Prepacked bankruptcies, or “prepacks,” are considered a hybrid form of distressed restructuring because they share certain characteristics with both of the widely used alternatives for reorganizing distressed companies—out-of-court restructurings (OCRs) and traditional Chapter 11 reorganizations. Prepacks are similar to OCRs in that creditors and the debtor agree to the major terms of the reorganization outside of the court. Prepacks are similar to traditional Chapter 11 filings in that the reorganization occurs under court supervision, confirmation of the plan requires approval by two-thirds in amount and one-half in number by each class of claimholder, and all claimholders must exchange their old securities in accordance with the terms of the plan. In a prepack, the Chapter 11 bankruptcy petition and a plan of reorganization are filed simultaneously with the court. In a 1991 article in this journal, John McConnell and Henri Servaes laid out a number of hypotheses as to why distressed firms might use prepackaged bankruptcies to reorganize.1 At the time of their article, however, prepacks were still relatively uncommon and these authors were limited to an “anecdotal” discussion of four cases to make their points. With the passage of time and the growth in the number of prepacks, we have been able to assemble data for a substantial sample of prepacks. Our study of prepacks complements a growing literature on the outcomes of various forms of distressed reorganization. A significant concern in this literature is whether the various reorganization procedures are efficient. Inefficient reorganization procedures can result in excessively high direct costs or sub-optimal financing and investment decisions by firms. The most efficient organization procedure is the one that creates the greatest value for the firm, net of all costs. Although efficiency cannot be observed directly, we provide evidence on a number of indirect measures of efficiency—for instance, the time required to reorganize, the cost of reorganizing, and the recovery rates by creditors. Where the data are available, we compare prepacks to OCRs and traditional Chapter 11s. We find that on most dimensions considered, prepacks lie between the two alternative means of reorganizing financially distressed firms. For example, prepacks have higher costs of reorganizing (as a fraction of assets) than OCRs, but lower costs than conventional bankruptcies. These findings buttress the idea that prepacks are a hybrid form of reorganization that contain some aspects of both OCRs and traditional Chapter 11s. Our sample consists of 49 financially distressed firms that filed prepacks over the period 1986 through June 1993. Crystal Oil, which filed a prepack in 1986, is widely regarded as the first prepack of a large firm. Following Crystal Oil, the next two prepacks in our sample occurred in 1989 with combined assets of $1.7 billion. In the years thereafter, four took place in 1990 with combined assets of $3.6 billion, 13 in 1991 with assets of $5.2 billion, 17 in 1992 with assets of $11.2 billion, and 12 took place through the first six months of 1993 (the cutoff point of our study) with total assets exceeding $5.5 billion. In 1993, moreover, 12 (or over 50%) of the 22 publicly traded firms with assets exceeding $100 million that filed for Chapter 11 filed a prepack (using our definition of the term). In 1994, 11 of 29 (or 38% of) such firms filed a prepack.2 Our sample includes two types of prepacks—“pre-voted” and “post-voted” prepacks.3 In a pre-voted prepack, claimholders vote on the plan of reorganization before the Chapter 11 bankruptcy petition is filed with the court. The bankruptcy petition and the voting results are then filed along with a plan of reorganization. Absent improper disclosure or voting irregularities, the pre-filing vote is binding upon all claimholders. In a post-voted prepack, the bankruptcy petition and the plan of reorganization are filed simultaneously, but prior to a formal vote by claimholders. A vote is then conducted under the jurisdiction of the court. In our sample of prepacks, 32 are pre-voted and 17 are post-voted. All 49 of these firms eventually reorganized and emerged from Chapter 11. As might be anticipated, pre-voted prepacks require less time in Chapter 11 than post-voted prepacks. It turns out that pre-voted and post-voted prepacks differ in other ways as well. In particular, pre-voted prepacks involve larger firms, involve a longer time for pre-filing negotiations, incur lower proportional fees, provide a higher recovery rate for creditors, have greater dollar percentage deviations from absolute priority, and provide for lower post-reorganization equity ownership for creditors. We now turn to our proxy measures of reorganization efficiency. To put our investigation in context, we compare measures for our proxies for efficiency with similar statistics generated for OCRs and traditional Chapter 11 reorganizations as reported in other studies.4 To determine the time required to reorganize using a prepack, we begin with the first date on which we could identify any public indication that the firm had begun negotiations with creditors. We end with the date on which the plan of reorganization is confirmed by the court. We split this interval into two components: the time spent in pre-filing negotiations and the time spent in Chapter 11. Overall, total reorganization time is shortest for OCRs and longest for traditional Chapter 11s. However, the data suggest that different methods of distressed restructuring involve trade-offs between time negotiating outside of court and time spent in Chapter 11. For example, as shown in Table 1, more time is devoted to pre-filing negotiations in pre-voted than in post-voted prepacks—20.0 months versus 14.9 months—but more time is spent under court supervision in post-voted than in pre-voted prepacks—6.0 months versus 1.9 months. On average, however, the two alternatives require the same amount of time from start to finish—about 22 months.5 Thus, the total length of time required to complete the reorganization by a prepack does not appear to be affected by whether the final vote on the plan occurs out of court or in court. Comparable data on reorganization time for OCRs and traditional Chapter 11s also are presented in Table 1. A comparison between the negotiation time and time in bankruptcy in prepacks and conventional Chapter 11s supports the observation that distressed firms trade off time in negotiating out of court for time spent in Chapter 11. In comparison with traditional Chapter 11s, more time is devoted to pre-filing negotiations in a prepack (18.3 months vs. 8.1 months) and less time is spent in court (3.3 months vs. 23.2 months). But, from start to finish the complete process takes less time with a prepack (21.6 months vs. 28.5 months) than with a traditional Chapter 11. In the case of OCRs, the time required to complete the process is just over 15 months. Overall, the total time required to complete a prepack lies near the midpoint between the time to complete an OCR and the time required for a traditional Chapter 11. We define the direct costs of reorganizing to include court costs and professional fees, the bulk of which go to financial advisors. As shown in Table 1, the direct costs of $8.5 million for pre-voted prepacks are greater than those of $3.8 million for post-voted prepacks. But, as a percentage of assets, the direct costs of pre-voted prepacks are marginally lower than those of post-voted prepacks. They average 1.65% of assets for pre-voted and 2.31% for post-voted prepacks. As a percentage of assets, the direct costs of both pre- and post-voted prepacks lie between the costs of 0.65% for OCRs and 2.80% for traditional Chapter 11s. The recovery rate for creditors and preferred stockholders, which is the total payoff to these classes of claimholders divided by the amount of their claims, is displayed in the bottom row of Table 1. The recovery rate for all classes of creditors and preferred stockholders is 75.1% in pre-voted prepacks and 68.2% in post-voted prepacks.6 These levels compare with the recovery rates of 80.1% and 50.9% for OCRs and traditional Chapter 11s, respectively. Thus, on this statistic as well, both types of prepacks lie between OCRs and traditional Chapter 11s. The intermediate recovery rate for prepacks, considered together with their intermediate levels of direct costs and total reorganization time (which is likely to capture indirect costs), suggests that the financial health of a firm plays an important role in its choice of a reorganization method. That is, our findings reinforce the suggestion that firms with the lowest degree of distress reorganize by means of an OCR, firms with an intermediate degree of distress choose prepacks, and firms in the greatest distress reorganize by means of Chapter 11 (or are liquidated in Chapter 7). Since our data include only prepacks, we are unable to test this conjecture directly. Absolute priority is upheld in a reorganization when a class of securities receives a payoff with a market value at least equal to the face value of its claim before a junior class receives any distribution. The degree to which absolute priority is preserved in a reorganization is an indicator of the degree to which the terms of debt contracts are upheld and creditors are protected from expropriation in the bankruptcy process. For our sample of prepacks, absolute priority is upheld in 22% of the cases; priority is upheld for secured creditors but not for unsecured creditors in 47% of the cases. For the remaining 31% of the firms, priority is violated for secured creditors. When we distinguish between pre- and post-voted prepacks, we find that absolute priority is upheld more frequently in post-voted than in pre-voted prepacks. In traditional Chapter 11 reorganizations, absolute priority is upheld 22% of the time. Priority is upheld for secured, but not unsecured creditors 70% of the time, and priority is violated for secured creditors just 8% of the time. Our data indicate that violations of absolute priority are frequent in prepacks. To determine the economic significance of these violations, we calculate percentage dollar deviations from absolute priority as the dollar amount of the deviation divided by the dollar amount paid to all claimholders in the reorganization.7 As reported in Table 2,8 the percentage dollar deviations from absolute priority are greater, on average, in pre-voted than in post-voted prepacks. Nevertheless, in both types of prepacks the percentage dollar deviations from absolute priority are quite small. For example, for secured creditors the percentage dollar deviation is –0.91% in pre-voted prepacks and it is +0.09% in post-voted prepacks. In comparison, the percentage dollar deviations for secured creditors are –3.54% and –2.63%, respectively, for OCRs and traditional Chapter 11s. The positive deviation of 0.20% for common stockholders in post-voted prepacks is smaller than the +9.51% and +2.28% positive deviations, respectively, in OCRs and traditional Chapter 11s. The positive deviation of +2.59% in pre-voted prepack is about the same as in traditional Chapter 11s and much smaller than that in OCRs. One interpretation of these results is that percentage dollar deviations are smaller for prepacks than for other forms of distressed restructurings. Alternatively, there may be small differences in the methodology or in the sample period that result in higher deviations in other studies. What does appear consistently across other studies and our own, however, is that compliance with absolute priority is greater when firms end up in court, or when voting occurs under the auspices of the court. Indeed, for the 11 firms in our prepack sample that modified the original reorganization plan while under court supervision, two firms made changes that did not affect the deviations from priority and the remaining nine made changes that resulted in smaller deviations from priority. Table 3 summarizes post-reorganization equity ownership for the firms in our sample in comparison with OCRs and traditional Chapter 11s. Creditors end up with 61.2% of the equity in pre-voted and 70.5% of the equity in post-voted prepacks. Both these numbers lie between the 41.9% for OCRs and the 79.2% for traditional Chapter 11s. In 76% of the prepack cases, creditors end up with at least 50% of the equity. Thus, in a typical prepack, control is transferred to creditors. On average, shareholders retain 21.6% of the equity in the reorganized firms. An important measure of how shareholders fare in a reorganization is the change in the market value of securities at critical points during the reorganization. We have stock price data for 21 of the prepack firms. For these companies, we calculate the excess stock returns around three announcements: (1) the initial indication of a restructuring attempt, (2) the Chapter 11 filing date, and (3) the confirmation date of the plan.9 The average announcement day excess returns are –3.9%, +3.2%, and +7.6%, all of which are statistically significant. Accordingly, the initial announcement of restructuring is bad news, while the prepack filing and confirmation are good news for shareholders. For traditional Chapter 11 filings, excess returns on the first two of these event dates are reported as –6.3% and –16.7%, both of which are significantly negative. These results suggest prepacks are good news relative to traditional Chapter 11s. For OCRs excess returns of –1.6% and +0.7% are reported at the initial restructuring and resolution dates, respectively.10 While prepack returns have the same signs as OCRs, neither of the OCR excess returns is statistically significant. Still, prepacks generate market reactions more similar to OCRs than traditional Chapter 11s. The reason for the small sample in the stock price study is that of the 49 firms in our sample, 44 had at least one publicly traded security, but only 23 had publicly-traded common stock (two firms’ publicly traded common stock had ceased trading by the initial restructuring announcement). The largest firm in the sample, Southland Corp., had total assets of $3.4 billion, but did not have publicly- traded common stock; the smallest firm, ARIX Corp., had assets of $9.7 million and did have publicly traded stock. The mean and median book value of total assets for firms in our sample are $570 million and $313 million, respectively. Thus, the relative frequency of firms with privately-held stock should not be attributed to the small size of the firms. The disproportionate number of private companies in our sample is more likely due to the fact that 22 of the firms underwent a leveraged buyout (LBO) within the 7-year period prior to the prepack. While the high number of former LBO firms in our sample suggests a link between the organizational form of the firm and the decision to reorganize using a prepack, we were unable to discover any statistical difference in any dimension we examined between those firms that had undergone an LBO and those that had not. All else equal, an efficient reorganization process will require a shorter time, have a lower cost, and will result in higher recovery rates than a less efficient reorganization process. Furthermore, an efficient process should result in low deviations from priority. With the exception of deviations from priority, the statistics in the tables suggest that OCRs are the most efficient form of reorganization, followed by prepacks, while traditional Chapter 11s are the least efficient form of distressed restructuring. Since prepack firms engage in extended pre-filing negotiations, why do these firms not merely reorganize out of court rather than file a prepack? McConnell and Servaes suggest three possible reasons. Two of these reasons relate to solving the holdout and free-rider problems that can arise in OCRs.11 Our sample provides some evidence on these points. For an OCR to be successful, significant debt relief must be achieved. Most OCRs specify that 90% or 95% of creditors must participate in order for the plan be implemented. The level of support necessary for a bankruptcy plan to be confirmed is much lower and, if confirmed, 100% of creditors must participate. Furthermore, the court can “cram down” the plan on especially recalcitrant creditors. Thus, the cramdown provision under Chapter 11 can resolve even the most severe holdout problem where either one powerful creditor or a group of creditors blocks a reorganization plan that has broad support among the remaining creditors.12 Although the cram-down provision has been invoked relatively often for equity-holders in Chapter 11 bankruptcies, the provision has seldom been used for creditors. However, for two firms in our sample of 49 prepacks, the reorganization plan was crammed down on creditors, as well.13 The case of E-II Holdings, Inc. illustrates how the cram-down provision of the Bankruptcy Code can be used to solve severe holdout problems. E-II was spun off in a 1987 leveraged buyout of Beatrice Companies. In 1991, E-II announced that it would stop paying interest on its bonds. After extended negotiations with an unofficial creditors’ committee, a plan of reorganization was proposed that provided debtholders a substantial equity stake in the firm. However, there was a major disagreement between senior and junior debtholders about the valuation of the firm. Senior debtholders favored a relatively conservative estimate of post-emergence value, which provided them a larger share of the firm's equity. Junior debtholders favored a higher valuation of the firm, which would reduce the proportion of equity required to pay senior debtholders in full. During the restructuring discussions, two investors who specialize in trading securities in financially distressed firms (“vulture” investors) took substantial positions in E-II's two debt issues. Carl Icahn acquired 31% of the junior issue and Leon Black's Apollo Advisors acquired 24% of the senior debt issue and 27% of the junior debt issue. Thus, either Icahn or Black could effectively block any out-of-court reorganization. In June 1992, E-II announced that an agreement in principle on a plan of reorganization had been reached with the creditors’ committee and filed its plan of reorganization. Although 90.5% in number of the voting junior debtholders subsequently supported the plan, clearing the 50% hurdle, only 59.5% in dollar amount cast favorable votes, thus falling short of the two-thirds requirement. Later E-II submitted a second plan wherein the estimated value of the firm was increased, thereby improving the apparent recovery rate for junior debtholders. The plan also gave senior debtholders the right to receive payment in equity rather than debt, thus giving them an option to maintain control of the firm. But, Icahn did not support the second plan because it did not give him a controlling equity position; therefore, the plan again failed to achieve the required level of support for confirmation. In the confirmation hearing, however, the court crammed down the firm's plan of reorganization on the dissenting junior debtholders, thereby circumventing the junior debtholders who had held out against the second plan. Because all security holders must participate in any exchange of securities in Chapter 11, a bankruptcy reorganization can help to solve the free-rider problem that can arise in an OCR. Creditors have an incentive not to exchange their old securities for new ones with less favorable terms and, thereby, to “freeride” on the concessions granted by other creditors even though the exchange would benefit all creditors collectively. Because all creditors must exchange securities in a Chapter 11 reorganization, bankruptcy can resolve the free-rider problem by removing the incentive to free-ride on the concessions of others. Our sample offers some insights into the way in which prepacks may provide a low-cost mechanism for solving the free-rider problem. In nine of the prepacks in our sample, the firm simultaneously mailed to creditors both a solicitation for an out-of-court exchange offer and a ballot for a prepackaged reorganization. The terms of the out-of-court restructuring and the prepack were identical. In each case, the firm indicated that the reorganization would be completed out of court if the exchange offer received sufficient participation. Because each of these firms ended up in our prepack sample, the OCR attempt obviously failed. In each of the four cases where we could obtain data, at least one class of claimholders gave a higher level of support for the prepackaged plan than for the proposed exchange offer. Apparently, the claimholders were more willing to participate in the prepack, which ensured 100% participation by claimholders, than in the identical exchange offer, which did not guarantee 100% participation. Consider the specific case of Gaylord Container. Gaylord Acquisition Corp., subsequently Gaylord Container Corp., was formed in 1986 to acquire assets in the paper industry. Gaylord made numerous acquisitions between 1986 and 1989. In 1990 Moody's lowered its rating on the firm's debt, citing higher-than-expected operating costs and high debt levels. Gaylord suspended interest payments on its subordinated notes in 1991. In 1992, following negotiations with creditors, Gaylord filed a registration statement with the SEC for an exchange offer with a back-up plan for a prepack should the exchange offer fail. The firm determined that to restructure successfully, 95% (in amount) of the subordinated debtholders must tender in the exchange offer. When the solicitation period expired, only 89% of the subordinated debt was tendered; but, at the same time, holders of 97% of this debt consented to the prepack. The firm's CEO concluded, “Clearly, holders of the subordinated debt opted for the prepackaged plan alternative that binds 100% of the bondholders and, therefore, treats all holders equally.”14 In sum, prepacks appear to provide a means of resolving the free-rider problem in reorganizations of financially distressed firms.15 Both the time spent in bankruptcy and the total time spent in reorganizing the firm are less with a prepack than with a traditional Chapter 11; The average direct cost of resolving financial distress is less in a prepack than in a traditional Chapter 11; The recovery rate by creditors is higher in a prepack than in a traditional Chapter 11; The incidence of violations of strict absolute priority is roughly the same as in traditional Chapter 11 bankruptcies; and The transfer of control to creditors in prepacks is similar to the transfer of control accomplished in traditional Chapter 11 reorganizations. But if these results suggest that prepacks are a “cheap” substitute for traditional Chapter 11 filings, there are also good reasons to believe that prepacks are not an option for many of the firms that end up in Chapter 11. For example, a large percentage of the firms that chose a traditional Chapter 11 reorganization may simply have lacked the financial resources necessary to continue operations throughout the relatively long pre-filing negotiation period (18 months, on average) typical of a prepack. Alternatively, our findings could also be used to support the case that prepacks are substitutes for OCRs in that prepacks offer an inexpensive solution to holdout and free-rider problems. This reading of the evidence would suggest that, before the rise of prepacks, if the costs of a traditional Chapter 11 were sufficiently greater than the costs of an OCR, even hesitant creditors may have been coerced into an out-of-court restructuring. But, insofar as a prepack now offers a low-cost mechanism for pressuring all creditors to participate, the firm may elect a prepack rather than an OCR. Our best guess is that both arguments have some validity. That is, we believe that some firms that would have chosen an OCR in the past now choose to reorganize by means of a prepack, while other firms that would have opted for a traditional Chapter 11 also choose to reorganize with a prepack. Lending support to this view are our findings that pre-voted prepacks have outcomes much closer to those of OCRs while post-voted prepacks look more like traditional Chapter 11s. In this sense, the advent of prepacks can be viewed as providing yet another option for financially distressed firms. Indeed, our findings can be interpreted as suggesting that prepacks themselves add to a range of possible solutions to financial distress that stretches between OCRs at one extreme and traditional Chapter 11s at the other. In each case, the solution reflects the working out of a process in which creditors and debtors are free to choose the method of reorganization that provides the greatest benefit at the lowest cost given their particular circumstances. It is our hope that the data provided here will be of use to debtors and creditors confronted with these choices.