-------

-------
I.  GENERAL ANALYTIC APPROACH




A.  Introduction




     The  financial  responsibility regulations  under RCRA reoroposed  on




May 19,  1980  included  a financial test as an option  for demonstrating a




firm's  capability  to meet  the  costs of  its  facility closure and  post-




closure obligations.   The  proposed financial test for closure and  oost-




closure care  included elements designed to verify  the current ability  of




a firm  to pay such costs  (net  working capital  equal to or greater  than




twice the  estimated  closure and/or post-closure costs: a minimum of S10




million in net worth), and a  financial  ratio  designed  to indicate the




overall viability of a firm (a  total  liabilities to net worth ratio  of




less than 3 to 1).   The financial  test  provision prompted many public




comments.   Several  commenters expressed  general reservations about the




validity  of  the financial  ratio  used  as an  indicator of viability,  or




questioned  the net working capital  or  net  worth  reauirements,  and




suggested  that an  evaluation  of  financial  test  performance  should  be




carried out  to determine   the  effectiveness of  financial tests.   The




Agency agreed  with  this proposal, and conducted  a rigorous and compre-




hensive evaluation of over  300 alternative financial  tests.




     This Appendix describes the methodology employed in performing this




evaluation and summarizes   the results  of the analysis which provide  a




set  of   "best" candidate   tests.    The  results  of  special  analyses,




performed  to  determine  the comparative effectiveness of one-year versus




three-year  eligibility  requirements,   and   the need for  specialized,







                                  1-1   o

-------
industry-specific  ratio tests,  are also  summarized  in  this  Appendix.




The  "best"  candidate  tests  were  further  evaluated  against  a  set  of




algorithms  designed to  estimate  the  total  direct  public  and  private




costs that  would  result from their  implementation.   The results of  the




cost calculations performed, and their implications  for the  final selec-




tion of a financial test, are discussed in Appendix  B of  this  Background



Document.




B.  Possible Elements of Financial Tests




     The  ideal  financial test would pass  all  firms capable of meeting




their closure and  post-closure obligations  while failing all  firms  that




would  enter bankruptcy and  not  be able  to  meet  those   obligations.




Unfortunately,   no  method   of   financial   forecasting  is  capable   of




achieving  this  ideal.   It  is  therefore useful  to  examine  a ranee  of




possible  financial  tests.    Some of these  tests have the  advantage  of




eliminating almost  all  firms  that  would  be  unable .to  meet their closure




and post-closure obligations, but at the same time they also eliminate a




large number  of firms  that could  meet  those  obligations:  other  tests




have the advantage of passing a much higher percentage of the  firms  that



could meet  those  obligations but they eliminate  fewer firms  that  could




not meet their obligations.



     A  firm  may  be  unable  to  meet  its  closure and   post-closure




obligations  for a  variety  of  reasons.   In  addition to  the ordinary




business  misfortunes  which  may lead  a  firm  into   bankruptcy,  a   firm




owning a  hazardous  waste treatment,  storage or disposal facility (TSDF)



might  fail  to  meet those  specific obligations  as  a result  of unique






                                   1-2

-------
events.    For  example,  a firm  that might  otherwise  not have  entered




bankruptcy  may be  forced  into  bankruptcy  if  its permit  is  suddenly




withdrawn  and  it  suddenly has to put up the necessary  funds  for closure




and post-closure.   A  firm also might enter bankruptcy  unexpectedly  as  a




result  of  large  cleanup  or  repair  expenditures unique  to a  hazardous




waste TSDF or of a large liability judgment.




     This Appendix analyzes the  performance of different  possible  finan-




cial tests, with respect to the  percentage of firms which pass  the tests




and could  meet  their  financial obligations,  and the percentage  of firms




that could not  meet their  obligations but still pass a given  test.   The




question of what  test is  to  be preferred must then be  determined  by the




relative costs associated with the different tests.




     Three types  of  elements which might  be  used  as criteria  in^finan-




cial tests are considered in  this Appendix:




     (1)  Financial ratios:   These are ratios of the financial  variables




found in the income statements or balance sheets of firms.




     (2)   Multiples:   In this text,  these are measures  of unencumbered




assets and/or  liquidity as multiples of  the  specific  financial obliga-




tion considered (closure and  post-closure).




     (3)  Net worth:  This is a requirement, independent  of the  multiple




requirement described above,  that  a firm must  have net worth above  a




specified level.




     Each of these possible  elements  of  financial  tests has a  ootential




role in limiting the number of firms that will fail to meet their  finan-




cial obligations.  By requiring firms to meet specific  requirements with







                                   1-3

-------
respect to  their  financial  ratios  in order to pass a  financial  test,  it




is possible  to  limit the  percentage of firms  which pass the financial




test but which  ultimately enter bankruptcy.   Multiples serve to  ensure




that the financial obligation itself does not? have so  great  an effect  on




the firm that it is a cause for bankruptcy.  Without multiples as  a  part




of the test, a small firm with excellent financial ratios might  still  be




unable to meet  its closure and post-closure  obligations because  of  the




large  size   of  their  obligations  relative  to  the  size  of  the  firm




itself.  A  minimum net  worth requirement serves  two  purposes:    (1)  it




has been found  that  the larger the  firm the  less likely it is  to fail,




and  (2)  significant   size   enables  the   firm  to  withstand  unusual



contingencies associated with the  ownership  of  hazardous  waste  TSDFs.




Financial ratios  and minimum  net  worth requirements  both  can  help  to




ensure that the failure  rate  of  the firms  which will  pass the financial



test is lower than the average failure rate for all firms.   The  multiple




element of  the test and the minimum net worth requirement help to  ensure




that the failure  rate of  the firms considered is  not higher than  that




for firms in all  industries,  due  to the unusual  size of these special



financial obligations relative to the size of the firm or due to contin-




gencies associated with the ownership of hazardous waste TSDFs.



C.  Overview of the Methodology




     This Section, and Figure  1-1,  provide  an overview of this  Appendix




by outlining  the   major  steps of  the methodology employed.    Detailed



discussions of the topics outlined in this section are then  presented  in




the following  sections  of  this Appendix.    Figure  1-1  illustrates  in






                                  1-4

-------
             Effect of
             Multiples
 I
Ul
                                        Share of
                                       All Klrms
                                       Wlilch Fall
                                          (F)
       F x
 Percentage
of Bankrupt
Sample That
 Pass Test
 (Up = Mls-
 classlfled
  bankrupt
   firms)
              Average
              Failure
              Rate  for
             All Firms
         I
       FxM,,
         + (F x Mp)
(Firms which do go
   bankrupt as a
share of all firms
   passing test)
                    x 10,000
              Effect of
              Net Worth
              Requirement
\
Share of
All Firms ..
Percentage
of Non-Bank-
nv . rupt Sample
Not Falling l '"NB That Pass
(1-F) Test
(ANB>
                                                                                                       Bankrupt
                                                                                                         Firm
                                                                                                        Sample
                                        Tests
                           Ratios
                          Employed
                        - in tests -
                          and Cut-
                         off Points
Evaluation
    of
Individual
  Ratios
 Against
 Samples
                                                                                                      Comments
                                                   Non-
                                                 Bankrupt
                                                   Firm
                                                  Sample
                                                                            FIGURE  1-1

                                                                    APPENDIX A METHODOLOGY

-------
particular the steps which were used  to  derive  the  effectiveness measure




(Ep)  of any  given test,  that, is,  the most  probable  estimate of  the




number  of  firms  per 10,000 which  pass  a given financial  test  but  later




fail without providing alternative financial assurance.            ,




1.  Determination of Baseline Failure Rate




     In  order  to  compare the  effectiveness  of  various  tests,   with




respect to the relative  costs  associated with these tests, it  is neces-




sary to determine  the  failure  rates of the set of  firms which  may  use a




financial  test  as  evidence of  financial responsibility.   The  baseline




failure  rate  (F) is  derived  by  first determining  the  average  failure




rate for all  firms  based on the historical data, and then by  adjusting



this rate  to  account  for  the  effect of a  net  worth requirement on  the




average  failure  rate  for  firms  which  may  be eligible  for a  financial



test.  In addition  to determining  the failure rate  of  firms, a  financial




test includes  a  multiple requirement  to provide assurance  that at  the




time a  firm  ceases  to pass the  test it will continue to  possess assets




sufficient  so that it  is able  to  pay for  closure and post-closure




costs.   Otherwise,  it is  possible that the attempt to  pay these  costs



will  be sufficient to  force  the firm  into  bankruptcy  before these




closure  and   post-closure   costs   have  been  fully  paid.      In   the



nomenclature adopted in this Appendix, the fraction of all firms  consid-




ered which will  go bankrupt  is  designated as F;  the  fraction of  all



firms considered which will not go bankrupt is equal to  1-F.
                                  1-6

-------
2.  Selection of Financial Ratios




     The  financial  ratios to  be  evaluated as  candidate  ratios for  the




financial test were  selected  from three sources:  (1) comments  received




on  the  May  19,  1980 regulation;  (2)  surveys of opinions of  bond  ratine




services and credit  analysts;  and (3) existing literature on bankruptcy




forecasting.   The  list of  ratios assembled  was extensive  and  varied




widely in content and estimated effectiveness.




     An  initial  set of  financial  ratios was  chosen  for  further  evalu-




ation because they  satisfied  three basic conditions:   (1) they  produced




significant  predictive  results in the prior  literature;  (2) they were




frequently identified by bond rating services  and credit analysts  as  key




Darameters:  and  (3) their values  were readily available  from  corporate




balance sheet data.




3.  Sample Selection




     In  order  to  determine  the  effect of  financial ratio  tests,  the




Agency  gathered  samples of  non-bankrupt  and  bankrupt   firms.   Non-




bankrupt firms, also referred to as viable firms, are firms that did  not




enter bankruptcy  within three years  of the time  period  for which data




were gathered.   For  bankrupt firms,  financial data were  gathered  for




three years  prior  to the time the  firm entered bankruptcy.   The  sample




of bankrupt firms was gathered from a  variety  of sources.  The  sample of




non-bankrupt firms  was  gathered   using the  Industry Index  of Moody's




Industrial Manual  to determine  members  of  industrial categories that




generate, treat, store,  or disnose of hazardous waste on-site.
                                  1-7

-------
4.  Evaluation of Financial Ratios




     These  ratios  were  first tested  individually  against  the  primary




bankrupt  and  non-bankrupt  firm  samples,  using a  variety of  pass-fail




cutoff  points for  each ratio derived from  the  bankruptcy  forecasting




literature.   A  second  set  of  financial  ratios  was  then selected  and




tested  against  a sample of  bankrupt and  non-bankrupt firms which  were




consistently misdassified by the original ratios,  to  determine if there




were  additional financial  ratios  which  would improve  the predictive




ability of a test.




5.  Design of Financial Tests




     The  most  promising  ratios  were  combined to form  151  alternate




multi-ratio tests.   The ratios  were combined  in  three  ways:   (1)  two-




ratio  tests  (firms must  pass both  elements  of the test  to pass);  (2)




three-ratio tests  (firms must pass all  three  elements of  the test  to




pass);  and (3)  three-ratio  contingent  tests  (firms  must  pass two  of




three  conditions  to pass).   The  tests  were then  retested  against  the




primary sample to derive for  each test the percentage of  bankrupt firms




of the sample that  pass  the  test and are therefore misclassified by  the



test  (Mp)  and the  percentage of  the  non-bankrupt  firms  sampled which




pass the test (Ajjg).



6.  Determining the Effectiveness Measure  (Ep)  for  a Given Test




     The effectiveness measure for any given test  is  calculated accord-




ing to a  two-step  process  (see Figure  1-1 above.)  First, the  fraction



of all firms which  will  go  bankrupt (F)  is multiplied by  the percentage




of the bankrupt firm sample that  is misclassified  by a given test (Mp),






                                  1-8

-------
that  Is,  pass the  test  but later enter  bankruptcy.   This  provides  the




fraction of  all  firms considered that  pass  a particular test  and  later




go bankrupt  (F x Mp).   Similarly,  the  share of all firms  that will  not




fail  (1-F)  is  multiplied by  the percentage  of  the  non-bankrupt firm




sample which  pass  the test (A^).   This  provides the fraction of  firms




that pass a given test and do not go bankrupt  (l-F)AIjB.




     Finally,  the  Agency  uses F  x Mp and (l-F)ANg  to calculate what




fraction of  all firms that  pass  a  given test are firms  which will  go




bankrupt (in  Figure  1-1  this  operation  is represented by F x Mp divided




by the  total of all  firms which pass  the test ((1-F)ANB  + (F x  Mp)).




This fraction  is then  multiplied  by 10,000 to yield the number of  firms




per 10,000 that pass a given financial test that will  go bankrupt.  This




number (Ep) is the measure of effectiveness of a given test.




7.  Use of Performance Curves to Select the Best Financial  Test




     As  discussed  above,  a  financial  test  has  two  basic performance




measures:   (1) the percentage  of non-bankrupt firms  that  can  pass  the




test (Ajjj):  and (2) the  number of firms  per  10,000  that  pass  the test




that will later enter bankruptcy without  providing alternative  financial




assurance (Ep).   The Agency developed  performance curves   consisting  of




those tests which pass the maximum percentage of non-bankrupt  firms  for




any given percentage  of  firms  that  pass  the  test and later enter  bank-




ruptcy.   The  Agency  used these  curves   in  the  analysis  described   in




Appendix B  to evaluate the  relative costs of  alternative  tests and  to




determine a best single financial test.
                                  1-9

-------
8.  Comparison to Other Tests




     The  resulting  performance  curves for the  tests developed  In  this




study  were  then  compared  to  the May  19,  1980 proposed financial  tests




and the results of other bankruptcy  forecasting  studies.




9.  Industry-Specific Problems




     Although  a  given  financial  test may  serve  quite  well  for  most




industries,  it  is  possible that  unique  characteristics  of a  specific



industry  may  cause  a  high  percentage  of  the viable  firms  in  that




industry to  fail a  given  test.   In order to determine  the  importance  of




such  problems,   the  Agency  examined  the  financial  characteristics  of




firms in various  manufacturing industries likely to dispose  of hazardous



waste, electric utilities, and hazardous waste management  firms.




10.  The Role of  Bond Ratings in a Financial Test




     Section VII of  this  Appendix discusses  the use  of bond ratings  as




either a substitute for or a  supplement to other elements of the finan-




cial tests considered elsewhere in this Appendix.




D.  Comparison to Alternative Methodological Approaches




     The methodology adopted by  the  Agency and outlined in  Section  I.C.



is not  the  only  methodology that  might be employed  to derive  a set  of



financial tests.   This Section  considers  three  alternative approaches:



(1) the use  of functions  of financial  ratios  rather  than cutoff points



for financial ratios: (2) the development of industry-specific financial




tests rather  than  single  tests  for  all industries:  and  (3) the use  of



population sampling  instead of  the  sampling  procedure  employed in  this




analysis.
                                   1-10

-------
1.  The Use of Functions of Financial Ratios




     The approach  used  by the Agency was  to examine a series of  finan-




cial tests consisting of  requirements  that specific financial ratios  be




at certain levels.  An alternative approach would be to design financial




tests that consisted of linear or nonlinear functions of a set of  finan-




cial ratios (for example, a linear function of financial ratios would  be




a test that required  that (a  x Rj) + (b x R2) + (c x R3)  > d, where  a,




b, c, and d are constants and Rp R2, and  R-j are ratios).  Three methods




have been proposed in the literature for developing financial tests  that




consist of functions of financial ratios.




     One approach  is  to rely on expert  judgment to decide which  ratios




should be  included and  what values  should be  associated with them  (see




Tamari,  1966).    This  approach  has  the disadvantage  that  there  is  no




statistical method of determining whether the  ratings  or the financial




ratios chosen have been appropriately determined.




     A second approach that has been proposed is to develop  a functional




model of the factors that lead to business failure and to use this model




to  estimate  the   probabilities  of  business failure  for  any specific




firm.    Though  theoretically  elegant,   this   approach  has only  been




attempted  once in the  literature reviewed  for  this  study,  and  that




attempt did not include any statistical validation of the model (Wilcox,




1971, as discussed in McWilliams,  1977).




     The  Agency  decided  not  to  develop  a  functional  model  for  two




reasons.  First, the time available in which to  carry out the study  of a




financial test  option was relatively  short and might not  have allowed
                                  1-11

-------
 sufficient  research  and  refinement  of  the model  to  produce  useful




 results.   Second, because  the  technique was  untested  in this  applica-




 tion,  the  Agency was concerned  that  it might prove  unsatisfactory,  and




 that,  if it were not  effective,  insufficient time would  be  available to




 carry  out an alternative analysis.




     The third  approach,  and that most commonly used in the  bankruptcy




 forecasting literature, is that of multi-discriminant analysis.   This is




 a  statistical  search  technique for arriving  at  the  best linear  or non-




 linear function  for a set  of chosen financial ratios for discriminating




 between bankrupt and non-bankrupt firms.  This is  the most statistically




 sophisticated approach  to the  problem.   The use of this  approach  was




 rejected in  this study  for two reasons:   (1)  a  function  of financial




 ratios would be more difficult to compute, more  difficult to administer,




 and more difficult to check  quickly than a simple set of ratio  require-




ments; and (2)  it has been  pointed out several  times in the  literature




on  bankruptcy  forecasting  that multi-discriminant  analysis  has   not




yielded more accurate forecasts  of  bankruptcy than the simpler approach




adopted here (Deakin,  1972 and McWilliams,  1977).   A comparison of  the




results of the  financial  tests  developed in  this  Appendix  to financial




 tests  developed  in  bankruptcy forecasting  studies that employed multi-




discriminant analysis is included in Section V.




2.  Use of  Industry-Specific Tests




     A number  of commenters suggested  that  the  Agency  adopt specific




tests  for different industries.   The problem  with this  approach is  the




extreme difficulty of gathering  sizable samples of bankrupt  firms.   An
                                  1-12

-------
extensive  search  (described in  Section III)  for firms  which had gone




bankrupt and which had publicly available financial statements yielded a




total of  only 95 firms,  of  which many were  retail  firms and none were




electric utilities.  To further subdivide this sample into, for example,



two-digit SIC codes would result  in samples of bankrupt  firms too  small




to be significant.  Thus, while industry-specific tests might be theore-



tically  desirable,  they  cannot  be developed  from the  data currently



available.  A review of the bankruptcy forecasting literature found only




one  study which had  developed industry-specific  tests (Altman,  1973).




This study was  of  the  extremely bankruptcy-prone railroad industry, and



still had to use  bankruptcies over  a period of  50 years  in order to



develop an adequate sample size.



3.  Use of Population Sampling



     As  outlined  in Section I.e.,  the performance of  a financial test




was  developed  in  two  steps:    first,  the  development  of  a baseline




failure rate, and second, the determination of how this baseline failure



rate was reduced by the performance of a specific test.   The performance




of  each  test  was  measured  against  the samples  of bankrupt  and non-




bankrupt  firms.   An alternative  procedure  would be  to gather relevent



data for  a large population of firms, determine which ones later entered




bankruptcy and  which  ones did  not, and then examine  each test against




this entire  sample  for  its  ability to forecast bankruptcy.  The problem




with this  approach  is  that bankruptcy  is a relatively rare event.  The




average  business  failure  rate  in the  post-war  period  has been approxi-




mately  44 per  10,000  firms  per  year.   This implies   that  in order to






                                   1-13

-------
obtain a  sample  of bankrupt firms  as  large as that used in  this  study,




it would have been  necessary to obtain data on a  total of  15,000  firms.




Such an effort would have been  beyond  the scope of this study, and  has




not been attempted in any previous study of bankruptcy forecasting.   The




structured  sample  employed  in  the  study  should  provide,  however,




approximately equivalent results.
                                  1-14

-------
II.  BASELINE FAILURE RATES AND DETERMINATION OF MULTIPLES

A.  Baseline Failure Rates

     This Section  explains  how the Agency determined a baseline failure

rate applicable  to large firms from the  existing  data on failure rates

by  size  of  firm.   For the purposes  of  this study,  "large  firms" were

considered to be firms with over $10 million in net worth.

     Table  II-l  is  a  compilation by Dun &  Bradstreet  of  historical

failure rates  per  10,000 firms,  incorporating  all sizes  of  firms.   As

shown in  the Table,  the arithmetic mean  for  the period 1950-1978 is 44

business failures  per year  per 10,000  firms.   This is probably too high

a baseline failure rate for the kinds  of  large firms considered in this

study.  Greater assets and net worth reduce the probability of corporate

failure.  The  failure  rates  shown in the Table for  all  firms  thus may

overstate the rate of failure  for large firms.  Altman suggests a reason

for this phenomenon:

   [Mjany of these companies are not permitted to fail.  Except in
   the event of  fraud,  or where the failing company is simply too
   large, we  rarely  observe  in the decades before  1970  firms  of
   over $25 million  in assets  actually  going bankrupt.   In many
   cases, the financially troubled firms can expect to be wooed by
   a highly liquid or managerially rich firm, usually resulting in
   a  merger  absorption  before insolvency  in a bankruptcy  sense
   occurs. (Altman, 1971)

During  the  period from 1969  to  1974, for example,  Douglas  Aircraft,

Ling-Temco-Vought,  Lockheed Aircraft, and Mohawk Data Sciences came very

close to  failing,  but failure was  prevented  by  government intervention

or private sector efforts (Altman et al.,  1977).

     Precise quantitative data on  failure  rates  by size of firm are not

available.  Dun & Bradstreet, which maintains the only extensive failure

                                 II-l

-------
                       TABLE II-l
       ANNUAL BUSINESS FAILURE PER  10,000 FIRMS
Years
1970-78
1960-69
1950-59
1940-49
1930-39
1950-1973
1930-1978
Arithmetic
Mean
36
52
42
26
86
44
49
Maximum
Year
44
64
56
63
154
64
154
Number
of Years
Over 60
0
2
0
1
8
2
11
SOURCE:  Derivation from Dun & Bradstreet (1979).
                          II-2

-------
rate  data now  published,  does  not regularly  provide data  on failure




rates by  size  of assets, sales, or net  worth.   Although a good deal of




indirect  evidence  suggests  that  larger firms  have substantially lower




failure rates,  a combination or comparison of  the  different  data bases




is difficult.   The estimate in this study  of the baseline failure rate




for large firms was derived from the following sources:




   (1)   Dun  &  Bradstreet  publications  on  the  relationship  between



         credit ratings and failure rates;




   (2)   Quantitative  data from Altman (1971)  combined  with IRS




         statistics on the number  of  corporations  above a given




         size;  and



   (3)   A study by Richard  C.  Edwards on long-term failure rates




         for very large firms.



1.  Failure Rates by Credit Ratings and Firm Size




     Dun  &  Bradstreet credit  ratings  are  composed of two  elements —




size and  creditworthiness.   Size is defined  as  the companies' tangible



net  worth.   Creditworthiness  is  rated from  1  (high) to  4  (limited).




Table 11-2 provides the key to the ratings classifications.



     A study of the relationship between business failures and these Dun




&  Bradstreet  credit  ratings  has  been published.   The  results  of  the



study  are presented  in  Table  I1-3.    Because  the  size  categories  are




approximately equal to the tangible net worth  of  the  firm, two conclu-




sions can be drawn  from  the  results in this Table.  First, firms in the




top  two  categories  for financial strength  (representing  firms of above




$50  million  in  tangible net worth and  firms  of  between $10  and  $50







                                 II-3

-------
                             TABLE II-2




                           KEY TO RATINGS


















- •
5A
4A
3A
2A
1A
BA
BB
CB
CC
DC
DD
EE
FF
GG
HH
-Estimated Financial
Over
$10,000,000 to
1,000,000 to
750,000 to
500,000 to
300,000 to
200,000 to
125,000 to
75,000 to
50,000 to
35,000 to
20,000 to
10,000 to
5,000 to
Up to
Strength
$50,000,000
50,000,000
10,000,000
1,000,000
750,000
500,000
300,000
200,000
125,000
75,000
50,000
35,000
20,000
10,000
5,000
Composite
High
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Good
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Credit
Appraisal
Fair Limited
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
SOURCE:  Dun & Bradstreet.
                                II-4

-------
                              TABLE II-3

        RELATIONSHIP OF D&B CREDIT RATINGS WITH BUSINESS FAILURES
Rating
Category
5A1/
4Ar,
3A-
2A
1A
BA
BB
CB
CC
DC
DD
EE
FF
GG
HH
Percent of
Failures
Rated High
1971
0.00%
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.01
0.01
0.00
0.05
0.00
1972
0.00%
0.00
0.01
0.00
0.00
0.00
0.00
0.01
0.01
0.00
0.02
0.00
0.02
0.06
0.00
Percent of
Failures
Rated Good
1971
0.00%
0.00
0.00
0.09
0.00
0.05
0.02
0.04
0.05
0.05
0.06
0.08
0.08
0.08
0.07
1972
0.00%
0.00
0.02
0.04
0.00
0.03
0.02
0.05
0.04
0.05
0.06
0.08
0.09
0.08
0.06
Percent of
Failures
Rated Fair
1971
0.00%
0.00
0.33
0.37
0.36
0.30
0.35
0.31
0.39
0.37
0.41
0.46
0.46
0.43
0.21
1972
0.00%
0.00
0.50
0.23
0.25
0.15
0.18
0.28
0.36
0.43
0.52
0.52
0.49
3.63
3.50
Percent of
Failures
Rated Limited
1971
0.00%
0.00
0.00
0.00
4.00
0.31
1.50
1.17
1.65
1.46
2.05
1.89
1.70
1.33
0.50
1972
0.00%
0.00
0.00
0.00
0.00
0.00
0.00
1.02
1.48
2.46
2.04
0.00
1.70
1.40
3.49
                               SUMMARY

                  Percent Failures to Listings Rated

                                           1971
            1972
          High
          Good
          Fair
          Limited
          Numeral
          Blank
          Total Failures
            per 10,000 listings
0.004%
0.07
0.40
1.42
0.06
0.75
0.007%
0.06
0.40
1.51
0.07
0.64
42
38
   U,
   -Category represents firms with over $50 million in  tangible
net worth.
   2/
   -Represents firms with between $10 and $50 million in  tangible
net worth.
   -Represents firms with between $1 and $10 million in tangible
net worth.

SOURCE:  Sanzo (1974).
                                   II-5

-------
million in net worth) have substantially fewer  failures  for  every  credit




appraisal rating than do firms in lower categories, including firms with




from $1  to  $10 million in net worth.   The Table shows  that the percent




of failures is .00 percent for the top two credit ratings.   However, the




failure  rate  for  .firms  in  the  financial strength  category of  3A or




lower,  i.e.,  firms with  less  than  $10  million in  tangible net worth,




increases, especially for those  firms  with a "fair" credit  rating.  The




"limited"  rating is  given  to  firms  for  which  insufficient data  are




available to  provide  a more specific  rating.  As  the Table shows,  the




majority of firms with a "limited" rating are smaller firms  in the lower




financial strength categories.



     One problem in using these data is that when very large firms, such




as W.  T.  Grant, are  believed  to be approaching failure, they  are  not



given a "limited" rating but rather are given a special  rating indicated




by a  double  hyphen.  This  means that if  Dun & Bradstreet  thinks that



failure is  a  possibility  for a very large  firm,  it  is not rated  at




all.   Nevertheless, this tends  to be a relatively small category (Backer




and Gosman, 1978).



     Table II-3 also shows that  Dun  &  Bradstreet haa greater difficulty




accurately rating  small  firms  than  large  firms.    For every  credit



rating, there  are  significantly more  failures  in  the  small-firm  cate-




gories than in the  respective large-firm categories.




2.  Quantitative Data from Altman Analysis  and IRS Statistics



     In 1970,  12 firms of  over  $10 million  in assets  filed  for  bank-



ruptcy (Altman 1971).   At  that time,  there were  23,300  corporations  of
                                 II-6

-------
over $10 million in assets (U.S. Internal Revenue Service, Statistics of

Income,  Corporate  Income Tax Returns).   Thus,  the  implied bankruptcy

rate in  1970 for firms  of  over $10 million in assets is approximately 5

per 10,000.   Given that the  failure  rate  for all firms  is  over 44 per

10,000, this implies that the failure rate for firms of over $10 million

in assets is only about 11 percent of that reported for all firms.

3.  Long-Term Failure Rates for Very Large Firms;  Data from Edwards
    Analysis

     In a study conducted in  1975,  Richard C. Edwards traced the corpo-

rate development of 225 very  large firms  from the years  1919  to 1969.

Very large  firms  in the Edwards  study were  defined  as  firms  with over

$50 million (in" 1967  dollars) in assets  in the year  1919.    This  is

approximately equivalent to $100 million in assets in 1980 dollars.  The

purpose  of  Edwards' study  was to  determine  the extent  to  which large

size  alone  might  effectively  prevent  the  possibility  of  business

failure.  Table  II-4  shows  the results  of  this  study.   For comparison,

the average failure rate for  all  firms for this period, as derived from

Dun &  Bradstreet  data,  is 61  firms per  10,000.   As  is shown  in the

Table,  the  failure rate for large  firms as  a percentage of the failure

rate  for all firms  varies  from  18 to  32  percent,  depending  on which

class  of  firms  is   examined.   This wide variance is  largely due to the

inclusion in  the  sample  of  15 urban  transit  firms, virtually  all  of

which failed during the 1920's and  1930's.

4.  Summary of Failure Rate Data Bases and Conclusions

     Table  11-5  summarizes  the results  discussed above.   These results

strongly suggest significantly lower failure rates for larger firms, but

                                 II-7

-------
                           TABLE II-4



               FAILURE RATES OF VERY LARGE  FIRMS



Class of Firm
Manufacturing
All Firms Except
Urban Transit
All Large Firms



Number
°f I/
Firms^
110
210

225



Number
°f I/
Failures^
8
11

22



Implied
Annual
Rate Per
10,000
Firms^/
15
11

19.5
Implied
Failure
Rate as a
Percentage
of Dun &
Bradstreet
Failure Rate
For all Firms
(61/1Q,000>3/
24
18

32
— Number of firms and failures are from Edwards  (1975).


— Implied failure rate calculated as:



                         number of failures
                                              1/50
                     1 -
                          number of firms
3/
— Derivation from Dun & Bradstreet data.
 SOURCE:  Derivation from Edwards, 1975.
                              II-8

-------
                           TABLE II-5

          SUMMARY OF EVIDENCE ON FIRM FAILURE RATES
                    BY SIZE OF NET WORTH
         Size Category
 Failure  Rate
as a Percentage
of Failure Rate
 for All Firms
+$10 Million in Net Worth with a D&B Fair Credit
  Rating (1971-1972)                '.  •

+$25 Million in Net Worth (1970)

+$100 Million in Assets (1919-1969)
      11

     18-32

-------
a  precise  estimate is impossible to  determine  given the quality of  the




data bases.   For example, the estimate  of  a 1  percent failure rate  for




firms with over  $10  million  in net  worth as a percentage of  the failure




rate for all  firms based  on  firms with a Dun & Bradstreet "fair" credit




rating is probably an underestimate.  Since Dun & Bradstreet  provides no




data on the distribution of firms by size, it is impossible to determine




from Table II-3  if the  entries of .00 percent,  which Dun and Bradstreet




states may  be any percentage less than  or  equal to  .004,  represent a




significant number of  firms.   Further uncertainty  arises  from the fact




that large firms that are expected to fail are not rated at all.



     Although  it is  impossible  to derive  a precise failure  rate  for




large firms  from the data available,  all of the data  suggest that  the




failure rate  is  at least  50  percent lower than the failure rate for  all




firms.  The Agency concluded  that an  estimated  baseline failure rate of




22 per 10,000 for firms with  over  $10 million  in net worth was reason-




able, assuming that  the failure  rate for large  firms would  be at least




50 percent  lower than  the estimated  failure rate of 44 per 10,000  for




all firms.   Considering the  quality of  the  data,  however,  even smaller



failure rates than 22 per 10,000 are possible.




B.  Ability to Pay



     The fact  that a firm has not  actually  entered bankruptcy does not




ensure that it has adequate funds available  to  meet financial responsi-




bility obligations.   This  Section therefore  describes how  the  Agency




determined if adequate funds  would be  available to  those  financially



weakest  firms which pass  a  financial  test  to meet  their financial






                                 11-10

-------
responsibility  obligations.    If  a  non-bankrupt  firm  does  not  have




adequate  funds,  then  such  obligations  may  themselves  be  a  cause  of




bankruptcy.  This  Section  also  describes  how the Agency determined what




additional requirements  are necessary  to ensure  that  such obligations




can be  met without increasing  the failure rate.   These determinations




were made  primarily through  the examination of  the  characteristics of




several firms  that would  pass  many  pos.sible financial  tests  and enter




bankruptcy within  three years of passing  the test.  Such firms were the




financially weakest of all firms that could pass financial tests.




1.  Availability of Funds




     To analyze the adequacy  of the  funds  available to financially weak




firms immediately  prior  to bankruptcy,  the Agency  examined  a sample of




12 bankrupt  firms.  These  firms passed many ratio tests (see Sections




III  and  IV)  three  years  prior to  bankruptcy,  and  tests which  were




stringent enough to eliminate them also eliminated a large percentage of




viable  firms.   The  fact  that  several of  the  12  firms also  passed  a




number  of  tests  two  years  prior  to bankruptcy  indicates  that  their




financial  status  deteriorated  very  rapidly.   They therefore represent




the type of firms most likely to fail a financial test only at the point




where  they lack  sufficient  funds  to  establish an  alternate  form  of




financial assurance.  The Agency examined financial data for these firms




to determine whether they would have been able to meet financial respon-




sibility obligations  at the  time  they were  eliminated by  a financial




test.
                                 11-11

-------
     Table  II-6 shows,  for the  12 firms,  the changes  in net  working
capital,  net worth  and  quick assets  occurring  over the  three  years
preceding  bankruptcy.   As  can be  seen from this Table,  even if  these
firms did not fail a financial test until two years prior  to bankruptcy,
all of  the  firms  still would have had positive net worth  and  some  quick
assets.   In  fact,  the quick  assets  of  most  of  these  firms increased
between  the third and  second  year prior to  bankruptcy.   Therefore,  if
these firms  had failed a financial test  two  years prior  to bankruptcy,
they would have had significant amounts of cash and other  liquid  assets.
Current assets would have exceeded current liabilities for two-thirds  of
the firms, and  total assets would have exceeded total  liabilities for  11
of the 12 firms.
     If the establishment of an alternative means of financial assurance
for these  firms were  delayed by, for  example,  legal  proceedings,  Table
II-6 shows  that 66  percent of the  firms  (8  out of 12) still would have
had positive net working capital  one year prior to bankruptcy.   All
except  one  firm would  have  had  positive net worth,  and  all would have
had significant quick  assets.   Therefore,  with a delay up  to one year
prior  to bankruptcy,  66 percent  of  the firms  would have  maintained
adequate  cash and net  working capital  to  pay at  least   some  of  their
obligations.   Furthermore,  negative net working capital  alone does not
necessarily  imply that a firm cannot  pay any of its  debts.   A conser-
vative  conclusion would be  that if legal proceedings  normally took one
year or less, at least  50 percent  of  the firms that pass the test up  to
two years  prior  to bankruptcy  would  still have  sufficient  funds   to
establish an alternative form of financial assurance.
                                 11-12

-------
                                              TABLE  I1-6



                      CHANGES  IN KEY  FINANCIAL PARAMETERS  FOR  12  BANKRUPT FIRMS  LIKELY

                                TO BE MISCLASSIFIED  BY A  FINANCIAL TEST
Company
Name
Walt ham
Bowmar
Polarad
Beck
Unishops
Uniservices
Remco
Maule Industries
M.H. Fishman
National Video
Dolly Madison
King Resources
Percent of
Decline From
3 Yrs To 2 Yrs
Prior to Bankruptcy
NWK
+ ±'
+
60.9
+
+
+
39.7
-
4.9
NA
+
89.5
NW
+
+
41.9
+
11.5
+
9.0
+
5.7
+
+
+
QA
+
+
+
+
+
+
+
+
13.6
51.6
+
NA
Percent of
Decline From
2 Yrs to 1 Yr
Prior to Bankruptcy
NWK
_ I/
+
+
+
-
66.7
75.6
-
2.6
21.1
83.8
-
NW
92.8
76.4
16.7
+
-
48.5
84.0
11.6
4.6
21.2
5.8
65.7
QA
35.5
12.9
26.1
+
+
27.3
12.3
50.5
+
26.9
+
NA
Percent of
Decline From
3 Yrs to 1 Yr
Prior to Bankruptcy
NWK
—
+
34.8
+
-
40.0
85.3
-
7.4
21.1
75.7
-
NW
90.2
50.7
51.6
+
-
32.0
85.4
-»
10.0
18.9
+
38.6
QA
7.3
+
15.0
+
-
20.0
6.5
38.1
+
64.6
+
NA
I
M
Lo
          ~ A plus sign indicates that there was an increase in value rather than a decline.

          21
          — A negative sign indicates that there was .a negative value so calculating the

       percentage decline is meaningless.
          NA:  Not available

-------
     To  test  this  hypothesis,   the  Agency  examined  the  net working




capital and net  worth  of  the entire sample of 66 bankrupt firmsi  Table




II-7  shows that  over  88  percent of  all of  these bankrupt  firms   had




positive net working capital and positive net worth three years prior to




bankruptcy.   In  fact,' even  one year  prior  to  bankruptcy,  89 percent




still  had positive  net  worth and 72 percent  still  had  positive   net




working capital.




2.  Multiple Requirements




     Adequate funds  will  only be available if  the  net working capital,




net  worth and quick assets  of  a  firm are  large  enough to  cover   its



required  financial responsibility obligation,  without the expenditures




themselves causing bankruptcy.   Financial test  criteria which require




net  working  capital  and  net  worth  to be equal  to  some  multiple of   the




costs of closure and post-closure would help  ensure the availability  of




adequate funds.   Table II-6, which gives  the rates of deterioration  of




net  working  capital,  net worth  and  quick assets  for the  12 bankrupt



firms  discussed  above, shows  that  the  quick assets  for most of  these




firms  deteriorated relatively slowly.   This  suggests that  a financial



test  requiring   minimum   levels  of  quick assets  will not  necessarily




ensure that  adequate funds are  available when needed.   The percentage



declines  in  net  working  capital  and  net worth  are significant,   and




suggest that  a  test  should  require  that net  working capital  and   net




worth be multiples of closure and post-closure costs.




     Under normal  circumstances,  a  firm with net working capital  equal




to twice  the  estimated closure and post-closure  costs would be able  to
                                 11-14

-------
             TABLE II-7

GENERAL ABILITY TO PAY CHARACTERISTICS
   OF FIRMS APPROACHING BANKRUPTCY
Percentage of All Bankrupt
Firms with Positive Net
Working Capital
Percentage of All
Bankrupt Firms With
Positive Net Worth
Years Prior to Bankruptcy
x-3
88
97
x-2
79
97
x-1
72
89
                 11-15

-------
pay these costs without  threatening  its  continued  viability.   Inspection

of  the  financial data  of bankrupt  firms,  however,  indicates that  net

working  capital may  often decline  precipitously  as  bankruptcy  nears.

Furthermore,  firms  approaching  bankruptcy  may  frequently   experience

negative cash flow, which will place additional  burdens  on  available  net

working capital.   Nine of the 12  firms  sampled, for example, had nega-

tive cash flows in the year prior  to bankruptcy.   Therefore,  a financial

test should  require  that net  working  capital be  a multiple  of closure

and post-closure costs,  in order to provide the  Agency with a reasonable

probability that a firm will have  resources available for the establish-

ment of an alternative form of financial assurance.

     The Agency determined  the appropriate  multiple, by assuming that  if

a  firm  met  this multiple  requirement  three  years  prior to  bankruptcy,

then it would  still have  sufficient funds for  its financial responsi-

bility obligations one  year prior to bankruptcy.   This requirement  can

be derived in the following way:

   (1)    V3 >_  M x FR

   (2)    VL 2.  FR,

where V-,  is  the value  of  net  worth or  net working capital  three years

prior to bankruptcy, V^ is the value of net worth or net working capital

one year prior  to bankruptcy,  M  is the value of the multiple, and FR is

the value  of the  financial responsibility  obligation  (closure  costs,

post-closure  costs,   and/or liability   requirements).    Solving   for M

yields:

          V.
   (3)    -:p  >  M
          vl  ""
                                 11-16

-------
     This multiple  has  been calculated for each  of  the 12 firms in the




sample  discussed above,  with  the  results shown in  Table II-8.   The




average value  of the multiple for the  firms  that experienced a decline




in net  working  capital  but still had  positive net working  capital is




2.8.   That  is,  if  net  working  capital had been equal  to  2.8 times the




cost of financial responsibility  obligations  three years prior to bank-



ruptcy, then  adequate  funds  would still  have  been available  one year




prior  to  bankruptcy  (i.e.,  after the  firm's  net working  capital  had




deteriorated).   The average multiple for  net worth  for firms which had




positive net worth remaining one year prior to bankruptcy was 3.3.



     These  multiples only  ensure  that these firms  would  have  had  net




working capital  and net worth exactly  equal  to  their  financial respon-




sibility obligations.  As noted above,  this provides no margin of safety




after the financial responsibility obligations  have been  met.   If both



of  the  above  multiples  are  rounded  to 3,  and  then multiplied  by  2 to




provide a reasonable margin of safety  and to ensure  that the firm has




some  funds   remaining  to  cover  other  liabilities which may  not  be



recorded  on the balance  sheet,   then  multiples  of 6  result.   These




multiples provide  a margin of  safety to  ensure  that  the  failure rates




are not  greater than  those estimated  in  the previous  Section  (22  per




10,000), and  to ensure  that  adequate  funds  are  available even in the




event of lengthy legal proceedings to obtain the funds.
                                 11-17

-------
                               TABLE  I1-8

         MULTIPLES  OF  FINANCIAL PARAMETERS  NECESSARY TO INSURE
                        AGAINST FIRM  DETERIORATION
Company Name
Waltham
Bowraar
Polarad
Beck
Unishops
Uniservices
Remco
Maule Industries
M. H. Fishman
National Video
Dolly Madison
King Resources
Average Multiple
No. of firms that
decline in value
% of firms that
decline in value
Net Working Capital
JY
£
1.5
+
-
1.7
6.8
-
1.1
'1.3
4.1
-
2.8
10/12
83%
Net Worth
10.2
2.0
2.1
+
-
1.5
6.3
+
1.1
1.2
+
1.6
3.3
9/12
75%
Quick Assets
1.1
+
1.2
+
+
1.3
1.1
1.6
+
2.8
rf
NA
1.5
6/12
50%
— Negative sign indicates negative value
2/
— A plus sign indicates an increase in value from three
  years to one year prior to bankruptcy

NA:  Not available
                                  11-18

-------
III.  PROCEDURES USED IN SELECTING FIRM SAMPLES AND CANDIDATE
      FINANCIAL RATIOS

A.  Sample Selection

     The Agency assembled an initial sample of 95 firms that filed under

either Chapter  10  or Chapter  11  of the  Federal  Bankruptcy Act between

1966  and  1979.   These  firms  were  identified in  two ways.   Previous

articles  on  bankruptcy  forecasting by Altman,  Haldeman  and  Narayanan

(1977) and Elam (1975) listed the bankrupt entities investigated in each

study.  The Agency  included  these  firms  in its sample if financial data

for the three-year  period preceding  bankruptcy could be located.  Addi-

tional bankrupt firms  were  identified through a  two-step  process.   The

Agency compiled a list of all  companies  that had ceased to be listed in

Moody's Industrial  Manual between 1968  and  1978 (excluding  all  firms

specifically  identified  as   having  undergone name   changes,  mergers,

etc.).  The  fate of  these  firms  was then determined  by  consulting the

company section  of  Funk  and Scott's Annual  Periodical Index.   Actual

bankruptcy filing dates for  firms  in  this sample  were identified either

from Funk and Scott or  the Bankruptcy heading of  the annual Wall Street

Journal Index.

     The Agency  then  compiled  financial data  for each sampled firm for

the  three years  immediately  prior  to  bankruptcy.    These   data  were

collected primarily from Moody's Industrial Manual and supplemented when

necessary by  the company Form 10-K  reports  submitted annually to  the

Securities and Exchange  Commission.   As a result of  serious data gaps,

three firms were dropped from the original list of 95.  Still later, the

Agency  decided  to  exclude   all  firms  primarily  engaged  in  wholesale,

                                III-l

-------
retail  and/or transportation service  activities  from the sample,  since




these are not likely to be TSDFs, further reducing  the final sample size




to 66 firms.




     The  average net  worth of  the 66  bankrupt  firms  sampled (in  the




third year  prior to bankruptcy) was $14.5  million; only one non-retail




firm had a  net  worth greater than $100 million.  The average asset size




of the  bankrupt  firm sample  was $46.5  million.  Thus, as shown in  Table




III-l,  the  Agency  obtained  a sample  with  a  higher  average  asset  size




than the samples in almost all  previous  studies  (Beaver, 1963; Altaian,




1968; Altman, et al. 1977; Deakin, 1972).




     The Agency  then assembled  a sample  of  190 non-bankrupt firms  iden-



tified  by the industry  index of Moody's  Industrial Manual as members of




industrial  categories  that  generate .and dispose of large quantities of




hazardous  waste  on-site  (e.g.,  primary  metals,  petroleum   refining,




chemical and  plastics  manufacturing,  textiles).    Data  were   collected



from Moody's. for these  firms  for the  three-year  period 1973-1975.  The




recession year of 1975  was  deliberately chosen to  evaluate  the effects




of business cycle  fluctuations on financial  ratio  performance.  Twelve



of these firms were  later eliminated because they were primarily engaged



in wholesale, retail, or transportation service activities.




     Of the 178  firms in the  final non-bankrupt sample,  slightly more




than half had net  worth (in 1973)  of  less  than $100 million.   For the




smaller firms,  the   averages  for net  worth and asset  size were  $33.6




million and $69.8  million,  respectively; for the  larger  firms,  $519.5



and $985.7  million,  respectively.







                                III-2

-------
                             TABLE III-l
             AVERAGE ASSET SIZE OF BANKRUPT FIRM SAMPLES

                        (In Millions of Dollars)
Sample
EPA

Altaian,
Deakin,
Altman,
Beaver,


et al. 1977
1972
1968
1966
Number of
Firms
66

53
79
31
79
Average
Asset Size
46.5
I/
96.0-
20.0
6.4
6.3
— Five non-bankrupt firms (Douglas Aircraft, Ling-Tempco-Vought,
 Lockheed Aircraft, Memorex Corp., and Mohawk Data Sciences) with
 very large assets were included in the bankrupt firm sample because
 the firms remained non-bankrupt only due to extraordinary external
 support.
                                III-3

-------
     Several bankruptcy  forecasting studies have utilized matched  bank-




rupt and  non-bankrupt firm samples.   Each firm  in the bankrupt sample




had a counterpart in  the non-bankrupt sample that had  a  relatively  equal




asset size  and  belonged to  the  same industry  category.  After careful




deliberation, the Agency decided that use of such a matched  sample  would




not satisfy the needs of  this  evaluation.   Although the sample of  bank-




rupt  firms  assembled  for  this study  is one of  the  most comprehensive




ever developed, it contains very few firms with greater  than $50 million




in net worth, because very few firms of this size  fail.   Several of  the




most highly publicized business failures of the last decade  (W.T. Grant,




Penn Central,  United Merchants & Manufacturers)  were  excluded from  the




sample because  those firms were primarily  involved in activities  other




than manufacturing,  and  this  reduced still further the number of  large




firms.    As  a  consequence,  matched  sample techniques  would  not  have



provided  information  about the effectiveness  of the various  ratio  tests




in accurately  classifying  the large,  viable  firms most  likely, to make




use of a  financial  test  provision.   The Agency did add  a  representative




sample of large  firms to  the non-bankrupt group,  in  order to verify



whether test options  were  effective over all size ranges.




     The  Agency also chose not  to  undertake a one-for-one  matching of



sampled firms by  industrial category.   In  order to assemble  a sample of




bankrupt  firms  of sufficient  size  to  provide  statistically  significant




results,  it was  necessary  to include  a   substantial number  of   firms



involved  in  manufacturing  activities not commonly  associated with haz-




ardous waste generation  (such  as  bakeries).   The Agency determined that






                                  III-4

-------
Che  improvements  in  classification  accuracy that  would  result  from




evaluating  test  alternatives against  a  set of firms  likely to seek  to




use  a  financial  test  greatly  outweighed  any  potential  inaccuracies




engendered  by  the use of  an unmatched sample.   To verify  this conclu-




sion,  the  Agency  performed  a  supplementary  comparison  of   industry




financial ratio averages, as described in Section VI of this Appendix.



     Many  commenters  on the  May  19,  1980  regulation  suggested  that




electric utilities and hazardous waste management firms possessed  unique




financial characteristics and  should be  provided with a separate  finan-




cial test.  To test  these  premises,  special samples of 26 utilities and




two representative large hazardous waste management firms were  examined.



These results are described in Section VI.




     The  fundamental limitation  of   this  analysis  with  respect to the



samples of  firms, particularly the  non-bankrupt  firm  sample,  is  uncer-




tainty with respect to how well they match  the financial characteristics




of firms that own or operate TSDFs.   In the absence of information about



the population which  the samples  are to  represent,  there is no  real way




to check for biases created by the methods used to assemble  the  samples.




     The  firms  in  the  non-bankrupt  sample were not  randomly  selected




from firms listed in Moody's.  Judgment was applied to ensure that firms



were chosen from  the specific industries considered most likely to own




or operate TSDFs and to ensure a greater representation of smaller firms



than a random  sample from  Moody's  would  have  produced.   Because the



sample was not randomly drawn, the possibility exists that the judgment
                                 III-5

-------
 employed may  have  inadvertently  resulted  in  the  selection of  firms




 exhibiting  unusually strong or weak financial  characteristics.  To check




 this  possibility,   the  Agency  compared the  mean  values  of the  non-




 bankrupt firm sample to  those of other samples used  in  past  studies of




 this kind.   (This  full  comparison is  given  in Section V.D.2.)   Where




 data on comparable  ratios  were available,  the average values of  ratios




 for the  non-bankrupt sample  in this study was either very  similar to




 that for other  studies  (within  10  to  20  percent), or fell between the




 values given by other studies  (average values  for cash flow  to total




 liabilities vary widely  among  samples  in  other studies).   As a further



.check for  possible  bias,  the  distribution  of  specific  ratios for the




 non-bankrupt  firm sample was compared to the  population  of  all firms in




 specific industries  reportd in the Expanded  Coverage section of the 1980



 Moody's  Industrial Manual  (Volume I) (see  Section VI.A of this  Appen-




 dix) .    This  comparison  showed that most industries  had from 5  to  10




 percent  more  firms meeting a given  ratio cutoff  point than  are found in




 the non-bankrupt firm sample for  1975.  This  would indicate that  either




 due to changes  in industry financial performance between  1975 and 1980,



 or  due to differences between the two samples,  the percentage of  viable




 firms  passing a  test could be higher than  that reported in this study.




 B.   Holdout Procedures



      In  an analysis  in  which numerous combinations of financial  ratios




 are tested  on a randomly drawn  set  of  firms, it is  important to  avoid



 the possibility  of search bias.   If  many combinations of  financial data




 are tested, the  test which is evaluated  as the most effective  may  attain
                                III-6

-------
this  result  because it  is  peculiarly sensitive  to  the precise  charac-




teristics  of the  sample being  analyzed.   To ensure  that  such search




biases  did not  influence  the final-conclusions of  this  analysis,  the




Agency  created  a "holdout" sample  comprised  of  firms randomly selected




by  a  computer  from  the  original  bankrupt  and  non-bankrupt  samples




(Frank,  Massy,  and  Morrison, 1965).   The  testing procedure  was  then




conducted  in two phases.  All tests were  evaluated initially against  the




136 non-bankrupt and 42 bankrupt firms which  remained in the  "primary"




sample.  Those tests which provided the best results were then  validated




against the 42 non-bankrupt and 24 bankrupt firms in the holdout  sample.



C.  Selection of Financial Variables to be Tested      :




     In  selecting  the financial ratios  to be tested  in this  analysis,




the Agency consulted  a broad spectrum  of opinion,  placing particular




weight  on  available  empirical results.   Candidate financial ratios were




drawn from three principal sources:   the comments  received  on the May




19,  1980  proposed   regulation,  the  surveyed  opinions  of  bond  rating




service  personnel  and credit  analysts,  and the  existing  literature  on




bankruptcy forecasting.  The  recommendations  from each of these  sources




are summarized below.




1.  Recommendations of Commenters




     Several commenters  proposed other financial ratios as better indi-




cators  of  firm  viability  than the  total liabilities/net  worth  ratio.




Alternative  ratios  most frequently mentioned involved cash  flow,  net




income, net  worth,  total liabilities,  debt and  equity.   Other measures




identified were  the  quick ratio,  estimates  of  return on  investment,
                                III-7

-------
total  fixed  assets,  retained  earnings,  and net  sales.   In addition,  a

number  of  comnenters  proposed  that  qualitative  measures of  a firm's

financial  position,  such  as  corporate  bond ratings  or  an affirmative

auditor's statement in the company's annual Form  10-K  report, be used  to

supplement or  to  substitute  for  a ratio test.  None of these commenters

provided quantitative  data  on the expected  classification accuracy  of

their  preferred  alternatives,  and only  a few suggested  what  an appro-

priate pass-fail  cutoff point  would be  for the ratios proposed.  Conse-

quently, these comments were used to  identify candidate ratios, but did

not play a significant role in the critical assessment process.

2.  Bond Rating and Credit Analysts

     Detailed  interviews  with senior  financial  analysts  from  the bond

rating agencies,  leading investment banking firms, and a number of major

urban  banks  conducted  for a  study  (Backer  and  Gosman,   1978)  for the

National  Association  of  Accountants  identified  the financial  ratios

those  analysts  considered  most  important  to  their  credit  decisions.

Those ratios are listed in Table III-2.  The same authors also conducted

an empirical study  of  the rate  of deterioration  in financial variables

for 57  firms whose  bond,  trade credit,  or bank loan ratings  had been

downgraded.  This analysis indicated that the ratios of cash flow /total

debt,  total  debt/total net worth, and  return on  sales showed  the most

significant  deterioration  in  the two years  prior to  downgrading,  and

hence  had  the highest correlation  with  actual  credit  actions.    The
      flow in this Appendix refers to the sum of net income, deprecia-
tion, depletion, and amortization.


                                III-8

-------
                             TABLE III-2

          FINANCIAL RATIOS CONSIDERED MOST SIGNIFICANT BY
                           CREDIT ANALYSTS
                    (based on interview results)


                              Short-Term    Intermediate Term  -Long-Term

                         Bank Loans   Trade    Bank Loans
Financial Ratios          Seasonal    Credit      Terra           Bonds

Accounts Receivable
  Turnover                   X
Cash Flow to Total
  Debt*                                             X              X
Current Liabilities to
  Tangible Net Worth                    X
Current Ratio                X          X
Fixed-Charge Coverage                                              X
Inventory to Working
  Capital                               X
Inventory Turnover           X
Long-Term Debt to
  Working Capital                       X
Quick Ratio                  X          X
Return on Sales                                     X
Total Debt to Tangible
  Net Worth*                 X          X           X              X
Working Capital to Sales                            X
Source:  Backer & Gosman (1978).
*We have substituted the word "debt" for "liabilities"
 because the great majority of interviewees indicated
 that they exclude Deferred Taxes and Minority
 Interest from liabilities, thereby approximating debt.
                              III-9

-------
analysis  also  revealed  that many  other ratios  not identified  in the

interview  process showed  much  greater deterioration  for  the sampled

firms than the ratios listed in Table III-2.

3.  Results of Prior Bankruptcy Forecasting Studies

     The Agency carefully reviewed the existing literature on bankruptcy

forecasting and identified  a large array of  financial  ratios  that have

previously been analyzed for their predictive  abilities.   These ratios

can  generally  be  organized  into  four basic  categories of  financial

information,  as described below.

a.  Profitability (Income or Cash Flow Based) Ratios

     Profitability  measures  reflect   the  current  response of  market

forces  to  the  products  manufactured  and/or  sold  by the  firm,  and may

indicate the ongoing  profit-making potential of an  operation.   Since a

firm's  ultimate existence relies  on  the earning  power of  its assets,

measures of  this  type  should  reflect  the  future stability of  a firm.

The  most  straightforward  measure  of  profitability  is   net  after-tax

income (NI).   Beaver  (1968)  tested the predictive power of  these ratios

of net  income  to sales,  total assets,  net worth, and total  debt, and

found the net income/total asset  (NI/TA)  ratio  to have the most discri-

minating power.   NI/TA has also  been  employed  in  a  number  of  other

studies (Deakin 1972, 1977;  McWilliams, 1977).  Beaver identified cutoff

points for this ratio ranging  from 0  to .04, depending on the number of

years prior to bankruptcy the test was applied.   McWilliams,  in deriving

qualitative descriptors  for  financial  ratio data, assigned  a  plus  to  a

firm with NI/TA greater than .04,  and  a minus to those with NI/TA below

0.
                                111-10

-------
     Altman  (1968,  1977) proposed  the use of  earnings  before interest




and taxes  CEBIT)  as a  substitute  for net  income.   Theoretically, EBIT




has  the   advantage   of  excluding  discontinuities  in  income   trends




resulting  from  one-time extraordinary charges.   However,  EBIT fails to




take into  account the  penalties  to profitability  suffered  by a  highly




leveraged  firm, as  the  result of interest  payments.   More importantly,




Altman has never  published results which  compare the actual  predictive




abilities  of EBIT versus NI.   Since  EBIT  values  are not always clearly




identified  in  income account statements  (particularly  in the  case of




smaller firms), EBIT was not included in the list of variables tested.




     Measures of  cash flow (CF) have been observed  in  past  studies to




provide forecasting ability superior to net income ratios.  The two most




popular formulations of cash flow ratios are cash flow/total liabilities




(in which CF = NI + depreciation), and cash flow/total debt (in which CF




= NI + depreciation - dividends, and  total debt  =  long-term debt + the




current portion of long-term debt).  Preliminary analysis indicated that




certain  key industries  (most significantly,  electric  utilities) per-




formed much worse on the  latter  of  these two  ratios.  Consequently, the




CF/TL ratio was  used  in the  preliminary  evaluation of  tests.    Beaver




used cutoff  points  for cash  flow/total  debt  of  .03  to  .11;  Me Williams




considered the cutoff points to be  .15. to .35.




b.  Liquidity Ratios




     Liquidity  ratios  define  the  capability  of  a  firm to  respond  to




short-term or emergency expenditure needs  without borrowing.   Since the




Agency may call  on firms  to close  their facilities immediately,  the
                                III-ll

-------
availability  of  liquid  assets  to  cover  these  costs  is particularly




relevant.   Four measures of liquid  assets  are commonly employed:   cur-




rent  assets (CA); quick assets (QA),  defined as  current assets minus




inventories; net  working capital  (NWK), defined as current assets minus




current  liabilities  (CL);  and cash  (including  marketable securities).




These measures  are commonly contrasted to either current  liabilities or




total assets (TA).




     Altman et al. in 1977 found the current ratio  (CA/CL, or CURRAT) to




be the most accurate predictive measure of firm stability, while in  1968




Altman used  the ratio NWK/TA  as  his  liquidity measure.   Beaver (1968)




obtained equivalent  predictive results using  both  measures,  as well as




the quick  ratio (QA/CL,  or  QRAT).   Deakin (1972)  used all seven vari-




ables (CURRAT, QRAT,  Cash/CL, CA/TA, NWK/TA, Cash/TA) in his 14-variable




multi-discriminant equation.  In his refined five-variable model (1977),




only CURRAT, CA/TA, and Cash/TA are used.




c.  Leverage Ratios




     Leverage ratios illustrate the  percentage of a firm's total assets




that  are  tied  to  short-term or   long-term  liabilities.    A  highly




leveraged  company may have substantial  liquid  or long-term  assets on




hand, but they may be secured by other creditors if the firm's financial




condition  begins  to  deteriorate.   There  are  several  alternate measures




of leverage.  Beaver (1968) looked at a number of ratios that contrasted




current  and  other liabilities  to  total  assets,  and found  total debt/




total assets  to be  the  most effective.   Altman (1968) found  that  the




ratio  of  the  market  value of  common  and preferred  stock  to  equity
                                111-12

-------
provided  better  predictive  results  than  other  debt/equity  measures,




although  the  relative performance  of the  tested ratios was  not docu-




mented.   Long-term  bond  analysts  cited  the total  liability (TL)/net




worth (NW) ratio  as  the  most  important  variable influencing their judg-




ment  (Backer  and Gosman,  1978).    The  Agency used  a TL/NW ratio as a




measure of a firm's leverage in its preliminary evaluations.  The  common




application in  the business  community  and ease  of  calculation  of this




ratio make it an appealing candidate for these analytical purposes.




d.  Turnover Ratios




     Both Beaver (1968)   and  Altman  (1968)  tested  a number  of   ratios




relating various  asset categories  to net  sales  figures.   Such turnover



ratios  are  commonly  used by  analysts  to measure  the sales-generating




capability of assets,  or to  indicate management's  effectiveness  in the




competitive marketplace.   Although these ratios have traditionally shown




very poor predictive power  when used  independently,  Altman,  in his 1968




Z-score  model,  included  a net sales (NS)/TA  ratio,  and rated  it the




second most important contributor.to the overall discriminating power of




his  function.    However,  research  by the  Agency revealed that  such a




ratio  severely  discriminates  against a  number  of  industry  categories




(e.g.,  electric  utilities;  firms  with  large  extractive  interests, such




as mining, gas,  oil and timber).




4.  Ratios Selected




     Seven financial ratios  were  selected  for  detailed  evaluation  as




candidates for inclusion  in a financial  test.   These ratios were chosen




from   the   many  identified   because  they   satisfied   three  primary







                                111-13

-------
conditions:   (1)  they had been  quantitatively  validated as significant




predictive  variables  by  previous  research  efforts;  (2) they  had been




frequently  identified by bond rating services and credit analysts  as key




parameters; and (3) their values could  be readily obtained from Consol-




idated  Income  Account or Consolidated  Balance  Sheet  data.   The seven




ratios selected were:




   Cash flow/total liabilities (CF/TL)




   Net income/total assets (NI/TA)




   Total liabilities/net worth (TL/NW)




   Current assets/current liabilities (CURRAT, or current ratio)



   Quick assets/current liabilities (QRAT, or quick ratio)




  . Net working capital/total assets (NWK/TA)




   Net sales/total assets (NS/TA)




D.  Summary Financial Data for the Firm Samples




     Financial data from the companies included in the bankrupt and non-




bankrupt  firm samples  are  summarized  in  Tables  III-3  through  III-6.




Table  III-3 summarizes  the  primary bankrupt  sample;  Table  III-4,  the




primary non-bankrupt  sample;  Table III-5,  the  holdout  bankrupt sample;



and Table  III-6,  the  holdout non-bankrupt  sample.   The  three  years of




data collected for each  sample  are hereafter referred  to  as  years x-1,



x-2, and  x-3.   For each  sample,  values  for  eight  financial  parameters




are detailed  in  the  tables:  net  income (NI), cash  flow (CF),  quick




assets (QA), current assets  (CA), total assets (TA), current liabilities




(CL),   total liabilities  (TL),   and  net  worth  (NW).    The  right-hand



portions of tne Tables for the primary  samples  (Tables  III-3  and III-4)







                                111-14

-------
indicate each firm's  performance  in the six major financial ratio cate-




gories used in this analysis:  cash flow/total liabilities, total liabi-




lities/net  worth,  net  fixed  assets/total  assets,  net  income/total




assets,  current  assets/current  liabilities,  and quick assets/current




liabilities.  Because the net income/total assets ratio




was  not  included  in any  of  the  tests  evaluated  against  the holdout




samples, this ratio is not included in Tables III-5 and  III-6.
                                111-15

-------
                                                       TABLE  I11-3
                           FINANCIAL  DATA  FOR FIRMS  IN  PRIMARY  BANKRUPT SAMPLE
Name ut
Plru
Alan Wood Steel


Ail led Leisure


American Beef
Packers

Anttiilciiii Bouk


Armac
Industries

Ast rciduta


Harrington
Industries

Hi shop
Industries

buwmar

Computer
Appl Icat Ions

Dlversa


Vear
I978(x-l)
(x-2)
(x-3)
1977(x-l)
(x-2)
(x-3)
1975(x-l)
(x-2)
(x-3)
1973U-1)
(x-2)
(x-3>
1976(x-l)
(x-2)
(x-3)
)970(x-l)
(x-2)
(x-3)
I9>6(x-l)
(x-2)
(x-3)
1970(x-l)
(x-2)
(x-3)
l»75(x-l)
(x-3)
I97()(x-l)
(x-2)
(x-3)
1969(x-l)
(x-2)
(x-3)
Nl
NA
-15.0
-9.4
-3.0
-0.2
-0.3
4.8
3.4
1.1
0
-2.4
0.1
NA
-6.8
2.4
0.2
-6.7
-5.9
NA
0.3
0.1
-1.4
-7.6
0.2
-25.6
2.0
-2.1
0.6
0.7
6.7
-7.3
-7.4
CK
NA
-1.2
-1.4
-2.8
-0.1
-0.2
7.1
5.3
2.3
1.2
-1.2
1.3
NA
NA
3.0
0.7
-6.4
0
NA
NA
NA
-3.0
-7.3
0.4
-24. 1
2.4
-1.4
1.2
1.6
6.9
-6.4
-6.8
QA
NA
18.7
18.4
0.3
1.2
1.3
47.7
49.6
31.3
b.a
11.0
13.6
NA
13.2
10.7
10.7
11.4
9.5
NA
3.6
3.6
4.7
4.5
12.2
20.9
10.7
14.4
14.2
9.9
0.7
17.1
6.0
CA
NA
11.7
34.1
2.2
2.3
3.2
78.2
82.0
40.4
9.'J
12.9
15.5
NA
29.5
30.5
10.6
11.7
16.7
NA
5.9
5.6
8.7
11.1
17.7
41.9
18.2
17.4
17.2
11.3
0.7
22.2
13.7
U.
NA
41.4
29.2
4.4
2.0
1.4
72.2
74.8
35.4
15.5
6.3
6.9
NA
24.0
17.7
5.8
18.0
17.4
NA
5.6
6.9
4.5
10.2
9.1
20.4
10.5
13.3
17.1
7.6
7.6
12.5
27.7
Tl.
NA
H8.0
79.0
4.4
2.0
1.4
97.4
90.3
45.7
12.2
17.5
16.5
NA
27.3
22.3
11.1
23.3
21.7
NA
7.7
9.6
14.0
16.3
15,2
51.5
13.4
33.2
23.0
14.6
39.3
73.1
80.3
NW
NA
32.0
47.0
-1.3
2.1
2.3
22.6
17.7
14.3
6.6
6.6
12.8
NA
5.9
11.7
2.3
-8.7
-1.0
HA
1.2
2.5
2.3
2.2
7.0
3.5
7.1
1.0
11.2
7.2
6.6
3.0
10.4
TA
NA
120.0
126.0
•J.I
4.0
3.7
120.0
108.0
60.0
18.8
24.1
29.3
NA
33.2
34.0
13.4
14.6
20.7
NA
8.9
12.1
16.3
18.5
22.2
55.0
20.5
34.3
34.2
21.8
46.4
76.1
90.7
NFA/TA-'
NA
P
P
r
f
f
?
23.6
P
P
P
P
NA
V
f
f
f
F
NA
F
F
F
F
F

F
'
F
23.6
25.2
1 f
f
T
CF/TIJ?'
NA
F
F
F
F
F
F
F
F
F
F
F
NA
F
13.5
F
V
f
NA
F
F
K
F
F
F
17.9
F
F
11.0
17.1
F
F
•tUmi-1
•NA
F
1.7
P
P
P
F
F
F
l.B
F
1.3
NA
F
1.9
F
F
F
NA
F
F
F
F
F
F
1.9
F
F
F
F
F
F
CURKAT*'
NA
F
F
F
F
P
F
F
F
F
P
P
NA
F
1.7
1.8
F
F
MA
V
F
L.9
F
1.9
P
1.7
F
F
F
F
F
K
QKAT-'
MA
F
F
F
F
F
f
f
F
F
P
P
NA
F
F
P
F
F
NA
F
F
1.0
F
P
1.0
1.0
1 . 1
F
P
F
F
F
NI/TA5'
NA
F
F
F
F
P
4.0
3.1
F
F
F
F
NA
F
P
F
F
F
NA
3.4
F
F
F
F
F
P
F
F
3.2
P
F
F
-'p  > 30.0 (passes all  tests); F < 20.0 (falls all  tests)
- t  > 20.0 (passes all  tests); F < IO.O (tails all  tests)
- P  <  1.0 (passes all  tests); F >  2.0 (falls all  tests)
4/
- P  >  2.0 (pusses all  tests); F <  1.5 (falls all  tests)
- P  >  1.2 (passes all  tents); F <  l.O (falls all  tests)
- P  >  4.0 (passes all  teats); F •  2.0 (falls ull  tests)

-------
                                                          TABLE III-3
                                                          (continued)
M
M
M
I
Niiou: of
Firm
Uully Madison


litigro
Corporation

Fotot:hromG
Inilust ries

General Alloys





Cray
Manufacturing

On.un
Industries

Harvard
Industries

King Resources


Maule
Industries

Michigan
Bakeries

Mllo
Electronics

National Radio

Year
1970(x-l)
(x-2)
(x-3)
1973(x-l)
(x-2)
(x-3)
1970(x-l)
(x-2)
(x-3)
I973(x-l)

-------
                                                       TABLE III-3

                                                       (continued)
M
M
i
i—•
do

Firm
National Video


Omega Alpha


Patterson
Purcluuent

Pernmnuer
Company

Photon


Pularad
Electruulcs

Heading
Industries


Corpuratlo n

Kemco Industries


Scottux


Sequuy.ih
Induutl liia

Sheffield
Watch

Stellar
Indust rles


Year
1969(x-l)
(x-2)
(x-3)
I971(x-l)
(x-2)
(x-3)
1974(x-l)
(x-2)
(x-3)
I976(x-l)
(x-2)
(x-3>
1972(x-l)
(x-2)
(x-3)
1970(x-l)
(x-2)
(x-3)
1976U-1)
9x-2)
(x-3)

(x-2)
(x-3)
197I(x-l)
(x-2)
(x-3)
1973(x-l)
(x-2)
(x-3)
1971U-I)

(x-3)
1971(x-l)
(x-2)
(x-3)
1975(x-l)
(x-2)
(x-3)

NI
-5.3
2.2
7.3
41.7
0.6
2.0
-2.2
-0.8
-1.3
-15.7
-17.2
-3.0
1.1
-1.0
1.0
0.4
-1.5
0.2
-2.1
-3.3
l.l

-4.8
-0.1
-10.8
-0.7
2.0
-2.4
1.1
0.8
-0.6
-9.5
-10.4
HA
0.7
0.4
-1.7
-4.0
-1.7

CF
-2.1
5.8
9.3
47.4
2.4
6.7
0
0
-0.5
-12.2
-13.6
-0.4
1.8
-0.6
1.2
0.6
-1.4
0.4
-1.2
-2.4
2.0

-4.6
0
-8.8
1-0
3.5
0
0
1.1
0.3
-8.3
-9,2
NA
0.8
0.7
-1.3
-3.5
-1.3

QA
5.7
7.8
16.1
27.9
54.4
39.1
4.4
5.0
6.1
7.8
14.1
18. 5
16.0
14.0
17.4
1.7
2.3
2.0
1.2
4.9
8.8

2.9
2.3
10.0
11.4
10.7
10.1
6.0
3.8
4.7
1.5
1S.O
NA
9.6
S.6
2.4
2.0
3.5

LA
12.7
13.3
23.6
51). 8
95.9
74.5
10.3
11.2
12.1
19.0
36.1
42.5
34.9
31.3
32.4
4.0
5.4
5.1
9.2
11.1
16.6

11.0
4.3
14.0
17.8
13.6
19.3
13.8
7.7
13.3
12.5
29.1
NA
21.6
15.6
3.8
3.9
7.0

CL
5.2
3.8
14.1
40.5
65.5
41.9
5.4
4.6
5.0
13.9
19.1
14.7
8.7
6.2
5.6
1.5
4.5
2.8
H.2
3.6
5.8

8.0
1.7
11.0
13.7
6.8
12.1
7.5
6.7
8.4
12.9
24.3
NA
12.2
9.3
4.5
4.7
6.6

Tl.
8.2
4.5
14.9
145.9
156.5
111.8
11.2
10.2
U.I
67.2
74.6
62.0
23.0
20. 5
21.9
1.6
4.6
2.9 .
16.5
16.9
20.1

a. 7
2.8
16.3
16.9
9.4
22.7
13.4
9.9
29. 0
27.0
39.7
HA
17. 2
10.7
7.9
10.4
10.8

NV
19.3
24.5
23.8
33.1
31.5
63.2
7.3
9.5
10.6
-11.6
4.1
21.3
17.5
15.1
15.7
1.5
1.8
3.1
1.8
5.9
9.2

3.9
1.7
1.1
8.1
8.9
8.0
10.4
2.9
2.8
3.8
15.5
NA
5.3
5.6
0.1
-0.8
2.7

TA
27.5
29.0
38.7
179.0
188.0
175.0
18.5
19.7
21.7
55.6
78.7
83.3
40.5
35.6
37.6
5.1
6.4
6.0
20.3
22.8
29.1

12.6
6.5
17.6
25.0
18.3
30.7
21.8
12.8
31.8
30.8
55.2
NA
22.5
16.3
8.0
9.6
13.5

NFA/TAi'
P
t
r
P
P
P
P
P
P
P
P
P
F
F
F
F
F
F
P
P
P

F
F
F
24.3
22.5






NA
F
F
P
P
28.6

CF/TI^'






F
F
F
F
F
F '
F
F
F
16.7
F
13.8
F
F
F

F
F
F
F
P
F
F
U.I
F
F
F
NA
F
F
F
F
F

Tt/NU?'
P
P
P
F
F
1.8
1.5
1.1
1.0
F
F
F
1.1
1.4
1.4
F
F
P
F
F
F

F
P
F
F
1.1
F
1.3
F
F
F
F
NA
F
1.9
F
F
F

CIJHRAT*-'
V
e
1.7
F
F
1.8
1.9
P
P
F
1.9
P
P
P
P
1.6
F
1.8
F
P
P

F
P
F
e
?
p
1.8
f
1.6
F
R
NA
l.B
1.7
F
F
F

(JRAT-5'
1.1
P
1.1
F
F
F
F
P
P
F
P
P
P
P
P
F
F
F
F
P
P

F
P
F
F
P
f
F
F
F
F
F
NA
F
F
F
F
F

Ml/TA^
F
P
P
P
F
F
F
F
t
f
F
F
1.2
F
2.7
P
F
1.3
F
F
1.8

F
F
F
F
P
F
P
P
F
F
F
NA
1.1
2.5
F
F
F

-------
TABLE III-3
(concluded)
Name of
Firm
Technical
Measurements

Trana- Beacon


Walt ham
Industries .

Wustgate
California

Year
1968(x-l)
(x-2)
(x-3)
1970U-1)
(x-2)
(x-3)
1971(x-l)
(x-2)
(x-3)
1972U-1)
(x-2)
(x-3)
Nl
1.9
-2.1
-1.8
NA
0.2
0.3
-13.9
0.3
1.6
-2.6
2.8
3.6
CF
2. B
-i.a
-1.5
MA
0.3
0.1
0
1.6
2.4
2.7
7.2
6.8
QA
6.7
2.6
4.9
NA
1.5
1.3
8.9
13.8 .
9.6
22.8
46.3
19.6
CA
10.0
5.0
9.1
NA
1.6
1.4
27.0
34.7
18.6
41.3
63.8
32.1
CL
2.7
4.5
5.6
NA
1.5
1.2
30.8
21.9
8.1
34.0
76.9
32.2
TL
3.1
6.0
7.5
NA
2.6
3.7
43.9
47.9
21.1
143.6
182.0
116.7
NU
16.5
0.5
5.5
NA
0.7
-0.8
1.1
15.3
11.2
30.4
33.0
34.3
TA
19.6
6.5
13.0
NA
3.3
2.9
45.0
63.2
32.3
174.0
215.0
151.0
NFA/TAr-
P
F
f
HA
25.2
P
26.8
28.1
29.2
P
P
P
CF/TL-7
P
F
F
NA
F
F
F
F
11.4
F
F
F
TL/NW^
P
F
1.3
NA
F
F
F
F
1.9
F
F
F
4/
CUKKAT-
P
F
1.6
HA
F
K
F-
1.6
P
F
K
F
QKA^'
P
F
F
NA
1.0
1.1
F
F
1.2
F
F
F
Nl/TA^'
P
F
F
NA
P
P
F
F
P
F
F
2.4

-------
                                                                             TABLE  III-A
                                           FINANCIAL DATA  FOR FJRMS  IN  THE PRIMARY NON BANKRUPT SAMPLE
I
N>
o
                        -'for all  firms,  x-l In 1975, x-2  Is 1974. and x-3  Is 1973
                        -'p  • 30.0 (passes all tests); F • 20.0 (falls all  tests)
                        -'P  ' 2U.O (passes all I oil:;); F < 10.0 (falls all  tests)
                        - P  < 1.0  (p:isse« all testa); f >  2.0 (falls nil test::)
                        - P  > 2.0  (passes all teals); F '  1.5 (falls all li-.sls)
                        -'e  '' 1.2  (pistes all lusts); F •  1.0 (falls all tests!
                        -'p  •' 4.0  (P..SSUU all tvato); F -'  2.0 ((alia all tests)
N.UQ.- l.f
Firm
Alc-.ii


Al lugliuny 1. ml tun


All mail


A 1 lima x


Ani.'tx


Ainu r on


AiuuLi:k

-
Ampeo-PUiiiburgli


/Vastcd


Ankuu


Artncn SlUiil


ASAKCO


Y..ri'
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-1)
HI
65
173
104
3D
45
31
1
.7
3
21
36
16
134
148
105
8
7
6
13
11
9
6
9
5
35
17
15
.7
.6
1
117
204
107
26
126
113
CK
235
338
263
50
64
49
2
2
3
23
36
16
193
194
142
14
12
11
16
15
13
H
10
6
45
26
24
1
1
2
209
290
190
62
161
140
'I*
400
474
453
119
153
135
8
13
9
117
99
81
451
261
417
41
48
35
43
32
10
14
16
13
98
71
63
4
4
4
438
601
438
195
182
171
CA
1(121
1040
905
325
379
320
16
26
21
217
179
139
677
404
535
70
87
60
85
78
70
3 8
43
35
157
129
117
7
7
7
1001
1124
851
442
408
364
r.i.
37/
472
308
141
184
146
3
12
8
54
1,1)
51
333
207
209
29
39
25
35
36
29
12
17
13
75
61
50
2
3
3
415
554
172
174
234
196
Tl.
1845
1651
1412
318
352
317
4
16
12
It) I
118
71
1116
826
812
67
82
56
54
57
48
15
22
18
79
65
61
2
3
3
1276
1268
1114
641
466
375
MM
1575
1547
1409
320
304
270
17
15
15
279
259
201
1364
942
840
68
62
58
67
59
53
50
45
38
179
153
144
10
9
9
1330
1274
1145
86)
863
774
TA
3420
3198
2821
63tt
656
588
21
31
27
462
377
272
2480
1768
1652
135
144
114
121
117
102
65
67
56
258
218
205
12
12
12
2606
2542
2259
1502
1329
1149
NFA/TA-'
P
P
V
V
r
V
22.3
F
20.9
f
F
f
P
P
P
P
P
P
29.6
P
29.2
25.8
20.7
21.4
P
P
P
P
P
P
P
P
P
P
P
P
CF/TI.-'
12. H
P
18.6
15. B
18.2
15.5
P
10.3
P
12.5
P
P
17.3
P
17.5
P
14.9
18.6
1'
P
P
P
P
P
Y
Y
Y
P
P
P
16.. 3
P
17.1
F
P
P
•ri./Nu^'
1.17
1.06
1.00
P
1.16
1.17
P
1.01
P
P
P
P
P
P
P
P
1.32
P
P
P
P
P
P
t
. t
t
P '
P
P
P
P
P
P
P
P
P
CUKKA r -
t
p
P
f
P
r
p
p
p
p
p
p
p
1.95
p
p
p
p
p
p
p
p
• p
p
1.54
p
p
p
p
p
p
p
p
p
1.74
1.86
ns&t-'
1.06
1.00
H
F
f
F
P
1.05
I. 10
P
P
P
P
P
P
P
P
P
P
f
1.03
P
1.04
1.02
P
1.16
P
V
P
P
1.06
1.08
1.18
1.12
K
F
NI/TA-'
F
P
3.7
P
P
P
P
2.3
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
F
P
P

-------
                                                      TABLE  I11-4

                                                      (continued)
I
ro
N;itnu of
Firm
ASPRO


Alhlone


A.I ft 1 c


Bernz-0-Matlc


Bethlehem Steal


Bliss & Laughlln


Brooks & Perkins


Buntly Corporal ion


Carpenter


Ceco


Cerro


i:FU


Cleveland-Cliffs


Yoari'
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-I)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
NI
2
2
2
12
13
7
3
4
2
-2
.8
2
21,2
342
207
10
9
8
1
.5
.5
4
.3
8
22
17
14
9
11
10
-2
16
15
38
35
14
31
26
21
r.r
4
3
3
15
16
10
5
6
5
-1
1
2
476
553
403
14
14
12
1
.8
.a
8
4
II
23
24
20
14
17
15
11
24
22
64
60
33
34
26
21
QA
13
11
10
43
50
32
10
14
11
4
7
6
768
1172
934
37
24
29
5
4
3
19
19
21
47
49
40
6(1
50
46
208
247
185
85
79
62
58
72
70
CA
23
19
18
85
96
77
25
27
24
12
14
14
1388
1682
1376
64
56
56
11
8
7
29
29
33
113
117
100
98
100
92
311
318
253
127
115
92
75
81
78
Cl.
12
9
a
32
45
30
8
9
a
4
5
6
719
1032
713
14
19
16
4
4
1
10
11
U
34
40
28
26
31
31
147
154
80
71
46
53
)2
30
18
Tl.
23
21
18
109
122
112
22
21
23
IU
11
9
1180
2023
1677
52
50
56
7
5
3
21
23
25
55
65
55
58
63
65
380
396
298
127
107
117
68
56
33
NU
15
13
12
50
48
40
27
25
22
8
9
9
2612
2490
2242
68
62
57
6
5
6
50
49
52
139
124
112
92
87
80
341
369
362
Ib2
181
168
216
203
185
TA
38
34
30
159
170
152
49
46
45
18
20
17
4592
4513
3919
120
112
113
13
10
8
71
72
78
194
189
167
150
150
145
/21
765
660
309
288
285
284
259
218
?/
NFA/TA-'
t
t
t
27.5
24.1
25.7
P
P
P
P
23.5
F
P
P
P
21.9
24.1
23.6
F
F
20.2
P
P
P
P
P
P
P
30.0
P
NA
F
F
P
P
P
F
F
F
CF/T1.2'
16.3
13.7
15.6
13.7
13.3
?
P
P
P
t
11.7
P
P
P
P
P
P
P
18.6
17.4
P
P
16.2
P
P
P
P
P
P
P
F
11.1
K
P
V
P
P
P
P
TI./NW-'
1.53
1.55
1.48
F
F
F
P
P
1.02
1.30
1.19
1.01
P
P
P
P
P
P
1.17
P
P
P
P
P
P
P
P
V
P
P
P
1.07
P
P
P
P
P
P
P
ClIHKAT-
1.95
P
P
P
P
P
P
P
"'
P
P
p
1.93
1.63
1.93
P
P
P
P
1.83
P
P
P
P
P
P
P
p
P
i>
P
I'
P
1.78
P
1.72
P
P
P
•IKA-i-fc'
1.06
P
P
P
1.11
1.07
P
P
P
1.05
P
1.11
1.07
1.14
P
P
P
P
1.15
F
P
r
p
p
p
p
p
p
p
p
p
p
p
1.20
p
1.15
P
P
P
NI/'IA?'
P
P
P
P
P
P
P
P
P
F
3.9
P
P
P
P
P
P
P
P
P
P
P
F
P
P
P
P
P
P
P
P
2.1
2.3
P
P
P
P
P
P

-------
                                                     TABLE  III.-4
                                                     (continued)
I
Ni
Hi) nit* of
Kinu
Coiiunerc J n 1 Mt-ta 1 s


Conayne


Copper Riinge


Co|>|i*-rwuld

1
t
Crown Industries

•
1 Cyclops

j
Driver Harris
.
.
: Earth Resources
I

tlASLO


Ea^turn


tl.'IKA


Florlilu Steel


Cultural Steel


rui.pl'
(x-l)
(x-2)
(»-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)

-------
                                                       TABLE  ITI-4

                                                       (continued)
i
NJ
u>
Naint: of
Firm
lUiimu


liar sco


llofitiaim
Indust r les

Houvur Ba 1 1
& Bt:arlnu

IMCO


Inland Steel


Insp i rat Ion
Consul Idated

Interl jke


Knistr Aluminum
& Chemical

Kaiser Industries





K.ity Industries


Kiiweckl-Burylco


vcar!'
(x-l)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x 3)
(x-1)
(x02)
(x-3)
(x-l)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)

(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
NI
47
21
23
36
25
22
.7
3
1
12
11
14
166
57
26
83
148
83
-4
10
15
34
39
17
95
104
45
79
31
45

67
53
10
10
12
-1
7
5
CF
57
30
33
59
48
42
2
3
2
21
17
21
199
79
41
164
222
154
7
20
21
54
55
32
149
157
95
213
91
97

111
96
15
15
17
3
10
a
QA
79
75
60
135
98
91
7
8
8
60
39
33
267
267
183
309
323
250
28
18
36
98
124
87
354
401
312
212
219
219

143
144
35
35
38
22
21
21
CA
100
97
81
224
194
162
18
16
16
87
72
58
469
346
260
627
664
515
60
39
47
216
250
164
825
799
618
305
334
313

244
224
75
75
70
74
72
61
Cl.
49
49
31
59
90
64
12
9
11
37
31
27
284
150
128
243
302
243
13
17
15
107
155
66
446
452
394
189
202
191

134
135
34
34
40
18
14
15
Tl.
92
114
86
152
137
107
16
14
16
69
70
59
630
435
305
896
819
710
54
65
68
216
254
148
1304
855
754
657
692
688

394
422
98
98
97
64
57
40
NW
293
283
274
206
182
164
14
13
11
79
73
62
458
309
250
9/1
939
849
115
91
88
264
242
214
798
1202
1060
632
577
550

395
336
68
68
60
53
55
50
TA
385
397
360
358
319
271
30
27
27
149 1
143
122 |
1088 1
744
555 1
1867
1758
1559
169
156
156
480
496
362
2102
2057
1814
1289
1269
1238

789
758
167
167
156
116
112
90
IIFA/TA-
F
F
F
P
P
P
P
P
P
P
P
f
P
P
'
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P

P
P
20.9
20.9
20.8
P
29.0
26.6
CF/TI.~
P
P
P
P
P
P
12.5
P
11.9
P
P
P
P
18.2
13.4
18.3
P
P
11.3
P
P
P
P
P
11.4
18.4
12.7
P
13.1
14.1

P
P
15.4
15.4
17.8
f
17.3
19.3
TL/NU^'





P
1.16
1.04
1.43
P
P
P
1.38
1.41
1.22
P
P
P
P
P
P
P
1.05
P
1.63
P
P
1.04
1.20
1.25

P
1.26
1.44
1.44
1.64
1.21
1.64
1.03
CURRAT-
P
1.96
P
P
P
P
1.5B
1.89
I. 56
P
P
P
1.65
P
P
P
P
P
P
P
P
P
1.61
r
.85
.77
.57
.61
.65
.64

.83
.66
P
P
1.74
P
P
P
QKA'I-
,.
P
P
P
1.09
P
F
F
F
P
P
P
F
P
P
P
1.07
1.03
P
1.07
P
F
F
P
F
F
F
1.12
1.08
1.15

1.07
1.07
1.02
1.02
F
P
P
P
NI/TA-7
P
P
P
P
P
P
2.3
P
P
P
P
P
P
P
P
P
P '
P
F
P
P
P
P
P
P
P
2.5
P
2.4
3.6





K
F
P
P

-------
                                                       TABLE  II1-4
                                                       (continued)
M
Is)
N^mu ut
Firm
Kumiccot t:


Keybtont?
Consul Idatttil

Kerr-HcCue


Ki-wanee


Kluibi-Tly-Cliirk


Kin-Ark


Koppurs


Kraft co


l.tjliigh Valley
Intlusl rlus

Lilly & Co.


I.TV Corporation


l.ubrlzol


likens SLbul


v I/
Year-
(x-l)
(x-2)
(x-3)
(x-1)
(x-2)
(x-1)
(x-1)
(x-2)
(x-3
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(«-l)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(xU3)
(x-1)
(x-2)
(x-3)
(x-l)
(x-2)
(x-1)
(x-1)
(x-2)
(x-3)
N[
49
211
1S9
11
10
5
131
116
63
28
29
17
103
95
82
.9
1
1
60
48
30
140
95
103
2
2
3
101
179
156
13
112
50
47
51
17
11
11
7
CF
71
302
240
20
15
12
190
187
120
43
42
29
150
143
124
2
2
2
91
Bl
SB
192
144
151
4
4
4
212
207
182
70
197
130
59
61
46
12
HA
HA
QA
171
317
331
41
39
17
338
298
200
101
66
49
237
249
228
4
4
3
214
193
1S1
318
337
295
8
a
a
611
553
450
367
645
498
116
112
101
31
39
31
CA
446
606
550
135
124
102
567
496
333
1H1
116
86
493
486
402
7
7
4
369
339
255
1024
1110
854
32
32
32
947
850
667
848
1097
871
193
191
138
60
73
58
1:1.
IB3
287
241
41
55
36
285
294
128
63
42
30
219
199
169
4
6
7
150
144
107
466
582
411
IB
IB
19
',16
390
268
585
647
515
64
80
49
27
16
25
Tl.
812
767
670
128
122
108
579
510
108
174
74
60
575
562
1.58
15
17
16
311
336
246
6 SB
860
583
53
53
55
476
417
290
1602
1686
1552
74
89
55
54
66
55
NM
1411
1442
1107
112
103
95
809
655
559
183
170
147
730
668
610
12
11
1O
368
312
274
1009
850
BOB
6
6
4
958
849
730
361
344
233
231
203
170
109
102
94
TA
2223
2209
1295
240
22S
202
I38B
1164
867
357
244
207
1305
1229
1068
27
27
25
680
648
520
1667
1710
1391
59
59
59
1434
1265
1020
1963
2031
1785
3IIS
292
22S
161
168
149
NFA/TA-
P
P
P
P
P
P
P
P
t
t
P
P
P
P
P
P
P
P
P
P
P
P
P
P
23.7
23.8
23.1
P
29.3
P
P
P
P
P
P
P
P
P
P
CF/Tl.-7
F
P
P
15.9
12.2
10. B
P
P
P
P
P
P
P
P
P
14.6
13.9
14.2
P
P
P
P
16.7
P
F
f
f
t
r
r
r
11.7
t
t
r
r
p
15.0
15.0
L 1
TI./HW-
P
P
P
1.14
1. IB
1.14
P
P
P
P
P
P
P
P
P
1.25
1.54
1.60
P
1.08
P
P
1.01
P
F
F
F
P
P
P
F
F
F
P
P
P
P
P
P
UWKAT-
P
P
P
P
P
P
1.99
1.69
P
P
P
P
P
P
P
1.61
F
F
P
P
P
P
1.94
P
l.Bl
1.81
1.74
P
P
P
F
1.69
1.69
P
P
P
P
P
P
QUA-!*'
F
1. 10
t
1.01
F
1.01
1.19
1.01
P
V
f
I1
I.OB
P
!•
1.08
F
F
P
P
P
F
F
F
F
F
F
P
P
P
F
f
F
P
P
P
P
1.09
P
HI/TA-'
F
P
P
P
P
2.4
P
P
P
P
P
P
P
P
P
3.3
P
P
P
P
P
P
>•
P
3.9
3.9
P
P
P
P
F
P
2.8
P
P
P
P
P
P

-------
                                                           TABLE  I.I 1-4

                                                           (continued)
i
NJ
in
Name of
Kiroi
MarAildrtrws
Forbus

M.mgood
Corporut Ion

Marathon
Manufacturing

Martin
Marietta

Mclntouh
Corporal ion

Mcl.onll!


Me He 11
Corporation

Mead


Merck


Mlirhluan Seamless
Tub.:

MiUlaiid-Koss


Miles
Laboratories

JM


Ycari'
(x-l>
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-1)
(x-1)
(x-2)
(x-3)
(x-1)
(x-2)
(x-1)
NI
3
2
6
.2
.1
.2
14
7
-19
55
81
57
1
4
4
6
22
16
4
5
6
51
82
50
229
211
183
10
7
4
21
21
12
15
16
17
262
102
296
CF
6
5
a
.5
F
F
22
13
-12
115
139
109
3
5
5
23
41
15
6
7
8
94
120
82
281
257
223
11
P
P
11
12
21
19
26
26
417
411
405
Q.A
7
8
10
4
6
5
42
42
52
207
215
225
10
10
10
55
75
68
11
37
30
245
219
190
447
312
135
14
15
1!)
140
•J9
76
ai
60
74
795
675
741
CA
28
26
27
1
9
7
126
lit)
106
393
192
371
15
18
15
136
110
118
95
92
77
171
181
136
857
721
580
12
29
26
209
186
141
144
149
124
1589
1577
1110
Cl.
2
10
12
4
5
3
64
68
66
161
169
145
5
9
5
79
65
55
16
15
27
192
211
152
155
161
238
15
10
9
104
89
57
77
105
80
620
921
565
Tl.
25
21
27
a
9
6
112
160
117
510
518
569
7
11
9
154
144
138
61
55
40
560
552
486
624
420
279
21
27
19
169
157
110
200
198
171
1194
1146
768
NU
11
29
28
4
4
5
81
68
6)
609
584
501
21
22
20
175
177
162
7)
69
66
512
509
459
950
821
710
42
14
29
182
167
152
146
.117
128
1821
1695
1511
TA
56
52
54
12
13
11
213
228
198
1139
1121
1072
10
36
28
129
321
299
131
124
106
1092
1061
945
1574
1241
989
65
60
48
151
124
282
146
315
299
3017
2841
2281
NFA/TA-'
V
t
t
f
V
V
r
p
p
p
p
p
p
p
p
p
p
p
25.1
23.2
23.9
P
P
P
V
V
P
P
P
!•
26.1
27.6
P
P
P
P
P
P
P
CK/T1.-'
P
P
P
F
F
F
16.4
F
F
P
P
19.2
P
P
P
14.8
P
P
10.5
13.0
P
16.8
P
16.8
P
P
P
P
P
P
18.5
P
15. B
13.1
13.2
15.3
P
P
P
TlW
P
P
t
1.86
F
1.11
1.63
F
F
P
P
1.13
P
P
P
P
P
P
P
P
P
1.05
1.09
1.06
P
P
P
P
P
P
P
P
P
1.37
1.45
1.34
P
P
P
C11RKAT-
P
P
P
P
1.84
P
1.97
P
1.61
P
P
P
P
P
P
1.70
P
P
P
P
P
1.91
1.79
P
P
1.99
P
P
P
P
P
P
P
1.88
F
1.55
P
1.71
P
QKAT-'
F
F
F
1.20
P
P
F
F
F
P
P
P
P
1.18
P
F
1.15
P
F
1.04
I. 11
P
1.01
P
p
F
p
F
P
P
P
1.11
P
I.Ob
F
K
P
F
P
NI/TA-"
P
P
P
F
F
F
P
3.1
F
P
P
P
P
P
P
P
P
P
2.7
1.7
P
P
P
P
P
P
P
P
P
P
P
P
P
p
P
P
p
P
P

-------
TABLE III-4
(continued)
Nuoiu of
firm
N t r i-o
Aluminum

Mobl 1 Oil


Mulyci>ri>


Morton Norwich


Nalco Chemical


Mat lonal
Clieoiuuarch

H.i I ional
Ultulllcrs

Nul ional
Industries

N.I 1 1 mi j)
SurvJce

National
Sturch & Chum.

National Steel


Nc-w £11^1 anij
tine 1 u.i r

N4:wim>nt Mining


Y.arl'
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3>
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
U-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
NI
3
3
4
BIO
1047
849
8
14
8
18
25
24
32
27
25
17
14
10
62
90
72
15
17
14
18
23
23
19
18
17
58
176
98
3
3
2
53
114
103
CF
5
5
6
1522
1560
1282
16
22
14
35
41
39
44
36
32
19
14
10
89
117
101
22
HA
NA
29
33
32
25
25
23
17k
286
207
4
4
3
85
140
123
QA
26
17
16
3954
3861
2986
10
12
14
107
105
96
77
58
71
64
50
42
328
315
34)
93
100
66
87
90
87
58
42
16
318
578
42J
8
II
9
129
123
111)
CA
51
49
43
6156
5827
3939
30
32
30
201
191
164
118
100
100
86
67
52
629
641
629
IB2
197
154
139
161
138
100
89
70
763
857
681
11
13
11
298
266
186
Cl.
10
9
18
5234
5206
3375
a
15
13
59
89
61
47
41
30
34
21
15
163
181
174
110
143
101
45
74
55
32
33
28
503
581
431
3
3
3
126
124
72
TL
28
27
22
8209
7638
4976
42
50
42
222
208
186
58
49
35
43
29
20
429
454
504
189
219
182
74
104
85
84
69
47
1201
1096
958
4
4
3
481
440
379
MU
43
42
40
6841
6436
5715
92
77
75
238
231
217
158
138
122
74
59
48
595
565
505
92
86
81
155
148
140
130
116
102
1209
1192
1067
19
16
13
648
637
564
TA
71
69
62
15050
14074
10690
134
127
117
460
439
403
216
187
157
117
88
68
1024
1019
1009
281
305
261
229
252
225
214
185
149
2410
2289
2024
23
20
16
1129
1077
944
NKA/TA?'
27.8
29.0
P
r
P
f
f
f
r
f
?
r
p
V
27.4
20.0
p
If
p
V
f
f
f
21.2
p
r
p
p
p
p
p
p
p
p
p
28.8
p
p
p
CF/TL-'
19.2
19. 1
P
18.5
P
P
P
P
P
15.7
19.7
P
P
P
P
P
P
P
P
P
P
11.7
10.0
10.0
P
P
P
P
P
P
14.2
P
P
P
P
P
17.7
P
P
TL/mi*'
P
P
P
1.20
1.19
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
F
F
F
P
P
P
P
P
P
P
P
P
P





CIIRHAT-'
P
P
P
F
F
F
P
P
P
P
P
P
P
P
P
P
P
P
P
]>
r
1.65
F
1.53
P
P
P
P
P
P
1.52
F
1.58
P
P
P
P
P
P
O.KAT-'
P
P
F
F
F
F
1.14
F
1.12
P
1.17
P
l>
P
P
P
p
P
P
P
P
F
F
F
P
P
P
P
P
P
F
F
F
P
P
P
1.02
F
P
NI/TA-'
P
P
P
P
P
P
P
P
P
3.8
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
2.4
P
P
P
P
P
P
P
P

-------
TABLE III-4
(continued)
Name of
f 1 rut
Nl. Iiulu-it r Jus


Northwest
Industries

NW Steel 6 Wirfc


Hurt on Co.


Nucor
Corporation

Oaklte
Products

Occidental
l*t-troluiuu

Ulln
Cttrporut iu n

P6F Industries


Park Chiiiolcul


Park
Elect rochciofcal

Peerless Tube


PenilUii 1 1


Yeari/
(x-l)
(x-2)
(x-J)
(x-l)
(x-2)
(x-1)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-i)
(x-2)
(x-3)
(x-l)

-------
                                                    TABLE II1-4
                                                    (conLinued)
I
oo
Name uf
Film
PeilllZOil

-
Phelpa Doil^u


Pgh-Dub Mollies
Sletl

I'll CalMH|;li
Forging*

Plant
Industries

Portec

j
Kedwan
Industtl us

Republic Sltul


Kevere ^oupfcr
& Brass

Reynolds MetalB


Kohl in Industries


Kusco Indutit r i*-'s


Season-All
Induut r Ics

Yea pi'
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
Nl
107
128
84
71
122
109
4
2
3
9
3
2
2
1
1
6
5
2
-23
-24
-15
72
171
87
-31
17
3
60
111
45
2
5
-3
.4
.2
.9
.8
.8
.6
CF
215
218
155
106
158
145
7
4
5
12
6
5
5
t
3
7
6
4
-22
-22
-13
157
252
166
-15
34
19
133
188
120
5
7
-2
1
1
2
1
1
1
QA
297
218
245
145
121
229
62
58
49
25
24
16
11
10
8
18
17
8
22
22
31
282
502
352
64
61
84
294
399
221
21
20
21
9
10
10
4
3
3
CA
445
338
321
336
324
362
73
73
57
52
52
37
20
IB
16
41
44
27
33
33
45
614
759
613
lay
194
173
845
902
717
31
32
37
21
22
19
8
7
6
Cl.
247
146
228
189
231
145
48
51
42
26
31
18
19
16
11
11
18
10
20
20
39
269
409
333
88
83
84
292
340
250
13
14
18
9
9
9
4
3
3
Tl.
1453
1281
1135
759
601
454
56
59
43
A2
42
30
33
26
20
22
28
19
89
89
110
792
810
755
321
329
333
1173
1525
1409
36
3D
49
22
24
22
6
6
5
MM
573
515
628
893
892
815
39
36
35
42
36.
32
16
16
16
35
29
24
11
11
35
1279
1232
1107
137
170
153
831
799
709
18
17
13
7
7
7
6
5
5
TA
2026
1798
1762
1652
1493
1269
96
95
78
84
78
62
48
44
36
56
57
13
100
101
146
20/1
2042
1862
458
499
486
2204
2324
2118
54
55
61
3O
32
29
12
12
10
NFA/TA^'
P
P
P
P
P
P
20.6
F
2O. 6
F
F
F
P
P
P
25.3
21.1
27.9
f
f
f
V
P
P
P
P
P
P
P
P
28.0
28.3
26.0
24.3
24.1
26.2
P
P
P
CF/TI.-'
14.8
17.0
13.7
13.9
P
P
11.7
F
10.9
P
13.8
18.2
LS.I.
12.4
13.5
P
P
18.9
r
f
F
19.8
P
P
F
10.4
F
F
12.3
F
12.8
17.2
F
F
F
F
20.0
16.7
19.6
TI./NW*'
F
F
1.81
P
P
P
1.42
1.64
1.23
1.00
1.18
P
F
1.81
1.28
P
P
P
F
F
F
P
P
P
F
1.94
F
1.65
1.91
1.99
1.97
F
F
F
F
F
1.05
1.21
1.11
OJKKAT"'
1.80
P
F
1.78
F
P
1.51
F
F
P
1.68
P
f
r
t
r
f
e
I.b8
1.65
F
P
1.86
1.84
P
P
P
P
P
P
P
P
1.99
P
P
P
P
P
1.84
(jHA-ri7
1.20
P
1.07
F
F
P
P
1.14
1.17
F
r
F
F
F
F
P
F
F
1.13
1. 10
F
1.05
P
1 . 06
K
V
1.01
1.01
1.17
P
P
P
1.14
i.t>:;
1.07
1.05
i. ii
F
F
NI/TA--'
2.7
P
P
P
P
P
P
2.0
3.5
P
3.5
3.9
f
3.2
3.1
P
t
r
K
K
F
3.5
P
P
F
3.4
F
P
P
2.1
P
P
F
F
F
3.1
P
P
P

-------
TABLE III-4
(continued)
Name of
Pirn
Sharon Si i*e 1


Signal Companies

1
PsiBnoile
' Corporation
i
; st. joe
Minerals

j Standard
1 Alliance

'.
Sid. Pressed
Steel
Sunbeam
• Corporation
j
1 Sunshine
1 Mining

SynaJ toy


Tf-iineUiice
Forging

Texasfculf


U.S. Reduction
Company

Utah
Intvrnut iona I

»«,*'
(x-l)
(x-2)
(x-l)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
NI
IS
47
12
1
176
59
17
24
22
82
89
38
2
3
2
li
23
31
28
4
5
3
3
2
.9
1
4
1
101
147
74
8
a
i
12
97
55
CF
25
58
22
78
201
80
32
39
35
113
113
56
2
3
2
8
16
12
38
45
41
5
6
4
3
2
2
2
.6
2
129
188
109
10
10
3
IbO
135
91
QA
89
140
62
485
464
254
70
69
75
126
146
82
10
13
12
26
37
27
199
188
152
26
21
24
9
8
6
7
9
4
118
229
103
14
22
8
118
80
60
CA
146
1U9
104
1149
1038
684
157
165
127
200
190
112
20
25
22
80
98
72
466
398
322
36
36
32
20
20
16
17
17
7
291
327
182
28
34
18
193
132
110
Cl.
50
86
49
443
403
273
39
54
39
107
119
65
10
12
11
27
43
34
213
198
126
9
11
9
6
8
5
a
13
6
105
142
68
16
23
12
176
118
84
Tl.
96
137
89
1058
747
672
99
115
100
195
197
151
19
22
21
56
71
54
340
272
207
31
34
30
12
15
13
29
24
13
527
417
336
19
29
19
505
461
428
NU
182
171
127
BOB
786
686
174
164
146
361
299
229
14
13
11
75
75
66
267
256
237
27
25
23
17
15
14
13
11
5
628
560
440
25
17
9
541
448
380
. 1
278
308
216 1
1866
1533
1358
273
279
247
556
496
380
33
35
32
131
146
120
607
528
444
58
59
53
30
30
26
42
35
18
1155
977
776
44
46
28
1046
909
828
NFA/TA-'
" P "
P
P
21.5
F
F
P
27.8
27.2
P
P
P
22.7
t
r
P
P
P
F
F
F
21.8
20.8
20.4
22.3
22.7
26.5
P
P
P
P
P
P
P
2.52
P
P
P
P
CF/TI.-'
P
P
P
F
P
11.9
P
P
P
P
P
P
r
11.2
11.3
13.8
P
P
11.2
16.4
19.7
16.8
17.9
14.1
P
15.4
13.6
F
P
15.3
P
P
P
P
P
13.5
P
P
P
ul
•n./nv-
f
p
p
1.31
p
p
p
p
P
p
p
p
1.38
1.76
1.99
P
P
P
1.27
1.06
P
1.15
1.33
1.28
P
P
P
F
F
F
P
P
P
P
1.69
F
P
1.03
1.13
CUHRAT--
P
P
P
P
P
P
P
t
P
.87
.60
.72
.99
.99
P
p
l>
p
P
P
P
P
P
P
P
P
P
F
F
P
P
P
1.78
F
F
F
F
F
QKAT-7
P
P
P
1.0'J
1.15
F
P
P
P
I. IB
P
P
F
1.01
I. 10
F
F
F
F
F
P
P
P
P
P
1.04
P
F
F
F
1.12
P
P
F
F
F
F
F
F
Nl/TA-'
P
P
P
2.2
P
P
P
P
P
P
P
P
P
P
P
•>
I
3.8
P
P
P
P
P
P
P
l>
3.1
P
P
P
P
P
P
P
P
P
P
P

-------
                                                   TABLE III-A
                                                   (concluded)
U)
o
Nairn: u f
Firm
Vu 1 Icy I n Jus t r luy


Vim Porn
Company

Vulcan
MatL*rtals

WuHhingCou
Steel

Uluniling-
Pittsburgh

Youngs town
Steul Door

»eari'
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
(x-l)
(»-2)
(x-l)
(x-l)
(x-2)
(x-3)
(x-l)
(x-2)
(x-3)
HI
IS
25
5
3
4
4
28
30
23
3
6
4
.6
73
7
1
3
2
CF
16
26
6
6
7
6
51
60
42
3
6
4
31
102
37
2
4
1

-------
                                                                             TABLE  II1-5
                                               FINANCIAL  DATA  FOR FIRMS  IN  THE HOLDOUT BANKRUF1'  SAMPLIi
 I
U)
Name of
firm
Ai-.ine-llainlltoii HI'B.


Aurodex


AIM Companies


Capeliart Corporation


Cuimnmltire Corporation


Commonweal til Oil
Refining

Dynamics Corporation


tcologlcal Science


lilcor Chemical


Klcctrospace


Kill. Inc.


C-arcia Corporation


Year
197/U-U
1976(x-2)
1975(x-3)
1975(x-l)
1974(x-2)
1973- 3)
1969U-1)
1968(x-2)
1967(x-3)
I970(x-l)
1969(x-2)
1968(x-3)
1972(x-l)
1971(x-2)
1970(x-3)
I977U-1)
1976(x-2)
1975(x-3)
1977(x-l)
!977(x-2)
197S(x-3)
NI
NA
-0.9
0.)
NA
.0(16
0.3
-2.9
1.3
-1.1
-4.0
3.2
1.8
1.1
0.8
-3.0
-18.0
-32.1
-24.2
-6.3
-9.9
2.3
2.8
2.5
0
-39.5
-8.8
-4.7
1.8
2.0
0.9
1.6
0.8
-8.7
-10.0
-4.5
0.3
CF
MA
-0.4
O.S
NA
0.6
1.0
NA
2.1
-0.6
-2.7
3.9
2.3
1.9
1.6
-2.2
-6.1
-19.8
-13.3
-4.3
-7.4
4.3
3.6
3.0
0.3
-38.6
-7.4
NA
4.0
2.5
1.2
3.6
2.9
-7.8
-7.6
-2.4
2.2
l)A
NA
5.1
4.0
NA
2.5
5.2
4.1
4.6
4.5
14.5
15.2
8.6
17.0
13.6
9.3
129.0
110.0
151.0
43.3
51.1
51.6
18.4
13.5
3.1
2.9
6.7
9.7
NA
11. 1
8.1
23.7
21.8
21.2
28.5
33.0
40.4
CA
HA
11.2
10.4
NA
8.8
11.2
13.4
a. 5
8.5
41.0
39.0
14.1
26.6
20.4
17.3
232.6
248.2
252.6
63.7
80.9
82.0
26.8
18.1
3.5
2.9
8.2
11.5
33.9
30.1
27.1
36.3
31.7
33.8
61.9
61.0
75.7
a.
NA
9.0
7.1
NA
6.5
7.8
9.6
4.0
4.5
43.1
32.9
7.9
21.4
14.2
12.0
401.9
221.2
208.5
39.8
51.7
40.9
16.9
7.9
1.2
11.4
11.6
12.4
19.4
18.3
15.8
14.0
12.7
15.9
55.7
46.5
60.0
TL
NA
15.3
13.0
NA
13.4
15.8
19.6
10.6
14.2
33.9
39.3
10.1
37.1
29.9
27.0
409.1
420.9
394.2
29.9
26.7
63.7
66.1
37.6
20.6
43.5
53.9
62.7
30.5
29.7
27.0
44.2
40.3
44.4
68.7
59.8
68.6
NW
NA
4.3
5.2
NA
0.9
0.8
6.9
7.6
6.4
7.1
11. 1
7.9
9.8
8.6
9.4
127.5
145.2
179.3
50.0
73.9
37.1
22.7
16.2
7.3
-32.0
6.5
3.9
13.8
11.9
10.1
10.8
10.4
9.8
12.1
22.1
26.7
TA
NA
19.6
18.2
NA
14.3
16.6
26.5
18.2
20.6
41.0
50.4
18.0
46.9
38.5
36.4
536.6
566.1
573.5
79.9
100.6
100.8
88.8
53.8
27.9
11.5
60.4
66.6
44.3
41.6
37.1
55.0
50.7
54.2
80.8
81.9
95.3
NFA/TAi'
NA
f ,
P
NA
F
22.9
21.7
25.6
24.8
f
f
r
f
r
r
f
t
t
F
F
r
V
f
V
e
r
t
r
f
r
r
t
20.7
r
f
r
CF/TI.*'
NA
F
F
NA
f
F
NA
19.8
F
f
F
f
F
F
F
F
F
F
f
F
F
F
F
F
F
F
NA
13.1
F
r
r
F
F
F
F
F
TI./NW -'
NA
F
F
NA
F
V
F
1.39
F
F
F
1.28
F
F
F
F
V
F
H
P
1.72
F
F
F
F
F
F
F
F
F
F
F
f
F
F
F
OIKIIAT-'
NA
F
F
NA
F
F
F
H
1.89
F
F
1.78
F
F
F
F
F
F
1.60
1.56
P
1.59
t
P
F
F
F
1.75
1.64
1.72
t
P
P
F
F
F
HUM-
MA
F
F
NA
F
F
F
1. 15
1.00
F
F
1.09
F
F
F
F
F
F
1.09
F
1.26
1.09
P
P
F
F
F .
F
F
F
P
P
P
F
F
F
                   •
• P "' 30.11  (passes all  Itstu); F < 2U.O (falla al
- P > 20.0  (p.iiisuu all  testa); F < 10.0 (falls al
— P < 1.0 (passes all tests); F > 2.0 (falls all
                                                           all lesttl)
                                                           all tests)
                                                             tests)
                   -v > 2.0 (passes all tents); F < 1.5 ((alls all  tests)
                   - P > 1.2 (passes all testu); F < 1.0 (falls all  tests)

-------
                                                      TABLE II1-5

                                                      (concluded)
M
ii
I

to
Niimf. of
Klrui
CKT Corporation


l.ulsure Croup, Inc.


M^istur Hr.in


Puller liiHlriiuiuiics


K. II. iu


Sltkln


TMA Co.


iranso«rain


Ul Icox-Clbbs


Huberts Company


Stulbi-r


Plcduitjnt Industrie



Year
1978(x-l)
1977(x-2)
1976(x-3)
I970(x-l)
1969(x-2)
1968(x-3)
19/OU-l)
1969(x-2)
1968(x-3)
1974(x-l)
1973(x-2)
1972(x-3)
1968(x-l)
1967(x-2)
1966U-3)
1977(x-l)
1976(x-2)
1975(x-3)
I970(x-l)
1969U-2)
I968(x*3)
1970
-------
 I
LO
U>
                                                                          TABLE II1-6
                                         FINANCIAL  DATA  FOR FIRMS  IN THE HOLDOUT  NON-BANKRUPT  SAMPLE
Naiuu of
Kliui
Attains Ml 1 1 is Corp.


Air Products & Chemicals


Akzoiia


Aiutitafie


American Pt*trofina


Alias Corp.


Avoiidale Mills


Hall Corporation


Bu 14 1 ng-llciul nway


boftj Warner


bucliLcr Corporation


Year"
x-l
x-2
x-J
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
HI
1.7
-5.9
1.3
54.2
39.7
21.1
7.9
33.4
35.7
8.1
13.5
9.8
40.2
86.7
37.0
3.0
1.6
-2.4
4.0
6.3
8.1
14.1
9.5
7.2
3.2
2.8
2.5
44.5
50.8
71.3
0.7
1.1
0.9
CF
3.8
-3.6
3.5
107. 6
85.5
61.6
41.3
63.4
63.6
15.5
20.3
16.7
71.3
IJ7.2
56.2
4.8.
3.1
-0.8
11.8
14.2
16.0
23.3
18.2
NA
5.1
4.7
4.2
87.3
93.9
111. 8
1.5
1.9
1.7
()A
11). 0
7.1
11.2
171.0
168.0
111.0
124.0
99.0
118.0
44.0
40.7
44.3
177.0
170.0
118. 0
12.9
12.7
9.5
40.0
49.8
38.4
511.6
40.6
35.7
20.2
15.1
18.0
253.0
245.0
271.0
3.5
4.5
4.5
CA
20. 3
19.8
24. 8
248.0
223.0
144.0
265.0
258.0
228.0
101.2
93.5
88.1
282.0
280.0
172.0
27.8
26.5
20.8
66.8
74.2
65.7
129.6
105.1
75.1
45.9
42.9
41.0
590.0
688. 0
642.0
11.8
12.7
9.6
Cl.
8.8
7.8
8.3
193.0
162.0
106.0
94.0
80.0
87.0
49.3
35.4
33.2
1S5.0
148.0
100.0
21.2
17.6
11.8
32.9
41.6
33.8
71.5
54.3
33.4
14.0
13.4
17.2
244.0
264.0
291.0
4.0
5.1
5.4
TL
16.6
24.0
26.5
475.0
411.0
309 . 0
341.0
301.0
264.0
92.6
79.1
75.2
238.0
230.0
175.0
25.6
22.1
16.0
35.4
43.8
36.0
122.3
89.4
64.8
29.0
29.4
29.7
504.0
615.0
530.0
10.5
12.3
10.4
NW
19.0
17.0
23.0
3U1.0
248.0
211.0
315.0
323.0
304 . 0
90.0
86.8
84. B
364.0
345.0
276.0
27.2
24.3
23.1
74.5
74.3
70.3
102.8
91.7
84.8
33.1
30.8
29.4
b89.0
670.0
642.0
5.3
4.6
3.5
TA
35.4
41.1
49.8
776.0
659.0
520.0
656.0
624.0
51.8.0
182.6
165.9
160.0
602.0
575.0
451.0
52.8
46.4
39.1
109.9
118.1
106.3
225.1
181.1
149.6
62.1
60.2
59.1
1193.0
1285.0
1172.0
15.8
16.9
13.9
2 /
NFA/TA-
P
f
r
f
r
r
t
p
p
p
p
p
p
p
p
p
p
p
p
p
t
p
p
p
20.7
22.5
25.3
P
P
P
23.0
22.0
26.0
CF/TL-'
P
F
11.5
P
P
19.9
12.1
P
P
16.7
P
P
P
P
P
18.8
14.0
F
P
32.4
P
19.0
20.4
11.1
17.6
16.0
P
17.3
15.3
P
14.3
15.4
16. 3
TI./NW4-'
I'
1.41
1.15
l.SB
1 .66
l.4(>
1 . 0«
I'
1'
i .03
P
V
1'
P
f
e
p
p
p
H
r
1.19
p
p
p
V
1.01
p
p
p
i.'ja
F
F
CUliRAT-'
P
P
P
F
F
F
P
P
P
P
|>
P
1.82
1.89
1.72
F
1.51
1.76
1'
1.79
1.94
1.81
1.94
P
P
|»
P
P
P
P
P
P
1.78
IJRA'I*'
P
F
I1
F
1.04
1.05
P
P
P
F
1.15
P
1.14
1.15
1.18
F
F
F
P
1.20
1.14
F
F
1.07
P
1.13
1.05
1.04
F
F
F
F
F
-For all  firms, x-l Is 1975.
-V > 30.0 (passes all testa)
- P > 20.0 (passes all tests)
- P < 1.0  (passes all tests);
-- P > 2.0  (passes all tests);
- I' > 1.2  (passes all tests);
                                               x-2 Is  1974, anil x-3 Is 19/3.
                                               F < 20.0 (falls all lur.ls)
                                               F < 10.0 (falls all tests)
                                               F > 2.0 (falls all tests)
                                               F < 1.5 (falls all tests)
                                               F < 1.0 (falls all lestii)

-------
                                                  TABLE IIT.-6


                                                  (continued)
V
CO
Name ul
t'iria
Hurl lug tun


Carter Wai lace


Chemut run


Clark Oil


CulL Industries


Coiicltuuico


Cone Hills Corp.


Crompcon Co.


Crown Central


Daliliitruni Corporation


Damon Corporal Ion


Daniel Industries


Dan River Inc.


ra.rl'
X-1
X-2
x-3
x-1
K-2
x-1
x-1
x-2
x-3
x-1
x-2
x-3
x-1
x-2
x-1
x-1
x-2
x-1
x-1
x-2
x-3
x-l
x-2
x-1
x-1 .
x-2
x-3
x-1
x-2
x-3
x-1
x-2
x-3
x-1
x-2
x-3
x-1
x-2
x-3
NI
39.8
9'J.5
82.4
8.4
3.1
10.4
10.4
22.3
10.4
5.2
-7.1
30.5
52.1
77.3
26.7
0.3
1.8
1.8
24.2
15.3
9.7
-2.6
U.I
5.8
5.5
10.2
8.4
0.9
1.8
-4.1
1.8
8.1
9.7
4.4
3.0
1.3
-2.9
7.0
10.4
CF
132.9
191.7
173.1
11.1
5.6
12.5
46.9
38.1
26.1
19.1
6.1
43.2
79.8
103.8
52.2
1. 3
2.7
2.6
37.8
27.3
21.6
1.5
2.5
7.9
13.0
16.6
14.4
1.5
2.2
-3.5
4.5
10.7
11.9
5.2
3.6
1.9
11. U
19.7
22.1
I(A
485.0
496.0
459.0
52.7
71.5
68.7
99.0
92.1
68.5
60.0
76.7
84.4
205. (I
227.0
205.0
11.1
11 .1
10.5
73.5
78.7
81.9
29.1
22.6
26.4
5O.6
50.0
'ia.3
4.5
3.5
S.4
52.3
59.0
55.5
13.0
9.6
7.8
89.4
79.2
88.3
CA
853.0
932.0
aao.o
83.0
94.0
99.0
190.2
173.5
146. 1
120.7
141.1
141.4
497.0
497.0
422.0
19.1
26.0
21.1
162.2
163.1
162.1
41.7
14.9
31.8
82.7
76.9
61.1
9.5
6.0
9.1
73.1
81.0
72.1
32.6
23.9
16.7
188.7
198.7
210.4
Cl.
237. 0
319.0
293.0
20.5
25.3
24.7
54.0
60.1
55.4
95.3
123.6
101.5
187.0
172.0
119.0
6.8
11.4
7.2
51.5
65.5
67.9
24.2
11.8
12.5
79.0
69.3
47.2
6.5
2.8
5.5
20.5
25.3
24.7
16.6
11.4
8.5
57.2
76.9
73.1
Tl.
671.0
757.0
756.0
30.9
19.9
22.5
161.9
158.6
141.4
208.3
202.9
253.5
491. 0
438.0
196.0
12.1
19.5
15.0
96.0
95.5
102.8
51.9
42.1
21.4
1 34 . 6
112.2
91.2
14.5
7.6
12.5
44.0
48.9
49.4
22.5
15.6
9.6
161.6
166.9
165.6
NH
H'i'j.O
887.0
825.0
124.3
119.0
118. /
209. 4
183.6
167.4
100.5
98.8
109.5
175.0
140.0
275.0
22.5
22.5
21.2
177.5
156.4
145.6
18.9
41.5
42.7
71.7
66.2
57.0
6.9
6.0
4.2
62.5
61. 8
52.7
21.9
16.9
14.3
144.0
148. 0
144.0
TA
1566.0
1644.0
1581.0
155.2
118.9
141.2
371.3
342.2
308.8
10H.8
301.7
263.0
866.0
778.0
671.0
34.8
42.0
36.2
273.5
241.9
248.4
90.8
81.6
64.1
206.3
178.4
148.2
21.4
11.6
16.7
106.5
110.7
102.1
44.4
12.5
21.9
308.0
315.0
310.0
NFA/TA2-
F
F
P
F
20.9
F
f
r
t
t
t
P
F
P
P
F
27.1
29.0
F
F
P
P
P
F
F
P
P
P
P
F
25.1
22.4
24.4
25.5
26.2
29.3
P
P
P
CF/TI.3'
19.8
25.1
22.9
t
t
e
r
p
18.4
F
F
17.0
16.3
P
13.2
F
13.8
17.3
P
P
21.0
F
F
P
F
14.8
15.8
10.3
F
F
10.2
F
P
P
V
19.8
6.7
11.8
13.3
TI./HW-'
F
f
V
r
f
p
f
p
p
F
F
F
1.31
1.29
1.44
P
F
P
P
P
F
1.33
1 .01
P
1 .SB
1.69
1.60
K
1.27
F
F
P
P
1.03
P
F
1.13
1.13
1.15
CUKUAT-'
P
P
P
F
F
P
F
P
P
F
F
F
P
V
l>
P
P
P
P
P
P
1.81
F
P
F
F
F
F
P
1.65
F
P
t
1.96
P
1.96
F
F
r
b/
O.KAT-
P
F
P
P
F
P
P
P
F
F
F
F
1.10
P
F
F
F
P
P
P
P
F
F
P
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F

-------
                                                        TABLE II1-6
                                                        (continued)
 I
(Jl
Num.: "f
Firm
l)e Sol,,, Inc.


IHvursu


KjBlu-Pithcr


Klcoi Clii.-mlcal (Jorp.


Kiuury liuliistriuB


KLhyl Curp.


Fab Ind.


Fairwuul Chemical


Fcrro


Flltrul


Fuule Mineral Co.






Y,ari'
x-l
x-2
x-l
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-l
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3

x-2
x-3

Nl
2.B
0.6
8.1
4.0
3.6
2.5
IB. 7
18. 0
14.8
5.2
2.5
0.4
10.7
10.2
7.0
61.0
74.3
52.9
2.1
0.9
1.9
0.1
1.1
0.2
15.2
19.9
15.9
5.5
4.4
3.1
9.8
11.3
2.8

-4.8
-1.0

CF
9.2
6.5
11.9
6.4
5.5
4.1
27.8
27.2
23.4
6.1
3.1
0.8
15.2
14.7
11.1
103.1
111. 1
B5.3
3.3
2.0
3.0
O.5
1.3
0.5
21.8
25.9
21.1
7.6
5.6
3.6
14.8
15.6
6.9

-4.6
-0.5

O.A
32.1
32.0
31.3
23.4
18.7
16.0
74.9
60.0
57.1
15.9
9.3
12.0
35.8
30.5
26.7
246.0
179.0
274.0
15.2
9.9
10.5
1.4
2.6
1.3
84.7
77.3
73.3
17.6
13.6
22.0
18.7
26.7
24.0

4.1
5.1

CA
75.4
87.5
97.0
15.6
32.2
24.6
IJI.l
129.0
111.2
16.7
10.0
12.1
61. 8
52. 1
45.7
371.0
128.0
365.0
24.7
20.5
19.7
2.9
4.2
2.4
I39.B
147.1
121.6
2b.O
20.0
25.2
49. B
45.2
45.0

5.4
6.4

Cl.
51.7
62.2
62.4
21.5
21.9
14.5
41.5
46.7
38.4
9.2
7.0
10.4
23.6
21.2
13.7
105.0
128.0
95.0
24.7
6.2
6.5
0.5
1.8
0.7
54.7
69.9
54.1
9.9
6.2
5.0
15. B
21.3
19.1

2.5
2.3

Tl.
64.9
75.7
74.8
33.3
31.2
23.0
81.8
BB.5
81.0
14.7 ••
14.2
21.7
61.4
49.1
40.8
440.0
383.0
348.0
10.2
11.7
11.9
0.7
2.1
l.l
88. 0
101.4
86.1
28.6
24.4
5.0
52.4
26.7
29.5

2.6
2.5

MU
70. 1
69.2
72.0
22.6
19.5
16.7
136.9
124.0
111.1
12.2
7.0
4.6
81. 7
74.3
67.4
436.0
400.0
348.0
14.6
17.8
17.2
4.0
3.8
2.8
126.3
115.1
99.4
31. b
27.9
26.1
74.4
67.2
58.7

4.4
9.1

TA
135.0
144.9
146.8
55.9
50.7
19.7
218.7
212.5
192.1
26.9
21.2
26.1
143.1
123.4
108.2
876.0
783.0
691.0
19.8
29.5
29.1
4.7
5.9
1.9
212.1
216.5
185.5
60.4
52.1
11.1
126.8
93.9
88.2

7.0
11.6

NFA/TA-'
P
t
V
24.5
21.5
22.4
P
P
P
t'
21.3
22.8
P
f
P
P
P
P
P
P
P
27.6
20.3
P
29.9
27.6
28.8
P
P
f
P
P
P

F
F

CF/T.,5-'
14.2
F
18.6
19.2
17.6
17.8
P
P
P
P
P
F
P
t
e
p
p
p
12.4
17.1
P
P
P
P
P
P
t
P
p
p
p
p
P

F
F

TI./MU--

.O'l
.04
.47
.6
.18
1'
p
P
1.21
F
F
V
p
p
1.01
P
1.00
p
p
P
P
p
P
P
P
P
V
P
P
P
P
P

P
P

UIKKAT-


.55
.06

.7
P
P
P
1.82
F
F
P
P
P
P
P
"'
P
P
P
P
P
P
P
P
p
P
P
P
P
P
''

F
P

WAT*7
F
F
F
1.09
F
1.1
P
P
P
P
F
F
P
P
P
P
P
P
P
P
P
P
P
l>
1>
1.11
P
P
P
P
1.18
P
P

P
P


-------
                                                      TABLE II1-6

                                                      (concluded)
M
M
i

LO
Nairn; uf
firm
(In If Resources 6
Chemical

tluiuly & Herman


Hunt Qiumical Corp.


Hy drome La la, Inc .


.lioi Walter


Youngs tuwu Sliect &
Tube

Yeari'
x-l
x-2
x-3
x-l
x-2
x-3
x-1
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
x-l
x-2
x-3
NI
28.7
36.2
7.4
12.7
12.2
5.4
3.6
6.1
6.L
3.4
3.7
1.0
69.3
62.9
53.6
40.9
96.4
43.3
CF
40.7
43.7
12.7
15.7
14.8
7.8
NA
NA
NA
4.8
5.2
4.5
98.4
88.6
76.2
105.2
156.8
97.9
QA
81). 6
79.0
36.9
56.6
54.5
50.1
19. 0
18. 0
15.2
14.4
12.8
15.3
618. 1)
618.0
528.0
208.0
319.0
251.0
CA
135.9
111.0
61.4
111.5
127.0
100.7
28.9
26.9
24.4
50.2
52.2
45.8
813.0
824.0
693.0
471.0
520.0
199.0
CL
63.1
46.6
24.3
59.6
82.0
72.0
6.5
6.9
5.6
14.3
18.7
13.8
519.0
627.0
498.0
250.0
304.0
198.0
TL
158.9
142.3
102.6
90.1
111.9
92.4
6.5
6.9
5.6
28.1
34.1
30.0
875. 0
862.0
731.0
546.0
560.0
561.0
IIW
118.4
76.1
41.7
56.4
45.4
35.1
38.7
36.7
32.1
33.2
29.7
26.5
432.0
398.0
351.0
713.0
692.0
664.0
TA
277.3
218.4
144.3
146.5
157.2
127.5
45.2
43.6
37.7
61.3
63.8
56.5
1309.0
1260.0
1082.0
1259.0
1252.0
1125.0
NFA/TA-
P
t
r
21.6
F
F
P
e
e
F
F
F
p
p
p
p
p
p
CF/TL-'
P
P
12.4
17.4
13.2
f
P
P
P
17.1
15.2
15.0
11.2
10.3
10.4
19.3
P
17.4
TL/MU-'
1.34
1.87
If
1.60
F
F
P
F
P
f
1.15
1.13
F
F
F
P
P
P
CIJKIIAT-
V
V
f
1.87
1.55
F
P
P
P
P
P
P
l.bO
F
F
l.BH
1.71
P
6/
(
-------
IV.  EVALUATION CRITERIA AND DEVELOPMENT OF RATIO TESTS




     All  test  alternatives  were  initially  evaluated  against  both the




primary  non-bankrupt  and  bankrupt  samples.   In  this  evaluation the




Agency assumed  that a  firm would have  to meet  the  financial criteria




required by a test  for its  most recent fiscal year in order to pass the




test  (i.e.,.a one-year eligibility  requirement).  The  seven financial




ratios selected  for detailed  analysis were  first  individually tested;




for each  ratio  examined several different pass-fail  cutoff  points were




investigated.   The most  promising   individual   ratio  tests  were then




combined  into  a series of  120 two-ratio  tests  (a firm  must  pass both




ratio elements  to  pass),  three-ratio tests (a firm must pass all ratio




elements  to  pass),  and contingent three-ratio  tests  (a  firm  must pass




two of three ratio elements to pass).




     To further validate  whether these tests represent  the  most effec-




tive  indicators  of  future  firm viability,  the  Agency  performed  a




supplementary analysis.   Other financial variables  identified  in the




preliminary literature  review  but not included in the  set  of 120 tests




were evaluated against a subset of 32 non-bankrupt and 12 bankrupt firms




from  the  primary   sample  that  had  proven  particularly difficult  to




classify.  The ratio that  performed best in this  supplementary analysis,




net fixed  assets/total assets,  was then combined with  the  other candi-




date  financial  ratios, and  31 additional  three- and  four-ratio  tests




were evaluated.




     All ratio tests were examined with two variants:   a one-year eligi-




bility requirement and a three-year eligibility requirement.  The former
                                 IV-1

-------
requires  that  a firm meet the requirements of a  financial  test based on

its  most  recent annual  report  in order  to pass the  test.  The  latter

requires  that  a firm meet the requirements of a  financial  test based on

its three most  recent annual reports in order to  pass  the test.

A.  Evaluation  Criteria

     Five evaluation criteria have been  developed  to assist the  Agency
in comparing the test results.  These measures are defined  and  described
below.

     (1)   ANB  represents the percentage  of  sampled non-bankrupt  firms

that pass a given  test.   For tests with a  one-year eligibility require-
ment,  A^JTJ is  measured  as the  percentage  of firms  in the  non-bankrupt

firm sample  which  passed the test  in  the  year 1975.   For  tests with  a
three-year eligibility requirement, Ajjg is  measured as  the  percentage of
firms  in  the  non-bankrupt firm sample which  passed  the test for all of

the years 1973-1975.
     Because the non-bankrupt firm sample  created  for this analysis  is

composed of companies from industry categories  which are most  likely  to
have  on-site   treatment,  storage  and/or  disposal  facilities  requiring
financial assurance, and  contains  a representative  sample of firms from
all size  classes,  the Agency believes that the  ANB  results can be used

as a  reasonable surrogate measure  of  the  overall percentage of viable
firms  that  would  be  able to pass  a  financial test.   Therefore,  tests
which  have  a.  very  low ANB  will generate  large  private sector expendi-

tures, because  many viable firms  will  be  required  to establish alter-
native forms of financial assurance.   Alternatively,  tests with a high

Ajjg will result in low private costs of complying with RCRA.
                                 IV-2

-------
     (2)   M represents  the  percentage of  sampled  bankrupt firms which

would fail a given financial test with insufficient time remaining prior

to bankruptcy to ensure  that alternative financial mechanisms are avail-

able for facility closure, post-closure, and liability requirements.

     Because firms  in financial distress often  suffer a rapid deterio-

ration in  their  liquid assets in the  two  to three years prior to bank-

ruptcy,  and  because  of  possible delays in enforcement  and litigation,

the Agency has  concluded that a one-year  lead  time would not be suffi-

cient to guarantee that  sufficient funds would be available to cover the

costs  of  closure,  post-closure,  and  liability  requirements.    Three

different assumptions as to what constituted "sufficient lead time" were

analyzed by the Agency:

    Best case - If a  test fails a firm at least two years prior to

    bankruptcy, there will still be  sufficient  time to ensure the

    funding of alternative mechanisms;  for  this  case the value of

    M is designated as Mg.

    Worst case - If a test fails a firm at  least three years prior

    to bankruptcy,  there will still be sufficient  time  to ensure

    the funding of alternative mechanisms;  for this case the value

    of M is designated as MW.

    Most probable case - All  firms  that are first eliminated by a

    test three years  prior to  bankruptcy  will provide alternative

    financial  assurance.  One-half  of the firms  that  are first

    eliminated two years prior to bankruptcy will provide alterna-

    tive financial  assurance.    For  this case  the value  of  M is

    designated as Mp.
    c' '                          IV-3

-------
     The  implications  of all  three  of these  assumptions  on the  inter-
pretation of  possible  classification results are presented in Table  IV-
1.   Although  the posibility of  injunctive  relief under Section 7003  of
RCRA might render all three of these assumptions  conservative estimates,
the  Agency  has  decided  to use  the  "most  probable  case"  definition  of
lead time requirements.  (See, Hazardous Waste Section, Land and Natural
Resources Division,  United States Department  of Justice,  Annual  Report
(October 1980.)
     (3)   C  represents  the percentage  of  sampled  bankrupt  firms that
fail a financial test with sufficient time remaining prior to bankruptcy
to  ensure  that  alternative  financial  mechanisms   are   available  for
facility closure,  post-closure,  and  liability requirements.   C is thus
measured as:
                              C = 100 - Mp
     (4)   D  reoresents  the difference  between  the  percentage  of non-
bankrupt  firms  passing  a  test  and  the percentage  of bankrupt  firms
passing the test, and is called the "discriminating power"  of a test.  D
is calculated according to the formula:
                              D = ANB ~ %
The higher the D score,  the better  the test discriminates  between bank-
rupt and  non-bankrupt  firms.  A test  receiving a D score of  100 would
perfectly discriminate between bankrupt and non-bankrupt firms, passing
all non-bankrupt firms and  failing  all bankrupt  firms.   A D  score  of
zero would  indicate  that the  same  percentage  of  bankrupt firms  pass a
test as non-bankrupt firms,  suggesting that the  test  does not  discrim-
inate  between bankrupt  and non-bankruot  firms.   A  negative  D  score
indicates that more bankrupt firms pass a test than non-bankrupt firms.
                                  IV-4

-------
                                              TABLE IV-1

          ESTIMATED PROBABILITY THAT A FIRM WILL NOT SET UP ALTERNATIVE FINANCIAL MECHANISMS
                     FOR VARIOUS PATTERNS OF PASSING AND FAILING FINANCIAL TESTS
Test Result—
1 Year
Prior to
Bankruptcy
P
P
P
P
F
F
F
F
2 Years
Prior to
Bankruptcy
P
P
F
F
P
P
F
F
3 Years
Prior to
Bankruptcy
P
F
P
F
P
F
P
F
Worst Case
Probability
for
One-Year
Eligibility
Requirement
1.0
1.0
1.0
1-0
1.0
1.0
1.0
0
Probability
for
Three-Year
Eligibility
Requirement
1.0
0
1.0
0
1.0
0
1.0
0
Most Probable Case
Probability
for
One-Year
Eligibility
Requirement
1.0
1.0
1.0
1.0
1.0
1.0
.5
0
Probability
for
Three-Year
Eligibility
Requirement
1.0
0
.5
0
1.0
0
.5
0
Best Case
Probability
for
One-Year
Eligibility
Requirement
1.0
1.0
1.0
1.0
1.0
1.0
0
0
Probability
for
Three-Year
Eligibility
Requirement
1.0
0
0
0
1.0
0
0
0
—'P = Passes Test; F = Fails Test.

-------
     (5)  E represents  the  number  of  firms  per  10,000  which pass a given




financial  test  and will  enter  bankruptcy without providing  alternative




financial assurances.   E, like M, can be calculated for  three different




assumptions as to what  constitutes sufficient lead time.




    Best case - If a test fails  a  firm at least two years  prior to




    bankruptcy, there will  still be  sufficient time to ensure the




    funding of alternative  mechanisms;  for this case  the  value of




    E is designated as  Eg.




    Worst case - If a test  fails a firm at  least three years  prior




    to  bankruptcy,  there will still  be  sufficient  time to ensure




    the funding of alternative mechanisms;  for  this case the  value




    of E is designated  as Ew.




    Most probable case  -  All firms that are first eliminated  by a




    test three years prior  to bankruptcy will  provide alternative




    financial  assurance.    One-half   of  the firms that  are  first




    eliminated two  years  prior  to bankruptcy  will  provide alter-




    native financial assurance.   For  this case, the value of  E is




    designated as Ep.




Ep is  used  for most analytic purposes.   A detailed formula for  Ep, as




noted in Section I, is:






                   Ep .      	F(MP)	
                       (U-F) x Aj^) -I- (F x Mp)




However,  because of  the  very  small  values  of  F,  a  less complicated




formula approximates  this completely  correct  measure with  an error of




less than 1 percent for  all  tests examined.   Ep is therefore  calculated




as:




                                 IV-6

-------
                         ANB

B.  Development of Ratio Tests

1.  Evaluation of Single Ratio Tests

     Each ratio listed in Section III.C.4 was  tested against  the  primary

bankrupt and non-bankrupt samples.  Several pass-fail  cutoff  points  were

analyzed for each ratio since the midpoint values used in  the historical

literature often vary  widely  between  studies.  These  single-ratio tests

are summarized in Table IV-2.
            A.
     In choosing  ratios  and  cutoff  points to  employ  in further  tests,

the Agency did not  use the same  sample  as it  used  in the rest of  this

study.  As noted in Section III, a larger primary sample which contained

retail  firms  was employed  in the  earlier phases  of   this  study.    The

evaluations of individual  ratios and cutoff  points  shown in  Table  IV-2

were  conducted  using  this  larger sample  and are reported  in terms  of

that larger sample  in this Section.   This sample consisted  of 60 bank-

rupt firms and  147  non bankrupt firms.  M and C were  evaluated using  a

three-year eligibility assumption,  whereas  Ajjg  was  evaluated  using  a

one-year  eligibility  assumption.    However,  results for   the  tests

presented in Sections V and VIII are derived  from the  sample  with  retail

firms removed.

     As Table  IV-2  illustrates,  the cash  flow/total liability ratio  is

the  most  significantly  predictive  single  ratio,   attaining the   two

highest D  scores,  the  three  highest C  scores, and the  four highest  E

ratings  (i.e.,  the  lowest  number  of  firms  per  10,000  that will  fail

without  providing  alternative  financial  assurance).    The  margins  of
                                               i
                                  IV-7

-------
                            TABLE IV- 2
                 PERFORMANCE OF SINGLE -RATIO

Test M 21
Description ; "P

CF/TL > .05
CF/TL > .1
CF/TL > .15
CF/TL > .2
CF/TL > .3
TL/NW < 3.0
TL/NW < 2.0
TL/NW < 1.5
TL/NW < 1.2
TL/NW 1.0
NI/TA > 0
NI/TA > .02
NI/TA > .04
NWK/TA > 0
NWK/TA > .2
NWK/TA > .25.
CURRAT > 1.2
CURRAT > 1.5
CURRAT > 2.0
QRAT > 1.0
QRAT > 1.2

NS/TA > 1.0
2 /
^
t
30.4
12.5
5.4 '
5.4
5.4
61.7
35.0
26.7
16.7
11.7
50.0
21.7
11.6
83.3
46.7
30.0
66.7
36.7
18.3
18.7
13.6

71.7
NS/TA > 1.5 1 41.7
58.9
32.1
17.9
8.9
3.6
81.7
61.7
41.7
31.7
25.0
68.3
43.3
31.6
90.0
60.0
50.0
76.7
55.0
31.6
33.9
25.5

80.0
48.4
3/
^B

88.5
83.8
69.9
52.0
24.5
96.0
90.7
83.4
67.5
60.9
92.1
83.4
72.8
100.0
74.8
63.6
98.7
97.3
74.9
76.2
53.1

77.4
36.9
21
C

55.3
77.7*
88.3*
92.8*
95.5*
28.3
51.7
65.8
75.8*
81.6*
40.9
67.5
78.4*
13.3
46.6
60.0
28.3
54.1
75.0*
73.7*
80.4*

24.1
54.9
4/
D

29.6
51.7*
52.0*
43.1*
20.9
14.3
29.0
41.7*
35.8
35.9
23.8
40.1*
41.2
10.0
14.8
13.6
22.0
42.3*
43.3*
42.3*
27.6

[-2.6]
4/
P

11.1
5.9*
3.7*
3.0*
4.0*
16.4
11.7
9.0*
7.9*
6.6*
14.1
8.6*
6.5*
19.1
15.7
13.8
16.0
10.4*
7.3*
7.6*
8.1*

21.6
[-11.5] 26.9
 ^
      values expressed in % terms except Ep which represents the number
  of firms (per 10,000 passing the test) that will fail without providing
  alternate financial assurance.  All results are based on the original
  samples that included retail firms.
2/
—Based on three-year eligibility requirement.

-Based on one-year eligibility requirement.

-^Based on three-year eligibility requirement for M and one-year eligi-
  bility requirement for A^.

*Exceeds minimum performance cutoff point.
                                IV-8

-------
difference between the  scores  of  the best cash flow ratios  and  those  of




other  ratios  are  also  substantial.    Thirteen  of  the  ratios  tested




reduced  the  normal failure  rate  for  large  (greater than  $10  M in net




worth) firms by more than 50 percent (from 22 per 10,000 to  less  than  11




per  10,000).   Ten  ratios eliminated  more  than 70  percent of  bankrupt




firms  in  sufficient time  to establish  alternative  financial guarantee



mechanisms:  nine had  a  discriminatory  power  (D)   of  greater  than  40




percent.    If  these  three  conditions  are  viewed  to  be  the   minimum




acceptable values of test effectiveness, TL/NW, NI/TA, CURRAT,  and  ORAT




all have cutoff scores that satisfy these criteria; only the net  working




capital to  total  assets and net  sales to  total assets ratios failed  to



provide significant results.




     Those parametric  values which  satisfied  one  or more  of  the above




minimum criteria (13 in all) were used to develop multi-ratio tests; the



others were  dropped  from further consideration.   There were two  excep-




tions  to  this general  classification  rule:    (1)  the  cash flow/total




liability  greater than  .3  ratio was  dropped, because  the  E  ratings



demonstrated that this test was already dominated by the .2  and  .15  cash




flow  cutoff  points, and  its extremely  high  rate of  non-bankrupt  firm




rejections made it a poor candidate for use in multi-ratio  tests  and (2)



the  total  liabilities/net worth  cutoff point  of  2.0  was  added to the




list  of  ratios  for  detailed  consideration,  despite  the   fact  that  it




failed to  meet the  minimum reauirements, because preliminary investiga-



tions  revealed that this ratio was  one of  the few that  could  be used




effectively to classify electric  utilities (see Section IV.  B).
                                  IV-9

-------
2.  Evaluation of Multi-Ratio Tests




     The  13  financial  ratios  selected  from  the  initial  evaluation




process  were  linked in  various  combinations and  then retested against




the primary bankrupt and non-bankrupt samples.  The ratios were combined




in  three  ways:    two-ratio tests (firm  must pass both  elements  of the




test to pass  the  test),  three-ratio tests (firms must pass all elements




to  pass  the  test),  and three-ratio contingent  tests  (firms  must pass 2




of  3 elements to  pass  the  test).   Since CF/TL and NI/TA are alternative




methods of measuring a firm's rate of return on its assets, these ratios




were not included within the  same  test;  the same procedure was followed




with CURRAT and QRAT, the two liquidity measures being evaluated.



     In  all,  120  alternative tests  were  investigated  in  this  phase.




Section VIII  presents  the  results  of  all tests both with one-year and



three year liability requirements in Tables  VIII-1  to VIII-4 and VIII-5




to VIII-8, respectively.




     As illustrated  in these Tables, two-ratio  tests are  more accurate




than single-ratio tests in  the C and E ratings.  This pattern continues,




at a somewhat lower rate of increase, for the more stringent three-ratio



tests (passage of all elements required to pass).  The three-ratio (pass



2 out of  3)  tests evaluated barely out-performed  the single-ratio cash




flow tests on the C, 0, and E measures.   Although test  formulations of



the three-ratio  type greatly increase the eligibility of firms  in the




non-bankrupt sample, they  also increase  misclassifications  of  bankrupt



firms by a similar (and sometimes greater) amount.
                                 IV-10

-------
     The  major problem  with  the  multi-ratio  tests evaluated  in  this




phase is that  the  discriminating power (D score) of the best tests  does




aot  increase  greatly.   The greater  levels  of  bankrupt  firm identifi-




cation  and  overall  test  effectiveness  indicated  in  these  results




apparently are attained  mainly by  excluding a  larger  number of viable




firms.  Consequently, the Agency performed supplementary tests to estab-




lish whether there were other financial variables that could be added to




these tests to improve their  discriminating  power,  while retaining  high




levels of bankrupt firm detection.




3.  Test of Supplementary Financial Ratios




     Based on  the  results of the initial  round of  testing,  a set of 32




non-bankrupt and 12 bankrupt firms  was identified that were consistently




misclassified  by  the tests evaluated  in the first  phase.   The results




achieved in testing  this  set  of firms against a second set of financial




ratios are summarized below.




     The following ratios used  in  this auxiliary analysis were selected




from  the  financial  ratios  tested  in  prior  bankruptcy  forecasting




studies:




    Retained earnings/total assets (RE/TA)




    Earnings before interest and taxes/total assets (EBIT/TA)




    Cash flow/net sales (CF/NS)




    Balance sheet value of preferred and common stock/net worth (PC/NW)




    Balance sheet value of  preferred  and common stock/current and long-




    term debt  (PC/CL -1- LTL)




    Net fixed assets/total assets  (NFA/TA)
                                 IV-11

-------
    Cash/total assets (Cash/TA)




    Cash/current liabilities (Cash/CL)




    Net sales/total assets (NS/TA)




    Net working capital/total assets (NWK/TA)




Values  of  each  of  the  ratios  were computed  for the  firms described




above.   The data  were  then analyzed  to  determine the  values  for each




ratio that would provide the most accurate overall classification of the




tested  firms.   The percentage  of  bankrupt and  non-bankrupt  firms that




passed  the  resulting test were  then compared to  derive an estimate of




the discriminating  power (D1)  of  each ratio.   For  this  analysis, the




value of  D' , defined as the difference  between A^g and  Mp,  was  used.




(This is  similar  to D,  but  substitutes  Mp for  My).   These results are




presented in Table IV-3.




     In many  cases, the  ratio  tested had higher  values  of  Mp  than of




Ajjg, resuitng  in  negative values  for  D' .   A single ratio test  with a




negative discriminating power would  not be likely  to add to the overall




effectiveness of financial tests combining several ratios.  Three ratios




— RE/TA, NS/TA,  and Cash/CL — had some  positive impact on firm  cate-




gorization,  with D' values of 15 to 24 percent.  These results, however,




were far  exceeded by the classification  results achieved  by  using the




NFA/TA  ratio.   Three different  cutoff points  for NFA/TA  were  tested,




with resulting D'  ratings of 30  to 44  percent.  More significantly, the




ratio of NFA/TA greater  than .3 correctly classified  9  of the 12 bank-




rupt firms as non-viable three years prior to bankruptcy.
                                 IV-12

-------
                             TABLE IV-3
             TESTS OF  SUPPLEMENTARY  FINANCIAL RATIOS-1
                                                     II
Test
Variables
RE/TA >
RE/TA >
EBIT/TA >
NS/TA > 1
NS/TA > 1
CF/NS >
CF/NS >
PC/NW >
PC/TL >
NFA/TA >
NFA/TA >
NFA/TA >
Cash/TA >
Cash/CL >
Cash/CL >
NWK/TA >
.2
.25
.12
.25
.4
.04
.05
.1
.1
.15
.25 -
.3
.04
.15
.2
.25
»P
41.7
25.0
41.7
41.7
41.7
91.7
83.3
66.7
50.0
50.0
33.3
25.0
50.0
50.0
33.3
75.0
v •
62.5
37.5
43.8
65.6
50.0
62.5
43.8
40.6
40.6
93.5
61.3
54.8
65.6
46.9
34.4
56.3
D'^/
20.8
12.5
2.1
23.9
8.3
-29.2
-39.5
-26.1
-9.4
43.5
28.0
29.8
15.6
-3.1
1.1
-18.7
I/
 Tests run against a portion of the EPA primary bankrupt and non-
bankrupt samples comprising 32 non-bankrupt and 12 bankrupt firms
that were frequently misclassified by the initial set of candidate
financial tests.

I/
 D1 = Difference between the percentage of viable firms passing a
test and the percentage of non-viable firms passing the same test
in year x-3.
                                IV-13

-------
     The NFA/TA ratio represents  the  portion of  a firm's assets tied to

long-term tangible property, excluding the portion of these fixed assets

that has already been depreciated.   It  represents the reserve of assets

that a company  can call on in a  time of  financial difficulties, either

as a reliable source of earnings or a. potential source (through sale) of


needed  capital.    This  ratio has  not  been  extensively evaluated  in

previous bankruptcy  forecasting  studies;  however,  several  studies have
                            •>.
examined the ratio that represents to some extent its inverse — current

assets/total assets (CA/TA), the fraction of total assets in the form of

cash, inventories, and short term receivables.  Both Edmister (1972) and

Deakin (1972) have  found CA/TA to be a significant indicator of finan-

cial stability;  and both  studies  have indicated  that  a high  CA/TA is

negatively correlated with  continued  firm solvency.   A  NFA/TA test also

can be  readily  passed by  almost  all  electric utilities, a  category of

hazardous  waste disposers  that  encounters  great  problems   in  passing

candidate  tests  using other  more  common  financial  ratios  (see Section

VI).   For  these  reasons,   the Agency decided to  evaluate  a number of

tests  incorporating NFA/TA  as a  variable   against  the  entire  primary

sample.  The results of  these tests with  one-year and three-year eligi-

bility requirements  are presented  in Section VIII, Tables  VIII-4  and

VIII-8, respectively.  As will be discussed  in the next  Section, most of

the tests  incorporating  NFA/TA (Tests 121-151) dominate  tests  which do

not incorporate this ratio.
                                 IV-14

-------
V.  TEST PERFORMANCE




A.  Construction of the Performance Curves




     The performance of a test is measured by two values: Ep, the number




of  firms  per  10,000  which  pass  the  test  that later  enter bankruptcy




without providing alternative financial assurance; and ANB,  the percent-




age of  non-bankrupt firms passing  the test.   Both  of  these values are




used to determine the  relative  costs  of alternative tests.  In order to




determine the  best  tests,  the Agency  ascertained,  for particular values




of  Ep,  the  test which allows  the  highest value of  ANB.   A test which,




for any  given value  of  Ep  has  the highest  value  of Ajjg,  is  termed a




dominant test.




     Using the  results  for all tests-of  a  given type (see Section VIII




for complete  test  results),  a graph was  constructed  for  each  of the




eligibility requirements,  depicting on one axis  the number of bankrupt




firms per  10,000 which pass  the  test, (E?)  and  on the  other  axis the




percentage of non-bankrupt firms which pass the test (ANB).




     Figure V-l illustrates the method used to determine if a point C is




on  the performance  curve.   If a test  is to be on the performance curve,




there must be no other tests which dominate that test.
                                  V-l

-------
                ANB
                             BETTER
                           DIFFERENT
DIFFERENT
 WORSE
                               FIGURE V-l

    TECHNIQUE FOR DETERMINING IF A POINT IS ON THE PERFORMANCE CURVE




     For example, a  particular  test which passes 6  per  10,000 of bank-

rupt firms  and  passes 70  percent  of non-bankrupt firms  would have the

coordinates of  (6,  70%)  in  Figure  V-l.  To compare the  performance of

other tests to this test, as illustrated in the Figure, any tests repre-

sented by  points  in  the  northwest quadrant, using point  C as the point

of origin, dominate point C because both the value of Ep is lower (i.e.,

fewer bankrupt  firms  pass the  test)  and  the  value  of  AJJB  is  higher
                                                   i
(i.e., a higher percentage of non-bankrupt firms pass the test.)  A test

with the coordinates  (4, 80%) would fall in the northwest quadrant, and

would indicate  that  4 of  10,000 bankrupt firms and 80  percent of non-


bankrupt firms pass the test.  This makes it superior to the test repre-

sented  by  point C.   Thus, if there were any  tests  with coordinates in

the northwest quadrant,  the  test represented by point C would not be on
                                  V-2

-------
the performance  curve.   Similarly, any  point  in the southeast quadrant




is absolutely inferior to point C.




     Points  in  the other  two quadrants  cannot  be  classified  as domi-




nating point C, but can only be classified as different.  For example, a




point with  coordinates  (8, 80%)  means  that 80  percent of non-bankrupt




firms pass the test, which is absolutely superior to point C; however, 8




out of  10,000 bankrupt firms  also pass, which  is  inferior  to point C.




There is  thus no way to classify  any points  in  the northeast or south-




west quadrants as dominant to point C.  Points in those quadrants may or




may not be on the  performance curve;  they would  have to be evaluated in




the same manner as point C.



     Once the  performance  curve is constructed,  it is  sometimes useful




to compare  specific tests to  the performance curve as a whole.   If a




test is plotted  above the  performance curve,  then it dominates at least




one test  on  the performance  curve.    If  it  is below  the  performance




curve,  then  it  is  dominated by  at least one  test on the  performance




curve.




B.  Performance Curves for the Primary Sample




     Figure V-2  shows the  performance  curve  for the  set of  151 ratio




tests  with  one-year  eligibility  requirements.    Table  V-l  lists  the




components of  the tests  and the  values  of  Ep  and ANB  for  each test.




Seventeen tests  lie  on  the performance  curve, thus  reducing  the set of




tests that  need  to be further  considered from 151  to  17.   The perfor-




mance curve  includes  tests ranging from  a  test  with an Ep of  0  and an




    of 49 percent to one with an Ep of 10.1 and an ANB of 96 percent.







                                  V-3

-------
<5
         100-

         95-
     NB
          50-
          45-

          40-

          35-
              139
           FIGURE V-2

PERFORMANCIi CURVE FOR  TESTS WITH
ONE-YEAR ELIGIBILITY REQUIREMENTS
                                       143
                           T22
Ho   .4
2.'o  2'.5  3.'o
5.'o  5.S
65
                                         7.s
                                                                                                          9 .*5   lo'.O lo'.5

-------
             TABLE V-l




          DOMINANT TESTS




(One-Year Eligibility Requirement)
Test
Number
139


122


141


143


138


151



134


113


150




Test
Variables
NFA/TA > . 3
CF/TL > .1
CURRAT >2.0
NFA/TA > .3
CF/TL > .1
TL/NW < 1.2
NFA/TA > .2
CF/TL > . 1
CURRAT > 1.5
NFA/TA > .2
TL/NW < 1.2
QRAT >1.0
NFA/TA > .2
CF/TL > .1
CURRAT >1.5
NFA/TA > . 3
CF/TL > .1
TL/NW < 1.2
CURRAT >2.0
NFA/TA > .3
CF/TL > .1
TL/NW < 1.0
CF/TL > .1
TL/NW < 1.0
QRAT > 1.0
NFA/TA > .3
CF/TL > .1
TL/NW < 1.5
CURRAT >2.0

Number of
Variables
Required
To Pass
All


All


All


All


All


3 of 4



2 of 3


2 of 3


3 of 4




EP
0


1.4


1.9


2.6


2.8


3.7



4.6


5.2


5.5




^B
(%)
49


53


55


58


76


77



79


80


81


1
1
              V-5

-------
 TABLE V-l
(concluded)
Test
Number
135


149



136


120


137


98


146


100


Test
Variables
NFA/TA > .3
CF/TL > .1
TL/NW < 1.2
NFA/TA > .3
CF/TL > .1
TL/NW <2.0
CURRAT >2.0
NFA/TA > .3
CF/TL > . 1
TL/NW < 1.5
CF/TL > . 1
TL/NW < 2.0
QRAT > 1.0
NFA/TA > .3
CF/TL > . 1
TL/NW < 2.0
CF/TL > . 1
TL/NW < 2.0
CURRAT > 2.0
NFA/TA > .2
CF/TL > . 1
CURBAT > 2.0
CF/TL > . 1
TL/NW < 2.0
CURRAT > 1.5
Number of
Variables
Required
To Pass
2 of 3


3 of 4



2 of 3


2 of 3


2 of 3


2 of 3


2 of 3


2 of 3


EP
5.6


6.0



7.9


9.1


9-1


9.3


9.9


10.1


\B
(Z)
83


87



89


92


92


93


95


96


     V-6

-------
     The  tests  on the  performance curve had  certain ratios  in  common.




All  tests  on  the  performance  curve  included  the  cash  flow to  total




liabilities  ratio.   Thirteen  of the  17  tests on the  performance  curve




included the net fixed assets  to  total assets  ratio.   Thirteen out  of .17




tests included a total liabilities  to  net worth ratio.




     Figure  V-3 shows  the  performance   curve (indicated  by the  heavy




line)  for  the  set  of  151  ratio  tests  with  a three-year  eligibility




reauirement.   The  lighter line  shows, for  purposes  of comparison,  the




performance  curve  for  tests  with  a  one-year eligibility  requirement.




Descriptions of these tests  and  their  values for Ep and ANB are  given in




Table V-2.   Nine  tests  with a three-year eligibility  requirement lie on




.this performance curve.  The tests  range  from  one with  an  Ep of  0 and an




ANB °^ -^ percent to one with  an  Ep.of 7.3  and an Ajjg  of 86 percent.  All




of these  tests  include  the cash flow to  total  liabilities  ratio  and  the




net fixed assets to total  assets  ratio.




     Comparing  these two  curves  shows   that  the  general  effect  of  a




three-year  eligibility  requirement  on any  given test  is  to lower  the




values of  both Ep  and  ANB.   The overall  effect is  relatively  slight,




with the  following  two  exceptions:   (1)  Test  139 markedly  improves with




a  one-year  eligibility  requirement:  (2)  only  tests   with  a one-year




eligibility  requirement  can  achieve  values   of Ajjg  greater  than   84




percent.




     The  choice  between a one-year and  three-year eligibility  require-




ment cannot  be  based solely on performance,  however,  since the Agency




must also  consider  the  administrative  burdens associated with different
                                  V-7

-------
                                                         FIGURE V-3
<
CO
            100-

             95-

             90-

             85-

             80-

             75-
         NB
    COMPARISON OF PERFORMANCE CURVE FOR TESTS WITH
 THREE-YEAR ELIGIBILITY REQUIREMENTS WITH PERFORMANCE
CURVE FOR TESTS WITH ONE-YEAR ELIGIBILITY REQUIREMENTS
                                                                               Key:  Dark Line - Three-Year Tests
                                                                                     Light Line - One-Year Tests
                         1.0  l!5  2.0  2.5  3:0  3.5  AiO 4.5  5.0  5.5  6iO 6.5   7.0  7.5  8.0  8.5  9.0  9.5  10

-------
                TABLE V-2




             DOMINANT TESTS




(Three-Year Eligibility Requirement)
Test
Number
139


141


125


138


127


134


135


136


137


Test
Variables
NFA/TA > .3
CF/TL > .1
CURRAT > 2.0
NFA/TA > .2
CF/TL > .1
CURRAT > 2.0
NFA/TA > .25
CF/TL > .1
TL/NW < 1.5
NFA/TA > .2
CF/TL - > .1
CURRAT > 1.5
NFA/TA > .2
. CF/TL > . 1
TL/NW < 2.0
NFA/TA > .3
CF/TL > . 1
TL/NW < 1.0
NFA/TA > .3
CF/TL > . 1
TL/NW < 1.2
NFA/TA > .3
CF/TL > .1
TL/NW < 1.5
NFA/TA > .3
CF/TL > .1
TL/NW < 2.0
Number of
Variables
Required
To Pass
All


All


All


All


All


2 of 3


2 of 3


2 of 3


2 of 3


EP
0


1.2


1.4


1.6


1.9


2.9


4.1


5.4


7.3


**B
(%)
32


44


56


64


67


72


77


82


86


                   V-9

-------
tests.   In order  to  assess the performance  of  one-year and  three-year




tests  together,  a merged  performance  curve,  consisting  of  all the




dominant  tests  without  regard  to  eligibility  requirements,  was  con-




structed.   Such a  curve is the envelope  curve of  the two performance




curves shown in Figure V-3.  Figure V-4 presents this envelope  curve and




Table V-3  presents  a  full description of  the tests on  this curve.  This




merged performance  curve contains  16 tests,  of which four have  a  three-




year  eligibility  requirement,  and  12  have  a  one-year  eligibility




requirement.




G.  Performance with Respect to the Holdout Sample




     As noted in Section III.B. the Agency used a holdout sample in this




study to help ensure the statistical validity of the results.  Figure




V-5 compares  the performance on the primary sample  and  on the holdout



sample of  those one-year eligibility  tests that  appeared on the primary




sample performance  curve;  Figure V-6 provides this  same comparison for




three-year eligibility  tests.   table V-4 compares  the results  of all




tests with a one-year eligibility requirement and Table V-5 of all tests




with a three-year  eligibility  requirement.   Tables  V-6 and V-7 provide



complete holdout sample results for a larger set of tests.




     For one-year  eligibility  tests, 11 of  the  17  tests  yield results



for the  holdout sample  which  lie above  the performance curve  for the




primary sample.   The  remaining  six  tests are only  slightly  below the



primary sample  performance  curve.   This  result strongly  suggests that




search bias has not led to  tests which are incapable of validly discrim-



inating between bankrupt and non-bankrupt firms  when applied to samples
                                  V-10

-------
f
              100-

               95-


               90-

               85-

               80-

               75-
          V
                    39 (1)
               45-


               40-

               35-
                                                              FIGURE V-4

                                         PERFORMANCE CURVE OF 1JEST FINANCIAL TESTS WITH EITHER
                                            ONE-YEAR OR THREE-YEAR ELIGIBILITY REQUIREMENTS
                                                                            9U(1)
                                                                           (1)
                                                                       120(1)
0
2.*0  2*.5  3.'o  3?
                                                                  5.'o  5.'s  6.'o  6.*5  ?!o  7.*5  S.'o  8.!5  9.'o  9 .*5  lo'.O lo'.5

-------
                  TABLE V-3

DOMINANT FINANCIAL TESTS WITH EITHER ONE-YEAR
   OR THREE-YEAR ELIGIBILITY REQUIREMENTS
Test
Number
139


125


138


127


138


151



134


136


135


Number
Of Years
Required
To 3e
Eligible
1


3


3


3


1


1



1


3


1


Test
Variables
NFA/TA > . 3
CF/TL > .1
CURRAT > 2.0
NFA/TA > . 25
CF/TL > . 1
TL/NW < 1.5
NFA/TA > . 2
CF/TL > . 1
CURRAT > 1.5
NFA/TA > . 2
CF/TL > . 1
TL/NW < 2.0
NFA/TA > .2
CF/TL > . 1
CURRAT > 1.5
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2
CURRAT > 2.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.5
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2
Number of
Variables
Required
To Pass
All


All


All


All


All


3 of 4



2 of 3


2 of 3


2 of 3


SP
0


1.4


1.6


1.9


2.8


3.7



4.6


5.4


5.6


ANB
49


56


64


67


76


77



79


82


83


                     7-12

-------
 TABLE V-3
(concluded)
Test
Number
149



136


120


137


98


146


100


Number
Of Years
Required
To Be
Eligible
1



1


1


1


1


1


1


Test
Variables
NFA/TA > .3
CF/TL > . 1
TL/NW < 2.0
CURRAT > 2.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.5
CF/TL > . 1
TL/NW < 2.0
QRAT > 1.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 2.0
CF/TL > . 1
TL/NW < 2.0
CURRAT > 2.0
NFA/TA > . 2
CF/TL > . 1
CURRAT > 2.0
CF/TL > .1
TL/NW < 2.0
CURRAT > 1.5
Number of
Variables
Required
To Pass
3 of 4



2 of 3


2 of 3


2 of 3


2 of 3


2 of 3


2 of 3


EP
6.0



7.9


9.1


9.1


9.3


9.9


10.1


^B
87



89


92


92


93


95


96


    V-13

-------
                                                          FIGURE V-5
f
        100-


         95-
     NBH
         45-
PERFORMANCE CURVE V. HOLDOUT SAMPLE  PERFORMANCE
FOR TESTS WITH ONE-YEAR ELIGIBILITY  REQUIREMENTS

           • 137
                                          • 136
                                                          120
                                  143
                                                                                                               146)
                                                                 137)
                                                                 (120)
                                                                                                          (98)
                                                                             (100)
            •  122
         40-
          35-
              139
H0   .5
2.'0  2*5  3.*0
5.'o  5.*5
6.*5
                                                7.*5
                                                                                                     9.'o 9 .'s  lo'.O  lo'.5

-------
          100-

           95
                     FIGURE V-6

PERFORMANCE CURVE V.  HOLDOUT SAMPLE PERFORMANCE FOR
  TESTS WITH THREE-YEAR ELIGIBILITY REQUIREMENTS
<
           90

           85-

           80-

           75-
« 136
                                                  (137)
                 134*127
                    • 138
                  .5   1.0  15  20 2.5  3.0  3.5  410  4.5 5.'0  5.5  6.0  615  710 7/5  8JO  8.5  9.0  915  10.0 10.5

-------
                                                    TABLE V-4
f
          COMPARISON OF  PRIMARY AND HOLDOUT  SAMPLE PERFORMANCE FOR TESTS ON THE ONE-YEAR  PERFORMANCE CURVE
            (Subscript 11 Indicates Holdout Sample Result, No Subscript  Indicates Priniary  Sample  Result)
Test
Number
98


100


113


120


122


134


135


136


137


US


139


Test
Variables
CF/TL > . 1
T1./NW < 2.0
CURRAT > 2.0
CF/TL > . 1
TL/NW < 2.0
CURRAT > 1.5
CF/TL > . 1
TL/NW < 1.0
QRAT > 1.0
CF/TL > . 1
TI./NU < 2.0
QKAT > 1.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2
NFA/TA > . 3
CF/TL > . 1'
TL/NW < 1.5
NFA/TA > . 3
CF/TL > . 1
TL/NW < 2.0
NFA/TA > . 2
CK/TL > . 1
CURRAT > 1.5
NFA/TA > .3
CF/TL > . 1
CURRAT > 2.0
Number of
Variables
Required
To Pass
2 of 3


2 of 3


2 of 3


2 of 3


All


2 of 3


2 of 3


2 of 3


2 of 3


All


All


•Vii
20.8


31.3


14.6


18.8


0


4.2


4.2


12.5


16.7


10.4


0


"P
39.2


44.0


19.0


38.0


3.5


16.6


21.4


32. 1


38.0


9.5


0


Net
Change*
(V'W
+ 18.4


+ 12.7'


+4.4


+ 19.2


+3.5


+12.4


+17.2


+19.6


+11.3


-0.9


0


*NBI1
84


91


74


91


44


81


86


91


98


70


37


^B
93


96


80


92


53


79


83


89


92


76


49


Net
Change*
(ANBirANB>
-9


-5


-6


_,


-9


+2


+3


+2


+6


-6


-12


EP»
5.4


7.6


4.3


4.5


0


1. )


1. 1


3.0


3.7

:
3.3

,
i 0


El-
9.3


10.1


5.2


9. 1


1.4


4.6


5.6


7.9


9.1


2.8


U


Net
Change*

+3.9


+2.5


+0.9


+4.6


+1.4


+3.5


+4.5


+4.9


+5.4


-0.5


0



-------
                                                                        TABLE V-4

                                                                       (concluded)
I
M
-~J
Test
Number
141


143


146


149



150



151



Test
Variables
NFA/TA > .2
CF/TL > . 1
CURRAT > 2.0
NFA/TA > . 2
TL/NW < 1.2
QRAT > 1.0
NFA/TA > .2
CF/TL > . 1
CURRAT > 2.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 2.0
CURRAT > 2.0
NKA/TA > . 3
CF/TL > . 1
TL/NW < 1.5
CURHAT > 2.0
NFA/TA > . 3
CF/TL > . 1
TL/NU < 1.2
CURRAT > 2.0
Number of
Variables
Required
To Pass
All


All


2 of 3


3 of 4



3 of 4



3 of 4



MPII
4.2


4.2


31.3


6.3



6.3



U



»P
4.7


7.1


42.8


23.8



20.2



13.0



Net
Change*
(V^,,)
+0.5


+2.9


+ 11.5


+ 17.5



+ 13.9



+ 13.0



*NBII
51


47


93


86



74



72



ANB
55


58


95


87



81



77



Net
Change*
(ANBlfANB)
-4


-11


-2


-1



-7



-5



EPII
1.8


2.0


7.4


1.6



1.8



0



EP
1.9


2.6


9.9


6.0



5.5



3.7



Net
Change*

+0.1


+0.6


+2.5


+4.4



+3.7



+3.7



              Mn all cases, a positive value in the Nee Change  column represents an improvement in test  performance in  the holdout over  the primary sample.

-------
                                                             TABLE V-5


                               COMPARISON OF  PRIMARY  AND HOLDOUT  SAMPLE PERFORMANCE FOR TESTS
                                        ON THE THREE-YEAR ELIGIBILITY PERFORMANCE CURVE
Test
Number
125


127


134


135


136


137


138


139


141


Test
Variables
NFA/TA > . 25
CF/TL > . 1
TL/NW < 1.5
NFA/TA > . 2
CF/TL > . 1
TL/NW < 2.0
NFA/TA > . 3
CF/TL > . 1
TI./NW < 1.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.5
NFA/TA > . 3
CF/TL > . 1
TL/NW < 2.0
•
NFA/TA > .2
CF/TL > . 1
CIJKRAT > 1.5
NFA/TA > . 3
CF/TL > . 1
CURHAT > 2.0
NFA/TA > . 2
CF/TL > . 1
CURKAT > 2.0
Number of
Variables
Requl red
To Pass
All


All


2 of 3


2 of 3


2 of 3


2 of 3


All


All


All


M.'l.
2.1


2.1


0


4.2


0.3


12.5


2.1


0


0


MP
3.5


5.9


9.5


14.3


20.2


28.5


4.7


0


2.3


Nee
Change*

+ 1.4


+3.8


+9.5


+10.1


+ 11.9


+ 16.0


+2.6


0


+2.3


ANBH
51


63


63


70


81


82


60


33


44


*NB
56


67


72


77


82


86


64


32


44


Net
Change*
(ANBH~*NB)
-5


-4


-9


-7


-1


-4


-4


+ 1


0


EHI1
0.9


0.7


0


1.3


2.3


3.4


0.8


0


0


EP
1.4


1.9


2.9


4. 1


5.4


7.3


1.6


0


1.2


Net
Change*
(VW
+0.5


+ 1.2


+2.9


+2.8


+3.1


+3.9


+0.8


0


+1.2


M
00
           *lu all  cases, a positive value in Che Net Change column  represents an improvement in test performance in the holdout  over the primary sample.

-------
                     TABLE V-6
PERFORMANCE OF BEST TESTS AGAINST HOLDOUT SAMPLE
         (One-Year Eligibility Requirement)
Test
Number
11

12

26

67


68


98


100


110


113


120


Test
Variables
CF/TL > .1
TL/NW < 1.5
CF/TL > .1
TL/NW < 2.0
CF/TL > .1
CURRAT > 1 . 5
CF/TL > .1
TL/NW < 1.5
CURRAT > 1.5
CF/TL > .1
TL/NW < 2.0
CURRAT > 1.5
CF/TL > .1
TL/NW < 2.0
CURRAT > 2.0
CF/TL > .1
TL/NW < 2.0
CURRAT> 1.5
CF/TL > .1
TL/NW < 2.0
QRAT > 1.2
CF/TL > .1
TL/NW< 1.0
QRAT > 1.0
CF/TL > . 1
TL/NW < 2.0
QRAT > 1.0
Number of
Variables
Required
To Pass
All

All

All

All


All


2 of 3


2 of 3


2 of 3


2 of 3


2 of 3


E3
2.4

2.2

3.5

2.5


2.4


4.4


6.0


2.0


3.7


3.0


"P.
3.6

3.3

5.2

3.8


3.6


5.4


7.6


4.0


4.3


4.5


.=»
4.8

4.4

7.0

5.1


4.8


6.5


9.1


6.0


5.0


6.0


*NB
77

84

79

72


77


84


91


91


74


91


                          V-19

-------
 TABLE V-6
(continued)
Test
Number
122
.


125



127


132


134



135





136



137


138


Test
Variables
NFA/TA > .3
CF/TL > . 1
TL/NW < 1.2

NFA/TA > .25
CF/TL > . 1
TL/NW < 1.5

NFA/TA > . 2
CF/TL > .1
TL/NW < 2.0
NFA/TA > .3
CF/TL > .15
TL/NW < 1.2
NFA/TA > . 3
CF/TL > .1
TL/NW < 1.0

NFA/TA > .3
CF/TL > .1
TL/NW < 1.2



NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.5

NFA/TA > .3
CF/TL > .1
TL/NW < 2.0
NFA/TA > . 2
CF/TL > .1
CURRAT> 1.5
Number of
Variables
Required
To Pass
All



All



All


2 of 3


2 of 3



2 of 3





2 of 3



2 of 3


All


EB
0



1.6



2.5


1.3


1.1



1.1





2.0



2.8


2.6


i
i
EP
0



2.4



3.1


1.3


1.1



1.1





3.0



3.7


3.3


*V
0



3.1



3-7


1.3


1.1



1.1





4.0



4.7


3.9


^B
44



58



74


72


81



86
•

i
1

91
i
i
i
98


70


         V-20

-------
 TABLE V-6
(concluded)
Test
Number
139


141


143


146


149



150



151



Test
Variables
NFA/TA > .3
CF/TL > . 1
CURRAT > 2.0
NFA/TA > .2
CF/TL > . 1
CURRAT > 2 . 0
NFA/TA > .2
TL/NW < 1.2
QRAT > 1.0
NFA/TA > .2
CF/TL > .1
CURRAT > 1.5
NFA/TA > .3
CF/TL > .1
TL/NW < 2.0
CURRAT > 2.0
NFA/TA > . 3
CF/TL > .1
TL/NW < 1.5
CURRAT > 2.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2
CURRAT > 2.0
Number of
Variables
Required
To Pass
All


All


All


2 of 3


3 of 4



3 of 4



3 of 4



EB
0


1.8


2.0


4.9


1.1



1.2



0



EP
0


1.8


2.0


7.4


1.6



1.8



0



^
0


1.8


2.0


10.8


2.1



2.5



0



^B
37


51


47


93


86



74



72



      V-21

-------
                   TABLE V-7
PERFORMANCE OF BEST TESTS AGAINST HOLDOUT SAMPLE
     (Three-Year Eligibility Requirement)
Test
Number
i
11

12

26

67


68


98


100


110


113


120


122


Test
Variables
CF/TL > .1
TL/NW <1.5
CF/TL > .1
TL/NW <2.0
CF/TL > .1
CURRAT >1.5
CF/TL > .1
TL/NW <1.5
CURRAT > 1 . 5
CF/TL > .1
TL/NW <2.0
CURRAT >1.5
CF/TL > .1
TL/NW <2.0
CURRAT >1.5
CF/TL > .1
TL/NW <2.0
CURRAT >1.5
CF/TL > .1
TL/NW <2.0
QRAT >1.2
CF/TL > .1
TL/NW <1.0
QRAT >1.0
CF/TL > .1
TL/NW <2.0
QRAT >1.0
NFA/TA > .3
CF/TL > .1
TL/NW <1.2
Number of
Variables
Required
To Pass
All

All

All

All


All


2 of 3


2 of 3


2 of 3


2 of 3


2 of 3


.All


EB
1.5

1.4

0

0


0


3.6


4.5


1.2


1.7


1.2


0


EP
2.9

2.7

2.1

1.5


1.5


4.8


6.2


3.7


2.5


4.3


0


*W
4.4

4.1

4.2

3.0


3.0


5.9


7.9


6.2


3.3


6.2


0


^B
63

67

65

60


60


77


81


74


56


74


40


                    V-22

-------
TABLE V-7 (Continued)
Test
Number
125


127


132


134


135


136


137


138


139


Test
Variables
NFA/TA > .25
CF/TL >' .1
TL/NW <1.5
NFA/TA > .2
CF/TL > . 1
TL/NW <2.0
NFA/TA > .3 '.
CF/TL > . 15
TL/NW <1.2
NFA/TA > . 3
CF/TL > .1
TL/NW <1.0
NFA/TA > .3
CF/TL > .1
TL/NW <1.2
NFA/TA > .3
CF/TL > . 1
TL/NW <1.5
NFA/TA > .3
CF/TL > .1
TL/NW <2.0
NFA/TA > .2
CF/TL > .1
CURRAT >1.5
NFA/TA > .3
CF/TL > . 1
CURRAT >2.0
Number of
Variables
Required
To Pass
All


All
•

2 of 3


2 of 3


2 of 3


2 of 3


2 of 3


All


All


EB
0


0


0


0


1.3


1.1


2.2


0


0


EP
90


.73


0


0


1.3


2.3


3.4


.77


0


A
1.8


1.5


0


0


1.3


3.4


4.5


1.5


0


**
51


63


58


63


70


81


82


60


33


           V-23

-------
TABLE V-7 (Concluded)
Test
Number
141


143


146


149



150



151



Test
Variables
NFA/TA > .2
CF/TL > .1
CURRAT >2.0
NFA/TA > .2
TL/NW <1.2
QRAT >1 . 0
NFA/TA > .2
CF/TL > . 1
CURRAT >1.5
NFA/TA > .3
CF/TL > . 1
TL/NW <2.0
CURRAT >2.0
NFA/TA > .3
CF/TL > .1
TL/NW <1.5
CURRAT >2 . 0
NFA/TA > .3
CF/TL > . 1
TL/NW <1.2
CURRAT <2 . 0
Number of
Variables
Required
To Pass
All


All


2 of 3


3 of 4



3 of 4



3 of 4



EB
0


2.3


4.5


0



0



0



EP
0


2.3


7.4


.66



.71



0



*W
0


2.3


10.2


1.3



1.4



0



^B
44


40


81


70



65



63



         V-24

-------
other than  the  sample used in  the  test selection.  The results for  the




three-year  eligibility  tests  further confirm  the  validity of the  tests




selected.   All  nine  three-year  eligibility tests for the holdout  sample




lie above ,the primary sample performance curve.   For  the tests on  the




merged performance curve,  only  three of the 16 tests on the  performance




curve lie below the merged primary sample performance curve.




     Although the  holdout  sample results attest  to the validity of  the




selection  procedure   for  the tests  on  the primary  sample  performance




curve, they are less  encouraging with  respect to  the  accuracy of  the




estimates  of  Ep   and  A^g.    When  one-year  eligibility   tests  on  the




primary sample  performance curve are tested against the holdout sample,




ANB falls  by  an average of four percentage points while Ep falls  by an




average of  2.6.  Similar  results are obtained  for tests  with a  three-




year eligibility requirement.   While small changes in the values of  Ajjg




are not  surprising given  the sample size, the  changes  in Ep are  unex-




pectedly high,  given  that  the primary and  holdout sample  were randomly




assigned  from  the  same  original  sample  of  firms.    Apparently,  the




resulting  holdout  sample  contained  firms with  much  weaker financial




performance two to  three  years  prior  to  bankruptcy than  the primary




sample.




     If the results of  the primary  sample are used, Test 100 (one-year)




has the highest value of Ajjg of any of the ratio tests examined.  If  the




results of  the  holdout  sample are used, Test 137 (one-year) is superior




to Test 100 (one-year), with both a higher Ajjg and a lower Ep.
                                  V-25

-------
     As  a  further check on the  relative  performance of these  tests,  Ep




and A"B  were  calculated  using  a weighted average of the performance  for




the primary  and holdout samples.   These results  are  shown in Table  V-




8.  Using these weighted average results, all three  tests would lie on a




performance curve, with none absolutely dominating the  others.




D.  Comparison to the Results of Other Tests




1.  Other tests examined




     This Section reports  the  performance  of  the other  types of tests




which were examined in this study in addition to the ratio  tests.




     One of the other tests evaluated was the financial test proposed  by




the Agency  on  May  19,  1980.   That  test  consisted   of  the  following




elements:




     Total liabilities to net worth < 3




     Net working capital > 2 X closure plus post-closure costs




     Net worth > $10 million




This test had a one-year eligibility requirement.  The  effect of the net




worth requirement and the  total  liabilities to  net worth ratio in elim-




inating  bankrupt  firms while  admitting  viable  firms  was evaluated using




the same methodology used for  deriving  the performance   of  the  ratio




tests.  The effect of the multiple requirement on the performance of the




test  is  more difficult  to evaluate.   A requirement   that net working




capital  be a multiple  of total  financial responsibility obligations




means  that net  working  capital must be positive.  As  a result, the May




19 test  was evaluated as a test which required  a ratio of  total liabil-




ities  to net  worth of  less  than three and a  current  ratio  of greater
                                  V-26

-------
                  TABLE V-8
 PERFORMANCE OF SELECTED LESS STRINGENT TESTS
USING WEIGHTED AVERAGES OF HOLDOUT AND PRIMARY
        SAMPLE TO DETERMINE Ep AND



Test
Number
137


146


100


Number
of Years
Required
To Be
Eligible
1


1


1



*

Test
Variables
NFA/TA > . 3
CF/TL > . 1
TL/NW < 2.0
NFA/TA > . 2
CF/TL > .1 .
CURRAT > 2.0
CF/TL > . 1
TL/NW < 2.0
CURRAT > 1.5

Number of
Variables
Required
To Pass
2 of 3


2 of 3


2 of 3 :





Tf
EP
7.1


9.0


9.2






NB
93


94


95


                      V-27

-------
chan one  (i.e.,  the  ratio  of current assets to current liabilities must




be greater than one for there to be positive net working capital).




     The  Agency  also  evaluated  tests  which  were  composed  only   of




multiple  elements.    In this  report,  the term  "Ability to  Pay Test"




refers  to those  tests  that required  that net  worth and  net working




capital each  exceed  six times the  total  financial responsibility obli-




gation to be covered through a financial test.  The performance  of those




tests was evaluated in the same manner as the performance of the May  19,




1980 tests.   That is,  the net worth  multiple  was ignored as  implicitly




accounted  for  by  the $10  million  in  net worth requirement,  and   the




current ratio was required to be greater than one.  The Agency evaluated




the Ability to  Pay  test using both one-year and three-year eligibility




requirements.




     The  performances  of  the May 19, 1980  and  Ability to Pay tests  are




shown in  Table V-9.  Neither the proposed May 19, 1980 test nor the  two




Ability to Pay tests are dominated by other tests, and therefore all  are




added  to  the  performance  curve.   These  tests  pass  almost  all viable




firms,  but also have very high values of Ep»




     The presence in a test of multiple elements of the type included  in




the Ability to Pay Tests does not  influence the Ep or A^ results.   All




ratio tests of  any  importance fail firms  that  would have failed a test




including a requirement that  the current  ratio  be greater than  1.  As a




result, the  addition  of  such a requirement  to a  ratio  test  does   not




alter either the A*,,, or E- of the test.
                                  V-28

-------
           TABLE V-9

  PERFORMANCE OF SELECTED TESTS
WITH RESPECT TO THE PRIMARY SAMPLES
Test Description
May 19, 1980 Test
Ability to Pay Test
Ability to Pay Test
Parameters
Tested
TL/NW < 1
OJRRAT > 1
CURRAT > 1
CURRAT > 1
Eligibility
Requirement
One-Year
One-Year
Three-Year
^
97
100
99
EP
15.1
20.2
18.4
               V-29

-------
2.  Comparison to Results Obtained by Other Investigators
     As noted in Section  II,  a number of studies have been published on
methods for forecasting bankruptcies.   The  most commonly used method in
these studies has been multi-discriminant analysis.  The disadvantage to
this approach is that tests based upon multi-discriminant analysis would
require much more extensive reporting forms  and would be more difficult
to check.  Nevertheless, the Agency concluded that if multi-discriminant
analysis would  lead to results  clearly superior to  those  of the tests
adopted in this study, it should consider adopting one of the tests from
the published literature  rather  than  one of  the tests it had developed.
The Agency therefore compared its results to the results reported in the
literature.
     Four of the studies which utilized multi-discriminant analysis were
further examined:   Altman (1968), Altman  et al. (1977), Deakin (1972)
and Deakin (1976).   All of these studies examined the same basic problem
as that  examined here:   forecasting  bankruptcies for  relatively large
firms engaged in manufacturing and, in some cases, retail trade.
     The first  step  in  the  comparison was to  examine the  results  of
these studies as reported, and to use the baseline failure rate employed
in this  study to  derive a  measure   of E? comparable  to  that  used  in
reporting  the  results  of  this  study.   Multi-discriminant  analysis  is
particularly prone  to search bias, with the result that performance with
respect to the holdout sample is almost always weaker than that with the
primary sample.  Whenever possible,  the performance  reported for these
studies is  that  obtained  against a holdout sample.   (This  was not pos-
sible for Altman et al. (1977) because no holdout sample was employed.)
                                  V-30

-------
     The results for these studies are shown in Table V-10.  The ANB and




Ep for each  study  are plotted on a  graph showing the performance curve




and holdout  results for one-year eligibility  tests in  Figure  V-7.   As




can be seen  from the  Figure,  the reported result  for  Deakin (1976)tis




dominated by  both  the primary  sample  performance curve  and  by many of




the results  from  the holdout  sample.   The  Deakin  (1972)  result  is,




however,  extremely strong and would be a dominant test even on a perfor-




mance curve consisting only of results from the holdout sample.




     The problem with these comparisons is that each study reported used




a different sample  of  bankrupt and non-bankrupt  firms,  with  the result




that  comparisons of  performance may  simply  reflect  different  sample




characteristics  rather  than  the actual .relative  strength of  the tests.




Table V-ll  compares the means of the  samples used  in various studies




with the  mean of the samples for the Agency primary and holdout samples.




As shown in the  Table,  the  means  of  the key financial variables for the




Agency bankrupt firm sample are higher than those of firms used in other




studies,  which  indicates that the firms used  by  the  Agency had greater




average financial strength.  This difference could be expected to result




in higher Ep  scores for the other tests  if  the Agency bankrupt samples




had been used in evaluating  those other  tests.  The non-bankrupt sample




of firms shows  more mixed  results.   In some cases  the means  of the EPA




samples show  greater  financial strength;  in other  cases, other studies




show greater financial strength.  Thus  no strong conclusion is possible




about how  A™  for  other studies  would change  if applied to  the Agency




samples.
                                  V-31

-------
                             TABLE V-10

               ESTIMATED PERFORMANCE OF TESTS DEVELOPED
                BY PRIOR BANKRUPTCY FORECASTING STUDIES
Study Name
Altaian (1968)-
Altman (1977)-
Deakin (1972)-
Deakin (1976)-
Type of Test
5-Variable MDA
7-Variable MDA
14-Variable MDA
5- Variable MDA
MB
17.0
15.1
8.0
NX
U.
w •
29.4
25.5
18.0
17.0
-SB
87.1
89.1
94.0
75.0
EP
5.8
5.0
3.0
5.0
— Performance based upon results reported in Altman (1977).  No Holdout
  Sample was used in Altman (1968), but this test was evaluated against
  the new sample developed in Altman (1977).  Note that Altman (1977)
  Bankrupt Sample contains 5 firms which did not fail.
2/
— Primary Sample Performance from Altman (1977).  No Holdout Sample
  was used.

— Holdout Sample results are reported in Deakin (1972).

— M,, based on results reported for M  for all firms passing the test
  and entering bankruptcy from 1-3 years following passing the test.
                                 V-32

-------
                                                           FIGURE V-7
LO
CO
          100-

           95-

           90-

           85-

           80-
           75-
              •  151
       UUII
70-

65-

60-

55-


50

45-
                 122
           40-
           35-
              •  139
                           RESULTS OF OTHER STUDIES PLOTTED WITH EPA ONE-YEAR
                               PERFORMANCE CURVE AND HOLDOUT SAMPLE RESULTS

                                       • 137
                               \» 1 36
                            D'72 •
• 120/        .
     • A'77   /
        A'68/ ,
                                                                                                      146)
             •135  • 149
             •  134
                                     (138)
                                • 150
                                                                                                         (100)
                     Key:  D172 -  Deakin (1972)
                          D'76 -  Deakin (1976)
                          A168 -  Altman (1968)
                          A177 -  Altman (1977)
                                    143
30-,
.4
2.'0  2'.5  3.'o
    5.0
                                                        5.S  6.'0  6*5
                                                                                                      9*0  9 .'s  lo'.O

-------
                                                                 TABLE V-ll

                                            COMPARISON OF SAMPLE MEANS  FOR  EPA AND OTHER
                                                     BANKRUPTCY FORECASTING STUDIES
               A.  Bankrupt Firms Samples;
f

Financial


CF/TL
TI./NW
CUKRAT
QRAT
NWK/TA
Due a Year
x-1
EPA
Primary
-.046
7.78
1.51
.77
.13
EPA
Holdout
-. 117
3.22
1.35
.67
.09
Deakin
(1972)
-.088
-51.00
.70
.35
-.21
Altman
(1968)
NA
26.78
1.33
NA
-.06
x-2
EPA
Primary
.017
3.67
1.58
.87
.18
EPA
Holdout
-.007
3.67
1.60
.77.
.22
Do akin
(1972)
-.052
10.66
.89
.48
-.07
A 1 tman
(1968)
NA
3.35
1.31
NA
.02
x-3
EPA
Primary
.049
1.84
1.97
1.07
.25
EPA
Holdout
.008
3.99
1.72
.90
.23
Deakin
(1972)
-.001
8.52
1.44
.53
. 18
Al ttoan
(1968)
NA
1.58
1.62
NA
.18
               B.  Non-Bankrupt  Finn Samples:
Financial


CK/TL
TL/HU
ClIKKAT
QRAT
NWK/TA

x-l
EPA
Primary
.261
1.08
2.75
1.38
.30
EPA
Holdout
.239
.997
2.61
1.40
.32
Dc-akin
(1972)
.132
1.15
2.33
1.13
.33
Al troun
(1968)
NA
NA
NA
NA
.41
Al tman
(1977)
.314
NA
2.60
NA
NA
Data Year

man
»77)
314
JA
60
W
JA
x-2
EPA
Primary
.278
1.18
2.29
1.22
.29
EPA
Holdout
.172
1.07
2.26
1.26
.30
Deakin
(1972)
.150
.89
2.37
1.18
.33
EPA
Utilities
.114
1.41
1.26
.91
.02
x-3
EPA
Primary
.271
1.20
2.42
1.37
.28
EPA
Holdout
.232
1. 17
2.50
1.45
.32
Dunk in
(1972)
.115
.91
2.51
1.24
.35
EPA
Utilities
. 113
1.41
1. 11
.74
.01
               SOURCE:  Hoody's Industrial Manual and company annual  Form 10-K reports for EPA samples; original articles cited for other samples.

-------
     In  order  to  provide  better  methods  of  comparing  tests,  Deakin




(1972)  tests  were  evaluated  using the  Agency  holdout  sample  using  Ew




rather  than Ep.   The Agency  used  E^  because it required gathering  only




one year  rather  than two years of  additional  data.  The result of  this




evaluation is shown in Figure V-8 and the  results of  the Deakin  test are




summarized in  tabular form in  Table  V-12.  When both the Agency's  and




Deakin1s  tests  are  compared against the same  sample of firms,  the  per-




formance  is  comparable.   Seven  of the Agency's  tests fall  below  the




performance curve for Deakin1s  test and seven are above it.  Two  of the




one-year eligibility tests are  not  comparable.   (If the line connecting




the last  two  points  representing Deakin's  tests  were extended, both  of




these noncomparable .tests  would be above  this  hypothetical performance




curve.)
                                  V-35

-------
                                                           FIGURE  V-8


                             COMPARISON OF EPA AND DEAKIN  (1972)  TESTS  USING EPA HOLDOUT SAMPLE
f
to
        NBH
                122
            40-
            35-
                139
                                                              137
                                                                                                         • 100
                                                                                Key:  D1 - Deakin (1972)
                                                                                                                       146*
3t"b   .5   llfl
2*.0
3.0
5.o
                                                                      EW
7.*5  8.'o  8J5  9.'o
                                                                                                                 Ifl'.O

-------
                              TABLE V-12


               PERFORMANCE OF DEAKIN 14-VARIABLE TEST

                     VERSUS EPA HOLDOUT SAMPLES*
Test Cutoff
Score
20.0
25.0
27.0
30.0
EPH
(x-3)
0
9.4
12.9
31.2
^SlBH
49
74
81
38
D
49
66
68
55
%
0
2.4
3.5
8.3
*G and Ep scores cannot be computed because only one year of
data, x-3, was analyzed against the Deakin test.
                                V-37

-------
VI.  THE NEED FOR ALTERNATIVE TESTS FOR SPECIFIC INDUSTRIES




     A number of  commentsrs  argued chat no  single  financial test could




be  selected as  an  appropriate  indicator  of  firm  viability,  because




financial ratios  varied greatly  among  industries.   Many  of these same




commenters  proposed  that separate financial tests  should be formulated




for each industry category.   The Agency was generally opposed to such a




concept, because such a structure would greatly increase the administra-




tive complexity of the financial responsibility regulations.  Initially,




it would be a  difficult task to  determine the number and type of cate-




gories that would be appropriate for this purpose.  For many industries,




data on bankrupt firms would be so limited that the statistical validity




of a  test  derived  from such data  would  be very questionable.  Further-




more, even  if  a framework and  set of  tests could  be established, many




large firms  with  diversified facilities would not  fit easily  into  any




classification scheme.  For example, if a parent firm primarily involved




in chemicals manufacturing attempted to use  the financial test provision




to guarantee the waste  disposal operations  of  its subsidiary,  a pulp




mill, a  decision rule  would  have to  be  developed  on whether  the firm




should be classified under the  chemicals  or paper category.  The Agency




believes that the addition of such complexities could only be justified




if  there  is compelling evidence  that  the  financial  test alternatives




evaluated   in   this   analysis    all   discriminated   against   specific




industries.




     This  Section examines  the  need  for  industry specific  tests  for




three categories  of industry:    manufacturing firms  by  two digit  SIC




classification;  utilities;  and hazardous waste management firms.



                                 VI-1

-------
A.  Manufacturing Industries




     To  test  the hypothesis that  specific manufacturing industries may




be unable to pass a.  general financial test, the Agency collected finan-




cial data for  all the firms listed  in the Expanded Coverage section of




the  1980 Moody's  Industrial  Manual  (Volume  1).   Using  the  Dun and




Bradstreet Million Dollar Directory,  the  Agency  identified the 2-digit




SIC  classifications  corresponding  to  each firm's  major  manufacturing




activities.   Table  VI-1 shows  the number  of  firms  examined by 2-digit




SIC  classification   activities.    The  Agency  then  tested  this sample




against  the  four financial ratios which  consistently  provided the best




classification performance  in  the  initial evaluation of  test alterna-



tives:    cash flow/total  liabilities, total liabilities/net worth, net




fixed assets/total assets, and  current assets/current liabilities.



     The results of  this evaluation, as shown in Tables VI-2 through VI-




5, do not in any way support the  argument for industry-specific tests.



Although there are some isolated differences between industries in cases




where stringent ratio cutoff points are assumed,  each of the four ratios




has at  least  one commonly  used  cutoff value (NFA/TA >  .2,  CF/TL > .1,



TL/NW <  2.0,  CURRAT >  1.5)  which at  least 75 percent of  the firms  in




each 2-digit SIC can  pass.   The  discrimination among  industry groups




resulting from  the  use  of  a single  test  therefore appears  slight and




well within the  boundaries  of difference  illustrated by  other financial




indicators (e.g.,  bond  ratings).   Consequently,  there  is  no justifi-




cation   for   requiring  industry   specific  tests   for   manufacturing




industries.
                                 VI-2

-------
             TABLE VI-1

NUMBER OF FIRMS BY SIC CLASSIFICATION
  USED IN TABLES VI-2 THROUGH VI-6
SIC
20
22
24
26
28
29
30
32
33
34
35
36
37
38
39
Description of Industry
Food and kindred products
Textile mill products
Lumber and wood products, except furniture
Paper and allied products
Chemicals and allied products
Petroleum refining and related industries
Rubber and miscellaneous plastics products
Stone, clay, glass, and concrete products
Primary metal industries
Fabricated metal products, except machinery
and transportation products
Machinery, except electrical
Electrical and electronic machinery, equipment
and supplies
Transportation equipment
Measuring, analyzing, and controlling instru-
ments; photographic, medical and optical
goods; watches and clocks
Miscellaneous manufacturing industries
Number
of Firms
33
8
15
17
51
13
29
19
26
44
65
38
33
20
14
                VI-3

-------
                             TABLE VI-2
PERCENTAGES OF FIRMS IN MANUFACTURING INDUSTRIES PASSING OR FAILING
       GIVEN CASH FLOW/TOTAL LIABILITY (CF/TL) REQUIREMENTS
                (Based on Sample Described in Text)
SIC
20
22
24

26
28
29

30

32

33
34


.35
36


37
38




39

Description of Industry
Food and kindred products
Textile mill products
Lumber and wood products,
except furniture
Paper and allied products
Chemicals and allied products
Petroleum refining and
related industries
Rubber and miscellaneous
plastics products
Stone, clay, glass, and
concrete products
Primary metal industries
Fabricated metal products,
except machinery and
transportation products
Machinery, except electrical
Electrical and electronic
machinery, equipment, and
supplies
Transportation equipment
Measuring, analyzing, and
controlling instruments;
photographic, medical and
optical goods; watches and
clocks
Miscellaneous manufacturing
industries
CF/TL
< .10
9.1
18.7
13.3

-
9.8
7.7

3.4

10.5

3.8
6.8


7.7
10.5


9.1
-




7.1

>.1Q
90.9
81.3
86.7

100.0
90.2
92.3

96.6

89.5

96.2
. 93.2


92.3
89.5


90.9
100.0




92.9

>.15
72.7
62.5
73.3

88.2
72.5
76.9

75.9

84.2

61.5
68.2


72.3
81.6


66.7
75.0




64.3

>.20
48.5
50.0
60.0

64.7
47.1
53.8

55.2

68.4

42.3
40.9


50.8
57.9


24.2
55.0




57.1

                                VI-4

-------
                             TABLE VI-3

PERCENTAGE OF FIRMS IN MANUFACTURING INDUSTRIES PASSING OR FAILING
     GIVEN NET FIXED ASSETS/TOTAL ASSETS (NFA/TA) REQUIREMENTS
                (Based on Sample Described in Text)
SIC
20
22
24

26
28
29

30

32

33
34


35
36


37
38




39

Description of Industry
Food and kindred products
Textile mill products
Lumber and wood products,
except furniture
Paper and allied products
Chemicals and allied products
Petroleum refining and
related industries
Rubber and miscellaneous
plastics products
Stone, clay, glass, and
concrete products
Primary metal industries
Fabricated metal products,
except machinery and
transportation products
Machinery, except electrical
Electrical and electronic
machinery , equipment , and
supplies
Transportation equipment
Measuring, analyzing, and
controlling instruments;
photographic, medical and
optical goods; watches and
clocks
Miscellaneous manufacturing
industries
NFA/TA
< .20
9.1
-
13.3

-
7.8
-

10.3

10.5

3.8
13.6


18.5
21.1


12.1
20.0




28.6

>.20
90.9
100.0
86.7

100.0
92.2
100.0

89.7

89.5

96.2
86.4


81.5
78.9


87.9
80.0




71.4

>.25
75.8
93.8
73.3

100.0
80.4
92.3

89.7

78.9

84.6
72.7


70.8
71.1


75 .'8
65.0




50.0

>.'30
66.7
87.5
66.7

94.1
72.5
92.3

86.2

78.9

76.9
63.6


50.8
47.4


51.5
55.0




42.9

                                VI-5

-------
                             TABLE VI-4
PERCENTAGE OF FIRMS IN MANUFACTURING INDUSTRIES PASSING OR FAILING
      GIVEN TOTAL LIABILITIES/NET WORTH (TL/NW) REQUIREMENTS
                (Based on Sample Described in Text)
SIC
20
22
24

26
28
29

30

32

33
34


35
36


37
38




39

Description of Industry
Food and kindred products
Textile mill products
Lumber and wood products,
except furniture
Paper and allied products
Chemicals and allied products
Petroleum refining and
related industries
Rubber and miscellaneous
plastics products
Stone, clay, glass, and
concrete products
Primary metal industries
Fabricated metal products,
except machinery and
transportation products
Machinery, except electrical
Electrical and electronic
machinery, equipment, and
supplies
Transportation equipment
Measuring, analyzing, and
controlling instruments;
photographic, medical and
optical goods; watches and
clocks
Miscellaneous manufacturing
industries
TL/NW
>2.0
9.1
-
20.0

5.9
3.9
7.7

3.4

10.5

3.8
6.8


6.2
5.3


6.1
-




14.3

< 2.0
90.9
100.0
80.0

94.1
96.1
92.3

96.6

89.5

96.2
93.2


93.8
94.7


93.9
100.0




85.7

< 1.5
78.8
93.8
73.3

88.2
80.4
61.5

86.2

84.2

73.1
75.0


83.1
81.6


69.7
85.0




78.6

•<1.2
69.7
75.0
73.3

76.5
47.1
46.2

65.5

63.2

57.7
52.3


58.5
63.2


39.4
80.0




64.3

< 1.0
42.4
56.3
46.7

52.9
31.4
15.4

48.3

57.9

38.5
36.4


46.2
44.7


24.2
40.0




57.1

                               VI-6

-------
                       TABLE VI-5

PERCENTAGE' OF FIRMS IN MANUFACTURING INDUSTRIES PASSING
 OR FAILING GIVEN CURRENT RATIO (CURRAT) REQUIREMENTS
          (Based on Sample Described in Text)
SIC
20
22
24
26
28
29
30
32
33
34
35
36
37
38
39
Description of Industry
Food and kindred products
Textile mill products
Lumber and wood products, except
furniture
Paper and allied products
Chemicals and allied products
Petroleum refining and related
industries
Rubber and miscellaneous plastics
products
Stone, clay, glass, and concrete
products
Primary metal industries
Fabricated metal products, except
machinery and transportation
products
Machinery, except electrical
Electrical and electronic
machinery , equipment , and
supplies
Transportation equipment
Measuring, analyzing, and control-
ling instruments; photographic,
medical and optical goods;
watches and clocks
Miscellaneous manufacturing
industries
CURRAT
< 1.5
6.1
6.2
6.7
5.9
9.8
23.1
3.4 .
5.3
3.8
6.8
9.2
13.2
15.2


>1.5
93.9
93.8
93.3
94.1
90.2
76.9
96.6
94.7
96.2
93.2
90.8
86.8
84.8
100.0
100.0
>2.0
48.5
68.8
53.3
70.6
56.9
15.4
69.0
68.4
53.8
68.2
58.5
50.0
54.5
70.0
64.3
                        VI-7

-------
B.  Electric Utilities




     In  their  comments  on the proposed financial  test of May 19, 1980,




firms in the electric utility industry asserted that utilities could not




be  analyzed  using  "normal" financial logic, and  should be judged under




alternative  criteria.    Utilities  objected  to  the  use of  net  working




capital as a required component of a financial test, since many firms in




each  category  operate  regularly  from  a  negative net  working  capital




position.  Both  the  data submitted by the commenters and an independent




check by the Agency verified that utilities do operate with negative net




working capital.  As a  result,  any test which requires that net working




capital  be  some multiple  of  financial responsibility obligations will




fail most utilities.




     The  Agency also  examined  the  question  of  the  extent  to  which




specific  ratio  requirements might  consistently fail  viable utilities.




To examine this issue,  a sample of 26 electric utilities were drawn from




the  1978  and  1979  Moody's Public Utilities Manual  (see Table  VI-6).




Table VI-7  illustrates  the performance of this sample  against  the  key




single-ratio tests identified in Section IV.  As  can  be seen,  a number




of  these  ratios are totally  unsuited for  analyzing  utility viability.




The following  financial ratios  and  pass-fail  cutoff  points were  iden-




tified as potentially acceptable tests:




    CF/TS > .1




    TL/NW < 2.0, < 1.5




    NI/TA > .02, > .04




    NFA/TA > .2, > .25,  > .3
                                 VI-8

-------
                                                               TABLE  VT-6
                                        FINANCIAL DATA  FOR THE ELECTRIC  UTILITY SAMPLE
Maim: of
Firm
Hlai:k III UK
I'nwtT .mil l.lp.lit

lloHton Kill Him


Carol Ilia 1'ower
anil Unlit

iA'iitral Hudson 
P
P
F
K
F
F
F
F
•  I'   III.II (p.-iiisns .-ill LCKts);  F '-  2O.O  (l.illa nil teals)
21V •  2(1.0 (II.-IK.SCH all ti:HlH);  F •  10.0  (falls all tuKtu)
  I1   1.0 (II.-IHSUK all lasts);  F > 2.0  (falls all tests)
- P   2.0 (|>a:;s.'H all testa);  F
V,.
      1.2  (passes all tests);  F
1.5 (falls all tests)
I.0 (lalIs al1 tests)

-------
TABUS VI-6 (Continued)
Nami- ul
!•' 1 riu
I'lorlila I'uwi.-r ami l.l|;lil:


1 :<•!!,• r a 1 I'ulil lc
Ul 1 1 lly Corp.

01,1,. lull son


IV.uusy 1 van la I'owor
ami I.I t;lit

IVnplirs lila KliKtrlc
C.mi|Mny

I'lfdmoiiL Nat u nil lias





I'url laml CentT.il
lil.tclrl.-

I'oLtimai: Flecirli:
l'uw<:| til.

PiiMIc Sitrvlci; Ci>. uf
Cu l<> ratio

Public Sm'vlcti I'o. ol
1 mil ana

Pill, II.- S.-rvli:,.. <:,.. of
Hi-w llam|i.slilri:

Y.:ar
1977(x-l)
l976(x-2)
1975(x-3)
1977(x-l)
1976(x-2)
1975(x-3)
1977(x-l)
1976(x-2)
!975(x-3)
1978(x-l)
1977(x-2)
1976(x-3)
1978(x-l)
1977(x-2)
1976(x-3)
1978(x-l)
1977(x-2)
1976(x-3)
1978
-------
                                                                                TAIil.li  VI-6
N.-IIIH! of
Hriu
San l)frt;o (>:IH
anj K|i;t:lrlr

Saviiniuili Klurirlt:
Power i.'u.

Ye.ir
I.978U-1)
1977(x-2)
1976(x-3)
1978(x-l)
1977(x-2)
1976(x-3)
Nl
66.8
60.2
50.5
11.4
8.7
6.4
C.F
1.02 . 4
92.8
80.5
17.4
14.3
11.2
QA
NA
NA -
NA
NA
NA
NA
CA
199.0
187.0
146.0
15.8
11.1
10.8
Cl.
163.0
213.0
181.0
16.7
11.0
8.7
1
86(
83
701
16'
13'
12
Tl.
860.0
831.0
706.0
169.0
134.0
127.0
NW
694.0
582.0
481.0
79.0
73.0
63.0
TA
1554.0
1413.0
1187.0
248.0'
207.0
190.0
NFA/TAi/
P
P
P
P
P •
P
CK/T,,^
11.9
11.2
11.4
10.3
10.7
f
TI./NW-'
1.24
1.43
1.47
K
1.84
K
CUKKAT-
f
K
K
F
K
K
(JRAT-
NA
K
K
K
NA
NA
<5
M

-------
                          TABLE VI-7

         APPLICABILITY OF FINANCIAL TESTS TO UTILITIES
                      , (26 Firm Samples)
1
1
Single
CF/TL
CF/TL
CF/TL
TL/NW
TL/NW
TL/NW
TL/NW
NI/TA
NI/TA
NWK/TA
CURRAT
CURRAT
QRAT
QRAT
NFA/TA
NFA/TA
Ratio Tests
> .1
> .15
> .2
< 2.0
< 1.5
< 1.2
< 1.0
> .02
> .25
> .25
> 1.5
> 2.0
> 1.0
> 1.2
> .25
> .3
w
(One- Year
Eligibility
Requirement)
77
4
0
81
46
12
0
100
54 '
0
12
4
8
4
100
100
^JBU
(Three-Year
Eligibility
Requirement)
54
4
0
77
38
4
0
100
46
0
8
4
4
4
100
100
I/
  A™.. = Percent of viable utilities passing a given test.
                             VI-12

-------
     Given  this  reduced list of  acceptable  parameters, only  28  of the




151 multi-ratio tests were classified as applicable to utilities.  These




tests were  then  evaluated against the utility  sample,  with the results




summarized  in  Table VI-8  and  VI-9.   For many  of  the  higher Ep score




values, the tests that  can be  effectively applied to utilities are also




those whi.ch are  most  effective  for other  industries.  If,  however,  a




greater  than  90  percent  reduction  in the overall  rate of  large firm




bankruptcy is sought, the tests which permit an acceptable percentage of




utilities to pass also reject a higher percentage of other viable firms.




This disparity is  illustrated  in Figure  VI-1,  which contrasts one-year



tests on the performance curve  with utility applicable tests, using only




tests applicable to utilities derived in Section V.




C.  Hazardous Waste Management  Firms




     Because  the majority  of  firms  currently  in  the  hazardous  waste




management  business  are either  privately held  or  single-site entities




with assets less than  $1  million (Foster D.  Snell, 1976),  it  was not




possible to assemble a  sample  of  hazardous waste management firms which




could be used to evaluate test  applicability.




     Table  VT-10 shows average  ratios  for  hazardous  waste  management




firms from  1971-1975.   These ratios suggest that the average hazardous




waste management firm  would have little  difficulty with  possible  net




fixed assets to  total assets requirements.  However the average firm in




this industry would  have had  trouble  in many of these  years  with pos-




sible current ratio and total  liability  to net  worth requirements.  The




average profits  to  total  debt  ratio suggests an uncertain  picture with






                                 VI-13

-------
                TABLE VI-8
OVERALL EFFECTIVENESS OF UTILITY-APPLICABLE
              FINANCIAL TESTS
      (One—Year  Eligibility  Requirement)
Test
Number

11

12

15

16

19

20


97


98


99


100


109


110


Test
Description
Pass All:
CF/TL > .1
TL/NW < 1.5
CF/TL > . 1
TL/NW < 2.0
NI/TA > .04
TL/NW < 1.5
NI/TA > .04
TL/NW < 2.0
NI/TA > .02
TL/NW < 1.5
NI/TA > .02
TL/NW < 2.0
2 out of 3:
CF/TL > .1
TL/NW < 1.5
CURRAT > 2.0
CF/TL > .1
TL/NW < 2.0
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.5
CURRAT > 1.5
CF/TL > .1
TL/NW < 2.0
CURRAT > 1.5
CF/TL > .1
TL/NW < 1.5
QRAT > 1.2
CF/TL > .1
TL/NW < 2.0
QRAT > 1.2

EP

6.1

6.8

6.3

7.3

6.5

7.4


9.7


9.3


10.1


10.1


8.3


8.4



*NB

77

81

66

72

77

82


86


93


91


96


82


87



\BU

42

69

31

46

46

81


42


69


42


69


42


69


                      71-14

-------
TABLE VI-8 (Continued)
Test
Number
119


120



123


124


125


126


127



130


133


134


135


Test
Description
CF/TL > .1
TL/NW < 1.5
QRAT > 1.0
CF/TL > . 1
TL/NW < .2.0
QRAT > 1.0
Pass All:
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.5
NFA/TA > . 3
CF/TL > .1
TL/NW < 2.0
NFA/TA > . 25
CF/TL > .1
TL/NW < 1.5
NFA/TA > .25
CF/TL > .1
TL/NW < 2.0
NFA/TA > . 2
CF/TL > .1
TL/NW < 2.0
2 out of 3:
NFA/TA > . 3
CF/TL > .2
TL/NW < 1.5
HFA/TA > . 3
CF/TL > .15
TL/NW < 1.5
NEA/TA > . 3
CF/TL > . 1
TL/NW < 1.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2
EP
9.5


9.1



3.3


3.1


2.9


3.6


3.8



7.1


6.8


4.6


5.6


SB
88


92



55


59


62


66


75



73


80


79


83


SBU
42


69



42


69


42


69


69



46


46


77


77


             "1-15

-------
TABLE VI-S (Concluded)
Test
Number
136


137


146



149



150



Test
Description
NFA/TA > .3
CF/TL > .1
TL/NW < 1.5
NFA/TA > . 3
CF/TL > .1
TL/NW < 2.0
NFA/TA > . 2
CF/TL > .1
CURRAT > 2.0
3 out of 4:
NFA/TA > . 3
CF/TL > .1
CURRAT > 2.0
TL/NW < 1.5
NFA/TA > . 3
CF/TL > .1
CURRAT > 2.0
TL/NW < 2.0

EP
7.9


9.1


9.9



6.0



5.5




\B
89


92


95



87



81




NBU
81


88


77



69



42



       VI-16

-------
                TABLE VI-9
OVERALL EFFECTIVENESS OF UTILITY-APPLICABLE
              FINANCIAL TESTS
   (Three-Year Eligibility Requirement)
Test
Number

11

12

15

16

19

20


97


98


99


100


109


110


Test
Description
Pass All:
CF/TL > .1
TL/NW < 1.5
CF/TL > .1
TL/NW < 2.0
NI/TA > .04 .
TL/NW < 1.5
NI/TA > .04
TL/NW < 2.0
NI/TA > .02
TL/NW < 1.5
NI/TA > .02
TL/NW < 2.0
2 out of 3:
CF/TL > .1
TL/NW < 1.5
CURRAT > 2.0
CF/TL > .1
TL/NW < 2.0
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.5
CURRAT > 1.5
CF/TL > .1
TL/NW < 2.0
CURRAT > 1.5
CF/TL > .1
TL/NW < 1.5
QRAT > 1.2
CF/TL > .1
TL/NW < 2.0
QRAT > 1.2

%

3.7

5.2

4.7

6.1

4.9

6.6


7.7


8.6


8.2


9.7


6.5


7.4



^B

70

76

56

60

69

75


75


82


80


84


72


78



\BU

35

58

<20

23

38

77


35


58


35


58


35


58


                      VI-17

-------
TABLE VI-9 (Continued)
1
Test
Number
119


120



123


124


125


126


127



130


133


134


135



Test
Description
CF/TL > .1
TL/NW < 1.5
QRAT > 1.0
CF/TL > .1
TL/NW < 2.0
QRAT > 1.0
Pass All:
NFA/TA > . 3
CF/TL > .1
TL/NW < 1.5
NFA/TA > . 3
CF/TL > .1
TL/NW < 2.0
NFA/TA > .25
CF/TL > .1
TL/NW < 1.5
NFA/TA > .25
CF/TL > .1
TL/NW < 2.0
NFA/TA > . 2
CF/TL > .1
TL/NW < 2.0
2 out of 3:
NFA/TA > .3
CF/TL > .2
TL/NW < 1.5
NFA/TA > .3
CF/TL > .15
TL/NW < 1.5
NEA/TA > . 3
CF/TL > . 1
TL/NW < 1.0
NFA/TA > . 3
CF/TL > . 1
TL/NW < 1.2

EP
8.0


8.2



1.6


1.5


1.4


2.2


1.9



4.9


4.7


2.9


4.1



V
78


83



49


53


56


59


67



64


72


72


77



\BU
35


58



35


58


35


58


58



38


42


62


62


              VI-18

-------
TABLE VI-9 (Concluded)
Test
Number
136


137


146



149



150



Test
Description
NFA/TA > .3
CF/TL > .1
TL/NW < 1.5
NFA/TA > .3
CF/TL > .1
TL/NW < 2.0
NFA/TA > . 2
CF/TL > .1
CURRAT > 2.0
3 out of 4:
NFA/TA > . 3
CF/TL > .1
CURRAT > 2.0
TL/NW < 1.5
NFA/TA > .3 .
CF/TL > .1
CURRAT > 2.0
TL/NW < 2.0

EP
5.4


7.3


9.5



5.0



4.1




\B
82


86


83



74



71




\BU
65


81


62



58



35



           VI-19

-------
                                                   FIGURE VI-1
                       PERFORMANCE CURVE FOR TESTS WITH ONE-YEAR ELIGIBILITY REQUIREMENTS
                                    WITH AND WITHOUT UTILITY APPLICABLE TESTS
NB
                                                                         Key:  Dark  Line -  Utility Applicable
                                                                                            Tests
                                                                               Light Line - One-Year Tests
    40-

    35-
O A
  0
.4
!5  2.'0   2*.5 3.'o
                                              !o  4.*5 5.'o  S.'s  6.'o  6,'s   ?!o  7.'s  s!o  8.*5  9.'o  9 .'s  lo'.O id. 5
                                                         c.p

-------
                                                   TABLE VI-10

                          AVERAGE RATIOS OF HAZARDOUS WASTE MANAGEMENT FIRMS,  1971-1975

                                                           1971        1972        1973        1974        1975

       Current Ratio

         a.  Firms whose main objective is hazardous       1.58        1.77        1.42        1.33        1.28
             waste management
         b.  Corporations having divisions or sites        1.28        1.61        1.81        2.04        1.83
             engaged in hazardous waste management

       Total Debt/Equity

         a.  Firms whose main objective is hazardous       1.20        1.19        1.18        1.60        1.64
             waste management
         b.  Corporations having divisions or sites        1.37        1.66        2.43        1.96         NA
M            engaged in hazardous waste management

*"      Profits/Total Debt

         a.  Firms whose main objective is hazardous       NA          .096        .102        .134        .182
             waste management
         b.  Corporations having divisions or sites        .127        .098        .056        .076        .075
             engaged in hazardous waste management

       Net Fixed Assets/Total Assets

         a.  Firms whose main objective is hazardous       .65         .64         .62         .555        .60
             waste management
         b.  Corporations having divisions or sites        .62         .57         .63         .60         .62
             engaged in hazardous waste management


       SOURCE:  Foster D. Snell,  1976.

-------
respect  to  the ability of  the  average  firm in the industry  to  meet the

possible  cash flow/total  liability requirements.    This  ability  would

depend  in at least  some  years on  total  depreciation, amortization and

depletion allowance.

     Thus,  firms in  the hazardous  waste management  industry  could  be

expected  to  have difficulty meeting many possible financial  tests  even

without  a $10 million in net worth requirement.  However,  this  reflects

the financial weakness of this industry  during  this  period  rather  than

some unique financial  feature of the industry.  The study from which the

ratios in Table VI-10  are drawn states:

    Since  1973,  long  term  debt   has   remained   stable  for the
    industry  as  financial   institutions  have  been  reluctant  to
    invest  capital  in  the  industry  because:    hazardous  waste
    management  firms  in  general  did  not  have  strong financial
    positions, and  the growth of  the industry was deemed risky  in
    future periods.   Short  debt financing has increased from $3.2
    million in 1971  to $7.2 million in 1975.  This high cost debt
    financing additionally  points  out  the current problems of the
    industry  in  obtaining lower cost  long-term  capital.    (Foster
    D. Snell, 1976).

     The  share of hazardous waste facilities owned  by hazardous waste

management firms  with over  $10 million  in net worth and  independently

audited  is  small.   Only 26 facilities   are  owned by  the  four largest

hazardous waste  management  firms  which are  independently  audited.   The

next eight to nine largest firms are all  privately held and are  unlikely

to be  independently audited.   The majority of off-site  facilities are

owned by relatively  small firms with less than $10 million in net worth

(Booz, Allen  & Hamilton  and Putnam, Hayes  &  Bartlett, 1980).   The  26

facilities which  are  owned  by  large,  independently  audited  firms   thus

represent only about  1 percent of the 2,500  facilities  owned by firms

eligible for  a financial test.

                                  71-22

-------
     A careful  examination of the  balance  sheets of  two  of  the larger




firms in the industry which own 12 of the 26 facilities, Browning Ferris




Industries  (BFI)  and SCA  Services,  Inc. showed  their  performance  with




respect to  the relevent ratios as  summarized  in Table VI-11.   BFI would




pass  13  of the  16  financial tests  on  the  performance curve  for tests




with one-year or three-year elibibility requirements.  SCA would pass 12




of these 16 tests.
                                 VI-23

-------
                            TABLE VI-11


            RATIOS FOR BROWNING FERRIS INDUSTRIES  (BFI)

                      AND SCA  SERVICES,  INC.
Firm Name
BFI


SCA

Year of Data
(x-1) 1978
(x-2) 1977
(x-3) 1976
(x-1) 1978
(x-2) 1977
CF/TL
17.5
15.5
15.7
15.4
14.9
TL/NW
1.13
1.10
1.12
1.52
1.55
NFA/TA
^
P
P
P
P
CURRAT
1.60
F*'
1.64
1.65
F
QRAT
P
P
P
P
P
NI/TA
P
P
P
4.0
3.0
— P = Passes all tests using  this variable.

2/
— F = Fails all tests using this variable.
                               VI-24

-------
VII.  BOND RATINGS




     Several commenters on the May 19, 1980 reproposed financial  respon-




sibility  regulations  suggested  that corporate  bond ratings  should be




used as  an alternative or  substitute for the  proposed  financial  test.




The Agency therefore analyzed  the  possibility of  using corporate  bond




ratings as an assurance of financial responsibility.




     Bonds  are  a  form  of long-term debt.   They may be  unsecured, or




secured by collateral such as equipment, marketable  securities, or  fixed




assets.   Certain  bonds  include specific promises  concerning the use of




special sinking  funds.   Repayment  of other  categories  of  bonds may be




subordinated to repayment of other debt instruments.




     Bond  ratings  provide an  appraisal  of  the ability of firms to  repay




these  long-term  debts.   They  are  used  extensively  by  the  financial




community  for  the purpose of  providing  a long-term credit risk evalu-




ation  of   the  relative likelihood  of  timely payment  of  interest and




repayment  of  principal of   specific  bond  instruments.    Many  banks,




including  large  ones,  rely  almost exclusively on rating service evalu-




ations  of bonds  and  do not  have  internal  groups  which  independently




evaluate  the riskiness  of  their  bond  portfolios  (Kaplan  and  Urwitz,




1979).   Because  bond  ratings measure  the ability  of  firms  to  meet a




long-term  debt,  they are  better adapted  to an  analysis  of  long-term




financial responsibility than are ratings  of short-term credit and  trade




credit.




     The Agency did  not attempt to  evaluate  the  performance of  ratings




of  specific  types  of  bonds.    Rather,  it  considered  the  ratings
                                VII-1

-------
performance of all categories of long-term debt that were (1) identified




as bonds by either  of  the two major bond ratings services and (2) rated




by either of them.




     The two major ratings services are Moody's Investors Services, Inc.




(Moody's) and  Standard  and  Poor's  Corporation (Standard  and  Poor's).




Each  has  long  experience  in  rating  corporate  bonds.    Moody's  and




Standard and Poor's rate almost  all corporate bond issues of $ 5 million




and over  in principal amount with or without  request  from the issuer.




The bonds of certain issuers  (including  newly created companies in most




cases) are  not  rated,  and  ratings  of  other issuances are  carried out




only upon request and are not  published except upon  permission  of the



issuer.




     Both major ratings  services provide nine basic ratings of corporate




bonds.  Figure VII-1 provides a  listing  of  the ratings given by Moody's




and the  interpretation of each rating as suggested  by  Moody's.   Figure




VII-2  provides  the  ratings and  their interpretations used  by  Standard




and Poor's.    The ratings classifications  are highly  qualitative,  and




neither rating service suggests any specific quantitative equivalent for



the ratings assigned.




     The top four ratings of both services are assigned to bonds consid-



ered to be of investment grade,  while the remaining ratings are assigned




to bonds  that  are considered to  be speculative.  Ratings  of  each bond




issuance are  reviewed  periodically  and  may be upgraded  or downgraded.




Table VII-1  provides a distribution of corporate bond ratings issued by
                                VII-2

-------
                           Aaa
     Bonds which  are  rated  Aaa are judged to be of the hest
quality.   They  carry  the  smallest degree of investment risk
and are generally referred  to as "gilt edge."  Interest payments
are protected by  a large  or by an exceptionally stable margin
and principal is  secure.  While the various protective elements
are likely to change,  such  changes as can be visualized are
must unlikely to  impair  the fundamentally strong position
of such issues.
                             Aa

     Bonds which are rated Aa are Judged to be of high quality
by all standards.  Together with the Aaa group they comprise
what are generally known as high grade bonds.  They are
rated lower than the best bonds because margins of protec-
tion may not be as large as in Aaa  securities or fluctuation
or protective elements may be of greater amplitude or  there
may be other elements present which make the long term
risks appear somewhat larger than in Aaa securities.
any great length of  time.   Such bonds lack outstanding invest-
ment characteristics and in fact have speculative characteristics
as well.
                                 Ba

     Bonds which are rated Ba are judged to have speculative
elements;  their future cannot be considered us well assured.
Often the  protection of interest and principal payments may
be very moderate and thereby not well safeguarded during both
good and bad times over the future.  Uncertainty of position
characterizes bonds in this class.
                                  B

     Bonds which are rated B generally lack characteristics of the
desirable investment.  Assurance of interest and principal payments
or of maintenance of other terms of the contract over any long
period of time may be small.
     Bonds which are rated A possess  many  favorable  invest-
ment attributes and are to be considered as upper  medium
grade obligations.   Factors giving  security to  principal  and
interest are considered adequate but  elements may  be present
which suggest a susceptibility to impairment sometime in  the
future.
                            Baa

     Bonds which are rated Baa are  considered  as  medium
grade obligations, i.e., they are neither  highly  protected  nor
poorly secured.  Interest payments  and principal  security ap-
pear adequate for the present but certain  protective elements
may be lacking or may be characteristically  unreliable over
                                 Caa

     Bonds which are rated Caa are of pour standing.  Such issues
may be in default or there may be present elements of danger with
respect to principal or interest.


                                  Ca

     Bonds which are rated Ca represent obligations which are
speculative in a high degree.  Such Issues arc often in default
or have other marked shortcomings.


                                   C

     Bonds which are rated C are the lowest rated class of bonds and
issues so rated can be regarded as having extremely poor prospects of
ever attaining any real investment standing.
                                                             FIGURE VII-1

                                                   KEY  TO  MOODY'S  BOND  RATINGS

-------
H
M
A Standard & Poor's corporate or municipal bond rating is a current assessment of
the creditworlhiness of an obligor with respect to a specific ilcbi obligation.  I'll is
assessment may lake into consideration obligors sncli as guarantors, insurers, or
lessees.
   The  bund rating is  not a recommendation to purchase, sell, or hold a security
inasmuch as it  does not comment as to market price or suitability for a particular
investor.
   The ratings are based on current information furnished by the issuer or obtained
by Standard £ I'oor's from other sources we consider reliable. We do not perform an
audit in connection with any rating and may, on occasion, rely on unaudited financial
information. The ratings may  be changed, suspended, or withdrawn as a result of
changes in, or unavailability of, such information, or for other reasons.
   The ratings arc based, in varying degrees, on the following considerations:

        I. Ukelihood of default—capacity and willingness of the obligor as to
          the timely payment of interest and repayment of principal in accor-
          dance with the terms of the obligation

      II. Nature of and provisions of the obligation.

      HI. Protection afforded by, and relative position of, the obligation in the
          event of bankruptcy, reorganization, or other arrangement under
          the laws of bankruptcy and other laws a flea ing creditor's rights

       Bonds rated AAA have the highest rating assigned by Standard & I'oor's to
AAA  a debt obligation. Capacity  to pay interest and repay  principal is extremely
       strong.

       Bonds rated AA have a very strong capacity to pay interest and repay prin-
       cipal and differ from the highest-rated issues only in a small degree.

       Bonds rated A have a strong capacity to pay interest and repay principal,
       although  they  are  somewhat  more  susceptible  to the adverse  effects of
       changes in circumstances and economic conditions than  bonds in higher-
       rated categories.

       Bonds rated BBB are regarded as having an adequate capacity to pay interest
       and repay principal. Although they  normally  exhibit adequate protection
       parameters, adverse economic  conditions or changing circumstances  are
       more likely to lead to a  weakened capacity to pay interest and repay principal
       for bonds in this category than for bonds in higher-rated categories.
              AA
                                                                                            BB
                                                                                             B

                                                                                          CCC
                                                                                            CC

                                                                                             C

                                                                                             D
                                                                                            NR
             BBB
Bonds rated  UB, B. CCC. and CC arc regarded, on balance, as predomj.
nanlly s|>eculalive with respect to capacity to pay interest and repay principal
in accordance with the terms of the obligation. UU indicates (lie lowest degree
of S|x.-cnlation and CC the highest degree of s|x:culaiion. While such bond*
will likely have some quality  and protective characteristics,  these are out-
weighed by large uncertainties or major risk exposures to adverse conditions.

The rating C is reserved for income lionds on which no interest is being paid.
Bonds rated O are in default, and payment of interest and/or repayment of
principal is in arrears.
Plus (+) or Minus (-): The ratings from  AA to UB may lie modified by the
addition of a plus or minus sign to show relative standing  within the major
rating categories.
Provisional Ratings: The letter "p" indicates that the rating is provisional.
A provisional rating assumes the  successful completion oi  the project
financed by the bonds being rated and indicates that payment of debt service
requirements is largely or entirely dependent  upon the successful and lime!)
completion of the project. This rating, however, while addressing credit qual-
ity subsequent to completion of the project, makes no comment on the like-
lihood of, or the risk of default UJKIII failure of, such completion. The inves-
tor should exercise his or her  own judgment  with respect to  such likelihood
and risk.
Indicates that no rating has been requested,  thai there is insufficient infor-
mation on which to base a rating, or that  Standard & Poor's  does not rate a
particular tyjx: of obligation as a matter of policy.

Debt obligations of issuers outside the United Stales and  its territories are
rated on the same basis  as domestic corporate and  municipal  issues. The
ratings measure  the credilworlhiness of  the  obligor, but  do not take into
account currency exchange and other unccriaiiuies.

Bond Investment Quality Standards: Under present commercial bank reg-
ulations issued by the Comptroller of the Currency, lx>nds rated in the top
four categories (AAA. AA. A, BBB, commonly known as "investment grade"
ratings) are generally regarded as eligible for bank investment. In addition,
the legal investment laws of various stales impose certain ratings or  other
standards for obligations eligible  for investment by savings banks, trust com-
 panies, insurance companies, and fiduciaries generally.
                                                                   FIGURE VII-2

                                            STANDARD &  POOR'S CORPORATE  RATING DEFINITIONS

-------
                                TABLE VII-1
               PERCENTAGE DISTRIBUTION OF PUBLICLY OFFERED
                        STRAIGHT CORPORATE BONDS
Amount
in billions
Aaa
A
Baa
Not
& Aa

& below
rated
1966
7.8
48%
19
27
6
1967
14.8
42%
19
30
9
1968
10.9
48%
17
23
12
1969
9.5
53%
27
20
0
1970
23.1
50%
38
11
1
1971
21.3
53%
32
11
1
1972
12.3
60%
28
8
4
1973
12.3
61%
29
5
5
1974
24.8
64%
29
6
1
1975-
31.1
57%
36
7
3
                                                                         I/
— 11 months.
SOURCE:  Backer and Gosman  (1978).
                                  VII-5

-------
Moody1 s for  recent years.  Table VII-2  shows  the number of outstanding

bond issues rated  by Moody's at a point in time in July 1981.  According

to  this  Table,  86  percent  of  the  outstanding bond  issues  rated by

Moody's as  of  that  time would  be acceptable  for satisfying  the  bond

rating portion of  the financial test.

     Both ratings  services  insist  that  a bond  rating is  based  on the

specific  bond  instrument in question, and  does not  imply  a rating for

other instruments  of the  firm  or a  rating of the firm as a whole.  This

is due,  in  part,  to the  fact  that different bonds are  secured by dif-

ferent types and amounts  of collateral,  and  by  different types of other

so-called "structural  credit  supports" such as  third party guarantees,

segregated cash  flow,  or insurance.   However,  the rating  agencies do

carry out an  analysis  of the  situation  of both  the industry and of the

company within the industry and a financial analysis of the company, and

these analyses  also affect  the rating  assigned to  the specific bond

instrument.

     According to  Standard and Poor's  Ratings Guide,  ratings are based,

to varying degrees, on the following considerations:

      I.   Likelihood of  default—capacity and willingness  of  the
           obligor  as   to  the timely payment  of  interest  and
           repayment of  principal in  accordance  with  the terms of
           the obligation

     II.   Nature of and provisions of the obligation

    III.   Protection afforded by,  and relative position of,  the
           obligation in  the event  of bankruptcy,  reorganization,
           or other arrangement under the laws  of  bankruptcy  and
           other laws affecting creditor's rights

                           (Standard and  Poor's  Ratings Guide,  1979)
                                VII-6

-------
                            TABLE VII-2

                     OUTSTANDING BOND ISSUES*
                             JULY 1981
Moody 's Ratings
Investment Grade
Aaa
Aa
A
Baa
Total Investment Grade
Below Investment Grade
Ba
B
Caa
Ca
C
Total Below Investment Grade
Number
of Issues

750
1140
2029
1012
4931

339
320
95
38
10
802
Percentage
of Issues

13
20
35
18
86

6
6
2
1
0
15
*Includes:  Industrials, Utilities, Transportation, Banks and Finance
 Companies.

 Note: Totals may exceed 1013% due. to rounding error.

 SOURCE:   Ron Lewis of Moody's Corporate Bond Department.
                              VII-7

-------
     The  wide  scope  of  bond  rating  agency  analysis  can be  better

appreciated from  the  description given  by  Brenton W.  Harris,  a former

President of  Standard and  Poor's,  when listing  the factors used  as a

basis for assigning bond ratings:

    1.   Issuing  documents:  ...  In  determining a  rating,  the
         indenture  is  far  less  important  than  the  company's
         earning power, financial  resources,  and property protec-
         tion.  This  is not to  say, however,  that  the indenture
         does not have a great bearing on a bond rating...

    2.   Earnings:  the past record and foreseeable potential are,
         in most cases,  the single most important factor in credit
         rating.   High  levels  of  earnings  frequently  preclude
         liquidity  problems  because  access  to  short-term  cash
         needs can be  readily accommodated.   Remembering that bond
         ratings  turn  on   timely   repayment  of  principal  and
         interest, strong  cash flows  generated by high  and con-
         tinued earnings, combined with  an  adequate depreciation,
         depletion,  and  amortization  policy  where  applicable,
         contribute a healthy  plus factor to  the determination of
         a bond rating...

    3.   Asset  protection:   Asset  protection  generally  is  more
         important  as  a  long-term  consideration  than  as  one
         influencing  immediate  liquidity.   The analysis  here  is
         primarily statistical and,  hence,  highly objective.   Of
         primary interest are the  ratio  of  its  working  capital to
         its debt; the  ratio of its  debt  to  its  equity;  and the
         ratio of its  total net  tangible assets to its  debt.  The
         relative importance of  these  major ratios  depends on the
         type of industry the company is in...

    4.   Management:    Evaluating management  is one  of  the  most
         difficullt chores  a rating  agency  faces.   But  management
         is  one of the  most  important  facets  of  a  successful
         operation.   It is  my opinion  that  the  single most impor-
         tant reason for  the failure of  the Penn  Central  was the
         inability of  its management to deal with its problems...

    5.   Financial resources  are,  of  course,  the largest single
         area in which liquidity has a direct impact  on long-term
         debt  rating.   In looking at the financial  resources  of  a.
         company,  we are  concerned not only with  a company's cash
         position but  also with its  ability to  obtain cash.   This
         area of financial resources, which  constitutes  one of the
         five   fundamental  areas  of  investigation  to   determine


                                VII-8

-------
         long-term  debt  rating,   is   the  test  of  liquidity   ...
         Briefly stated,  the financial resources  we consider are
         those  alternative  sources of  borrowing that  a firm may
         use  to raise  cash for either  long- or  short-term debt
         repayment.   They include the amount  of cash reserves on
         hand   including   salable  receivables;   the  short-term
         borrowing potential, particularly bank  lines; the ability
         of. the company to  tap  the long-term debt market, partic-
         ularly at  the  time of its choosing;  the  ease with which
         the  company  could  sell stock; and  the  potential sale of
         assets—obviously the weakest alternative.

                               (Quoted in Backer and Gosman, 1978)

     The Agency is  satisfied that the actual  establishment  of a rating

is the  result of  the informed  judgment of  trained  analysts  based  on a

detailed study  of  all  of the above  factors,  and does  not  involve the

mechanical application  of formulae.   Bond ratings are  therefore  a  dif-

ferent  type  of  analysis  of  financial liability than  a  financial test.

Spokesmen for the  ratings  services emphasize the point that bond ratings

are based on  expert judgment (Backer and Gosman,  1978).   Although  some

investigators have questioned this assertion (Ross 1976, Sherwood 1976),

arguing  that  bond ratings  services  rely on explicit cutoff  points  of

particular financial  ratios,  a  thorough recent  study of  bond  and other

financial ratings  procedures prepared  for   the  National  Association  of

Accountants reported, with respect to the importance of financial ratios

in credit rating evaluations, that:

    All interviewees  stated  the belief  that  financial ratios were
    of moderate to strong importance in their credit rating evalu-
    ations.    However, almost  all  those interviewed cautioned that
    such financial measures were not the only thing of importance;
    additional  considerations    frequently   mentioned   included
    quality   of  management,   future   for  product,   and  general
    economic conditions.  (Backer  and Gosman, 1978).
                                VII-9

-------
     The  same  study  conducted a statistical analysis to determine which




ratios  seemed  to be most  important by comparing  a  group  of bonds that




had  not been  downgraded  and group  of bonds  that  had  been downgraded.




The  most  statistically significant  ratio  was  found to  be  cash flow to




long-term debt.   However,  the statistical  results also indicated that a




number  of other  ratios  could also  be of  some  importance:   return on




sales,  return  on total assets, return on  tangible net worth, long-term




debt to capitalization, net tangible assets to long-term debt, long-term




debt  to property  and  equipment,  and  cash  flow to  total  debt.   This




statistical examination of downgraded  bonds compared to bonds that were



not  downgraded would probably have  revealed a specific formula had one




existed.  No such specific formula was found.




     The  Agency, in order  to  assess  the actual  performance  of  bond




ratings in  predicting  default,  first  conducted  a  literature search for




previous studies of  performance.  A variety of sources suggest that bond




ratings can  be used as  a reasonably  good indicator of the  quality of



corporate debt.   (See Lev,  1974;  Fraine and  Mills,  1961;  and Fiedler,




1971.)  The  only sources  for detailed statistical  data supporting this



conclusion,  however, are  two studies by  W.B.  Hickman, Corporate Bond




Quality and Investor  Experience  (1958),  and  Statistical   Measures  of




Corporate Bond Financing Since 1900 (1960).



     Although  Hickman1s  work is  somewhat dated,  covering  the  period




1900-1943,  it   does  examine  a  period  of  economic  volatility  which



provides  a  good  test of  the results of ratings services.   The ratings




used are  those assigned  at  the  time of issue.   Thus,  if  the  bond was







                                VII-10

-------
downgraded prior to default but after issue, this would not be reflected

in these ratings.  Given the volatility of the economy for the period in

question, and the fact that changes in ratings are not accounted for use

of these  data results in  a  strict judgment of bond  ratings.   In addi-

tion,  statistical  analysis  of  the   performance of bond  ratings  is

difficult for recent  periods because  so  few firms with rated bonds have

gone  bankrupt that creation  of a  useful  sample is  almost  impossible.

The Agency noted,  for example, that none  of the  firms in its  sample of

bankrupt firms between 1966 and 1979  had an investment grade rated bond

issuance.

     In a study incorporating Hickman's findings, T.B. Atkinson compared

prewar and postwar (through 1965)  corporate  bond  quality.  He  concluded

that:

    U.S.  corporate  bonds  defaulting  in  the postwar  period  (from
    1945  through  1965)  averaged  less  than  0.1  percent of  the
    volume of outstanding, or  about  one-half billion dollars in
    all.  This compares  with 1.7  percent of the  outstanding  bonds
    which defaulted  during  1900-43.   The postwar defaults  were
    concentrated in bonds  of the  railroad  industry,  many of  which
    had been outstanding before 1920 or had been refundings of  the
    original bonds issued prior to that year.  (Atkinson, 1967)

It should  be noted  that  these percentages  of outstanding  bonds  which

defaulted include bonds of investment  and speculative quality.

     In  order  to  analyze  the  effectiveness  of  agency  ratings  in

predicting defaults,  the Agency applied  a general procedure similar to

that used to calculate the effectiveness  (E) of various financial tests.

Hickman  reported  in  his  study the percentage  of the  total  value  of

outstanding bonds of a particular rating  defaulted upon by firms  with
                                VII-11

-------
these  bonds  and  the  percentage  of   the  total  value  not  defaulted.




Although  there is  no direct  evidence correlation  between the  default




rates  of  bonds and firms,  it  was assumed that as a  rough  approximation




the  accuracy  of   predicting  defaults  on  bonds  is  equivalent  to  the




accuracy  in identifying  defaulting firms.   Therefore,  the  percentages




reported  by Hickraan were used in  columns 1 and 2 of Table VII-3 as  the




equivalents  of  the  Mp  and  A.™  measures obtained  when  a  particular




financial test  is  applied to the  bankrupt and to the  non-bankrupt  firm




samples developed  by  the Agency.   Using two alternative estimates of  the




baseline  failure  rate, effectiveness  measures (E)  were calculated  for




each grade  of  bond ratings in the top four categories.  The  results of




this analysis  are  given  in Table  VII-3.   The first  assumption is  the




baseline  of 22 per  10,000  used  earlier  in  this  report.  The second




reflects the possibility that the  baseline failure rate for the  kinds of




firms  which would  have  rated  bonds might  be  as  low  as 11 per  10,000.




Public  bond   ratings  are   generally   given  to   bond   issuances   from




substantial firms.  A significant  number  of large public bond  issues  are




not  rated  at all.   This category  of  unrated  public debt  issues has  a



much  higher failure  rate  than  the   failure  rate  of   those  which  are




rated.  The unrated issues are  excluded  from  the entire rating system,




and from the percentages shown in  the  Table.




     As can be noted  from the  Table, the results  are quite different




depending upon whether one considers  the ratings for  public  utilities,




industrials, and transportation  companies combined,  or only the ratings




for industrials (and  public  utilities, though  they are not shown).   One




important reason  why bond  ratings, taken  as  a whole,  did not perform



                                   VII-12

-------
                                                    TABLE VI1-3


                                                  : OF AGLNCY  UAT1NGS  (1900-1943)

-------
better  during  this period is  that  transportation company bonds, speci-




fically  those  of  railroads,  were  overrated  and  suffered  very  high




failure rates, even for firms  in the highest rating categories.




     The  column  of Table VII-3  labeled  "Percentage  of  Non-Defaulting




Bonds  With  Rating"  shows  the  proportion  of  non-defaulting  bonds  by




dollar value included in  each  set  of  rating categories.  The investment




grade ratings  I-IV  include the vast majority  of  all rated bonds.  They




include 94  percent  of all rated corporate  bonds and  93  percent of all




rated industrial  bonds.   Thus, about  6  percent of corporate bonds and 7




percent of  industrial bonds  are eliminated by insisting  on a rating of



investment  grade  or better.   This means that over  94  percent  of the




dollar value of all corporate bonds rated were rated as investment grade




or above.



     The  top  two  classes of  investment grade   ratings  are  much  more




restrictive.   For all bonds,  only 46  percent had  one of  the  top two




ratings,  and only 19  percent had the top rating.  The Agency therefore



concluded that  the  top  two  bond ratings  are  roughly  equivalent to the




effectiveness  of  a relatively  stringent  financial  test,  assuming  that



the baseline failure rate is approximately 11 per  10,000 for the firms



likely  to  use  the test rather than 22 per  10,000.   A rating  of IV or




above, in contrast, is about equivalent  to the effectiveness  of a less




stringent test.



     The Agency also  performed a special study to  determine the effec-




tiveness  of  bond  ratings as  a  means  of financial responsibility  for




utilities  in  particular.    A  sample of  113  utilities  was  drawn  from







                                VII-14

-------
Moody's  Public Utilities  Manual.   All  were large  (greater  than $10




million in net  worth),  and were either electric  or gas utilities.  The




mix of bond ratings for these firms was as follows:









                              TABLE VI1-4




                          UTILITY BOND RATINGS
RATING
Aaa
Aa
A
Baa
Unrated
NUMBER OF
UTILITIES
4
22
50
25
12
113
PERCENTAGE
OF TOTAL
3.5
19.4
44.2
22.1
10.7
100.0
PERCENTAG
RATED UTIL
3.9
21.7
49.6
24.8
—
100.
     As can be seen, requiring that firms have Aaa or Aa ratings to pass




the test would allow only  23-26  percent  of  utilities to pass this test.




If utilities  with A  ratings  are  included,  67-75 percent  of  utilities




would be able to  pass,  and a  requirement that a firm have an investment




grade  bond rating  would allow  about 89 percent or  more  to  pass  the




test.   However,  because the  sample  obtained did not  contain  any util-




ities with a  rating lower  than  investment  grade,  the Agency  could not




determine the precise effect of requiring an investment grade rating.
                                VII-15

-------
VIII.  RESULTS FOR ALL RATIO TESTS




     This Section presents the Tables showing  the  performance  of  all of




the  ratio  tests with  respect  to the evaluation  criteria described  in




Section IV.A for the primary sample.  Tables  VIII-1  through  VIII-4 give




results  for  tests  with  a one-year  eligibility  requirement and  Tables




VIII-5  through  VIII-8  give  results   for   tests  with  a   three-year




eligibility requirement.
                               VIII-1

-------
            TABLE VIII-1
  PERFORMANCE OF TWO-RATIO TESTS
(One-Year Eligibility Requirement)
           (Tests 1-56)
Test
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Test
Variables
CF/TL > .2
TL/NW < 1.0
CF/TL > .2
TL/NW < 1.2
CF/TL > .2
TL/NW < 1.5
CF/TL > .2
TL/NW < 2.0
CF/TL > .15
TL/NW < 1.0
•CF/TL > .15
TL/NW < 1.2
CF/TL > .15
TL/NW < 1.5
CF/TL > .15
TL/NW < 2.0
CF/TL > .1
TL/NW < 1.0
CF/TL > .1
TL/NW < 1.2
CF/TL > .1
TL/NW < 1.5
CF/TL > .1
TL/NW < 2.0
NI/TA > .04
TL/NW < 1.0
NI/TA > .04
TL/NW < 1.2
NI/TA > .04
TL/NW < 1.5
*B
4.7
4.7
9.5
9.5
4.7
4.7
9.5
7.1
4.7
4.7
11.9
11.9
4.7
4.7
11.9
KW
11.9
14.2
16.6
16.6
14.2
16.6
19.0
21.4
16.6
21.4
30.9
38.0
14.2
19.0
26.1
^B
38
41
41
41
52
59
62
63
58
71
77
81
54
64
66
C
91.7*
90.5*
87.0
87.0
90.5*
89.3
85.8
85.7
89.3
87.0
78.6
75.0
90.5*
88.1
81.0
D
26.1
26.8
24.4
24.4
37.8
42.4
43.0
41.6.
41.4
49.6*
46.1
43.0
39.8
45.0
39.9
EP
4.8*
5.1*
7.0.
7.0
4.0*
4.0*
5.1*
5.0*
4.1*
4.1*
6.1
6.8
3.9*
4.1*
6.3
                  VIII-2

-------
TABLE VIII-1 (Continued)
Test
Number
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Test
Variables
NI/TA > .04
TL/NW < 2.0
NI/TA > .02
TL/NW < 1.0
NI/TA > .02
TL/NW < 1.2
NI/TA > .02
TL/NW < 1.5
NI/TA > .02
TL/NW < 2.0
CF/TL > .2
CURRAT > 2.0
CF/TL > .15
CURRAT > 2.0
CF/TL > .1
CURRAT > 2.0
CF/TL > .2
CURRAT > 1.5
CF/TL > .15
CURRAT > 1.5
CF/TL > . 1
CURRAT > 1.5
CF/TL > .2
QRAT > 1.2
CF/TL > .15
QRAT > 1.2
CF/TL > .1
QRAT > 1.2
CF/TL > .2
QRAT > 1.0
CF/TL > .15
QRAT > 1.0
*B
11.9
4.7
4.7
11.9
11.9
4.7
4.7
4.7
4.7
7.1
11.9
4.7
4.7
4.7
4.7
4.7
*v
35.7
16.6
21.4
33.3
42.8
9.5
14.2
19.0
9.5
19.0
33.3
9.5
11.9
16.6
9.5
14.2
^B
72
58
71
77
82
35
54
63
41
62
81
28
42
50
44
55
C
76.2
89.3
87.0
77.4
72.6
92.9*
90.5*
88.1
92.9*
87.0
77.4
92.9*
91.7*
89.4
92.9*
90.5*
D
36.3
41.4
49.6*
43.7
39.2
25.5
39.8
44.0
31.5
43.0
47.7*
18.5
30.1
33.4
34.5
40.8
*P
7.3
4.1*
4.1*
6.5
7.4
4.5*
3.9*
4.2*
3.8*
4.6*
6.1
5.6*
4.3*
4.7*
3.6*
3.8*
      VIII-3

-------
TABLE VIII-1 (Continued)
Test
Number
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Test
Variables
CF/TL > .1
QRAT > 1.0
TL/NW < 1.0
CURRAT > 2.0
TL/NW < 1.2
CURRAT > 2.0
TL/NW < 1.5
CURRAT > 2.0
TL/NW < 2.0
CURRAT > 2.0
TL/NW < 1.0
CURRAT > 1.5
TL/NW < 1.2
CURRAT > 1.5
TL/NW < 1.5
CURRAT > 1.5
TL/NW < 2.0
CURRAT > 1.5
TL/NW < 1.0
QRAT > 1.2
TL/NW <.1.2
QRAI > 1.2
TL/NW < 1.5
QRAT > 1.2
TL/NW < 2.0
QRAT > 1.2
TL/NW < 1.0
QRAT > 1.0
TL/NW < 1.2
QRAT > 1.0
TL/NW < 1.5
QRAT > 1.0
«,
7.1
7.1
11.9
16.6
16.6
4.7
9.5
19.0
23.8
7.1
7.1
9.5
9.5
7.1
9.5
14.2
M..
26.1
16.6
23.8
30.9
35.7
21.4
28.5
45.2
57.1
11.9
19.0
23.8
26.1
11.9
19.0
26.1
V
66
54
65
69
72
62
76
84 .
91
42
49
51
54
52
63
68
C
83.4
88.1
82.2
76.2
73.9
87.0
81.0
67.9
59.6
90.5*
86.9
83.4
82.2
90.5*
85.8
79.8
D
39.9
37.4
41.2
38.1
36.3
40.6
47.5*
38.8
33.9
30.1
30.0
27.2
27.9
40.1
44.0
41.9
EP
5.6*
4.8*
6.0
7.6
8.0
4.6*
5. "5*
8.4
9.8
5.0*
5.9
7.2
7.3
4.0*
5.0*
6.5
           VIII-4

-------
                       TABLE  VIII-1  (Concluded)
Test
Number
48
49
50
51
52
53
54
55
56
Test
Variables
TL/NW < 2.0
QRAT > 1.0
NI/TA > .04
CURRAT > 2.0
NI/TA > .04
CURRAT > 1.5
NI/TA > .02
CURRAT > 2.0
NI/TA > .02
CURRAT > 1.5
NI/TA > .04
QRAT > 1.2
NI/TA > .02
QRAT > 1.2
NI/TA > .04
QRAT > 1.0
NI/TA > .02
QRAT > 1.0
KB
14.2
4.7
9.5
4.7
9.5
4.7
7.1
4.7
7.1
*W
33.3
19.0
30.9
26.1
45.2
14.2
23.8
19.0
30.9
^B
74
57
73
64
83
46
49
62
67
C
76.2
88.1
79.8
84.6
72.7
90.5*
84.6
88.1
81.0
D
40.7
38.0
42.1
.37.9
37.8
31.8
25.2
43.0
36.1
=P
7.1
4.6*
6.1
5.3*
7.2
4.5*
6.9
4.2*
6.2
*Exceeds performance of best single ratio tests.
                                VIII-5

-------
                  TABLE VIII-2

PERFORMANCE OF THREE-RATIO TESTS:   PASS  ALL TO
        (One-Year Eligibility  Requirement)
                  (Tests 57-88)
PASS
Test
Number
57


58


59


60


61


62


63


64


65


66


67


Test
Variables
CF/TL > .2
TL/NW < 1.0
CURRAT > 2.0
CF/.TL > .15
TL/NW < 1.0
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.0
CURRAT > 2.0
CF/TL > .2
TL/NW < 1.2
CURRAT > 2.0
CF/TL > .2
TL/NW < 1.5
CURRAT > 2.0
CF/TL > .15
TL/NW < 1.2
CURRAT > 2.0
CF/TL > .15
TL/NW <.1.5
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.2
CURRAT > 2.0 i
CF/TL > .1
TL/NW < 1.5
CURRAT > 2.0
CF/TL > .1
TL/NW < 2.0
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.5
CURRAT > 1.5
*B
4.7


4.7


4.7


4.7


4.7


4.7


4.7


4.7


4.7


4.7


7.1


"w
7.1


9.5


9.5


9.5


9.5


11.9


14.2


11.9


14.2


19.0


23.8


^B
33


46


50


34


34


51


52


59


62


62


76


C
94.1*


92.9*


92.9*


92.9*


92.9*


91.7*


90.5*


91.7*


90.5*


88.1


84.6


D
25.9


36.5


40.5


24.5


24.5


39.1


37.8


47.1*


47.8*


43.0


52.2


S
3.9*


3.4*


3.1*


4.6*


4.6*


3.6*


4.0*


3.1*


3.4*


4.2*


4.5*


                        VIII-6

-------
TABLE VIII-2 (Continued)
Test
Number
68


69


70


71


72


73


74


75


76 .


77


78


79


Test
Variables
CF/TL > .1
TL/NW < 2.0
CUREAT > 1.5
CF/TL > .2
TL/NW < 1.0
QRAT > 1.2
CF/TL > .15
TL/NW < 1.0
QRAT > 1.2
CF/TL > .1
TL/NW < 1.0
QRAT > 1.2
CF/TL > .2
TL/NW < 1.2
QRAT > 1.2
CF/TL > .2
TL/NW < 1.5
QRAT > 1.2
CF/TL > .15
TL/NW < 1.2
QRAT > 1.2
CF/TL > .15
TL/NW < 1.5
QRAT > 1.2
CF/TL > . 1
TL/NW < 1.2
QRAT > 1.2
CF/TL > .1
TL/NW < 1.5
QRAT > 1.2
CF/TL > .1
TL/NW < 2.0
QRAT > 1.2
CF/TL > .2
TL/NW < 1.0
QRAT > 1.0
"B
7.1


4.7


4.7


4.7


4.7


4.7


4-7


4.7


4.7


4.7


. 4.7


4.7


Mw
33.3


7.1


9.5


9.5


9.5


9.5


11.9


11.9


14.2


14.2


16.6


7.1


^B
80


27


36


38


28


28


41


41


45


46


47


34


C
79.8


94.1*


92.9*


92.9*


92.9*


92.9*


91.7*


91.7*


90.5*


90.5*


89.3


94.1*


D
46.7*


19.9


26.5


28.5


18.5


18.5


29.1


29.1


30.8


31.8


30.4


26.9


*P
5.6*


4.8*


4.3*


4.1*


5.6*


5.6*


4.5*


4.5*


4.6*


4.5*


5.0*


3.8*


           VIII-7

-------
                       TABLE VIII-2 (Concluded)
Test
Number
80


81


82


83


84


85


86


87


88


Test
Variables
CF/TL > .15
TL/NW < 1.0
QRAT > 1.0
CF/TL > .1
TL/NW < 1.0
QRAT > 1.0
CF/TL > .2
TL/NW < 1.2
QRAT > 1.0
CF/TL > .2
TL/NW < 1.5
QRAT > 1.0
CF/TL > .15
TL/NW < 1.2
QRAT > 1.0 |
CF/TL > .15
TL/NW < 1.5
QRAT > 1.0
CF/TL > .1
TL/NW < 1.2
QRAT > 1.0
CF/TL > .1
TL/NW < 1.5
QRAT > 1.0
CF/TL > .1
TL/NW < 2.0
QRAT > 1.0
"B
4.7


4.7


4.7


4.7


4.7


4.7


4.7


4.7


4.7


KW
9.5


9.5


9.5


9.5


11.9


11.9


14.2


16.6


23.8


*m
46


49


36


36


52


54


59


62


&4


C
92.9*


92.9*


92.9*


92.9*


91.7*


91.7*


90.5*


89.3


85.8


D
36.5


' 39.5


26.5


26.5


40.1


42.1


44.8


45.4


40.2


EP
3.4*


3.2*


4.3*


4.3*


3.5*


3.4*


5.3*


3.8*


4.9*


*Exceeds performance of best single ratio tests.
                                 VIII-8

-------
                   TABLE VIII-3
PERFORMANCE OF THREE-RATIO  TESTS:   PASS  2  OF 3  TO PASS

         (One-Year Eligibility Requirement)
                  (Tests 89-120)
Test
Number
89


90


91


92


93


94


95


96


97


98


.99


Test
Variables
CF/TL > .2
TL/NW < 1.0
CURBAT > 2.0
CF/TL > .15
TL/NW < 1.0
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.0
CURRAT > 2.0
CF/TL > .2
TL/NW < 1.2
CURRAT > 2.0
CF/TL > .2
TL/NW < 1.5
CURRAT > 2.0
CF/TL > .15
TL/NW < 1.2
CURRAT > 2.0
CF/TL > .15
TL/NW < 1.5
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.2
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.5
CURRAT > 2.0
CF/TL > .1
TL/NW < 2.0
CURRAT > 2.0
CF/TL > .1
TL/NW < 1.5
CURRAT > 1.5
«B
7.1


7.1


7.1


11.9


23.8


11.9


11.9


23.8


26.1


26.1


23.8


*V
21.4


26.2


33.3


26.2


35.7


30.9


35.7


38.0


50.0


52.3


59.5


\B
62


70


76


72


77


74


78


80


86


93


91


C
85.8


83.3


79.8


81.0


70.3


78.6


76.2


69.1


62.0


60.8


58.4


D
40.6


43.8


42.7


45.8


41.3


43.1


42.3


42.0


36.0


40.7


31.5


Sp
5.0*


5.2*


5..8*


5.8*


8.5


6.4


6.7


8.5


9.7


9.3


10.1


                        VIII-9

-------
TABLE VIII-3 (Continued)
Test
Number
100


101


102


103


104


105


106


107


108


109


110


111


Test
. Variables
CF/TL > .1
TL/NW < 2.0
CURRAT > 1.5
CF/TL > .2
TL/NW < 1.0
QRAT > 1.2
CF/TL > .15
TL/NW < 1.0
QRAT > 1.2
CF/TL > .1
TL/NW < 1.0
QRAT > 1.2
CF/TL > .2
TL/NW < 1.2
QRAT > 1.2
CF/TL > . 2
TL/NW < 1.5
QRAT > 1.2
CF/TL > .15
TL/NW < 1.2
QRAT > 1.2
CF/TL > .15
TL/NW < 1.5
QRAT > 1.2
CF/TL > . 1
TL/NW < 1.2
QRAT > 1.2
CF/TL > .1
TL/NW < 1.5
QRAT > 1.2
CF/TL > .1
TL/NW < 2.0
QRAT > 1.2
CF/TL > .2
TL/NW < 1.0
QRAT > 1.0

" :3
28.5


7.1


7.1


7.1


7.1


14.2


7.1


14.2


7.1


19.0


19.0


7.1



\
59.5


16.6


19.0


26 ..1


21.4


28.5


23.8


30.9


28.5


42.8


47.6


16.6



ANB
96


54


59


62


64


70


69


72


77


82


87


59



C
56.0


88.1


' 87.0


83.4


85.8


78.6


84.6


77.4


82.2


69.1


66.7


88.1



D
36.5


37.4


40.0


35.9


42.6


41.5


45.2


41.1


48.5*


39.2


39.4


42.4



**
10.1


4.8*


4.8*


5.9


4.9*


6.7


4.9*


6.9


5.1*


8.3

-
8.4


4.4

i
           VIII-10

-------
                      TABLE  VIII-3  (Concluded)
Test
Number
112


113


114


115


116


117


118


119


120


Test
Variables
CF/TL > .15
TL/NW < 1.0
QRAT > 1.0
CF/TL > .1
TL/NW < 1.0
QRAT > 1.0
CF/TL > .2
TL/NW < 1.2
QRAT > 1.0
CF/TL > .2
TL/NW < 1.5
QRAT > 1.0
CF/TL > .15
TL/NW < 1.2
QRAT > 1.0
CF/TL > .15
TL/NW < 1.5
QRAT > 1.0
CF/TL > .1
TL/NW < 1.2
QRAT > 1.0
CF/TL > .1
TL/NW < 1.5
QRAT > 1.0
CF/TL > .1
TL/NW < 2.0
QRAT > 1.0
KB
7.1


7.1


9.5


19.0


9.5


19.0


11.9


26.1


26.1


\
21.4


30.9


21.4


30.9


26.1


35.7


38.0


50.0


50.0


^B
67


80


68


74


75


79


83


88


92


1
C
85.8


81.0


84.6


75.0


82.2


72.7


75.0


62.0


62.0


D
45.6


49.1


46.6*


43.1


48.9*


43.3


45.0


38.0


42.0


S
4.7*


5.2*


4.9*


7.4


5.2*


7.6


6.6


9.5


9.1


*Exceeds performance of best single ratio tests.
                                VIII-11

-------
               TEST VIII-4
PERFORMANCE OF MULTI-RATIO FINANCIAL TESTS
 INCLUDING NET FIXED ASSETS/TOTAL ASSETS
     (One-Year  Eligibility  Requirement)
              (Tests 121-151)
Test
Number
121
122
123
124
125
126
127
128
129
130

Test
Variables-
Pass All:
NFA/TA • > -3
CF/TL > .1
TL/NW < 1.0
NFA/TA > .3
CF/TL > .1
TL/NW < 1.2
NFA/TA > . 3
CF/TL > .1
TL/NW < 1.5
NFA/TA > . 3
CF/TL > .1
TL/NW < 2.0
NFA/TA > . 25
CF/TL > .1
TL/NW < 1.5
NFA/TA > . 25
CF/TL > .1
TL/NW < 2.0
NFA/TA "> .2
CF/TL > .1
TL/NW < 2.0
Pass 2 of 3:
NFA/TA . > .3
CF/TL > .2
TL/NW < 1.0
NFA/TA > .3
CF/TL > .2
TL/NW < 1.2
^ O
TL/NW < 1.5

^
2.3
2.3
4.7
4.7
4.7
7.1
9.5
7.1
9.5
19.0


^
4.7
4.7
11.9
11.9
11.9
14.2
16.6
11.9
16.6
28.6


^B
43
53
55
59
62
66
75
62
70
73


C
96.5*
96.5*
91.7*
91.7*
91.7*
89.3
87.0
90.5*
87.0
76.2


D
38.3
48.3*
43.1
47.1*
50.1*
51.8*
58.4*
50.1*
53.4*
44.4


i*
1.7*
1.4*
3.3*
3.1*
2.9*
3.6*
3.8*
3.3*
4.0*
7.1

                    VIII-12

-------
TABLE VIII-4 (Continued)
Test
Number
131


132


133


134


135


136


137



138


139


140


141


Test
Variables
NFA/TA . > .3
CF/TL > .15
TL/NW < 1.0
NFA/TA > . 3
CF/TL > .15
TL/NW < 1.2
NFA/TA > • 3
CF/TL > .15
TL/NW < 1.5
NFA/TA > • 3
CF/TL > .1
TL/NW < 1.0
NFA/TA > *3
CF/TL > .1
TL/NW < 1.2
NFA/TA > -3
CF/TL > .1
TL/NW < 1.5
NFA/TA > -3
CF/TL > .1
TL/NW < 2.0
Pass All:
NFA/TA > -2
CF/TL > .1
CURRAT > 1.5
NFA/TA > • 3
CF/TL > .1
CURRAT > 2.0
NFA/TA > • 3
CF/TL > .2
CURRAT > 1.5
NFA/TA > • 2
CF/TL > .1
CURRAT > 2.0

^B
7.1


9.5


19.0


9.5


11.9


19.0


21.4



4.7


0


2.3


2.3



\
14.2


19.0


30.9


23.8


30.9


45.2


54.7



14.2


0


2.3


7.1



^B
71


76


80


79


83


89


92



76


49


31


55



C
89.3


85.8


75.0


83.4


78.6


67.9


62.0



90.5*


100.0*


97.7*


95.3*



D
56.8*


57.0*


49.1*


55.2*


52.1*


43.8


37.3



61.8*


49.0*


'28.7


47.9*


n
EP
3.3*


4.1*


6.8


4.6* .


5.6*


7.9


9.1



2.8*'


0*


1.6*


1.9*


              VIII-13

-------
                      TABLE VIII-4 (Concluded)
Test
Number
142


143


144


145



146


147


148



149



150



151



Test
Variables.
NFA/TA > -2
CF/TL > .2
CURRAT > 2.0
NFA/TA > -2
TL/NW < 1.2
QRAT > 1.0
NFA/TA > • 3
TL/NW < 1.0
CURRAT > 1.5
NFA/TA > • 3
TL/NW < 1.5
CURRAT > 2.0
Pass 2 of 3:
NFA/TA > -2
CF/TL > .1
CURRAT > 1.5
NFA/TA > • 3
CF/TL > .2
CURRAT > 2.0
NFA/TA > • 3
CF/TL > .2
CURRAT > 1.5
Pass 3 of 4:
NFA/TA • > .3
CF/TL > .1
TL/NW < 2,0
CURRAT > 2.0
NFA/TA > • 3
CF/TL > .1
TL/NW < 1.5
CURRAT > 2.0
NFA/TA > -3
CF/TL > .1
TL/NW < 1.2
CURRAT > 2.0

*B
2.3


4.7


2.3


4.7



21.4


16.6


21.4



14.2
-


11.9



7.1




*W
4.7


9.5


4.7


7.1



64.2


23.8


30.9



33.3



28.5



19.0




**B
. 32


58


49


49



95


70


80



87



81



77




C
96.5*


92.9*


96.5*


94.1*



57.2


79.8


73.9



76.2



79.8



87.0




D
27.3


48.5*


44.3


41.9



30.8


46.2


49.1*



53.7*



52.5*



58.0*




EP
2.4*


2.6*


1.5*


2.6*



9.9


6.3


7.1



6.0



5.5*



3.7*



*Exceeds performance of best single ratio tests.
                                VIII-

-------
            TABLE VIII-5
    PERFORMANCE OF TWO-RATIO TESTS
(Three-Year Eligibility Requirement)
           (Tests 1-56)
Test
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Test
Variables
CF/TL > .2
TL/NW < 1.0
CF/TL > .2
TL/NW < 1.2
CF/TL > .2
TL/NW < 1.5
CF/TL > .2
TL/NW < 2.0
CF/TL > .15
TL/NW < 1.0
CF/TL > .15
TL/NW < 1.2
CF/TL > .15
TL/NW < 1.5
CF/TL > .15
TL/NW < 2.0
CF/TL > .1
TL/NW < 1.0
CF/TL > .1
TL/NW < 1.2
CF/TL > .1
TL/NW < 1.5 .
CF/TL > .1
TL/NW < 2.0
NI/TA > .04
TL/NW < 1.0
NI/TA > .04
TL/NW < 1.2
NI/TA > .04
TL/NW < 1.5
KB
2.3
2.3
4.7
4.7
2.3
2.3
4,7
4.7
2.3
2.3
2.3
4.7
2.3
2.3
4.7
Mw'
9.5
11.9
11.9
11.9
11.9
14.2
14.2
19.0
14.2
19.0
21.4
30.9
11.9
16.6
19.0
SB
31
34
35
36
43
50
52
54
49
62
70
76
41
51
56
C
94.1
92.9
91.7
91.7
92.9
91.7
90.5
88.1
91.7
89.4
88.1
82.2
92.9
90.5
88.1
D
40.5
22.1
23.1
24.1
31.1
35.8
37.8
35.0
34.8
43.0
48.6
45.1
29.1
34.4
37.0
EP
4.2
4.6
5.2
5.1
3.6
3.7
4.0
4.8
3.7
3.8
3.7
.5.2
3.8
4.1
4.7
                  VIII-15

-------
TABLE VIII-5 (Continued)
Test
Number
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Test
Variables
NI/TA > .04
TL/NW < 2.0
NI/TA > .02
TL/NW < 1.0
NI/TA > .02
TL/NW < 1.2
NI/TA > .02
TL/NW < 1.5
NI/TA > .02
TL/NW < 2.0
CF/TL > .2
CURRAT > 2.0
CF/TL > .15
CURRAT > 2.0
CF/TL > .1
CURRAT > 2.0
CF/TL > .2
CURRAT > 1.5
CF/TL > .15
CURRAT > 1.5
CF/TL > .1
CURRAT > 1.5
CF/TL > .2
QRAT > 1.2
CF/TL > .15
QRAT > 1.2
CF/TL > .1
QRAT > 1.2
CF/TL > .2
QRAT > 1.0
CF/TL > .15
QRAT > 1.0
XB
4.7
2.3
2.3
4.7
7.1
0
0
0
2.3
2.3
4.7
0
0
0
2.3
2.3
KW
28.5
' 14.2
19.0
26.1
38.0
4.7
9.5
14.2
7.1
14.2
26.1
4.7
7.1
11.9
7.1
11.9
^B
60
49
60
69
75
25
37
49
34
46
64
20
27
30
32
43
C
83.4
91.7
89.4
84.6
77.4
97.7*
95.3
92.9
95.3
91.7
84.6
97.7*
96.5*
94.1
95.3
92.9
D
31.5
34.8
41.0
42.9
37.0
20.3 .
27.5
34.8
26.9
31.8
37.9
10.5
15.1
18.1
24.9
31.1
EP
6.1
3.7
3.9
4.9
6.6
2.0*
2.8*
3.2*
3.0*
4.0
5.3
2.5*
2.9*
4.3
3.2*
3.6
           VIII-16

-------
TABLE VIII-5 (Continued)
Test
Number
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Test
Variables
CF/TL > .1
QRAT > 1.0
TL/NW < 1.0
CURRAT > 2.0
TL/NW < 1.2
CURRAT > 2.0
TL/NW < 1.5
CURRAT > 2.0
TL/NW < 2.0
CURRAT > 2.0
TL/NW < 1.0
CURRAT > 1.5
TL/NW < 1.2
CURRAT > 1.5
TL/NW < 1.5
CURRAT > 1.5
TL/NW < 2.0
CURRAT > 1.5
TL/NW < 1.0
QRAT > 1.2
TL/NW <-1.2
QRAT > 1.2
TL/NW < 1.5
QRAT > 1.2
TL/NW < 2.0
QRAT > 1.2
TL/NW < 1.0
QRAT > 1.0
TL/NW < 1.2
QRAT > 1.0
TL/NW < 1.5
QRAT > 1.0
KB
4.7
2.3
7.1
9.5
11.9
2.3
7.1
9.5
16.6
2.3
2.3
4.7
4.7
4.7
7.1
11.9
*w*
23.8
11.9
19.0
23.8
30.9
19.0
26.1
35.7
50.0
7.1
14.2
19.0
21.4
9.5
16.6
23.8
ANB
52
42
50
55
56
53
66
76
80
26
30
33
33
46
50
5A
C
85.8
92.9
87.0
83.4
78.6
89.4
83.4
77.4
66.7
95.3
91.7
88.1
86.9
92.9
88.1
82.2
D
28.2
30.1
31.0
31.2
25.1
34.0
39.9
40.3
30.0
18.9
15.8
14.0
11.6
36.5
33.4
30.2
EP
6.1
3.7
5.8
6.7
8.4
4.4
5.5
6.6
9.2
4.0
6.1
7.9
8.7
3.4
5.2
7.3
           VIII-17

-------