THe-sas L;_._-;~bs;. «,5 a E .' ¢ 4-3- . -—- -£ 1 r 3-1....- ‘.up n . ' I . . .. ‘ -------'!" ‘v-v ‘3 ’—.~— -- fl. 1._‘_-¢-..~:..‘ n J ‘7" l This is to certify that the dissertation entitled THE EFFECT OF CHAPTER 11 REORGANIZATION ON THE RISK AND RETURN CHARACTERISTICS OF COMMON EQUITY ISSUES presented by Susan E. Moeller has been accepted towards fulfillment of the requirements for Ph . D. , Finance degree m WSW Major professor Date W MSU is an Affirmatiw Action/Equal Opportunity Institution O~ 12771 TV1ESI_J RETURNING MATERIALS: Place in book drop to LJBRAfiJES remove this checkout from “ your record. FINES will be charged if book is returned after the date stamped below. THE EFFECT OF CHAPTER 11 REORGANIZATION ON THE RISK AND RETURN CHARACTERISTICS OF COMMON EQUITY ISSUES BY Susan E. Moeller A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Finance and Insurance 1985 Copyright by SUSAN E. MOELLER 1985 ABSTRACT THE EFFECT OF CHAPTER 11 REORGANIZATION ON THE RISK AND RETURN CHARACTERISTICS OF COMMON EQUITY ISSUES BY Susan E. Moeller The objective of this research was to measure the effect of reorganization on common equity issues. Capital market data were analyzed to determine how the risk and return characteristics changed for firms that filed under the 1978 Chapter 11 code. The study examined the performance of the stocks of bankrupt firms after filing. Prior studies looked at the reaction to the news that a firm was filing for bankruptcy. The sample for this study was taken from a list of firms which filed under Chapter 11 after October 1, 1979. Based on the availability of stock return data for each firm, the sample was divided into two groups: Group I with daily data available and Group II with monthly data available. Estimates of systematic risk and abnormal return performance were obtained from the two-parameter and mar- ket models for each firm. Average risk premia over the Susan E. Moeller market index were also calculated as risk proxies. T-tests were done to determine how these risk estimates changed from before to after filing and if abnormal returns existed after filing. Results showed that 79% of Group I and 60% of Group II experienced a decline in systematic risk. How- ever, the variance of the stock returns increased for all of Group I and 58% of Group II after filing. Although the risk premia did not change for 62.5% of Group I and 85% of Group II firms, 50% of Group I and 62% of Group II earned more than the market, on average, after filing. Abnormal returns, as measured by the two-parameter model, were not obtained for either group. ‘ While this study did not determine if the returns on stocks of firms under Chapter 11 were enough to compen- sate investors for their risks, the data did show that for some firms large absolute returns were available to investors who purchased their stocks after filing. DEDICATION To Molly Moeller ACKNOWLEDGMENTS I would like to thank my committee members, Drs. Myles Delano and George Kutner for their assistance with this research. I am grateful for all the valuable advice and guidance provided by my committee members throughout all phases of the study. Thanks are due Mrs. Jo McKenzie for typing the preliminary drafts as well as the final copy of the thesis and Ms. Pam Moore, who provided me with excellent computer assistance. Finally, I would like to thank Dr. Ronald F. Cichy for providing me with the inspiration to finish this dissertation. iv TABLE OF CONTENTS LIST OF TABLES. O O O O O O O O O I O O O O O O 0 LIST OF FIGURES O O O O O O O O O O O O O O O O 0 Chapter I INTRODUCTION I O O O O O O O O O 0 O I 0 Purpose of Study . . . . . . . . . . . Institutional Setting. . . . . . . . . Investing in Chapter 11 Firms. . . . . An Overview of Macro- and Micro- Environmental Factors Affecting Reorganizing Firms. . . . . . . . . . II LITERATURE REVIEW. . . . . . . . . . . . Empirical Bankruptcy Research. . . . Theoretical Implications of Chapter 11 Petitions on the Performance and Risk of Firms' Secturities. . . . . . Summary. . . . . . . . . . . . . . . . III RESEARCH DESIGN AND METHODOLOGY. . . . . Objectives . . . . . . . . . . . . . . Assumptions. . . . . . . . . . . . . . Sample Selection . . . . . . . . . . . Estimates of Systematic Risk and Abnormal Returns. . . . . . . . . Statistical Testing Procedures . . . . IV RESEARCH RESULTS . . . . . . . . . . . . Summary of Results for Groups I and II. . . . . . . . . . . . . . . . Analysis of Group I Daily Return Data . . . . . . . . . . Analysis of Group II Monthly Return Data . . . . . . . . . Page vii ix 1? 27 27 35 45 46 46 49 59 63 70 70 71 81 Chapter V SUMMARY AND CONCLUSIONS. . . . . . REFERENCES. APPENDICES. Implications of the Results. . . Limitation of the Study. . . . . Suggestions for Future Research. vi 95 97 98 100 106 Table II III IV VI VII VIII IX XI XII XIII XIV LIST OF TABLES Exchange Listing Prior to Filing . Size of Sample Firms at Filing . . Industry Membership. . . . . . . . Age of Firm at Filing. . . . . . . State of Balance Sheet at Filing . Summary of Characteristics . . . . Test for Differences Among Groups. Group I - Change in Risk Premia over Time I O O O O O O O O O O 0 Group I - Characteristics of Risk- Premia Subgroups . . . . . . . . Group I - Change in Risk Premia - Timing and Direction of Absolute Group I - No Significant Change in Risk Premia-Timing and Direction of Absolute Change . . . . . . . Group I — Percentage of Firms with 8. Estimates Significantly Different 3 From zero. 0 O O O O O O O O O C Group I - Change in Average 8. Pre- to Post-Filing. . . . . . . Group I - Explanatory Power of Regressions. . . . . . . . . . . Group I - Percentage of Firms with Abnormal Returns Significantly Different From Zero. . . . . . . Group II - Change in Risk Premia Over Time. . . . . . . . . . . . vii 56 56 57 57 77 77 78 78 79 79 80 80 86 Table XVII XVIII XIX XX XXI XXII XXIII Group II - Characteristics of Risk Premia Subgroups. . . . . . . . Group II - Change in Risk Premia- Timing and Direction of Absolute Change. . . . . . . . . . . Group II - No Significant Change in Risk Premia - Timing and Direction of Absolute Change . . . . . . . . . Group II - Percentage of Firms with B. Estimates Significantly Different From Zero. . . . . . . . . . . . . . Group II - Change in Average 8. Pre- to Post-Filing. . . . .3. . . . Group II - Explanatory Power of Regressions. . . . . . . . . . . . . Group II - Percentage of Firms with Abnormal Returns Significantly Different From Zero. . . . . . . . . viii 87 87 88 88 89 89 Figure 1 LIST OF FIGURES Macro View of Events Affecting Chapter 11 Firms . . . . . . Reorganization . . . . . . . . Micro-Environment of Chapter 11 Firms. . . . . . . . . . . . ix CHAPTER I INTRODUCTION Historically, the topic of bankruptcy and its economic ramifications on a firm's value has been studied within the context of determining bankruptcy costs and their impact on a firm's risk and return. These studies were basically concerned with explaining the ex ante effect of bankruptcy on the value of a firm. Beginning with the early 1970's and the bankruptcy of the Penn- Central Railroad, some economists expressed concern about the impact of corporate failures on our economic system. Analyses were undertaken in an attempt to understand how U.S. bankruptcy laws affected the redistribution of in— vestor wealth. From these studies the concensus came that the laws needed revision since they tended to delay the recovery of financially distressed firms and often resulted in the inefficient use and reallocation of resources. On July 14, 1978, the National Bankruptcy Act was amended to become the Bankruptcy Reform Act of 1978. This new code treats the common equityholders of distressed firms more leniently than the old law and makes it easier for these firms to file for reorganization. 2 Purpose of Study The study measured the effect of reorganization under the 1978 Bankruptcy Reform Act on firms' equity issues. The return and risk characteristics of such firms' stocks were examined to measure the direction and magnitude of any risk changes over time and also to deter- mine whether investors who purchased the common stock of these firms after they filed for reorganization earned abnormal returns. While much research has been done that supports the efficient Operation of the capital markets with respect to firms filing for reorganization under the old law, little has been done to examine the effect of re- organization under the new code on the performance and risk measures of firms' debt and equity issues. Many studies, using pre-l978 data, have shown that -- ex post -- the capital markets react efficiently to announcements of a firm's financial distress.1 In addition, Aharony, Jones, and Swary (1980) showed that the total risk of a firm's returns (measured by the variance of returns) and its unsystematic risk component showed a dramatic increase when compared to similar successful firms. Their analysis did not extend beyond the filing dates of the sampled firms. The sample for this study, on the other hand, included those firms that filed Chapter 11 lSee Altman (1969a), Westerfield (1970), Aharony, Jones, and Swary (1980) and Clark and Weinstein (1983). 3 petitions for reorganization after the new code became effective on October 1, 1979 and survived for at least a year after filing. The data set was obtained by taking price information for all sample firms after filing from either the Wall Street Journal, from the pink sheets obtained from Merrill, Lynch, Fenner, and Pierce or from Standard and Poor's Daily OTC Stock Price Record. The formal research hypotheses were stated as follows: 1. Ho: The systematic risk of Chapter 11 firms did not change significantly after filing. H1: The systematic risk of Chapter 11 firms changed significantly after filing. 2. H : The purchase of Chapter 11 firms' equities did not result in abnormal positive returns for investors. H1: The purchase of Chapter 11 firms' equities did result in abnormal positive returns for investors. Institutional Setting The original National Bankruptcy Law required that large public firms filing for bankruptcy do so under Chapter X of the code. Under Chapter X, a firm's management was replaced with a trustee who devised 4 a reorganization plan for the firm. Allocation of funds to creditors was done according to the absolute-priority rule, which stated that all senior claimants be paid in full before junior creditors received anything. To avoid these rigid, formal requirements, firms customarily attempted to file under old Chapter XI instead of Chapter X. The former was designed to be used by small non-public firms--usually solvent--needing protection from creditors and time to evaluate their unsecured debts and deve10p a repayment plan. Informal negotiations between creditors and debtor firms accelerated the reorganization process, permitting management to retain control of the firm and operations to continue. Because of the leniency with which debtor firms were treated under Chapter XI, many public firms sought protection under the code. Although this situation offered more potential benefits for firms' shareholders than under Chapter X, it caused difficulties for its creditors in three main areas.2 First, secured debt and equityholders' rights were not well protected under the Chapter XI code.3 Trost states the issue very clearly in comparing proceedings under Chapter XI and Chapter X:4 2Schnepper, The New Bankruptcy Law (Addison- Wesley, 1981). 31bid. 4Trost, "Corporate Bankruptcy Reorganizations: For the Benefit of Creditors or Stockholders?", 21 U.C.L.A. Law Review, p. 540-541 (1973). 5 "Under Chapter X, any plan which provided cred- itors with as much or more than they would receive upon liquidation meets the statute's command that the plan must be in the "best interests" of creditors. Thus, shareholders of an insolvent Chapter XI corporation may retain an interest in the reorganized corpo- ration even though creditors are not paid in full. But in Chapter X reorganizations the court applies a quite different rule: Unless the corporation is solvent, shareholders can- not participate in the fruits of the reorgani- zation because of the application of the so— called "absolute priority" doctrine." Second, creditors were hurt because the SEC was empowered by the Courts to determine the value of a bank- rupt firm. To protect shareholders from losing their investment entirely, the Commission tended to overestimate the value of the firm as a reorganized entity. Trost dis- cusses the difficulties that resulted from this procedure.5 "Corporations in reorganization are valued as a going concern by capitalizing the prospec- tive earnings of the rehabilitated corpora- tion. Although market values, liquidation values, and past earnings records may be relevant, they are not determinative. . . . Some courts and commentators are more candid than others about the difficulties inherent in the valuation problem. . . . Particularly troublesome is the choice of the proper cap- italization rate. Assuming that future earnings can somehow be forecast, a change in the cap- italization rate one-half point up or down can have momentous consequences for junior interests . . . . By a slight change of the capitalization rate, an insolvent company in which shareholders are denied participation becomes a solvent com- pany in which shareholders are entitled to some kind of interest. It seems not likely that implicit in the choice of a rate may be the 51bid., 545-546 quoting Gardner, "The sec and Valuation Under Chapter X," 91 U. PA. Law Review, p. 440—4-3 (1943). 6 desire to rationalize a predetermined result, as for instance, the retention of the common shareholders in the enterprise." Finally, many firms wished to file under Chapter XI because management was allowed to remain in control of the operations, whereas under Chapter x they are automat- ically ousted. Meckling points out:6 "There are apparently occasions on which it pays creditors to concede something to shareholders in order to keep the current man- agement . . . . It is an open question whether the managerial retention case is empirically important. As a corporation approaches in- solvency, it will become more and more attrac- tive to the stockholders to have management convert assets into cash and pay the cash out as dividends. There are constraints on what management can do to serve these interests, but such constraints cannot be fully effective. One conversion opportunity which is very diffi- cult to monitor is the abandonment of mainte- nance. In the case of Penn Central, for example, it was practically impossible to tell whether the failure to maintain track beds was efficient or represented a conversion of assets into cash flow for the benefit of stockholders. Given that stockholders have such opportunities, creditors may find it desirable to concede positive shares in bankruptcy to them as a way of discouraging exploitation." Reform of the National Bankruptcy Act addressed two of the above problems creditors faced when firms filed under Chapter XI instead of Chapter X by combining the procedures of both codes into a new form of Chapter 11. Its new provisions can be summarized as follows:7 6Meckling, "Financial Markets, Default, and Bank- ruptcy: The Role of the State," 41 Law and Contemporary Problems, p. 36-37 (1977). 7 Op. cit., Schnepper. 7 1. Firms do not have to demonstrate balance sheet insolvency in order to petition for reorganization, but can now file based on an inability to pay their general creditors. This amendment allows firms to file more quickly and before large—scale erosion of their financial positions occurs. 2. Management is allowed to retain control of the firm, sustain operations, and provide a return on investment for shareholders. 3. Not all reorganization plans need to adhere to the absolute-priority rule. More compromise and in- formal negotiation can now take place between creditors and debtors. 4. The role of the Securities and Exchange Com- mission (SEC) in reorganizations has been expanded. To help understand the expanded role of the SEC and why the new law appears to give filing firms greater benefits than the old law, an overview of the court's procedural system follows. The involvement of the SEC in reorganization cases was as important under the old law as it is under the new one. However, the form of its authority has changed. The Chandler Act of 1938 provided the procedural means for the SEC to participate in bankruptcy corporate reorganization proceedings to provide protection for public investors. Under the new Chapter 11, a disclosure approach--analogous to that under the securities law--is followed in lieu of 8 the former approach involving scrutiny and approval of a plan proposal's merits by the SEC and the court. Under the old law, the SEC had to be invited by the judge or interested parties to participate in a reorganization proceeding. Now Section 1109 (a) of Chapter 11 gives the SEC authorization to intervene at any time and be heard on any issue.8 In addition, Congress determined that -- as under the federal securities law -- the parties of interest should be given adequate information to enable them to make their own informed judgments as to whether they should accept or reject a reorganization plan. The amount of disclosure required is flexible, taking into account the facts of a particular case, and is determined by the ' SEC. In the interest of expediency, reorganized plans, under the new law, can be approved without an extensive investigation of the firm’s financial situation.9 In the past, Chapter X required that the trustee transmit to creditors, stockholders, and other interested parties a report or summary of an investigation into a firm's financial state. One purpose of that report was to provide information to those who could submit 8Corroto, A. R. and Pickard, I. H. "Business Reorganizations Under the Bankruptcy Reform Act of 1978 -- A New Approach to Investor Protections and the Role of the SEC." 28 DePaul Law Review (Summer 1979), p. 962-967. 9 Ibid., p. 987-991. 9 suggestions in response to the trustee's request. Under Chapter XI, the debtor submitted an arrangement which, in most cases, had been negotiated with the official creditors' committee. The new Chapter 11, however, provides that a debtor can file a plan at any time, even before an examiner's report is completed. Thus, it will be diffi- cult for other interested parties to comment on the plan since they will not have received all the information available about a debtor's financial state. In summary, it is possible that creditors and stockholders will be asked to vote on a reorganization plan having only limited information about a firm's financial state.10 Not only the lack of required information, but also the acceptance standards, tend to favor the adoption of management's proposed reorganization plan. Under former Chapter X, the "fair and equitable" standard, otherwise known as the absolute-priority rule, was one of the key public investor protection features. Since the former Chapter XI confirmation standard of "best interests of creditors" meant that creditors received more than in liquidation, a negotiated arrangement usually worked to retain enhanced value for the shareholders at the expense 10Ibid., p. 983-984. (Since the appointment of trustees is not required under the new law, the court sometimes appoints an examiner to investigate a debtor firm. The court decides on a case-by-case basis the nature and extent of the investigation. Also, it may not be necessary for the court to consider the report of an examiner before the reorganization can proceed.) 10 of the unsecured creditors. Thus public debenture holders, under former Chapter XI, could receive less than their absolute priority vis-a-vis stockholders. Con- sidering that the debtor was the only one that could pro— pose a plan, creditors had to negotiate with management for a larger share of the company. Currently, under Chapter 11, both sets of standards are applied. The key provision of the law provides that the court shall confirm a plan if either "each class" (or claimant) has accepted or is not impaired under the plan.11 The court must examine the acceptance of the plan by each impaired class in order to determine if each member of the class has accepted the plan, or, if not, whether a person rejecting the plan will receive under the plan "not less than the amount" he or she would have received in liquidation. An accepting class majority can bind the dissenting minority members to a distribution differing from that which the "fair and equitable" standard would otherwise mandate as long as it is not less than what the minority member would have received if the debtor were liquidated. This approach will have a significant impact on public classes of debt securityholders because they will be bound to a distribution where only 2/3 in amount and 1/2 in number of allowed claims of the class actually voting can decide to accept the plan proponent's offer. llIbid., p. 999-1000. 11 Thus, the interests of a public investor class may be con- trolled by the interests of the few, larger class claimants.12 If a plan is rejected by one or more classes, the proponent of the plan may request that the court still confirm the plan if it is found to be "fair and equitable" to each such class. The importance of adequate and effec- tive public investor representation and participation in a reorganization case is underscored by the new act's approach to confirmation. Under former Chapter X, public investors had the independent trustee, the SEC, and the court to ensure them of a fair and equitable participation under the plan. Thus, they were not left to negotiate a relative participation on their own. The act now places this responsibility on the investors with a "fairness" backstOp possible only if the class dissents and the proponents elect to go forward and not withdraw.13 In summary, with the Bankruptcy Reform Act of 1978, Congress adopted a new approach to the means by which in- vestor protections are achieved in business reorganization of public entities. The prior focus on the disinterested trustee, the required investigation, the independent formu- lation of a plan, SEC scrutiny, and court approval of the plan is now shifted to an emphasis on investor initiative lzIbid., p. 1002. 13Ibid., p. 1004. 12 and vigilance and to the disclosure of "adequate informa- tion" in order that the interested parties themselves can reach informed judgments. The flexibility of former Chapter XI and the rigidity of former Chapter X have given way to a single reorganization chapter under which the presumption is that the debtor generally will remain in possession. However, with disclosure of "adequate infor- mation" drawing closer to the securities laws concept than was the case under former law, debtor control of plan formulation and confirmation will have to be balanced against the protection of investors. SEC participation in the disclosure process and prior court approval of the disclosure statement are two important investor protection features designed to assure continued investor protec— tion.14 One additional problem that the new law does not address is the valuation of a reorganizing firm by the SEC and the reliance of the court on this estimate of firm value. Meckling expresses his concern with the law's revisions when he states:15 "In particular, the revisions would signifi- cantly soften rigid adherence to absolute priority in settlement claims under Chapter X. This will tend to give shareholders in publicly held corporations more power in bankruptcy than they currently possess. Meanwhile the revisions more or less ignore the problem of valuation of l4Ibid., p. 1005. 150p. cit., Meckling, p. 37-38. 13 a firm . . . . As a result of intensive research over the last decade or so, we now know a great deal about how the value of firms is determined in financial markets. Most of us would have little faith that estimates derived by the SEC staff would be superior to those that markets would generate." As the new code has only been in effect a short while, no consensus exists in the literature as to whether the new law is more efficient than the one it replaced. The current opinion of some bankruptcy experts is that, on balance, the new law is easier to apply and less costly for all involved.16 Lack of data and confounding economic variables make any empirical comparison of the old law versus the 17 new code extremely nebulous. For example, Dun and Bradstreet's 1982 Failure Record reported that the number of firms filing for new Chapter 11 protection rose since the law became effective in 1979. The reasons for this dramatic increase in firm failures may have been totally unrelated to the law's revision. Instead, high interest rates, a long, persistent recession, and climbing debt 18 Given this ratios may have been the principal culprits. conflict, the aim of this study was not to determine if investors were better off investing in firms which filed 16Altman, E. 1. "Discussion on Behavior of Firms in Financial Distress." Journal of Finance, May 1983, p0 517-5180 17 Ibid., p. 517. lBIbid., p. 517. 14 under the old law. Instead, the implications of the new law were taken as the legal environment in which all Chapter 11 firms must operate. The importance of the law for this study were twofold: 1. The law allowed distressed firms the oppor- tunity to file quickly and halt basic deterioration of their capital structures. It seems evident that the more quickly they were able to file and start reorganization proceedings the more likely they were to emerge success- fully from Chapter 11. 2. There seems to be a significant difference in bankruptcy costs under the old versus the new law. White (1983) argues that aggregate bankruptcy costs are lower under the new law. If market efficiency holds for Chapter 11 securities, the effect of these lower costs should be incorporated into their prices. If market participants are unaware of these cost changes, or if aware, uncertain as to their measurement, then prices of Chapter 11 firms may not reflect their true values. In conclusion, although the new Chapter 11 law attempts to correct some of the delays in resolution of bankruptcy cases, it appears to provide less protection for creditors. In addition, new shareholders of a newly reorganized firm may be able to earn positive abnormal returns which are simply the reflection of negative abnormal returns to other claimants. 15 Investing in Chapter 11 Firms Although many investment houses have bankruptcy specialists who trade Chapter 11 firms' securities, there were only two publicly available mutual funds in the busi— ness at the time of the study. Investors attracted by the opportunities afforded by reorganizing companies, but not by all the laborious work that goes into analyzing them, could purchase shares in either Max Heine's Mutual Shares Corporation or Merrill Lynch's Phoenix Fund.19 Of these two funds, Mutual Shares is the older and has the better performance record. Holding $370 million in assets,this no-load mutual fund rose 15.5% for the twelve months ending September 30, 1984. Lipper Analytical Services rated its performance fourth out of 700 funds. In comparison, the Dow Jones Industrials rose only 2.8% and the Standard and Poor's 500 (S&P 500) gained 4.6%. Over a ten-year period, Mutual Shares investors enjoyed even greater returns. It had an impressive 893% gain, ranking 17th on the Lipper list, versus 241% for the Dow and 326% for the S&P 500. A review of Mutual Share's annual report provided some insight into what type of securities Mutual Shares Corporation purchased to obtain such a strong performance. Most of the investment in reorganizing companies was done 19All of the following information was obtained from the annual reports of the two funds and from conver- sations with the managers of the funds. 16 through their secured senior securities. Dun's Business Month (December 1984) reported in an interview with Heine that the fund usually gets more than par value out of the securities. In addition, Heine claims he buys the equities of bankrupts on "rare occasions." Using funda- mental analysis, Heine buys on news rather than rumor. As of the date of this study Mutual Shares had large positions in the bonds of AM International and Wickes Corporation and in the equity of John Mannville. Finally, it should be noted that the fund does not hold only securities of re- organizing firms, but has invested in undervalued, neglected stocks. Merger situations, bankruptcies and other special situations account for one-third of the Mutual Shares Portfolio. The rest is in undervalued stocks. As a newcomer to the high risk securities game, Merrill Lynch's Phoenix Fund has had an interesting per- formance record. Organized in November of 1982, the fund earned a 31.05% return in 1983, but only 9.93% in 1984. The Phoenix fund mainly invests in "troubled" companies with an objective of achieving long-term capital growth. The mix of securities includes convertible issues, corpo- rate bonds and municipal bonds as well as equities. The fund's investment strategy is similar to Mutual Shares. In the case of a company experiencing cash flow problems, an investment is made in its common stock. Filing for Chapter 11, however, makes the senior debt of a firm more attractive to the fund. 17 In analyzing the performance of the Phoenix fund over the last two years, it is apparent why there was such a return differential. In 1983 the successful recovery of Ford Motor Company was a major factor, as it represented one of the fund's largest holdings. On the other hand, the 1984 collapse and default of the Washington Public Power Supply system caused the major setback in the fund's asset value. A sixty million dollar drop occurred in the value of these bonds held by the Phoenix Fund Port- folio. The current interest of investors in Chapter 11 securities makes the topic of this study very timely. Two important concerns of these investors are: Have Chapter 11 equities earned abnormal returns in the past? And how has the risk of their equity changed? An Overview of Macro- and Micro-Environmental Factors Affecting Reorganizing Firms Macro Analysis Although many factors have been cited as causes of firm failure, what is of interest to potential in- vestors in Chapter 11 firms is the macro-environment con- fronting a firm entering reorganization. First of all, no matter what the cause of a firm's difficulties, cer- tain responses can be expected from its management. Figuresl and 2 show schematically the consequences of such responses. 18 Figure 1 Macro view of events affecting Chapter 11 firms Costs Causes 4%1, Poor Management Expansion.._____’ 0‘ (‘ Competition Technology...._...e’ Failure 4%, Economic Conditions Borrowing Enters Pressure Costs Rise 01 stress by Creditor Change Out of i" Violates covenants “5" Omits dividends CW" Reorgani- Cuts costs ' zation Reorganizes finances Reaction Changes Policies 0 Liquida- y tion Market Files for Chapter 11 Liquida- tion Process Reorganization Process l9 Figure 2 Reorganization REORGANIZATION Liquida- Smaller tion Version Under New Name Merged Smaller Hith Version Another of Firm Old Firm Sold to Another Firm 20 As creditors begin to pressure delinquent firms for compensation, management reacts by instituting a variety of measures. They may violate debt covenants by defaulting on interest payments. Shareholder dividends may be omitted, although this is usually not done until the firm has absorbed months of negative cash flow. Management may attempt to reduce costs by obtaining concessions from its labor force and suppliers. Failing in that approach, it may begin to close plants, lay off salaried workers, sell off unprofitable divisions or take other actions in an attempt to reduce overhead costs. In addition, management begins to look for new financial support from its bankers and other general creditors. This is because the key to survival for a Chapter 11 firm tends to be in its ability to reorganize its capital structure. Finally, a shift in policy may be made by management as it seeks new ways to increase cash flow. Different types of projects or lines-of- business may be taken on under such circumstances. The capital markets respond to these actions by re-evaluating the risk and return relationship of these firms' securities. In most cases, the trading in these securities shifts from the major exchanges to regional exchanges and over-the-counter (OTC). Institutional in- vestors usually lead the way in selling off these securi- ties as they become more speculative. Credit markets demand higher interest rates on new loans (of course, 21 in some cases, no loans will be forthcoming).20 Other factors which have been documented in the literature as affecting financially distressed firms are bankruptcy costs and increased risk of a firm's expected cash flows. Although bankruptcy costs have been empirically shown to exist, there has been debate in the literature as to whether they are high or low and as to how the risk of future cash flows really changes.21 Financial distress for a firm may not be fatal. Many large firms (e.g. Chrysler, Massey-Ferguson, Inter- national Harvester) have been able to reorganize their capital structures through out-of—court settlements. Some firms, however, must file for Chapter 11 protection or be forced into liquidation by creditors. Such firms were used in this study. Since the purpose of reorganization is to allow negotiations to occur between debtor firms and their creditors under the supervision of the bankruptcy court, the effect of this process on a firm's future earning power is important to investors and management. At any point during the negotiations, the courts can rule that the firm must be liquidated. The length of time a firm remains in reorganization depends on the extent of its 20This summary of responses was collected by actual observation of distressed firms' behavior as cited in the financial press and Moody's manuals. 21See Haugen and Senbet (1978) and Warner (1977a) and Warner (1977b). 22 problems and the willingness of its creditors and the courts to support it until reorganization. The Chapter 11 firms that emerge from reorganization are usually smaller, with little debt, and sometimes have a new name. Others are acquired by another firm. Micro Analysis Understanding the micro-environment faced by a Chapter 11 firm is important for potential investors. Figure 3 illustrates the usual reactions of investors to the plight of a Chapter 11 firm. As the company enters financial distress, the market quickly reacts to this potential 'bankruptcy effect' and re-evaluates its expec- tations with regard to the risk and return relationship of a bankrupt company's securities.22 Aharony, et. a1. (1980) showed that the total variance of Chapter X firms' returns drastically increased as they approach formal bankruptcy. The basic question this study addressed is how the following characteristics changed during and after filing for Chapter 11. l. The systematic and unsystematic risk of a firm's returns. 2. The risk and return relationship, described by abnormal returns, of a firm's equity. 22See Altman (1969a), Westerfield (1970), Aharony et. a1. (1980), and Clark and Weinstein (1983) for a summary of bankruptcy event studies. 23 Figure 3 Micro- Environment of Chapter 1 1 Firms Efficient Reaction to Information ---—>l Expectations Change th Retained Riskier Policy “ecisions Files for Chapter 11 Court Valuation of Firm In Reorganization Process Effect on i Unsyste- matic Risk? Enters Financial Distress Expectations of Future Cash Flows Change Wealth Transfer Abnormal Equity ? Returns 24 Throughout the reorganization literature, a variety of reasons have been postulated as to why the risk and return tradeoff of Chapter 11 firms may change and give rise to abnormal returns. A summary of the current thought in this area is given below.23 The return on a Chapter 11 firm's equity is a function of the following variables: 1. Market forces 2. Reorganization risk (uncertainty surrounding the court proceedings). 3. Firm-specific risk changes. Market Forces One of the market forces which could account for Chapter 11 stocks being undervalued is that the market for these stocks may be inefficiently priced. One reason for the lack of efficient pricing is that fewer buyers than sellers are interested in transactions with such securi- ties. Institutional investors sell such stocks because the funds they manage often have restrictions against holding such securities. In addition, Chapter 11 firms' stocks trade infrequently and are often only offered by dealers. 23A more detailed economic analysis is undertaken in Chapter 2. 25 Reorganization Risk Reorganization risk is defined as the uncertainty surrounding the reorganization process. This multi- faceted risk, imposed by the bankruptcy court, is beyond the control of management. One facet of this risk is the negotiating process. There is a great deal of uncertainty concerning the effect an agreement between creditors, man- agement, and shareholders will have on the division of a firm's assets. Another facet of reorganization risk is the choice of valuation principles adopted by the bank- ruptcy court. The court determines how the wealth of the firm will be allocated among individual claimants. Finally, the uncertainty surrounding bankruptcy costs is another facet of reorganization risk. Bankruptcy costs include lost revenues resulting from a slow judicial process and foregone opportunities for the disposal of lines of busi- ness and other assets. Firm Specific Risk Firm-specific risk is defined as the risk arising from factors that are characteristic to a particular come pany and, by definition, are unrelated to other companies. The intensity with which specific risk will affect returns of Chapter 11 firms depends on their financial strength at the time of filing. If there exists a higher earnings potential after filing, or if their managements change the thrust of investment policies to include less risky 26 projects, then their specific risk will not affect returns as much as if no changes in policy occurred after filing. In summary, this section has provided a brief overview of the macro- and micro-environments faced by Chapter 11 firms. A more detailed discussion of the specific economic consequences of reorganization on firm values follows in the succeeding sections. CHAPTER II LITERATURE REVIEW As the new Chapter 11 reorganization code has only been in effect since October 1, 1979, most literature on the topic of firm failure has dealt with firm bankruptcy under the old Chapter X. These studies were concerned with explaining the ex ante effect of bankruptcy on a firm's value. Once a Chapter X firm was technically bank- rupt or in reorganization, most researchers were no longer interested in studying its risk and return characteristics. Reorganizing firms were viewed as losers which did not survive in a competitive business environment and whose fate would be decided by the judicial system. Although the author has not uncovered any research which analyzed the effect of the new Chapter 11 code on a firm's stocks, many empirical and theoretical studies have looked at the effect of the Chapter X code on a firm's value and performance. A review of this literature follows. Empirical Bankruptcy Research Past empirical research on bankruptcy can be grouped into two general categories: the deve10pment of 27 28 bankruptcy forecasting models and the testing of the infor- mation effects of financial distress on a firm's security prices. The first type of research has received much atten- tion in the literature. These bankruptcy forecasting models have been validated and were able to predict corpo- rate failures. Altman (1968) made the most significant contribution to the bankruptcy literature. Using financial ratios and discriminant analysis, he succeeded in con- structing a regression model that classified any firm into one of two categories: those most likely to survive and those most likely to fail. Since the publication of his original research, Altman and other investigators have refined his methodology by developing.more sophisticated forecasting models and by demonstrating the usefulness of the models.1 The second main form of bankruptcy research focuses on the speed with which the market reacts to the news that a firm is approaching insolvency. In separate studies, Altman (1969), Westerfield (1970), Aharony et. a1. (1980), and Clark and Weinstein (1983) concluded that there were market reassessments of the solvency posi- tion of failed firms four to six years prior to their filing for bankruptcy under Chapter X. However, their 1For an excellent summary of these models see Altman, "Bankruptcy and Reorganization," The Financial Handbook, Chapter 35, 1981, p. 32-37. 29 research findings indicate that failure itself was not expected by the market. Instead, the market displayed a strong negative reaction to a firm's stock when it finally announced it was filing for Chapter X. While past research provided insights into how investors can predict corporate failure and how the market reacted to the news of financial distress, most literature was concerned with measuring the post-bankruptcy perform- ance of bonds. The first studies in this area were done by Calkins (1948) and Hickman (1958). Calkins studied eight cases of bankruptcy reorganization during the period 1941-1946. He found that if the claims were discharged under the principle of absolute-priority then the market value of newly issued securities in subsequent years generally exceeded the amount of the original claim. Calkins concluded that this occurred because of low esti- mates of earning power by the SEC and determined that the senior bondholders benefitted at the expense of junior creditors and equityholders. This loss was partially offset by low SEC estimates of a firm's discount rate. Hickman (1958) studied the behavior of defaulted bonds between 1900 and 1943. He found that given an average coupon rate of 6.9% for industrial firms' bonds, their realized yield at the time of default was negative 7.5%. If, on the other hand, an investor had purchased these bonds at default and held them to maturity, their return would have been 25.8%. Hickman concluded that 30 the market undervalued bonds at or near the date of default with two results: 1. Capital losses were extremely heavy on bonds that were purchased at offering and sold at default. 2. Returns were large to investors who bought at default and held to maturity (or when the issue was extinguished). In response to Hickman's conclusions, Baskins and Crooch (1968) studied the rates of return on flat bonds. Their sample focused on two forms of railroad bonds: income and defaulted. They compared the annual rates of returns on these bond groups with the returns associated with common stocks from the 1968 Fischer and Lorie study.2 They found that all bonds issued between 1950 and 1960 earned an average return of 6.8%. The range of all the returns, however, was from a negative 1.9% to a positive 32.4% with the maximum yield of 88% on a single bond and a 100% loss on another. They also concluded that the returns on defaulted railroad securities were highly correlated with the overall return on the common stocks from the Fischer and Lorie study. In a more recent paper, Warner (1977b) examined the effect of bankruptcy on the bond returns of 20 2Fischer, L. and Lorie, J. H. "Rates of Return on Investments in Common Stock -- The Year-by-Year Record, 1926-1965," Journal of Business, 1968. 31 bankrupt railroads during the time period 1930 to 1955. His measure of abnormal performance was the difference between the return on the defaulted bond portfolio and the return on another portfolio of securities with the same level of systematic risk as that of the defaulted bond portfolio. The following results were observed: 0 The systematic risk of the railroad bonds increased prior to filing for bankruptcy. 0 Negative abnormal returns occurred during the month the bankruptcy petition was filed. 0 In the post—bankruptcy period, investors who purchased this portfolio of bonds on the date of the bankruptcy petition appeared to earn significant abnormal returns. Unfortunately, it is difficult to generalize from Warner's results to the performance of all defaulted bonds. Federal regulators tend to encourage the continued oper- ation and successful reorganization of bankrupt railroads. An extensive literature search, up to the time of this study, revealed that only one researcher studied the post-bankruptcy experience of common shareholders. Altman (1969) developed a model to quantify the risk/return per- formance of a bankrupt firm's stocks. He then empirically tested the model with data taken from the time period 1941-1965. The general form of the model was: n Rt 2 t t=0 (1+k) I SPI = 32 where SPI = Stockholder Profitability Index R = Returns from New (or Old) Securities (includes dividends and capital gains) k = Stockholder Opportunity Cost I = Investment n = Number of years The variables were determined as follows: 1. Returns (R) were defined as the cash flows accruing to shareholders in the form of dividends plus capital gains. 2. Shareholder investment (I) was determined as the amount an investor could have received had he sold the security or would have invested had he purchased the security prior to bankruptcy filing. 3. The opportunity cost of capital (k) was set at the average rate of return on all New York Stock Exchange stocks during the period in question. Altman used the above model to test the hypothesis that the SPI of bankrupt firms would not be significantly different from 1.0. This testing procedure indicated whether an investor received returns commensurate with his expectations of a firm's risk return level. During the analysis, Altman adjusted his model to include the losses and gains due to taxes on shareholders. He used a 25% tax rate on capital gains and losses. He believed that this addition to the model would correct for any 33 statistical bias that resulted from the SPI's of the firms not following a normal distribution. Altman's results showed that common shareholders of a bankrupt firm tended to earn as much return as all other equityholders of New York Stock Exchange stock, if the reorganized firm survived for at least 5 years after its filing date. However, this result only applied to 33% of the sample. When summarizing his results, Altman offers advice to those interested in bankrupt stocks:3 "The trick is to determine which firms are likely candidates for a successful reorganiza- tion and then to wait at least one month after the petition date to purchase the securities. It was found that the price of bankrupt firm equities falls on average 25% from one month before failure to one month after. This drop in price, sometimes referred to as the bank- ruptcy information effect, implies that the market was not totally anticipating the bank- ruptcy, or else the price would have been fully discounted." Altman's thesis that the stock of some bankrupt firms is a bargain for an astute investor may be true; but the mistakes he made in formulating his methodology may discount the importance of his results. Hanna (1972) summarized Altman's methodological errors as follows: 1. The rates of return of New York Stock Exchange stocks were used as a surro- gate for the required rates of return of all bankrupt firms, even though Altman acknowledged that the two groups had 3Op. cit., Atlman, p. 37. 34 different risk characteristics. The rates of return of the New York Stock Exchange stocks did not include dividend income in their calculations. It was not appropriate to use a measure of portfolio return as a proxy for the opportunity cost of investing in a single stock. The sample of bankrupt firms was biased, because firms whose equity became worth- less during the testing period were gradually excluded from the sample. Tests for normality of returns were not made, yet statistical tests requiring such an assumption were used. His conclusions did not coincide with the design of his research. While he claimed to test for the post-bankruptcy performance of a firm's equity, he began the testing period one month before filing instead of at the filing month (this was after the final marketplace reaction had occurred). His goal was to prove incorrect the general- ization that reduced share values result from the bankruptcy process itself. Due to methodological problems, however, he does nothing more than confirm the obvious: 35 that equity share values of a firm can be expected to fall in bankruptcy. Theoretical Implications of Chapter 11 Petitions on the Performance and Risk of Firms' Securities Since Altman's study, more theorists have been interested in analyzing the economic ramifications of bankruptcy on firms and their investors. The 1970 bank- ruptcy of the Penn Central Railroad sparked the American Finance Association to devote a large part of its proceed- ings that year to this problem. Gordon (1971) presented a paper outlining the theory that, as a firm's financial position becomes tenuous, the firm's increased risk results in a wealth transfer from bondholders to shareholders. He examined the values of bankrupt railroads' debt and equity over the period 1966 to 1970 and discovered that, although the value of equity fell to one-third its original value, the value of senior secured debt fell more drastically during this time period. Financial distress affected common stock returns first, and then it affected debt returns. However, after bankruptcy, the prices of the common stocks rose before any other security. Gordon also found that, while technically the market value of a bankrupt firm's equity should be zero when the market value of firm falls below the par value of its senior debt securities, this was not the case. Although pre-bank- ruptcy owners lost absolute wealth, their stock was not 36 worthless. In fact, the equity value of these railroads increased as a percentage of total firm value. Gordon concluded that a wealth transfer to shareholders had occurred in the case of these railroads. Since the time of Gordon's paper, others have advanced reasons why a wealth transfer may occur to the shareholders of bankrupt firms from the creditors. In 1977 a seminar on the economics of bankruptcy reform was held at the University of Miami's School of Law to discuss the wealth transfer question and others related to the pending revision of U.S. bankruptcy law. Moore (1977) summarized the resulting arguments as follows: 1. Some practices in corporate bankruptcy proceedings seem to favor debtors over creditors. The shareholders and manage- ments of firms in Chapter 11 proceedings may have an advantage over creditors be- cause of the management powers they retain. 2. When shareholders are left in control of an insolvent firm during a Chapter 11 pro- ceeding, the incentive to make riskier policy decisions may lead to a transfer of wealth from creditors to debtor shareholders. 3. The valuation practices of the SEC in bank- ruptcy proceedings tends to subvert the absolute-priority rule in favor of debtors. 37 Little empirical research has been done to support the above list of assertions. Any attempt at further research, however,.must consider the above arguments, as they constitute the present core of expert thought on the effect of bankruptcy on a firm's value. Each idea repre- sents a reason new shareholders of Chapter 11 firms may, in general, be able to earn excess returns. An analysis of the above three contentions follows. 1. Retention of Management In a discussion of the causes of firm failure, Dewing (1931) attributed most bankruptcies to some form of management incompetence. Argenti (1976) summarized such management deficiencies as follows: 1. Ignoring product technology changes. 2. Fraud. 3. Maintaining a poor intercompany communica- cation system. 4. Reluctance to establish tight cost control standards. 5. Ignoring cash flow when making investment decisions. 6. Expanding the firm too quickly by floating excessive amounts of debt. 7. Increasing sales at the expense of profits. 8. Concentrating too many resources into one large project. Thus, if poor management is a main cause of firm failure, it is logical to question whether changing man- agement will result in a favorable response by the market 38 toward a firm's predicament.4 Although not yet empiri- cally tested, it can be hypothesized that if a change in management results in a more efficient operation, the mar- ket would react favorably. Stocks, which may be correctly priced at the time of default, may increase in value due to new positive information. This situation would not result in excess returns assuming market efficiency exists, as a bankrupt firm's stock would already have had incorpo- rated into its price the effect of a change in management. On the other hand, there are several circumstances which might have allowed Chapter 11 stocks to earn excess returns, even if poor management were retained. First, the personal appeal of a firm's chief operating officer could be a main reason the financial markets agree to support a firm's reorganization plan.5 Second, creditors may support retention of the current management and its reorganization plan. Their motive could be to prevent the conversion of assets to cash dividends by a manage- ment expecting to be ousted in the event of filing.6 While retaining management does not, as an act in and of itself, lead one to believe that a Chapter 11 firm could earn excess returns, it is the final result that occurs from retaining management that allows one to make certain 4Arthur S. Dewing, Corporation Finance (N.Y.: Ronald Press, 1931). 5Ibid., Dewing. 6 Op. cit., Meckling, p. 37. 39 assertions. If retaining management allows a quick, inexpensive reorganization plan to be accepted by all parties, the firm has a chance to emerge from Chapter 11 financially strong. Existing management will have another chance to run a successful operation. On the other hand, Altman (1983) noted, that in the case of mismanagement, old management is rarely retained. Management is often forced to resign by creditors. They bring about executive changes by threaten- ing to force a firm to file involuntarily for Chapter 11. In certain cases, if old management retains control, a firm's old problems will likely remain. In conclusion, it appears that retaining manage- ment as a reason for a firm's securities earning excess returns is dependent upon the firm's characteristics. No generalizations can be made as to whether retaining or ousting management would in every case cause a firm's securities to behave in a certain manner. 2. Riskier Policy Decisions Using the option pricing model, Miller (1977) provided a clear analysis of why Chapter 11 firm's share- holders and management may have another advantage over its creditors. In general, shareholders can be viewed as holding a call option on a levered firm, If they could increase the riskiness of the firm's cash flow, the value of their option would increase. This change would result 40 in a wealth transfer from bondholders. Bondholders have committed their funds to projects with certain expected risk/return characteristics. If the monies obtained from the bond issue are used to finance riskier projects than expected, this wealth transfer would occur because share- holders would be the recipients of excess returns reaped from the riskier project. If the riskier project fails, the value of the firm will decline, and stockholders will not exercise their Option. Rather, they will allow the firm to become bankrupt and let creditors have the remaining assets. Creditors do attempt to restrain share- holders from behaving in this manner by placing covenants on debt issues and surveilling management's practices. However, if a firm begins to default and it can get court protection from foreclosure, Miller claimed the value of the call Option is restored and was not subjected to any creditor restrictions. He stated:7 "Gambling with other people's money would indeed be an artistic way of making a living, if only one could find the other people to supply the bankroll at the riskless rate of interest. In general one can't. But if a corporation is close to default, and if it can get court pro— tection from foreclosure, it may have a close equivalent." Miller argued that court protection will give value to a shareholder's worthless option on a Chapter 11 firm. He 7Merton Miller, "The Wealth Transfers of Bank- ruptcy: Some Illustrative Examples," Law and Contempo- rary Problems, p. 39-46 (1977). 41 showed that the expected value of the firm may be less if shareholders are allowed to continue operations versus an immediate liquidation of assets. This decrease occurs because the spread between the upper and lower boundaries of a firm's potential cash flow increases due to the under- taking of more aggressive projects. Miller also described this process for a viable firm that has become temporarily insolvent. Shareholders allowed by the courts to continue making policy decisions can attempt to increase their expected return by accepting riskier projects. Of course, bondholders will be forced to assume more risk, given their fixed level of return. When changes in shareholder risk levels occur, investors may not fully anticipate the resulting wealth transfer; rather, they may undervalue Chapter 11 stocks if they do not realize that the risk for the shareholder has declined. The most recent work in this area has been done by Golbe (1981), who analyzed this problem and discovered that shareholders of bankrupt firms prefer a higher variance of firm returns. He showed that increasing the variance of a firm's returns increased shareholder's expected returns, but not their downside risk. He analyzed the distribution of shareholder returns, and showed that they were truncated from below. Even though such share- holders are facing a total loss of their invested capital, their loss is limited. Thus, they have nothing more to lose, but everything to gain, by allowing management to 42 make aggressive policy decisions. Altman (1983) disagreed with Golbe's arguments when he stated: "I am not persuaded that managers increase their risk-taking as bankruptcy approaches. Debts pile up but not voluntarily, and if any- thing, conservative strategies may be adopted in order to perpetuate the business. It is usually the abnormal risks that are taken long before bankruptcy is apparent that are the fundamental causes of failure . . . creditor attempts to reduce their losses will also likely discourage these risky managerial attempts. . . ." In summary, Golbe's research provided evidence that shareholders can increase their return by increasing the riskiness of the firm's cash flow. One wonders whether a new investor purchasing shares in Chapter 11 firms under the continued control of prior management is receiving a bargain, given the changing risk/return characteristics of the debtor firm. It is difficult for the market to eval- uate the change in magnitude and direction Of this risk/ return relationship. This study provided an analysis of how the components of a firm's risk measure changed after a reorganization petition was filed. If a downward shift in a Chapter 11 firm's risk occurred, without an appro- priate market response, the possibility exists that new shareholders earned excess returns. Miller added support for the hypothesis that new shareholders of Chapter 11 firms could earn excess returns when he said:8 81bid., Miller, p. 44. 43 "The wealth transfers (and associated ineffi- ciencies) that we have been discussing have occurred in cases where the rules have changed, in effect, after the game has started. In particular, I have tried to show how permitting stockholders to claim court protection and thereby retain control of a corporation in default would amount to giving them a call Option at the expense of the creditors. Furthermore, the cost to the creditors may be because the arrangement may encourage wasteful business policies." 3. The Problem of Valuation According to Weston (1977), a bankrupt firm can be viewed as a new organization which must readjust to its new circumstances. New management, capital, and Objectives will alter its identity, and it will take time for this new firm to become fully efficient in the marketplace. For this reason, valuation of bankrupt firms is difficult and surrounded by a large degree of uncertainty. Weston stated:9 "The market lacks a clear view not only of the future values of the firm under various states of its environment, but also of those possible future states as they affect the particular firm." Even though the courts possess no better informa- tion on the expected future earnings of bankrupt firms than the marketplace, they have usually followed only SEC guidelines when determining the reorganization value. Meckling claimed that courts have "systematically over- estimated the value of the reorganized firm in order to 9Fred Weston, "Some Economic Fundamentals for an Analysis of Bankruptcy," Law and Contemporary Problems, V. 41 (Autumn, 1977), p. 63. 44 give something to junior creditors and stockholders." Schuchman (1977) outlined some reasons why the legal system believes in protecting the shareholders of bank— rupt firms: 1. Before the adoption of the original Chapter X and XI bankruptcy codes, the courts sought to protect junior creditors and small groups of unorganized investors believed to be treated inequitably by those creditors in a position to take advantage of liquidation. 2. It was argued that business investment would decline if shareholders were not granted any rights in the event of firm failure. Because future growth of equity investment would be inhibited, all businesses would suffer as a result. 3. The marketplace can not be used to determine a value for a failed firm because the courts believe that such an estimate would only reflect a specific point in time. The courts have stated they believe the SEC does a better job at incorporating future expected earnings estimates into the valuation equation than does the marketplace. 4. The courts claim that senior secured credi- tors possess the real power in reorganizations. If they do not assert their rights, this does not prove shareholders have too much power or that they abuse what power they retain. Thus, the courts dismiss the concept that the bankruptcy law itself allows a wealth transfer 45 to occur. 5. Finally, the courts claim no empirical evi- dence has been analyzed to determine whether such a wealth transfer to shareholders occurs, or if shareholders end up with more wealth than they should from a bankruptcy pro- ceeding. The above arguments are counter to basic corporate financial theory in terms of how a firm should best be valued. It is not important at this juncture to argue who is correct, but rather to determine if empirical evidence can be found that shows SEC valuation procedures allow shareholders to earn excess returns. Summary In summary, many arguments have been posited as to why shareholders may be able to earn abnormal returns by investing in reorganizing firms' securities. Models have been developed to facilitate the prediction of corporate failures. Tests have been done to determine the market's pre-bankruptcy reaction to share prices. Since much theoretical development has occurred in the formation of these arguments, but little has been done empirically to test their theories, this research attempted to show whether investing in Chapter 11 firms' stocks can be a viable investment strategy. CHAPTER III RESEARCH DESIGN AND METHODOLOGY Objectives The objectives of this research were two-fold. First, capital market data were analyzed to determine if a change occurred in the risk measures of Chapter 11 firms. Second, tests were done to ascertain whether in- vestors purchasing Chapter 11 firms' stocks after their filing dates could earn abnormal returns. By measuring for shifts in the market measures of risk and for the existence of post-filing abnormal returns for these firms' stocks, this study attempted to determine if the markets for Chapter 11 firms' stocks were efficient. The potential for abnormal returns is important for in- vestors who wish to purchase these stocks after their com- panies have filed for Chapter 11. In addition, the results of this study provide information for the bankruptcy courts. Since the courts must determine the value of a reorganizing firm, the effect of any risk change on a firm's discount rate could result in a change in the wealth distribution of a firm's assets. 46 47 Prior to 1978, studies cited in Chapter 2 provided evidence that the courts under-estimated a firm's discount rate when determining its value. It is also important for other claimants in the Chapter 11 proceedings to have more information about how court protection affects a firm's risk. This information may change their negotiating strategies with debtor firms' management, especially given the fact many firms are solvent when filing. Abnormal returns were not used in this case to determine how far in advance or how quickly market par- ticipants reacted to a firm's financial distress. Instead, tests were conducted to ascertain whether abnormal returns could be earned by new investors purchasing Chapter 11 stocks after filing. Such tests on the post-filing returns were tests of the semi-strong form of market efficiency. Assumptions Several assumptions were made regarding the treat- ment of Chapter 11 firms in this study. The first assump- tion was that a bankruptcy announcement effect existed and was properly incorporated into Chapter 11 firms' stock prices. Supporting documentation for this assumption comes from Altman (1969), Aharony, et. al. (1980), Altman and Brenner (1981), and Clark and Weinstein (1983). Using cumulative average residual analysis, their studies showed that prior to filing, the market for bankrupt stocks be- haved efficiently. It should be noted that all of the 48 above studies used data from firms that had filed under the old law. The second assumption made in this study was that Chapter 11 firms were considered to be different firms after filing. Viewing these firms as new entities pro- vided a way to test for any abnormal returns that existed after filing and for risk changes pre- to post-filing. The reasons for viewing firms from this perspective were: 1. When firms sought protection from cred- itors under Chapter 11, they were essen- tially asking the courts for permission to start over as new businesses. The law gave these firms time to reorganize their operation, financial structures, and corporate policies, without the interference of unpaid creditors demanding compensation. Many financially distressed firms emerged from reorganization proceedings smaller, having divested themselves of unprofitable divisions or subsidiaries, and with healthy balance sheets. Finally, it can be argued that Chapter 11 firms have quite different Objectives, finan- cial structures, and policy decisions than before filing. Weston (1977) supported the treatment of Chapter 11 firms in this manner, but evidence also comes from actual 49 reorganization cases. After filing, many Chapter 11 firms immediately began to sell off unprofitable divisions, to close down excess capacity, and to hire turn-around specialists with reputations for bringing firms out of reorganization. An example of a company that employed this strategy is Wickes of Santa Monica, California. Wickes, a diversified building materials retailer, filed for Chapter 11 protec- tion in April 1982 with $2 billion worth of liabilities and $2 billion in assets. Wickes brought in a turn- around specialist to serve as its new chairman and presi- dent. After filing Wickes sold its agricultural unit, closed down or sold 103 of its 3000 stores, and cut the staff by 10%. Another Chapter 11 firm, Lionel, followed a similar plan and, thus, was able to start over as a new firm.1 Sample Selection Collection of Data A sample of 63 firms was selected from a list of approximately 200 firms that filed for Chapter 11 between October 1, 1979, and December 31, 1982. This list, which represented the population of Chapter 11 firms existing 1Most of this information was obtained from current business periodicals dated around the time of the filings of these firms. Information was taken from Financial World (11/15/82), Forbes (3/15/82), and Dun's Business Month (September 1982). 50 during this time period, was obtained from the Corporate Reorganization Department of the Securities and Exchange Commission. The sample firms were specifically chosen from the population of Chapter 11 firms based on the following criteria: 1. Return or price data was publicly avail- able on a daily or monthly basis. 2. Financial information concerning the firm was obtainable from Moody's Industrial Manuals. One of the problems with studying Chapter 11 firms was data availability. The problem resulted from a lack of trading interest on the part of investors and the specu- lative nature of the stocks. Because of this problem the sample firms were divided into three groups, based on the frequency of the return data available for each firm (monthly, daily, or inconsistently available). This approach allowed an analysis to determine if the three groups had different financial characteristics. Sample Firms' Characteristics Prior to Chapter 11 filing, the sample firms' equities were listed on the New York Stock Exchange (NYSE), the American Stock Exchange (AMEX), or were traded over- the-counter (OTC). Pre-filing daily return data for the NYSE and AMEX-listed stocks was obtained from the Center for Research on Securities Prices (CRSP) tapes available 51 from the University of Chicago. Daily return data for the OTC stocks was computed from daily price quotes secured from the Daily OTC Stock Price Record. Because of the data availability problem for some firms the original 63 firm sample was divided into the following three groups: 1. The first group consisted of firms for which daily return or price data was avail- able after filing and two years before filing. After filing data was available either until December 31, 1983, a reorgan- ization plan was approved, a firm was acquired by another firm, or liquidation occurred. (Group I = 24 firms) The second group of firms had end-of-the- month return or price data available after filing and two years before filing. After filing data was again available either until December 31, 1983, a reorganization plan was approved, a firm was acquired by another firm, or liquidation occurred. These firms did not trade frequently enough after filing to be included in the daily sample. (Group II = 26 firms) The third group of thirteen firms was deleted from the study for the following reasons: 52 i) They did not have consistent monthly price or return data available after filing. ii) They had some monthly data available during this time period but were missing more than three consecutive data points. Post-filing return data availability was a problem for the study. All but three of the NYSE firms had trading suspended by the exchange. They began trading on the Pacific Stock Exchange (PSE) or over-the-counter. Daily price data for those trading on the PSE after filing was collected from daily editions of the Wall Street Journal, while daily OTC price data was collected from the Daily OTC Stock Price Record. For some firms, daily price data was not available from any source after filing.2 Monthly price data was collected from pink sheets provided by Merrill, Lynch, Fenner, and Pierce. Also for these firms, monthly CRSP return data was collected for the period before filing. Since the original data set was divided into three groups, it was of interest to determine if the groups had common characteristics. A cross-sectional, descriptive analysis was compiled on a number of variables to discern 2Daily data had to be available from the time a firm filed through its reorganization, acquisition, or liquidation or it was placed in Group II. 53 if any one variable appeared to distinguish one group from another. The variables chosen to discriminate between groups were exchange listing, size, industry membership, age, and a firm's financial solvency at filing. Tables I through V provide a summary of these variables by group membership. A detailed list of the firms in each group and their characteristics can also be found in Appendices A, B, and C. Each of the summary tables shows the percentages of firms within each group that possessed certain char- acteristics. Group I firms, for which the most return data were available, were mainly listed on the NYSE prior- to-filing (58%), had over 100 million dollars worth in assets at filing (50%), and were over ten years old (71%). On the other hand, Group III firms -- those deleted from the study because of data unavailability —- were mainly listed on the AMEX prior-to-filing (62%), and were rela- tively small in size with assets under $25 million (54%). The age of Group III firms varied with 54% over ten years Old and 31% under 5 years Old. Group II firms traded primarily on the AMEX and OTC prior-to-filing (46% and 42% respectively), were small to medium in size (35% and 46% respectively), and were almost all over ten years Old at the time of filing. Table VI summarizes these findings. Finally, neither industry membership or solvency at the time of filing differentiated firms within the groups. Over fifty percent of the firms in all three groups were 54 solvent at filing. This outcome was probably the result of a more lenient Chapter 11 code. To determine if any one variable differentiated better between firms than another, the Chi-square test was used. The results of this analysis are shown on Table VII. Only exchange listing and age at filing are variables that are significantly different at the .05 significance level for all three samples. Size, industry membership, and state of solvency at filing were not significant factors. The results of this analysis suggest that size, industry and state of solvency may not be linked to data avail- ability, which was the main factor for delineation of the sample into 3 groups. Age and exchange listing, on the other hand, were factors that could explain the differences in data availability. Firms which had been in existence for a long time and which traded on the NYSE originally had daily data available more often than smaller firms which traded on the OTC or AMEX. In conclusion, the results of the statistical tests on the data of Groups I and II can only be generalized to similar firms within each group. The smaller and younger firms of Group III were deleted from the study because of lack of data. This did not present a problem because most investors would not be able to purchase these securities as no market existed for the stocks after filing, often for a long period of time. 55 TABLE I EXCHANGE LISTING PRIOR TO FILING Group I Group II Group III NYSE 58% 12% 15% AMEX 0% 46% 62% OTC 42% 42% 23% TABLE II SIZE OF SAMPLE FIRMS AT FILING* Group I Group II. Group III Over $100m 50% 19% 15% $25 - 100m 25% 46% 31% Under $25m 25% 35% 54% *Size is measured by amount of total assets at the time a firm files for Chapter 11. 56 TABLE III INDUSTRY MEMBERSHIP Group I Group II Group III Manufacturing 42% 58% 46% Retailing 33% 15% 15% Service 25% 27% 39% TABLE IV AGE OF FIRM AT FILING* Group I Group II Group III Over 10 years 71% 96% 54% 5 - 10 years 12% 4% 15% 0 - 5 years 17% 0% 31% *Years since founding of firm or a previous reorganization. TABLE V STATE OF BALANCE SHEET OF FILING Group I Group II Group III Technically Insolvent 75% 88% 77% Bankrupt 25% 12% 23% 57 TABLE VI SUMMARY OF CHARACTERISTICS Group I Group II Group III Exchange NYSE AMEX , OTC AMEX (58%) (46%) (42%) (62%) Age (Years) Over 10 Over 10 Over 10 (71%) (96%) (54%) Under 5 (31%) TABLE VII TEST FOR DIFFERENCES AMONGZGROUPS CHI-SQUARE FACTORS (X ) AT A SIGNIFICANCE LEVEL OF .05 V Degrees P-Value/ x2 Decision* of Fractile Freedom Exchange .001/18.46 24.26 Reject 4 Size .10/7.78 8.90 Fail to Reject 4 Industry .50/3.36 3.38 Fail to Reject 4 Age .05/9.46 10.60 Reject 4 Solvency .50/1.39 2.44 Fail to Reject 2 *HO: There are no differences among groups on specific characteristics. H1: There are differences among groups on specific characteristics. 58 Calculation of Returns Except for the return data obtained from the CRSP tapes, all other data was in the form of stock prices. Daily and monthly returns were calculated as the change in price over a single holding period plus any dividends received. (Most Chapter 11 firms did not pay dividends within the time period of interest to this study because their earnings were negative for some time prior to filing as well as during reorganization.) When available, the asking price was used in the return calculations. This procedure was chosen for two reasons. First, this study was concerned with the returns new investors may earn on equity purchases, and second, such investors would pay the ask—price when purchasing these securities. For some Group I firms, trading was suspended for a time after filing and before trading began again on a different exchange. This occurrence required data- smoothing procedures to cover the problem time period. For days where price data was unavailable, both the missing day and the first day a stock resumed trading were deleted. The procedure was necessary since the first return calculation following the missing data was a multiple-day return, and as such, it could not be matched with a single-day market return. For firms that were liquidated during reorgani- zation or acquired by another firm, the final value to equityholders was included as its last closing price. 59 However, if a name change occurred, its return data was obtained under the new name. Given that the firms in the sample filed on earlier dates than others, some firms had more data avail- able. In addition, some firms emerged from reorganization, while others were still under court protection. Some firms had returns available under a different information set than others. Therefore, each firm's stock risk and return measures were analyzed individually rather than grouping them into portfolios. Although the ability to generalize this study's results was reduced, the literature indicates that Chapter 11 companies have such firm-specific problems, that it would be difficult to do more than describe the behavior of each firm's return and risk measures. Daily and monthly return data from the Standard and Poors 500 (S&P 500) Index was used as surrogates for the market portfolio returns. This data was Obtained from the CRSP tapes. Ninety-one-day Treasury bill rates were used as surrogates for the returns on the riskless asset. These rates were obtained from the Federal Reserve Bank of St. Louis. Estimates of Systematic Risk and Abnormal Returns The Two-Parameter Model The two-parameter model proposed by Sharpe (1964) and Lintner (1965) established a method of pricing risky assets in equilibrium. In ex ante form, the 60 Sharpe—Lintner model can be expressed as (1) E<fij> = Rf + IE(Rm) — Rf] ej where E(Rj) = the expected return on any asset j; E(Rm) = the expected return on the market portfolio; Rf = the expected return on a riskless asset; and Bj = cov :aé, am) 0 (Rm) Black (1972) varies the Sharpe-Lintner assumptions to develop models in which the intercept term is set equal to the expected return on a portfolio whose return is un- correlated with Rm. This portfolio, E(Rz), is called the zero-beta portfolio. Major empirical tests of this model using Rf and E(Rz) as the intercept terms were published by Blume and Friend (1970); Black, Jensen, and Scholes (1972); Miller and Scholes (1972); and Fama and McBeth (1973). With few exceptions, the studies agree that the intercept term, E(Rz), is significantly greater than Rf when Rf is approxi- mated using the 91-day Treasury bill rate. However, since the current research used the empirical form of this model to determine abnormal returns, it was not necessary to 61 distinguish between these two versions.3 This study also made use of the empirical results cited by Black, Jensen, and Scholes (1972), and Fama and McBeth (1973) which state that systematic risk, Bj, is the only risk measure affecting expected returns, E(Rj), and that a positive linear tradeoff exists between 8j and E(Rj).4 Assuming that equation (1) was the correct specifi- cation of the return generating process, we used its ex post form to determine the systematic risk of Chapter 11 firms' stocks across specified time periods. Estimates of a stock's abnormal returns were determined simultaneously. The ex post form can be written as: (2) Rjt ' th = (Rmt " th)8j + Ejt where Rjt = the actual return on security j at time t. R = the actual return on a riskless asset ft . at time t. R = the actual return on the market port- folio at time t. 8- = the systematic risk of security j. e. = residual term (having mean zero in the absence of abnormal returns). 3Copeland, Thomas E. and J. Fred Weston, Financial Theory and Corporate Policy (Addison-Wesley, 1980), pp. 180-181. 4A comprehensive summary of tests of the two- parameter model can be found in Jensen (1972) with a more recent critique available in Roll (1976). 62 Adding an additional term to equation (2) allows us to partition the cjt's into a random noise coefficient (ejt) and an abnormal performance measure (aj) such that: (3) (Rjt - th) = aj + (Rmt - th)8j + ejt where e. is independently, identically, and normally distributed with E(e = 0 -) = 2 J and VAR(ej) Ge . a. is a measure of a security's abnormal performance. The market equilibrium model described by equation (1) says that aj = 0 for all j. Thus estimates Of “j in equation (3) provide measures of "abnormal performance" given that (1) is the appropriate model of market equili- brium. Estimates of Bj and “j were derived simultaneously using equation 3 and were used to test for shifts in the risk of Chapter 11 firms' stocks and for the existence of abnormal returns across time. In addition, Markowitz's work (1959) on portfolio selection resulted in the concept that the only risk which investors will pay a premium to avoid is systematic risk. The total risk of any individual security can be parti- tioned into two parts, systematic and unsystematic such that Total = Systematic Unsystematic risk risk risk . 63 In equation form, this statement has been shown to be equivalent to:5 E 2 g 2 (4) OR. — 8. OR + 0“ J J m 3 where 2 OR. = the variance of a firm's stock 3 returns. 8. = the systematic risk of a firm's 3 stock. 2 0 OR = the variance of the market port- m folio's returns. a: = the unsystematic risk of a firm's j stock (the variance of the regression residuals from equation 3). Thus estimates of 0%. were also derived using equ- 3 ation 3 and used to test for shifts in the specific risk of Chapter 11 firms' stocks across time. Statistical Testing Procedures Regression analysis applied to the ex post form of Sharpe and Lintner's two-parameter model (equation 3) was used to estimate a sample firm's Bj and a.. The time periods used to run the regressions, however, were different for the firms in Group I and II. 5Fama, E. Foundations of Finance (Basic Books, New York, 1976), p. 70. 64 Estimation of Systematic Risk for Group I For Group I the parameters were estimated using ten and twenty data points. These data points were chosen from nonoverlapping time intervals. First, ordinary least squares regressions were run using (R. ) as the jt-th dependent variable and (Rmt-th) as the independent vari- able. Careful matching was done to ensure that the return of a stock on a certain calendar day matched exactly with the return on the S&P 500 and with the return on the appropriate 91-day Treasury Bill for the same day. were esti- The regression coefficients a. and B jt jt mated simultaneously, where t represents the time interval associated with each data point grouping. As a result of this procedure, for each firm in Group I, three time series were constructed: One repre- sented the behavior of a firm's systematic risk over time (Bjt), another represented the behavior of any abnormal returns accruing to a firm's shareholders over time (ajt), and the third represented the behavior of a firm's unsyste- matic risk over time (oéjt) . The regression results indicated that the Sharpe- ILintner model did not explain much of the variance in each individual firmfs returns. The coefficients of determin- artion were all less than 20% to as low as 0% for most of 'bhe regressions. In addition, there was no pattern to the results. The coefficients of determination fluctuated randomly over time. Ten or twenty data points per 65 regression did not appear to be enough to provide good predictability of returns. A visual inspection of the 8. and 02 series 3t ejt indicated that uncorrelated behavior was also evident. There appeared to be no discernible pattern for the Bjt's of each individual firm nor was there a pattern among firms' Bjt series. The Bjt's appeared to fluctuate inde- pendently from extremely high values to negative ones. The 2 Ejt was questionable because of the poor fit of the regression o '3 exhibited more stable behavior, but their relevance models. Plots of autocorrelation functions with lags up to 36 confirmed what was suggested by visual inspection. There was no autocorrelation present in either time series for any firm; however, the data did exhibit the properties of stationarity and constant variance over time. In conclusion, the Bjt and oéjt series for each firm behaved in an uncorrelated manner. There was no consistent behavior over time for individual firms or between firms. It was not possible to discern whether a certain characteristic caused a specific group of firms' time series to behave in a consistent manner. Alternate Risk Measures for Group I Given the lack of conclusive results stemming from the use of systematic risk as the only measure of Chapter 11 firms' risk, risk premia over time were used as an 66 alternate proxy for risk. Risk premia were defined as the difference between a firm's return and the return on the S&P 500. One of the reasons the previous regression results were unsatisfactory was that the interval of time for which the Bjt estimates were derived was too short. Changes in the variance of returns was not taken into account. To compensate for these problems, F-tests were done to deter- mine where the variance of the risk premia time series changed significantly. The time series for each individual firm was divided into phases as illustrated below: I II III IV filing Divisions were based on the occurrence of a change in vari- ance not on specific calendar days. Visual inspection was used to determine appropriate dividing points between phases. Then a F-test was done to determine if the vari- ances were significantly different. If not, then the phases were combined into one larger phase. Phases were combined until groups of data across time were identified that had a change in variance. If and when the phase- variances were different, then future testing was done. For each firm the number and time length of phases differed depending on the return characteristics of a specific firm. 67 Before parametric statistical testing can be done to see if there are differences between phases, a number of assumptions must be met. To use t-tests to determine differences between means, the data must display normal- ity, homogeneity of variance, and independence. Based on Fama's work, data was assumed to be normally dis- tributed. Homogeneity of variance existed by virtue of the method used in dividing the data into phases. To test for independence, autocorrelation functions were cal- culated for each phase for each firm. If the return data within a phase was not autocorrelated then t-tests were done to determine if the mean risk premium changed between two sequential phases. If, however, autocorrelation of a particular phase was shown to exist, the data was modelled using time series methodology.6 For most phases displaying an autocorrelation problem, adding a moving average term of order 1 to the time series removed the autocorrelation. The model's constant term, which is equivalent to the mean of the time series, was then compared, using t-tests, to the next phase's mean risk premium. Given that the risk premia analysis indicated whether a stock earned more or less than the market, was important to determine the systematic risk component for each phase. This was done to determine how the average 6Box and Jenkins, Time Series Analysis, Forecasting and Control (San Francisco: Holden Day, 1970). 68 systematic risk,8j, Of each phase differed from the syste- matic risk of the market, which is equal to 1.0. The Ej of each phase was estimated using the mar- ket model: R. = a. + b. R + e. 3 J J m 3 where Rj = return on stock bj = estimate of systematic risk Rm = return of S&P 500 ej = residuals. T-tests were used to test for a significant change in bj's over time. Group II Estimate of Systematic Risk For Group II, equation 3 was used to estimate BjP'Oéjp' and ajp' where p is equal to the estimation period. Since the data for each firm in this group is monthly, ordinary least squares were run using approxi- mately two years worth of data. The time periods of interest were: 1. The 24 months before a firm filed for Chapter 11 (Period I). 2. The months covering the interval from filing to December 31, 1983, or when the firm was liquidated, merged into another firm, or reorganized (Period II). 69 Thus for each firm a B. , a. , and a 02 were JP JP ejp estimated for Period I and II. In addition, the variance of returns, 0%, each firm for Period I and II. and a mean return, ER, were calculated for These parameters can be defined as the average 5 stematic risk, . , y B3p average unsystematic risk, 6:. , during each time period. JP Statistical testing was done to see how each para- average abnormal return, ajp’ and meter differed between Period I and II for each firm in Group II. CHAPTER IV RESEARCH RESULTS Summary of Results for Groups I and II Similar analytical techniques were applied to Groups I and II. The basic difference between the two groups is that Group I firms had daily data available for the time periods in question while Group II had only monthly data available. First, t-tests were used to test for a change in average risk premia from before to after filing. In addi- tion, F-tests were used to test for a change in the vari- ance of the premia from before to after filing. In sum- mary, 62.5% of Group I and 85% of Group II did not experi- ence a statistically significant change in risk premia pre- to post-filing at the .05 level. All of the firms in Group I showed an increase in premia variance after filing, while only 58% of Group II firms had a similar experience. In absolute terms, however, 50% of Group I and 62% of Group II firms earned more than the market return on average after filing. Next, t-tests were used to test for a change in average systematic risk pre- to post-filing. It was 70 71 determined that 96% of Group I and 85% of Group II experi- enced a change. Two subgroups of the original samples, 79% of Group I and 61% of Group II firms, experienced a decline in systematic risk after filing. A concern with the interpretation of this result was that the explanatory power of the regressions for both groups declined signifi- cantly after filing. The market and capital asset pricing models did a poor job in explaining the return of Chapter 11 firms after filing. Finally, using the capital asset pricing model to model the return generating process of these firms, it was not possible for investors purchasing these securities after filing to earn abnormal returns. A detailed description of the above results with their testing procedures follows. Analysis of Group I Daily Return Data The first part of this research involved analyzing the daily return data of Group I firms. A time series of daily risk premia was derived for each firm by subtracting the daily market return from its corresponding daily firm return. Each risk premia time series was then divided into phases based on when a change in variance occurred. F-tests were used to identify the appropriate dividing points between phases for each time series. Tests were done to determine if differences existed in the mean daily risk premia and the average systematic risk from before 72 to immediately following Chapter 11 filing. Also, tests were done to determine if abnormal returns existed for investors purchasing these stocks post-filing. Discussion of Risk Premia Time Series Plots Visual inspection of each firm's time series plots provided some evidence of consistent risk premia behavior between firms. Prior-to-filing, there was a similar pat- tern to the movement of each stock's risk premia. The risk premia declined steadily, until at some point in time, the premia dropped 30-50% within one day. This drop in premia occurred at different points in time relative to filing for each firm. Another 30-50% drop in risk premia occurred again for all stocks within a few days after filing. A rationale for this behavior is that when a firm's financial woes first become public, before the actual filing, the market reacts by selling the stock. Institu- tional investors are sensitive to 'bad news', especially if it results in a dividend cut or elimination. Altman (1969) found the same type of drop in post- filing return. In ides study, he discovered that the mar- ket seemed surprised when a firm finally filed for Chapter 11. It is also possible that the market overreacts to a firm's filing announcement. For every firm in Group I, the risk premia in- creased between 30-50% within a week after filing. Although it is plausible to assume that the investors 73 in these firms were speculators, this return spike after filing may indicate the market overreacted to the filing announcement and needed time to reevaluate the financial fundamentals of each firm. After filing, the variance of the risk premia of all firms increased. This high variance may have occurred as the market attempted to assess what kind of reorganiza- tion plan would be worked out by management in court. Any news caused a movement in price. Analysis Of Risk Premia Over Time Using the procedure described in Chapter 3, the mean risk premium for each time series phase was calcu- lated. The parametric t-test was used to determine if a significant difference occurred between phases. Given that the data in each phase displayed the properties of independence and constant variance, the t-test was appropriate. The results indicated in Table VIII show that 62.5% of the firms' risk premia did not change significantly at the .05 level over time, while 37.5% of the firms exhibited a change. Using the Chi-square test, three characteristics -- exchange listing, age, and size -- were tested to see if significant differences existed between these two sub- groups: those with a change in risk premia and those without. It was determined that none of the character- istics differentiated between subsamples. The percentages 74 of Group I firms in each category are listed on Table IX. Although 62.5% of Group I firms did not have a significant change in average risk premium over time, this finding may not be as enlightening for investors as dis- covering what the absolute change in risk premia was from before to after filing. Tables X and XI illustrate the timing and direction of the absolute changes in risk premia around firm filing dates. For both subgroups, firms had a decline in risk premia from two months before filing to one month after filing. Seventy-eight percent of the group with a significant change in risk premia and 87% of the group with no significant change experienced a decline in absolute risk premia. However, after filing the risk premia was negative for only 50% of all of Group I firms. Estimates of Systematic Risk Over Time Estimates of average systematic risk were derived using the market model. One regression was run per phase for each individual firm. The resulting Bj were tested for significant differences at the .05 level pre- to post- filing using t-tests. In addition, t-tests were used to determine what percentage of firms had Bj estimates sig- nificantly different from zero at the .05 level before and after filing. Table XII illustrates that before filing 62.5% of Group I firms had 8. J at the .05 level, while after filing only 17% had Bj's 's significantly different from zero 75 different from zero. Overall Bj's appear to have declined from before to after filing. This result must be qualified by the fact that the fit of the market model to the return data was very poor after filing. Table XIII summarizes whether a change, if any, took place in Bj and in what direction it occurred. The results are quite conclusive in that 96% of the firms exhibited a significant change pre- to post-filing. Among Group I firms, 79% had a de- crease in Bj while 17% had an increase in Bj post-filing. In summary, it appears that the majority of Group I firms experienced a decline in systematic risk after filing. However, the variance of return for all the firms in Group I increased after filing. If these firms were included in an investor's well-diversified portfolio, and if systematic risk was the only relevant risk measure then it might be possible to say that the risk of Chapter 11 firms with Group I characteristics declined after filing. Otherwise, all that can be said is that the total risk of the firms increased after filing. Another interesting result is that the fit of the market model to Group I firms' data was better using before filing data than after. Statistical analysis indicated that the explanatory power of the regressions pre-filing were much higher than post-filing. F-tests were performed and yielded the following results which are displayed in Table XIV. 76 1. Before filing, the market return was a good predictor of a firm's stock return for 83% of the firms in Group I. 2. After filing, however, the market return was a poor explanatory variable for 75% of Group I's stock returns. The market model was less able to explain a Chapter 11 firm's stock return post-filing than it was pre- filing, due to changes that Chapter 11 firms experienced at that time. The next chapter will outline some possible reasons why the market model was less able to explain a Chapter 11 firm's stock returns after filing. Presence of Abnormal Return Behavior The presence of abnormal returns for these firms post-filing is shown in Table XV. The percentage of firms with abnormal returns not significantly different from zero post-filing was 100%. Given that the two-parameter model correctly predicted the returns of the firms and that Bj was an appropriate risk measure, Group I firms did not earn abnormal returns, even though their systematic risk declined. 77 TABLE VIII GROUP I CHANGE IN RISK PREMIA OVER TIME Percentage of Firms Tested Using t-ratios at a Sig- nificance Level Of .05 No Significant Significant Change Change 62.5% 37.5% TABLE IX GROUP I CHARACTERISTICS OF RISK-PREMIA SUBGROUPS Percentage of Firms Exchange Change No Change NYSE 67% 53% OTC 33% 47% Size y__ Change No Change Over $100m 56% 47% $25-$100m 33% 20% Under $25m 11% 33% Age Change No Change Over 10 years 67% 73% 5 - 10 years 22% 7% Under 5 years 11% 20% 78 TABLE X GROUP I CHANGE IN RISK PREMIA TIMING AND DIRECTION OF ABSOLUTE CHANGE Percentage of Firms Timin Direction 9 Increase Decrease At Filing 11% 44% One Month Before 11% 11% Two Months Before .;: ‘22% Total 22% 77% Conclusions: 1. 56% were earning less than the market after filing. 2. 44% were earning more than the market after filing. NO SIGNIFICANT CHANGE IN RISK PREMIA TABLE XI GROUP I TIMING AND DIRECTION OF ABSOLUTE CHANGE Percentage of Firms Timing D1rection Increase Decrease At Filing 6.5% 40% One Month Before - 27% Two Months Before 6.5% 13% One Month After - _1% Total 13% 87% Conclusions: 1. 40% were earning less than the market after filing. 2. 6.5% were earning the same as the market. 3. 53% were earning more than the market. 79 TABLE XII GROUP I PERCENTAGE or FIRMS WITH 8. ESTIMATES SIGNIFICANTLY DIFFERENT JFROM ZERO Tested Using t-ratios at a Significance Level of .05 Pre-filing Post-filing Different From Zero 62.5% 17% Not Different From Zero 37.5% 83% TABLE XIII GROUP I CHANGE IN AVERAGE B. PRE- TO POST-FILINGj Tested Using t-ratios at a Significance Level of .05 Significant Change Occurred Direction Yes No Increase 17% Decrease 79% Total 96% 4% 80 TABLE XIV GROUP I EXPLANATORY POWER OF REGRESSIONS SIGNIFICANT F-TEST Percentage of Firms with Models Having Significant Explanatory Power Significance Level of .05 Before Filing After Filing Significant 83% 25% Insignificant 17% 75% TABLE XV GROUP I PERCENTAGE OF FIRMS WITH ABNORMAL RETURNS SIGNIFICANTLY DIFFERENT FROM ZERO Tested Using t-ratios at a Significance Level of .05 Pre-Filing Post-Filing Different From Zero Positive 0 0 Negative 21% 0 Not Different From Zero 79% 100% 81 Analysis of Group II Monthly Return Data The second part of this research involved analyzing the monthly return data of the 26 firms in Group II. All statistical testing was done on a pre/post-filing basis. Average measures of specific parameters were calculated using pre-filing return data and were compared to average measures of the same parameters calculated using post- filing return data. The following results were obtained from this analysis. Discussion of Risk Premia Time Series Plots Visual inspection of the time series plots for each firm's stock risk premia did not provide evi- dence of consistent behavior among these firms. However, some general comments about each firm's pre- and post- filing mean risk premia and variance of premia follow: 1. The variance of the return was higher post-filing for 58% of the firms when com- pared to their pre-filing variances. 2. For 96% of the firms, the average risk premium over the two years prior to filing was negative, while the average premium after filing was positive for 55% of the firms. 3. In addition, 69% of the firms had sub- stantially increased risk premia during the first month after filing. The durability of the increase over time varied among 82 firms,with some exhibiting a long-run increase while others began to decrease by the second month after filing. Of the four firms that had decreased premia during the first month after filing, two showed substantially increased risk premia during the second month while one had increased premia beginning three months after filing. Analysis of Risk Premia Over Time Using the procedure described in Chapter 3, the mean risk premium was calculated for the time periods be- fore and after filing. The parametric t-test was used to determine if a significant difference occurred from before to after filing. The results are shown in Table XVI and indicate that 85% of the firms did not display a significant change in risk premia. Only 15% of the firms had a change in risk premia, or four of the 26 firms examined. As with Group I firms, the Chi-square test was done to see if significant differences at the .05 level existed between the charac- teristics of the group with a change in premia and the characteristics of the group without a change. It was again determined that neither exchange listing, size, nor age of firm at filing differentiated between the groups. Table XVII illustrates the breakdown of characteristics. When looking at the absolute risk premia over the 83 market return, most of the firms earned more than the mar- ket after filing. Sixty-two percent of Group II firms earned more than the market on average after they filed. Tables XVIII and XIV show that 100% of the group with a change in risk premia had an increase in risk premia post- filing, while 77% of the group with no significant change had an increase in absolute premia. Estimation of Systematic Risk Using Sharpe and Lintner's two-Parameter model, systematic risk and abnormal return estimates were derived for each firm pre- and post-filing. A summary of the re- sults are presented in Tables XX, XXI, XXII, and XXIII. Two regressions were run using each firm's return data. Abnormal return and systematic risk estimates were obtained for the pre- and post-filing time periods. Pre- filing estimates were obtained for the two-year time period prior-to-filing. These parameter estimates were compared to post-filing estimates to determine if a significant difference occurred around an individual firm's Chapter 11 filing date. The explanatory power of the regression models was poor. The low values of the coefficients of determination (less than 25%) for many firms indicated that the market risk premium did not eXplain much of the variance in indi- vidual firm's risk premia. The Durbin-Watson statistic for most regressions was over 1.80. The evidence 84 indicated that in only two cases was there possible auto- correlation among residuals because of a Durbin Watson statistic below 1.80. Therefore, the existence of corre- lated residuals was not a valid explanation for the poor fitting regression models. Analysis of the systematic risk estimates shown on Table XX indicates that the percentage of firms with Bj estimates significantly different from zero altered con- siderably pre- to post-filing. Prior to filing 54% of Group II firms had systematic risk estimates statistically different from zero at the .05 level, while after filing 85% were not significantly different from zero at the .05 level. This change in the Bj estimates.may again be the result of poorly fitting regression models. The t-ratio was used as a test statistic to deter- mine if a change in the Bj estimates occurred pre- to post- filing. Table XXI highlights these results. Sixty- one percent of the firms showed a significant decrease at the .05 level in systematic risk. Twenty-three percent exhibited an increase in their estimates and 15% showed no change. The results of the analysis of Group II firms is similar to the analysis of Group I firms in that the majority of Group II firms experienced a decline in syste- matic risk after filing. However, Group II firms did not all display an increase in return variance after filing. Instead, the variance of return increased for 58% of the 85 firms, while it decreased for 42%. Of the firms with a decline in variance of return after filing, 73% also experienced a decline in systematic risk. In conclusion, the overall risk of the returns of 8 out of 26 in the sample declined after filing. Finally, as with Group I firms, the fit Of the regression models shown in Table XXII for Group II was better before than after filing. It appears that for firms engaged in the reorganization process that the mar- ket model and the capital asset pricing model are not very good models to use in predicting firm return after filing. Presence of Abnormal Returns The presence of abnormal returns for these firms post-filing is shown in Table XXIII. The percentage of firms with abnormal returns not significantly different from zero post-filing was 88.5%. While actual returns increased during the first month after filing, for many firms it does not appear to have been enough to compensate for the risk these firms faced. These statements assume the legitimacy of the two-parameter model and 8. as an 3 appropriate risk measure. 86 TABLE XVI GROUP II CHANGE IN RISK PREMIA OVER TIME Percentage of Firms Tested Using t-ratios at a Significance Level of .05 No Significant Significant Change Change 85% 15% TABLE XVII GROUP II CHARACTERISTICS OF RISK-PREMIA SUBGROUPS Percentage of Firms Exchange Change No Change NYSE 25% 9% AMEX 25% 50% OTC 50% 41% Size Change No Change Over $100m 25% 18% $25-100m 25% 50% Under $25m 50% 32% Age Change No Change Over 10 years 75% 100% 5-10 years 25% - Under 5 years 87 TABLE XVIII GROUP II CHANGE IN AVERAGE RISK PREMIA TIMING AND DIRECTION OF ABSOLUTE CHANGE Percentage of Firms Direction Increase Decrease Timing Before to After 100% 0% 100% were earning more than the market after filing. TABLE XIX GROUP II NO SIGNIFICANT CHANGE IN RISK PREMIA TIMING AND DIRECTION OF ABSOLUTE CHANGE Percentage of Firms Direction Increase Decrease Timing Before to After 77% 23% 55% were earning more than the market after filing. 88 TABLE XX GROUP II PERCENTAGE OF FIRMS WITH 8. ESTIMATES SIGNIFICANTLY DIFFERENT FROM ZERO Tested Using t-ratios at a Significance Level of .05 Pre-filing Post-filing Different From Zero 54% 15% Not Different from Zero 46% 85% TABLE XXI GROUP II CHANGE IN AVERAGE B. PRE-TO POST-FILING Tested Using t-ratios at a Sign1f1cance Level of .05 Significant Change Occurred Direction Yes No Increase 23% Decrease 62% ____ 85% 15% Total 89 TABLE XXII GROUP II EXPLANATORY POWER OF REGRESSIONS Percentage of Firms With Models Having Significant Explanatory Power Significance Level of .05 Before Filing After Filing Significant 77% 19% Insignificant 23% 81% TABLE XXIII GROUP II PERCENTAGE OF FIRMS WITH ABNORMAL RETURNS SIGNIFICANTLY DIFFERENT FROM ZERO Tested Using t-ratios at a Significance Level of .05 Pre-Filing Post-Filing Different From Zero Positive 0 3.8% Negative 38.5% 7.7% Not Different From Zero 61.5% 88.5% CHAPTER V SUMMARY AND CONCLUSIONS This investigation scrutinized the market risk and return characteristics of firms that filed under Chapter 11. These characteristics were examined for the two year period before filing and for a certain period after filing. The period of interest after filing covered the time interval from filing to the point at which a firm either was reorganized, liquidated, or purchased by another firm. If none Of these events occurred by December 31, 1983, then that date was used as the ending date of the after filing interval. The central objective of this research was to determine if the securities market efficiently priced the common stock of these firms after they petitioned the court for protection from creditors. This study differed from earlier studies in several ways. Past studies on bankrupt firms had concentrated on determining if the securities market reacted efficiently to the announcement of a bankruptcy under Chapter X of the Chandler Act. The studies indicated that the market for the common stocks of bankrupt firms were in fact efficiently priced prior to their filing for Chapter X 90 91 relief. In this investigation, the sample of firms was taken from those filing under the new Chapter 11 bank- ruptcy code that became law in 1979. Tests were carried out for changes in the market measures of risk and return of the sample firms. These tests were used to determine if there was a change in risk from the pre-filing period to the post-filing period and to ascertain the existence of abnormal returns. Previous studies had ignored post- filing changes. Analyses terminated at the time a firm entered bankruptcy protection under Chapter X. One possible reason for the absence of post-filing studies up to now was the lack of post-filing data. In this study, the data for the time period after filing was compiled from the Wall Street Journal, Merrill Lynch's pink sheets, and from Standard and Poor's Daily OTC StockPrice Record. Information derived from this research will be of interest to investors who are considering the purchase of common stocks of Chapter 11 firms. In their search for undervalued equities, some analysts have recommended the common stocks of Chapter 11 firms which in their Opinion have a high probability of recovery after reorganization. The conclusions offered in this study should allow analysts to make more informed judgments. In addition, the study will be of interest to the parties involved in the reorganization process. Given that 92 the court must determine the value of a Chapter 11 firm, it should be interested in finding out what effect reorgan- ization might have on the firm's cost of capital. How a firm's risk changes after filing will also be of interest to creditors. If the risk increases, then the present value per dollar of a firm's future earnings will be less. If the risk decreases, it will be greater. The larger the present value of future earnings, the more likely it is that the shareholders will be allowed to share in the dis- tribution of assets at the expense of creditors. In this paper, two groups of Chapter 11 firms were analyzed. Group I firms, for which daily data was avail- able, were mainly old firms originally listed on the New York Stock Exchange (NYSE). Group II firms, for which monthly data was available, were firms which mainly traded on the American Stock Exchange (AMEX) and over the counter (OTC) before filing. The results summarized below apply to each group of securities analyzed. 1. All firms experienced a decline in stock price, and thus return, around the time of Chapter 11 filing. After filing, all firms experienced a large jump in price, which resulted in an upward spike in the return profile when the returns of each firm were shown as a time series. This behavior did not, however, occur in a consistent pattern for each firm. The spike in the returns appeared at different points for each firm, and may have been the 93 result Of some specific 'news' announcement about the re- organization proceedings. 2. Tables XIII and XXI show that after filing 79% of Group I and 60% of Group II firms experienced a decline in systematic risk. In terms of overall risk, however, 100% of Group I firms had an increase in the variance of their stock returns after filing, while only 58% of Group II firms' stock returns experienced such a change.1 3. The poor explanatory power of the regressions on the post-filing data of both groups (demonstrated in Tables XIV and XXII) implies that either the model of the return generating process was misspecified or that the returns of the Chapter 11 firms were largely uncorrelated with the market. If the model was misspecified then the variance of the returns of a firm would be a better measure of the risk of a firm than its systematic risk measure. Tables XII and XX show that, after filing, the systematic risk estimates were zero for 83% of Group I firms and 85% of Group II firms. The fact that the value of the systematic risk measures was zero could mean that both the market model and the two-parameter model misspecified the return generating process of these firms during the reorganization process. On the other hand, it could be that the returns of Chapter 11 firms were not correlated with the market —v 1See Appendices D and J. 94 and have indeed no systematic risk. If the above asser- tion were true, then the stocks of Chapter 11 firms would be valuable assets, whgg added to a portfolio. They would have had the same risk as a riskless asset, but would have earned more than the market. In this research, 50% of Group I stocks and 62% of Group II stocks earned more than the market after filing. 4. Tables XV and XXIII show the results of using the two-parameter model to measure abnormal return per- formance. One hundred percent of Group I and 89% of Group II's firms did not experience abnormal returns. 5. Tables X, XI, XVIII, and XIX show the results of using the market return as an alternate measure of rela- tive risk. The risk premia performance before and after filing was examined. The tables show that 50% Of the Group I firms and 62% of the Group II firms earned more on average than the market return. In summary, the results of this investigation showed that all of the Chapter 11 firms exhibited a large positive return spike during reorganization. The existence of these spikes indicated that there was the potential for investors to earn large positive returns if they purchased the stocks of these firms after a Chapter 11 petition was filed. However, this study did not provide a guide to selecting the stocks that would earn these large absolute returns, since no consistent patterns existed among firms as to when this behavior occurred. Also, this study was 95 not able to determine if these returns were enough to compensate investors for their risks. In terms of describing the risk of the returns of a Chapter 11 stocks, the analysis of the after filing risk premia indicated that on average Chapter 11 firms earned more than the Standard and Poor's 500 Index. Also, the study revealed that after filing the variance of these stock returns increased for 100% of Group I and 58% of Group II firms. Interpretation of the systematic risk measures was more difficult. Since these risk estimates were equal to zero for most of the Chapter 11 firms, either the returns of these firms could not be modelled with the two-parameter and market models or they represented an asset that had returns which were uncorrelated with the market. Implications of the Results. The results of the study did not resolve the issue of whether the market for the common stocks of Chapter 11 firms were priced efficiently. The statistical analysis indicated that the common stocks did not earn abnormal returns after the firm filed for Chapter 11 protection. However, the absolute returns of the stocks after filing were greater than the market return in most cases. The difficulty in making statements about these returns was in determining the appropriate risk adjusted discount rate. The finding that the stocks did not provide after- 96 filing abnormal returns was consistent with the assump- tion of market efficiency. Any additional return that the common stocks earned over the market return may have been compensation for their increased risk. Since the increase in variance of returns was a measure of the increased total risk, changes in both the systematic and unsystematic risk could have occurred. The fact that the.measure of systematic risk declined in some cases after filing meant that the measure of unsystematic risk may have increased (depending on how the variance of the market return changed.) The unsystematic risk was a measure of firm specific risk and included reorganization risk. This reorganization risk resulted from the uncer- tainty surrounding the factors engendered by the judicial system and thus beyond the control of management. One of the factors that created reorganization risk was the negotiations between creditors, management, and share- holders who were searching for an acceptable compromise that would result in the liquidation of the assets of the firm. Another factor in the composition of reorganization risk was the uncertainty of how the courts would apply the new law to each firm's situation. The unresolved question during a court supervised reorganization was, "Who will benefit the most: creditors, shareholders, or some other constituent?" The uncertainty surrounding the distribution of the wealth of a firm will affect the way individual claimants perceive the risk of the firm. 97 However, the question whether the market overcom- pensates for the increased risk of these stocks remains unanswered. The market perceptions of both the underlying strength of a Chapter 11 firm at filing and the expected treatment by the court were reflected in the variance of the returns. But the large positive return, which was evi- dent in the return time series for all of these firms, indi- cated that the market may have undervalued these stocks by applying a discount rate that was too large when discounting their future cash flows. In summary, this investigation tested for changes in the risk and returns of firms filing under the new Chapter 11 code. Depending on the risk measure used, the research provided evidence of an increase or decrease in risk for firms filing under Chapter 11. It showed that investors who timed their purchases of Chapter 11 stocks could earn large absolute positive returns. One explana- tion for this large absolute return was that investors were not able to accurately judge the reorganization risk of these firms and accordingly the stocks were underpriced. Limitation of the Study The capital asset pricing and market models were used for testing purposes. Any study using these models to estimate systematic risk faces the possibility that the models misspecify the return generating process and there- fore confound the results. Within the context of this study the models fit the pre-filing data better than the post-filing data. 98 Suggestions for Future Research The main focus of future research might well be in two areas. One Objective of research would be the development of a model that could be used to predict which firms will emerge from reorganization successfully. Another goal for future research would be to determine what other variables could be incorporated into the capital asset pricing model to account more accurately for the risk of firms in reorganization. Much of the bankruptcy literature to date concen- trated on classifying firms by their probability of going bankrupt. This type of analysis could be applied to re- organizing firms to determine which variables seem to pre- dict a firm's successful emergence from its court pro- ceedings. Some factors to incorporate into such a model would include age at the time of filing, asset size at the time of filing, exchange listing, amount of debt, and other financial ratios. After studying these Chapter 11 firms, one con- cludes that additional exogenous variables besides the market return should be included in the capital asset pricing model as risk measures. Variables need to be identified that can be used as surrogates for reorganiza- tion risk. Given the tradeoff that exists during the court proceedings between bondholders and equityholders, the one variable that stands out as most important is the debt of a firm. Because of the prior claim of debt one 99 would expect firms with higher debt loads to be riskier. and provide less returns to Chapter 11 shareholders. How- ever, shareholders of firms with large amounts of debt may actually be the winners in a Chapter 11 filing since the court valuation process may result in a wealth transfer to shareholders from creditors. Research that sheds more light on how these markets operate should be welcome. 100 REFERENCES Aharony, J., Jones, Charles P. and Swary, Itzhak. "An Analysis Of Risk and Return Characteristics of Corporate Bankruptcy Using Capital Market Data." Journal of Finance, Vol. 35 (September 1980), pp. 1001-1016. Altman, E. I. "Discussion on Behavior of Firms in Finan- cial Distress." Journal of Finance, Vol. 38 (May 1983), pp. 517-522. The Financial Handbook, 1981, pp. 35.3-35.47. . Corporate Bankruptcy in America. Lexington, Mass.: Health Lexington Books, 1971. . "Corporate Bankruptcy Potential, Stockholder Returns and Share Valuation." Journal of Finance, Vol. 24 (December 1969), pp. 887-900. . "Bankrupt Firms' Equity Securities as an Investment Alternative." Financial Analysts Journal, Vol. 25 (July-August 1969), pp. 129-133. "Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy." Journal of Finance, Vol. 23 (September 1968), pp. 589-609. Altman, E. and Brenner, Menachem. "Information Effects and Stock Market Response to Signs of Firm Deterioration." Journal of Finance and Quantita- tive Analysis, Vol. 16 (1981), pp. 35-51. Argenti, John. Corporate Collapse: The Causes and Symptoms. London: McGraw-Hill, 1976. Asquith, Paul and Kim, E. Han. "The Impact of Merger Bids on the Participating Firms; Security Holders." Journal of Finance, Vol. 37 (December 1982), pp. 1209-1228. Baskin, E. and Crooch, G. "Historical Rates Of Return on Investments in Flat Bonds." Financial Analyst Journal, Vol. 24 (November-December 1968), pp. 95- 97. Banz, Rolf W. "The Relationship Between Return and Market Value of Common Stocks," Journal of Financial Economics, Vol. 9 (March 1981), pp. 5-18. 101 Bernstein, P. W. "Who Buys Corporate Losers," Fortune (January 26, 1981), p. 60. Black, F. "Capital Market Equilibrium with Restricted Borrowing," Journal of Business, Vol. 45 (July 1972), PP. 444-455. Black, F., Jensen, M. C., and Scholes, M. "The Capital Asset Pricing Model: Some Empirical Tests," reprinted in M. C. Jensen, ed., Studies in the Theory of Capital Markets. New YorE? Praeger, 1972, pp. 79-124. Blume, M. and Friend, I. "A New Look at the Capital Asset Pricing Model," Journal of Finance, Vol. 29 (March 1973). pp. 19-34. Brown, S. J. and Warner, J. B. "Measuring Security Price Performance." Journal of Financial Economics, Vol. 8 (September 1980): pp. 205-258. Box, G. E. P., and Jenkins, G. M. Time Series Analysis, Forecasting and Control. San Francisco: Holden Day, Inc., 1970. Box, G. E. P. and Tiao, G. C. "Intervention Analysis with Applications to Economic and Environmental Problems," Journal of the American Statistical Association, Vol. 70 (March 1975), pp. 70-79. Calkins, Francis. "Corporate Reorganization Under Chapter X: A Post-Mortem." Journal of Finance, Vol. 3 (June 1948), pp. 19-28. Clark, T. A. and Weinstein, M. I. "The Behavior of the Common Stock of Bankrupt Firms." Journal of Finance, Vol. 38 (May 1983), pp. 489-504. Copeland, R. E., and Weston, J. F. Financial Theory and Corporate Policy. Reading, Massachusetts: Addison-Wesley Publishing Company, 1981. Corotto, A. F. and Picard, I. H. "Business Reorganization Under the Bankruptcy Reform Act of 1978 -- A New Approach to Investor Protections and the Role of the SEC." DePaul Law Review, Vol. 28 (Summer 1979), pp. 961-1006. Deakin, E. B. "Accounting Reports, Policy Intervention and the Behavior of Securities Returns." The Accounting Review, Vol. 51 (July 1976), pp. 590- 603. 102 Dewing, Arthur A. Corporation Finance. New York: Ronald Press Company, 1931. Doherty, E. J. "Betting on Bankruptcies." Financial World (November 15, 1983), P. 47. Dun and Bradstreet. Failure Record, 1982. Dun's Business Monthly, "Business Bankruptcies Soar." April 1982, p. 29. Dun's Business Monthly, "How to Profit When Firms GO Bust." September 1982, p. 37. Fama, E. Foundations of Finance, New York: Basic Books, 1976. Fama, E. and McBeth, J. "Risk Return and Equilibrium: Empirical Tests," Journal of Political Economy (May/June 1973): pp. 607-636. Fischer, L. and Lorie, J. H. "Rates of Return on Invest- ments in Common Stock -- The Year-by-Year Record, 1926-1965." Journal of Business, Vol. 41 (July 1968), PP. 291-316. Greene, Richard and Bernstein, Aaron. "Congratulations You're Bankrupt. Forbes (March 15, 1982), p. 114. Golbe, Debra L. "The Effects of Imminent Bankruptcy on Stockholder Risk Preferences and Behavior." Bell Journal of Economics, Vol. 12 (Spring 1981), pp. 321-328. Gordon, M. J. "Towards a Theory of Financial Distress." Journal of Finance. Vol. 26 (May 1971), pp. 347- 356. Hanna, Mark. "Corporate Bankruptcy Potential, Stockholder Returns and Share Valuation: Comment." Journal of Finance, Vol. 27 (June 1972), pp. 711-717. Haugen, R., and Senbet, L. "The Insignificance of Bank- ruptcy Costs to the Theory of Optimal Capital Structure." Journal of Finance, Vol. 34 (May 1978). PP. . Hershman, Arlene. "The Big Bankruptcy Scare." Dun's Business Month (September 1982), p. 32. Hickman, W. B. Corporate Bond Qualityyand Investor Experi- ence. Princeton, N.J.: Princeton Universify Press, 1958. 103 Ibbotson, R. G. and Jaffe, J. F. "Hot Issues Markets." Journal of Finance, Vol. 30 (September 1974), pp. 1027-42. Ibbotson, R. G. "Price Performance of Common Stock New Issues." Journal of Financial Economics, Vol. 2 (1975), pp. 235-272. Jensen, M. C. "Capital Markets: Theory and Evidence." The Bell Journal of Economics and Management Science, Vol. (Autumn 1972), pp. 257-398. Larcher, D. F., Gordon, L. A., and Pinches, G. E. "Testing for Market Efficiency: A Comparison of the Cumulative Average Residual Methodology and Intervention Analysis." Journal of Financial and Quantitative Analysis, Vol. 15, June 1980, pp. 267-287. Lintner, J. "The Valuation of Risk Assets and the Selec- tion of Risky Investments in Stock Portfolios and Capital Budgets." Review of Economics and Statistics, Vol. 47 (1965), pp. 13-37. Markowitz, H. Portfolio Selection: Efficient Diversifica- tion of Investments. New York: Wiley, 1959. McClave, James T. and P. George Benson. Statistics for Business and Economics. San Francisco: Dellen Publishing, 1982. McDonald, J. G. and Fisher, A. K. "New-Issue Stock Price Behavior." Journal of Finance, Vol. 27 (March 1972). PP. 97-102. Meckling, William H. "Financial Markets, Default, and Bankruptcy: The Role of the State." Law and Contemporary Problems, Vol. 41 (Autumn 1977), pp. 13-38. Miller, Merton H. "The wealth Transfers of Bankruptcy: Some Illustrative Examples." Law and Contemporary Problems, Vol. 41 (Autumn 1977), pp. 39-46. Miller, M. H. and Scholes, M. "Rates of Return in Rela- tion to Risk: A Re-examination of Some Recent Findings," in M. C. Jensen, ed., Studies in The Theory of Capital Markets. New York: Praeger, (1972). PP. 47-78. Moody's Manuals: Industrials; Banks and Finance, and OTC Industrials. Moody's Investors Service, New York. 104 Moore, John. "Forward for Sessions on Economics and Bank- ruptcy Reform." Law and Contemporary Problems, Vol. 41 (Autumn 1977). PP. 1-12. Reilly, F. K. "New Issues Revisited." Financial Manage- ment, Vol. 6 (Winter 1977), pp. 28-42. Reinganum, M. R. "Misspecification of Capital Asset Pricing," Journal of Financial Economics, Vol. 8 (March 1981). PP. 19-26. Roll, R. "A Critique of the Asset Pricing Theory's Tests," Journal of Financial Economics, Vol. 4 (March 1977). pp. 129-176. Sanger, Gary. Stock Exchange Listings, Firm Value and Market Efficiency. Ann Arbor, Michigan: UMI Research Press, 1982. Schnepper, Jeff A. The New Bankruptgy Law. Addison- Wesley, 1981. Schuchman, Philip. "Theory and Reality in Bankruptcy: The Spherical Chicken." Law and Contemporary Problems, Vol. 41 (Autumn 1977), pp. 66-106. SEC. "List of Public Firms Which have Filed for Chapter 11 Since October 1, 1979. Sharpe, William F. "Capital Asset Prices: A Theory of Market Equilibrium Under Conditions of Risk." Journal of Finance, Vol. 19 (1964), pp. 425-442. Siegel, Sidney. Nonparametric Statistics. New York: McGraw-Hill, 1956. Stiglitz, J. E. "Some Aspects of the Pure Theory of Corpo- rate Finance: Bankruptcies and Take-Overs." Bell Journal of Economics and Management Science, Vol. (Autumn 1972). PP. 458-482. Trost, Ronald J. "Corporate Bankruptcy Reorganizations: For the Benefit of Creditors or Stockholders?" U.C.L.A. Review, Vol. 21 (1973), pp. 540-75. Warner, J. "Bankruptcy Costs: Some Evidence," Journal of Finance, Vol. 33 (May 1977a), pp. 337-347. . "Bankruptcy, Absolute Priority, and the Pricing of Risky Debt Claims." Journal of Financial Economics, Vol. 4 (May 1977b), pp. 239- 276. 105 Westerfield, R. "The Assessment of Market Risk and Corporate Failure." Wharton School Of Finance (August 1970). Weston, Fred J. "Some Economic Fundamentals for an Analysis of Bankruptcy." Law and Contemporary Problems, Vol. 41 (Autumn 1977), pp. 47-65. Wichern, D. W. and Jones, R. H. "Assessing the Impact of Market Disturbances Using Intervention Analysis," Management Science, Vol. 14 (November 1977), pp. 329-337. White, Michelle J. "Bankruptcy Costs and the New Bank- ruptcy Code." Journal of Finance, Vol. 38 (May 1983). pp. 477-488. APPENDICES 106 .wcunuu no oiuu or» an unsuxcon I m g“ u:o>—°UCH I N‘C .anau accuses coca peace: no Ou ouoa who: has» u“ uo mwo_ nuclease nozuuo cu ucqecu van med—«u - nouasgu among cucuon muse» can ucnccawoa luau some new panorama can: manage» >~uuec _ oo\~_ a.» as as.a asap»: a _.~n pee .uoam .acoaadz . ~u\o «us on ~a.o a==ax a a.~o- mm»: d..«>::dx ~o\n\n no .- pudenda ad sonata — ~o\~ a.» so so.~ Gaza: - n.~k_ mm>z ~a=o.a ~ _c\~ gas on o.~ .«aaox a a.» pee stanzas a co\o. an» SN A-.av mcaaaouoaa due: a n.~ use «unmade cox, acaoaauaa an nouaaou< . o~\o_ at» .a Aom.oc Apouaasooc baud: a o.om ope xd.oc=_ .muoa_ a aux." one". a» a -.~ m=._.d.da a ~.As~ mm»: ex: Noxa v_o=-»_ou->«.a «and». a ~o\~_ a.» o~ cm.o Asauadxc cacd: a “.4 sec Road;e-o:e coaamda noxo Beaches and: sauna H ~a\a a.» a .a._ med—aaco supposed : n._n use map a sac «usage ~a\.n\~— no a. use eon-up _ Noxs use on no._ A_aoad¢c used: a o..m mm»: madden «Anson _o\n~\~ co .- oauauaa co pounce _ ~o\~ a.» as o~.n once: a ~.c__ mm»: .acu d.ud.< ~ox~n\n no a. u...u-a co nuance ~ «exp at» am so._ used: a ~.omam mm»: _.uc_ a< coauuaaam goddamn nudmnmmxww. mu« wmwmmm .Innmmmmmmwlli Namwnmmwm, mummmmmm mama coca ans—«m \aAdc Au eschew one-am chauoz «x—«aa :u nova—ecu newcoasou no o_«u0om < x—ozmmm< 107 umxnxn no no uuuuuum co «outage ~w\n «cuccmxo; UFO co ACOnoooxv use: an: wamu voumvuocua ~m\N\o I ~—\w «catchy veocoeoso ~c\¢\o up we Oeo :6 save.» $5.22 .. ~33; weave: avocados» ~w\e~\- Co as who no pounce ~w\eN\~ yo no uuuuucm co powwow ~mxc~xo he no Ouuuooa co octane ~w\_—\w .zouoz ouq>uam an oouuscu< ~m\o—\~_ co mp oaoauam co vacate ow\o\m u —~\< «catch» covet; nose I: cm\s\m CO on who co vetenh noduosuuw aouuoam w up: .3 a: can: mod—um cho ocxw ~m\m Nxxn ~m\c_ —w\~ Nqu owxc— —w\~— o~\~— ~m\o— cwxe no: no» aha my» ~ m~ ON N .uaoo» moo—c a»; on» aha on; my» nu> mcz cn> m— an m an cm on ow¢ m~.c An~.m. —n.—o— ce.~ -.~ .wawmm \uaoe vascuucooln< .Ou acme—o: ucuuoxucz .909 ‘CEJ uses: but ucoauooacn .uceuh uate: uucuox UCEVSBUGOK «one: macs: uses: Nmmmwvco x—czmma< I n.n°n~ I o.~n~ I c.c~ I c.n~o I c.0cc I n.c- I n.— I o.nse I n.cn~m A nay—osnucz :2: aunI—u< suaom cinnamon canon .88?» can moo-um flaw nonnoo one: cues»: 0%KHG CECE ].C)8 dcxo coco: an pecusooc cwx— nox- I ~m\_— a cx\~— s usuncco ccx- owxm .oc~ convene Ou posse «no v.6» no >cz a_.sa. we was ~¢\o pco~ouu cu wcuoc nc\m ~¢\c— ~c\e wam ~¢\— .adxd _ to. F n¢\_ .utoda nx\- - a pee ~¢\_ we pandeaac.a ce\o oc\a oe\c_ ~¢\s spasms as po._=uu< ~m\s o~Ad_ad>< as dado sauna spas: aazccae «as: _ Oax— a _x\— — cc\m _ —x\¢ — ~¢\m _ ~w\c_ \c _m\__ ~ cx\—— u ccxo— m —c\~— a Ox\o H D¢\~ H ~c\c — —¢\n «ea: EMINWI once mas—Eu o—escm ¢>_;u:oz :« v0v3~u=~ oa—coniou no u—«uoum nu> cc> cu> cu> mu> max ass was rah our no» as» ww< oc «— mm mm ~— N— c— —N c. on <~ .xcu~«u mo elm» 0:» us unsuxcea I n v:- uco>uoncq I use .vo_ue nsoum auau cu aluuu use you». once—«ash no: can last Ind-cc co.— uses: Ao~.cv nouupuow «cameo; n.~ «cu—ueuaz o.m «cm—«ouoz cm.— mace: o.o_ 1.3oc .ou undead: Amo.nv ~mucoz ammo; .0uom N~.c uses: Cc.— couu=—~Om uu< I mace: me.c~ aaaucccsa Aew.a_o _«a¢-a=ecp we.» open: o~.m death pa< mo.~ ones< . cans: Ammwmm Nwmmmmmm \uaoa A_~ anodes n x—azuaa< I c.0n I 0.060— I o.sn I ¢.n—» A5 couscous; .ao so: co . ~w\c— un\~ .uuooz Oax— no vouucaav«a n Ow\¢ u A so on\~— mw\oN\c eco>ouw ma Ou v—Om —x\—~ pm .u«— .ooso .uu00u oz ~w\~— m¢\e cadue antes: ciao ~c\n oc\r~\_ mc\m~\~ Na edudpaaoa; - p¢\m no no cuocxc: cooupv< —x\—\— . ~w\__ use» so e0..=co< nw\~_\_ ~c\w\~ —c\mu\~ l lillllnllll ii )illlli illill unbouuo>< a— some >~uQO £0“: £&:Oh£.—. Once — ~ ~ — u “man to IV ocxo oc\a anxww ~c\— ~m\o chm ox\m Nxxn ow\~— —w\w~ ~¢x~ ~w\n page mc__.a Oh> ou> ns> an» an» an» mcz on» on» mu> mc> on» On ~— cw ~— ~— —— m— n— no mm on mum 0—.n— oh.¢ mo.— on.m Ge.— cc.~ mm.— —o.« NMflmmw \ucvo uses: uuouox uscmz macs: uses: uncut uses: ._axm mac a duo uate: uses: «acne: uses: Nmmnavcu vascuuCOUiln K~O2mma< I n.o~ I 0.0nn I ~.oss I n.- I ¢.¢_ » Namatmmmm UFO UhO xux< xuz< xmz< xux< UHO UFO war: war: 0 cqsouu coon: assessor + conch Sup—Uh. aucaoueeum ououaoaou nun—osm cues: ~.u=~ ceasuqu access «an no: ~009m guacaO: use: nuclease III: 110 l. i=2: no 0!: u. amass... I n can asap—02... I use .ouaca unzusal no u—«oe - co acuuuu —_ heunegu use». cusp accuuuuusocu be use voyeuoac ch— uuucs mod—«u usuuu unsuGOI popes» — O¢\~ an» an «v.0 ocluh I n.cu who u£lu0hh IOIuyO «Sp-.5 o: us 8 m “Na; 155.20; caucus »_.-p n_o\- ooze: nape.» ».:.=aa ”anxa ~aa== amends s_a-p _ om\o a.» me me.— cacax I _.omo mm»: page: coda: .Ou ucuufio: o~cs z—0uc>_un I one up ae\n .mccau .m¢\a uwuca acquuu noun: aoocuu >~u¢v n —o\~— aux n. Ano.~v .uscox nousaeou I a.n~ who suck cucum cuuom nudge. canon deputy a: a _e\o ace a. ~e.~s Gaza: I ~.an xuz< .ucu oceans: ~m\~— _«uc: uOu=Auuu «cu—«u sauce z—zucos coves» n ~m\m no» on .mc.mU noun nun-m Oua< I n.o xmz< vows wax ~c\~— —_u=: «cu—«u segue a—cucos eaves» ~ ~mx~ ocz n ca.m .ou acuv~o= ~«uuox I c.cn xuz< scam :Oqut n¢\o a wax» cassava novccu or u «xxx as _ nm.m med—«cues I c.on~ am»: you _c\m duos: acuduu sauce manna a—sucos a co auco papacy _ cx\n on» c ~O.n unuxucm you: I ~.n~ xmz< avenue we evacuasvc_ oucoqn no open _ Nm\m a»; o— o.n— cousbuuuouc I o.~n who anomauom eel—uu —m\o_ ~«uc: «cu—«u sauce wanes s—zuccs e :0 cho pooch» _ Ox\m at» a m~.— Cats: I c.n~ xux< .nuoo vac—use Ao_cc~«c>cc: >_«cov canon >_:ucos A._c>c«v e co mcxn oocuo voocuu go: an; O ~c\m was ~ nc.~ Case: I o._~ xwz< 95090 no acuuuu oocqa cocoon be: an; a ~cx- on» we Am~.wv coach I m.n~ xmz< uuouucn nonoou oe\o can“. covets so: an; _ cw\~ a.» an ~n.~ cued: I a » xuz< mean-om and»; scuuouoo courmmnwom HMNmLHmawfll mufl Nwmmrm NMuusv:~ A