Investing in bankrupt stocks: Trading, value, performance and losses

Available online at www.sciencedirect.
com
Journal of Financial Markets 16 (2013) 33–60

www.elsevier.com/locate/finmar
Investing in Chapter 11 stocks: Trading,

value, and performance$
Yuanzhi Lia,1, Zhaodong (Ken) Zhongb,n
a
Department of Finance, Fox Business School, Temple University, Philadelphia, PA 19122, United States
b
Department of Finance, Rutgers Business School, Rutgers University, Piscataway, NJ 08854, United States
Received 1 March 2012; received in revised form 10 September 2012; accepted 29 September 2012
Available online 31 October 2012
Abstract
We address questions about Chapter 11 stocks regarding their trading environment, fundamental
value, and performance. First, there exists active trading for Chapter 11 stocks throughout the
bankruptcy process. Second, equity value after filing is positively related to asset value, asset
volatility, risk-free rate, and expected duration and is negatively related to liabilities. Furthermore,
the return correlation between bankrupt stocks and their matching samples exhibits non-linearity
similar to out-of-money call options. Third, investing in Chapter 11 stocks incurs large losses.
Consistent with heterogeneous beliefs and limits to arbitrage, stocks with higher levels of information
uncertainty and more binding short-sale constraints experience more negative returns.
& 2012 Elsevier B.V. All rights reserved.
JEL classification: G33; G12; G14
Keywords: Chapter 11; Option theory; Heterogeneous beliefs; Limits to arbitrage
$
We want to thank Edward Altman for providing the bankruptcy resolution data and Carl Giangrasso from
Pink OTC Markets Inc. for providing the pricing data of delisted stocks. We also want to thank Edward Altman,
Nicholas Barberis, Hank Bessembinder, Kose John, Ronald Masulis, Stewart Mayhew, Robert Mooradian,
Stephen Treanor, Robert Whitelaw, Hong Yan, David Yermack, conference participants at the 2009 European
Finance Association annual meeting, and the seminar participants at Temple University for their helpful
comments. This research is supported in part by the David Whitcomb Center for Research in Financial Services
from the Rutgers Business School of Rutgers University. Special thanks go to an anonymous referee and Tarun
Chordia (the Editor) for their valuable suggestions that significantly improved the quality of our paper.
n
Corresponding author. Tel.: þ1 848 445 5109.
E-mail addresses: yuanzhi.li@temple.edu (Y. Li), zdzhong@business.rutgers.edu (Z. Zhong).
1
Tel.: þ1 215 204 8108.
1386-4181/$ - see front matter & 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.finmar.2012.09.006
34 Y. Li, Z. Zhong / Journal of Financial Markets 16 (2013) 33–60
1. Introduction
During the recent financial crisis, many firms fell into distress and filed for bankruptcy. The
number of total bankruptcy filings surged 32% in 2008 compared to 2007.2 Meanwhile, there
has been increasing coverage in the major media about investing in bankrupt equity after
Chapter 11 filing.3 However, this is an area that academic research has largely overlooked.
Many important questions involving the trading environment, fundamental value, and
performance of these stocks remain unanswered. Presumably, the lack of research is due to the
lack of data since most bankrupt stocks are delisted from major exchanges before or around
bankruptcy filings, rendering subsequent market data unavailable in traditional databases such
as CRSP. Using a unique dataset from Pink OTC Markets Inc., this paper sheds light on
these topics.
In our sample of 602 Chapter 11 filings from 1998 to early 2006, most stocks are delisted
from major exchanges and resume trading on Pink Sheets, an electronic quotation system
operated by Pink OTC Markets Inc. In contrast to the traditional view that trading is scarce
except for initial short-covering by previous short-sellers, we find that active trading activity
exists throughout the Chapter 11 process. Despite the initial decline in prices and widening
of bid–ask spreads, more than 50% of these stocks are traded on any given day, even for the
firms that have been in Chapter 11 for as long as three years. There is also a significant
decrease in the institutional ownership for these stocks accompanying the bankruptcy filings.
After the filing, more than 90% of the investors are individual investors.
Although in most cases shareholders eventually receive nothing in the final reorganiza-
tion or liquidation plans, our sample indicates that these stocks trade well above zero
immediately after Chapter 11 filings. The majority of previous studies (e.g., Franks and
Torous, 1989; Weiss, 1990, Eberhart, Moore, and Roenfeldt, 1990, Franks and Torous,
1994; Betker, 1995a) about bankrupt equity valuation focus on the violation of the
absolute priority rule (APR). APR violation occurs when creditors are not fully satisfied
before shareholders get any payments. In particular, Eberhart, Moore, and Roenfeldt
(1990) show that equity valuation right after filing reflects the market expectation of APR
violation. In contrast, we resort to the hypothesis that bankrupt equity is, although out-of-
money, a call option on firm assets. Chapter 11 allows the firm to continue its ongoing
business operation. It is possible that business might improve and the equity might be
evaluated in-the-money again.4 Applying the option theory enables us to expand the scope
of bankrupt equity valuation to include factors that have not been examined in the prior
literature, such as asset volatility, risk-free rate, and expected duration.
2
A more detailed description about the number of bankruptcy filings over time can be found at: http://www.
creditslips.org/creditslips/2009/01/bankruptcy-filings-rise-32-in-2008.html.
3
See ‘‘Betting on the Equity in Bankrupt Companies’’ by Aaron Pressman on June 11, 2009, in Business Week;
‘‘Zombie Stocks: Speculators Gamble on Firms Whose Shares May Be Worthless’’ by Ken Bensinger on
September 24, 2009, in Los Angeles Times; and ‘‘Bankrupt Firms: Who is Buying?’’ by Eugene Fama and Kenneth
French on February 23, 2010, at the online Fama/French Forum.
4
For example, ‘‘In the energy sector, some companies filed for bankruptcy after getting squeezed between the
credit crunch and plummeting oil prices. But oil prices have rebounded, increasing cash flow as well as the value of
potential drilling sitesy The shares have been trading for less than 25 cents since the filing; the analyst thinks they
could be worth $1.50 to $5y,’’ quoted from ‘‘Betting on the Equity in Bankrupt Companies’’ by Aaron Pressman
on June 11, 2009, in Business Week.
Y. Li, Z. Zhong / Journal of Financial Markets 16 (2013) 33–60 35
Using the Black-Scholes model to calculate the option values of bankrupt stocks, we are
able to explain about 25–35% of the cross-sectional variation of the observed market
values of these stocks. More specifically, we find that the bankrupt equity value right after
filing is positively related to asset value, asset volatility, risk-free rate, and expected
duration, and is negatively related to liabilities. This finding is robust even after controlling
for the expectation of APR violations.5 Turning to the payoff structure of investing in
these stocks, we show that buy-and-hold returns of bankrupt stocks from bankruptcy filing
to resolution are heavily right-skewed and resemble the payoff structures of out-of-money
call options: most of them die out without any payoff, while a few pay off tremendously.
To further investigate whether these payoffs are related to the underlying business
fundamentals in the way predicted by option theory, we construct a proxy for each firm’s
business performance using an industry-and-size matching sample over the Chapter 11
duration. We find that the return correlation between bankrupt stocks and their matching
samples exhibits strong non-linearity: it is significantly positive when the matching sample
performs well and is zero otherwise. This finding suggests that the payoff structure of
bankrupt stocks indeed has the same ‘‘hockey stick’’ shape as call options.
Examining the distribution of the buy-and-hold returns of Chapter 11 stocks, we find
that betting on these stocks on average generates large losses, both before and after risk
adjustments.6 Our sample’s median matching-sample-adjusted monthly return is 15%,
and market-adjusted monthly return is 14%. The negative abnormal returns do not
cluster in a particular year but persist over time. In addition, this finding is not attributable
to the negative delisting returns documented in the prior literature since more than half of
our sample firms are delisted two days before their bankruptcy filings. Therefore, our
finding of negative returns of Chapter 11 stocks from the first trading day after bankruptcy
filing to the final resolution is an indication of the poor performance during the Chapter 11
process, which can last from a few months to a few years.
It is surprising that investors lose so much money investing in Chapter 11 stocks, even given
the fact that shareholders are residual claim holders in bankruptcy. Thus, the finding that
Chapter 11 stocks underperform indicates the existence of market frictions. Our explanation
for the negative returns is motivated by the Miller (1977) theory, which argues that, when
investors have heterogeneous beliefs about the value of a risky asset in a market with restricted
short-selling, prices will reflect the more optimistic valuation.7 After Chapter 11 filings, these
stocks are mostly traded on Pink Sheets, which does not require information disclosure to
investors. Meanwhile, as the stock ownership data shows, institutional investors dramatically
reduce their stock holdings around bankruptcy filings, and more than 90% of the shareholders
post-filing are individual investors. Many analysts stop covering these stocks due to the lack
of interest from institutional investors. Individual investors are presumably less efficient in
5
It is worth mentioning that the option theory is not mutually exclusive with the APR literature. As
documented in prior studies, shareholders can engage in the bargaining process with creditors in the bankruptcy
court to lower the amount of debt paid back to creditors, which essentially lowers the strike price of the call option
and causes the option to be in-the-money.
6
Our finding of Chapter 11 stocks being out-of-money call options on firm assets is not inconsistent with the
negative returns. While the cross-sectional variation in the value of these stocks indicates they are call options on
firm assets, it does not rule out the possibility that they can be systematically mispriced, thus leading to negative
returns.
7
Another potential explanation is the behavioral model as in Barberis and Huang (2008), which we defer to
future research.
gathering information and interpreting the available information (Barber and Odean, 2000).
Therefore, the information uncertainty and the divergence of opinion regarding the true value
of these stocks increase dramatically after filing. In addition, low institutional ownership
produces binding short-sale constraints for these stocks.8 As a result, the high information
uncertainty and binding short sale constraints cause bankrupt stocks to be overvalued.9
Finally, a unique feature of Chapter 11 stocks is that, on the final resolution date, the
reorganization/liquidation plan states explicitly the true value of the stock.10 In other words, it
provides a condition for the stock prices to return to their fundamental values eventually.
Thus, the buy-and-hold return from filing to resolution becomes negative.
Many studies (e.g., Chen, Hong, and Stein, 2002, among others) develop price-optimism
models following the approach of Miller (1977). Two important implications of these
models are as follows: (1) the larger the disagreement about a stock’s value, the lower its
expected return; and (2) the more binding the short-sale constraint, the lower the expected
return. Using the bid–ask spread, intraday volatility, and turnover as proxies for
information uncertainty, and the percentage of institutional ownership and number of
institutional investors as proxies for short-sale constraints, we show that stocks with higher
levels of information uncertainty and more binding short-sale constraints experience more
negative returns during the Chapter 11 process.
The remainder of the paper is organized as follows: Section 2 provides a brief literature
review related to our paper. Section 3 describes the data used in this study. Section 4
investigates the trading environment and ownership of bankrupt stocks. Section 5 tests the
hypothesis that Chapter 11 stocks have fundamental values based on the call options on
firm assets. Section 6 tests the hypothesis that negative returns of Chapter 11 stocks are due
to the initial overvaluation by optimistic investors at filing in the context of heterogeneous
beliefs with short-sale constraints. We conclude in Section 7.
2. Related literature
The value of bankrupt equity has been scrutinized primarily under the framework of
violations of the absolute priority rule. Franks and Torous (1989) analyze a sample of
bankruptcies from 1970 through 1984 and find that 21 of 27 sample cases exhibit APR
violation. Weiss (1990) reports that priority of claims is violated in 27 cases in a sample of
8
D’Avolio (2002) shows that the main suppliers of stock loans are institutional investors, and Nagel (2005) finds
that stock loan supply tends to be sparse and short selling more expensive when institutional ownership is low.
9
Short-sellers may face various arbitrageur risks, even if they manage to borrow in the first place. Due to the
extreme lack of information about these stocks, rumors about bankrupt companies often lead to rapid increases in
the stock price without any justifiable reason, and the arbitrageurs will face high short-squeeze risk. The high
collateral requirement of short-selling relative to the low price level of these stocks imposes another type of
arbitrageur risk for bankrupt stocks. ‘‘The problem here is that an arbitrageur can tie up a lot of capital shorting
in-bankruptcy securities... if the stock price is $0.10 and the arbitrageur shorts 1 million shares, he receives only
$100 thousand from the sale but he must post collateral worth perhaps $1.2 million,’’ as quoted from ‘‘Bankrupt
Firms: Who is Buying?’’ by Eugene Fama and Kenneth French on February 23, 2010, at the Fama/French
Forum. Therefore, the difficulty to borrow Chapter 11 stocks and high arbitrageur risks explain why it is hard for
arbitrageurs to profit from the negative abnormal returns.
10
More precisely, this is true only when the equity is worthless, which accounts for most of sample cases. In
other cases, the value of old equity is specified as a conversion ratio to new stocks or warrants. Although the exact
value of new stocks and warrants is not known at the resolution time, it provides significantly more guidance to
value the old equity, compared to before the confirmation of a reorganization or liquidation plan.
37 bankruptcy filings between 1979 and 1986. Eberhart, Moore, and Roenfeldt (1990)
examine the amount paid to shareholders in excess of the amount they would have received
under the APR, and find that it represents an average of 7.6% of the total amount awarded
to all claimants from a sample of 30 filings from 1979 to 1986. They further show that
common share values reflect the value received in violation of APR, suggesting that
deviations from the rule are expected by the equity market. Betker (1995a) examines the
cross-sectional determinants of APR violations in 75 Chapter 11 bankruptcies during
1982–1990, and finds evidence that APR violations are larger when the firm is closer to
solvency, banks hold fewer claims, the CEO holds more shares, CEO pay and shareholder
wealth are positively related, and the firm retains the exclusive right to propose a
bankruptcy plan. Bharath, Werner, and Panchapegesan (2010) find that the APR
violations have become less common in recent periods, and they attribute it to the
development of debtor-in-possession (DIP) financing and key employee retention plans.
Several papers on bankruptcy also evaluate claims on bankrupt firms in an option
framework. Bebchuk (1988, 2000) proposes a new method of dividing the reorganization
pie among claim-holders in the Chapter 11 reorganization process by issuing tradable
options, replacing the current bargaining process. Bebchuk argues that when the true
firm value is unknown, bargaining is inefficient. However, the value of all claims on the
bankrupt firm can be synthesized using options with different strike prices, regardless
of the true value of the firm. Bebchuk and Chang (1992) develop a sequential bargaining
model of the negotiations between shareholders and creditors within Chapter 11.
Shareholders have incentives to delay the reorganization process to prolong the option
maturity they have on firm assets, while delayed reorganization can hurt overall firm value
due to additional financial distress costs. In the equilibrium, creditors are willing to
sacrifice some of their claims for shareholders to achieve faster reorganization resolution.
The argument is empirically supported by Franks and Torous (1994), who find that
Chapter 11 reorganizations have lower equity deviations from absolute priority and lower
creditor recovery rates, compared to out-of-court exchange offers. Building on the prior
literature, our paper conducts a comprehensive analysis with the newly available in-
bankruptcy data in the option framework.
There are a few papers studying stock performance before bankruptcy filing. For
example, Clark and Weinstein (1983) examine filing cases before the 1978 Bankruptcy
Reform Act and find evidence of large equity losses over long periods of time prior to
filing. Aharony, Jones, and Swary (1980) compare risk and return characteristics of
bankrupt stocks with matched non-bankrupt firms using pre-filing capital market data.
They find that the bankrupt sample displays significant negative cumulative returns
starting from four years before filing. Several studies have focused on the performance of
bankrupt firms and their securities after the bankruptcy resolutions. Hotchkiss (1995)
studies the accounting performance of bankrupt firms after they emerge from Chapter 11
and shows that more than 40% of them continue to experience operating losses in the
three years following emergence. Eberhart, Altman, and Aggarwal (1999) evaluate post-
bankruptcy performance of ‘‘orphan stocks’’—new issuance of equity for firms emerging
from bankruptcy. They report evidence of large positive excess returns in the 200 days
following emergence. Very few papers have examined the stock performance during the
Chapter 11 process. A few exceptions are Dawkins, Bhattacharya, and Bamber (2007),
who study the price patterns in the short period (one to two weeks) after Chapter 11 filings
and find a return reversal from the negative filing price reaction during bull markets, and
Morse and Shaw (1988), who examine a sample of 56 stocks traded after filing in the 1978–
1982 period and find no abnormal returns within three years. Our finding of substantial
negative returns for a large cross-section of firms during the Chapter 11 process contrasts
sharply with the findings in the prior literature.
Our paper is also related to the literature on heterogeneous beliefs and limits to
arbitrage. There have been both theoretical and empirical developments following the
seminal paper by Miller (1977). See, for example, Chen, Hong, and Stein (2002), Duffie,
Garleanu, and Pedersen (2002), and Hong, Scheinkman, and Xiong (2006), among others,
for new theoretical models, and see Figlewski (1981), Figlewski and Webb (1993), Diether,
Malloy, and Scherbina (2002), and Boehme, Danielsen, and Sorescu (2006), among others,
for empirical studies. Most previous empirical studies focus on one of the following two
conditions: the degree of disagreement and the short-sale constraints. Theoretically, the
overvaluation can last forever if the bubble persists over time. By studying the performance
of bankrupt stocks from the bankruptcy filing date to the final resolution date (when the
bubble ‘‘bursts’’), our paper provides a test that meets three conditions: the existence of
disagreement, short-sale constraints, and the eventual return of prices to their fundamental
values.
3. Data description
We first compile a sample of Chapter 11 filings between 1998 and early 2006 from an
online data vendor, http://www.bankruptcydata.com. The sample starts from 1998 since it
is when the daily trading information became available from Pink Sheets. On any given
day before or after Chapter 11 filings, if the stock is already delisted from major exchanges,
we collect the trading data from Pink Sheets, which provides the daily highest, lowest, and
closing prices, as well as bid–ask quotes and trading volume. If the stock is not delisted, we
collect the same data from CRSP. The final bankruptcy resolution dates and outcomes
are mainly provided by Edward Altman. We supplement the missing values with
Bankruptcydata.com and Lynn Lopucki’s Bankruptcy Research Database. If the firm is
successfully reorganized, the resolution date is the confirmation date of the reorganization
plan. If the firm is liquidated or converted to Chapter 7, the resolution date is the
confirmation date of the liquidation plan or the conversion date. In addition, whether a
firm negotiates its reorganization plan with creditors before filing (a ‘‘prepackaged’’ filing),
whether it receives additional financing after filing (DIP financing), and whether an equity
committee is formed are the important aspects of Chapter 11 filings as documented in the
bankruptcy literature (e.g., Betker, 1995b; Tashjian, Lease, and McConnell, 1996; Dahiya,
John, Puri, and Ramirez, 2003; Bharath, Werner, and Panchapegesan, 2010; among
others). Thus, we collect this information for all cases from three sources: http://www.
bankruptcydata.comLynn Lopucki’s Bankruptcy Research Database, and through news
search.11 We also gather the accounting information of these firms from the last quarterly
filings with the SEC before Chapter 11 filings. To guarantee that the accounting
information accurately describes sample firms’ conditions at the time of Chapter 11 filings,
we require all quarterly filings to be within 12 months of bankruptcy filing. Around 90% of
the sample firms have the last quarterly filing within six months before Chapter 11 filing.
11
In our sample, about 22% of cases are prepackaged filings, 41% of cases have DIP financing, and 27% of
cases have equity committees.
Table 1
Outcome and duration of chapter 11 filings.
Distribution of final outcomes and durations of Chapter 11 filings. The sample consists of 602 cases between
1998 and 2006. There are five categories of outcome: (1) Emerged or Reorganized; (2) Liquidated or Converted to
Chapter 7; (3) Acquired, Merged, or Sold; (4) Case Dismissed by the Court; and (5) Unknown. Duration is the
length of time from the Chapter 11 filing date to the final resolution date (reorganization confirmation date,
liquidation date, or the date of conversion to Chapter 7), which is measured in calendar days.
Outcome Number of Cases Percent Duration
Min Median Max Mean Std. Dev.
Emerged or Reorganized 260 43.2 29 266 1584 355 285

Liquidated or Converted to Ch7 227 37.7 13 401 1955 510 381
Acquired, Merged, or Sold 64 10.6 47 392 1495 457 286
Case Dismissed 33 5.5 11 490 2126 555 506
Unknown 18 3.0 43 299 1093 343 248
Total 602 100.0 11 350 2126 435 346
We also exclude cases from our study that do not have the same economic meaning as a
standard Chapter 11 filing, and the ones that are insignificant in terms of size. Specifically,
we drop firms in any of the following cases: (1) firms filing for Chapter 11 for non-distress
reasons such as litigations;12 (2) firms filing for Chapter 11 for the second time; (3) firms
with a pre-filing asset size of less than $1 million; and (4) firms with the first post-filing
trading price of less than $0.01. Our final sample consists of 602 Chapter 11 filing cases.
Table 1 shows the distribution of final resolution outcomes of Chapter 11 filings and
their durations. We classify the final outcomes into five categories: (1) Emerged or
Reorganized; (2) Liquidated or Converted to Chapter 7; (3) Acquired, Merged, or Sold; (4)
Case Dismissed by the court; and (5) Unknown. Note that even for the ‘‘Unknown’’ cases,
there is a confirmation date available in the data; only the outcome cannot be determined.
So these cases are by no means abandoned by the court or claim-holders. Among the 584
firms with known outcomes, successful reorganization cases account for slightly less than
half of the sample, while liquidation and acquisition cases account for the other half. We
also calculate the length of time (in terms of calendar days) from the bankruptcy filing
date to the final resolution date (reorganization confirmation date, liquidation date, or the
date of conversion to Chapter 7). The median duration of all bankruptcies is 350 days.
Interestingly, we find that firms that are reorganized successfully spend less time in
bankruptcy than the ones that are liquidated, with median durations of 266 days and 401
days, respectively. One reason is that the reorganized group includes more prepackaged
filings, which spend less time in bankruptcy. Our sample distribution of the resolution
outcome and duration is comparable to Bharath, Werner, and Panchapegesan (2010).
Their sample of Chapter 11 filings from 1979 to 2005 contains 626 filing cases, around 60%
of which are successfully reorganized, and the median duration is 14 months. The fact that
their sample firms are more likely to be reorganized successfully and tend to stay in
Chapter 11 longer is due to the larger asset size of their firms. The majority of their sample
12
There have been some Chapter 11 filings due to asbestos-related litigation. For example, ‘‘W. R. Grace & Co.
(NYSE: GRA) announced that the Company has voluntarily filed for reorganization under Chapter 11 of the
United States Bankruptcy Code in response to a sharply increasing number of asbestos claims on April 2, 2001,’’
as quoted from the official website of W. R. Grace & Co., http://www.grace.com.
firms are collected from Lynn Lopucki’s Bankruptcy Research Database, which only
covers filings by firms with asset size above $100 million.
Table 2 gives the descriptive statistics of variables reported or constructed with
information in the last quarterly filings with the SEC before Chapter 11 filings. We are able
to find or compute total assets, book equity, and leverage ratio for all sample firms. Only a
small number of firms have missing data for net income, tangible ratio, or return on assets
(ROA). Panel A provides the distribution summary statistics of these variables. Since we
require our sample firms to have a pre-filing asset size of more than $1 million, the
minimum total asset value is $1 million. The mean asset value of the sample is $857 million
with a standard deviation of $3,209 million, which means that the sample is not
concentrated on small or large firms. The averages of net income, book equity, and ROA
are negative. However, not all firms have negative net income, book equity, or ROA before
bankruptcy filings. The maximum net income is $48.21 million, the maximum book equity
is $6,584 million, and the maximum ROA is 10%. We hand-checked firms with positive net
income, book equity, or ROA to make sure that these firms indeed file for Chapter 11 due
Table 2
Firm characteristics.
Characteristics of firms that filed for Chapter 11 between 1998 and 2006. Panel A reports the summary
statistics. Panel B reports the correlation matrix (Pearson below the diagonal and Spearman above the diagonal).
Accounting information is from the last quarterly filing before Chapter 11 filings. Leverage is the ratio of total
liabilities over total assets; Tangible Ratio is the ratio of tangible assets over total assets; ROA is the ratio of
EBITDA (earnings before interest, taxes, depreciation, and amortization) over total assets.
Panel A: Summary statistics
N Min Q1 Median Q3 Max Mean Std. Dev.
Total Assets (MM$) 602 1.00 37.73 150.69 535.75 52285.60 856.63 3209.24
Net Income (MM$) 596 3197.00 38.77 9.25 2.67 48.21 61.88 226.19
Book Equity (MM$) 602 7316.70 37.60 0.11 27.46 6584.07 32.54 625.28
Leverage 602 0.06 0.76 0.96 1.25 12.53 1.16 0.98
Tangible Ratio 592 0.00 0.12 0.28 0.50 0.94 0.33 0.24
ROA 596 4.06 0.08 0.02 0.01 0.10 0.08 0.26
Panel B: Correlation matrix
Total Assets Net Income Book Equity Leverage Tangible Ratio ROA
Total Assets 1 0.60 0.06 0.01 0.26 0.50

(0.00) (0.12) (0.78) (0.00) (0.00)
Net Income 0.56 1 0.19 0.19 0.25 0.10
(0.00) (0.00) (0.00) (0.00) (0.02)
Book Equity 0.02 0.21 1 0.78 0.15 0.05
(0.65) (0.00) (0.00) (0.00) (0.26)
Leverage 0.04 0.01 0.15 1 0.11 0.00
(0.27) (0.76) (0.00) (0.01) (0.91)
Tangible Ratio 0.16 0.13 0.11 0.07 1 0.19
(0.00) (0.00) (0.01) (0.07) (0.00)
ROA 0.07 0.02 0.01 0.02 0.10 1
(0.07) (0.66) (0.77) (0.68) (0.02)
p-Values in parentheses.
to distress.13 Leverage is calculated as the ratio of total liabilities to total assets, with a
median value of 0.96.14 The tangible ratio ranges from 0% to 94%. The financial variables
indicate that sample firms are, on average, in bad operating condition and highly leveraged
at the time of their Chapter 11 filings. Panel B is the correlation matrix of firm
characteristics. It suggests that in our sample, there is no significant correlation between
size and leverage ratio. Larger firms tend to be more profitable and have higher tangible
asset ratios.
4. Trading of Chapter 11 stocks
In the bankruptcy literature, trading of distressed stocks after Chapter 11 filings is not
well studied. The common perception is that equity is wiped out upon Chapter 11 filings,
and thus trading must be scarce. The fact that traditional databases, such as CRSP, do not
cover delisted stocks also creates the problem of data availability. Using the daily trading
data provided by Pink OTC Markets Inc., our intent is to shed some light on this topic.
For each sample firm, we also collect the institutional ownership data before and after
the Chapter 11 filing from the Thomson-Reuters Institutional Holdings Database. This
dataset is based on Form 13F filed quarterly with the SEC by institutional managers with
$100 million or more in assets under management. With the available data, we mainly
investigate the following questions: (1) On any given day after filing until the final
bankruptcy resolution date, what is the percentage of Chapter 11 stocks that are traded?
(2) Who trades Chapter 11 stocks: institutional investors or individual investors?
To investigate the percentage of stocks traded in Chapter 11 on each trading day after
filing, we first calculate the total number of firms remaining in the Chapter 11 process after
filing. After 252 trading days (one calendar year), approximately 300 firms (around half of
the sample) have exited Chapter 11. This result is consistent with the overall duration
distribution in Table 1, which shows that the sample median duration is around 350
calendar days. We then use the number of firms remaining in the Chapter 11 process as the
denominator and the number of firms with available daily closing prices and trading
volume as the numerator to calculate the percentage of firms traded on each trading day.
Fig. 1 shows the percentage of traded Chapter 11 stocks on each trading day through the
three years following the Chapter 11 filing. Panel A suggests that, after a brief reduction of
trading in the few days immediately following Chapter 11 filing, trading picks up steadily
13
For example, the firm with the highest net income of $48.21 million is Waxman USA Inc., which filed for
Chapter 11 on October 2, 2000. Checking all 10-Q filings of the company before its Chapter 11 filing, we
discovered that the reported net incomes were negative except the last pre-Chapter 11 10-Q on September 30,
2000, which was because of the sale of a subsidiary. ‘‘The company completed the sale of all of the remaining
common stock of Barnett Inc. owned by Waxman USA Inc., a wholly owned subsidiary of the Company, and
applied the proceeds to repay Waxman USA’s 11 1/8% Senior Notes due 2001, reduce working capital
borrowings from Congress and to pay taxes and other expenses associated with the transaction,’’ as quoted from
its last 10-Q report. The firm with the highest ROA of 10% is Metrocall Inc., which filed for Chapter 11 on June 3,
2002. This is because we compute ROA using EBITDA, which is before depreciation. The company has a large
amount of depreciation of $18 million, which accounts for 10% of total assets. Thus, its net income is $29
million.
14
Marchfirst Inc. has the minimum leverage ratio of 6%. It filed for Chapter 11 on April 12, 2001. It is a short-
lived international systems integrator and Internet consulting company at the tail end of the dot-com boom. It was
a NASDAQ-traded public company whose peak stock price reached $52. By the time the company filed for
bankruptcy, it traded for pennies ($0.16 on March 28, 2001).
100%
90%
Percentage of Firms Traded

80%
70%
60%
50%
0 10 20 30 40 50
Event Date after Chapter 11 Filing
100%
90%
80%
Percrentage of Firms Traded
70%
60%
50%
40%
30%
20%
10%
0%
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750
Event Date after Chapter 11 Filing
Fig. 1. Percentage of Chapter 11 stocks traded after bankruptcy filing. Panel A: Up to 50 trading days after filing.
Panel B: Up to 750 trading days after filing.
within a month. Overall, as shown in Panel B, trading of Chapter 11 stocks happens far
more than what is generally perceived: the percentage is well above 50% on most trading
days, even after three years post-filing.
Given the frequent trading of Chapter 11 stocks, we turn to the question of who is
trading these stocks: Are they institutional investors or individual investors? We collect the
institutional ownership data in each quarter around the Chapter 11 filing time from the
Thomson-Reuters Institutional Holdings Database. Out of 602 sample firms, we are able
to find 592 firms covered at some point by this database. One relevant caveat of the
database is that it is only mandatory to report positions in exchange-traded securities.15
Thus, a firm can be missing from the dataset in any quarter for either of the two following
reasons: (1) the stock is exchange-traded, but the total institutional holding for the stock is
zero; or (2) the stock is delisted, thus institutional investors are not required by the SEC to
15
According to the SEC, ‘‘Section 13(f) securities generally include equity securities that trade on an exchange
or are quoted on the NASDAQ National Markety.’’ Official lists of Section 13(f) securities each quarter can be
found at: http://www.sec.gov/divisions/investment/13flists.htm.
report their holdings of the stock, which can be zero or positive. Since many of the sample
firms are delisted at some point around Chapter 11 filing, not all firms are covered in every
event quarter. To resolve this issue, we analyze the data in two ways.
As the first step of the analysis, we look at the number of firms found in the database
and the percentage of institutional ownership in each quarter around their Chapter 11
filings (see Panel A of Fig. 2). The x-axis is the event quarter relative to the Chapter 11
filing date, the left y-axis is the number of sample firms reported in Form 13F filings, and
the right y-axis is the percentage of institutional ownership. The number of sample firms
held and reported by institutional investors decreases sharply from well above 500 to
around 200 through one year before to one year after filing. Without making any
assumptions about the actual holdings for firms not found by the Form 13F Database, we
compute the sample median of percentage institutional ownership across reported firms
in each event quarter. Chapter 11 filings apparently have a significant negative effect
on institutional ownership. There is a persistent trend that institutional investors are unloading
the stocks as a firm approaches Chapter 11 filing. The median percentage decreases from
around 20% one year before filing to below 5% at the time of filing, and close to 1% one year
after filing.
600 0.20
Percentage of Institutional Ownership

500
No. of Firms with Institutional
0.15
400
Ownership Data
300 0.10
200
0.05
No. of Firms
100
Median of Reported Firms
0 0.00
-4 -3 -2 -1 0 1 2 3 4
Event Quarter Relative to Chapter 11 Filing
50 0.40
Percentage of Institutional Ownership
40
No. of Firms with Institutional
0.30
Ownership Data
30
0.20
20
0.10
10 No. of Firms
Median of Reported Firms
0 0.00
-4 -3 -2 -1 0 1
Event Quarter Relative to Chapter 11 Filing
Fig. 2. Institutional ownership of Chapter 11 stocks in event quarters. Panel A: Full sample. Panel B: Subsample
delisted one-quarter after Chapter 11 filing or never delisted.
In the second step, as shown in Panel B of Fig. 2, we conduct a more robust analysis by
studying only the subsample of 48 firms that remain exchange traded until the second
quarter after their Chapter 11 filings. Since it is mandatory to report positions in these
stocks on Form 13F through the period of quarter (4, 1) around filing time, the
institutional ownership data are robust in regards to the above-mentioned no-coverage
issue due to delisting. The median institutional ownership for these firms also decreases
monotonically from around 30% one year before filing to below 10% one-quarter after
filing. Considering that when institutional investors choose to hold Chapter 11 stocks, they
should prefer exchange-traded ones due to better liquidity, a conservative upper-bound for
average institutional ownership after Chapter 11 filing is 10%. Both panels in Fig. 2 thus
suggest that institutional investors start to unload bankrupt stocks before and around
Chapter 11 filings. This finding contrasts sharply with our earlier results on active trading
of Chapter 11 stocks, as shown in Fig. 1. Combining these results, we conclude that most
active trading of Chapter 11 stocks is by individual investors. It is worth mentioning that
our result is not inconsistent with the notion that institutional investors often engage in
‘‘vulture investing’’ by taking control of distressed firms. The latter is mostly done by
buying distressed debt rather than distressed equity (Gilson, 1990; Jiang, Li, and Wang,
2012).
We summarize our main findings on the trading of Chapter 11 stocks as follows:
(1) Active trading exists for Chapter 11 stocks on Pink Sheets. The percentage of stocks
traded is above 50%, even after the firms stay in Chapter 11 for three years. (2)
Institutional investors sell stocks upon the Chapter 11 filing news. Institutional ownership
of Chapter 11 stocks post-filing is less than 10%, suggesting that the active trading of
bankrupt stocks is mainly generated by individual investors.
5. The valuation of Chapter 11 stocks
As shown in the previous section, Chapter 11 filings have a significant negative effect on
stock prices. However, stock prices do not drop to zero after filing. Rather, they are mostly
traded well above zero, and trading remains active long after the filings. So what is the
fundamental value of these stocks? We argue that the value of Chapter 11 stocks comes
from the call option value on firm assets. Shareholders could end up getting no payoff in
the final reorganization plans or liquidation plans, but ex ante the option value should
be positive unless the chance of getting anything is zero. Chapter 11 allows the firm to
continue its business operations. It is possible that business might improve by the time the
firm is evaluated to determine payoffs to claim-holders. Meanwhile, as documented in the
APR violation literature, shareholders can negotiate with creditors in the bankruptcy court
to lower the amount of debt paid back, which essentially lowers the strike price of the call
option and causes the option to be in-the-money.
In the following sections, we use two different but related methods to test our option
pricing explanation of the valuation of Chapter 11 stocks. First, we use the Black-Scholes
pricing model to explain the initial value of Chapter 11 stock at the time of filing. Second,
we examine whether the payoff structure of buying the Chapter 11 stocks is related to the
firm’s fundamental value as predicted by the option theory. To obtain a useful sample for
the analysis in this section, we restrict the stocks to satisfy the following conditions: (1) the
first trading day after Chapter 11 filing is within 30 days after the filing date; (2) there is
trading on the bankruptcy resolution date, or within 30 days after the resolution date; and
(3) the time between the first trading day after Chapter 11 filing to the first trading day on
or after the bankruptcy resolution date is no less than 30 days.
5.1. The Black-Scholes call option value and market value of Chapter 11 stocks
Merton (1974) shows that when a firm’s debt is a zero-coupon bond, the equity value as
a contingent claim on firm value, V, can be solved using the Black-Scholes option formula,
where the underlying value process is the firm value process and the strike price is the face
value of the debt. When a firm needs to make interest payments before the debt maturity
date, or not all of its debts mature on the same date, this method is not very accurate.
However, there are no interest payments during the Chapter 11 process, and all debts need
to be paid off at the same time on the Chapter 11 resolution date. Thus, it is reasonable to
use the Black-Scholes formula to calculate the equity value after Chapter 11 filing as
follows:
CðV ,X ,r,T,s2 Þ ¼ VNðd1 ÞXerT Nðd2 Þ,
lnðV =X Þ þ ðr þ s2 =2ÞT
d1 ¼ pffiffiffiffi ,
pffiffiffiffi s T
d2 ¼ d1 s T , ð1Þ
where V is the current firm value, X is the strike price of the call option, s is the underlying
asset volatility, r is the risk-free rate, and T is the maturity of the option, i.e., the length of
time that the firm expects to stay in Chapter 11.
Measuring V, X, and r is relatively straightforward. Both V and X can be found in the
last pre-bankruptcy quarterly filing. We use total assets as firm value V.16 X is measured by
the face value of the sum of short-term debt and long-term debt. Risk-free rate r is the
annualized one-month T-bill rate in the month of Chapter 11 filing.17
The asset volatility s needs to capture the underlying business risk of the firm. We
compute it using a ‘‘matching firms’’ approach. We do not use any volatility measure of
the bankrupt firms themselves since the volatility of these firms can capture more of
the ‘‘filing uncertainty’’ rather than the underlying business risk. Thus, we use the average
asset volatility of matching firms for each bankrupt firm. The matching firms are
constructed by industry and size: for each bankrupt firm, at the time of Chapter 11 filing,
we select all the firms that are in the same Fama-French 48-industry group and in the same
decile sorted by book asset value. We calculate an annualized asset volatility for each firm
in the matching group of each bankrupt firm at the time of its Chapter 11 filing. The asset
volatility is constructed using the KMV-Merton model as in Bharath and Shumway
(2008).18 Then, the average of all matching firms’ asset volatilities is used as the proxy for
the bankrupt firm’s asset volatility s.
16
We also use an alternative measure defined as Firm Value ¼EVþShort Term Debtþ0.5nLong Term Debt,
where EV is the market value of equity one day after the Chapter 11 filing date. The results remain similar.
17
We get similar results using longer term Treasury rates.
18
Using daily data from the past one year, an initial annualized asset volatility is estimated as a standard
deviation of daily percentage change of firm asset value, which is assumed to be equal to daily equity market value
plus the book value of debt. Then the Black-Scholes formula is used to calculate the model implied daily firm asset
value given that the equity market value, which is the call option value, is observed each day. After that, a second
asset volatility is calculated with the model-implied daily firm asset value. The iteration continues until the two
asset volatility numbers converge (i.e., difference is smaller than 0.001).
The maturity of the call option T should be the expected Chapter 11 duration at the time
of the filing. Although we do observe realized durations of sample firms, they are not the
same as expected durations. Therefore, we approximate investors’ expectation with a
simple linear regression model to get the conditional mean. Denis and Rodgers (2007) find
that Chapter 11 duration is related to firm size, leverage, profitability, and industry. We
regress realized durations on these variables and use the fitted value as the expected durations.
The regression equation is: Duration ¼ a þ b1 Total Assets þ b2 Leverage Ratioþ b3 ROAþ
b4 Tangible Ratio þ GFama French Ten Industry Dummy Variables þ E. The accounting
variables are measured in the last quarterly filing before bankruptcy.
As a first-step analysis, we plot the observed market value of equity one day after
Chapter 11 filing against the Black-Scholes option value, as shown in Fig. 3. To
accommodate the heterogeneity of size among sample firms, we take the logarithm of the
variables. The scatter plot shows a significant relation between the observed market value
of Chapter 11 stocks and the theoretical Black-Scholes option value.
To further test the hypothesis that the market value of Chapter 11 stocks depends on the
intrinsic option value, we run a regression analyzing the relation of the two:
EV ¼ a þ b1 C þ e, ð2Þ
where EV is the logarithm of the market value of Chapter 11 stocks on the first trading day
after bankruptcy filing, and C is the logarithm of the Black-Scholes option value. In addition,
22
20
Market Equity Value
18
16
14
12
10
8
8 10 12 14 16 18 20 22
Black-Scholes Option Value
Fig. 3. Log market equity value against log Black-Scholes option value. The market equity value is observed one
day after Chapter 11 filings. The Black-Scholes option value is calculated using the call option formula:
CðV ,X ,r,T,s2 Þ ¼ VNðd1 ÞXerT Nðd2 Þ,
lnðV =X Þ þ ðr þ s2 =2ÞT
d1 ¼ pffiffiffiffi ,
s T
pffiffiffiffi
d2 ¼ d1 s T ,
where V is the value of total assets in the last pre-bankruptcy quarterly filing, X is the sum of short-term debt and
long-term debt in the last pre-bankruptcy quarterly filing, r is the risk-free rate at filing, s is the underlying asset
volatility constructed as the average of matching firms’ asset volatilities, and T is the predicted duration of the
Chapter 11 process.
we also expand the explanatory variables to the inputs of the Black-Scholes formula:
EV ¼ a þ b1 Assets þ b2 Volatility þ b3 Short-Term Debt þ b4 Long-Term Debt

þb5 Risk-Free þ b6 Duration þ Control Variables þ e, ð3Þ
where EV is the logarithm of the market value of Chapter 11 stocks on the first trading day
after bankruptcy filing, Assets is the logarithm of total assets, Volatility is the average annual
asset volatility of matching firms, Short-Term Debt is the logarithm of short-term debt, Long-
Term Debt is the logarithm of long-term debt, Risk-Free is the risk-free rate, and Duration is
the predicted duration. The Black-Scholes formula predicts that b1 40, b2 40, b3 o0, b4 o0,
b5 40, and b6 40.
It is worth noting that Franks and Torous (1994) consider the ‘‘option-to-delay,’’ which is a
form of threatening to enter Chapter 11 or once in Chapter 11, delaying the firm’s emergence
from reorganization. The main variable of interest in Franks and Torous is the ratio of debt to
value: options that are closer to being at-the-money, according to Franks and Torous,
increases shareholder bargaining power and causes more APR violations. In contrast, our
predictions of the option theory rely on the idea that the stock value upon filing also consists
of an embedded call option on firm assets. In other words, our predictions are based on a
collection of variables including total assets, debts, volatility, risk-free rate, and duration.
Therefore, the main difference distinguishing the option valuation we discuss here and the
APR violations related to option-to-delay discussed by Franks and Torous lies in the
predictions regarding the coefficients of volatility, risk-free rate, and duration.
To further control for the effect of APR violations on equity value, we use three dummy
variables: Prepack¼ 1 if the filing is pre-negotiated with creditors and 0 otherwise; DIP¼ 1 if
the firm receives DIP financing during bankruptcy and 0 otherwise; and Equity Committee¼ 1
if an equity committee is formed and 0 otherwise. Tashjian, Lease, and McConnell (1996)
document that the frequency of APR violations is higher for secured creditors in prepackaged
filings than in traditional Chapter 11 filings. Bharath, Werner, and Panchapegesan (2010) find
that the likelihood of an APR violation increases in prepackaged filings. They also suggest
that DIP financing could act as an effective deterrent to APR violations through stringent
restrictions for financing, and that the existence of an equity committee increases the probability
of an APR violation. If the expectation of an APR violation is a significant determinant of
stock value after bankruptcy filing, we expect the coefficient of Prepack to be positive, the
coefficient of DIP to be negative, and the coefficient of Equity Committee to be positive.
The estimation results are shown in Table 3. In model 1, the call option value C is
significantly related to the equity value EV and can explain 27.5% of the cross-sectional
variation of the latter. In model 2, all inputs in the Black-Scholes formula show the expected
signs as predicted by the option theory, and all of them are statistically significant except for
Short-Term Debt.19 The adjusted R-square further increases to 36.8%. It is worth noting that,
in any option pricing model, one of the most important determinants of option price is
volatility. Therefore, the success of our option-based explanation hinges on the coefficient for
Volatility, which turns out to be positive and significant in all specifications.20 Moreover, the
coefficients of Risk-Free and Duration also support the hypothesis that Chapter 11 stocks have
19
One possible explanation is that we use the last quarterly financial statement before the bankruptcy filing to
measure the liabilities at filing. Thus, long-term debt is a more reliable measure than short-term debt.
20
The fact that we have to use the volatility measure of matching firms as a proxy suggests that the results
should be even stronger if the asset volatility of the Chapter 11 firm is available.
Table 3
Explaining market value of Chapter 11 stocks with the Black-Scholes option value.
Coefficients and t-statistics (in parentheses) of the following regression equations:
EV ¼ a þ b1 C þ e,
where EV is the logarithm of the market value of Chapter 11 stocks on the first trading day after Chapter 11 filing,
and C is the logarithm of the Black-Scholes option value. The construction of C is described in Section 5.1.
EV ¼ a þ b1 Assets þ b2 Volatility þ b3 Short-Term Debt þ b4 Long-Term Debt þ b5 Risk-Free þ b6 Duration

þControl Variables þ e,
where Assets is the logarithm of total assets; Volatility is the average annual asset volatility of matching firms;
Short-Term Debt is the logarithm of short-term debt; Long-Term Debt is the logarithm of long-term debt; Risk-
Free is the risk-free rate; Duration (in calendar days) is the predicted duration with explanatory variables of total
assets, leverage ratio, ROA, tangible ratio, and the Fama-French 10 industry dummy variables; Prepack is a
dummy variable of whether the filing is pre-negotiated; DIP is a dummy variable of whether the firm receives
debtor-in-possession financing; and Equity Committee is a dummy variable of whether an equity committee is
formed.
Model 1 Model 2 Model 3 Model 4 Model 5
Intercept 7.075nnn 1.839n 2.039nn 2.677nnn 2.898nnn

(13.50) (1.83) (1.99) (2.86) (3.04)
C 0.411nnn
(14.03)
Assets 0.616nnn 0.600nnn 0.615nnn 0.599nnn
(13.87) (12.74) (13.79) (12.68)
Volatility 0.670nn 0.656nn 0.573nn 0.566nn
(2.50) (2.43) (2.16) (2.11)
Short-Term Debt 0.009 0.009 0.009 0.009
(0.78) (0.80) (0.80) (0.79)
Long-Term Debt 0.028nnn 0.028nnn 0.027nnn 0.027nnn
(3.48) (3.42) (3.36) (3.31)
Risk-Free Rate 7.880nn 8.414nn 8.071nn 8.442nn
(2.30) (2.44) (2.35) (2.44)
Duration 0.002nn 0.002nn
(2.20) (2.25)
Prepack 0.040 0.024
(0.28) (0.17)
DIP 0.050 0.028
(0.38) (0.21)
Equity Committee 0.265 0.262
(1.95)n (1.92)n
Adj-R2 27.5% 36.8% 36.9% 36.3% 36.4%
F-Value 196.93 51.14 34.59 59.95 37.98
N 518 518 518 518 518
n nn nnn
po0:10, po0:05, po0:01.
intrinsic option value. In model 3, we add the variables for APR violations to the regression
model. The coefficients of Prepack and DIP have the signs predicted by the APR literature,
but they are not significant. The coefficient of Equity Committee is positive and significant,
suggesting that APR violation expectations explain part of the equity value. More
importantly, after controlling for APR violation, the inputs of the option model remain
significant, suggesting that the option theory explains the valuation of bankrupt equity beyond
the expected APR violation.21
To address the concern that predicted duration as an explanatory variable could be
correlated with the other variables (debt and asset), we run robustness checks excluding the
predicted duration variable from the explanatory variables. The results, as shown in models 4
and 5 of Table 3, remain similar to those in models 2 and 3. The indicator of the success of the
option theory: the coefficient on the asset volatility variable is still significantly positive. The
coefficients of all other option variables also have the same signs as predicted.
5.2. The payoff structure of Chapter 11 stocks
If Chapter 11 stocks are indeed out-of-money call options on firm assets, the holding period
return of these stocks should depend on the underlying firm performance in a non-linear way:
the correlation of the two should be high when firm performance is high and zero otherwise.
Since the underlying firm performance is not directly observable, we construct a proxy for it
using the performance of an equal-weighted portfolio of industry- and size-matching firms
over the Chapter 11 duration of the bankrupt firm. The matching firms are constructed the
same way as in the asset volatility calculation described in Section 5.1. Since all matching firms
are in the same decile, the equal-weighted and value-weighted results are similar. Furthermore,
we calculate the buy-and-hold return of each matching firm before we take the average to
avoid rebalancing bias. When calculating the buy-and-hold return for each matching firm, we
take into account the delisting return following the method used in the prior literature.22
The holding period return of Chapter 11 stocks is defined as the return from the first
trading day after Chapter 11 filing to the bankruptcy resolution date. If there is no trading
on the bankruptcy resolution date, we use the price on the first trading day after the
resolution date. We consider this period to be the holding period. Denote holding period
returns (HPR) by23
HPR ¼ ðPlast Pfirst Þ=Pfirst , ð4Þ
where Plast is the price on the last trading day of the holding period, and Pfirst is the price
on the first trading day of the holding period.
Fig. 4 shows the scatter plot of the HPR of Chapter 11 stocks against the buy-and-hold
return of an equal-weighted portfolio of matching firms over the same period.24 There are
quite a few stocks with very large positive HPR on the right-hand side of the graph (where the
performance of matching firms is relatively good), while most stocks have returns close to
100% on the left-hand side. To systematically test the relation between the performance of
Chapter 11 stocks and the matching firms, we estimate a piecewise linear regression:
HPRi ¼ ð1yi Þnðglow þ blow nHPRmatch,i Þ þ yi nðghigh þ bhigh nHPRmatch,i Þ, ð5Þ
21
Our results remain the same if we use moneyness (the ratio of debt over total assets) to replace the level of
debts in the analysis.
22
When a stock is delisted, we include its delisting return in CRSP if it is available. If the delisting return is
missing in CRSP and the delisting code is either 500 or falls between 505 and 588, we assume the stock is delisted
for negative reasons. For these stocks, we replace the missing delisting return with 30% for NYSE/AMEX
stocks and 55% for NASDAQ stocks.
23
Since there is no dividend for Chapter 11 stocks, the only gain/loss is the price appreciation or depreciation.
24
Due to scale limitation, a few extreme values, such as [1.01, 141], are not shown in the graph.
15
14
13
12
11
Holding Period Return

10
9
8
7
6
5
4
3
2
1
0
-1
-1 -0.5 0 0.5 1 1.5
Matching Sample Return
Fig. 4. Holding period return of Chapter 11 stocks against matching sample returns. The holding period return of
Chapter 11 stocks is defined as: HPR ¼ ðPlast Pfirst Þ=Pfirst , where Plast is the price on the last trading day of the
holding period, and Pfirst is the price on the first trading day of the holding period. The holding period is from the
first trading day after filing to the resolution date, or the first trading day after the resolution date if there is no
trading on the resolution date. The matching sample return is the buy-and-hold return of an equal-weighted
portfolio matched by industry and size over the same holding period.
where HPRi is the holding period return for the bankrupt stock i; HPRmatch,i is the buy-and-
hold return of an equal-weighted portfolio of matching firms for bankrupt stock i over the
same holding period; and yi is a dummy variable (1 if HPRmatch,i is above a certain threshold,
and 0 otherwise).
To ensure continuity, we impose the following restriction on glow , blow , ghigh and bhigh :
glow þ blow nThreshold ¼ ghigh þ bhigh nThreshold: ð6Þ
We set Threshold ¼ 0 exogenously to estimate the parameters. The regression results show
that blow ¼ 0:10 with t-statistic ¼ 0.36, and bhigh ¼ 2:40 with t-statistic¼ 2.46. Panel A of
Fig. 5 plots the fitted lines of the piece-wise regression for the entire sample. The return
correlation between a bankrupt stock and its matching firms exhibits strong non-linearity:
the correlation is significantly positive when the return of matching firms is positive and is
zero otherwise. This finding suggests that the returns to Chapter 11 stocks are similar to
those obtained from buying out-of-money call options on equal-weighted portfolios of
matching firms.
We further conjecture that the option payoff structure could be more pronounced for
some Chapter 11 stocks than others, depending on whether a firm is expected to be
reorganized. The option value should only be viable if the firm is expected to continue its
business, and there is a time value regarding its future business success. If a firm is expected
to be liquidated, then there is hardly any option value embedded in equity. In this case, the
return correlation of the stock and matching firms should be close to zero regardless of the
performance of the latter. Panels B and C in Fig. 5 show the fitted lines of piece-wise
regressions for two subsamples: the ones successfully reorganized and the ones liquidated
or converted to Chapter 7. For the subsample that is reorganized, the resemblance to
option payoffs is more dramatic: blow ¼ 0:34 with t-statistic ¼ 0.99, and bhigh ¼ 2:23 with
t-statistic ¼ 3.22. For the subsample that is eventually liquidated or converted to Chapter 7,
it is almost one flat line: blow ¼ 0:11 with t-statistic ¼ 0.18, and bhigh ¼ 0:04 with
t-statistic ¼ 0.26.
4
Matching Sample Return
5 0.2
4 0
Matching Sample Return Matching Sample Return
Fig. 5. Piecewise linear regression of Chapter 11 stocks’ holding period returns on matching sample returns. The
regression equation is as follows:
HPRi ¼ ð1yi Þnðglow þ blow nHPRmatch,i Þ þ yi nðghigh þ bhigh nHPRmatch,i Þ,

with the restriction of glow þ blow nThreshold ¼ ghigh þ bhigh nThreshold, where HPRi is the holding period return for
the bankrupt stock i; HPRmatch,i is the buy-and-hold return of an equal-weighted portfolio of matching firms for
bankrupt stock i over the same holding period; and yi is a dummy variable (1 if HPRmatch,i is above a threshold,
and 0 otherwise). We set Threshold¼ 0. Panel A: Full sample. Panel B: Emerged or reorganized. Panel C:
Liquidated or converted to Chapter 7.
In summary, the analysis on the value of Chapter 11 stocks and their return correlation
with underlying firm performance supports the hypothesis that Chapter 11 stocks are
traded well above zero because they have a fundamental option value on firm assets.
6. The performance of Chapter 11 stocks
We investigate the performance of Chapter 11 stocks in the holding period as defined in

Section 5.2. Since some Chapter 11 cases get resolved earlier than others, HPR of different
stocks represents performances over different time horizons. Thus, we calculate a standardized
Table 4
Performance of Chapter 11 stocks.
Return distributions of Chapter 11 stocks. The holding period is defined as the number of days from the first
trading day after filing to the resolution date, or the first trading day after the resolution date if there is no trading
on the resolution date. The holding period return (HPR) is defined as: HPR ¼ ðPlast Pfirst Þ=Pfirst , where Plast is the
price on the last trading day of the holding period, and Pfirst is the price on the first trading day of the holding
period. The average monthly return (MHPR) is defined as: MHPR ¼ ð1 þ HPRÞ21=t , where t is the number of
trading days during the holding period. The market-adjusted average monthly return (MHPRMarketAdj) is defined
as: MHPRMarketAdj ¼ ð1 þ HPRÞ21=t ð1 þ HPRMarket Þ21=t , where HPRMarket is the buy-and-hold return of the
value-weighted NYSE/AMEX/NASDAQ market index over the holding period. The industry- and size-adjusted
average monthly return (MHPRMatchAdj) is defined as: MHPRMatchAdj ¼ ð1 þ HPRÞ21=t ð1 þ HPRMatch Þ21=t , where
HPRMatch is the buy-and-hold return of an equal-weighted portfolio matched by industry and size over the
holding period.
HPR MHPR MHPRMarketAdj MHPRMatchAdj
Min 1.00 0.96 0.98 1.10

Median 0.88 0.15 0.14 0.15
Max 141.00 0.45 0.43 0.51
Mean 0.06 0.16 0.16 0.16
Std. Dev. 6.45 0.19 0.19 0.19
Skewness 20.08 0.37 0.45 0.53
Kurtosis 435.16 1.83 1.86 2.43
t-Value 0.22 19.48 19.52 19.88
p-Value of student’s t 0.83 0.00 0.00 0.00
p-Value of sign 0.00 0.00 0.00 0.00
p-Value of signed rank 0.00 0.00 0.00 0.00
N 531 531 531 531
monthly holding period return, MHPR ¼ ð1 þ HPRÞ21=t , where t is the number of trading
days in the holding period. We then calculate the risk-adjusted returns in two ways: using the
buy-and-hold return of the value-weighted NYSE/AMEX/NASDAQ market index, and using
the buy-and-hold return of an equal-weighted portfolio matched by industry and size over the
holding period as described in the previous section.25 They are defined as follows:
MHPRMarketAdj ¼ ð1 þ HPRÞ21=t ð1 þ HPRMarket Þ21=t , ð7Þ
MHPRMatchAdj ¼ ð1 þ HPRÞ21=t ð1 þ HPRMatch Þ21=t : ð8Þ

Table 4 gives distribution statistics of HPR, MHPR, MHPRMarketAdj, and MHPRMatchAdj for
sample stocks. The mean and median for all return measures are negative. Specifically, the
median market-adjusted monthly return is 14% and the industry- and size-adjusted monthly
return is 15%. Because HPR is heavily left-skewed, we conduct the Wilcoxon rank tests for
all return measures to examine whether they are different from zero. The p-values are all close
to zero, suggesting that the performance of Chapter 11 stocks is significantly negative both
before and after the risk adjustment.
Table 5 describes the distribution statistics of HPR and MHPRMatchAdj sorted by the initial
price level, final outcome, and delisting time. The initial price level is a proxy for firm size. The
final outcome is relevant because successfully reorganized firms are supposedly the ones with
25
Using equal-weighted market index return or value-weighted matching portfolio return does not change the
results.
Table 5
Distribution statistics of HPR and MHPRMatchAdj sorted by price, outcome, and delisting time.
Distribution statistics of HPR and MHPRMatchAdj sorted by the initial price level right before filing, resolution
outcome, and delisting time. The holding period return is defined as: HPR ¼ ðPlast Pfirst Þ=Pfirst , where Plast is the
price on the last trading day of the holding period, and Pfirst is the price on the first trading day of the holding
period. The industry- and size-adjusted average monthly return is defined as: MHPRMatchAdj ¼ ð1 þ HPRÞ21=t
ð1 þ HPRMatch Þ21=t , where HPRMatch is the buy-and-hold return of an equal-weighted portfolio matched by
industry and size. There are three sub-groups by the initial price level: (0, $0.1), ($0.1, $0.5), and ($0.5, $10).
Resolution outcomes are classified as either (1) Successfully Reorganized or (2) Others. Delisting time is classified
as either (1) Delisted within the holding period or (2) Others.
Panel A: Distribution statistics of HPR sorted by price, outcome, and delisting time
N Mean Std. Dev. Min Median Max
(0, 0.1) 170 0.28 2.31 1.00 0.71 27.47

Initial Price ($1) (0.1, 0.5) 237 0.25 9.40 1.00 0.90 141.00
(0.5, 10) 124 0.36 1.40 1.00 0.92 9.96
Resolution Outcome Successfully Reorganized 237 0.06 2.12 1.00 0.63 21.31
Others 294 0.06 8.46 1.00 0.95 141.00
Delisting Time Delisted within Holding Period 163 0.43 1.29 1.00 0.91 9.96
Others 368 0.10 7.69 1.00 0.83 141.00
Panel B: Distribution statistics of MHPRMatchAdj sorted by price, outcome, and delisting time
N Mean Std. Dev. Min Median Max
(0, 0.1) 170 0.16 0.21 1.10 0.13 0.38

Initial Price ($1) (0.1, 0.5) 237 0.17 0.18 0.75 0.16 0.51
(0.5, 10) 124 0.15 0.17 0.82 0.15 0.33
Resolution Outcome Successfully Reorganized 237 0.12 0.17 0.66 0.12 0.51
Others 294 0.20 0.19 1.10 0.18 0.30
Delisting Time Delisted within Holding Period 163 0.16 0.17 0.82 0.16 0.33
Others 368 0.16 0.19 1.10 0.15 0.51
better operating performance during Chapter 11. We expect these stocks to exhibit better
returns, as long as the resolution outcome is not 100% predictable at the time of filing.
Delisting is generally a result of bad performance, and in return it also affects stock prices
adversely, as shown in Harris, Panchapagesan, and Werner (2008). Results shown in Table 5
confirm that firms eventually reorganized experience less negative returns, while firms delisted
within Chapter 11 have more negative returns. However, returns do not seem to depend on the
initial price level of the stocks. In addition, untabulated results indicate that the negative
returns are persistent over time and do not cluster in a particular year.
Given the observation of large negative abnormal returns of Chapter 11 stocks in the
sample, we investigate two related questions. First, why do the negative abnormal returns
exist in the first place? Second, given the existence of the negative abnormal returns, why
do not arbitrageurs short the stocks and correct the prices? Our answers to these questions
are motivated by the information uncertainty surrounding the Chapter 11 stocks at
bankruptcy filings and the difficulty of short-selling these stocks. Miller (1977) argues that
the very uncertain nature of financial markets implies heterogeneous evaluations of the
same financial asset. Therefore, in a market with restricted short selling and when investors
have heterogeneous beliefs about the value of a risky asset, prices will reflect a more
optimistic valuation. In the next section, we show that this is exactly the case for the
valuation of stocks after Chapter 11 filings. Furthermore, we address a unique feature of
Chapter 11 stocks: at the final bankruptcy resolution date, the reorganization/liquidation
plan states explicitly the true value of the stock. At this point, fundamental value is
revealed, and prices are corrected to reflect the true fundamental value. As a result, the
buy-and-hold return from filing to resolution becomes negative.
6.1. Information uncertainty, short-sale constraints, and arbitrageur risk of Chapter 11 stocks
When a firm files for Chapter 11, it significantly increases the information uncertainty of its
stock. There are several reasons for this effect. First, most stocks are delisted before or around
Chapter 11 filings. They can resume trading on Pink Sheets, where listing rules are far less
strict than in the major exchanges. SEC filings and disclosures are not mandatory for these
stocks. Second, most institutional investors, such as mutual funds and pension funds, cannot
hold bankrupt stocks, as stipulated in their investment mandates. Thus, sell-side analysts lack
the incentive to research and publish on these stocks.26 Therefore, the problem of scarce
information available for Chapter 11 stocks is further exacerbated. Third, as the stock
ownership data shows, more than 90% of the investors are individual investors post-filing.
Presumably, individual investors are less efficient in gathering information and less adept at
correctly interpreting the available information (Barber and Odean, 2000). The increased
information uncertainty can clearly be seen in the data of average quote spread and intraday
volatility around the filing date. Fig. 6 plots the time series of quoted spreads and intraday
volatilities from one month (21 trading days) before to one month after the filing date. The
large spike for both series centering around the Chapter 11 filing date shows dramatically
increased information uncertainty of stocks due to filings.
Short-selling is generally very difficult for Chapter 11 stocks traded on Pink Sheets, and
short-sellers face high arbitrageur risks even if they manage to short these stocks in the first
place. First, most institutional investors do not hold these stocks. As shown in D’Avolio
(2002), the main suppliers of stock loans are institutional investors. In addition, Nagel (2005)
finds that stock loan supply tends to be sparse and short-selling more expensive when
institutional ownership is low. Therefore, the substantially increased search and borrowing
costs make bankrupt stocks much less attractive to arbitrageurs. Second, Pink Sheets is not as
liquid as major exchanges, especially for bankrupt stocks. Quoted spreads for some can be
as high as 30%, as shown in Fig. 6. Therefore, the recalling risk from share lenders is
substantially higher for these stocks: If the borrowed shares are called, it may be difficult for
the arbitrageur to find a new lender, and it will be expensive to buy the shares in the market.27
6.2. The cross-section of Chapter 11 stock returns
If the negative abnormal returns of Chapter 11 stocks are indeed caused by the initial
overvaluation by optimistic investors at the time of filing, cross-sectionally stocks with higher
26
As pointed out by Harris, Panchapagesan, and Werner (2008), there is limited analyst coverage on stocks
trading on Pink Sheets.
27
Short-sellers of Chapter 11 stocks are vulnerable to a short squeeze because rumors about bankrupt
companies often lead to a rapid increase in the stock price without any justifiable reason. In addition, the high
collateral requirement of short-selling relative to the low price level of these stocks imposes another type of
arbitrageur risk for bankrupt stocks.
0.35
0.3
Average Quoted Spread
0.25
0.2
0.15
Trading Days around Chapter 11 Filing
0.8
Average Intraday Volatility
0.6
0.4
0.2
Trading Days around Chapter 11 Filing
Fig. 6. Average quoted spread and intraday volatility around Chapter 11 filing. Average Quoted Spread is the
cross-sectional average of quoted spread (bid–ask spread divided by the bid–ask midpoint); and Average Intraday
Volatility is the cross-sectional average of intraday volatility (difference between the daily highest price and lowest
price divided by the closing price). Panel A: Quoted spread. Panel B: Intraday volatility.
information uncertainty and more severe short-sale constraints should experience more
negative returns after filing. Glosten and Milgrom (1985) show that dealer bid–ask spreads are
related to heterogeneous information among traders. Volatility is used as a standard measure
of information uncertainty in the literature (see Zhang, 2006, among others). Scheinkman and
Xiong (2004) provide a detailed survey of literature analyzing how heterogeneous beliefs
among investors generate speculation and trading volume. Following the previous literature,
we measure the level of information uncertainty with the following variables: the average daily
bid–ask spread, the average intraday volatility, and the average trading turnover over the
Chapter 11 duration. To measure the short sale constraint, we use the proportion of shares
held by institutional investors. Following Chen, Hong, and Stein (2002), we also use the
breadth of institutional ownership as another measure of short sale constraint, which is the
number of institutional investors holding the stocks. We estimate the following regression:
MHPRMatchAdj ¼ a þ b1 Quoted Spread þ b2 Intraday Volatility þ b3 Turnover
þb4 Institutional Ownership þ Control Variables þ e, ð9Þ
where MHPRMatchAdj is the industry- and size-adjusted monthly return; Quoted Spread is the
average daily quoted spread (bid–ask spread divided by the bid–ask midpoint) over the
holding period for each firm; Intraday Volatility is the average intraday volatility (difference
between the daily highest price and lowest price divided by the closing price) over the holding
period for each firm; Turnover is the average trading volume over shares outstanding over the
holding period; and Institutional Ownership is measured by either the proportion of shares held
by institutional investors at the end of the quarter when the firm files for Chapter 11, or the
logarithm of one plus the number of institutional investors.
The results, shown in Table 6, confirm our hypothesis that stocks with higher quoted
spreads, higher intraday volatilities, and higher turnover have significantly lower returns,
and that stocks with lower institutional ownership (more severe short-sale constraints)
have lower subsequent returns in Chapter 11.28 Our proxies for information uncertainty
and short-sale constraints alone can explain a substantial amount of the cross-sectional
variation in stock returns, with adjusted R-squared equal to about 17%. Moreover, all of
our proxies for information uncertainty and the percentage of institutional ownership
remain highly significant after controlling for firm characteristics and industry-year fixed
effects. Examining the effects of control variables on post-filing performance, we find that
filing cases that are prepackaged have better returns during Chapter 11. While such a
finding is consistent with the APR literature prediction, it also may be that these firms
have less information uncertainty, and so the optimistic bias of investors is less severe.
Consistent with the APR literature prediction, the existence of an equity committee is
positively related to return. Finally, other pre-filing firm characteristics, such as size,
leverage, and profitability, have little explanatory power after considering the information
variables and short-sale constraint proxies. Overall, our findings indicate that the negative
returns of Chapter 11 stocks are mainly caused by initial overvaluation by optimistic
investors in the context of heterogeneous beliefs with short-sale constraints.
Since average daily bid–ask spreads and average turnover are also liquidity measures,
Table 6 can be interpreted as controlling for the liquidity condition of the stock,
information uncertainty proxied by the average intraday volatility is negatively related to
returns in Chapter 11. Additionally, as a robustness check, we run the same regression as
in Table 6 excluding these two variables, and use the average intraday volatility as the sole
measure for information uncertainty. The results essentially remain the same, as shown in
Table 7. The magnitude and statistical significance of the coefficient on Intraday Volatility
are both larger as compared to in Table 6, suggesting the other two measures also correlate
with the level of information uncertainty.
Our findings that institutional investors are dropping Chapter 11 stocks imply that the
individual investors may have beliefs that are different from institutional investors
regarding the value of these stocks. There are many well-documented behavioral biases
related to individual investors (see a survey by Barberis and Thaler, 2003). In particular,
Barberis and Huang (2008) rely on an idea in the field of psychology that the brain weights
probabilities in a nonlinear way. They show that, in a financial market, such psychology
bias will cause some investors to overvalue risky assets with lottery-like payoffs. Given our
findings that Chapter 11 stocks resemble out-of-money call options, these stocks provide
lottery-like payoffs to investors. Therefore, our results are consistent with the idea that
individual investors are more likely than institutional investors to suffer from the probability
28
Results using market-adjusted monthly returns are similar and available upon request.
Table 6
Explaining the performance of Chapter 11 stocks.
Coefficients and t-statistics (in parentheses) of the following regression equation:
MHPRMatchAdj ¼ a þ b1 Quoted Spread þ b2 Intraday Volatility þ b3 Turnover þ b4 Institutional Ownership

þControl Variables þ e,
where MHPRMatchAdj is the industry- and size-adjusted monthly return; Quoted Spread is the average daily quoted
spread (bid–ask spread divided by the bid–ask midpoint) over the holding period for each firm; Intraday Volatility
is the average intraday volatility (difference between the daily highest price and lowest price divided by the closing
price) over the holding period; Turnover is the average trading volume over shares outstanding over the holding
period; Institutional Ownership is measured by either the proportion of shares held by institutional investors at the
end of the quarter when the firm files for Chapter 11, or the logarithm of one plus the number of institutional
investors; Delisted is a dummy variable of the delisting time (1 if the stock is delisted within the Chapter 11
process, and 0 otherwise); Prepack is a dummy variable of whether the filing is pre-negotiated; DIP is a dummy
variable of whether the firm receives debtor-in-possession financing; and Equity Committee is a dummy variable of
whether an equity committee is formed. Other control variables are defined as in Table 2.
Model 1 Model 2 Model 3 Model 4
Intercept 0.090nnn 0.107nnn 0.132 0.096

(6.18) (4.55) (1.26) (0.91)
Quoted Spread 0.125nnn 0.114nnn 0.090nnn 0.088nnn
(5.19) (4.50) (3.24) (3.13)
Intraday Volatility 0.057nnn 0.062nnn 0.067nnn 0.070nnn
(4.00) (4.30) (4.60) (4.74)
Turnover 1.691nnn 1.730nnn 1.594nnn 1.669nnn
(4.45) (4.48) (3.80) (3.93)
Percentage of Institutional Ownership 0.252nnn 0.282nnn
(2.70) (2.82)
Breadth of Institutional Ownership 0.017n 0.011
(1.75) (0.89)
Log(Total Assets) 0.002 0.003
(0.31) (0.36)
Leverage 0.005 0.002
(0.43) (0.16)
Tangible Ratio 0.049 0.049
(1.30) (1.30)
ROA 0.018 0.014
(0.61) (0.48)
Delisted 0.002 0.002
(0.10) (0.09)
Prepack 0.056nnn 0.052nn
(2.78) (2.58)
DIP 0.006 0.006
(0.30) (0.30)
Equity Committee 0.034n 0.032n
(1.83) (1.72)
Industry and Year Dummies Included Included
Adj-R2 17.0% 16.2% 21.0% 19.6%
F-Value 23.23 21.96 4.88 4.55
N 436 436 424 424
n nn nnn
po0:10, po0:05, po0:01.
weighting bias, thus causing the heterogeneous beliefs regarding Chapter 11 stocks. A detailed
analysis of the behavioral biases related to individual investors is outside the scope of our
paper, and we will leave it for future research.
Table 7
Explaining the performance of Chapter 11 stocks—robustness checks.
Coefficients and t-statistics (in parentheses) of the following regression equation:
MHPRMatchAdj ¼ a þ b1 Intraday Volatility þ b2 Institutional Ownership þ Control Variables þ e,

where MHPRMatchAdj is the industry- and size-adjusted monthly return; Intraday Volatility is the average intraday
volatility (difference between the daily highest price and lowest price divided by the closing price) over the holding
period; Institutional Ownership is measured by either the proportion of shares held by institutional investors at the
end of the quarter when the firm files for Chapter 11, or the logarithm of one plus the number of institutional
investors; Delisted is a dummy variable of the delisting time (1 if the stock is delisted within the Chapter 11
process, and 0 otherwise); Prepack is a dummy variable of whether the filing is pre-negotiated; DIP is a dummy
variable of whether the firm receives debtor-in-possession financing; and Equity Committee is a dummy variable of
whether an equity committee is formed. Other control variables are defined as in Table 2.
Model 1 Model 2 Model 3 Model 4
Intercept 0.146nnn 0.172nnn 0.233nn 0.201n

(12.55) (8.78) (2.18) (1.88)
Intraday Volatility 0.091nnn 0.094nnn 0.090nnn 0.092nnn
(6.89) (7.03) (6.65) (6.76)
Percentage of Institutional Ownership 0.279nnn 0.282nnn
(3.07) (2.86)
Breadth of Institutional Ownership 0.025nnn 0.008
(2.73) (0.61)
Log(Total Assets) 0.006 0.007
(0.98) (0.91)
Leverage 0.010 0.008
(0.92) (0.70)
Tangible Ratio 0.046 0.045
(1.19) (1.17)
ROA 0.022 0.021
(0.75) (0.67)
Delisted 0.017 0.010
(0.93) (0.54)
Prepack 0.051nn 0.049nn
(2.54) (2.40)
DIP 0.004 0.006
(0.22) (0.31)
Equity Committee 0.035n 0.034n
(1.86) (1.79)
Industry and Year Dummies Included Included
Adj-R2 11.7% 11.4% 18.0% 16.4%
F-Value 31.06 29.96 4.56 4.19
N 453 453 441 441
n nn nnn
po0:10, po0:05, po0:01.
7. Conclusions
Our paper provides a thorough understanding regarding the trading, value, and performance
of the stocks in Chapter 11 using a unique dataset from Pink OTC Markets Inc. First, we show
that there is far more active trading of these stocks through the Chapter 11 process than is
generally perceived. Institutional investors unload the stocks when firms approach bankruptcy
filings, and around 90% of the investors are individual investors post-filing. Second, we show
that Chapter 11 stocks have positive fundamental values because they are call options on firm
assets. The option value calculated using the Black-Scholes model can explain a substantial
portion of the cross-sectional variation in the observed market equity values right after filing.
Even after controlling for the expectation of APR violations, the equity value after filing is
positively related to asset value, asset volatility, risk-free rate, and expected duration, and it is
negatively related to liabilities. Moreover, the return correlation between Chapter 11 stocks and
an industry- and size-matching sample exhibits strong non-linearity as predicted by the option
theory. Third, we document large negative returns of Chapter 11 stocks, both before and after
risk adjustments. Recognizing the increased information uncertainty and tightened short-sale
constraints accompanying the bankruptcy filing, as well as the unique feature of Chapter 11
stocks that the true value is revealed upon bankruptcy resolution, we provide an explanation for
the negative abnormal returns in the context of heterogeneous beliefs with short-sale constraints.
Consistent with our hypothesis, we show that stocks with more information uncertainty and
lower institutional ownership, and thus more binding short-sale constraints, tend to have more
negative returns over the duration of the Chapter 11 process.
References
Aharony, J., Jones, C.P., Swary, I., 1980. An analysis of risk and return characteristics of corporate bankruptcy
using capital market data. Journal of Finance 35, 1001–1016.
Barber, B.M., Odean, T., 2000. Trading is hazardous to your wealth: the common stock investment performance
of individual investors. Journal of Finance 55, 773–806.
Barberis, N., Huang, M., 2008. Stocks as lotteries: the implications of probability weighting for security prices.
American Economic Review 98, 2066–2100.
Barberis, N., Thaler, R., 2003. A survey of behavioral finance. In: Harris, M., Constantinides, G., Stulz, R. (Eds.),
Handbook of the Economics of Finance. North Holland, Amsterdam.
Bebchuk, L.A., 1988. A new approach to corporate reorganizations. Harvard Law Review 101, 775–804.
Bebchuk, L.A., 2000. Using options to divide value in corporate bankruptcy. European Economic Review 44,
829–843.
Bebchuk, L.A., Chang, H.F., 1992. Bargaining and the division of value in corporate reorganization. Journal of
Law, Economics, and Organization 8, 253–279.
Betker, B.L., 1995a. Management’s incentives, equity’s bargaining power, and deviations from absolute priority in
chapter 11 bankruptcies. Journal of Business 68, 161–183.
Betker, B.L., 1995b. An empirical examination of prepackaged bankruptcy. Financial Management 24, 3–18.
Bharath, S.T., Werner, I.M., Panchapegesan, V., 2010. The changing nature of chapter 11. Working paper. Ohio
State University.
Bharath, S.T., Shumway, T., 2008. Forecasting default with the Merton distance to default model. Review of
Financial Studies 21, 1339–1369.
Boehme, R.D., Danielsen, B.R., Sorescu, S.M., 2006. Short-sale constraints, differences of opinion, and
overvaluation. Journal of Financial and Quantitative Analysis 41, 455–487.
Chen, J., Hong, H., Stein, J.C., 2002. Breadth of ownership and stock returns. Journal of Financial Economics 66,
171–205.
Clark, T.A., Weinstein, M.I., 1983. The behavior of the common stock of bankrupt firms. Journal of Finance 38,
489–504.
Dahiya, S., John, K., Puri, M., Ramirez, G., 2003. Debtor-in-possession financing and bankruptcy resolution:
empirical evidence. Journal of Financial Economics 69, 259–280.
D’Avolio, G., 2002. The market for borrowing stocks. Journal of Financial Economics 66, 271–306.
Dawkins, M.C., Bhattacharya, N., Bamber, L.S., 2007. Systematic share price fluctuations after bankruptcy
filings and the investors who drive them. Journal of Financial and Quantitative Analysis 42, 399–420.
Denis, D.K., Rodgers, K.J., 2007. Chapter 11: duration, outcome, and post-reorganization performance. Journal
of Financial and Quantitative Analysis 42, 101–118.
Diether, K.B., Malloy, C.J., Scherbina, A., 2002. Differences of opinion and the cross section of stock returns.
Journal of Finance 57, 2113–2141.
Duffie, D., Garleanu, N., Pedersen, L.H., 2002. Securities lending, shorting, and pricing. Journal of Financial
Economics 66, 307–339.
Eberhart, A.C., Altman, E.I., Aggarwal, R., 1999. The equity performance of firms emerging from bankruptcy.
Journal of Finance 54, 1855–1868.
Eberhart, A.C., Moore, W.T., Roenfeldt, R.L., 1990. Security pricing and deviations from the absolute priority
rule in bankruptcy proceedings. Journal of Finance 45, 1457–1469.
Figlewski, S., 1981. The informational effects of restrictions on short sales: some empirical evidence. Journal of
Financial and Quantitative Analysis 16, 463–476.
Figlewski, S., Webb, G.P., 1993. Options, short sales, and market completeness. Journal of Finance 48, 761–777.
Franks, J.R., Torous, W.N., 1989. An empirical investigation of U.S. firms in reorganization. Journal of Finance
44, 747–769.
Franks, J.R., Torous, W.N., 1994. A comparison of financial recontracting in distressed exchanges and chapter 11
reorganizations. Journal of Financial Economics 35, 349–370.
Gilson, S.C., 1990. Bankruptcy, boards, banks, and blockholders: evidence on changes in corporate ownership
and control when firms default. Journal of Financial Economics 27, 355–387.
Glosten, L.R., Milgrom, P.R., 1985. Bid, ask and transaction prices in a specialist market with heterogeneously
informed traders. Journal of Financial Economics 14, 71–100.
Harris, J.H., Panchapagesan, V., Werner, I.M., 2008. Off but not gone: a study of Nasdaq delistings. Working
paper. Ohio State University.
Hong, H., Scheinkman, J., Xiong, W., 2006. Asset float and speculative bubbles. Journal of Finance 61,
1073–1117.
Hotchkiss, E.S., 1995. Postbankruptcy performance and management turnover. Journal of Finance 50, 3–21.
Jiang, W., Li, K., Wang, W., 2012. Hedge funds and chapter 11. Journal of Finance 67, 513–560.
Merton, R.C., 1974. On the pricing of corporate debt: the risk structure of interest rates. Journal of Finance 29,
449–470.
Miller, E.M., 1977. Risk, uncertainty and divergence of opinion. Journal of Finance 32, 1151–1168.
Morse, D., Shaw, W., 1988. Investing in bankrupt firms. Journal of Finance 43, 1193–1206.
Nagel, S., 2005. Short sales, institutional investors, and the cross-section of stock returns. Journal of Financial
Scheinkman, J., Xiong, W., 2004. Heterogeneous beliefs, speculation and trading in financial markets. In: Paris-
Princeton Lectures on Mathematical Finance 2003, Lecture Notes in Mathematics, Springer-Verlag, Berlin.
Tashjian, E., Lease, R.C., McConnell, J.J., 1996. Prepacks: an empirical analysis of prepackaged bankruptcies.
Journal of Financial Economics 40, 135–162.
Weiss, L.A., 1990. Bankruptcy resolution—direct costs and violation of priority of claims. Journal of Financial
Zhang, X.F., 2006. Information uncertainty and stock returns. Journal of Finance 61, 105–137.

Investing in bankrupt stocks: Trading, value, performance and losses

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Investing in bankrupt stocks: Trading, value, performance and losses

Uploaded by

Copyright:

Available Formats

Available online at www.sciencedirect.

Journal of Financial Markets 16 (2013) 33–60

Investing in Chapter 11 stocks: Trading,

JEL classification: G33; G12; G14

Keywords: Chapter 11; Option theory; Heterogeneous beliefs; Limits to arbitrage

Outcome Number of Cases Percent Duration

Min Median Max Mean Std. Dev.

Emerged or Reorganized 260 43.2 29 266 1584 355 285

Panel A: Summary statistics

N Min Q1 Median Q3 Max Mean Std. Dev.

Panel B: Correlation matrix

Total Assets 1 0.60 0.06 0.01 0.26 0.50

4. Trading of Chapter 11 stocks

Percentage of Firms Traded

Percentage of Institutional Ownership

5. The valuation of Chapter 11 stocks

CðV ,X ,r,T,s2 Þ ¼ VNðd1 ÞXerT Nðd2 Þ,

EV ¼ a þ b1 Assets þ b2 Volatility þ b3 Short-Term Debt þ b4 Long-Term Debt

EV ¼ a þ b1 Assets þ b2 Volatility þ b3 Short-Term Debt þ b4 Long-Term Debt þ b5 Risk-Free þ b6 Duration

Model 1 Model 2 Model 3 Model 4 Model 5

Intercept 7.075nnn 1.839n 2.039nn 2.677nnn 2.898nnn

5.2. The payoff structure of Chapter 11 stocks

Holding Period Return

Matching Sample Return

Holding Period Return

Matching Sample Return Matching Sample Return

HPRi ¼ ð1yi Þnðglow þ blow nHPRmatch,i Þ þ yi nðghigh þ bhigh nHPRmatch,i Þ,

6. The performance of Chapter 11 stocks

We investigate the performance of Chapter 11 stocks in the holding period as deﬁned in

HPR MHPR MHPRMarketAdj MHPRMatchAdj

Min 1.00 0.96 0.98 1.10

MHPRMatchAdj ¼ ð1 þ HPRÞ21=t ð1 þ HPRMatch Þ21=t : ð8Þ

N Mean Std. Dev. Min Median Max

(0, 0.1) 170 0.28 2.31 1.00 0.71 27.47

N Mean Std. Dev. Min Median Max

(0, 0.1) 170 0.16 0.21 1.10 0.13 0.38

6.2. The cross-section of Chapter 11 stock returns

Average Quoted Spread

Trading Days around Chapter 11 Filing

Trading Days around Chapter 11 Filing

MHPRMatchAdj ¼ a þ b1 Quoted Spread þ b2 Intraday Volatility þ b3 Turnover þ b4 Institutional Ownership

Model 1 Model 2 Model 3 Model 4

Intercept 0.090nnn 0.107nnn 0.132 0.096

MHPRMatchAdj ¼ a þ b1 Intraday Volatility þ b2 Institutional Ownership þ Control Variables þ e,

Model 1 Model 2 Model 3 Model 4

Intercept 0.146nnn 0.172nnn 0.233nn 0.201n

You might also like