Theory and Evidence on the Bankruptcy Initiation Problem

Transcription

1 Theory and Evidence on the Bankruptcy Initiation Problem Barry E. Adler a, Vedran Capkun b and Lawrence A. Weiss c 1st September 2005 a New York University School of Law, 40 Washington Square South, #424, New York, NY 10012, USA tel. +1 (212) , barry.adler@nyu.edu. b University of Lausanne, HEC Lausanne, BFSH1, 1015 Lausanne, Switzerland tel. +41 (21) , vedran.capkun@unil.ch c University of Lausanne, HEC Lausanne, BFSH1, 1015 Lausanne, Switzerland tel. +41 (21) , larry.weiss@unil.ch Abstract The bankruptcy act of 1978 placed corporate managers (as debtor in possession) in control of the bankruptcy process. Between 2000 and 2001 managers lost such control to creditors. This study examines financial ratios of firms filing for bankruptcy between 1993 and 2004 and tests the hypothesis that the change from manager to creditor control created or exacerbated managerial (and dominant creditor) incentive to delay bankruptcy filing. We find a clear deterioration in the financial conditions of firms filing after 2001, which suggests that managers may now postpone filings to avoid creditor control. We also observe patterns of operating losses and liquidations that suggest adverse economic consequences from such delay. FIRST DRAFT 1

2 1 Introduction A financially distressed firm may be economically viable and require debt relief to permit the effective application of its resources. Alternatively, the firm may be economically inviable with liquidation providing the highest value for its resources. Managers, who have some flexibility in the timing of the firm s filing for bankruptcy, do not always have the same incentives as the firm s investors or creditors. In particular, managers will not benefit from an efficient resolution of a debtor s distress if that resolution includes a new management team or liquidation and no management team. Additionally, the debtor s shareholders (whom the managers represent) will not benefit from an efficient disposition when it extinguishes equity s call option on the value of the debtor s assets. All else equal, the harsher the bankruptcy process is for managers or shareholders the longer managers can be expected to delay the onset of bankruptcy even at the price of inefficient continuation. Current US bankruptcy law provides a debtor s managers with an exclusive period in bankruptcy during which only they may propose a plan to reorganize the debtor. Traditionally, this exclusive period has afforded managers a strong bargaining position from which they might wrest concessions from creditors to their own or the shareholders benefit. Recently, changes in the Uniform Commercial Code, a new judicial attitude toward management, and the secured creditor monopoly on debtors source of liquidity, have combined in many cases to shift control of the bankruptcy process from the management (debtor-in-possession) to a secured creditor (secured-creditor-in-possession). A shift of control from management to creditor should alter managers incentives to file for bankruptcy. Managers should now have a greater incentive for delay even when the debtor will operate inefficiently and the firm s value will decline. The secured creditor may also have an incentive to delay the filing if he gains some control of a debtor and can exercise such control to assure repayment prior to the debtor s bankruptcy filing. This study tests this prediction and finds a clear deterioration in a variety of financial ratios of firms filing for bankruptcy in recent years. Firms that suffered such financial deterioration include those that suffered continual operating losses for at least a full year prior to the bankruptcy filing and those that ultimately stemmed such losses through liquidation in bankruptcy. 2

3 These results support two important hypotheses. First, managers react to changes in the bankruptcy law, which is thus shown to matter as a source of ex ante incentives. In this case, the result is delayed resolution of financial distress. Second, many firms in financial distress are also in economic distress and so exacerbation of the bankruptcy initiation problem has economic consequences. It follows that even for those who believe the modern trend of creditor control in Chapter 11 is a positive one, permitting optimal disposition of a distressed firm, the effects of the trend are self-limited. Efficient reform of the bankruptcy process should address the initiation problem as well. In the next section, we review the academic literature and discuss recent changes in bankruptcy practice as well as how these have reduced the incentives of managers to file for bankruptcy early. In Section 3 we present our data and methodology. Our main results are contained in Section 4. Finally, Section 5 provides our conclusions. 2 Background A standard, if imprecise, characterization of the relationship between corporate debt and equity is to say that equity is a call option on the firm s assets with the strike price equal to the amount of the debt. Bankruptcy accelerates the maturity of debt, and thus of equity s option. Consequently, one might expect that neither equity nor its agent, corporate management, has an incentive to bring a debtor into bankruptcy, particularly if the debtor is insolvent and equity s option thus out of the money. From these premises (and because the Bankruptcy Code 302(h) permits the debtor to quash a creditor s bankruptcy petition unless a debtor fails to pay its debts when due), Jackson (1986) reasons that one should expect firms to enter bankruptcy late even if an earlier bankruptcy filing would permit (or require) the firm to allocate its resources more efficiently, either in reinvestment of funds otherwise devoted to debt service or in liquidation of an unprofitable business venture. Studies by Bris et al. (2004) and Donoher (2004) support the supposition that the option value of 3

4 equity induces managers to delay the filing of a bankruptcy petition. To date, however, evidence of the potential harm caused by such delay has been sparse and mixed. Standard finance theory, dating from Jensen and Meckling (1976), predicts that managers of an insolvent firm will substitute unjustifiably risky projects for safer more profitable ventures, a phenomenon known as the debt overhang problem or as overinvestment (in the risky projects) and underinvestment (in the safer ones). In an attempt to verify this theory, researchers have examined firms to see if they alter their business practices as they face financial distress. These studies have concluded that firms do not alter their practices. See, e.g., Dahiya et al. (2003); Eckbo and Thorburn (2003); Andrade and Kaplan (1998). These results, however, do not imply the absence of a debt overhang problem, or that a delay in bankruptcy filing does not exacerbate any such problem, because an insolvent debtor s baseline of business as usual may itself indicate over- and underinvestment. Consider, for example, the following illustration. Suppose a firm must choose among its current activity but with an increased investment in inventory, the status quo of operation despite thin inventory, and liquidation where all of its assets are sold off to other firms. Without a recapitalization, and cancellation of equity, the firm might be unable to finance the new inventory and liquidation might directly eliminate equity s stake. Consequently, the firm might choose the status quo even if, from an efficiency perspective, the status quo is the worst of the three options. Yet a study equating debt-overhang inefficiency with a change from the status quo would (falsely) find no inefficiency. One might indirectly examine the economic consequences of financial distress. If financial distress induces firms to continue inefficient operations to serve the interest of the firms managers or shareholders, then the financially distressed firms will operate less productively than their financially sound counterparts, which presumably do not face the same perverse investment incentives. Research has tested just this proposition. Hotchkiss (1995) examines publicly traded firms filing for bankruptcy between 1979 and 1988 and finds these firms suffer operating losses in the years preceding (and following) bankruptcy. She concludes that these firms are not merely financially distressed but economically distressed as well. This supposition seems reasonable to us. To be sure, a firm might suffer operating 4

5 losses as part of an inevitably difficult process towards profitability, and undoubtedly this is true for many start-up firms. However, the 1980s experiment with highly leveraged transactions notwithstanding, we do not suppose investors will set up their firms to fail financially while they are economically sound, and so we do not believe a firm experiencing both financial distress and continuing operating losses is proceeding according to expectations. As suggested by Jensen (1988) and observed by Adler (1998), while financial distress can be the result of a single or limited number of exogenous shocks, continuing losses are more likely to be the product of an endogenous defect in the firm. More recent studies ostensibly call Hotchkiss (1995) results into question. For example, Andrade and Kaplan (1998), cited above, examine firms that became highly leveraged in the 1980s and then filed for bankruptcy. The authors observe that these firms operated efficiently up until and through their bankruptcy filings. But Andrade and Kaplan specifically chose highly leveraged firms to isolate the effects of financial distress from those of economic distress. Thus, their paper does not offer (or intend to offer) any insight into whether as a general matter financially distressed firms are also economically distressed. Another example is Maksimovic and Phillips (1998) who study manufacturing firm bankruptcies for the same period studied by Hotchkiss (1995). They find no difference in the operating efficiency of manufacturing plants of firms that filed for bankruptcy or their financially sound counterparts. This study, however, does not attempt to examine other aspects of a firm s efficiency, including a firm s distribution or sales operation. It may well be that the difference between a viable firm and an inviable one, particularly in a competitive industry, turns on these or related factors rather than on a manufacturer s ability to churn out product. To illustrate this point, albeit at the risk of caricature, one would not be surprised if a stapler fastened paper together as well at Enron as it did at Microsoft, yet one would not conclude from this that the two firms were equally efficient. Consequently, one might conclude, as did Hotchkiss (1995), that many debtors in bankruptcy are not only financially distressed but economically distressed as well. While the literature does address the questions of whether financially distressed firms delay bankruptcy filings and whether such firms suffer economic distress at the time of such filings, to the best of our 5

6 knowledge, no prior study has examined the important question of whether bankruptcy law or its process affects the timing of bankruptcy filings. Recent changes in United States commercial law and bankruptcy practice create an opportunity to answer this question. Between 1978-when the current Bankruptcy Code was enacted-and about 2001, a bankruptcy petition did not mean the end of management control over the debtor. Unlike the law of other countries (e.g. Canada and the U.K.) the U.S. Bankruptcy Code does not mandate that trustee takes over the control of a firm upon the filing of a bankruptcy petition. Quite the contrary, Code 1104 provides for the appointment of a trustee only after a hearing establishes that such an appointment is in the interest of the creditors or shareholders. Appointment of a trustee is an unusual event. Instead it is typical for the firm to remain under the control of the debtor s managers, referred to in the Bankruptcy Code as simply the debtor or the debtor-in-possession. Significantly, such control ordinarily extends beyond management of the firm s operations but to the reorganization process itself. The Code, 1121, grants the debtor the exclusive right for 120 days to propose a plan of reorganization and an additional 60 days to have the plan approved by the affected classes of claims or interests. Moreover, as documented by Weiss and Wruck (1998), at least until recently, judges often extended the exclusive period. 1 These rules had significant consequences, not only for managers but for shareholders whose option on a debtor s assets expire in bankruptcy. Bebchuk and Chang (1992) model how management s ability to delay resolution of the bankruptcy process can lead to concessions by creditors to managers and shareholders, and there is evidence to support the validity of this model. Although Gilson (1990) reports that 57% of CEOs and 54% of directors do lose their jobs when a debtor files for bankruptcy (in a 5 year period starting with the year of bankruptcy filing), the residual who retain their jobs do much better, of course, than they would under a more draconian system, such as those of other countries, that ousts management upon the firm s bankruptcy petition. 2 Regarding a firm s equity owners, Weiss (1990), among others, documents concessions to shareholders who received a return even when assets were insufficient to 1 The Bankruptcy Abuse Prevention and Consumer Protection Act of 2005 now limits the court s discretion to grant extensions, but for reasons that should become clear below, this recent development is inapposite to the discussion of this paper. 2 Gilson (1990) finds that only 17% of CEOs and and 15% of board members leave firm in the year of bankruptcy filing compared to the year before the filing. 6

7 repay creditors in full. 3 Thus, until recent changes, the bankruptcy process represented a relatively soft landing for corporate managers and the shareholders they represent. Prompted in part by a change in the Uniform Commercial Code, in or around the year 2000 a sea change occurred away from debtor control, and toward creditor control of the Chapter 11 bankruptcy process. As adopted by the states in the late 1990s the new UCC permits a creditor to take a security interest in a debtor s bank accounts. Armed with this new tool, it is now common for a debtor s dominant secured creditor to secure pledges of a debtor s liquid assets as the firm firm moves toward insolvency or becomes more deeply insolvent. Consequently, debtors now frequently enter bankruptcy with little or no liquid assets and require an immediate debtor-in-possession (or DIP ) loan to continue in business, with the pre-bankruptcy secured lender, who already has an interest in and knowledge of the debtor, in a unique position to provide such a loan. Given the strong creditor position, and a new judicial attitude that favors creditor control, secured lenders are as a practical matter increasingly able to dictate the outcome of the reorganization process. The Bankruptcy Code continues formally to grant control of the debtor and the reorganization process to the DIP, but as a condition of the DIP loan, the secured creditor demands that the debtor s managers contractually cede such control. These demands are frequently met, often before the bankruptcy petition is even filed, and courts now enforce such bargains. Not only do managers lose control of the debtor and reorganization process, but once in control creditor concessions to shareholders become unnecessary. Since the beginning of this decade, therefore, for many or most publicly traded firms the bankruptcy process has represented a hard landing as compared to earlier times. These changes have been documented, prominently, by other scholars. Skeel (2003) describes the new practice as adjustments creditors have made in an effort to reassert control in bankruptcy. Baird and Rasmussen (2002, 2003) argue that DIP lenders have come effectively to control the bankruptcy process. There is disagreement over whether the changes have been for better or worse. Baird and Rasmussen (2002, 2003), for example, conclude that creditor control has been largely beneficial, permitting creditors efficiently to screen viable firms from inviable ones without interference 3 See also Franks and Torous (1989) and Eberhart et al. (1990). 7

8 from managers or shareholders who have a perverse investment incentive. In contrast, Skeel (2003), as well as LoPucki (2003) and Warren and Westbrook (2003), believe that creditor control unduly permits creditor destruction of viable firms. More specifically, Warren and Westbrook (2003) decry the shift from the DIP to what they call the SPIP (secured-party-in-possession), and lament that Chapter 11 has become a sort of Super Article 9, invoked for the benefit of a single secured creditor, a development they see as a problem that requires congressional response. Regardless of the academic debate over the merits of the shift in bankruptcy practice, no one disputes that such a shift has occurred. The increased role creditors now have in Chapter 11 leaves managers and shareholders more at the mercy of creditors than was the case under prior practice. This leads to the proposition we test here. If managers are rational, all else equal, they should have been more willing to bring a firm under the supervision of the bankruptcy process-where the capital structure can be corrected and the perverse investment incentives of debt overhang eliminated-in the pre 2000 soft landing period than under current practice. This relationship should hold true even for firms that are economically distressed. As we explain more fully below, our data support this proposition. Even after controlling for economy-wide and industry-wide effects, financial ratios of firms filing for bankruptcy prior to 2000 were significantly better than the same ratios for firms filing after This suggests that in the more recent period managers are resisting bankruptcy longer than in the past all while the firm deteriorates. It is also possible a dominant secured creditor, who gained control of the debtor prior to the bankruptcy, may delay to satisfy its own claims in advance of the bankruptcy. This possibility supports our conclusions since in the the pre-2000 period the debtor s managers might have filed for bankruptcy to prevent the secured creditor from gaining control. Our data also isolates firms that suffer continual operating losses at the time of their bankruptcy filings (operating losses for a year prior to the bankruptcy filing). Consistent with Hotchkiss (1995), most firms in our sample that file for bankruptcy have suffered continued operating losses. For these firms, which are destroying value as they operate, worsening financial ratios reflect not only accruing 8

9 unpaid interest but a reduction in asset value. We find relative deterioration in financial ratios in the later (post 2001) period compared to the earlier (pre 2000) period even for these firms. Moreover, to a much greater extent in the latter period than in the former, liquidation is the resolution of insolvency in bankruptcy. Although we have not conducted a formal analysis of continuation or liquidation decisions in bankruptcy, and although it is thus possible that a delay in filing does not imply a delay in liquidation, an observation of increased liquidations is consistent with the hypothesis that delay in bankruptcy initiation permits inefficient continuation that might have been avoided by an earlier filing; certainly it seems plausible to assume that at least some inviable firms liquidate later because their managers delay bankruptcy and creditor control that may result in swift liquidation. These results challenge, or at least limit, the conclusion, in Baird and Rasmussen (2002, 2003), e.g., that the new hard landing regime in Chapter 11 is beneficial as it allows creditors quickly to liquidate firms that should not continue under current management. See also Jensen (1991), which argues that private workouts, such as those that are now routine within Chapter 11 practice, are likely to enhance efficiency. Regarding these last points, on the potential benefits of the new Chapter 11, creditor control of Chapter 11 may permit efficient resolutions once a debtor enters bankruptcy. However, the very creditor control yielding such a benefit ex post may exacerbate losses ex ante due to the incentive it provides managers of those firms to delay bankruptcy filing. Put simply, although creditor control may facilitate beneficial liquidation, it may be that by the time creditors gain control there is less to liquidate; it is not certain as a theoretical matter whether the old, soft landing regime or the new, hard landing regime is more efficient, and finance economists might note that it is also uncertain whether, or by how much, interest rates on risky debt should change as a result of the new regime. (For other work discussing the ex ante versus ex post tradeoffs between a hard and soft-landing regime, see Bebchuk (2002), Povel (1999), and Adler (1992)). A more complete solution to debtor misbehavior should include not only creditor control of the bankruptcy process but, as proposed by Adler (1998), some remedy to the bankruptcy initiation problem, such as a rule that would impede out-of-bankruptcy loans to an insolvent firm.. 9

10 3 Data and Methodology In this section we describe the collection of the sample of bankrupt firms, our choice of financial ratios, and our methodology for analyzing the data. We also document several characteristics of changes in bankrupt firms over time compared to the changes in the economic environment and non-bankrupt firms. 3.1 Firms Filing for Chapter 11 We identify 827 firms filing for Chapter 11 over the 12 year period from January 1993 to March 2004 using Edward Altman s database of corporate bankruptcies. This database provides a comprehensive list of all firms filing for bankruptcy. Next, we obtained the date of the last reported financial statements with the SEC from EDGAR - the SEC database. Financial information about these firms we obtained from Standard and Poor s Compustat database. We eliminated all financial institutions (they have unique financial statements that are not comparable with other industrial firms), all firms that had not published an annual or quarterly financial report within 365 days prior to the Chapter 11 filing, and all firms with total liabilities (according to the last report filed prior to the bankruptcy) less than $ 100 million. Our final sample contains 423 firms. We then split our sample into two sub-samples: Service and non-service firms. The Compustat database provides the SIC codes of the firms in our sample and we classified firms with SIC codes above 6000 as service firms and all the others as non-service firms. Table 1 presents the sample and two sub-samples including their last reported assets and liabilities prior to Chapter 11 filing. Out of the total number of firms in the sample, 345 are non-service firms and 78 are service firms. On average, firms in the sample have total assets and total liabilities prior to bankruptcy of just under $1.5 billion and just over $1.36 billion respectively. The range includes firms with assets from $33 million to $104 billion and with debt from $103 million to $52 billion prior to the bankruptcy filing. The largest non-service firms are substantially larger than the largest service firms both in totals 10

11 (the largest service firms have assets of $6.8 billion and debt of $6.2 billion versus $104 billion and $52 billion for non service firms) and their means ($1.02 and $1.1 billion versus $1.6 and $1.42 billion respectively). The mean of total assets and total liabilities of non-service firms are considerably higher than the ones of service firms, but median values show a smaller difference. Table 2 presents the annual distribution of Chapter 11 filings. As expected the largest number of bankruptcies were filed in 2001 (99) and 2002 (76) equaling 23% and 18% of the sample respectively. This is consistent with what Altman and Jha (2003) find for defaulted and distressed securities. Our analysis is done based on the last reported period (quarterly, semi annual, annual) prior to the bankruptcy filing. These reports serve as the basis for computation of financial ratios. The last financial report filed with the SEC prior to the bankruptcy filing was filed from 0 to 362 days prior to the bankruptcy filing, with a mean (median) of 114 (78) days. If the report was filed for a period ending up to 15 days after the bankruptcy filing we include the report and list this as 0 days prior to the bankruptcy filing Financial Ratios We predict the financial health of firms filing for bankruptcy will deteriorate in the SPIP period as managers will delay filing. To test this we use financial reports to compute six financial ratios for every firm in the sample as follows: Predicted change in financial ratios is shown by and arrows. 1. Return on assets (ROA) IncomeBef oreextraordinaryitems T otalassets (1) ROA proxies for the true profitability of a firm. As profitability increases, the firm will have 4 There were 9 firms for which we used these reports. Our results are not significantly changed if we use the prior period reports or if we exclude these firms from the sample. 11

12 less difficulties meeting its obligations, and it is predicted that the probability of bankruptcy will decrease. Also, large firms may be better able to withstand short-term downturns in their operations. 2. Current Ratio (LIQ) CurrentAssets CurrentLiabilities (2) LIQ proxies for the ability of a firm to meet liabilities as they come due. As liquidity increases, the ability of the firm to meet its-current obligations- increases, and the probability of bankruptcy is predicted to decrease. 3. Leverage (LEV) T otalliabilities T otalassets (3) LEV proxies for the indebtedness of a firm. As the percentage of the firm s financing by debt increases, its ability to meet the interest and capital repayments becomes less likely, and the probability of bankruptcy is predicted to increase. 4. Days receivable (DAR) AverageAccountsReceivable Sales 360 (4) DAR gives the days the firm needs to collect the money for goods delivered or services rendered. It shows the period the firm finances its customers. We predict an increase in this ratio as the firm approaches bankruptcy. This is the consequence of customers delaying paying for goods and services as they become aware of firm s problems. 5. Days Inventory (DIN) AverageInventory COGS 360 (5) DIN is the number of days a firms inventory is sufficient to meet its sales. We assume that as a firm approaches bankruptcy suppliers will be increasingly reluctant to sell the firm goods and materials. 12

13 6. Days Payable (DAP) AccountsP ayable 360 (6) Inventory + COGS + DepreciationAndAmortization DAP is the number of days the firm is financed by its suppliers. As the firm approaches bankruptcy this ratio increases because the firm delays payments to its suppliers. These ratios are standard ones that reflect the profitability, risk (short and long term) and activity covering the overall financial health of a firm. Needless to say there are hundreds of possible ratios. Our goal is not to find the best ratio, merely to determine if there is a significant change in the financial health of the firm. In order to compute ratios from quarterly financial statements we use 12 month moving cumulative sales, costs of goods sold, depreciation and amortization, as well as income before extraordinary items, and the beginning and ending 12 month period balances for all the other accounts. Due to data availability, not all the ratios could be computed for all the firms in the sample. Since 187 firms had negative equity according to the last report filed before bankruptcy, the return on equity ratio (ROE) was omitted from the presentation, but we will continue to discuss the number of firms having negative equity in the paper as one of the indicators of firms health. Table 3 presents the ratios for the firms in our sample. On average, firms filing for Chapter 11 have Return on Assets of % in the period of 12 months prior to filing for Chapter 11. The mean Current Ratio is 0.95 and the mean Leverage is The mean of Days Receivable is 51 days, Days Inventory 58 days and Days Payable is 48 days. Service firms have lower average Return on Assets than non-service firms (-30.51% versus 23.07%). The Current Ratio of service firms is lower (0.69 versus 0.99), and the Leverage higher (1.36 versus 1.17) than for non-service firms. Days Receivable are higher, but Days Payable lower for service than for non-service firms (63.18 versus and versus respectively). Days Inventory are much smaller for service firms than for non-service firms (12.13 versus 67.8), however, this is largely due to the lack of inventory held by service firms. 13

14 Overall, all our financial ratios are consistent with service firms entering bankruptcy proceedings on average in a worse financial position than non-service firms. Following on Hotchkiss (1995) we analyze the operating income of firms during the one year prior to bankruptcy filing. We use two measures; operating income before depreciation and amortization and operating income after depreciation and amortization. We define operating income as follows: Revenues - Cost of Goods Sold or Services Rendered - Sales, marketing and administrative expenses - Depreciation and Amortization (D&A) Operating income before depreciation and amortization serves as a pre-interest proxy for cash flow available to stakeholders and operating income after depreciation and amortization as a proxy for existence of economic distress. Operating loss, as a pre-interest measure, is a strong indicator of economic distress of a firm because it does not include costs of financing and is limited to firms operating activities. Operating losses also indicates that the firm is destroying assets. Restructuring or liquidation in bankruptcy is often cited as necessary to reverse the poor investment decisions that lead a firm to deplete its assets. Out of 423 firms in our sample, data on operating income was available for 399 firms. In the period one year prior to the bankruptcy filing, 147 (36.84%) firms suffered operating losses before D&A and 257 (64.41%) firms suffered operating losses after D&A. We run a separate search in Lexis-Nexis merger and acquisition database to find firms that liquidated some or all of their assets. Out of 423 firms in our sample 168 (39.71%) firms liquidated at least some of their assets (120 (28.37%) firms liquidated part of their assets and 48 (11.34%) firms were taken over after declaring bankruptcy). 14

15 4 Empirical Analysis, Results and Discussion We divide our 12 year sample of Chapter 11 filings into two sub-samples according to the changes in bankruptcy practices that occurred in years 2000 and 2001: (Debtor in Possession) and (Secured Party in Possession). Firms filing for bankruptcy protection under Chapter 11 during 2000 and 2001 were not included in this analysis as this is the period during which the change occurred. We perform the analysis of all the firms in the sample as well as for the non-service and service firms separately, but we present the result for all firms and non-service firms only. Table 4 presents the financial ratios of firms filing under Chapter 11 for the two sub-samples (DIP and SPIP) and also presents the results of the t-test and Mann-Whitney test of the difference between the means of two sub-samples for the six financial ratios. As can be seen from Table 4 the means of all the financial ratios are worse in the SPIP period. Return on Assets decreased by 7% from -22.1% to -23.6%. The Current Ratio decreased 35% from 1.22 in the period before the change in to 0.79 after the change in bankruptcy practices. Leverage worsened as well increasing by 24% to reach 1.35 in the SPIP period from 1.08 in the DIP period. Days Receivables increased 17% from 45 to 53 days while Days Inventory decreased 17% from 71 day before the change to 59 days after the change. Days payable did not change and stayed at 46 days. The percentage of firms filing for Chapter 11 having negative equity increased from 29% before the change to 59% after the change in bankruptcy practices. The change is statically significant for Current Ratio, Leverage and Days Receivable. This indicates that both short term and long term risk of the firms filing for bankruptcy increased in the SPIP period compared to DIP period and supports the claim that managers delay filing for bankruptcy until the firm has no choice but to file. The incentives apparently change towards filing for bankruptcy rather later than sooner. Unlike the DIP period, in the SPIP period most firms that file for bankruptcy have negative equity, and substantially higher leverage. The average Current Ratio falls under 1 in the SPIP period indicating increased short term liquidity problems. On average in the DIP period firms could cover their short term debt with their short term assets. This is not true in the SPIP period. We argue that the increase in days receivables is due to the change in behavior of customers as the 15

16 firm approaches bankruptcy. If the firm delays filing for bankruptcy more customers become aware of the problems the firm has and try to delay paying for goods delivered or services rendered. We repeat the test for the non-service firms sample. Table 5 presents the results of non-service firms divided into two sub-samples (DIP and SPIP). The results for non-service firms reflect the pattern noted for the entire sample above. Return on Assets decreased from -20.5% to -22.4% (9% change). The Current Ratio decreased by 37% from 1.27 to Leverage increased by 23% from 1.05 to While Days Receivable increased by 22% from 43 to 53 days, Days Inventory decrease by 18% from 79 to 65 days. Days Payable decreased by 7% from 48 to 45 days. The percentage of non-service firms filing for Chapter 11 protection having negative equity increased from 29% in DIP period to 57% in SPIP 2001 period. Changes in Return on Assets and Days Inventory are statistically significant at the 10% level (Mann-Whitney). As for the whole sample, the change in Current Ratio, Leverage and Days Receivable was statistically significant. When we run this test for service firms, we find that these firms do not show statistically significant change in their financial ratios, but the number of firms in the sub-samples is very small compared to the non-service sub-samples. (The samples consist of as few as 14 ratios.) Out of 128 firms for which the data was available in the pre 2000 period, 51 firms (39.84%) suffered operating losses before D&A and 85 (66.38%) operating losses after D&A. We find a similar result for the post 2001 period where out of 130 firms for which the data was available 50 (38.46%) suffered operating losses before D&A and 85 (65.38%) firms operating losses after D&A. This is consistent with our arguments about the presence of not only financial distress but also economic distress for firms filing for bankruptcy. As previously discussed, the financial ratios in the pre-2000 and post-2001 differ significantly. Our results do not change when we isolate firms with operating losses from those with operating profits. Financial ratios deteriorate in the SPIP period compared to the DIP period even for the operating losses sub-sample. In the pre-2000 period 23 (17.29%) firms out of 133 firms liquidated at least some of their assets (15 (11.28%) firms only part of their assets and 8 (6.01%) firms were taken over after the bankruptcy 16

17 declaration). The number of firms liquidated in part or in whole increases substantially in the post period to 51.45% (71 out of 138 firms). Fifty firms (36.23%) liquidated some of their assets and 21 (15.22%) firms were taken over after filing for bankruptcy. This is consistent with the findings of Aug et al. (2003) that in 84% of all chapter 11s from 2002, the investors entered bankruptcy with a deal in hand or used it to sell the assets of the business. These data indicate that the purpose of Chapter 11 bankruptcy shifted from reorganization to liquidation (and this is the case even if one chooses not to characterize as a liquidation those cases that involve the takeover of a going concern). The reason for such shift toward liquidation may be that firms in the latter period as well as in the former were frequently inviable, but in the latter period creditor control of the bankruptcy process permitted liquidation while in the former debtor control did not. Or the reason may be that firms in the latter period more than in the former deteriorated beyond salvation by the time of bankruptcy. In either case, however, the delay in bankruptcy initiation implies economic losses as well as financial decline. 5 Sensitivity of Results to Various Assumptions 5.1 Changes in the Economic Environment The changes in the financial ratios from the DIP to SPIP period could be due, at least in part, to changes in the economic environment. To examine the changes in the economic environment we compute financial ratios of all firms in the Compustat database to see how they changed over our time period, and specifically in the DIP versus SPIP periods. 5 For our sample of all firms, we used the population of US incorporated firms in the Compustat database and then eliminated all firms with total liabilities lower than $100 million. We then eliminated the financial firms, and further separated service and non-service firms (matching our procedures for 5 Removing the bankrupt firms from the total sample has no significant impact on the results, as might be expected due to the small number of bankruptcy firms in comparison to the entire population of bankrupt and non-bankrupt firms 17

18 our sample of bankrupt firms). We then computed the differences between the financial ratios of a bankrupt firm and median ratio of all non bankrupt firms in the same quarter to obtain the median adjusted ratio (MA). The median adjusted ratios show the distance of the bankrupt firm ratio to the median of all firms in the population. The increase in distance implies worsening of a financial ratio with respect to the general economic environment. Table 6 shows the changes in median adjusted financial ratios, the results of the t-test and Mann- Whitney test between sub-samples of adjusted ratios. Firms are then divided by date into our DIP and SPIP times periods and we again excluded the years 2000 and MA Return on Assets decreased by 4% from -27% to -26%. MA Current Ratios decreased (the distance increased) by 64% from to -0.95, MA Leverage increased by 49% from 0.52 to 0.78 and the MA Days Receivable increased by 94% from -5 to 0 days. MA Days Inventory decreased by 29% from 24 to 17 days and MA Days payable stayed unchanged at 12 days. The change is statistically significant only for MA Current Ratio and MA Leverage. This supports our claim that the change in financial ratios of bankrupt firms happened due to the changes in the legal framework rather than the change in the economic environment. We then repeat the analysis for non-service firms only. The results presented in Table 7 confirm our previous findings. MA Return on Assets increased by 3% from -26 to -25%. MA Current ratio decreased (the distance increased) by 77% from in the DIP to in the SPIP period. MA Leverage increased from 0.49 to 0.72 (by 47%). While MA Days Receivable increased by 94% (from -7 to 0), MA Days Inventory decreased from 32 to 23 days (by 28%). MA Days Payable decreased by 22% from 14 to 11 days. Non-service firms exhibit statistically significant worsening of the Current Ratio, and Leverage also when adjusted for economic trends. Next, we compute median statistics of industries (GICS four digit industry code) to test for industry effects over time. For every firm we compute the difference between the pre-bankruptcy ratios and the industry medians for the same accounting period. The results confirm the results for the non adjusted and median adjusted ratios. Table 8 shows the results for industry median adjusted (IMA) financial 18

19 ratios of bankrupt firms. IMA Return on Assets increased by 9% reaching -25% in the SPIP period compared with -28% in the DIP period. IMA Current Ratio decreased (distance to industry median increased) from to (by 76%). IMA Leverage increased by 46% from 0.57 to IMA Days Receivable decreased by 30% from 6 to 4 days, while IMA Days Inventory decreased by 16% from 14 to 12 days. IMA Days Payable stayed unchanged at 11 days. Only changes in IMA Current Ratio and Leverage are significant, confirms the previous results. 5.2 Period of Changes in the Legal System and Practices As discussed above, the changes in the legal system occurred during the year The aforementioned expansion of secured creditor rights under Article 9 of the Uniform Commercial Code (UCC) took effect on July 1, 2001 in most states (46 + DC) and by January 1, 2002 in the remaining 4 states. The changes in bankruptcy proceedings, however, were underway in the year In our main results, we decided to exclude years 2000 and 2001 from the analysis. In order to examine whether our results change if the sample is divided in some other way, we tested the difference between financial ratios of firms that filed for bankruptcy pre and post the second half of This excludes the period of legal changes only without taking into account the period before (when managers may have already had information about the coming changes). Our results do not change when we divide the sample this way. Bankrupt firms and median adjusted bankrupt firms Current Ratio, Leverage and Days Receivable show statistically significant change for the whole sample, as well as for the non-service firms only. When we exclude the entire year 2001 from the analysis (as the year when the changes of Article 9 took effect) the results also remain the same. Comparing years 1999 and 2002 also supports our main findings and show statistically significant differences between DIP and SPIP Current Ratio and Leverage. 19

20 5.3 Using all Financial Ratios vs Using Companies In our analysis we used all available financial ratios of firms that filed for bankruptcy. We included firms even if we only had the data on one ratio for the firm. We also run a separate test where we include only those firms that had all 6 financial ratios. Out of total number of firms (423), 133 belong in the DIP and 138 in the SPIP period. When we eliminate all the firms for which we do not have all the financial ratios, our sample is reduced to 115 firms in the DIP and 110 firms in the SPIP period. We repeat the procedure for non-service firms, and find a sample of 103 firms in the DIP and 98 firms in SPIP period (vs 116 and 118). Our results do not change if we use only the firms for which we had all 6 financial ratios available. 6 Conclusion Longstanding theory predicts that an insolvent debtor will file for bankruptcy only when default is imminent, even where an earlier filing might have stemmed deterioration of the firm s value. The bankruptcy reform act of 1978 built in a relatively soft landing for the debtor s managers (debtorin-possession) and thus its owners. Generous bankruptcy treatment of managers and owners could encourage managers to file somewhat earlier as such treatment leaves managers relatively less incentive to permit a decline in firm value in the hope of an unlikely reversal of fortune. Recently, it has become possible to test managerial reaction to bankruptcy incentives. Between 2000 and 2001, perhaps prompted by a change in the Uniform Commercial Code, bankruptcy practice shifted from debtor control of the bankruptcy process to control by a secured creditor, which reduced if not outright eliminated any incentive for the debtor s managers and shareholders to file early. This paper examines the financial and economic conditions of firms that filed for bankruptcy from 1993 through Our investigation reveals a clear deterioration in the financial health of firms at the time of their bankruptcy filing in the post 2001 SPIP (Secured Party in Possession) period, which cannot be explained by general economic changes or related changes of firms not filing for bankruptcy. We also 20

21 find that many filing firms suffered operating losses in both periods and that in the latter period more than the former liquidation was the the eventual bankruptcy outcome for many firms. These results support two important hypotheses. First, managers react to changes in the bankruptcy law, which is thus shown to matter as a source of ex ante incentives; in this case, the result is delayed resolution of financial distress. Managers increasingly file for bankruptcy only after financial distress becomes acute, or they cede the filing decision to a dominant secured creditor who delays filing until the debtor s finances are desperate. Second, many firms in financial distress are also in economic distress and so exacerbation of the bankruptcy initiation problem has economic consequences. Specifically if bankruptcy corrects inefficient investment incentives through recapitalization or liquidation, a delay in bankruptcy may also delay such correction. These findings also call into question the belief that bankruptcy practice in the SPIP period has effectively cured the debt overhang problem. Even for those who believe that the modern trend of creditor control in Chapter 11 is a positive one, encouraging eventual optimal disposition of a distressed firm, the effects of the trend are self-limited. As such our findings have implications for legislative policy, specifically for bankruptcy law as regulation of capital structure and economic distress. In particular, our results support the need for Congress to address the bankruptcy initiation problem, so that firms cannot waste assets before filing for bankruptcy just as they may have formerly wasted them afterward. 21

22 References Adler, B.E Bankruptcy and Risk Allocation. Cornell Law Review 77: Accelerated Resolution of Financial Distress. Washington University Law Quarterly 76: Altman, E.I., and S. Jha Report on Market Size and Investment Performance of Defaulted Bonds and Bank Loans: Technical Report, New York University Salomon Center, Leonard N. Stern School of Business. Andrade, G., and S. N. Kaplan How Costly Is Financial (Not Economic) Distress? Evidence from Highly Leveraged Transactions That Became Distressed. Journal of Finance 53: Aug, J.V., L.R. Ahern, R.M. Meth, and J. Greenstone Miller The Plan of Reorganization: A Thing of the Past. Journal of Bankruptcy Law and Practice 13:3 60. Baird, D.G., and R.K. Rasmussen The End of Bankruptcy. Stanford Law Review 55 (3): Chapter 11 at Twilight. Stanford Law Review 56 (3): Bebchuk, L.A Ex Ante Costs of Violating Absolute Priority in Bankruptcy. Journal of Finance 57: Bebchuk, L.A., and H. Chang Bargaining and the Division of Value in Corporate Reorganization. Journal of Law Economics and Organization 8: Bris, A., I. Welch, and N. Zhu The Costs of Bankruptcy: Chapter 7 Cash Auctions vs. Chapter 11 Bargaining. Yale ICF Working Paper No Dahiya, S., K. John, M. Puri, and G. Ramírez Debtor-in-possession financing and bankruptcy resolution: Empirical evidence. Journal of Financial Economics 69: Donoher, W.J To File Or Not To File? Systemic Incentives, Corporate Control, and the Bankruptcy Decision. Journal of Management 30:

23 Eberhart, A.C., W.T. Moore, and R.L. Roenfeldt Security Pricing and Deviations from the Absolute Priority Rule in Bankruptcy Proceedings. Journal of Finance 45: Eckbo, E.B., and K.S. Thorburn Control Benefits and CEO Discipline in Automatic Bankruptcy Auctions. Journal of Financial Economics 69: Franks, J.R., and W.N. Torous An Empirical Investigation of U.S. Firms in Reorganization. Journal of Finance 44: Gilson, S.C Bankruptcy, boards, banks, and blockholders : Evidence on changes in corporate ownership and control when firms default. Journal of Financial Economics 27: Hotchkiss, E.S Postbankruptcy Performance and Management Turnover. Journal of Finance 50:3 21. Jackson, T.H The Logic and Limits of Bankruptcy Law. Harvard University Press. Jensen, M.C Their Causes and Consequences. Journal of Economic Perspectives 2: Corporate Control and the Politics of Finance. Journal of Applied Corporate Finance 4: Jensen, M.C., and W.H. Meckling The Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics 3: LoPucki, L.M The Nature of the Bankrupt Firm: A Response to Baird and Rasmussen s The End of Bankruptcy. Stanford Law Review 56: Maksimovic, V., and G. Phillips Asset Efficiency and Reallocation Decisions of Bankrupt Firms. Journal of Finance 53: Povel, P Optimal Soft or Tough Bankruptcy Procedures. Journal of Law Economics and Organization 15: Skeel, D.A Creditors Ball: The New New Corporate Governance in Chapter 11. University of Pennsylvania Law Review 152: Warren, E., and J.L. Westbrook Secured Party in Possession. American Bankruptcy Institute Journal, September. 23

24 Weiss, L.A Bankruptcy Resolution: Direct Costs and Violation of Priority of Claims. Journal of Financial Economics 27: Weiss, L.A., and K.Wruck Information Problems, Conflicts of Interest and Asset Stripping: Chapter 11 s Failure in the Case of Eastern Airlines. Journal of Financial Economics 48: