729 391 21 90091 247933 17842 6640 233 0 970 141 4 13490 38080 60162 130039 189316 1087 32 1 2 3 18808. 41 20868. 07 209.6 7 414.4 6013.27 17749. 29 11340 7 83920 0137 327644 190608 642621 493300 0 28700 0 207600 0 87600 0 266100 0 874000 0 4294.6 18371 8 13460 00 20868. 07 136 73 11943 7 18200 3100 472200 0 16700 11178 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 93

4 66624 40320 26304 243 692 31868 19204 127336 4340 708 3314 12707 20438 1901 1933 20428 46647 26219 82 110 6 40773 188843 261930 8191 38940 382499 728982 34643 8400 3286 7 8 339 9664 630 82339 497 270 1230 9960 12 1972 9 71791 64474 63117 449 437367 108816 1 06314 4 397498 3 86000 172 60 131798 704 76094 8160 9813 70937 23687 16938 0 120 61 1891 27001 11110 394 383 8448 24099 3247 442 229 62 46 28394 27062 12230 9210 42313 821 40202 110 14 63 137313 4298 9471 632 13294 119140 32976 21062 1400 1166 64 120219 83293 36926 1410 27 2466 13710 111639 0 74 6 6829 27737 40792 327 2208 6388 98020 41632 1100 0 66 43462 46819 337 122704 6220 93 93272 3060 100 16 67 19740 33407 163998 83338 4246 1680 277434 11184 6311 809 68 8267 197094 111827 193 323 66296 180620 1 114090 69 10117. 8 7143.1 2 2974.4 2 417.01 7 1066.73 8 1792. 07 34769.0 6 16844 20 1204 70 2197.6 4230. 86 2032.9 3 6711.9 1 1707.69 6 6337.0 22 37880.8 4 3143. 81 1300 66 71 2820.2 74 4740.1 13 1919.8 4 602.1 21 2222.78 7 14888. 66 3108.8 3 16620. 16 230 66 72 6433.3 8 4097.1 79 2336.2 01 62.33 6 844.139 213.2 3 9988.44 7 477.2 17 133 49 73 18973. 41 6980. 86 38007. 4 16246 1898.2 4 6377. 1 72246.8 4 136003.9 487 2730 74 61083. 62 1261. 72 48467. 9 14944. 97 177.94 0229. 13 89333.3 39104. 22 1419 172 7 7401 316 708 48920 2330 8731 9047 316 3 666 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 94

4 years before bankruptcy Workin g Capital Retaine d Earning Operati ng Income Book Value of Equity DM U Current Assets 4 Current Liabiliti es 4 4 s 4 4 4 Total Assets 4 Total Liabiliti es 4 Employe es 4 Sharehold ers 4 1 29973 28023 190 134 641 672 81746 63382 0 0 2 69697 26976 42721 8192 830 1171 8237 31186 2736 89 3 6167. 7 6104.2 43 63.332 4 910 7767 1288 414.0 4 623.337 1421.3 79 800.16 7078.7 86 99 6 141062 3619 8867 302202 388069 2810 98 29918. 21 3234.9 08 26683. 31 10241. 246.66 31883. 88 3399. 6 31.6 83 110 170 6 12263 22319 1006 2747 461 62301 10463 167764 604 9 7 83672 134267 09 060 314 170012 718213 48201 10300 470 8 69399 4400 2394 197337 12301 16133 164786 326319 0 0 9 3221.6 67 1221.2 43 2000.4 24 1719. 2 1.372 280.1 66 4301.4 1 1721.2 8 18 94 10 442 4097 3280 1273 0 043 104484 49068 11 12646 344477 168169 4186 66233 173382 610042 436660 2000 8274 12 42210 28312 13898 4670 12861 48260 88126 39866 1729 141 13 302942 1068 19727 364491 4667 29697 402282 1068 7800 2000 14 4622 8417 38292 16630 42461 112624 6696 428972 20000 300 1 116.9 47 091.3 43 3934.4 4391.7 9 1487.89 3 901.4 34 21461.2 11812. 43 397 267 16 6262. 29 7203. 34 981.0 3297. 6 3296.6 24310. 26 264617. 1 288927.4 400 17 3382 44049 9333 061 1809 16810 62727 4917 163 60 18 246448 117471 128977 74912 27803 180211 378119 197908 9000 323 19 101840 0 881400 137000 700 138000 98100 13490 0 936800 24300 3744 20 2249 4869 2620 46973 6131 497 14409 0 76 0 21 89.2 38 7186.1 29 1290.8 9 16201. 3 1371.9 1100. 4 1410.3 9 14710. 7 0 0 22 3097 63032 3207 9784 694 202872 619997 41712 1000 2720 23 28003 3946 2943 466 3242 19019 476912 47893 17200 0 24 27238.7 90236. 39 182122.4 0373. 144472 71989. 26 38220. 8 31031. 10341 431 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 9

2 2 9988.0 76 1444.6 4 843.4 22 1161.2 3 1009.33 21768. 1 23311.1 7 142.6 4 400 84 26 27688 1737 1191 3182 12802 27298 64479 37181 200 0 27 1048.2 74 7101.4 7 603.2 9731.2 7 169.6 1913.0 71 12433.7 1 1020. 64 378 83 28 341 324 1817 344 10241 11908 1617 3709 111 113 29 880 33811 24994 7420 8113 39792 7727 393 0 0 30 18221 1619 20260 6314 14712 47982 234694 282676 296 2200 31 7306 1186 4280 2970 16771 86472 142921 6449 27 2100 32 182331 103 76778 43197 232 83327 213379 13002 2900 300 33 42492 097 1260 9417 8206 644 16104 1869 300 3000 34 20631 130229 76302 1628 1011 4967 243319 23083 496402 4908 3 108922 97192 11730 2372 179 26688 24748 228060 7217 36 1.13 233.7 2302.6 3 12874. 69 896.169 2829.9 69 862.234 3692.2 03 36 3342 18133 7439 43172 6643 23281 270 63 37 20671 876 1209 38 39 363968 197906 166062 103648 4880 203701 40208 198807 119 203 40 41 110491 110934 443 62339 973 621223 630796 400 42 130326 46874 8342 332 7069 3347 19966 146219 2037 1600 43 4888 7796 2908 279 1997 22001 3219 1018 1876 327 44 19741 28102 8361 9180 979 41612 6809 110121 200 2000 4 66427. 28 27766. 93 38660. 36 23909. 6 919.3 8 66144. 88 96662.0 9 3017. 21 4024 101 46 632 209 1116 300 1918 23 1094 341 170 1236 47 20788 178903 2868 10986 12288 44144 11820 6 114106 2 1000 3000 48 49 17868 6688 92180 80633 2898 116133 22211 106018 730 282 0 46063 33608 124 148301 384 72083 100600 171811 936 0 7393.1 1 632.0 9 4479.3 92 112.6 67 2012.8 3 412.8 98 1340.6 4 781.3 04 71 76 128896 2 89316 367467 27849 301071 170671 18983 0 769977 1300 2300 446000 24600 200400 443000 191200 709800 112290 412800 3 0 0 0 0 0 0 00 0 12000 104868 4 62618 40729 21889 13764 81 42324 166041 123717 4287 77 16389. 102. 11286. 277.4 102.27 11838. 2063.7 881.0 394 80 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 96

43 92 84 9 9 66 1 49 6 424182 18334 240828 14172 4008 39667 71608 3941 9100 3411 7 8 4016 9464 448 82642 90 4962 1603 11073 80 100 9 41190 37641 749 79814 279771 3486 199310 7 14824 2 9000 68 60 129111 0296 7881 62 776 69201 233279 164078 4600 1330 61 12974 30232 1728 3603 3698 13367 22998 3636 42 244 62 1783 2641 26142 18088 668 36398 77144 40746 1470 11 63 141661 260 116,011 4142 22443 12400 26938 14308 1012 1296 64 11400 68944 4061 7769 482 31136 124368 93232 0 0 6 6702 22701 44801 36041 98 6303 9861 380 120 0 66 3774 39822 2248 10406 12733 426 102361 289834 1600 12 67 172843 20889 1194 88106 791 170408 272333 10192 6364 101 68 18209 6714 4930 2029 16931 32411 7281 404040 69 388.2 88 327.8 61 312.42 7 1869.7 3 144.791 4684.3 67 1026.4 4 842.0 7 133 126 70 1240.4 79 387.3 24 2346.8 7216.9 3 301.744 431.1 73 2704.2 3 2214. 06 1300 68 71 279.0 64 4062.6 36 1483. 7 17.9 31 1994.07 6 14392. 33 31043.3 1660. 97 200 33 72 137.8 3 2020. 03 3117.3 221.18 4 321.204 4323.4 91 6443.99 4 2120. 03 1200 18 73 17627. 33 37818. 8 20191. 3 1347 1608.2 4443. 7 6266.4 1 117100.1 437 2730 74 74200. 07 6731.7 3 67468. 34 20074. 73 3306.16 1 963. 07 9698.4 7 36963. 4 177 189 7 6214 67 39 8808 481 667 1137 12042 3 18930 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 97

years before bankruptcy Workin g Capital Retaine d Earning s Operati ng Income Book Value of Equity DM U Current Assets Current Liabiliti es Total Assets Total Liabiliti es Employe es Sharehold ers 1 2 789 16921 40974 427 7781 47086 66067 18981 0 0 680.61 8 4118.7 3 187.3 39 867.9 7 7 172.88 1294.6 48 760.9 7 636.3 09 0 6 4 82270 8243 2973 1648 46084 109389 303812 413201 283 96 2888.2 3629.6 741.32 8776.3 267.0 6461.9 3786.8 9 17 7 698.74 4 11 71 0 80 6 14470 26701 12231 6880 013 2674 113788 166462 62 9 7 46477 1076 9099 37603 68669 18918 48348 32440 900 48 8 69309 47818 21491 191883 12667 148620 181098 329718 0 21 3242.2 1397.2 184.0 1666.3 229.8 402.6 1972.7 9 74 16 8 24 394.214 77 7 93 198 940 10 16693 1004 6639 3990 0 18736 31061 1232 11 46843 329466 138969 19640 270 108186 36793 428607 20000 61 12 47288 24146 23142 3880 922 737 91248 311 1627 1701 13 273780 11467 1910 329170 019 2961 374236 11467 700 2000 14 10098 239118 1 1299.4 88 3278.3 29 138133 162411 72631 112029 67097 372473 20300 17000 1978.8 3102.8 183.66 1011. 20819. 10308. 4 7 4 33 99 66 443 270 16 61987. 07 73364. 62 11377. 29017. 74 38637.0 7 20213. 01 284698.4 304911.4 4600 17 49722 3013 1969 4164 33168 30143 61800 3167 120 7 18 224212 101134 123078 63701 29170 124394 32686 228292 8600 30 19 846400 63400 211900 13900 103100 11400 121100 0 6600 22800 312 20 341 9 2414 3892 10889 3617 17676 0 0 0 1834.1 134.17 1 21 13028 1411.8 11616. 2 7 0 1307. 63 1891.8 244 0 22 26111 3698 2787 77668 843 177930 489383 31143 9874 763 23 1714 39210 21696 9434 28364 101 320027 304976 13200 0 24 336761 131680 20080 61888. 184037 26081 342044 1179 470 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 98

.6.4 07 3968.2.4.7.3 2 3418.9 96 1488.1 07 1930.8 89 38.90 1 467.343 72.6 44 7363.1 49 1610. 0 80 3 26 16172 33802 17630 62266 4062 1173 32073 43808 0 0 27 1388.1 63 2181.4 77 793.31 4 426.9 338.3 141.7 13 890. 31 7408.8 18 28 2 28 18104 6934 11170 29910 13302 1164 29696 1021 169 96 29 2963 21093 842 1819 093 1716 39808 2262 0 22 30 17672 209220 32648 6026 164 38868 222603 261471 2249 2300 31 973 1479 906 24316 34041 8170 114633 32883 0 2300 32 180210 94981 8229 6277 9392 101742 210636 108894 0 0 33 4968 4882 860 7707 8264 7913 170 16792 4100 3000 34 20817 127962 8019 1106 10372 484 243898 28637 023 191 3 36 134.64 1 1803.4 61 1668.8 2 988.1 01 1893.70 6 234.8 4 291.8 97 3186.7 1 60 3342 19912 6003 28913 9478 306 260 882 37 16974 832 8649 38 39 263149 134890 12829 70219 39630 164090 30042 13633 113 214 40 41 129836 10864 21182 2760 2043 63777 448342 432 74 42 12273 28406 94167 2793 12221 4482 183646 129164 2226 1700 43 3678 700 3822 2724 679 20212 3187 1164 197 349 44 128 2620 13692 9391 107 40098 6801 10899 400 2000 4 74. 93 24117. 27 31628. 66 178. 4 2.461 60030. 6 86442. 19 26411. 33 107 46 686 198 488 6933 162 3237 8822 8 0 1264 47 314269 174126 140143 824184 116236 824339 13947 6 204489 6 1000 2600 48 49 13464 83371 1193 80168 17223 71718 21298 14380 330 233 0 42307 41718 89 19894 12443 83822 92600 1764 1032 0 688.8 1 287.4 3 2344.1 4 2943.3 13 2968.42 4 004.8 07 1066. 73 061.3 6 6 6 120893 2 82682 29886 4096 27316 184207 490390 7 71847 13063 1700 370937 184212 186724 34069 1336 918 934171 33899 3 3 6 7 2 0 6 0 4 100000 78928 4 7783 37214 2069 3120 113 2941 168486 114 441 420 493.1 3372.0 163.0 21.04 1.437 170.0 688.7 4838.6 180 29 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 99

0 72 33 73 72 99 6 71310 437263 27842 311873 0 63192 13796 6 744041 1900 379 7 8 998 37604 27646 81707 9672 4713 43600 38887 900 100 9 313732 177417 13631 3391 234321 43128 143373 0 100220 2 23000 606 60 149228 317 9603 311 370 67987 260418 192431 400 140 61 13886 29643 177 32974 7280 10306 243 3741 418 262 62 2976 2434 28631 18787 236 3699 81410 4711 1420 169 63 12993 2909 100876 972 11278 12098 26066 139698 1000 0 64 94762 48218 4644 3400 914 3420 10012 6902 0 0 6 719 23146 48413 3464 1316 61848 101920 40072 120 0 66 7232 62387 9938 396 32936 760 1289 29008 0 0 67 162830 26717 136113 8773 12688 148642 272312 123670 7416 1071 68 10287 716 47229 2373 16891 32460 721372 396722 69 3268.9 76 1971.7 9 1297.1 81 208.0 6 1817.46 3769.0 26 6176. 23 2407.4 97 0 0 70 200.1 29 398.7 22 1098. 9 830.8 84.107 882.9 2201. 4 19318. 4 1200 700 71 2203.7 9 321.2 3 1047.4 4 4324.8 07 327.28 2 139. 21 23420. 98 9861.7 68 1900 27 72 4468.3 08 6.34 2 3902.9 66 37.321 897.92 411.6 04 4718.4 74 66.87 623 14 73 20968. 49 3299. 3 12027 12413 7668.39 23498. 8 86339. 92 109838.7 410 270 74 62040. 86 603.2 87 987. 7 17812. 3 9310.37 4 9196. 3 8094. 84 2178. 31 1666 183 7 13291 32208 18917 62883 7077 1034 37671 48016 931 27287 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 100

Group 2 1 year before bankruptcy Book Value of Equity DMU Curren t Assets 1 Curren t Liabiliti es 1 Workin g Capital 1 Retain ed Earning s 1 Operati ng Income 1 1 Total Assets 1 Total Liabiliti es 1 Employe es 1 Sharehold ers 1 1 64200 70100 900 38200 600 108800 300 246700 8688 72 1108.9 13 101.4 238.2 6011.1 9 1112.07 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 7123.2 6 23 1000 2 7.49 2 17913 120792 28607 144000 16374 30774 3 0 921414 28606 8 41227 0 7 7 7800 4013 4 48008 13099 34909 3377 67289 1869 241383 814 84 121 126880 74870 77240 200370 716400 0 2400 327100 219600 0 0 0 89100 277 6 70146 29112 41034 36681 1790 6142 96029 30887 2100 32 7 0.164 138. 138.33 6 10183. 7 344.272 137.07 6 1.424 138. 0 0 8 132.8 38 4134.4 87 2808.6 43719. 2 627.007 1936.3 76 9111.1 8 7174.7 82 127 89 9 394321 23294 161376 393213 70679 436263 784796 34833 29000 2383 10 284717 9 12919 9 18 0 19731 2937 23648 3 387133 3 1148 0 344 8827 11 7779. 94 2724.2 36 0.3 8 128.8 6 40.269 471. 29 104.3 9 793.8 6 42 703 12 396834 187844 208990 8397 64704 149036 8048 431449 6100 33 13 91.991 124.04 7 32.06 9.716 84.226 4.284 128.331 124.04 7 0 0 14 12018 4 110748 4 94370 37288 0193 797297 330980 3 2120 6 83000 6281 1 107196 344 72642 9042 1662 2841 639 394 137 144 16 372.92 8 224.68 2 148.24 6 742.93 6 211.169 16.93 3 3.323 369.39 36 17 49673 37166 1207 96442 32697 3917 7908 39901 3346 4 18 41701 3 29288 124642 8 3183 6 140226 49488 11377 4 88086 9 94000 1688 19 6628 9082 244 29407 3190 78 614 60976 2028 1002 20 101.0 68 206.02 7 809.04 1 2489.9 86 191.8 249.7 18 2836.26 9 340. 1 178 30 21 266 3949 12939 2129 14362 12421 20927 84106 7300 200 101

22 9. 68 31266. 79 24328. 89 6481.7 330.7 2 084. 62 8181.4 1 74044. 93 1463 9 23 36272 17707 186 10888 1339 44640 6341 20701 2300 20 24 187346 94910 92436 131698 1934 16283 21793 910 1193 118 2 3413.2 98 2947.2 98 466 3222 626.899 2130.7 2 688.8 7 4728.1 37 66 32 26 39880 41912 2032 1403 2204 48773 9833 4980 0 0 263.7 964.3 376.0 796.17 4191.0 27 9 9 829.06 79 8 99 46 129 617.08 9 322.8 79 2293.2 2376.7 22044. 2168.81 674.64 13192.1 1217. 28 7 04 83.434 08 2 7 7 2 2 383 114978 229027 114049 29 92794 269273 68321 381441 173494 3 6 3 9000 2468 30 8384 79617 21233 740 2386 4120 97339 93219 3900 400 3182. 3182. 3176. 3182. 31 0.789 91 1 11989 27786 9 6.027 91 1 400 32 8038 3974 4284 6781 12387 4914 97732 39840 326 16 33 14682. 82 18364. 21 3681.3 9 209.9 2 18632. 4 9.9 3989.73 36.4 6 1680.1 21279.9 9 2484. 4 939 10 34 10091. 8 1301. 71 6446.24 2 13660 1340. 12 238 818 3 63377 41027 2230 4680 68 74198 12046 0848 300 1343 36 332220 131788 200432 104 2779 132746 67768 44912 8100 4781 37 808 470 3308 20 39 061 1033 292 9 2202 38 23282 119373 113479 67807 3386 17839 313436 29062 612 2630 39 623.4 6 6407. 4 14.08 18179. 3 1764.27 4 61.98 9473.61 887.6 3 327 1009 40 1216 2946 14240 493 7139 8067 110873 118940 713 236 41 41280 17966 239314 264678 123181 440663 670710 230047 13600 3000 42 11403. 66 9828.3 7 17.3 01 200.7 6 966.471 8161. 2 2473.7 1674. 19 110 244 20986. 7436.3 130. 27333. 13779. 32283.7 860.6 43 7 1 39 9 7798.67 69 8 84 494 437 44 24309 43223 18914 6186 3267 2623 108000 10377 8167 26 2243.6 1792.6 41.0 6207.4 198.7 4079.93 2094.1 4 9 36 9 8 1102.12 48 2 84 126 7 46 17 1160 41 944 3491 800 009 48 0 102 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

10333 47 80917 14268 7331 16490 24628 88898 9361 18429 449 489 48 4764.4 68 363.1 8 1129.2 88 1167.4 7 1166.39 3 6326.8 96 16174.3 8 9847.4 82 172 80 49 61171 10247 41304 3240 238967 1114 124170 10906 266 684 0 2638.1 63 131.9 79 2493.8 2 19320. 4 7162.11 1464.9 3828.27 2 293.2 17 26 116 186487 1 163118 202799 7 4477 109982 2238 199112 8 22431 3 119 126 1887. 163.8 3276.2 16216. 374.89 3066.0 2379.1 686.6 2 43 22 8 9 1 87 9 8 20 2428 3486. 7218.8 3732.3 7236. 2886.1 4691.19 777.3 3 24 93 7 1.936 2 6 17 60 128 4 32326 263 773 8277 19078 19617 4830 68147 860 0 8864.1 19086. 10222. 227. 62662. 1278. 63123. 42 66 91 6902.78 4 6 1 1087 280 6 3976 10449 6473 4063 2962 2490 799 10449 4063 2490 7 9030 16091 7061 3196 480 10373 33944 2371 70 210 8 2377 8607 6230 12416 102 6230 2377 8607 0 477 9 123886 9 167497 8 436109 236679 22601 214603 18898 1 167497 8 11947 6628 60 17940 9 108212 3 497286 20939 1790 70189 16076 3 11676 4 164 3000 61 44 12226 7771 493 347 20038 49000 7249 3200 0 62 8917.1 82 21269. 14 1232 9498.7 3 2092.3 2172.2 37 4387.2 4321. 01 3399 470 63 70789 23786 47003 7727 326 29632 101783 7211 1613 1990 64 2379. 3 28241. 64 4662.1 1 908.3 6 2614.92 3 16333. 66 098.8 8 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 3462. 22 493 142 6 79240 60690 180 133083 1797 81984 103024 18008 880 717 66 7381 1789 10478 3709 10368 22760 63224 40464 980 3000 67 68 43.78 8 101.9 2 381.87 6 3.912 1330.7 1228.8 76 2 69.60 7 947.4 729.402 871.32 3 98.137 22.018 1742.69 7 126.18 7 871.37 4 0 69 104.1 69 3 220 103

69 69709 3339 16370 23814 4232 1382 106213 92631 102 63 70 290700 0 23100 0 6000 16100 0 426000 184900 0 800300 0 61400 0 69000 30207 71 13743 3323 10420 6049 10384 12421 22970 1049 106 43 72 1416 8869 68647 093 1376 20077 486810 28870 3700 43 73 174369 71001 103368 33181 2176 19893 336227 137292 347 1222 74 7480 4789 29061 1331 2406 31017 76806 4789 103 318 7 33700 7900 4800 478300 37200 184600 447100 631700 7400 0 76 776 37811 300 7673 4977 662 4490 38928 80 0 77 3628.4 6 471.0 41 1842. 8 24781. 8 906.327 28.94 1 11636.1 4 11377. 19 108 3280 78 329210 340181 10971 20049 28282 32033 381281 349248 1600 9 79 46.46 8 3120.9 78 80 44678 19826 81 82 1792. 14 31.1 26 47792. 49 907.1 8 2664. 1 11148 32000. 3 272.0 22766. 2 2978.64 963.97 1 3194.32 3 418.2 94 132 649 14194 19031 3798 160028 19826 274 32 3667. 691.72 2 7 24414 1711.72 3140. 7 170.7 8037.8 1394.0 3 4769. 27 4 403 907.1 8 929 330 109600 64023 602 100381 10983 2277 171 83 9111 10941 96830 2262.1 3442.4 1180.2 1726.0 1022. 18039.4 717.0 84 61 31 7 8 1680.46 41 3 2 40 10 8 11721 142641 9080 9197 100693 30717 760399 43242 0 0 86 8011 267987 312128 11068 228 39843 66630 267987 39 182 87 13113 117849 13664 76078 730 6979 199163 12984 4000 1118 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 104

2 years before bankruptcy Book Value of Equity DMU Curren t Assets 2 Curren t Liabiliti es 2 Workin g Capital 2 Retaine d Earning s 2 Operati ng Income 2 2 Total Assets 2 Total Liabiliti es 2 Employe es 2 Sharehold ers 2 1 9800 94700 1100 29100 38700 109800 487300 37700 108 71 2 121.0 7 1132. 6 1011.0 1820. 4 30.337 6499.0 3 1103.0 32 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 7602.0 8 27 1046 3 136112 9 97496 386173 229330 1 863880 96721 21316 9 310242 0 9000 3723 4 3728 10939 2689 1882 47147 8314 19062 10707 614 210 79900 112320 0 323700 86800 7700 600840 0 84890 0 2400 0 9800 273 6 1033 24379 2664 292 14313 4292 7072 2460 190 400 7 1.037 666.69 8 66.66 1 9847.3 231.923 66.66 1 1.037 666.69 8 0 0 8 72.0 4 6887. 0 613.4 32913 1693.44 8670.2 4 1217.2 66 9887. 0 64 83 9 36140 169266 192274 34814 30989 38372 684649 301077 12700 2642 10 294261 14061 9 1364 6 1408 9 2878 214217 4 3934 8 181317 4 9194 8308 11 813.1 76 6166.8 27 1968.3 49 300.72 8.706 4469.4 32 11400. 36 6930.9 29 40 698 12 399700 197164 20236 768 9207 149207 8483 43646 600 9 13 2137.3 8 64.02 2 173.3 63 21318. 8 97.626 1648.8 62 2212.8 84 64.02 2 0 0 14 122288 3 11247 8 9830 411861 104830 846192 3332 248933 3 80900 6890 1 10487 27113 77744 1047 2678 24332 644 32113 174 1629 67.38 16 346.62 9 166.20 3 180.42 6 1 1.48 237.11 9 712.26 3 47.14 4 0 0 17 6860 73276 7416 112324 2767 11209 9700 11236 3100 0 37094 2389 13238 276196 117234 46947 901232 43168 18 1 4 7 4 7 1 3 2 86000 1446 19 6317 104 4228 27864 3739 930 62188 6128 203 1020 791.76 180.76 611.00 2382.8 288. 2702.4 313.86 20 4 1 6 144.1 88 3 161 400 10

21 23670 31264 794 7146 30840 12411 202369 7824 800 2700 22 49490. 1 21431. 79 2808. 3 190.4 32 4473.60 6 1843. 43 72493. 69 660. 26 1600 90 23 40649 13837 26812 19376 22 340 70848 17398 2700 20 24 138946 63019 7927 13293 4472 112708 176012 63304 791 9 2 6092.9 11 4973.1 74 1119.7 37 147.7 8 37.966 211.6 39 1122. 6310.8 6 69 34 26 46974 169 468 38 10 47370 109307 61937 919 410 1831.4 3671.4 9766.0 14100. 4334.4 27 3 3997.0 11 46 92 121 2241.3 23 4072.7 7 26.4 1307.6 1248.8 2097. 2331.3 1687.9 12994. 11306. 28 96 82 14 8 9 69 77 8 2 306 106842 221021 114179 29 817738 27009 60729 31638 130687 3 8 9300 23 30 66627 76114 9487 7844 131 1930 109802 90497 4200 400 24966. 24917. 8469. 24917. 24966. 31 49.393 46 1 7 1236.14 1 0 46 1 600 32 943 20264 3679 666 404 39271 6396 20322 194 472 4662.3 33 18984. 16 23646. 48 1 14474 4767.66 93.14 4 27771. 8 27178. 71 1140 10 1723. 12407. 3316.0 389.08 8193.4 24444. 1620. 34 7 27 3 69.197 48 2 8 231 824 3 49327 22796 2631 43332 130 7001 10068 30184 3000 1400 36 37313 147398 22737 3003 18830 13297 728391 7094 10000 482 37 710 3643 307 271 2317 4801 964 4763 9 2167 38 284 14481 113973 19012 341 6698 397904 330411 6402 2966 39 644.9 46 9362.7 63 2907.8 2 19968. 88 4248.82 1281.1 2 9723.4 2 11004. 7 378 99 40 19440 38994 194 8680 1137 24266 13369 109429 0 0 41 381943 129832 22111 20147 13468 403894 63991 20097 12600 000 42 43 14847. 98 9928.6 13390. 79 78.6 47 44 6738 32697 4 2393.4 4 190.0 34 4919.3 84 832.1 43 261219 488.41 1 4878. 16007. 7 9 202.402 1108.06 9 386811 34720 4727.9 2 8.31 11071. 33 2737. 72 16466. 39 120 24 064.6 29 2110. 39 1608. 76 486 408 16264 173882 33636 994 117 2410.0 89 448.7 42 2138.6 3 210 20 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 106

46 1263 121 28 100727 1264 3837 8622 478 47 6900 47 13800 16161 23601 867 27823 24040 16724 180764 666 487 48 479.4 6 2399.8 83 3079. 73 2287. 92 93.313 6849.8 6 16384. 42 934. 7 14 71 49 8023 16129 64124 72931 8309 7228 99880 27622 0 0 0 440.18 3 3923.8 3483.6 2 1006. 1907.99 213.4 2 1824.7 86 3960.2 0 1 129 1 97242 164390 67148 79230 2404 114160 208020 9 196604 9 0 0 2 3126.6 3 281.2 2 214.6 12237. 1378.01 8 1312.1 43 4029.1 42 2716.9 99 41 1614 3 301.1 02 7634.7 44 4133.6 4 6102.8 1 03.00 1847.3 8 633.6 4 8201.0 24 66 124 4 66 0483 6082 73 43302 20 7898 73438 1024 0 112. 9 14330. 01 822.8 8 16976. 19 232.24 4 74319. 33 13726. 63207. 16 97 280 6 734 6310 1224 40862 7077 1711 13711 12000 40862 1711 7 93 1829 6276 26109 286 12299 3139 22840 726 208 8 240 8292 842 12028 106 842 240 8292 0 02 9 198221 4 947904 103431 0 371901 27020 827482 28301 9 173 7 11000 7200 60 14637 7 844692 61168 19477 6404 62899 14704 7 140102 8 149 3000 61 3892 8443 41 3199 3722 19363 4419 67244 3400 77 62 1181. 06 46616. 97 3476. 9 8877.2 9 686.187 201.3 46 63080. 61 6079. 27 398 800 63 6687 3624 30612 9030 2990 30117 96443 66326 1637 192 64 8989.8 13 976.3 78 86.6 838.2 1149.36 8 073.8 42 1726. 83 1062. 99 33 2200 6 6776 6303 423 121030 814 69991 9311 16302 11300 727 66 10490 7221 3269 2020 4706 32749 49263 1614 980 979 67 799.2 8 218.46 81.06 3 7.16 164.414 168.0 94 2269.3 06 701.21 2 0 0 167.62 874.37 8672.3 61.37 1879.3 1263.9 68 1042 8 322.4 2 63 91 30 200 69 9444 42440 17004 22688 861 2118 78110 7992 107 82 287300 283800 47700 240800 83300 9400 70 0 0 3000 0 20000 0 0 0 76000 31100 71 14649 216 12493 2314 7266 13103 23346 10243 826 48 107 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

72 12177 11134 40643 819 20736 1862 494034 307986 4300 400 73 1040 68616 81789 49230 1038 213667 29499 80932 337 1067 74 6397 41112 22863 710 124 24771 6883 41112 91 20 7 41300 7600 3200 387600 2400 90000 34300 624300 9000 0 76 109 6740 1631 0 49 1933 9003 7070 0 0 77 10347. 37 9286.4 09 1060.9 64 106. 6 82.88 6822.0 81 26483. 48 19661. 4 0 0 78 392878 34266 0213 9490 620 70662 434078 363416 0 80 79 980.63 7 334.1 84 2373. 19720. 7 3330.42 37.17 691.1 2 6628.2 9 336 619 80 6494 210303 13809 11804 942 309 207208 210303 6671 341 81 20344. 79 8903.8 83 11440. 91 3462. 7 6806.20 1 30824. 61 71772. 37 40374. 48 660 431 82 472. 1 7272. 2699.9 1189.9 4 2201.3 9900.9 8 17173. 46 7272. 779 330 83 14427 31178 1671 27737 21170 7626 16680 89424 2277 176 2178.6 6 84 2911.6 42 090.3 06 2961. 26 1088.63 8 11782. 49 21467. 26 9684.7 74 840 10 8 14062 1213 19270 100028 7009 311220 6973 3481 0 0 86 33004 270477 26227 80390 424 348789 619344 270 4133 268 87 102642 36172 66470 709 1162 98697 143726 4029 2800 28 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 108

3 years before bankruptcy Book Value of Equity DMU Curren t Assets 3 Curren t Liabiliti es 3 Workin g Capital 3 Retaine d Earning s 3 Operati ng Income 3 3 Total Assets 3 Total Liabiliti es 3 Employe es 3 Sharehold ers 3 1 9209 9619 36410 18461 870 223721 23214 299493 11991 14 2 201.2 2 882.64 9 681.39 7 18043. 4 68.9 6203.3 1316.2 98 719.6 46 33 77 3 101217 8 73893 273243 882028 607 266278 2471 1 22027 3 7600 3981 4 31260 10231 21029 9786 32018 72708 137868 6160 400 26 712600 113140 0 418800 10900 121700 61200 0 84080 0 24180 0 89700 169 6 39862 1966 20197 1661 333 36306 713 20829 177 600 7 12.819 427.12 2 414.30 3 9420 0.699 344.9 9 82.23 427.12 2 0 0 8 391.68 9 078.0 11 4686.3 2 30688. 6 3867.22 63.8 8 1222.1 33 778.0 11 4 69 9 30881 142673 163208 368289 29422 416810 709738 292928 12600 2839 10 23941 96380 143071 0 126747 6 23023 19013 0 34426 6 14013 6 4430 8296 11 926.7 77 7471.7 91 204.9 86 23.838 497.944 4991.6 14 13413. 24 8421.6 29 3 67 12 4186 23312 180274 8023 76286 134340 606389 472049 8000 62 13 206.03 103.29 9 102.73 6 2097. 2 10.73 1992.3 92 209.6 91 103.29 9 0 0 14 122403 7 113406 3 89974 413034 164391 83727 314174 1 230448 4 83400 7419 1 8183 20116 61737 12163 9089 866 33781 2116 111 111 498.00 17.0 200.78 26.0 364.2 16 3 6 64.723 9 348.063 9 1 40.03 0 0 17 7621 62466 131 11122 3020 111293 110640 139246 0 0 2868 17634 200638 3139 63446 320869 18 3 4 820339 4 83317 2 1 9 66000 13499 19 3243 42 2182 2718 76 1248 24011 22763 1217 1086 701.89 281.2 420.64 2442.2 2247.9 2668.4 420.49 20 4 1 3 29 21.7 61 6 9 63 400 21 22404 33282 10878 1930 8229 147629 29133 11104 00 126 22 40077. 16186. 23891. 3281.7 6104.80 1698. 240. 381. 130 96 109 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

67 08 6 73 4 46 07 61 23 48267 1886 29681 21273 2447 802 79337 233 2600 300 24 10707 40426 66649 19213 11672 8936 13030 41174 49 144 2 4999.3 79 328.0 6 1471.3 23 29.29 16.731 01.0 98 8783.2 41 3768.1 43 141 30 26 37229 32914 431 4203 2987 99904 160160 6026 1023 421 27 660.7 43 1308.2 3 297.4 9 237.38 9 108.67 1 13674. 83 1173. 22 1498.3 88 23 104 28 2794.6 62 6346.3 41 31.6 8 2107. 28 1284. 2 9234.6 4 14048. 86 23283. 1 32 39 29 7234 322243 430102 27114 13630 102838 0 216797 3 11399 3 11006 2822 30 7428 694 834 276 839 38881 106634 6773 4200 400 31 49.26 1109. 91 11460. 6 69180. 8 494.38 1141. 3 94.6 1109. 91 1 290 32 1336 1028 3071 44093 1367 980 21714 1040 163 281 33 20609. 48 1699. 96 4909. 16 4672. 1 60.293 10411. 8 34399. 7 23987. 9 1340 190 34 881.8 69 2361.6 6 320.2 09 133.00 3 368.12 6038.9 71 10496. 87 447.8 99 10 832 3 4743 20347 34396 34746 14766 66004 91298 2294 3100 100 36 378272 182088 196184 22367 8998 161433 74834 86921 11000 0 37 8996 6202 2794 2836 130 226 10337 8072 23 2191 38 2726 16893 109363 10177 1011 8207 41037 32794 6643 2904 39 7644.0 18 9238.9 77 194.9 6 14334. 6 266.31 3627.3 09 1887. 81 12260. 31 7 40 19487 38702 1921 648 370 4321 1878 11327 900 20 41 402381 12190 280791 1634 121734 392731 63410 260679 1271 33000 42 43 432.0 26 9034.3 12 1399.1 34 481.8 31 3132.8 92 442.4 81 44 63040 730 1226 128. 1 10104. 692.88 9169. 16 11318. 6 2149.1 34 0 0 9 1240. 202. 07 14139. 89 11637. 38 324 344 32334 70960 99377 22012 319889 11217 1088 373. 6 4 207.0 4 2173.6 333.39 714.067 247.1 8 10.0 69 2402.8 84 19 40 46 2747 2196 1 99161 2024 31 1079 228 4 3600 47 9642 96629 204 2248 1979 29417 126046 96629 130 278 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 110

48 49 0 7106.4 18 2476. 07 312.88 4 1281.6 02 228.3 3 2103.8 47 824.8 16 23217. 74 1790.9 6 2106.8 3 366.14 16046. 988.76 0 9 6773.6 4 2960.2 62.8 4 243. 32 1090. 7 1 42066 11680 74784 9320 24091 2023 113.9 761.0 237.0 10634. 1193.8 2 7 06 3 84 192.94 17 14074. 64 748.8 42 84 28212. 279.0 36 4 186 0 1204.6 04 166200 4 6467.1 46 229.1 78 9 146 14946 9 12000 141 273.3 29 32 1600 3 364.6 21 783.6 86 3929.0 7 38. 70.83 128.1 2 6846.6 68 8374.7 89 0 0 4 7804 84228 8424 9721 4402 38267 12484 86317 13 0 17679. 31 1189. 7 819. 64 1717. 23 6080.49 9 74402. 04 114149.3 39747. 22 997 278 6 823 399 428 32330 1008 10243 19928 968 32330 10243 7 10968 9964 1004 2110 1911 17349 4074 2340 81 216 8 219 7977 48 11644 64 48 219 7977 0 02 166414 224204 11690 9 4 88919 8022 26690 203838 72140 0 0 12100 800 113996 11097 109108 60 9 73626 403704 14716 7398 3360 7 7 141 3016 61 1630 6323 10207 764 410 3237 40969 77611 2940 4 12219. 1342. 39122. 4982.2 11.7 67446. 689. 62 3 14 8 1861.09 0 98 27 4000 800 63 71200 34477 36723 9244 918 29839 101078 71239 180 1200 64 7220.4 6 11349. 48 4129.0 1 8900.2 9 6440.32 1276.6 07 13672. 92 12396. 31 3 2200 6 2826 82998 30172 10860 64989 3939 96116 1347 0 0 66 24231 4177 2004 2183 2702 24880 2907 4177 120 3300 686.34 418.87 121.84 811.70 1286.3 474.6 67 8 267.47 8 8 180.884 2 7 0 0 34.99 392.99 1108.6 2487.4 1378.7 68 3 9 47.006 8162.1 34.784 47 92 33 204 69 62178 47393 1478 2404 1941 70 101020 100270 1600 79 29700 249100 447800 23600 789900 4300 70 0 0 106000 0 13000 0 0 0 70000 31200 71 1630 2623 13007 80 297 1331 24100 10749 716 2 72 41188 20128 21060 416 629 8616 107248 21083 1600 110 111 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

73 121493 70678 081 4424 19602 12181 231471 109620 3000 46 74 46043 31041 1002 803 3270 16030 47071 31041 80 201 7 44100 138400 94300 29300 4400 24900 68200 43300 10000 0 76 407.09 1 18218. 71 17811. 6 3081. 9 2623.46 410.7 3 1287. 97 18218. 71 203 33 77 9937.7 36 6407.9 91 329.7 4 13023. 6 0 88.1 4 21471. 08 1261. 93 98 2600 78 393672 3806 3607 14094 1973 6887 43072 364838 100 1 79 3390.8 68 126.8 8 2134.0 1 77. 4 1074.8 3723. 8 10097. 39 47621. 24 387 33 80 113071 169132 6061 7674 1778 109299 278431 169132 9097 313 81 19434. 66 1146. 1 7888. 09 4676. 4318.89 7 27399. 82 6228. 2 3460. 31 76 431 82 347.9 7146.0 84 3688.1 3 323.0 82 3697.26 12716. 72 19862. 81 7146.0 84 1188 3 83 16827 1663 174 20 423 103681 187236 83 2640 180 3788.1 9 84 39.6 6 9183.8 46 2918.2 2 162.13 11664. 17 2921. 74 1427. 7 4700 10 8 184093 91029 93064 7873 77413 277273 44808 2673 0 0 86 494828 223298 27130 117071 33093 384168 607842 223674 3481 272 87 390 28048 242 12613 3942 48291 82141 3380 1700 163 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 112

4 years before bankruptcy Book Value of Equity DMU Curren t Assets 4 Curren t Liabiliti es 4 Workin g Capital 4 Retaine d Earning s 4 Operati ng Income 4 4 Total Assets 4 Total Liabiliti es 4 Employe es 4 Sharehold ers 4 1 8839 88092 2923 18420 8192 12967 39449 409882 1484 184 2 243.89 99.1 1 71.61 6 17066. 6 117.339 693.6 7 1832.8 76 726. 44 67 66 3 42424 1617 262679 162060 111960 13874 648460 0971 2100 2304 4 20901. 17 7947.2 21 1293. 9 2860.7 96 1692. 9 64369. 0 98421. 11 3402. 07 300 26 631700 846200 21400 6980 0 39200 63790 0 827480 0 263690 0 80700 0 6 36828 20663 1616 1333 0 33008 01 22043 1800 600 7 0.31 294.2 293.89 9 9420 267.40 293.89 9 0.31 294.2 0 0 8 60.83 4 1981.2 2 137.3 9 26290. 8 827.87 2188.9 8 2292.2 44 4481.2 2 4 69 9 366226 142082 224144 3489 9920 42103 710397 289362 12100 3000 10 214634 9 90826 12402 3 11383 3 168244 173003 9 323170 1 10166 2 4662 7004 10662. 7300.6 3361.6 227.84 3776.8 1246. 8688. 11 3 32 7 6 43.28 34 36 21 4 60 12 37017 21288 1787 3102 3847 108800 7042 448242 8700 62 1483.2 268.36 1214.8 190. 3417. 378.9 368.36 13 36 9 67 1 1004.18 19 9 0 0 121722 2992 206862 14 7 9130 262097 4910 129 926632 3 1 80000 8029 1 6606 28397 28209 282 221 1801 46448 28397 121 1768 41.20 242.41 208.78 97.03 39.92 16 1 9 2 2.891 71. 7.109 1 2 12 8 17 80287 6281 17706 108088 13070 10789 113382 139332 0 0 210960 144932 1613 260060 21927 261866 18 2 0 660282 3 624114 9 7 8 60000 11334 19 2974 020 2046 2610 13 226 20110 1784 1121 1112 772.48 331.91 440.6 2288.6 2294.3 2907.7 613.39 20 7 8 2 24.463 84 76 2 0 0 21 19741 32317 1276 102 8231 146724 26177 11483 6000 109 113 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

22 41707. 2 17743. 46 23964. 06 2389.1 2 2720.83 9 1606. 81 020. 28 3894. 47 2263 102 23 0283 16788 3349 1990 390 416 78344 24188 2277 31 24 192 26208 666 137266 37227 9004 41630 30304 389 18 2 2939. 32 290.39 2649.1 42 146.73 2 39.418 3414. 49 370.4 39 290.89 0 0 26 3374 32762 983 47749 8798 88621 123068 34447 1093 421 27 16.36 884.6 2 368.20 2 2241.0 9 1764.1 2 66.86 178.4 7 2294.6 98 3873.1 63 362 17 28 1638.2 71 3879.3 61 10793. 4 2469.44 107.7 3 19142. 6 18066. 83 1270 379 190382 29 603298 23349 369749 22937 16121 963471 9 94038 1160 2673 30 7064 40921 16143 32922 1044 44601 99 134 4300 400 31 8.3 862.1 48 866.8 6882. 3 620.962 800.8 4 124.30 9 862.1 48 1 290 32 19202 12762 6440 29883 21447 12349 30401 13122 0 0 2210. 17 17809. 4700.6 69 4238.2 8 6496.03 9641.4 37 37388. 64 27747. 21 1140 160 33 34 3 3747 16000 2147 2018 162 36911 7319 20408 200 1600 36 48244 224040 2840 0 3630 183800 911341 72741 12782 0 37 440 031 409 4060 6 762 616 403 216 2203 38 23419 116734 13668 88 763 82439 3494 266481 479 2696 39 7673.2 14 9484.4 9 1811.2 8 1277. 41 1.738 3261.0 2 1027. 43 11766. 38 299 764 40 27083 110989 83906 41678 10004 79281 214170 134889 1100 261 41 28478 110386 17092 118721 90220 323048 70268 247220 112 18000 42 327.4 3 1306.2 02 978.74 9 170.4 9 862.3 9 800.4 809.912 7791.6 33 14447. 84 666.2 02 0 0 678.1 8148.6 1003.0 9378.6 837. 43 71 7 8 2926.27 36 18 82 30 260 44 67613 96028 2841 281762 4628 760 286633 344238 17813 1032 2294.1 2232.7 3008.3 263.6 136.7 2483.0 4 63 6 61.403 7 101.7 68 04 36 19 41 46 900 471 118 9683 41372 776 1107 731 107 3603 114 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

47 107074 10628 789 26484 2969 34062 140347 10628 194 97 48 3078.8 3 834.36 2244.4 88 197.6 33 2129.1 8 6229. 2 992.8 8 3363.3 33 42 63 49 4349.0 89 290.2 92 1443.7 97 607.7 23 4871.0 1663.1 69 461.0 33 2987.8 64 0 0 0 444.48 2 897.93 4 43.4 2 314.7 4 1933.33 1 182.14 1 1103.8 44 921.70 3 0 0 1 34149 3027 18878 82868 67870 3067 66060 3407 760 410 2722.9 6 2 4094.3 29 6817.2 87 9129.4 01 1186.47 2 113.36 6 6807. 89 6694.2 23 28 1239 3144.4 220.6 623.71 3209. 162.8 784.7 6228.9 3 09 93 6 1 1880.13 42 74 32 0 0 4 8044 7134 9091 2474 128 438 11209 7134 1200 0 2974. 1070. 1224. 14686. 7222. 39481. 99 6 3 1 1734.3 03 112004 98 941 272 6 14173 6097 8076 28411 19 14040 2677 12717 28411 14040 7 9479 6303 3176 1821 106 1899 3214 13186 616 27 8 23 7683 130 11316 0 130 23 7683 0 02 9 1249 4 66321 92243 1817 8843 637267 18726 0 121999 3 11200 11400 60 94419 60712 293447 9989 6311 66042 9610 88378 141 3046 61 1264 4671 3407 470 914 9742 16783 27439 0 0 62 09.20 4 2847.7 4 2338. 4 2020. 2 4.78 1414.7 1 222.0 33 3666.7 39 0 0 63 78333 33670 44663 446 23118 3373 112104 7831 1784 1800 167.3 9 64 6228.7 0 3066. 89 3162.1 16 241.389 326. 07 864.4 1 127.9 08 4 2200 6 66 18416 489 12927 22149 7733 24312 30110 798 0 0 67 421.6 279.04 1 142.61 4 296.39 7 299.431 200.60 3 499.80 3 299.2 0 0 68 311.8 4 383.90 3 72.319 7602.1 6 1937.3 1603.1 67 2942.0 8 1338.9 13 26 196 69 7182 39740 32112 4664 326 20099 1132 9343 1900 36 70 290400 0 23200 0 79000 461000 0 844000 3300 0 796300 0 442800 0 68000 31700 71 1394 270 11204 1676 4800 119 2248 10989 60 62 72 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 11

73 10371 6304 42326 3271 23920 10970 174020 6800 2400 00 74 31248 18434 12814 484 1112 12814 31248 18434 63 208 7 61200 113200 2000 230300 600 2000 06000 08000 10000 0 76 973.91 8 11392. 9 10418. 7 2463. 1 6000.71 06.70 4 14700. 98 13994. 28 267 33 77 8002.4 84 6997.3 88 100.0 96 329.3 82 0 1719. 84 2682. 74 9332.9 0 0 0 78 208717 226027 17310 966 10917 101 244298 233783 103 19 79 4879.8 91 41438. 37 368. 41119. 3 30000.7 2147. 9 21293. 79 42769. 74 70 33 80 60391 4261 17776 6806 804 90992 133607 4261 3333 121 81 993.4 61 180.3 24 4773.1 37 839.6 1 172.9 7 1043. 16 27439. 6 1238. 329 406 82 3117.3 91 6129.1 19 3011.7 3 736. 12 291.118 16936. 1 2306. 63 6129.1 19 1170 367 83 231 8871 620 416 803 43 161 1972 361 12 397.9 84 2802.9 06 6778.8 8 4344.7 19 210.34 3 13143. 18 242. 3 12399. 3 1400 10 8 112699 77272 3427 13076 44962 2046 398484 193019 0 0 86 34140 7744 269396 91808 7396 33993 416137 76184 3100 14 87 371 17070 2008 21612 312 39219 92 20333 0 0 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 116

years before bankruptcy Book Value of Equity DMU Curren t Assets Curren t Liabiliti es Workin g Capital Retaine d Earning s Operati ng Income Total Assets Total Liabiliti es Employe es Sharehold ers 1 62727 94970 32243 4884 8170 100799 61889 461090 0 0 2 2.69 141. 83 14926. 1 882.4 9 264.99 31.3 13 2710. 37 2039. 06 149 66 3 13733 43818 9317 3361 29209 28486 149006 12020 614 304 4 13944. 23 4632.6 49 9311. 81 2099.4 91 13328.7 6 329. 13 7238. 39 39979. 26 20 2 627900 713100 8200 76660 0 214600 761760 0 873100 0 111340 0 0 0 6 38364 23487 14877 1220 0 3190 6962 207 1700 600 7 83.148 49.7 7 376.60 9 867.3 7 19634. 1 1297.88 02.34 3 962.1 49.7 7 6 446 8 1341.8 64 927.73 2 414.13 2 7367.28 2416.6 63 3844.3 9 1427.7 32 93 80 9 4061 206104 19947 36192 111988 419029 681746 262717 14200 331 216313 132648 104801 16148 308117 146631 10 3 83661 2 4 220024 6 3 7 41742 8190 68.7 2329.7 3239.0 13.14 2864.7 67.0 3890.3 11 8 29 29 8 631.1 38 69 31 2 62 12 3749 210122 147337 27439 6164 99811 43992 444181 9300 62 13 4208.2 72 1770.7 89 2437.4 83 1739. 8 31.676 7177.8 81 21344. 14166. 62 2 973 14 123137 9 101600 21374 447768 169303 890072 300267 2 211260 0 84000 8808 1 2703 18880 33823 3427 8160 2026 39406 18880 96 1903 294.49 226.74 10.09 16 21.24 98.21 131.869 11.749 21.84 1 0 0 17 18146 13483 12488 221747 44349 221747 18 6 1 0293 8 70962 6 4 8 3492 11460 19 3807 4691 884 282 290 3300 19312 16012 122 123 20 21 20137 083 34946 18896 23286 136371 207881 7110 4300 70 39026. 16781. 22244. 3489.7 1996.17 17177. 2923. 374. 22 02 1 1 96 72 9 87 782 111 117 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms

23 47632 16380 3122 17004 3911 068 77964 27306 2308 313 24 18301 1996 16 98648 77 73267 2488 2392 0 0 2 960.77 1 9.83 4.941 110.11 2 144.622 131.63 2 1087.9 62 96.33 0 0 26 29840 23830 6010 36780 18096 72677 10084 32407 1288 39 120.27 27 28 27.63 1 1409.3 09 377.90 2 4919.3 8 310.0 7 1 20.182 646.831 8643.8 79 8631.70 9 18.96 8 32.44 8 1820 871.99 8 713.03 0 0 1830. 44 0 0 128962 29 471964 128110 34384 183887 9729 774196 9 1433 740 2184 30 387 44741 10646 2848 123 37417 92827 410 3900 400 31 32 69.87 4 13946. 21 13286. 3 404. 7404.94 13077. 9 868.32 2 13946. 21 1 290 33 1808. 42 12873. 14 212.2 83 210.26 3 89.909 14089. 98 2928. 49 1438. 1000 61 34 3 39083 14602 24481 16199 4673 22904 780 34946 2200 108 36 4097 31993 131004 238997 73689 289731 129311 8 100338 7 16700 1343 37 1037 600 387 374 242 1218 1140 10232 0 0 38 276307 111497 164810 24 947 89769 36260 272301 40 2777 39 832.1 73 8330.1 77 201.99 6 10923. 6 292.98 4416.8 12 17099. 97 12683. 16 329 764 40 32247 3161 20914 41804 10474 79402 232939 1337 13200 279 41 34713 100700 246813 81633 72700 227923 31096 303173 9399 22000 42 141.80 8 04.11 362.30 2 2.447 2.448 793. 78 1407. 69 6104.1 1 14 23 43 6611.4 08 664.8 19 46.89 446.6 374.72 172.8 47 9104.3 3 7290.8 26 260 4 44 3790 12860 92770 24271 38074 21137 348186 369323 0 997 4 3371.3 17 2091.9 86 1279.3 31 1948.6 8 484.03 3713.3 6 981.7 88 2268.4 32 230 29 46 9414 267 4147 4394 4932 4972 216 62 137 3989 47 160283 13984 24299 6869 11374 68663 206189 13726 190 643 48 49 448. 38 342.33 6 4116.2 02 3339.4 44 108.43 7383.7 3 897. 42 191.8 07 28 4 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 118

0 1 24106 32213 8107 6221 1247 283721 37819 94798 4227 30 2 2972.0 78 4683.7 04 1711.6 3 782.8 71.303 616.90 2 42.7 6 483.8 8 26 0 3 4 9939 18784 771 68111 2672 34864 110732 1496 1200 0 478. 32 29179. 66 1660. 66 16997. 16 6007.66 2 74734. 8 13279.4 8024. 8 104 270 6 1286 7981 487 2812 268 13991 21972 7981 2812 13991 7 12668 842 4243 17266 2733 19446 3407 1061 96 270 8 2724 726 4802 10988 37 4802 2724 726 0 06 9 124482 3 634729 610094 17830 18692 68093 1949 4 138146 1 12400 10000 60 66336 44739 217797 41 3262 43313 668992 62679 139 3133 61 62 162.39 169.4 94 133.1 942. 2 671.179 76.18 9 1800.3 0 26.4 94 13 68 63 6647 2378 42879 1622 4704 469 94007 3838 186 100 94.1 64 612.7 44 248.8 8 313.8 86 186.24 3619.3 84 8070.9 4 441. 6 0 0 6 66 6034 13020 6986 12642 9607 2982 43481 1366 0 0 327.68 318.61 324.09 333.16 67 7 9.072 14.404 1.983 6 8 9.072 0 0 68 2788.6 12 2421.4 16 367.19 6 39.2 1 1001.4 3663.0 13 7428. 39 376. 26 40 128 69 7847 332 2132 2733 3336 2192 121076 99124 0 0 70 31960 0 24070 0 618900 412010 0 960700 419060 0 802320 0 383260 0 63000 32300 71 1180 291 9214 4102 310 8616 19483 10867 06 68 72 73 103334 2408 0926 17964 16 88872 160029 7117 1900 440 74 21833 10218 1161 1073 1061 1161 21833 10218 186 7 7100 107200 3700 20200 7100 2700 48100 20600 11200 0 76 1319.6 01 323.3 3 191.7 3 11262. 12643.7 131.6 72 27067. 2 271. 8 0 482 77 7740.6 94 3144.6 97 49.9 97 144.3 99 231.13 12633. 9 32002. 26 19368. 31 0 0 78 82746 82946 200 7106 2973 7773 9491 86818 0 0 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 119

79 271.0 1 14876. 47 960.4 2 7861.1 6 1262.6 267.4 74 3822. 88 3298. 41 1096 40 80 3303 18018 1017 2186 20 6100 7918 18018 2047 79 81 7407.1 68 6181.2 44 122.9 24 8369.0 1 762.41 3961.2 47 11740. 4 7779.1 48 236 440 82 767.2 8 7439.3 04 1672.0 2 7302.6 9 4277.6 18229. 9 2668. 89 7439.3 04 1420 409 83 2 228 2003 27 1216 36 3631 3266 0 0 69.80 1 3971.1 2 84 31.62 7 1011.4 28 370.03 1830.4 19 4093.9 22 2263. 03 286 67 8 61628 6741 4887 2367 10098 146447 310604 16417 0 0 86 248768 3761 19007 46687 432 239646 29402 4406 0 0 87 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 120

Appendix C: List of paired companies Bankrupt Company Healthy Company AccuHealth New York Health Care Inc AHT Corp Pacific Health Care Organization Inc All Star Gas Corp Ferrellgas L.P. American Banknote ACG Holdings Avado Brands Carrols Corp Big Buck Brewery & Steakhouse Eat At Joes Ltd Big V Supermarkets American Consumers Briazz Inc Meritage Hospitality Group Inc Casual Male Corp Jos. A Bank Clothiers Inc CD Warehouse Hastings Entertainment Inc Cinemaster Luxury Theaters Inc Carmike Cinemas Inc Computer Learning Centers Hartcourt Companies Inc Converse LaCrosse Footwear Cooker Restaurant Corp ELXSI Corp Crown Books Corp Borders Group Inc Drug Emporium Inc BioScrip Inc Eagle Food Centers Homeland Holding Corp etoys Inc A.C. Moore Arts & Crafts Inc Florsheim Group Inc Rocky Brands Inc Furr's Restaurant Group Inc Mexican Restaurants Inc Gadzooks Inc Ascena Retail Group Gerald Stevens Inc 1800 Flowers.com Inc Healthcare Integrated Services RadNet Inc HeiligMeyers Company Jennifer Convertibles Inc Homeland Holding Corp Arden Group Horizon Pharmacies Inc Express Scripts Inc House2Home inc Home Depot Image Innovations Holding Inc Kolorfusion International Inc Integra Inc Sagemark Companies Ltd Jacobson Stores Inc BonTon Stores KushnerLocke International Inc Family Room Entertainment Corp Lamonts Apparel Inc Children's Place Retail Stores Inc Loews Cineplex Entertainment Corp AMC Entertainment Med/Waste Inc Commodore Applied Technologies Natural Wonders Inc FragranceNet.com Inc New York Bagel Enterprises Inc Star Buffet Inc One Price Clothing Inc Cache Inc Orbit Brands Corporation AllAmerican SportPark Inc Paper Warehouse Inc Kirkland's Inc Park Pharmacy Corp Standard Management Corp Pathmark Stores Inc Whole Foods Market Inc Paul Harris Stores Inc Charming Shoppes Inc Payless Cashways Inc Lowe's Companies Inc Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 121

PharMor Inc. Piccadilly Cafeterias Inc Planet Hollywood International Inc Platinum Entertainment Inc Play Co. Toys and Entertainement Corp Prandium Inc Premier Concepts Inc Prosoft Learning Corporation Provell Inc Quokka Sports Inc Regal Cinemas Inc Salex Holding Corp Samuels Jewelers Inc Schlotzskys Inc Sight Resource Corp Southern Investors Service Company Inc Spiegel Inc Strouds Inc Tender Loving Care Health Care Services Tesseract Group Inc UCI Medical Affiliates Inc Ultimate electronics Inc Unapix Entertainment United Petroleum Corporation USA Biomass Corp Valley Media Inc Video City Inc Video Update Inc VisionAmerica Inc Wall Street Deli West Coast Entertainment Inc WHSU Inc (fka Micro Warehouse Inc) Zany Brainy Inc Omnicare Inc Western Sizzlin Corp Steakhouse Partners Inc Internet Infinity Inc Toys R Us Inc Noble Roman's Inc Finlay Enterprises Inc New Horizons Worldwide Inc Amazon.com TIX Corp United Artists Theatre Circuit Inc Tilden Associates DGSE Companies AFC Enterprises Inc CareGuide Inc Cala Corp International Commercial Television Inc Pier 1 Imports Inc PHC Inc Questar Assessment Inc U.S. Physical Therapy Circuit City Stores Inc Image Entertainment Inc Great Atlantic & Pacific Tea Company Inc Pdg Environmental Rx for Africa Inc Netflix Movie Gallery Inc Amsurg Corp Million Dollar Saloon Inc Blockbuster Inc Overstock.com Michaels Stores Inc Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 122

Appendix D: List of Altman Z Scores Group 1 DMU Altman yr1 Altman yr2 Altman yr3 Altman yr4 Altman yr 1 1.02816 1.3672 4.11063 0.76143 2 6.19879.70409 6.317347.92714 7.674664 3 9.00843 1.60376 1.1903 0.84066 2.2762 4 2.38 1.88687 1.2939 0.22986 1.0984 26.1886.4806 82.34271 13.077 10.3648 6 1.3706.9379.29401 0.80399 0.4422 7 0.3246 0.3694 0.3229 0.90462 0.920296 8 3.2292 4.26238 1.7437 3.9144 2.67892 9 7.11427 6.468897 6.49289 6.17068.829379 10 0.2774 2.20707 4.0988 2.27892 3.978287 11 0.141417 1.7378 0.779773 3.244462 2.742414 12 6.736668 6.21101 6.720829.01742.672862 13.13131 3.971421 8.91708 9.89682 8.934337 14 0.869 0.368 0.296 1.1337 0.481912 1 431.007 6.3206 226.1103 0.927 0.47243 16 1.269728 1.431133 1.109106 1.146747 1.01706 17 1.6807 0.8474 0.41148 3.19477 2.72214 18 4.28027 4.44239 4.171217 4.333718 4.006009 19 2.11036 2.140143 1.9422 1.741777 2.161006 20 0.3996 33.0174 4.43081 1.4791 12.8607 21 18.3976 3.79269 1.00462 11.3964 836.224 22 0.2264 0.0623 0.083789 1.289669 0.86471 23 0.32387 0.02301 0.271017 0.106288 0.10682 24 2.712288 2.66709 2.61307 0.399376 2.816346 2.04746 0.43831.011888 17.6742 6.0682 26 6.3086 6.88793 4.74189 1.71196 9.3602 27 6.86679 3.73733 4.83274 6.4021 2.9372 28 12.7742 9.60688 7.824 7.6714 2.66306 29 0.043444 1.82603 3.214787 4.367226 3.2116 30 1.9112 4.76401 1.7412 0.09793 1.9983 31 4.10493 0.899649 1.67937 2.8704 4.7349 32 0.361 3.94179 4.60867 3.619916 4.90327 33 1.8181 0.070113 0.00477 0.31012 0.24989 34 0.14608 2.73372 3.19797 2.79967 2.841729 3 1.11147 0.967916 1.128284 1.14196 36 11.432 80.73848 41.24429 38.94817 11.6683 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 123

37 0.9388 1.02318 4.9764.18877 8.30988 38 8.3608 18.17397 11.8933 39 1.13440 1.079342 1.997124.437329.712811 40 6.1603 2.99366 22.221 41 0.338803 0.439627 1.1122 0.63731 1.23967 42 8.94664 0.7214 2.39963 3.30942 4.204237 43 4.74608 1.79046 0.0432 0.4947 1.6103 44 3.21063 2.09386 2.9397.8224 6.9388 4.97281 6.416937 6.70727 6.3418.46107 46 0.344 2.900321 2.191036 1.86331 0.40069 47 1.06773 8.0813 0.4378 0.9381 1.13049 48 13.2693 19.861 21.8277 49 2.743279 2.317738 3.08866.143134 3.837 0 2.71819 1.0189 1.1916 2.03911.1844 1 4.03239 3.06797 1.108341 0.27104 2.7209 2 2.77089 3.286811 3.80786.04401.48381 3 6.376194 6.79126.7024.40644.44998 4 0.82291 1.033611 0.818713 0.920783 1.226087 4.02822 0.78102 3.280972.372449 2.487977 6 1.77673 3.941107 3.911906 3.24174 3.69116 7 8.174982 0.26608 8 8.7463 18.9903 19.66 18.3126 19.3388 9 7.343 1.684379 0.979 1.648369 2.21293 60 3.12161 3.611141 3.6302 3.67232 3.387368 61 3.63281 7.29384 8.26607 11.4972 10.164 62 3.69399 3.039241 3.23461 2.968723 2.69266 63 6.1611 1.883 2.74273 4.331787 3.70836 64 2.00622 1.877387 1.67224 2.49779 3.88049 6 3.82499 3.728617.084192 6.41332 6.712236 66 1.0804 0.9242.1772 4.49114 2.1794 67.61848 6.27999 6.314837 6.66776.880931 68.0168 0.03423 0.401779 0.66737 0.701772 69 4.86461 3.702 1.8466 0.0012 0.063 70 0.3104 0.27761 0.4179 1.30282 0.49273 71 2.417061 1.49479 1.638347 1.67386 2.764327 72 1.3013 2.16702 3.432084.761171 14.42073 73 4.34003 8.31767 11.1013 12.38 6.42302 74 7.204426 6.124692.063728 7.183274 8.8836 7 2.4412 2.4609 14.79003 19.8393 10.2247 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 124

Group 2 DMU Altman yr1 Altman yr2 Altman yr3 Altman yr4 Altman yr 1 0.12648 0.69219 1.44216 0.881269 0.28288 2 6.8287 61.0344 49.301 33.2804 4.9106 3 7.16216.36122 1.9839 0.968178 2.31238 4 4.08128 3.711109 3.96424 4.100487 3.019346 0.639086 2.32712 2.440406 4.6106 10.1472 6 6.01364 6.877838.49148 4.289302 3.7134 7 2676.8 36671.7 410.039 98104 39.18 8 17.9044 131.479 129.16 8.9246 27.043 9 4.171199 4.3376 4.972779.10116 4.19339 10.07036.882896.67192.226163.69233 11 4.001944 1.728862 1.882641 2.410308 4.29886 12 3.849883 3.73839 3.40638 2.78979 2.938734 13 8.442 26.81 12.0893 6.77027 1.29022 14 0.91124 1.163994 1.3493 1.93201 1.77797 1 8.323421 10.7371 1.33292.18387 8.447183 16.03304 0.89748 12.3764 2.921694 3.98932 17 0.360289 3.41142 3.286 2.1214 18 3.06289 3.97874 3.787741 3.63816 3.74876 19 1.64032 1.43466 3.411678 3.934178.13781 20 6.111703 13.3768 10.26492 8.07396 21 0.71716 2.8398 1.2860 1.424691 1.948673 22 2.630468 3.318887 4.467919 3.764124 3.730267 23 1.746819 6.37902 6.02139 6.44616.626 24.281607 7.709208 10.203 18.8443 16.049 2 1.37373 1.76672 2.627149 16.23424 1.3974 26 1.1093 0.6278 0.797683 4.49897.028183 27 10.6 1.24003 12.40483 2.0378 4.238614 28 6.64438 7.2091 4.363167 1.9991 3.487712 29 3.46471 3.09689 3.0814 3.287793 4.29801 30 1.8007 0.42394 1.4917 4.187623 3.2633 31 138141 332.6 2214.4 366.677 32 6.743342 8.92074 12.34716 10.323 33.421 3.93104 1.06742 0.347 2.00201 34 8.08289 1.6241 3.899326 3 3.31392 36 2.1978 2.30112 1.991897 2.12077 0.748197 37 3.6418 6.02963 3.9677 2.736723 4.7888 38 1.003682 2.026224 2.17144 2.921427 3.0246 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 12

39 7.468806 7.791874 2.70602 2.1496 3.686131 40 0.63134 0.0881 0.12608 1.0049 0.841244 41.1166 6.612431 6.47783.128034.28933 42 0.18377 1.349717 4.407196 0.213424 1.16183 43 1.18191 0.024219 0.62782.86783 3.941 44 1.133 18.963 7.6369 4.13913 4.83918 4.3896 2.76436 1.77003 2.08439 1.16041 46 39.6268 38.4246 30.426 37.7328.934417 47 13.3977 1.71626 0.23846 1.4822 1.9412 48 1.886 2.8330 4.2863.64283 9.173044 49 20.13176 13.26182 19.28779 14.2201 0 33.6997 38.889 4.097 18.7282 1 4.90992 0.0188 0.03378 1.820964 4.447682 2 24.6337 9.198848 3.22613 2.936926 3.47617 3 11.6846 8.16773 7.20036 2.149 4 7.84623.40306 0.46988 1.10044 2.12022 0.24334 1.78986 3.148386 3.14378 2.894378 6 26.6224 12.4488 3.0277 0.1919 0.822 7 4.9849 3.7639 1.06342 0.08613 0.00109 8 37.2621 32.6774 30.1722 28.332 2.476 9 2.6042 4.294179 3.87002 3.269637 2.847264 60 2.12491 2.848434 2.437138 2.170189 2.268128 61 0.168701 0.428847 2.808671 0.0164 62 2.10746 3.9776 4.20666 11.7974 10.1116 63 3.3021 2.462094 2.46396 1.18274.412698 64 0.270 0.99162 7.16027 2.300429 2.8762 6 3.7483 4.93688 10.932 66 3.6244 0.219912 7.73346 3.09277 1.19336 67 3.3896 3.120 4.67884 3.38339 43.3212 68 28.1019 18.737 10.932 11.732 1.99113 69 0.6821 0.4363 0.86841 1.9663 1.7077 70 3.18364 2.727683 2.39301 3.914789 4.13281 71 7.440029 7.268184 6.332494 6.036963.83191 72 1.81672 1.08413.63116 73 2.4664.374372 3.80342 4.76424 4.4184 74 3.060142 3.099709 3.1421 3.608618 4.84978 7.1214 9.74638 2.78689 2.23641 1.6633 76.4678 1.31113 18.239 12.817 4.90618 77 8.4817 1.4384 0.16197 2.612311 2.261328 78 0.77043 1.13212 1.146772 0.011682 0.3621 79 3.2894 1.177 2.1812 27.28 2.3713 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 126

80 10.3712 9.8076 1.1638 3.68462.124089 81 9.60229 2.327326 1.879314 1.831167 0.66773 82 1.6798 0.23733 0.02218 3.001093 1.93272 83 14.232 1.17229 1.31139 0.64029 1.72407 84 0.13081 1.402186 0.1396 1.200272 3.97778 8 1.800474 2.3878 3.099 2.66128 0.72977 86 4.770182 4.09676.72779 10.846 10.294 87 2.30879.70602 3.3971 3.089878 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 127

Appendix E: List of DEA Scores for Original Model Group 1 DMU yr1 yr2 yr3 yr4 yr 1 0.84119 1 0.09 0.790418 2 0.967186 0.963 1 0.939966 0.984279 3 0.02972 0.216737 0.403041 0.7029 0.2866 4 0.176146 0.16832 0.188229 0.9293 0.360801 0.669401 0.77161 1 1 0.31388 6 0.028021 0.18611 1 0.226606 0.383206 7 0.182469 0.34909 0.386706 0.60849 0.69809 8 0.308 0.001744 0.1873 0.2117 0.41899 9 1 1 1 1 1 10 0.398876 0.68396 0.8361 0.83682 0.883961 11 0.398239 0.801861 0.7426 0.8239 0.827991 12 1 1 1 0.93077 0.91716 13 0.947033 0.939844 1 1 1 14 0.79793 0.6279 0.964 0.608926 0.66130 1 1 1 1 0.02327 0.49003 16 0.98129 0.70247 0.6106 0.677418 0.683233 17 0.287761 0.422078 0.68868 0.399799 0.498814 18 0.9412 1 0.83071 0.837282 0.811 19 0.8113 0.80889 0.81817 0.61191 0.67102 20 0.107668 0.261793 0.01344 0.0460 0.046327 21 0.229921 0.460274 0.461844 0.069074 1 22 0.386874 0.78696 0.437292 0.637398 0.627873 23 0.182142 0.347942 0.60802 0.427061 0.46414 24 0.722778 0.74987 0.646382 0.40 0.73947 2 1 0.7726 1 1 0.946113 26 1 0.826423 1 1 0.26168 27 0.272199 0.477671 0.464628 0.21744 0.268014 28 0.199103 0.2224 0.471948 0.1326 0.473 29 0.62707 0.799389 0.8423 0.89744 0.8676 30 0.008144 0.168302 0.17808 0.418493 0.17073 31 0.37884 0.8934 0.64069 0.748799 1 32 0.43028 0.89134 0.870682 0.6073 0.87019 33 0.0077 0.34924 0.3238 0.40481 0.6388 34 0.370793 0.809347 0.81723 0.783162 0.79289 3 0.3867 0.69368 0.73401 0.7481 36 1 1 1 1 1 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 128

37 1 0.79912 0.720847 0.7998 0.68211 38 1 1 1 39 0.6034 0.8948 0.773938 0.91871 0.940902 40 1 0.1027 1 41 0.8937 0.93696 0.4102 0.46874 0.791201 42 0.009787 0.38667 0.47606 0.663942 0.74382 43 0.36264 0.360627 0.478708 0.36111 0.71046 44 0.360663 0.17083 0.17642 0.013266 0.012894 4 0.832608 1 1 0.927786 0.801362 46 0.447839 1 0.64773 1 0.726278 47 0.94997 0.16711 0.406013 0.640844 0.438664 48 0.271919 0.3283 0.214718 49 0.794834 0.63291 0.63627 0.868421 0.80623 0 0.191327 0.4307 1 1 0.487794 1 0.229832 0.63412 0.439802 0.823401 1 2 0.899843 1 1 1 1 3 1 1 1 1 1 4 0.7089 0.6203 0.386766 0.397972 0.9947 0.183237 0.41020 0.40766 0.9424 0.646798 6 0.600448 1 0.912717 0.72093 0.89133 7 1 1 8 0.09404 0.46478 0.434701 0.448128 0.191043 9 0.168021 0.639844 0.771738 0.873476 0.918344 60 0.888182 0.872862 0.814319 0.80020 0.674898 61 0.227766 0.173476 0.028819 0.039133 0.033617 62 0.71647 0.672829 0.68881 0.68810 0.64468 63 0.188977 0.418818 0.63747 0.874902 0.874279 64 0.38344 0.0486 0.9261 0.46334 0.67076 6 0.71321 0.670287 0.72718 0.9103 0.948667 66 0.402914 0.383764 0.006173 0.008086 0.34988 67 1 1 1 1 1 68 0.199878 0.438319 0.46739 0.62601 0.674026 69 0.273186 0.281829 0.69303 0.649748 0.00071 70 0.437 0.389661 0.380407 0.21834 0.391218 71 0.8614 0.711869 0.673297 0.6212 0.744 72 0.17424 1 0.889381 1 1 73 0.182073 0.168966 0.17366 0.013914 0.010013 74 1 1 0.813373 1 1 7 0.13666 1 1 0.23798 0.02289 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 129

Group 2 DMU Year 1 Year 2 Year 3 Year 4 Year 1 0.421627 0.70217 0.73994 0.27841 0.378608 2 8.71E02 1 0.17721 0.237611 0.208037 3 0.241149 0.27378 0.390743 0.461921 0.27089 4 1 0.816024 0.99924 0.882211 0.931478 1 1 1 1 1 6 0.922678 0.842911 0.86037 0.860637 7 1 1 1 1 0.27493 8 0.220001 0.11324 8.62E02.81E02 0.0261 9 1 1 1 1 1 10 1 1 1 1 1 11 0.867021 0.60341 0.70748 0.627127 0.71386 12 0.87287 0.8396 0.832643 0.76422 0.841693 13 1 1 1 0.4201 14 0.761303 0.784691 0.79246 0.813748 0.80447 1 1 1 1 0.87983 1 16 1 1 1 1 1 17 1 0.26946 0.18667 0.39483 18 1 1 1 1 1 19 0.62336 0.96916 0.64479 0.73412 0.776612 20 1 1 1 1 21 0.704164 0.74116 0.73008 0.662284 0.683171 22 0.692238 0.764383 0.766388 0.8189 0.808194 23 0.79404 0.92232 0.9080 0.906783 24 1 1 1 1 1 2 0.4147 0.793029 0.804907 1 1 26 0.488694 0.64449 0.740736 1 1 27 0.289676 0.43223 1 0.287693 1 28 1 1 1 0.60073 0.63032 29 0.904798 0.934883 0.92126 0.946449 1 30 0.176736 0.384992 0.7620 0.840847 0.87229 31.80E02 1 0.18709 0.113239 6.4E02 32 1 1 1 1 33 7.71E03 0.18263 0.6469 0.400002 0.631049 34 7.89E03 0.763306 0.804913 3 0.93103 0.974016 1 0.936136 0.893092 36 0.64438 0.6610 0.63132 0.76427 0.9082 37 0.843727 0.837306 0.776779 0.814736 1 38 0.46696 0.602723 0.601789 0.6269 0.809123 39 1 1 0.632493 0.6442 1 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 130

40 0.19277 0.4226 0.6761 0.72877 0.83267 41 1 1 1 0.931818 1 42 0.42308 0.623702 0.67676 0.3242 0.33188 43 0.61469 0.83843 0.409782 0.206804 0.379282 44 0.171986 1.41E03 1.30E03 6.8E04 2.3E04 4 0.467722 0.44723 0.461689 0.48873 0.18423 46 0.43919 0.28098 0.41693 0.49363 1 47 1.08E03 0.23100 0.38106 0.39078 0.86892 48 0.772744 0.78804 0.790647 0.910869 1 49 1 1 1 1 0 1.4E02 9.87E02 0.1434 1 1 0.61644 0.16823 0.42624 0.6279 0.780909 2 1 1 0.741021 0.67167 0.763926 3 3.64E02 6.10E02 0.044937 0.39998 4 0.197218 0.3988 0.23994 0.601013 0.361631 0.490046 0.863223 0.89177 0.70121 0.833649 6 1.07E02 0.37346 0.449416 0.60146 0.694221 7 0.20341 0.272621 0.428 0.09213 0.496231 8 0.11194 0.146973 9.94E02 0.218199 4.93E02 9 0.199084 1 0.93138 0.86314 0.84827 60 0.83088 0.9434 0.86604 0.772388 0.77978 61 0.60042 0.614691 0.829469 0.607831 3.83E02 62 0.36413 0.34441 0.17441 7.36E02 0.869769 63 0.67721 0.607417 0.43134 0.46266 0.47132 64 0.42348 0.418779 0.1962 0.483046 6 0.374872 0.17724 3.01E03 66 0.27048 0.60302 0.99224 0.647733 0.3861 67 0.6738 1 1 1 68 0.22998 0.349231 0.334019 0.346401 0.467772 69 0.7366 0.379183 0.3906 0.91783 0.682 70 1 1 1 1 1 71 1 0.884701 0.80919 0.877784 0.946902 72 0.807696 0.79149 1 73 0.764801 1 0.876133 0.837 0.82731 74 0.678478 0.793736 0.743604 0.6423 0.709089 7 6.09E04 7.9E04 0.17631.28E04 0.1781 76 0.20326 0.391167 2.18E02 0.184869 0.177708 77 0.191726 0.403764 0.90711 0.87217 0.782383 78 0.189193 0.71632 0.7166 0.18227 0.4996 79 1.93E02 4.19E02 0.207338 7.01E03 0.19978 80 1.12E03 1.90E03 0.284823 0.88739 0.93247 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 131

81 2.02E03 0.643276 0.98107 0.69097 0.60614 82 0.344262 0.29033 0.489 0.91726 0.724 83 8.8E04 0.324914 0.68778 0.48209 0.462894 84 0.1788 0.68766 0.43233 0.641729 0.277643 8 0.834629 0.868182 0.910862 0.81191 0.463482 86 1 0.823986 1 1 1 87 0.4371 0.797106 0.674229 0.707381 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 132

Appendix F: List of DEA Scores for Revised Model Group 1 DMU yr1 yr2 yr3 yr4 yr 1 0.333164 1 0.30132 2 0.733747 0.820271 0.91014 0.740011 3 0.11931 0.327913 0.7324 0.66802 1 4 0.261671 0.34033 0.30208 1 0.44322 0.48249 1 1 1 0.39121 6 0.023642 0.26669 1 0.32786 0.841408 7 0.2361 0.671041 0.04232 0.7798 8 0.66223 0.138663 9 1 1 1 1 1 10 11 0.436613 0.8606 0.770341 0.83449 0.80989 12 0.831077 0.8644 0.944927 0.80167 0.931984 13 0.8183 0.89766 1 1 1 14 0.86984 0.71448 0.687331 0.79127 1 1 1 1 1 0.74896 0.631393 16 17 0.427388 1 1 1 1 18 0.734832 0.92297 0.888016 0.79909 0.823739 19 0.940711 0.87869 0.8869 0.64248 0.616629 20 0.041263 0.27224 0.382091 21 0.27802 0.974 22 0.301104 0.768 0.66702 0.934901 0.83411 23 24 0.622 0.720769 0.6723 0.449696 0.66076 2 0.0432 0.603291 1 1 1 26 1 0.623179 0.00666 27 0.283834 0.374311 0.36190 0.293741 0.399074 28 0.22083 0.376089 0.47439 0.488799 1 29 0.48442 0.797266 0.932329 30 0.079662 0.390781 0.21364 0.398828 0.441134 31 0.72979 0.726027 1 1 32 0.28297 0.610466 33 0.1037 0.41688 0.494094 0.471681 0.14733 34 0.22724 0.72136 0.72974 3 36 1 1 1 1 1 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 133

37 0.626741 0.696729 1 1 1 38 1 1 39 1 1 1 1 1 40 1 41 0.87877 42 0.096932 0.2446 0.490833 0.6072 1 43 0.296146 0.329848 0.47712 0.0844 0.26987 44 1 0.236472 0.22247 0.042927 0.032961 4 0.2898 0.802717 0.882447 0.74023 0.121 46 0.403109 0.63298 0.6296 0.46821 47 0.2178 0.349219 0.674407 48 1 0.466387 49 0.690762 0.60408 0.64961 0.800987 0.808998 0 0.31928 0.4307 0.32897 1 0.136693 1 1 1 1 2 1 1 1 1 1 3 1 1 1 1 1 4 0.07814 0.36641 0.28900 0.282908 0.01332 0.27604 0.38223 0.49024 0.872779 0.639227 6 0.421494 1 1 0.723877 0.86200 7 1 1 8 0.046187 0.3221 0.6012 0.621844 0.21712 9 0.2316 0.6020 0.80404 0.860101 1 60 0.783028 0.79807 0.691879 61 0.28891 0.321603 0.111907 0.08049 0.1130 62 0.4804 0.64304 0.63168 0.3819 0.644086 63 0.29146 0.89328 1 1 64 1 6 0.637323 0.7704 66 0.1177 0.1986 0.076209 0.079673 67 1 1 1 1 68 69 0.278096 0.372043 0.816891 0.670131 70 0.61310 0.4972 0.496829 0.301868 0.108 71 0.80466 0.748281 1 0.764068 0.81324 72 0.31683 0.31699 0.762346 0.768129 1 73 1 0.30623 0.30877 0.0871 0.1749 74 1 1 1 1 1 7 1 1 1 1 1 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 134

Group 2 DMU yr 1 yr 2 yr 3 yr 4 yr 1 0.46711 0.47923 0.88912 0.668094 2 1 1 1 0.297686 0.232162 3 0.193698 0.16812 0.43901 1 1 4 1 1 1 1 1 1 1 1 6 0.80338 0.7816 0.732 0.7349 0.741326 7 1 8 0.312393 0.11624 0.233742 0.07801 0.33178 9 0.781707 1 1 1 1 10 1 1 1 1 1 11 1 0.797979 0.88611 0.770433 1 12 0.833131 0.84364 0.86843 0.762287 0.81066 13 1 14 0.718111 0.721766 0.740193 0.774489 0.7224 1 1 1 1 1 1 16 1 1 17 0.07142 18 1 1 1 1 1 19 0.748376 0.7179 1 1 1 20 1 1 1 21 0.714944 0.790811 0.80608 0.81249 0.73619 22 0.4703 0.738177 0.742174 0.7397 0.78266 23 0.48908 0.96627 0.78374 0.7911 0.810278 24 0.96119 1 1 1 2 0.414281 0.831186 0.72337 26 0.72873 0.694043 0.81343 0.93736 27 0.27084 0.324079 1 0.30929 28 1 1 1 0.77046 29 1 1 1 1 1 30 0.22287 0.4287 0.673034 0.70411 0.718967 31 1 1 0.203669 0.227973 0.217421 32 0.983127 1 1 33 0.064207 0.24748 0.476434 0.298021 0.499008 34 0.08490 0.74892 1 3 0.742291 0.809662 0.82217 0.83986 0.832203 36 0.643387 0.63697 0.61421 37 1 1 1 1 38 0.37892 0.78292 0.60396 0.627077 0.889177 39 0.71172 0.738634 0.7629 0.78198 0.98268 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 13

40 0.2666 0.674921 0.66282 0.71100 41 1 1 1 1 1 42 0.382483 0.630338 1 43 0.417611 0.26932 0.326774 0.28866 0.2960 44 0.238602 0.012967 0.0422 0.039839 4 0.443642 0.397361 0.39868 0.36161 0.376091 46 1 0.31387 0.3607 1 47 0.080437 0.401284 0.01378 0.333207 0.7774 48 0.808178 0.833 1 1 1 49 1 0 0.11846 1 1 1 0.67934 0.697268 1 1 2 1 1 1 1 3 0.18269 0.120963 4 0.17783 0.821 1 0.686713 0.810434 6 1 1 1 1 1 7 0.2776 0.268074 0.284184 0.328874 0.31693 8 9 0.33897 1 1 1 1 60 1 1 1 1 1 61 0.64622 0.688836 0.764073 62 0.432096 0.4193 0.221398 0.0834 63 0.631682 0.37384 0.372621 0.416414 0.8699 64 0.492271 1 0.7127 1 6 0.286932 0.08319 66 0.377327 1 1 67 68 0.40739 1 1 1 0.3148 69 0.08636 0.294436 0.28481 0.66 70 1 1 1 1 1 71 1 1 1 1 1 72 0.77487 0.73438 0.8136 73 0.62393 0.84398 0.794234 0.760331 0.82383 74 1 1 0.969801 1 0.728429 7 76 0.067793 0.22922 77 1 1 78 0.330844 0.778292 0.747974 79 0.12437 0.128733 0.1282 0.02097 0.24076 80 0.01779 0.01736 0.267791 0.710941 0.7274 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 136

81 0.036496 0.609437 0.67894 0.79389 0.494676 82 0.268928 0.48391 0.46833 0.2832 0.24439 83 0.009016 0.31418 0.6709 0.63767 84 0.37039 0.70988 0.440723 0.672416 0.292482 8 86 0.960603 0.70908 1 1 87 0.293024 0.617344 0.4200 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 137

Appendix G: Tstatistics for comparison of means Group 1 Ttest Year 1 Year 2 Year 3 Year 4 Year Current Assets 1.1923 1.07729 1.00739 1.0623 0.890224 Current Liabilities 0.33820 1.046314 0.993181 1.182496 0.941828 Working Capital 1.9662 0.988972 0.9968 0.8769 0.804439 Retained Earnings 1.479387 1.31776 1.22643 1.2961 1.22838 Operating Income 0.9639 0.64266 0.74217 1.209921 1.33238 Book Value of Equity 1.212464 1.080198 1.08306 1.227896 1.300712 Total Assets 1.143004 0.92067 0.73822 1.04898 0.902211 Total Liabilities 0.783621 0.4181 0.1631 0.683649 0.199802 Employees 0.389899 0.48777 0.6691 0.70284 0.897813 Shareholders 1.27603 1.361882 1.223349 1.446163 1.328607 Group 2 Ttest Year 1 Year 2 Year 3 Year 4 Year Current Assets 2.438824 2.1984 2.2770 2.096074 2.002797 Current Liabilities 1.19228 2.309794 2.30446 2.113746 1.99883 Working Capital 2.441631 1.308733 1.331461 1.9821 1.22212 Retained Earnings 1.442274 1.418272 1.73748 1.929784 1.79811 Operating Income 1.83421 1.169864 1.087039 2.12213 1.963881 Book Value of Equity 2.08918 2.010937 2.210111 2.127467 1.9247 Total Assets 2.43716 2.3408 2.47121 2.312161 2.1122 Total Liabilities 2.18402 2.2496 2.31881 2.330803 2.01633 Employees 2.930714 2.8479 2.8886 2.7967 2.36637 Shareholders 2.224432 1.81917 1.836212 1.72836 1.610209 Data Envelopment Analysis of Corporate Failure for NonManufacturing Firms 138