Employment Patterns in Relation to Bankruptcy.doc

Document Sample
Employment Patterns in Relation to Bankruptcy.doc Powered By Docstoc
					      EMPLOYMENT PATTERNS IN RELATION TO BANKRUPTCY1
                   CHRYSTIN ONDERSMA

I.      INTRODUCTION

        A debate about the merits and shortcomings of Chapter 11 has raged
since the Bankruptcy Reform Act established the modern version of business
reorganization in 1978. The debate has been consistent in its fervor and
inconsistent in its factual rigor.2 With prodding from empirical scholars,3
Chapter 11 supporters and detractors alike have begun to recognize that
empirical data is foundational to the reorganization debate. Over time,
scholars have begun to develop a picture of what actually happens to large,
publicly traded companies that file for Chapter 11. This picture includes two
consistent features—companies emerge from reorganization with around 75%
of their pre-filing assets, and nearly all companies experience management
turnover.

        This study contributes to the understanding of the behavior or large,
publicly traded companies in Chapter 11 by examining their employment
patterns around the time of bankruptcy filing. Employment patterns are an
important part of the reorganization picture. Not only did Congress view job
protection as a key feature of Chapter 11,4 but jobs are also important
company assets whose fate in a corporate reorganization is worth tracking.
This study fills in the picture of patterns of publicly traded companies in
reorganization. In addition to asset reduction and management turnover,
another nearly universal feature of publicly traded companies that file for
Chapter 11 is prolonged—if not permanent—workforce reduction.
Companies decrease their workforces by nearly 50 percent around the time of
a Bankruptcy Filing, and do not increase their workforces at any point through
ten years after filing. Do these numbers indicate successful or unsuccessful
bankruptcy reorganizations? The answer to that question depends upon the


1
  I thank Professor Lynn LoPucki and Professor Elizabeth Warren for their extensive
feedback, comments, and guidance throughout the research and writing process,
Daniel DeVries IV for his help comprehending and navigating excel formulas and
statistics, and Lynda Zhang for guiding me through STATA.
2
  Teresa A. Sullivan, Elizabeth Warren, Jay Lawrence Westbrook, The Use of
Empirical Data in Formulating Bankruptcy Policy, 50 Law & Contemp. Prob. 195,
199.
3
  Id.
4
  See infra notes 21-33 and accompanying text.




                                         1
answer to more fundamental questions—such as whom reorganization should
serve and what purpose it should ultimately fulfill.

II.      BACKGROUND

        Chapter 11 has both fervent supporters and passionate critics.
Critics have argued that Chapter 11 is both inefficient5 and
unnecessary.6 Supporters have responded that Chapter 11 is more
efficient and more feasible than any proposed alternative,7 and that it




5
  Barry E. Adler, Bankruptcy and Risk Allocation, 77 Cornell L. Rev 439, 447-48,
465-468 (1992) (arguing that the opportunity costs, uncertainty costs, disruption
costs of Chapter 11 are not offset by any corresponding benefit since Chapter 11, he
claims, merely reallocates benefits from senior to junior claimants.); Michael
Bradley & Michael Rosenzweig, The Untenable Case for Chapter 11, 101 Yale L.J.
1043, 1067-1072 (1992) (arguing that Chapter 11 is inefficient because stockholders
and bondholders are worse off after bankruptcy filing according to their study of a
sample of public companies.); Douglas G. Baird, The Uneasy Case for Corporate
Reorganizations, 15 J. Legal Stud. 127, 139-145 (1986) (arguing that outright sale is
preferable to owners and less costly than reorganization).
6
  Douglas G. Baird & Robert K. Rasmussen, The End of Bankruptcy, 55 Stan. L.
Rev. 751, 753-54 (2002) (arguing that Chapter 11 is unnecessary since, they claim,
firms in financial distress are unlikely to have going-concern surplus, assets need not
remain with a particular firm, and investors contract for control rights that anticipate
corporate distress and facilitate sale).
7
  See, e.g., Lynn M. LoPucki, The Nature of the Bankrupt Firm: A Response to
Baird and Rasmussen’s The End of Bankruptcy, 56 Stan L. Rev. 645, 661-664
(2003) [hereinafter Response] (arguing that contracts are not appropriate control-
shifting mechanisms because of the impossibility of identifying a single residual
owner); id. at 666-668 (arguing that at the founding of the firm shareholders and
creditors contracted for the more efficient mechanism of handing power to a board of
directors who would make sale decisions in bankruptcy); Elizabeth Warren and Jay
Lawrence Westbrook, Contracting Out of Bankruptcy: An Empirical Intervention,
118 Harvard L. Rev. 1197, 1214-1254 (2005) (demonstrating that attempts to
contract around bankruptcy would be inefficient due to the large number of
maladjusting creditors, such as involuntary and quasi-involuntary creditors,
unsophisticated creditors, and creditors with small claims); id. at 1218 (arguing that
the transaction costs imposed by a contractualist system make such a system less
efficient than the standardized reorganization system under the Code); Elizabeth
Warren, The Untenable Case for Repeal of Chapter 11, 102 Yale L.J. 437, 475-76
(1992 [hereinafter Untenable Case]) (arguing that information asymmetry and
transaction costs make Chapter 11 a faster, lower cost alternative to state court
resolution of contract and other legal disputes).




                                           2
provides important protection for many different constituencies that
are affected by a corporation’s collapse.8

        Despite the vigor of the debate, surprisingly little is known
about what actually happens to companies that reorganize under
Chapter 11. As Professor Lynn LoPucki and Professor William
Whitford suggested, “any critique of Chapter 11 should begin with an
understanding of what is actually occurring in the cases.”9 Persuasive
theory and sound policy will always be well-grounded in fact.10
Professors Sullivan, Warren and Westbrook pointed out that
bankruptcy policy is too often “based on assumptions of fact about the
behavior or debtors and creditors . . . and the rest of the cast.”11 If
those factual assumptions are misguided, then the policies themselves
will also be misguided. An optimal Chapter 11 policy must be based
on a solid understanding of what happens in Chapter 11 cases.

        When it comes to the empirical study of companies in financial
distress, one size does not fit all. Chapter 11 governs a myriad of
businesses—from garage start-ups to behemoths such as Enron and
Worldcom. Scholars have pointed out that there are crucial
differences between companies at the two extremes of the size
spectrum, suggesting that their bankruptcy experiences are likely to be
quite different.12



8
  Elizabeth Warren, Bankruptcy Policymaking in an Imperfect World, 92 Mich. L.
Rev. 336, 354-357 (1993) (pointing out that bankruptcy is designed to benefit a
diversity of creditors, including involuntary creditors, as well as employees,
customers, and communities); Untenable Case, supra note 4, at 468 (explaining that
the Code considers the interests of “employees, taxing authorities, U.S. fishermen,
farmers, landlords, business partners, parties to executory contracts, beneficiaries of
statutory liens, ordinary course creditors, and creditors with setoff-rights-to name
just a few,” and that Congress made “repeated references to protecting jobs and
saving troubled businesses.”).
9
  Lynn M. LoPucki and William C. Whitford, Patterns in the Bankruptcy
Reorganization of Large, Publicly Held Companies, 78 Cornell L. Rev. 597, 597
(1993) [hereinafter Patterns].
10
   See id. at 201.
11
   Teresa A. Sullivan, Elizabeth Warren, Jay Lawrence Westbrook, The Use of
Empirical Data in Formulating Bankruptcy Policy, 50 Law & Contemp. Prob. 195,
199.
12
   See, e.g., Elizabeth Warren, Untenable Case, supra note 4, at 441-443 (1992).




                                           3
         Although large, publicly traded companies make up only a
small percentage of all companies that file for Chapter 11, these
reorganizations face unique and complex legal issues. The
reverberations of even a single such failure can be felt in international
stock markets as much as in local economies, impacting the lives of
literally millions of employees, creditors, and shareholders. Much of
the scholarly evaluation of the current Chapter 11 system is tightly
focused on large, publicly traded companies that reorganize. While an
empirical exploration of this subset of companies cannot shed light on
the merits of Chapter 11 as it affects closely held companies or
individual entrepreneurs, the financial reorganization of large, publicly
traded companies is important, and a better empirical understanding of
those mega-bankruptcy cases is critical to any meaningful evaluation
of the bankruptcy reorganization system.

        As scholars have worked to provide a reliable picture of what
happens in the large Chapter 11 cases,13 two patterns have emerged
that document important changes when these companies attempt to
reorganize: Companies experience a reduction in asset size and high
management turnover.

        With regard to asset size, Professor LoPucki found that
publicly traded firms filing for Chapter 11 between January 1, 1982
and March 15, 1988 emerged on average at 44% of their filing size.14
A decade later, the reduction was smaller, but nonetheless
pronounced.15 Professor Joseph Doherty and Professor LoPucki also
found that the 98 publicly traded firms that filed for Chapter 11
between 1991 and 1996 emerged on average at 77% of their prefiling
size.16 In his 2001 study, Professor LoPucki found that public
companies that file for Chapter 11 emerged at around 75% of their



13
   See, e.g., Patterns, supra note 6 at 597 (studying patterns in the bankruptcy
reorganization of the forty-three Chapter 11 cases that filed before March 15, 1988).
Others have taken a more expansive approach, examining both private and public
corporations. See Empirical Intervention, supra note 1, at 1208-1210 (studying 40%
of business cases in twenty-three districts, where debtors are a “mix of human
beings, partnerships, corporations, and other forms of legal entities.”).
14
   Patterns, supra note 6, at 605-606.
15
   Id.
16
   Response, supra note 4, at 657.




                                          4
prefiling size. Almost all public firms that file for Chapter 11
experience some significant asset reduction.17

        Management turnover is another nearly universal characteristic
of public firms that file for Chapter 11. Professor LoPucki and
Professor Whitford found that, during the two-year period between
eighteen months prior to bankruptcy filing and six months after
bankruptcy plan confirmation, 90% of public companies that filed for
Chapter 11 before 1988 changed CEO’s at least once.18 In a study of
202 public companies, Professor Brian Betker found that only 8% of
the top managers who held office two years before filing retained their
positions one year after plan confirmation.19 Professor Stuart Gilson
studied 409 public companies that filed for bankruptcy between 1979
and 1984, and found that 71% of managers lost their jobs within two
years of bankruptcy filing.20

        This study contributes to the picture of companies in
bankruptcy by studying patterns of employment in large, publicly
traded companies that filed for Chapter 11. Employment patterns in
these large Chapter 11 cases are important because 1) employee
relationships are valuable company assets, 2) employees comprise an
important part of the corporate team on whose behalf management
acts, and 3) the legislative history of the 1978 Bankruptcy Reform Act
indicates that employee protection is one of the key purposes of
Chapter 11. Including data on employment patterns enhances the
picture of Chapter 11 gleaned from the study of corporate assets and
management turnover, providing a more complete picture of the
overall process.

       Employee relationships are a key component of the going-
concern value of a business. Indeed, some scholars claim that going
concern value, which is the value preserved in a reorganization rather

17
   Patterns, supra, note 6, at 605.
18
   Lynn M. LoPucki & William C. Whitford, Corporate Governance in the
Bankruptcy Reorganization of Large, Publicly Held Companies, 141 U. Pa. L. Rev.
669, 723 (1993).
19
   Brian L. Betker, Management Changes, Equity’s Bargaining Power and
Deviations from Absolute Priority in Chapter 11 Bankruptcies, The Journal of
Business, Vol. 68, No. 2 (Apr. 1995), 161-183, at 161.
20
   Stuart C. Gilson, Management Turnover and Financial Distress, 25 J. Fin. Econ.
241, 247 (1989).




                                        5
than liquidation, “resides principally in relationships.”21 Employees
have valuable relationships with other people, such as creditors,
customers, and suppliers, from which the business profits.22 They also
have relationships with assets, such as knowledge about computer
software or the products they manufacture.23 These relationships take
time to build, and it is costly to rebuild them when they are broken
apart.24 Further, the “know-how and expertise” of the workforce is an
essential company asset.25 Any company that reduces its workforce
significantly may be cutting costs, but it is also forfeiting substantial
corporate assets at the same time. While the company may decide that
cutting provides the right balance, such a move involves the
disposition of an asset worth noting in the reorganization of a business.

        Second, employees are part of the corporate team to whom
management is accountable.26 Under this conception of corporations,
employees, along with stockholders, creditors, local governments,
comprise “a team of people who enter into a complex agreement to
work together for their mutual gain.”27 Professor LoPucki applied this
idea to the bankruptcy context, arguing that members of the team
intentionally incorporated bankruptcy law at the time they
contracted—if team members do not take steps to contract around
bankruptcy, they are affirmatively selecting to be bound by the default
rule of bankruptcy reorganization.28 The entire team agrees to permit
the company to remain in business at the time of bankruptcy as long as
it “benefits the members of the team in the aggregate.”29 Because
employees are members of the team, any evaluation of Chapter 11
must include an analysis of what happens to them in bankruptcy.

       Finally, Congress intended for employees to be beneficiaries of
the Chapter 11 process. Legislative history to the 1978 Bankruptcy

21
   Response, supra note 4, at 652.
22
   Id.
23
   Id.
24
   Id.
25
   Id.
26
   Margaret M. Blair & Lynn A. Stout, A Team Production Theory of Corporate
Law, 85 Va. L. Rev. 247, 271-77 (1999).
27
   Id. at 278.
28
   Lynn M LoPucki, A Team Production Theory of Bankruptcy Reorganization, 57
Vand. L. Rev. 741, 756 (2004).
29
   Id. at 764.




                                      6
Reform Act suggests that job preservation is one of the key goals of
Chapter 11. A key House judicial committee report states, “The
purpose of a business reorganization case, unlike a liquidation case, is
to restructure a business’ finances so that it may continue to operate,
provide its employees with jobs, pay its creditors, and produce a return
for its stockholders…It is more economically efficient to reorganize
than to liquidate, because it preserves jobs and assets.”30 Professor
Elizabeth Warren picks up the same theme, pointing out that
“Congress singled out a number of beneficiaries of its distributional
decisions, making repeated references to protecting jobs and saving
troubled businesses.”31 The Supreme Court also acknowledged that
“[t]he fundamental purpose of reorganization is to prevent a debtor
from going into liquidation, with an attendant loss of jobs and possible
misuse of economic resources.”32 Indeed, Chapter 11 is designed to
take employees into account: “By giving businesses the opportunity to
reorganize, the Bankruptcy Code may permit employees to keep their
jobs. This provision creates an alternative method to reallocate the
risks of business failure away from the employees toward other
lenders who can better diversify their investment portfolios.”33

        The notion that employees are a critical part of the
reorganization process is not universally accepted. Professor Charles
Mooney admonishes the courts to disregard the interests of employees
in business reorganization. He asserts that to exercise such concerns at
the expense of shareholders is “prima facie theft.”34 Nor is Professor
Mooney alone in this vision. Professor LoPucki notes, “The currently
prevailing contractarian theories of the firm and of bankruptcy
recommend cutting labor costs first. Bankruptcy, these theorists
postulate, exists solely for the benefit of the creditors and shareholders
of the firm.”35 Even for those who think employees should be

30
   H.R. Rep. No. 95-595, at 220 (1977).
31
    Untenable Case, supra note 4, at 468 (citing 124 Cong. Rec. 32,392 (1978)
(statement of Rep. Edwards); 124 Cong. Rec. 33,990 (1978) (statement of Sen.
DeConcini)).
32
   NLRB v. Bildisco & Bildisco, 465 U.S. 513, 528 (1984).
33
   Elizabeth Warren, Bankruptcy Policymaking in an Imperfect World, 92 Mich. L.
Rev. 336, 357 (1993).
34
   See, e.g., Charles W. Mooney, Jr. A Normative Theory of Bankruptcy Law:
Bankruptcy Law As (Is) Civil Procedure, 61 Wash. & Lee L. Rev. 931, 963-964.
35
   Lynn M. LoPucki, A Team Production Theory of Bankruptcy Reorganization, 57
Vand. L. Rev. 741, 742 (2004) [hereinafter Team Production Theory].




                                       7
disregarded, however, the data about what happens to those employees
in Chapter 11 will be useful. Researches of every normative bent can
use these data to argue whether too little or too much protection has
been offered to employees when their companies reorganize.

        Understanding patterns of employment in reorganizing
companies is essential to understanding—and eventually evaluating—
the Chapter 11 system. By reporting on patterns of employment in
large, publicly traded companies that filed for Chapter 11, this
research helps fill in the picture of what actually happens to companies
that reorganize under Chapter 11.

III.    METHODOLOGY

        This project examined the employment patterns of 590 large, publicly
traded companies that filed for reorganization after the 1978 Bankruptcy
Reform Act became effective. Professor LoPucki’s Bankruptcy Research
Database (“BRD”) contains data for all 723 publicly traded companies that
filed for Chapter 11 since 1978. Because the database includes every Chapter
11 case that meets the size requirements, no sampling was involved in this
project.

        In addition to drawing upon the BRD, this project also relied on data
supplied by Compustat. Information about the number of employees in
publicly traded companies is available on Compustat.36 Compustat provides
the number of employees employed at a given company in a given year
(“Number of Company Employees.”). For this study, Numbers of Company
Employees were available for at least some years in 590 of the 723 companies
in the BRD. Whenever possible, the data were collected for a twenty-one
year period, spanning the time from ten years prior to bankruptcy filing to ten
years after bankruptcy filing. Despite its preeminence as the leading
corporate database in the country, the Compustat data were surprisingly
sparse. Only twenty-two companies had Compustat-reported Numbers of
Company Employees each year for the twenty-one year span from ten years
prior to filing to ten years following filing. When the time period is
shortened, Numbers of Company Employees are available for more
companies. Sixty-two companies had Compustat-reported Numbers of

36
  Compustat is a service that obtains and organizes data from companies’ filings
with the Securities and Exchange Commission.




                                         8
Company Employees for each year from five years prior to filing through five
years after filing.

        There are several possible reasons for the sparseness of the data. First,
the design of the study necessarily limits the available data. Companies that
did not file prior to 1995 would not have data for ten years following filing,
and companies that did not file prior to 2000 would not have data for five
years following filing. Aside from this obvious limitation, missing numbers
could be the result of Compustat’s failure to record the company’s data during
certain years or some circumstance within the company. For example,
Compustat data are taken from 10k reports filed with the SEC, so any
company that failed to file an annual 10k report with the SEC would not have
complete data. Moreover, a company that lost its publicly traded status or
merged with another corporate entity would no longer be represented in
Compustat.

        Because so few companies had Numbers of Company Employees for
each year, two analyses were conducted. The first study (“Complete Data
Study”) examines the Numbers of Company Employees at the sixty-two
companies that had Compustat-reported Numbers of Company Employees for
each year from five prior to filing to five years following filing. 37 Because
this study contains only sixty-two cases and because the time period is limited
to only five years before and after filing, the usefulness of the Complete Data
Study may be somewhat limited. In order to expand the number of cases
studies, it was possible to construct a second dataset (“Partial Data Study”).
The Partial Data Study collected data for the twenty-one year period from ten
years prior to filing, to the year of filing, to ten years after filing, but includes
companies for which some years of employment were missing. The Partial
Data Study includes 590 cases.

         In order to qualify for inclusion in the Partial Data Study for a given
year, a company from the BRD list of bankrupt companies must have had a
listing under Number of Company Employees for both that year and the
following year. For example, the list of companies included in the Partial
Data Study for the second year after bankruptcy filing must all have a Number
of Company Employees for the second year after bankruptcy filing and the
third year after bankruptcy filing. That is, if company X only has data for the
year of bankruptcy filing and three years following filing, it would be
37
  Because the Ns were so small, I opted not to study the twenty-two companies with
complete data from ten years prior to filing to ten years following filing.




                                        9
included in the study for the year of filing, the year after filing, and two years
after filing—but would not be included for the third year after filing. This
ensured that a disproportionately large company’s disappearance from the
data set would not misleadingly suggest that many jobs had disappeared when
Compustat did not or could not record the data for that year.

        The first level of data analysis was confined to the sixty-two
companies in the Complete Data Study. I then repeated the same analysis
with the larger Partial Data Study, taking care to adjust the calculations to
account for the fact that the number and identity of companies included in the
Partial Data Study varied from year to year.

IV.      EMPLOYMENT PATTERNS IN RELATION TO BANKRUPTCY

         The Complete Data Study, represented in Figure 1, and the
Partial Data Study, represented in Figure 2, support two key
findings.38 First, companies experience a reduction of about 50
percent in Numbers of Company Employees in the years surrounding
bankruptcy filing. Second, the Numbers of Company Employees do
not increase at any point after bankruptcy filing, through ten years
after filing.


38
   In addition to the overall pattern represented here, some companies experienced
alternative patterns. I examined individually all 129 companies that had data
through the year of bankruptcy, and at least five years of data prior to filing or after
filing. Nineteen companies experienced an increase in Number of Company
Employees through the year prior to filing, and then decreased the year of filing and
did not increase again. Thirty-seven companies experienced an increase in Number
of Company Employees through 3 to 5 years prior to filing and then began to drop in
Number of Company Employees. Thirty-five companies experienced an increase in
Number of Company Employees between 2 and 5 years prior to filing, and then a
decrease in Number of Company Employees around the year of filing, and then
experienced some increase in Number of Company Employees at some point after
filing. Seventeen companies experienced no increase in Number of Company
Employees prior to bankruptcy filing, and experienced a slow drop Number of
Company Employees that continued after bankruptcy filing. The remaining
companies did not follow any of these patterns. I chose not to report on these
patterns because there was no discernible trend regarding which companies fell into
which pattern. I coded the companies according to the patterns described above, and
sorted them in Professor LoPucki’s Bankruptcy Research Database in an attempt to
unearth a trend. I sorted according to year, state of filing, whether the bankruptcy
was prepackaged, assets, and industry code. There were no discernible trends; the
companies fitting each pattern were randomly distributed throughout each category.




                                          10
       A.          Analysis of the Data

        The number of company employees at the sixty-two companies
comprises the Complete Data Study. For each company, it was
possible to calculate the sum of the number of company employees for
each year from five years prior to filing until five years following
filing. Figure 1 displays the change in the sum of Number of
Company Employees from five years prior to filing to five years after
filing.

                                        Figure 1: Complete Data Study
                             Total Number of Company Employees for 62 Companies
         900,000



         800,000



         700,000



         600,000



         500,000



         400,000



         300,000



         200,000



         100,000



               0
                   -5   -4        -3     -2     -1     bkcy    1      2      3    4   5




         Because the Partial Data Study does not include the same
company for every year, it was not possible to reproduce the
calculations of the Complete Data Study. The goal was to conduct a
calculation that would accurately reflect the same movement of
increase or decrease in employee numbers that is represented in the
Complete Data Study. The Partial Data Study calculated the
cumulative year-to-year change in employee numbers for companies
from ten years prior to filing to ten years after filing. The Partial Data
Study had four steps. The study began with a list of companies for
each year (from ten years prior to filing to ten years after filing) that
had a Number of Company Employees for both that year and the
following year. Second, I calculated the change in the Number of
Company Employees from one year to the next for each company on
that list (“Change in Employees”). Third, for each year, I divided the




                                               11
sum of the Change in Employees by the sum of the Number of
Company Employees, resulting in the average change in Number of
Company Employees (“Average Change in Employees”) for each
year, from ten years prior to filing to ten years after filing. Finally,
using ten years prior to filing as a starting point, the Average Change
in Employees for each year was added to the sum of the Average
Change in Number of Employees from the previous years. This
revealed the cumulative average change in the Number of Company
Employees (Cumulative Average Change in Employees)—from ten
years prior to filing to each subsequent year—through ten years after
filing.39 The Cumulative Average change in Number of Company
Employees relative to ten years prior to filing is shown in Figure 2.40




39
   Without taking the numbers cumulatively, the graph would represent a rate of
change rather than a change in number of company employees over time. While the
method is imperfect due to missing Compustat data, there is no other way to
represent a change over time that parallels the Complete Data Study without
distorting the data.
40
   Negative 100% was chosen as the base point of the y axis for the Partial Data
Study to avoid misleading the reader (by making the change in number of company
employees appear greater than it actually was) and to match the Complete Data
Study as closely as possible. Graph 1 has a y axis base point of zero (to reflect the
possibility that the total Number of Company Employees could have decreased to
zero), and negative 100% is the point that most closely reflects zero for the Partial
Data Study. Because the graph for the Partial Data Study is cumulative, it is
possible that a given company could decrease below 100% (an 80% drop one year
followed by a 50% drop the following year), but since the Cumulative Average
Change did not drop below 40%, a base point of negative 100% was sufficient to
roughly preserve the integrity of the scale.




                                         12
                                       Figure 2: Partial Data Study
                                      Cumulatve Average Change in




               20%




                0%




              -20%




              -40%




              -60%




              -80%




              -100%
                      -10   -8   -6        -4      -2     BKC Y       2   4   6   8   10




         B.           Jobs Decrease Around the Time of a Bankruptcy Filing

         Both the Complete Data Study and the Partial Data Study
support the conclusion that the Number of Company Employees in
companies that file for Chapter 11 reorganization decrease around the
time of bankruptcy filing. As Figure 1 shows, in the Complete Data
Study the companies collectively experienced a 48 percent decrease in
total Number of Company Employees from two years prior to filing to
two years after filing.41 The year-to-year decrease was statistically
significant for each of the two years prior to filing.42 The Partial Data
Study revealed a 52 percent decrease in the Cumulative Average
Change in Employees between two years prior to filing and four years
after filing, as appears in Figure 2. The decrease in Cumulative
Average Change in Employees was statistically significant for each of
the years from two years prior to filing to three years after filing.43

41
   The sum of the Number of Company Employees at the 62 companies for the year
two years after filing was 48 percent lower than the sum of the Number of Company
Employees at the 62 companies for the year two years prior to filing.
42
   Statistical significance in decline was tested according to a simple t-test (t-
test[variable]=0), which tests whether the decline is statistically significant from
zero. The increase in total company employees prior to filing was not statistically
significant. Appendix A shows the total Number of Company Employees for each
year and indicates the level of statistical significance for the decline in total Number
of Company Employees from the previous year.
43
   As with the Complete Data Study, statistical significance in decline was tested
according to a simple t-test (t-test[variable]=0), which tests whether the change from




                                                13
Because this method did not include the same companies for every
year tested, it was possible to test statistical significance in decline
only from one year to the next. This means that it is possible that the
overall decline after bankruptcy in the Partial Data Study was
statistically significant after the third year post-filing (for example,
from the third year after filing to the fifth year after filing), even
though the change from the third to the fourth year was not statistically
significant and the change from the fourth to the fifth year was not
statistically significant.

        While commentators might expect also to see a decline in
employee numbers in the ten years prior to Chapter 11 filing, the
Partial Data Study revealed a statistically significant increase in the
Average Change in Employees for several of the years prior to filing,
including the change from the tenth to the ninth year prior to filing.
This long run up prior to filing may reflect a general trend rather than
a phenomenon unique to companies filing for bankruptcy. Because
the ten-year time periods are different for different companies,
comparisons are difficult. Nonetheless, it is notable that employment
in the U.S. from 1970-2005, the time period included in the Partial
Data Study, employment grew by a modest 1.6% annually.

        While general trends may undercut the significance of the
growth in the number of employees pre-bankruptcy, it also serves to
accentuate the decline in the number of employees around the time of
a bankruptcy filing. If companies outside bankruptcy generally hire
more employees over time, the decrease in employee numbers for
firms filing for bankruptcy is actually greater relative to normal
employment patterns.

        The Complete Data Study also revealed an increase in
Numbers of Company Employees in the five years prior to filing, but
this increase was not statistically significant.

        The longer Partial Employment Study largely mirrors the
patterns revealed in the shorter Complete Data Study. In the Partial

the pervious year is statistically significant from zero. Appendix B shows the
cumulative Average Change in Employees for each year and the level of statistical
significance for the Average Change in Employees from the previous year.




                                        14
Data Study, the Cumulative Average Change in Employees continued
to decrease each year through five years after bankruptcy filing,
although the decrease was not statistically significant in any year after
the third year following filing. Companies in the Partial Data Study
experienced the decrease in Cumulative Average Change in
Employees between two years prior to filing and four years after
filing, whereas the sixty-two companies with complete data
collectively experienced all of the decrease in Number of Company
Employees between two years prior to filing and two years after filing,
as shown in Figure 1. The patterns of employment evident in Figures
1 and 2 bear important similarities—both studies reveal nearly a 50
percent decrease in employees in the years surrounding bankruptcy
filing.

       C.      Companies Do Not Regain Employees After Bankruptcy
               Filing

         Not only do companies that file for Chapter 11 reorganization
decrease in employee numbers around the time of bankruptcy filing,
but the data also indicate that these companies do not regain a
significant number of jobs at any point through ten years after filing.
Both the Complete Data Study, shown in Figure 1, and the Partial Data
Study, shown in Figure 2, support this conclusion. The drop in the
number of employees typically began two years prior to bankruptcy
filing. For the Complete Data Study, the total Number of Company
Employees for the fifth year after bankruptcy filing had declined
another 25 percent from the total Number of Company Employees for
the year of bankruptcy filing. Similarly, in the Partial Data Study, five
years after filing the Cumulative Average Change in Employees was
29 percent lower than the Cumulative Average Change in Employees
for the year of filing. The Partial Data Study displayed no statistically
significant increase in the number of jobs at any point after bankruptcy
filing—through ten years after bankruptcy filing. By the tenth year
after filing the Cumulative Average Change in Employees remained
26.6 percent below the Cumulative Average Change in Employees for
the year of bankruptcy filing. Both studies demonstrate that the
Number of Company Employees does not increase during the period
after bankruptcy filing.

       D.      Summary




                                   15
       The Complete Data Study and the Partial Data Study both
support two conclusions: that companies experience around a 50
percent reduction in Numbers of Company Employees in the years
surrounding bankruptcy filing, and that Numbers of Company
Employees five and ten years after bankruptcy remain 25 percent
lower than the Numbers of Company Employees for the year of filing.


V.      DISCUSSION

        One of the liveliest debates about Chapter 11 is the degree of
“success” in Chapter 11 cases.44 Scholars disagree sharply about the
metric: What constitutes a successful reorganization? Bradley and
Rosenzweig, for example, would deem reorganization successful when
it increases the wealth of shareholders and bondholders,45 a position
likely shared by Professor Mooney.46 Professor Alan Schwartz might
deem reorganization successful when it maximizes the value of the
debtor’s estate for the benefit of creditors.47 LoPucki and Warren
might deem a reorganization successful if it makes corporate
constituencies, including creditors, owners, customers, and employees,
collectively better off than they would be in liquidation.48 The debate
over when reorganization is successful is tightly interwoven with the
debate over whom reorganization should serve. A key point of
contention is whether Chapter 11 can legitimately serve any interest
other than the interests of creditors and owners.49

        A.       Did the Companies Get it Right?

       Whether the preservation of jobs is a legitimate goal of Chapter
11 or whether it is “prima facie theft,” what happens to jobs in Chapter
11 should play a role in the debate. Do companies that decrease

44
   See, e.g., Michael Bradley & Michael Rosenzweig, supra note 2, at 1067-1072;
Untenable Case, supra note 4, at 475-76; Patterns, supra note 6, at 598.
45
   Michael Bradley & Michael Rosenzweig, supra note 2, at 1067-1072.
46
   See note 31 and accompanying text.
47
   Schwartz, supra note 2, at 1849-50.
48
   Team Production Theory, supra note 31, at 765-67; Untenable Case, supra note 4,
at 469-474.
49
   See, e.g., Charles W. Mooney,supra note 30, at 963-964; Team Production
Theory, supra note 6, at 742; Untenable Case, supra note 4, at 469-474.




                                       16
around 50 percent of their workforce—and keep their workforce at this
reduced level—qualify as successful reorganizations? Is the reduction
insufficient or too dramatic—or did the companies get it right?

           Story #1: Companies Insufficiently Reduce their Workforces

         Some aspects of the data may suggest that employees receive
too much protection in Chapter 11 filings. Employees, overall, are
much better protected than management. Between 71 percent and 92
percent of managers lose their jobs in the years immediately following
reorganization.50 Half of employees still have their jobs two years
after filing—and those jobs remain viable even ten years after filing.
For those who suggest that management is the only component of
employment that constitutes a valuable asset,51 this result may seem
skewed in the wrong direction.

       While it is clear that companies in bankruptcy cut their
workforces, these data do not reveal whether companies cut their
workforces the optimal amount. Companies that cut their workforces
more than average may be more successful than those that try to hang
on to more employees, or the reverse may be true. The data provided
here open a new line of inquiry. Subsequent studies might explore
whether companies that deviate from the pattern described here—
whether by retaining more employees or firing more employees—are
more valuable five or ten years after bankruptcy. Comparing post-
bankruptcy common stock returns to rates of employment would
provide a direct comparison.

        Establishing association is not, of course, the same as
establishing causation. It may be that stronger post-bankruptcy
companies can afford to keep more employees, or that companies that
shrink intensify their return on investment. But the employee data
permit a first look at the next level of empirical analysis, an analysis
that accounts for both change in assets and change in workforce and
permits an exploration of the dynamic interplay between the two.

           Story #2: Companies Reduce their Workforces Excessively

50
     See supra notes 15-17 and accompanying text.
51
     Baird and Rasmussen, supra note 3, at 773-76.




                                          17
         Other aspects of the data support a contradictory story. For
example, the data suggest that while some resources may be directed
toward job preservation, most reorganization efforts are directed
elsewhere. Companies are now emerging with around 75 percent of
their assets, up from 40 percent. Job reduction, on the other hand, has
remained around 50 percent from the time the 1978 Bankruptcy
Reform Act was passed to the present.52 Indeed, these two trendlines
suggest that, in a rough sense, the ratio of employees to assets in post-
bankruptcy businesses has declined over time. Imagine a company
with 100 employees whose assets are valued at 100 million. In the late
eighties, when that company’s Number of Company Employees
decreased to 50, the assets decreased to 40 million. Now, in a
company where the assets only drop to 75 million, Numbers of
Company Employees still decrease by 50 percent. This is hardly the
ringing endorsement of Chapter 11 that Congress seemed to hope for
when it declared a “purpose” of Chapter 11 to be permitting
businesses to “continue to operate” and “provide its employees with
jobs . . . .”53

        The employment data may signal other problems in the
Chapter 11 system. The fact that employee numbers remain at the
same diminished level from bankruptcy filing through ten years after
filing may indicate companies that reorganize continue to struggle
long after filing, and may never recover. The dream of Chapter 11
was that companies would “rise phoenix-like from their failure.”54
Reduced employment levels may suggest effective cost-cutting, but
the data do not suggest post-bankruptcy flight.

         Additional empirical work could support or contradict this
story. If companies that reduce their workforces by 50 percent or
more are healthy according to other indicia of success at the time of
the reduction, this may indicate unnecessary cutting. A better test
would be a long-term analysis, comparing an indicia such as common
stock returns with employee numbers. If companies that reduce their
workforces least have higher common stock returns ten years after
filings than companies that substantially reduce their workforces, this
could indicate that the average reduction is excessive.

52
   No statistically significant change from 1978 to 2005.
53
   H.R. Rep. No. 95-595, at 220 (1977).
54
   Use of Empirical Data, at 201.




                                         18
       Story #3: Companies Appropriately Reduce their Workforces

        Some of the evidence from the Complete Data Study and
Partial Data Study supports the conclusion that companies
appropriately reduce their workforces. A fifty percent reduction is
substantial, but it is not total, as it would be in liquidation. Further,
the fact that the companies remaining ten years after filing remain at
this employment level may suggest that these companies found the
appropriate equilibrium. It is true that the data do not suggest that
companies filing for Chapter 11 experience a miraculous, phoenix-like
recovery. On the other hand, even if they are not soaring, these
companies have survived—a fact that may mean very good news for
the current employees, as well as for suppliers, customers, and perhaps
even shareholders.

        Other data suggest that companies are not sloughing off
employees unnecessarily. There is a statistically significant
correlation between employee numbers and sales activity. That is,
most companies in the Complete Data Study that experienced a
substantial drop in Numbers of Company Employees experienced a
corresponding drop in sales activity. From five years prior to filing to
five years after filing, twenty-two of the thirty companies in which
employees decreased by 50 percent or more also experienced at least a
50 percent decrease in sales. Of the twenty-two companies that
decreased in employee numbers by over 75 percent, all but one
experienced at least a 50 percent decrease in sales activity.

       Additional data analysis could provide further support for this story
about employment patterns in companies that reorganize under Chapter 11.
Again, a comparison of employee levels with another indicia of success such
as common stock returns would help. If companies around a 50 percent
employee reduction level ten years after filing demonstrate a greater
improvement in common stock returns than companies that reduce their
workforce by a much greater or lesser amount, this could suggest that the
average company is doing the right thing with respect to hiring decisions.

       B.      Does Chapter 11 Help Save Jobs?




                                   19
         In addition to asking whether job reduction corresponds to
other objective indicia of success, there is a more basic question: is
this degree of permanent job reduction evidence that Chapter 11 helps
save jobs? The companies studied in this paper did not liquidate, but
that is hardly enough to classify Chapter 11 a job savior. After
bankruptcy, jobs are reduced substantially and they do not reappear.
Some economists posit that workers overall would have been better off
had the company reformed, sold itself to another company, or
liquidated and made room for new business growth elsewhere. It is a
variation on a question that economists have posed with increasing
fervor: are we better off if we leave law out of the market?

         Empirical studies suggest while the market may be a powerful
force, it is not necessarily a benevolent one. America’s workers have
felt the sting of a changing economy as companies place the risks of
uncertainty on the shoulders of their employees.55 As Jacob Hacker
points out in The Great Risk Shift, this burden “is being borne almost
entirely by workers and their family on their own, rather than with the
help of corporate or government measures that might pool and manage
these growing risks.”56 Even if it is true that the death of one
workforce enables the birth of jobs in other areas of the economy, new
jobs are likely to offer lower wages and fewer benefits. The nation’s
largest employer has changed from General Motors—which paid its
workers $29,000 a year in today’s dollars and provided substantial
health and retirement benefits—to Wal-Mart, which pays its workers
$17,000 a year and “offers no guaranteed pension, and covers less than
half its workers through its health plan.”57 When a manufacturing
company liquidates, its workers are often forced to look for jobs in the
service sector. Hacker points out that these jobs “feature relatively
low pay and limited benefits, with restricted opportunities for
advancement.” He further notes that these companies rely “on part-
time and contingent workers . . .”. In previous decades workers who
lost jobs could often by re-employed in similar lines of work, but
today unemployment is often permanent, and may end “only when
workers accept a new job that often implies major cuts in pay, hours,


55
   Jacob Hacker, The Great Risk Shift, at 65 (2006).
56
   Id. at 68.
57
   Id. at 80 (citing Paul Krugman: “Always Low Wages, Always,” New York Times,
May 13, 2005, 23).




                                      20
or both.”58 Workers train for a specific company, and, as Hacker
explains, “when those firms, industries, or occupations go belly-up,
workers cannot easily exploit the skills they have gained and they face
inevitable costs—the costs of retraining, the costs of taking a job that
does not require their skills, or, worst of all, the costs of leaving the
workforce altogether.”59 The employment model has shifted from
“shared fate to individual gain,” and workers are paying for it.60

        Given this reality, if the availability of reorganization prompts
companies to choose to keep assets—and jobs—in one place rather
than sell divisions or be subsumed by another company, this could be
extremely valuable to employees. The costs of relocating or searching
for a new job are immense. Perhaps more importantly, as Professor
Hacker’s research indicates, the new employment is not likely to offer
the same pay and benefits. If Chapter 11 can play a role in curbing the
negative effects of our society’s increasing attraction to risk-taking, it
may be doing just what Congress hoped for it.

VI.       CONCLUSION

        This paper explored patterns of employment in large publicly
traded companies that filed for Chapter 11 reorganization. The
Complete Data Study and Partial Data Study taken together indicate
that companies decrease their workforces by about half in the years
around bankruptcy filing. For companies in the Complete Data Study
as well as for companies in the Partial Data Study, this decrease begins
two years prior to bankruptcy filing. Although the year-to-year
decrease in employees after bankruptcy filing is only statistically
significant for companies in the Partial Data Study, neither Study
revealed any significant increase in employment levels after Chapter
11 filing—employment levels remain about 25% lower than the level
at the year of filing. These data help fill in the picture about what
happens to large, publicly traded companies that reorganize under
Chapter 11. We know that these companies experience a significant
rate of asset reduction (around 25%), an even great rate of workforce
reduction (around 50%), and an even higher rate of management
turnover (71 to 92%). Reorganization is a period during which tough

58
   Id. at 68.
59
   Id. at 79.
60
   Id. at 67.




                                   21
choices must be made. Because scholars disagree about what
constitutes a “successful” reorganization, they will also disagree about
which choices are the right choices for achieving successful
reorganization. But regardless of ideology, scholars should agree that
knowledge about the actual behavior of corporations in Chapter 11
precede both praise and critique of the system. This paper offers a
contribution to that requisite understanding.




                                   22
Appendix A: Complete Data Study


                                          Significance of
                          Sum of          Decline from
                          Number of       Prior Year
                          Company         Relative to
       Year               Employees       Zero

       5 Years Prior to
       Filing                   675,489   p>.05

       4 Years Prior to
       Filing                   742,370   p>.05
       3 Year Prior to
       Filing                   798,205   p>.05

       2 Years Prior to
       Filing                   817,073   p>.05
       Year Prior to
       Filing                   729,057   p=.038

       Year of Filing           554,019   p=.047
       Year After
       Filing                   480,222   p>.05
       2 Years After
       Filing                   423,057   p>.05
       3 Years After
       Filing                   443,909   p>.05
       4 Years After
       Filing                   421,648   p>.05
       5 Years After
       Filing                   414,490   p>.05

      Appendix B: Partial Data Study




                                            23
                               Significance of
                 Cumulative    Change from
                 Average       Prior Year
                 Chanve in     Relative to
Year             Employees     Zero


10 Years Prior
to Filing to 9
Years Prior to
Filing           Plus 3%       p=.034
9 Prior to 8
Prior            No Change     p>.05
8 Prior to 7
Prior            Plus 6%       p=.032
7 Prior to 6
Prior            Plus 6.3%     p>.05
6 Prior to 5
Prior            Plus 10.2%    p=.037
5 Prior to 4
Prior            Plus 16.2%    p=.031
4 Prior to 3
Prior            Plus 22.7%    p=.029
3 Prior to 2
Prior            Plus 23.8%    p>.05
2 Prior to 1
Prior            Plus 17.7%    p=.029


Year before
Filing to Year
of Filing        Plus .06%     p=.024


Year of Filing
to Year after
Filing           Minus 9%      p=.024
Year After
Filing to 2
After            Minus 19.4%   p=.023
2 After to 3
After            Minus 24.7%   p>.05
3 After to 4
After            Minus 28.8%   p>.05




                                 24
4 After to 5
After           Minus 29.2%   p>.05
5 After to 6
After           Minus 24.2%   p>.05
6 After to 7
After           Minus 25.9%   p>.05
7 After to 8
After           Minus 26.8%   p>.05
8 After to 9
After           Minus 24.6%   p>.05
9 After to 10
After           Minus 27.2%   p>.05




                                25

				
DOCUMENT INFO
Categories:
Tags:
Stats:
views:3
posted:5/24/2012
language:
pages:25
shenreng9qgrg132 shenreng9qgrg132 http://
About