Griffin v. Bolger Brief for Plaintiffs-Appellants
Public Court Documents
May 3, 1984
Cite this item
-
Brief Collection, LDF Court Filings. Griffin v. Bolger Brief for Plaintiffs-Appellants, 1984. 0672f9ac-b49a-ee11-be36-6045bdeb8873. LDF Archives, Thurgood Marshall Institute. https://ldfrecollection.org/archives/archives-search/archives-item/f6d2075d-1ea4-44f4-b626-2ec2da340e8f/griffin-v-bolger-brief-for-plaintiffs-appellants. Accessed October 26, 2025.
Copied!
4
IN THE
UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
No. 84-3070
ERNEST L. GRIFFIN, et a i .,
Plaintiffs-Appellants,
v .
WILLIAM F. BOLGER,
Postmaster General,
Def endant-A.ppellee.
On Appeal From The United States District Court
for The Middle District of Florida
Jacksonville Division
BRIEF FOR PLAINTIFFS-APPELLANTS
'I
JACK GREENBERG
GAIL J. WRIGHT
CHARLES STEPHEN RALSTON
PENDA HAIR
99 Hudson Street
16th Floor
New York, New York 10013
W. BENJAMIN KYLE
1248 W. Edgewood Avenue
Jacksonville, Florida 32208
Attorneys for Plaint iffs-Appellants
IN THE
UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
No. 84-3070
ERNEST L. GRIFFIN, et. a l . ,
Plaintiffs-Appellants,
v.
WILLIAM F. BOLGER,
Postmaster General,
Defendant-Appellee.
On Appeal From The United States District Court
for The Middle District of Florida
Jacksonville Division
Certificate of Interested Persons
The undersigned counsel of record for plaintiffs-appellants,
certifies that the persons listed below have an interest in the
outcome of this action:
I . As plaintiffs-appellants
A. The named plaintiffs:
1. Ernest L. Griffin
2. D'Alver L. Wilson
3. Charles C. McRae
4. Richard Deloney, Jr.
*5. Samuel George
6. Alphonso West
7. Erno H. Denefield
l
8. Thaddeus E. Raysor
9. Margie L. Raysor
10. Joe Bailey, Jr.
1 1 . Andrew Edwards
12. Claude L. Smith
13. Smith M. Morgan
14. Jesse L. Wilcox
15. Harvey J. Harper
16. Joyce A. Scales
17. Albert Jackson, Jr.
18. Kenneth A. Rosier, Jr
19. Andrew D. Martin, Jr.
20. James Williams
21 . John H. Fowler, and
*22. Doris D. Galvin
*These persons are now deceased
B. The class of Blacks now employed or formerly employed
by the United States Postal Service, Jacksonville, Florida.
I I . As defendant-appellee;
1. Elmer T. Klassen, former Postmaster General.
2. Benjamin F. Bailer, former Postmaster General.
3. William F. Bolger, Postmaster General.
4. James E. Workman, Postmaster of the Jacksonville,
Florida Post Office.
5. Melvin Trescott, former Director, Office of Personnel,
and his agents.
These representations are made in order that judges of this
Court, inter alia, may evaluate possible disqualification or
recusal pursuant to Rule 22(f)(2) of the Local Rules for the
United States Court of Appeals for the Eleventh Circuit.
Respectfully submitted,
/ c •I V ■U. V
Counsel for Plaihtiffs-Appellants
11
IN THE
UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
NO. 84-3070
ERNEST L. GRIFFIN, et a l .,
Plaint iffs-Appellants,
v.
WILLIAM F. BOLGER,
Postmaster General,
Defendant-Appellee.
On Appeal From The United States District Court
for The Middle District of Florida
Jacksonville Division
STATEMENT REGARDING PREFERENCE
In accordance with Local Rule 22(f)(3) plaintiffs-
appellants state that this case is not entitled to preference
in disposition and processing.
Respectfully submitted,
f
Counsel for Plaintiffs-Appellants
- iii -
IN THE
ERNEST L. GRIFFIN, et a l .,
Plaintiffs-Appellants,
v.
WILLIAM F. BOLGER,
Postmaster General,
Defendant-Appellee.
UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
No. 84-3070
On Appeal From The United States District Court
for The Middle District of Florida
Jacksonville Division
Statement Regarding Oral Argument
Pursuant to Local Rule 22(f)(4) plaint iffs-appellants
respectfully request that this appeal be orally argued.
This matter involves numerous complex issues with regard to
actions instituted pursuant to Title VII of the Civil Rights
Act, as amended. First, the appeal addresses the proper
scope of Title VII lawsuits as based upon administrative
complaints. Further, the appeal concerns the appropriate
standards of proof in pattern and practice class actions.
In addition, this matter concerns the nature of individual
IV
determinations, in employment discrimination cases. Plaintffs-
appellants contend that the district court found no discrimina
tion on the basis of erroneous evidentiary standards, and that
the named plaintiffs and class members are entitled to a trial
on their individual claims. Finally, this appeal concerns the
appropriateness of assessing costs against unsuccessful
plaintiffs in Title VII lawsuits when the federal government
is the defendant.
The record in this appeal is voluminous. The tran
script, which resulted from a four and one half week trial,
is 4,500 pages; and the lower courts' opinion is in excess of
300 pages. Plaintiffs-appellants submit that oral argument
would clarify the presentation of the facts and facilitate the
resolution of the legal arguments.
1 t t r< i t V-vra 1 P
Counsel for Plaintiffs-Appellants
v
TABLE OF CONTENTS
Page
CERTIFICATE OF INTERESTED PERSONS ............................ i
STATEMENT REGARDING PREFERENCE ............................... iii
STATEMENT REGARDING ORAL ARGUMENT ............................. iv
TABLE OF CONTENTS ................................................ vi
STATEMENT OF THE ISSUES ........................................ 1
STATEMENT OF THE CASE ........................................... 2
A. Course of The Proceeding and Disposition
in The Court Below ................................ 2
B. Statement of the Facts ........................... 4
C. Standard of Review ................................ 33
SUMMARY OF ARGUMENT ............................................. 34
STATEMENT OF JURISDICTION ...................................... 35
ARGUMENT ........................................................... 35
I. The District Court's Refusal to Allow Plaintiffs
to Challenge Defendant's Written Personnel Test
Requirement Was in Error ................................. 35
II. The Statistical Evidence Establishes Discrimina
tion in Promotions, Details, Discipline, And
Awards ....................................................... 44
A. Introduction ...................................... 44
B. Supervisory Positions and Details ............. 49
C. Discipline .......................................... 51
III. The District Court Improperly Dismissed Plaintiffs'
Disparate Impact Claims .................................. 55
IV. The District Court's Finding with Regard to Class
Members Should Be Reversed And Remanded ............... 63
V. The Court's Award of Costs to The Defendant Was
Improper .................................................... 66
A. It Is Unjust to Tax Costs Against The
Unsuccessful Plaintiffs in This Action ....... 68
vi
B . Defendants Are Precluded from Seeking
Costs Because of A Justice Department
Directive ........................................... 71
Conclusion ........................................................ 74
Certificate of Service
Appendix
Table of Authorities
Cases: Pa9es
Albemarle Paper Co. v. Moody, 422 U.S. 405 (1975) ......... 50
Alexander v. Louisiana, 405 U.S. 625 ( 1972) ................ 48,50
Andersen v. Clear Ridge Aviation, 9 F.R.D. 50 (D.C.
Neb. 1949) ..................................................... 68
Barnett v. W. T. Grant Co., 518 F.2d 543 (4th Cir.
1975) ........................................................... 47
Bonner v. City of Pritchard, 661 F.2d 1206 (11th
Cir. 1981) ..................................................... 57
Bradley v. School Board of Richmond, 416 U.S. 696
(1974) ......................................................... 70
Capaci v. Katz & Besthoff, Inc., 711 F .2d 647
(5th Cir. 1983) .............................................. 33,48
*Carroll v. Sears Roebuck & Co., 708 F.2d 130 (5th
Cir. 1983) ..................................................... 45,49,57
Chicago Sugar Co. v. American Sugar Refinery, 173 F.2d
1 (7th Cir. 1949) ............................................ 68
C h i s h o l m v. U.S. Postal Service, 516 F. Supp. 810
(W.D.N.C. 1980), aff'd, 665 F.2d 482 (4th Cir. 1981) .. 35,42,43,44
45,49
*Christiansburg Garment Co. v. EEOC, 434 U.S. 412
(1978) .......................................................... 70,71,73,74
C o nn ec ti cu t v. Teal, 457 U.S. 440 (1982) ................... 36,50,59,
60,66
Cooper v. Federal Reserve Bank of Richmond, 83-185 ......... 66
Page
* Cases principally relied upon.
- vii -
Cases: Pages
County of Suffolk v. Secretary of the Interior, 76
F.R.D. 469 (E.D.N.Y. 1977) ................................. 70
Crawford v. Western Elec. Co., Inc., 614 F.2d 1300
(5th Cir. 1980) .............................................. 57
Crown, Cork & Seal Co. v. Parker, U.S. , 76 L.Ed.2d
628 (1983) ....................... 7 7 ____7 7 ................... 66
Davis v. Califano, 613 F.2d 957 (D.C. Cir. 1979) .......... 45,47,48
*Delta Airlines, Inc. v. August, 450 U.S. 346 (1981) ..... 67
Donaldson v. Pillsbury Co., 554 F.2d 825 (8th Cir. 1977) . 64
Dual v. Cleland, 79 F.R.D. 696 (D.D.C. 1978) ............... 73
*Eastland v. Tennessee Valley Authority, 704 F.2d 613, 714
F.2d 1066 (11th Cir. 1983), cert, denied, 52 U.S.L.W.
3631 (1984) .................................................. 41 ,43,44,57
*EEOC v. American National Bank, 652 F.2a 1176 (4th
Cir. 1981) ..................................................... 28,48
Elreage v. Carpenters, 46 Calif. Counties, 83 F.R.D. 136
(N.D. Cal. 1979) ............................................. 70
Evans v. U.S. Pipe & Foundry Co., 696 F.2d 925 (11th
Cir. 1983) ..................................................... 41
Farmer v. Arabian American Oil Co., 379 U.S. 227 (1964) .. 67
Gilbert v. City of Little Rock, Ark., 722 F.2d 1390
(8th Cir. 1983) .............................................. 61
Grant v. Bethlehem Steel Corp., 635 F.2d 1007 (2nd
Cir. 1980) ..................................................... 47,48
*Griggs v. Duke Power Co., 401 U.S. 424 (1971) ............ 50,55,58,
60,
Harrell v. Northern Electric Co., 672 F.2d 444 (5th
Cir. 1982) .................................................... 57
Harrison v. Lewis, 559 F. Supp. 943 (D.D.C. 1983),
Memorandum Opinion (D.D.C. June 7, 1982), aff'd,
F .2d (D.C. Cir. 1984) ................................. 63
Hazelwood School District v. United States, 433 U.S. 299
(1977) ......................................................... 47,49,53
- viii -
Hoffman v. Boeing, 596 F.2d 683 (5th Cir. 1979 ) ........... 42
la Power & Light Co. v. Burlington Northern Inc., 647
F . 2d 796 ( 8th Cir. 1981) ................................... 74
In re Northern Indiana Oil Co., 192 F.2d 139 (7th Cir.
1951) ........................................................... 68
Jaspers v. Alexander, 15 F.E.P. Cases 1238 (D.D.C. 1977) . 72
*Johnson v. Uncle Ben's, Inc., 628 F.2d 419 (5th Cir.
1980) ........................................................... 57,58,62
Lawler v. Alexander, 698 F.2d 439 (11th Cir. 1983) ....... 43
Lawrence v. Staats, 665 F.2d 1256 (D.C. Cir. 1981) ....... 74
Lea v. Cone Mills Corp. 438 F.2d 86 (4th Cir. 1971) ...... 70
Lichter Foundation, Inc. v. Welch, 269 F.2d 142 (6th
Cir. 1959) ................................................. 68
*Lilly v. Harris-Teeter Supermarket, 720 F.2d 326
(4th Cir. 1983) .............................................. 45,53,55 ,
Maldonado v. Parasole, 66 F.R.D. 388 (S.D.N.Y. 1975) ..... 70
Mangiapane v. Adams, 661 F.2d 1388 (D.C. Cir. 1981) ...... 42
Martin v. Frontier Federal Savings and Loan Assoc.,
510 F. Supp. 1062 (W.D. Okla. 1981) ...................... 70
Miller v. Amusement Enterprises, 426 F .2d 534
(5th Cir. 1970) .............................................. 90
Miller v. International Paper Co., 408 F.2d 283
(5th Cir. 1969) .............................................. 69
Mizrany v. Texas Rehabilitation Commission, 522 F. Supp.
611 (S.D. Tex. 1981) ....................................... 73
Mortensen v. Callaway, 672 F.2d 822 (10th Cir. 1982) .... 57
Neal v. Delaware, 103 U.S. 370 (1881) ...................... 48
Newman v. Piggie Park Enterprises, 390 U.S. 400 (1968) ... 70
Ong v. Cleland, 642 F.2d 316 (9th Cir. 1981) ............... 41
Cases; Pages
IX
Page v. U.S. Industries, Inc., 726 F.2d 1038 (5th Cir.
1984) ........................................................... 45,57
Parker v. Califano, 561 F.2d 320 (D.C. Cir. 1977) ......... 71
Patton v. Mississippi, 332 U.S. 463 ( 1947) ................. 48,54
Payne v. McLemore's Wholesale & Retail Stores, 654
F .2d 1130 (5th Cir. 1981) .................................. 65
*Payne v. Travenol Laboratories, 673 F.2d 798 (5th Cir.
1982) ........................................................... 45,65
Peaues v. Mississippi State Employment Serv., 699 F.2d
760 (5th Cir. 1983) .......................................... 48,53
Pope v. City of Hickory, N.C., 679 F.2d 20 (4th Cir.
1982) ........................................................... 57
Pouncy v. Prudential Ins. Company of America, 668 F.2d
795 (5th Cir. 1982) .......................................... 56,57,58 ,
62
Pullman-Standard Co. v. Swint, 456 U.S. 273 (1982) ....... 61
Rowe v. Cleveland Pneumatic Co., Numerical Control, 690
F . 2d 88 (6th Cir. 1982) ..................................... 57
Rowe v. General Motors Corporation, 457 F.2d 348 (5th
Cir. 1972) ..................................................... 57
Rule v. I.A.B.S.O.I., Local Union No. 396, 568 F.2d
558 (8th Cir. 1977) ........................................... 58
Sampson v. Civiletti, 632 F.2d 860 (9th Cir. 1980) ....... 42
Schaulius v. CTB/McGraw-Hill, Inc., 496 F. Supp. 666
(N.D. Ca. 1980) .............................................. 70
Segar v. Civiletti, 508 F. Supp. 609 (D.D.C. 1981) ....... 47,48,63
Service v. Dulles, 354 U.S. 363 (1957) ...................... 72
Siegel v. Kreps, 654 F.2d 773 (D.C. Cir. 1981) ............ 42
*Teamsters v. United States, 431 U.S. 324 (1977) .......... 47,48,49,
55
Texas Dept, of Community Affairs v. Burdine, 450 U.S.
248 ( 1982) ..................................................... 50
Thomas v. Department of the Navy, No. 77-1965 ............. 72
Cases; Pages
x
Cases: Pages
Torres v. Claytor, 25 F.E.P. Cases 998 (S.D. Cal. 1978) .. 73
United States v. Nixon, 418 U.S. 683 ( 1974) ................ 72
United States ex rel Accardi v. Shaughnessy, 347 U.S.
260 (1954) ..................................................... 7 2
Valentino v. U.S.P.S., 674 F.2d 56 (D.C. Cir. 1982) ...... 49
Vitarelli v. Seaton, 359 U.S. 535 ( 1959) ................... 72
Walker v. Jefferson County Home, ___ F.2d ___ , 34 F.E.P.
Cases 465 (5th Cir. 1984) .................................. 51
Other Authorities:
CCh Fair Employment Practices 1i 5083 (1978) ................ 71
5 C.F.R. § 713.216 (a) .......................................... 43
H. Rep. No. 92-238 (92nd Cong., 1st Sess., 1971) .......... 46,61
Recommendations on Pretrial Proceedings In Cases With
Voluminous D a t a , 39 The Record of the Association of
the Bar of the City of New York 49 ( 1 984) ............... 32
Rule 54(d), F.R. Civ. Proc...................................... 67
Schlei & Grossman, Employment Discrimination Law
(2nd Ed. 1983) ................................................ 71
S. Rep. No. 92-415 (92nd Cong., 1st Sess., 1971) ........... 46,47,61,
29 C.F.R. § 1607 ................................................. 62
Uniform Guidelines on Employee Selection ................... 62
xi
IN THE
ERNEST L. GRIFFIN, et a l .,
Plaintiffs-Appellants,
v.
WILLIAM F. BOLGER,
Postmaster General,
Defendant-Appellee.
UNITED STATES COURT OF APPEALS
FOR THE ELEVENTH CIRCUIT
NO. 84-3070
On Appeal From The United States District Court
for The Middle District of Florida
Jacksonville Division
BRIEF FOR PLAINTIFFS-APPELLANTS
Statement of The Issues
I. whether the district court erred in refusing to allow
plaintiffs to challenge defendant's written tests?
II. Whether the district court applied the proper legal
standards in assessing the statistical evidence?
III. whether the disparate impact model of proof of discrimina
tion is limited to an isolated, objectively scored com
ponent of a selection process?
IV. Whether the district court's findings that none of the
individual class members had suffered racial discrimination
were erroneous?
V. Whether the class members are entitled to a trial of their
individual claims?
VI. Whether the defendant is entitled to costs?
STATEMENT OF THE CASE*
A. Course of The Proceeding and Disposition in the
Court Below
This action was instituted by ERNEST L. GRIFFIN, a
Black former employee of the United States Postal Service,
Jacksonville and twenty-two other plaintiffs to enforce
rights granted by Title VII of the Civil Rights Act of 1964,
as amended, 42 U.S.C. § 2000e-16. On August 29, 1971 plaintiff
Griffin filed a third party complaint with the United States
Civil Service Commission stating that Blacks were being dis
criminated against due to their race. (RE 149).
The Complaint was filed on July 7, 1972 in the District
Court for the Middle District of Florida. (R 1-11). It was
amended on November 6, 1972 (R 1-11, 128-131), and a Consoli
dated Amended Complaint was filed on November 12, 1981 (RE
V Throughout the Brief plaintiffs use the following references:
R ........................... Record
RE ........................... Record Excerpts
TR ........................... Trial Transcript
PX ........................... Plaintiff's Exhibit
DX ........................... Defendant's Exhibit
- 2 -
20-29).- The action, brought on behalf of the class of all
Black employees of the United States Postal Service, Jackson
ville, Florida, alleges that the defendant engaged in racially
discriminatory employment practices in violation of 42 U.S.C.
§ 2000e-16(a ) . Plaintiffs sought injunctive and affirmative
relief for themselves and the class which they represented.
In an Order dated January 9, 1973 the district court
denied defendant's motion to preclude plaintiffs from maintain
ing a class action and "authorized plaintiffs to proceed as
representatives of a class." (R 164). On September 8, 1982
the court below denied the defendants' Renewed Motion to
V This lawsuit, which was one of the first major class
actions lodged against a federal government agency, was
instituted at a time when the law in the area was un-established
or irresolute. Litigation activities were suspended or delayed
from 1972 until 1976 pending Chandler v. Roudebush, 425
U.S. 840 (1976) (plaintiffs suing the federal government
pursuant to § 717 of Title VII are entitled to a trial de
novo); Brown v. General Services Administration, 425 U.S. 820
(1976)(Title VII provides the exclusive judicial remedy for
claims of discrimination in federal employment); and Place v .
Weinberger, 426 U.S. 932 (1976)(Title VII applies retroactively
to administrative complaints which were pending on March 24,
1972, the effective date of the amendment to the Act). (R
4-5-25; 4050).
Further, litigation activities inclusive of discovery,
were abated pending the court's determination of the defen
dant's motion to dismiss, which was filed on September 29,
1972 (R 34-107). Judge George Tjoflat's January 9, 1973 Order
reserved ruling on the question of whether the complaint was
timely filed and therefore properly before the court, pending
an evidentiary hearing. (R 163-164). On January 17, 1978 the
case was transferred to the Honorable George C. Carr, and on
August 17, 1978 the lawsuit was re-assigned to the Honorable
Susan Black. (RE 3-4) The evidentiary hearing as directed by
Judge Tjoflat in 1973 was not conducted until 1980, and a ruling
was rendered on April 15, 1980. (R 801-806). Thus, discovery
could not be pursued until 1980, seven years after the case was
filed. (RE 5-15) .
3
Dismiss plaintiffs' class claims filed on July 13, 1982 (R
2890-2901), and the case proceeded as a class action.
(RE 151-52).
The case was tried before the Honorable Judge Susan Black
from September 13 to October 21, 1982. Plaintiffs chal
lenged the defendant's employment policies, practices and
2/
procedures which affected Black employees.— They offered
comprehensive and extensive evidence in the form of statis
tics, documents and testimony to support their claim that
defendants engaged in racial discrimination.
In an Opinion dated November 23, 1983 the trial court held
that plaintiffs had not prevailed on any of their claims. (RE
419). This appeal from the district court's judgment (RE 123)
was timely filed.-^(r 4449-4451).
B . Statement of the Facts
1. ERNEST L. GRIFFIN, the lead plaintiff, was employed
by the United States Postal Service in Jacksonville as a Dis
tribution Clerk from April, 1966 until his termination in Decem
ber, 1976. (R 1-11). Twenty-two additional class members were
2/ The court's statement that it was not until the eve of trial
when plaintiffs advised that they were not addressing the issue
of hiring or representing the class of applicants is incorrect.
(RE 161). It was evident from numerous pre-trial pleadings that
applicants and hiring were not at issue and that defendants had
notice of this fact. (S e e , e . g . , R 1525). The misunderstanding
may have resulted from plaintiffs' challenge to defendant's
practices with regard to initial job assignments of persons
hired, as opposed to the issue of hiring itself.
3/ Costs were awarded in favor of defendant; a Bill of
Costs was filed on February 6, 1984, subsequent to plaintiff's
filing their notice of appeal.
4
named in the Complaint.-7 (R 1-11). The current Postmaster
General, William F. Bolger, as the head of the United States
Postal Service, Jacksonville is the sole proper defendant under
42 U.S.C. §2000e-16(c).
2. Organization and Function of the Post Office
The United States Postal Service was created by the Postal
Reorganization Act of 1970 as an independent establishment in
the executive branch (Title 39 U.S.C. § 3201 United States Code).
(RE 24). The Post Office in Jacksonville covers the metropolitan
area and provides service primarily for Duval County. (RE
154) .
The Post Office, one of the major federal employers in
Jacksonville, employed 1,839 persons at the time this action
was initiated — 1,240 (67%) were White, 589 (32%) were Black
and 10 (1%) were other. (Id.) The average estimated workforce
during the time covered by this lawsuit has been 1,880 persons,
approximately 36% of whom were Black (TR 2430-31).
The Post Office maintains twenty stations and branches
throughout the Jacksonville area. (TR 2157). The major
office is the General Mail Center, commonly referred to as "GMC".
4/ (1) D'Alver L. Wilson, Distribution Clerk; (2) Charles C.
McRae, Clerk; (3) Richard Deloney, Mailhander; (4) Samuel
George, Clerk; (5) Alphonso West, Clerk; (6) Erno Denefield,
Mailnandler; (7) Thaddeus E. Raysor, Clerk; (8) Margie L.
Raysor, Distribution Clerk; (9) Joe Bailey, Jr., Clerk; (10)
Andrew Edwaras, Carrier; (11) Claude L. Smith, Clerk Technician;
(12) Smith M. Morgan, Clerk Technician; (13) Jesse L. Wilcox,
Clerk; (14) Harvey J. Harper, Clerk; (15) Joyce A. Scales,
Clerk; (17) Albert Jackson, Jr., Mailhandler; (18) Kenneth A.
Rosier, Distribution Clerk; (19) Andrew D. Martin, Jr., Clerk;
(20) James Williams, Clerk; (21) John H. Fowler, Mechanic; (22)
Doris D. Galvin, Relief Window Clerk.
5
(TR 2161). The main stations are Jacksonville Beach, Pottsburg,
Arlington, Mandarin, Lake Shore, Murray Hill, Lake Forest,
Oceanway and Carver. (TR 2157).
The Post Office has four major organizational divisions.
(UX 8) These divisions perform the following functions:
The Mail Processing Division is responsible for distribut
ing, processing, and dispensing the mail, and plant maintenance,
inclusive of custodial and building maintenance and machinery
repair (TR 2158). Approximately 800 persons are employed in
this division, primarily as clerks and mailhandlers. (TR 2158;
TR 2755, TR 2161 ) .
The Customer Service Division is responsible for the
dispatch and delivery of mail to customers, window services and
related activities involving direct contact with customers.
(PX 98-100, 106; TR 2159-2165). Persons employed within this
division include city carriers, window clerks, and motor
vechile operators or mechanics. (DX 8; TR 2169-70).
The Finance or Support Division is responsible for provid
ing finance administration, data systems information, accounting
and costing, auditing and administrative services. (PX 98-100,
106; TR 2169-70; 4311). The craft employees in this division
include accounting clerks and data technicians. (TR 2167.)
The Employee and Labor Relations Division is responsible
for the administration of personnel policies including hiring,
training, promotions, discipline and the administration of the
collective bargaining agreements, health insurance, and life
insurance. (PX 99-100, 106; TR 2160-2161).
6
The Post Office is headed by a Postmaster, James E.
Workman, a White male, who held the position of Acting Post
master from March 1961 until his permanent appointment in the
mid-1070's. The four organizational divisions are each headed
by a "Director" who reports to the Postmaster. Under the
division director are superintendents, managers, supervisors and
initial level supervisors who administer operational and per
sonnel policies and practices (PX 98-100; 106; PX 2; DX 13; TR
2157, TR 2425)
3. Description of the Positions and Pay Schedules
Eighty-five to ninety percent of the total workforce are
employed in one of several collective bargaining units or
crafts. (PX 1 1, DX 13, TR 2214-2220). The remaining ten
percent are non-bargaining unit employees, supervisors, or
management personnel (_Id.) Approximately fifty (50) percent of
the craft employees are clerks. (PX 1 Table 1.3; TR 2176).
(TR 2161-68). Thirty (30) percent of the craft employees are
carriers. Ten (10) percent are mailhandlers. (PX 2, 3, 7, TR
2167). Maintenance employees include mechanics responsible for
repairing the buildings and equipment, and custodians who clean
the facility. (TR 2176).
Craft employees are designated by Postal Field Service
(PF S) or by Postal Service (PS) pay levels. (DX 13). Each
position is assigned a pay level, and the higher the level the
5/ In 1970 the title "foreman" was revised to "supervisor".
7
greater the base rate of pay. (_Id. ) At the time this lawsuit
began clerks were generally at Levels 4 and 5 carriers were
Level 4; and journeymen mailhandlers Level 4. (PX 98-100,
106). Levels 1, 2 ana 3 are maintenance employees. These
levels were raised by one level (number) sometime in the
1970's (_Id. ) Each pay level is further aivided into "steps";
and tne higher the step the greater the salary, with a top step
of 12 (Id., DX 13; Chap. 4, 422.3).
Currently, the majority of employees in the clerk craft
are level 4, 5 and 6. The majority of employees are carrier
craft level 5, with the exception of a few level 6 carrier
technician positions (TR 2168, 2176). Mailhandlers are gener
ally level 4, with some at level 5 positions. (TR 2168; 2176).
Maintenance employees who perform custodial services are levels
1 to 3; mechanics are level 4 to 9; and technical mechanical
positions begin at level 6. (TR 2880; 2882).
Non-craft positions include initial level supervisors,
certain non-technical positions and management positions. (PX
2, DX 13). Initial level supervisory positions and non-techni
cal positions were PS level 7 or 9 until 1973, when they were
redesignated Postal Management Schedule (PMS) levels 15, 16,
17. (DX 13, TR 2223). As of 1973 management supervisory
positions were designated PS 10 and above. (PX 2, 3; DX 13).
In 1973 these positions were designated Postal Executive
Schedule (PES) at levels 18 and above (I_d. ) In 1976 the
Technical, Administrative and Clerical (PTAC) Schedule was
8
established. (DX 1 , 3 ) . In 1978 the PFS and PTAC Schedules
were joined into the Executive and Administrative Schedule
(EAS). (Id.)
4. Personnel Procedures
a) Governing guidelines
The procedures for all of defendant's personnel practices
at the Jacksonville Postal Service are governed by regulations
of the Office of Personnel Management (formerly the United States
Civil Service Commission), the Federal Personnel Manual, and imp
lementing directives. Supplementing these regulations are
6/
internal directives of the United States Postal Service.—
Chapter 5 of the Personnel Handbook, P-11 establishes policies
and procedures for assignment, reassignment and promotion. (PX
2). Employment practices relating to employee classification,
salaries and wages, job posting, reassignments, discipline,
promotions, details, and grievances have also been affected
by both national and local agreements between the United
States Postal Service and four national unions, which each
represent a craft.— ^ (DX 14). (Id . ) .
6/ The primary documents controlling selection, assignment,
promotions, performance evaluations, detailing, step increases,
awards and discipline have been the P-11 Handbook; P-1 Handbook
and amendments; the Employee Labor Relations Manual; Handbook
X-118B; Personnel Handbook, P-12 and the union agreements. (PX
2, 3, and DX 13, 14). The promotion procedures in existence
from 1968 until 1981 are compiled in Plaintiffs Exhibit 9,
"Promotions Procedures."
7/ The major unions are the American Postal Workers Union,
AFL-CIO (APWU), which represents employees in the clerk, motor
vehicle, maintenance, and special delivery crafts; the National
Association of Letter Carriers AFL-CIO (NALC) which represents
letter carriers; the National Post Office Mail Handlers, Watchmen,
Messengers and Group Leaders Division of the Laborers' Interna-
9
b. Advancement Within Crafts
Pursuant to the collective bargaining agreements there are
two types of craft positions or bargaining unit positions which
employees bid upon: the "senior qualified," which are filled
according to seniority, and "best qualified," which considers
qualifications. (PX 2, Section 520; TR 2196). Ninety percent
of the advancements within craft positions are made on the
g ̂
basis of "senior qualified". (TR 2201)— Positions above
Level 6 and positions which involve technical or special
knowledge or skills are determined on the basis of the best
9 /
qualified. (TR 2205).— Applicants submit a letter expressing
interest in the postion and listing their qualifications. (RE
167) Management uses its discretion in the selection of
employees for best qualified positions. (I d .).
c. Promotions for Non-Bargaining Unit Employees
Non-bargaining unit employees are categorized in three
groups: (1) initial level supervisors, including the line
supervisors, who supervise the work of the bargaining unit
7/ continued
tional Union of North America, AFL-CIO, which represents the
mailhandler craft; and the National Rural Carriers Associa
tion, which represents the rural carrier craft. (DX 13, 14,
TR 2169) .
8/ Senior qualified positions in the clerk craft are both
level 5 and 6 craft expediter, distribution expediter, window
clerk, finance clerk, window technician and LSM operator. The
only senior qualified position in the carrier craft is carrier
technician. The senior qualified positions for the mailhandler
craft are mailhandler technician, sorting machine operator and
mail processing machine operator. (TR 2197, 2199, 2220).
9/ Best qualified positions include postal source data
technician, accounting technician, training technician and data
collection technician. (TR 2206).
10
employees; (2) managers above the initial level supervisors,
including subordinate supervisors, tour superintendents,
station managers, and managers who supervise supervisors and
subordinate managers. (TR 2220-2222); and (3) non-craft,
non-supervisory or managerial employees including personnel
assistant, safety assistant, PAR counselor, E.E.O. counselor and
customer service representative. (TR 2223-2224).
d.(i) Promotion to Initial Level Supervisory
Positions
The promotion process underwent two significant changes
since 1970. (PX 9) In 1968 the Post Office used two
written examinations, the OS-100 for the post office branch
and the VS for vehicle services, for the selection of
initial level supervisors. (RE 169).— '7 In oraer to be
placed on the supervisory register, which is a list of employees
who passed the written examination, an employee had to attain a
score of 70. (PX 9) One half point was added to the score
for each year of the employee's service and employees were
allowed to receive up to 13 and 1/2 service credits. (Iji.).
The top 15 percent of these employees, based upon the exami
nation score and service credits, were placed in the "zone of
consideration". (_Id. , TR 2225). ^ Only those employees
10/ Eligibility to compete for supervisory positions was
TTmited to those career employees in the post office branch of
the postal service at level PFS-4 or higher who had a total of
4 or more years of service in the clerk, letter carrier,
mailhandler, rural carrier, special delivery messenger craft,
or a combination of those crafts. (PX 9).
11/ Promotions to initial level supervisory positions in
vehicle services were taken from the top 5U% of the persons on
the register. (RE 22).
who were in the zone of consideration on the supervisory
register were notified of vacancies; persons interested in
being considered had to return a form expressing their
interest. (PX 9; TR 2231). In 1970 the OS 100 examination
was administered again. (RE 169). The old register was
abolished and a new register of eligibles was developed
based upon the same criteria as had been used in 1968 . (I_d.)
In June 1972 the "zone of consideration" standard was
eliminated and eligibles on the supervisory register who
had received an examination score of 70 or better were eval
uated ana rated by their supervisors. (PX 9). Employees
receiving an "A" rating were eligible for immediate promotion;
employees receiving a "B" rating were to be re-evaluated in
6 months; and those receiving a "C" rating were to be
reevaluated in twelve months. (PX 9). In 1973, the OS 100
examination was administered. (RE 170). However, eligible
employees on the 1970 register were not required to re-take
the examination. (Id.) The passing score was lowered to
55, service points were added, (Id_. ) and the use of the ABC
supervisor evaluation process remained in effect. (L d .)
In 1974 a new OS 100 examination was administered,
and a new register was established. (RE 170) Employees
with one year of service were eligible to take the examination,
and employees obtaining an examination score of 50 or better
could be considered. (PX 97) Service credits were not con-
s iaered. (I d .)
12
In June 15, 1976 the qualifying examination was eliminated;
instead employees were required to complete a training program
as a pre-condition for promotion. (PX 9.) Persons with a
minimum of 1 and 1/2 years of postal service employment who had
fulfilled the supervisory training requirement were eligible for
promotion. Eligible applicants were subsequently rated by
their supervisors pursuant to the "ABC" rating system. (I d .)
On February 27, 1978 the Profile Assessment System of
Supervisors (PASS), the second significant change in the process,
1 2 /
was introduced and remains in effect.— Employees with one
year of service are eligible to participate in the program. The
initial level supervisor positions covered by PASS are divided
into 10 "families" of jobs. (PX 2). In order to be eligible for
consideration for a specific initial level supervisory position,
an employee must be on the register for the family of jobs which
includes the position. Both the applicant and his/ her supervisor
evaluate the candidate's qualifications. Based on the self-assess
ment and supervisory assessment, candidates receive ratings of
"excellent," "good," or "some" from the National Test Administra
tion Center (NT AC). (PX 2; 9; 10).
Candidates receiving an "excellent" rating (or a "good"
rating when there are not three candidates in the "excellent"
category) are reviewed by a Promotion Advisory Board, which
makes its selection of the three best qualified candidates. (PX
2, Chapter 530). The PASS program does not provide written
12/ The Jacksonville Post Office established its initial
PASS register on March 2, 1979. (PX 9).
13
standards to the promotion board. (RE 173). The Board submits
the names of the candidates to the Postmaster who selects the
candidate. (PX 2, Chapters 537.431 and 537.433).
(ii) Promotions to Upper Level Supervisory and
Managerial Positions, and Non-Bargaining
Unit Positions
The X-118 Handoook and Personnel Handbook P-12A, and
Personnel Handbook P-11 prescribe the processes and require
ments for upper level supervisory and managerial positions. (PX
•
2, Chapter 510). In order to be eligible for an upper level
supervisory position an employee must have first served as an
initial level supervisor. These positions can be filled competi
tively by management nomination and employee application. (PX 2,
544.111). An eligible employee submits a Form 991-A, and his/her
immediate supervisor completes a Form 991-B. (PX 2, 544.3) A
review committee of at least three persons is established to
select three candidates for recommendation to the selecting
official. (PX 2, 544.5). The committee recommends at least three
but not more than five candidates to to the selecting official
who then makes the final selection.
(iii) The Role of Promotion Advisory Boards
In The Promotion Process
The Post Office uses "Promotion Advisory Boards" ("PAB")
which are responsible for selecting candidates for promotions to
supervisor. The Board is generally composed of four permanent
members selected by the Postmaster and his staff, and ad hoc
members chosen from among the managers in the unit where the
1 4
supervisory vacancy exists. (Id^ ) .— Until recently the PAB ' s
have been comprised primarily of Whites. (TR 1533; TR 2797-99;
TR 3802; TR 3121; TR 3761; PX 98). The Board reviews all of the
applicants for the position, rejects those whom it feels should
not be interviewed and interviews the remaining applicants. The
initial review is standardless. (TR 2733). There are no written
directives governing the operation of the PAB (TR 3814; TR 3170;
TR 3815) and members of the PAB are not required to utilize any
standard forms or guidelines in developing their evaluations of
an employee. (TR 2548; TR 2733). There are no written directives
governing the operation of the Board (TR 3814; TR 3170; TR
3815). There are no guidelines as to what type of questions to
ask during the interview. (Tr 2733-34).— ^
Postmaster Workman makes the final decision as to which
candidate to promote to supervisor. (TR 2449; TR 2231-2232).
He does not refer to standard guidelines or procedures in making
13 /
13/ There are no requirements governing what documentation is
to be reviewed. Boards are not necessarily furnished copies of
the job description for the position being filled. (TR 2729)
Some Boards review only the applications, while others review
official personnel folders. (TR 3815; TR 3132) The criteria
which are considered vary and are based on personal judgments.
(TR 2548-49; TR 3814; 3834-36; 3808, 3812, 3813). Board members
base their recommendations on a review of subjective factors as
diverse as work habits; poise; attitude; ambition; temperament;
assertiveness, membership in organizations; or ability to
communicate; (TR 2700-5; TR 2826; TR 3125-26; TR 3107; TR 2763).
During interviews PAB members consider "personality type",
"manner of breeding", or "social amenities", or the wearing of a
suit. (TR 2764; 3838; TR 529-31, 534). As defendant's wit
nesses so aptly stated "It's a judgment call" (TR 2734) or a
matter of "intuition". (TR 3833). Selections may be based on
undocumented informal discussions or casual observations
outside the scope of the interview. (TR 3808, 3812, 3813).
16
his selections.— 7 t ™ 2451-52, 255-<2551 ; 2574). Prior to
the PASS program, the Promotion Advisory Boards recommended
, 15/
applicants in numerical order, according to their rank,
d . Components of the Personnel System
(i) Performance Evaluations
Chapter 370 of the Employee and Labor Relations Manual
establishes the standards governing performance evaluations for
bargaining unit employees (DX 13). Since 1971 employees have
received evaluations during their ninety day probationary period
at 30, 60, and 90 day intervals. (DX 13, 376, 377, 378; TR 4323).
Evaluations which are made after the probationary period are used
to form the basis for promotion recommendations and step increase
recommendations. (DX 13, 378.2). Employees are rated outstanding,
satisfactory, or unsatisfactory. (^d. 373.21). However, since
1971 the only formal performance evaluation craft employees have
received has been during the probationary period. (TR 2258).
Subsequently, there is no particular standard a supervisor uses
to evaluate an employee, and supervisors aiscretionarily
evaluate employees "whenever they feel the need." (TR 2658; TR
4471). Under Chapter 4 of the Employee and Labor Relations
Manual, non-bargaining unit employees in the supervisory, profes
sional, clerical, administrative and EAS Schedule are to be
14/ Workman testified that he bases his decisions as to which
candidate to promote on his personal knowledge of the employee.
(TR 2581-82) .
15/ This procedure is in violation of postal service regulations
wnich prohibit the use of rank order. See generally William v .
Vukonich, 720 F.2d 909 (6th Cir. 1983).
17
evaluated by management on a regular basis according to the Merit
Performance Evaluation Program. (DX 13, 416.1; PX 14).
(ii) Details
Detailing is a temporary assignment to a higher position
in which a vacancy exists. (PX 20, DX 13). Details to supervi
sory level positions, "204B details", provide employees with an
opportunity to acquire experience, thereby enchancing their
qualifications and making them more attractive candidates for
promotions. (I d ; TR 2728; TR 2809; 2834; TR 3806— 07; 3814).
There is no application process to be considered for a detail (TR
743, 745-76 4312); thus employees do not bid on them (RE 174),
and no records of the names of employees requesting details are
retained. (TR 4321). There are no standards or written guidlines
used to select employees for details. (TR 2696, 2728). Selections
are left up to the individual supervisor. (TR 2696, 2728; 2809;
2834; TR 3806-07) .
(iii) Awards
16/There are numerous awards programs:— 7 Supervisors deter
mine, subject to the approval of their superiors, that the
employee's performance merits an award. They do not utilize any
16/ Service Award for productivity, and efficiency, inventions
arid outstanding achievements; suggestion and incentive award for
submission of ideas for improving efficiency; safety award for
accident free records; superior accomplishment award for specific
achievement in an area; superior performance award for above
average or outstanding performance; and quality step increase in
recognition of extra competence. (DX 13, Chapter 6, PX 2). The
amounts of the awards range from $25.00 to $5,000.00. (!£• ) •
18
standards or written guidelines in determining award recipients.
(TR 2041). In theory, managers review the award recommendations,
but since they do not have knowledge of the employee's work,
they cannot know whether a recommendation or failure to recom
mend is correct. (TR 3040). Personnel does not review awards
to determine whether they were appropriately and fairly given.
(TR 2309 ) .
(iv) Discipline
Article XVI of the National Agreement, the Employee Leave
Policy and Attendance Control Program, and the Work Rules, set
forth the major standards relating to discipline. (DX 14; 27;
28). Tne personnel handbooks for various positions and the
Employee and Labor Relations Manual contain general provisions
relating to discipline. (DX 41; 13). The categories of offenses
are: unsatisfactory performance of duties; attendance problems;
improper conduct; failure to obey instructions; insubordination;
and offenses involving motor vechcles. (DX 41; 13). Disci
plinary action runs the gamit in severity from oral counseling
or a letter of warning to suspension or dismissal. (DX 14).
Barbara Kawchak, a White female, has been a labor relations
assistant in the Office of Employee and Labor Relations since
December 1979. (TR 2933). Ms. Kawchak reviews and processes
disciplinary actions, handles the grievance machinery, and
represents management in arbitration proceedings and Merit
System Protection Board matters. (I_d.) . Kawchak independently
determines the nature of the disciplinary offense by considering
the seriousness of the offense. (TR 2942). She does not refer
19
to the Employee and Labor Relations Manual in performing her
tasks. (TR 3008). She also independently determines what type
of offenses are serious based solely on her experience in labor
relations. She does not refer to the Employee and Labor Rela
tions Manual in performing her tasks. (TR 3008; TR 2942-43).
Kawchak did not obtained the training crucial to her
position until 1981, after she began performing these functions.
(TR 2971-72). Moreover, she rarely if ever objects to or
rejects supervisors' disciplinary recommendations. (I d .,
17/2973-3017).— Postal Service officials have not met with
the supervisors who initiate disciplinary actions to discuss
when or what type of disciplinary action is appropriate. (TR
2973). No one else in the personnel office reviews supervisors'
determinations. (TR 2973). Kawchak's trial testimony that
decisions are reviewed by the Director of Employee and Labor
Relations was inconsistent with her deposition testimony that no
one reviews her decisions. (TR 2974).
The rules involving discipline do not specify: a) what
constitutes an infraction; b) how the matter should be pro
cessed; and c) what penalty should apply. (TR 2978-79).
The personnel handbooks merely state that disciplinary action
can result from violations of any rules, but do not state what
specific type of action results from an infraction of a rule.
(TR 2976). The "Work Rules" do not specify the penalty which
17/ The first of Kawchak's two meetings with supervisors on how
they should discipline is appropriate was conducted in 1980.
(TR 2976-77) .
20
will be incurred if the rule is violated. (TR 2992). There are
no standards or guidelines which determine penalties (TR 29736;
or the seriousness of the offense (TR 2973). Disciplinary in
fractions are undefined (TR 2976). "Insubordination", "improper
conduct", or "unacceptable behavior" cover a multitude of unde-
18/rined behavior. (TR 2944).— ' Performance related disciplinary
actions have been been developed on an ad hoc, and standardless
basis (TR 4314-15).
Supervisors unilaterally determine whether to initiate
discipline and what penalty to associate with the disciplinary
19/
action. (TR 3011-12).— Employees are supposed to have an
opportunity to give a factual explanation as to the circumstances
surrounding the proposed action. However, the forms are no more
than statements supplied by the supervisor. (TR 3000-01).
Although supervisors are supposed to consider the mitigating
circumstances for an employee's behavior (TR 3010), employees
infrequently provide personal statements as to what occurred.
(TR 2998 , 3062) .
In theory, upper level supervisors review the recommenda
tions of subordinate supervisors to verify their appropriateness.
18/ Thus, supervisors have disciplined employees for "sitting
improperly on test bars", (TR 3148) or "remonstrating" (TR 1729)
or "loitering at the coke machine". (TR 1410-12).
19/ Supervisors have total discretion in determining whether an
employee's request for emergency leave is acceptable. (TR
2991), whether to accept a request for excusable tardiness, (TR
2725-26; TR 2781) and which employees to place on sick leave
restriction. (TR 3011). Although there has been a computerized
system for monitoring the timeliness of employees since the
middle seventies (TR 2723), supervisors do not uniformly rely
upon it, but discretionarily determine when to notify an employee
or discipline an employee who has been late. (TR 1726).
21
However, upper level managers rarely, if ever, revise or reject
these recommendations. (TR 3054-56). Whether the manager
reviews or investigates the supervisor's recommendation depends
strictly upon the manager's perception of the occurrence. (TR
20/
2312; 3151) .— /
The only avenues to challenge unfair disciplinary actions
are to file a grievance; appeal to the MSPB; or file an EEO
complaint. (TR 2944, 2949-51; TR 2693). However, the first two
processes are not available to probationary employees. (TR 4323).
Discipline as implemented by the Jacksonville Postal
Service has a cumulative negative effect on Black employees,
since the disciplinary record is considered each time an
employee is disciplined in order to determine what type of
disciplinary action is to be taken. (TR 2944). Defendant's
entire disciplinary procedure is discretionary controlled and
arbitrarily implemented by supervisors who subjectively and
unilaterally determine what type of conduct warrants discipline
and what type of penalty to impose.
5. The Statistical Evidence
a. Introduction
The plaintiffs introduced into evidence a variety
of statistical data. The statistics were either "descrip-
20/ The managerial staff, which has no direct contact with^
employees, may by-pass the supervisor in order to take disci
plinary action against Blacks. (TR 3053; TR 3101). In the
cases of class members Leroy Robinson and Kenneth Rosier the
supervisors who would have been most affected by their conduct
did not concur in the respective decisions to terminate them.
(TR 3101; TR 3052).
22
The descriptive statistics show2 V
tive" or "inferential",
the distribution of Blacks and non-Blacks among different levels
and positions; persons who are supervisors, who receive promo
tions, details, and awards year by year; and the frequency,
extent, and nature of discipline. (PX 1; 120; 121).
Data was coded from employee record (SF-7) cards and from
disciplinary, detail, ana training logs and key punched into
22/
computer readable form. (TR 30-36) plaintiffs also relied
on other sources of statistical information, including original
records, Affirmative Action Plans developed by the Post Office,
and other reports reflecting the workforce. (PX 22; 23).
b. Graae Level Distribution ana Job Assignment
Blacks are far more likely to be at Level 4 or lower jobs
̂. . 23/
than Whites, and far less likely to be in Level 7 or higher.
The Affirmative Action Plans prepared by the defendant confirms
that Blacks have been concentrated in the lower levels throughout
the period. (S e e , e.g., PX 22 Equal Employment Opportunity Plan
for FX 1976, Attachment #1). Blacks consistently have occupied
21/ Compare EEOC v. American Nat'l Bank, 652 F.2d 1176,
TT 89 -1190 (4th Cir. 1981), with Hazelwood School District v .
United States, 433 U.S. 299, 308, n.14 (1977).
22/ The analyses were developed by Dr. Carl Hoffman, a recognized
expert in the field of statistics and statistical methods
applied to employment discrimination. (PX 1d; TR 11-16).
23/ (PX 1, Table 1.2; TR 261). Applying the chi-square test, a
standard method of statistical analysis, it is clear it is
highly improbable (1 chance in 10,000) that the differences in
grade distribution that are shown by Table 1.2 could have
occurred by chance alone. (TR 261-62).
23
the lower level and lower paying jobs. The probabilities that
the observed distributions would occur by chance are extremely
low, only 1 in 10,000 in each year. (PX 1, Table 1.3; TR
263-65) /
Of particular importance is the under-representation
of Blacks in supervisory positions. In 1969 only 1% of Blacks
held supervisory positions, compared to 12% of Whites. In 1981,
only 5% of Blacks compared to 10% of Whites held such positions.
Thus, while Blacks have consistently been 35% of the workforce,
in 1969 they held only 5% of all supervisory jobs and in 1981
only 21%. (PX 1, Table 1.3).
c. Promotions to Supervisory Positions
When the average number of Blacks and Whites
in the crafts is used as the pool, Blacks were promoted to
supervisory positions in numbers far lower than expected in the
periods 1969-72 and 1973-76? for the former period the prob
ability of the low rate of Blacks being promoted occurring by
chance was only .0036, or less than 4 times in 1,000, and for
the latter period the probability was only .0149, or less than
15 times in 1,000 25/
The Jacksonville Post Office promotes persons to supervisory
24/ The statistical evidence also shows that Blacks were less
likely to be hired directly into career appointments than
Whites. Moreover, of those persons initially receiving career
appointments, Blacks were more likely than Whites to be hired
into levels 4 or lower. (PX 1; Tables 2.3, 2.4, 2.5, 2.6, 2.7,
2.8, 2.9.) The differences in a number of years were statisti
cally significant.
25/ It was not until the period 1977-81 that Blacks began to
be promoted in numbers comparable to their representation in the
craft population. (PX 1, Table 3b; TR 316.)
24
positions almost entirely from within its workforce. The key to
the underrepresentation of Blacks in supervisory positions was
their high underrepresentation on the supervisory registers from
which supervisors were selected. The probabilities that the
numoers of Blacks on the registers, given their 35% representa
tion in the craft workforce, could have occurred by chance
ranged from .00000000000015, or 15 chances in 100 trillion, in
1968, to .00067, or 67 chances in 100,000 in December, 1977.
26/
(TR 311-12.)— The underrepresentation existed whether
supervisory registers were developed from the use of tests
alone (1968-73), tests plus the ABC rating system (1973-1975),
or the taking of supervisory training (1976-78). Since the
underrepresentation resulted in the under-selection of Blacks
for supervisory positions during most of the period at issue
26/ Table 1
Representation on Register, by Register and Race,
Compared to Representation in Craft
Course: PX 1, Table 2.
Craft Registers Probability
Register White Black White Black
1968 1084 583 86 3 . 1 4809484 D-12*
1 9 7 CJ 1070 580 123 16 .41599607 D-10
1973 1053 617 216 49 .12062558 D-1 1
1 a74 1055 652 155 71 .13990509 D-0 1
1975 982 627 230 84 .32455825 D-G 6
May'77 949 641 81 26 .24358886 D-0 3
D e c . ' 77 949 641 89 32 .66999744 D-0 3
1978 PM 933 611 14 5 .17084724
Vab 933 61 1 16 9 .44126389
CS 933 611 105 59 .18112493
MP 933 611 116 93 .94911634
* D-12 indicates that in he number to the left the decimal point
should be followed by 12 zeroes, and so on.
25
here; the result of this under-selection is that Blacks continue
to remain substantially underrepresented in supervisory positions.
Defendants' statistics are fully consistent with plaintiffs'
and show a substantial disparity between persons in the work
force and persons on the supervisory registers. Thus, Table 2
of defendant's expert's report (RE 456-57), shows that Blacks
represented 3.3% on the registers in 1968, 6% in 1970, 16% in
1973, 31% in 1974, 27% in 1975, and 24% and 26% in 1977,
tne same figures shown in PX 1, Table 2 (n. 25, supra).
Blacks received an extremely disproportionate share of
discipline (see infra). Disciplinary action can lead to an
individual being barred or removed from the supervisory register.
In PX 121, Table H, plaintiffs showed that in the period 1972-74,
persons who were detailed, whether or not they were on a register,
were more than twice as likely to take the supervisory examina
tions. (TR 4057-58.) Since Blacks were given details signifi
cantly less often than were Whites (see infra). As a direct
result they took the examination less often. (TR 4058.)
d . Details
With regard to details, which have been a crucial factor in
employees progression (see supra), there has been a consistent
pattern of statistically significant differences in the assignment
of Blacks to details as compared to Whites in the periods 1968-70
and 1973-76, and the years 1972, 1973, and 1974 for details from
26-
levels 5 and 6 (PX 1, Tables 12b and 7; TR 321-22, 339-40). The
evidence shows that in a typical year (1973) approximately the
same number of persons were detailed from registers as off
registers, and that the probability of seeing the low number of
Blacks detailed was only .01 (1 in 100) (PX 1, Table 8; PX 121,
Table C ) .
The defendants had no explanation for these differences
except for the speculations of their expert. Dr. Beckett opined
that details were not made between stations, a speculation not
supported by the record, and that there might be differences in
levels of interest in details. (TR 3314-15). He never verified
this proposition (TR 3573-77) and it was not borne out by the
evidence. The defendants' showing relating to details has
a serious methodological flaw, which Dr. Beckett admitted
on cross-examination. (TR 3593-94; 3571-72). In DX 25, Table
12 (see RE 467), the probabilities are calculated by using
all craft employees, both those on and off the supervisory
register, as the pool. This, however, skewed the results since
Blacks were severely underrepresented on the register. (TR
3567-69). When the figure for "expected Blacks detailed" is
computed using the proper source pool, the figure dramatically
rises and the levels of probability dramatically drop. (Compare
DX 25, Table 12, with PX 120, Table R ) .
e . Awards
Blacks were less likely to receive superior accomplishment
and suggestion awards. When the figures are aggregated from
1973-81, the probability is only .0244, or less than 25 in 1000,
that the low number of awards to Blacks could have occurred by
27
chance. (PX 1, Table 7.1; PX 121, Table D ) . Black employees
were less likely to receive quality step increases, to a level
of high statistical significance, recovery only 1 out of the 26
QSI's given. Since Blacks averaged 35% of craft workers, the
probability of this occurring by chance is .0002. (PX 1, Table
7.1; PX 121, Table E ) .
f . Discipline
Set out in the margin are figures demonstrating a consistent
statistically significant over-disciplining of Blacks in compari
son to their numbers in the relevent workforce. Blacks were
35-36% of the workforce but wre more than 50% of those dis
ciplined to a high of 67% in one year. The differences in
the level of discipline, whether one examines the number of
disciplinary actions or the number of individuals disciplined,
28/
are at high levels of significance in each of the years 1969-81.
28/ Taple 2
Incidences of Disciplinary Action by Race, Percentage of
Discipline and Year. Source: PX 1, Table 9.1
Year Black Other Total Probability
# % # %
1969 66 54% 57 46% 123 ,001
1970 77 66% 39 34% 116 .001
1971 63 65 % 34 35% 97 .001
1972 51 65% 28 35% 79 .001
1973 38 55% 31 45% 69 .001
1974 100 51% 98 49% 198 .001
1975 324 64% 182 36% 36 .001
1976 161 56% 124 44% 285 .001
1977 22 67% 11 33% 33 .001
1978 204 58% 144 41% 348 .001
1979 154 55% 124 45% 278 .001
1980 178 52% 165 48% 343 .001
1981 305 52% 285 48% 590 .001
28
When the type of discipline is examined Blacks are consistently
subjected to harsher disciplinary actions than are Whites, at
levels of statistical significance for the periods 1969-70,
1971-74, and 1975-78. (PX 1, Table 9.3; TR 358).— '/
The defendant's expert conceded that Blacks are disciplined
28/ continued
Table 3
Individuals Disciplined by Race, Percentage of
Discipline, and Year. Source: PX 1, Table 9.2
Year Black Other Total Probability
# % # %
1969 58 52% 54 48% 112 .01
1970 69 67% 34 33% 103 .01
1971 50 60% 33 40% 83 .01
1972 44 62% 27 38% 71 .01
1973 30 51% 29 49% 59 .01
1974 70 50% 70 50% 140 .01
1975 185 59% 128 41% 313 .01
1976 143 52% 133 48% 276 .01
1977 108 53% 97 47% 205 .01
1978 117 54% 101 46% 218 .01
1979 110 50% 110 50% 220 .01
1980 131 50% 133 50% 264 .01
1981 207 50% 210 50% 417 .01
(TR 358; 360-362).
29/ The data also establishes that in each year from 1969-79,
Blacks received a greater number of step deferments a .01 level
of probability, i.e., at a high level of statistical significance.
(PX 1, Table 27). These differences persisted even when those
persons who had received suspensions or leaves without pay were
excluded. (PX 121, Table A.) The higher rate of step deferments
is related to the lower rate of pay of comparable Blacks and
Whites, as well as to the underrepresentation of Blacks on
supervisory registers. Defendant argued that the difference in
length of time to receipt of step increases was not great. (DX
25, Tables 22-23). However, the defendant's expert's own data
establishes that such differences exist. (TR 4157). Inaeed,
the differences are such that they are often near statistical
significance despite the small numbers involved. (See DX 25,
Table 23).
and admittedmore often than Whites (TR 3366; 3367; 3374)
that he did not determine whether Blacks in fact committed more
infractions that could lead to discipline than did Whites (TR
3363; 3661-62); that is, there was no way to determine whether
the conduct of Blacks and Whites is different, and thereby to
explain the differences in levels of discipline. The frequency
and the severity of disciplinary actions could be determined,
but not whether disciplinary actions were warranted or whether
supervisors initiated the same discipline when faced with the
same conduct of Blacks and Whites.
Defendant compared all persons disciplined with persons
receiving discipline related to attendance, in an attempt to
show that the higher level of Blacks disciplined was caused by
their poor attendance. The analysis, however, was flawed
because it was circular. The defendant's expert defined the
expected percentage of Blacks available to receive other dis
ciplinary actions as equivalent to those who received attendance
30/
30/ Defendant produced tables which examine the severity of
discipline based on the number of disciplinary actions taken.
(DX 25, Tables 24, 25, and 26). These show, e. g. , that persons
with a third disciplinary offense are disciplined more severely
whether they are Black or White. This simply re-emphasizes that
Blacks are disciplined more often, but does not explain the
phenomenon.
31/ Dr. Beckett also argued that Blacks used more leave than
Whites. (DX 68, Tables 48-49). However, the leave was that to
which employees were entitled; that is, neither group came close
to using up their authorized leave. No showing was made that
there was any relationship between the proper use of annual
and/or sick leave and discipline. (TR 4202).
30
receive it, or, conversely, that all persons who should have
received attendance related discipline did in fact receive it.
33/(TR 3651-53; 3654; 4457).— On rebuttal, plaintiffs demon
strated that Blacks who had no absence related discipline action
taken against them, received subjective disciplinary actions at
a grossly disproportionate rate. (PX 120, Table 0; TR 4456).
Defendant's expert acknowledged that he could have developed
an objective basis for determining whether attendance related
discipline was even-handedly imposed by examining time cards
and time records and thereby to overcome the inference of racial
discrimination as the cause of the observed disparity. (TR 4453).
The failure to do so leads to the inference that the data would
not have supported such a conclusion. Indeed, the defendant
showed that discipline for the first attendance related offense
varies widely from case to case. (DX 25, Table 24; RE 479).
Finally, even defendant's approach showed that the probabil
ity of the differences in levels of discipline (DX 25, Table 37
(RE 493)) occurring by chance was .0144, a level that establishes
statistical significance. Moreover, Blacks are disciplined far
less often for offenses related to objective job-performance as
compared to actions involving subjective decisions making. For
example, Blacks are disciplined less often for mail-handling
irregularities, safety, and scheme failure offenses, and far
33/ The Jacksonville Post Office had no objective centralized
mechanism for determining attendance related offenses and
ensuring that all persons who committed infractions received the
same, or any, disciplinary action. Even with regard to attendance
related offenses, the ultimate discretion whether to initiate
disciplinary action still rests with the first-line supervisor.
31
more often for such vague, open-ended offenses as "failure to
follow instructions," "improper conduct," and "miscellaneous."
(Compare, DX 25, Tables 31, 33, and 34, (RE 486, 489, and 490)
34/
with Tables 29, 30 and 32 (RE 484, 485, and 487)).—
h. The Defendants' Attempted Rebuttal
As described above defendant attempted to rebut the prima
facie case developed by plaintiffs through its own statistical
presentation.— ^ The evidence established an underlying
34/ In rebuttal, plaintiffs produced a further analysis of
defendants' discipline studies that re-affirmed that Blacks were
disciplined more often and with greater severity than were
Whites. (PX 1, Tables 24, 25, and 26; PX 120, Table Q ) . Plain
tiffs also demonstrated that, even accepting defendants' figures
in DX 25, Tables 27-37, discipline for attendance related
offenses tends to follow other discipline. That is, defendants'
study is based on the proposition that persons who have attendance
problems have performance problems; therefore, the higher
overall rate of discipline of Blacks is caused by their own
actions in coming in late, being A.W.G.L., etc. (TR 3360-62;
3648-50). The studies of plaintiff, however, show that persons
who have attendance problems are those who were previously
disciplined for alleged performance deficiencies. Thus, PX 120,
Table P, shows that 67.2% of employees who are disciplined for
attendance-related reasons are so disciplined after being
disciplined for a subjective performance related offense. Since
Blacks are disciplined far more often than Whites, the higher
rate of discipline of Blacks for attendance infractions is
caused by their becoming discouraged and offended by having been
subjected to other discipline far more often than warranted by
the predominantly White supervisory staff. (TR 4179-82.)
35/ Plaintiffs were seriously hampered in their ability to
explore the accuracy of defendant's statistical case by the
trial court's refusal to require the production of the computer
data base used to prepare it. (RE 154). It is literally
impossible to determine whether complex statistical analyses are
correct without access to the underlying computerized form of
the data from which the analyses have been extracted. See
Recommendations On Pretrial Proceedings In Cases With Voluminous
D a t a , 39 The Record of the Association of the Bar Of the City of
New York 49, 52-54 (1984).
32
heflaw in the approach of the defendants' statistician; viz.,
developed models of how the employment system should work based
on unverified information obtained from persons in the personnel
office. At time did he attempt to verify this information indepen
dently, although there were a number of opportunities to do so.
For example, with regard to details, Dr. Beckett speculated,
without verifying, that the differences in the rates of details
between Whites and Blacks could be explained by Blacks being
less inclined to sign up for details, or by Blacks being in
locations where there were fewer details. On cross-examination,
in instance after instance Dr. Beckett acknowledged that his
studies were based on unverified assumptions as to the way in
which the personnel system operated based on information provided
by defendants. (See, e .g ., TR 3521-23; 3603; 3624-25; 3629-30;
3638-40; 4455-56.) However, he also testified that in order to
develop an accurate model for statistical purposes a statistician
36/must understand and establish how a personnel system works.— '
C . Standard of Review
1. With regard to Arguments I, II, III, and V, the
district court erred as a matter of law.
36/ Finally, in a number of instances defendant's expert failed
to aggregate results in ways that could easily be done. For
example, Dr. Beckett looked at Quality Step Increases year-by-year
Because of the small number of Q.S.I.'s given each year, it is
not suprising that even though in every year but one Black
received no Q.S.I.s (the "inexorable zero"), in no single
instance were the results statistically significant. However,
when the results are cumulated over the entire period during
which Q.S.I.'s were given (1976-81), the under-representa
tion of Blacks (1 out of 26 Q.S.I.'s) becomes highly significant
at the .0002 level. (PX 121, Table E ) . See Capaci v. Katz &
Besthoff, Inc., 711 F.2a 647, 655-56 (5th Cir. 1983), for a
cogent criticism of disaggregation by looking at data year-by-year
33
2. With regard to Argument IV the district court
erred as a matter of law and some of its factual findings were
clearly erroneous.
SUMMARY OF ARGUMENT
I. The plaintiff properly exhausted the issue of testing
during the administrative process. The denial of promotions to
Blacks to higher level positions for which tests were required
was raised and was a subject of the administrative investigation.
Therefore, it was error to exclude the issue from this action.
II. The statistical evidence established a pattern of dis
crimination against Blacks in higher level promotions, discipline,
details and awards. The court erred in casting upon plaintiffs
the burden of eliminating all possible non-racial reasons for
the disparities shown and accepting various speculative reasons
advanced by defendants. It was error to rely on the effect of
tests to explain the underrepresentation of Blacks in supervisory
positions.
III. The district court erred in dismissing plaintiffs'
disparate impact claims. Decisions of the Supreme Court make it
clear that personnel systems may be challenged under a disparate
impact theory.
IV. The district court's findings with regard to class
members must be vacated for reconsideration under Teamsters-
Franks standards. In addition, a number of findings with regard
34
visory personnel."— ^ Moreover, the trial court itself held
that Dr. Hoffman, plaintiffs' expert statistician, failed to
account for the impact of the tests. (RE 247).
Thus, the key question on appeal is whether the district
court's exclusion of testing as an issue was in error. Plaintiffs
submit that even assuming that all other aspects of the defendants'
promotion scheme were proven to be nondiscriminatory, if the test
had a disparate impact there has been a violation of Title VII.
Connecticut v. T e a l , 457 U.S. 440 (1982).
In paragraph 35(e) of the Complaint plaintiffs claimed that
the defendant's:
(e) "Use of written personnel tests as a condition of
promotion which tests disqualify Black employees in sub
stantially greater proportions than they disqualify white
employees and which have not been shown on the basis of
professional validation studies to be predictive of job
performance."
(R 8). On September 29, 1972 defendant filed a Motion to Dismiss
on the ground that "The Court lacks jurisdiction over the subject
matter because Plaintiffs failed to exhaust available administra
tive remedies . . . ." (R 108-117). Plaintiffs, in response,
argued that the plaintiff Ernest L. Griffin alleged in his admin
istrative charge of August 29, 1971, that he was denied a promo
tion to the position of "Serviceman", which was conditioned upon
his ability to pass the written examination (R 148). On January
37/ The district court also rejected individual claims of dis
crimination on the ground that the witness had not scored high
enough on a test to be eligible for promotion. RE 270; 313; 341;
362.
36
9, 1973, the court below granted defendants' motion as to para
graph 35 (e), thereby excluding the issue of testing from the
case. (RE 124-25) .
In a motion dated January 19, 1973, to amend or alter the
January 9th Order, plaintiffs argued that the administrative
complaint supported the allegation in paragraph 35(e) with
respect to testing. (R 166). Further, plaintiffs sought an
appeal from the order on the ground that a reinstatement of the
testing issue "may materially advance the ultimate determination
of the litigation." (R 172). The lower court denied plaintiffs'
motion on January 26, 1973 (RE 126). On January 28th the district
court entered an Order confirming its denial of plaintiffs'
motion to amend, and further denying plaintiffs' motion to permit
plaintiffs to take an immediate appeal pursuant to 28 U.S.C. §
1292(b) (R 182). On March 27th plaintiffs filed a motion to
reinstate paragraphs 35(b), 35(e) and 35(h) of the Complaint
(R 177; 271-76). This renewed motion was made consequent to
the deposition of Clarence H. Featherson, the Director of the
Office of Equal Employment Compliance of the Postal Service, in
charge of the administrative processing of Griffin's case. (R
194)
Plaintiffs urged the judges who subsequently presided over
the lawsuit to reconsider Judge Tjoflat's exclusion of the
testing issue (R 436-451; 714-783). They affirmed the original
order on the ground that plaintiffs had failed to exhaust their
administrative remedies as to this allegation. (RE 132). Thus,
this issue was not addressed at the trial.
37
During the course of the administrative investigation
plaintiff Griffin clearly challenged those aspects of defendant's
employment practices and policies which would reasonably include
testing requirements. The administrative record, inclusive of
Griffin's administrative charge and the investigative report,
evidence that the written test requirement was an integral part
of defendants' promotional scheme and was or should have been
encompassed in a reasonable investigation of a charge of systematic
discrimination in promotional practices.
By letter dated August 29, 1971 plaintiff Griffin filed a
third party discrimination complaint against the defendant
stating that "this discriminatory complaint is based on race
since qualified blacks were and are still being systematically
excluded in training ano development and opportunities for
advancement." — ^(R 27). One of the allegations was that:
"Only token blacks are appointed to level seven [the first
supervisory level] and above." (R 63).
The administrative investigation spanned eight days and
included an extensive review and analysis of Griffin's allegations,
numerous interviews, and the compilation of a voluminous mass of
documents. (R 63-107). The Investigation Report states in its
"Analysis and Findings of Fact" that "while the scope of the
investigation centered around Mr. Griffin's thirteen allegations
of racial discrimination, the investigation was not limited to
38/ The Complaint was forwarded to Clarence H. Featherson,
Director of Equal Employment Opportunity Compliance, United
States Postal Service, who accepted it as a proper complaint. (R.
38) .
38
of racial discrimination, the investigation was not limited to
these allegations." (R 95). Further, the Report confirms that
the testing issue was addressed during the administrative investi-
. • 39/gation.—
The purpose of the requirement of administrative exhaustion
is to put the agency on notice of all issues in contention, prior
to allowing the plaintiff to raise those claims in court. Plain
tiffs submit that during the course of the administrative investi
gation, plaintiff Griffin identified aspects of defendant's em
ployment practices which would reasonably include the written test
requirements. The practice was an integral part of defendants'
promotional scheme and should have been encompassed in a diligent
investigation of a charge of systematic discrimination in promo
tions. In fact, Clarence Featherson, the Director of Equal Employ
ment Compliance and who was in charge of Griffin's complaint,
testified that he would expect the investigator to investigate
39/ Exhibits produced during the investigation included 1968 as
well as current supervisory registers of persons who passed the
written supervisory examination. (R. 68). Twenty-five Black
employees were interviewed and affidavits were obtained from each
(r . 71-72). The supervisor's examination was mentioned in almost
half of these interviews. (R. 72-83). For example, the report of
the interview with Joseph Dean, states "Mr. Dean added that the
jobs in which he was interested had as one of their qualifications
that an applicant must be on the supervisor's eligibility roster.
As a result of this, he did not bid on them." (R. 78). Others
attested that they had not advanced because they had not passed
the supervisory examination (R 78, 80). The investigation uncov
ered that Blacks were underrepresented on the supervisory roster,
comprised of employees who were successful on the examination.
(R 93, 95). The investigator in his analysis and findings repeat
edly identified the supervisory examination as an impediment to
the advancement of Blacks in response to allegations 1, 5, 6, 10,
11, 13 (R 95, 97-99, 104, 105, 106-107). The investigation un-
equivocaly stated "Failing to be listed on the supervisory roster
limited those employees' chances for promotion." (R 99).
39
all elements of the promotion procedure including the racial
40/
impact of the supervisory examination.
Further, the "Regional Instructions. Subject: Updated
Procedures for Processing Equal Employment Opportunity Complaints
required "that the scope of the investigation include" at least
"an analysis of any policies or practices related to discrimina
tion even though they have not been expressly cited by the
complaint." (R. 206).— '/ Thus, it is clear that on the basis
40/ Q. If you receive a complaint ... alleging racial
discrimination in promotions ... would you expect your
investigating officer to try to determine whether there is
any element of the promotion procedure that may be discrimi
nating against applicants on the basis of race?
A. Yes, we would.
Q. Since an applicant for a job at level 7 or
above is required to take and pass a supervisory
exam, would your investigating officer be expected to
look at the racial impact of that examination as I
described racial impact before?
A. Of the exam?
Q. Yes.
* * * *
0. Would the investigating officer look at the
supervisory exam as it had been given, perhaps two or
three times, try to determine how many whites and
blacks took the exam and how many whites and how many
Dlacks passed the exam?
A. Re would do that, yes.
Featherson Deposition at 31-32. (R. 195-196).
41/ Featherson testified that the instructions issued on December
TT, 1972 were substantially the same as those in effect at the
time Griffin's complaint was investigated.
40
of Griffin's charge, it could have been reasonably expected
, 4 2 /
that the issue of testing would have oeen investigated.— 7
This Court has recognized that, "The starting point for
determining the permissible scope of the judicial complaint is
the EEOC charge and investigation." Evans v. U.S. Pipe & Foundry
C o . , 696 F .2d 925, 927 (11th Cir. 1983). Eastland v. Tennessee
Valley Authority, 714 F.2d 1066 (11th Cir. 1983). Thus, a determi
nation of whether a plaintiff has exhausted administrative
remedies depends on an analysis of the "fit" between the adminis
trative charge and the subsequent judicial complaint. Ong v .
Cleland, 642 F.2d 316 (9th Cir. 1981).
The record in the case demonstrates that the plaintiff
presented thirteen instances of claims of denials of
promotional opportunities to supervisory positions conditioned
upon successfully taking the supervisory examination. Further,
the record reveals that during the administrative investigation,
the testing issue was specifically raised and addressed. (See
supra.) Clearly, plaintiff's allegations that he had been denied
promotions to supervisory level positions and that there were
only token numbers of Blacks in upper level jobs was sufficient
to put the agency on notice that the written test requirement was
42/ Information from the defendant during discovery confirms
that written tests are an integral part of the promotional
process, and that the testing issue is "like or related to
Griffin's allegations of discrimination." In response to inter
rogatories propounded by plaintiffs, defendants stated that "Any
employee who wishes to be promoted must meet the requirements for
the position he seeks which are listed in Personnel Handbook p.
11, . . . A written test is required for consideration for
promotion to all initial level supervisory positions." (Def.
Ans. to Pi. First Interr., Interr. No. 16).
41
at issue. In analagous situations, but where the facts are not
as compelling as those presented here, the courts have reasoned
that an employee need not identify a particular position or
specific employment device at the administrative level in order
to exhaust administrative remedies. S e e , e,g. , Mangiapane v.
Ad am s, 661 F.2d 1388 (D.C. Cir. 1981). Here, plaintiff described
the bases of his claims with sufficient clarity to put the agency
on notice that he was challenging all mechanisms, including
written tests, which precluded Blacks from advancing to super
visory level positions.
This is not a case where the plaintiff failed to pursue any
43/of the administrative remedies available.— As the investi
gator's report points out, Griffin actively participated in the
administrative process and assisted the investigator in his efforts
to obtain witnesses and relevant data. He was cooperative during
the investigation and expended tremendous energy in attempting to
resolve the matter. Thus it cannot be disputed but that Griffin,
exhausted the administrative remedies that were the prerequisite
to filing an action in federal court. S e e , e .g ♦, Chisholm v.
U.S. Postal Service, 516 F Supp. 810, af f’d. in part, vacated in
pa rt, 665 F.2d 482 (4th Cir. 1981).
Moreover, to the extent that the testing issue was not
expressly raised, Title VII actions may embrace not only the
discrimination expressly disclosed in the charge or actually
43/ Compare Siegel v. Kreps, 654 F.2d 773 (D.C. Cir. 1981);
Sampson v. Civiletti, 632 F.2d 860 (9th Cir. 1980). Hoffman v.
Boeing, 596 F.2d 683 (5th Cir. 1979).
42
developed during the investigation, but also discrimination like
or related" to such discrimination. S e e , e .g. , Eastland v.
Tennessee Valley Authority, 714 F.2d at 1067. Therefore, assuming
arguendo, that the testing issue was not considered during the
administrative process, plaintiffs should not be penalized as a
result of the defendant's dereliction in performing its duty to
conduct a reasonably comprehensive investigation. Defendant, who
controlled the administrative process, should not be allowed to
limit the scope of the lawsuit in this action where it was
apparent that the defendant's use of written tests was at issue.
To the extent that it can be argued that the administrative
agency chose not to interpret Griffin's complaint as raising the
issue of testing, this Court should permit those issues to be
addressed. This is particularly true since the governing regula
tions require that the investigation include "...any policies and
practices related to work situations wnich may constitute, or
appear to constitute, discrimination even chough they have not
been expressly cited by the complainant." 5 C.F.R. § 713.216(a).
The Court of Appeals for the Fourth Circuit, when faced with
the very issue before this Court concluded that the plaintiff's
administrative complaint alleging discrimination in promotion
was sufficient to put the Post Office on notice that the entire
promotional system was being challenged to permit the litigation
of testing. Chisholm v. United States Postal Service, supra.
See also Lawler v. Alexander, 698 F .2d 439, 442 (11th Cir. 1983)
(challenge to federal agency promotions included "any discrimina-
43
tory practices in the entire promotion process.") For the reasons
discussed above, we urge this Court to follow the reasoning set
forth in Eastland and adopt the conclusion rendered in Chisholm.
II. THE STATISTICAL EVIDENCE ESTABLISHES DISCRIMINATION IN
PROMOTIONS, DETAILS, DISCIPLINE, AND AWARDS.
A. Introduction
The district court below found against the class on the
basis of an analysis of the statistical evidence that was
seriously flawed by legal error. The statistics of both parties
established the following:
1. Although Blacks comprised 36% of the workforce they
have held supervisory positions at far lower levels. (PX 1,
Table 1.3.)
2. Blacks have been underrepresented on the supervisory
registers from which supervisors are selected at statistically
significant levels. (PX 1 Table 2; DX 25, Table 2; RE 456-7.)
3. Blacks have been underrepresented in details which
could give them experience for future promotions to supervisory
positions. (DX 25, Table 12; PX 1, Tables 12b, 7; RE 467.)
4. Blacks have been overdisciplined at statistically
significant levels. (DX 25, Table 37; PX 1, Tables 9.1, 9.2; RE
493 . )
Movement into supervisory craft positions is entirely from
internal promotions out of the craft workforce. Eligibility for
consideration for supervisory positions is conditioned on workers
44
being placed on the supervisory registers. There have been a
variety of ways by which placement on the supervisory registers
was achieved, including the use of tests. (See Chisholm v. U.S.
Postal Service, 665 F.2d 482 (4th Cir. 1981).
It is well established that if promotions into supervisory
positions are from the craft workforce, a showing that Blacks are
underrepresented in supervisory positions at statistically signi
ficant levels casts the burden on the defendant employer to
explain the underrepresentation. Payne v. Travenol Laboratories,
673 F .2d 798 (5th Cir. 1982); Carroll v. Sears Roebuck & C o . , 708
F .2d 130, (5th Cir. 1983). Thus, the appropriate comparison in
an internal promotion case is the workforce from which promotions
ordinarily take place. See also Davis v. Califano, 613 F.2d 957
(D.C. Cir. 1979); Page v. U.S. Industries, Inc., 726 F.2d 1038,
1047 (5th Cir. 1984) .
With regard to discipline, courts have also held that once a
statistically significant overdisciplining of Blacks has been
shown, the employer may not rely on assumptions that the disci
pline has occurred because the Blacks' conduct deserves it. See
Lilly v. Harris-Teeter Supermarket, 720 F.2d 326, 336, 337, n. 22
(4th Cir. 1983); Chisholm v. U.S. Postal Service, 516 F. Supp.
810, 850 (W.D.N.C. 1980) aff'd, 665 F .2d at 496-97. The burden
is on the defendant to estabish a non-discriminatory reason for
the imbalance.
The district court, however, consistently cast on plaintiffs
the burden of proving that there were no non-racial reasons for
45
In so doing, the court failed to appre-
44/
the disparities.
ciate the history and purpose of Title VII and Section 717 of the
Equal Employment Opportunity Act.
When Congress passed the Equal Employment Opportunity Act of
1972 it recognized that the issue of employment discrimination
was more complex, far reaching, and entrenched than had been per-
45/
ceived in 1964.— With regard to agencies of the federal
government Congress found in the concentration of Blacks in the
lower grade levels evidence both of employment discrimination and
of the failure of existing programs to bring about equal employment
46/opportunity.— The present case presents the same pattern
44/ See, e. g., RE 434; 438-440.
45/ In 1964, employment discrimination tended to be viewed
as a series of isolated and distinguishable events, for the
most part due to ill-will on the part of some identifiable
individual or organization . . . Experience has shown this
view to be false.
Employment discrimination as viewed today is a far more
complex and pervasive phenomenon.
S. Rep. No. 92-415 (92nd Cong., 1st Sess., 1971) p. 5.
46/ The house Report stated:
Statistical evidence shows that minorities and women continue
to be excluded from large numbers of government jobs,
particularly at the higher government levels ....
This disproportionate distribution of minorities and
women throughout the Federal bureaucracy and their exclusion
from higher level policy-making and supervisory positions
indicates the government's failure to pursue its policy of
equal opportunity.
H. Rep. No. 92-238 (92nd Cong., 1st Sess., 1971), p. 23. The
Senate report also included statistics which showed the concen
tration of minorities in the lower grade levels, and concluded
that this indicated that their ability to advance to the higher
grade levels had been restricted. S. Rep. No. 92415 (92nd
Cong., 1st Sess.) pp. 13-14.
46
that led Congress to extend Title VII to federal agencies;
Blacks are largely relegated to lower positions.
Title VII, of course, is based on the proposition that in an
employment system that is fair and neutral with regard to race,
one would expect to see persons receiving employment benefits on
an equal basis irrespective of their race. Thus, if the issue is
hiring, one would expect to see a workforce reflective of the
workforce from which employees are hired. Teamsters v. United
States, 431 U.S. 324, 339 n. 20 (1977). If the issue is internal
promotions one would expect over a period of time to see Blacks
distributed fairly through the workforce. Davis v. Califano, 613
F.2d 957, 963-64 (D.C. Cir. 1979). Indeed, it was this expecta
tion and its disappointment that led Congress to conclude that
minority federal employees suffered from employment discrimination
, . 47/
and that corrective action was needed.—
The burden on plaintiffs in a Title VII action is not to
disprove every conceivable explanation for a maldistribution
of Blacks in the workforce, but to show patterns which demonstrate
that the underlying presumptions of Title VII are not met. Blacks
are disproportionately in lower grades; fewer are promoted and
• • 48/ ,
they are underrepresented in supervisory positions. The
burden then shifts to the defendant employer to come forward with
47/ See S. Rep. No. 92-415, supra, pp. 5-6.
48/ See Hazelwood School District v. United States, 433 U.S.
299, 307 (1977); Teamsters v. United States, 431 U.S. 324,
336-338 (1977); Barnett v. W. T. Grant C o . , 518 F.2d 543, 549
(4th Cir. 1975); Segar v. Civiletti, 508 F. Supp. 690 (D.D.C.
1981); Grant v. Bethlehem Steel Corp., 635 F.2d 1007, 1018 (2nd
Cir. 1980).
47
legally sufficient reasons for these disparities and maldistribu
tions .
The principle that the alleged discriminator may not simply
rely on assumptions that Blacks are less qualified or are more
likely to have engaged in disqualifying behavior has its source
49/
in jury discrimination cases.— As long ago as Neal v. Delaware,
103 U.S. 370, 397 (1881), the Supreme Court held that a prima
facie showing of discrimination through statistical evidence
could not be overcome by indulging in the "violent presumption"
that Blacks lacked the qualifications for jury service. In
Patton v. Mississippi, 332 U.S. 463 (1947), the Court rejected
the argument that a higher crime rate among Blacks could be
presumed to explain their disqualification from jury service.
Similarly, an employer cannot simply sit back and demand that the
plaintiffs counter every speculative explanation that may be
■ 50/
invented by a fertile mind.
In sum, Title VII provides an effective remedy to
correct the historical denial to Blacks of equal opportunity and
a fair share of employment benefits. It is a remedial statute
and must be construed in light of the problems it was passed to
49/ The jury cases are applicable to disparate treatment claims
Tn a Title VII case. S e e , e.g., Teamsters v. United States, 431
U.S. 324, 342, n. 24 (1977), citing Alexander v. Louisiana, 405
U.S. 625, 632 (1972).
50/ See Segar v. Civiletti, 508 F. Supp. 690, 712 (D.D.C. 1981),
citinq Davis v. Califano, 613 F.2d at 964. Grant v. Bethlehem
Steel Corp., 635 F.2d 1007, 1015 (2nd Cir. 1980); Pegues v.
Mississippi State Employment Service, 699 F.2d 760, 769 (5th Cir.
1983); EEOC v. American National B a n k , 652 F.2d 1176, 1186-89
(4th Cir. 1981); Capaci v. Katz & Besthoff, Inc., 711 F .2d 647,
654 ( 5th Cir. 1 983) .
48
address and correct. The statistical evidence in this case
estaolishes a consistent pattern of discrimination and disparate
treatment of Black employees at the Jacksonville Post Office
which requires the conclusion that Title VII has been violated.
B . Supervisory Positions and Details
The evidence establishes that Blacks are underrepresented
in supervisory positions when compared with their representation
in the eligible workforce. The underrepresentation has at all
times been at statistically significant levels and has ranged
from 5% in 1969 to 21% in 1981.
When, as here, the basic qualification for promotions to
supervisory positions is experience in the basic craft workforce,
a showing of disparities such as here is casts upon the employer
the burden of coming forward with a legally sufficient explanation
for the d i f f e r e n c e ' See, e.g., Carroll v. Sears Roebuck &
C o . , 708 F .2d 130 (5th Cir. 1983). The only qualification in
addition to coming from the craft workforce for eligibility for
supervisory positions was that employees must be on the supervisory
register.— ^ Persons not on the register cannot apply for a
position, or if they do they are automatically disqualified.
Defendant's statistics purport to show that selections
from the supervisory registers are in appropriate amounts. How-
51/ Thus, the court's heavy reliance on Valentino v. U.S. Postal
Service, 674 F.2d 56 (D.C. Cir 1982) was misplaced. The positions
at issue in Valentino were not promotions out of industrial-type
jobs, but promotions into high level administrative, professional,
and managerial positions. This case is much more like Teamsters
than Hazlewood.
52/ RE 169-70. See also Chisholm v. U.S. Postal Service, 665
F.2d at 495, n. 18.
49
ever, the defendants have failed to explain the underrepresenta
tion of Blacks on the registers themselves, except to point to
the possibility of a lower pass rate of Blacks on the various
tests that were used to select persons for the supervisory
registers. (S e e , e . g . , R. 3349.) However, defendants are
precluded from relying on these tests as an explanation.
In Part I of the argument we have demonstrated that it
was error for the court below to have precluded plaintiffs from
challenging the tests. A showing that the tests had a disparate
impact would compel a finding that Title VII was violated, even
if the rest of the promotion process did not exclude Blacks
disproportionately. Connecticut v. Te al, 457 U.S. 440 (1982).
Assuming arguendo, that plaintiffs were barred from challeng
ing the tests affirmatively, this would not mean that the tests
could be used by defendant to defend the observed disparity in
promotion rates since defendant can only use an explanation
demonstrated to be lawful and non-discriminatory. S e e , e, g.,
Alexander v. Louisiana, supra; c f ., Texas Department of Community
Affairs v. Burdine, 450 U.S. 248 (1982). This can only be done
by showing that the tests in fact did not result in a lower pass
rate of Blacks, or that the tests are job related (Griggs v. Duke
Power C o . , 401 U.S. 424 (1971); Albemarle Paper Co. v. Mo od y, 422
U.S. 425 (1975)), neither of which has been done. Moreover, as
pointed out above, the underrepresentation of Blacks persisted
even when tests plus a rating system were used, and when tests
were not used at all.
50
The district court therefore erred when it relied on the
effect of the examinations to explain the low rate of selection
of Blacks to supervisory positions until 1977 (RE 246-48). More
over, it was incorrect to rely on defendant's data on applicants
for promotion (RE 246-48; 428; 436), since only persons who were
on the supervisory registers could apply and those persons who
passed the tests were placed on the registers. (RE 169-70.)
Therefore, the disparity between Blacks in the pool.of craft
workers and those holding supervisory positions remains unex
plained .
With regard to details, which provide supervisory exper-
53/ience,— the evidence shows that Blacks have not received
details in numbers representative of their percentage in the
workforce which is eligible for such details. It is also clear
that assignment to details is in the uncontrolled and unreviewed
discretion of the immediate supervisor. Also, there is no
central register of employees wishing to be considered for details.
The only explanations advanced by defendant for the underrepresen
tation of Blacks in details were wholly speculative. Thus,
defendant's expert speculated that Blacks were less interested in
details, even though there is no data whatsoever to support such
a speculation, and that details were only to the immediate office.
C . Discipline
Both parties and the court recognized that Blacks were
subjected to disciplinary action with far greater frequency than
53/ See Walker v. Jefferson County Ho m e , ___ F.2a ___ , 34 F.E.P.
Cases 465 (5th Cir. 1984).
51
were Whites. Thus, Blacks received more incidents of discipline
ana, since the level of discipline is dependent on the number of
disciplinary actions against an individual, Blacks received more
severe discipline. The differences, regardless of the methodology
used, were at statistically significant levels.
The district court accepted the statistical analysis of
discipline done by defendant's expert, Dr. Beckett, even though
his analysis withstands scrutiny only if one is prepared to
engage in circular reasoning and implausible assumptions. The
first branch of his analysis proceeds on the assumptions —
without proof — that persons who receive attendance related
discipline are more likely to receive performance related disci
pline and that attendance related discipline was meted out uni
formly and in a non-discriminatory fashion; ergo, prior to test
ing to determine if statistically significant differences exist
in performance related discipline, the data must be adjusted to
. , . 54/
eliminate differences in attendance related discipline.
(Tr. 3644, 3647.) As a result of this sleight of hand, Dr.
Beckett was able to dramatically increase the number of attendance
related disciplines Blacks might be expected to receive from 288
(35%, a number equal to the proportion of Blacks in craft jobs)
to 492 ( 60%) /
54/ Dr. Beckett made this adjustment by artificially setting
the expected number of attendance related discipline equal to
the actual number of such discipline. (Tr. 3644, 3647.)
55/ The entire purpose of the statistical analysis is to measure
the difference between the expected number of personnel actions
and the actual number of those actions to determine whether that
52
Despite this adjustment, a statistically significant level
of disparity remained in the amount of discipline received by
black and white employees. (RE 493.) Dr. Beckett sought to
explain away the remaining difference by a number of untested
assumptions. First, he asserted that Blacks receive more
discipline because of their relative youth, without demonstrating
any causal relation. He then assumed that the difference in
discipline could be explained by an actual difference in conduct,
although he investigated no data to support that assumption. He
also assumed a causal relation between level of attendance related
discipline and performance related discipline without establishing
55/ continued
difference could be explained by chance. See Hazelwood School
District v. United States, 433 U.S. 299, 308, n. 14 (1977);
Pegues v. Mississippi State Employment Service, 699 F.2d 760
(5th Cir. 1983); Lilly v. Harris-Teeter Supermarket, 720 F.2d
326, 336, n. 19 (4th Cir. 1983). If one sets the expected
number equal to the actual number, as did Dr. Beckett, there
will be no difference, and no standard deviations.
If the Hazlewood Pegues-Lilly calculation is made using 288
expected Black attendance related discipline, the result is 14.9
standard deviations, substantially greater than in Lilly (9.71).
Using the formula as set out in Lilly (720 F.2d at 336, n. 19)
with the figures in DX 25, Table 37 (RE 493):
n = total employees disciplined for attendance violations
p = black percentage of the workforce
1-p = white percentage of the workforce
Q = actual number of Blacks disciplined
E = expected number of Blacks disciplined = np
S = standard deviations = >/np (1-p)
number of standard deviations = Q - E
S
Here, n = 825, p = .35, 1-p = .65, Q = 492, E = np =228,
S = 13.68, and number of standard deviations = Q -E
S
= 492 - 288
13.68 = 14.9
53
such a link and sought to draw an inference from the fact that
Blacks were absent more often, even though they were well within
the permissible levels of sick leave and annual leave.
The district court sought to bolster its acceptance of Dr.
Beckett's conclusions by assertions that the disciplinary system
contained checks and balances that would guard against arbitrary
disciplinary actions.— ^ This reasoning fails, however. As
explained in the Statement of Facts, there were no controls on
the crucial decision of supervisors whether to initiate discipline.
Even in the case of attendance related discipline, which was sus
ceptible to a controlled, objective procedure,there was no check
to ensure that all persons with similar attendance violations
were treated the same. With regard to subjective disciplinary
actions there were no controls whatsoever, and defendant's own
statistics demonstrated that Blacks were far more likely to receive
subjective disciplinary actions than discipline for objective
. 57/violations relating to work performance.
Of course, the same principle — that statistical disparities
cannot be explained away by unsupported assumptions — applies to
discipline as well as to other actions. A direct parallel can be
found in the case of Patton v. Mississippi, supra, in which the
Supreme Court rejected a proffered explanation that a difference
in conduct between Blacks and Whites could explain a difference
in selection for jury service in the absence of proof by the state
56/ As described in the Statement of Facts, a number of the
dTstrict court's findings are contradicted by the evidence.
57/ Compare, RE 484, 485, and 497, with 486, 489, and 490.
54
that such differences existed and that there was a causal connec
tion. Similarly, Lilly v. Harris-Teeter Supermarkets, 720 F.2d
at 336, n. 20, 337, n. 22. rejects an employer's attempt to explain
away differences in levels of treatment by assumptions that the
conduct of Blacks was the reason. Such a non—explanation must be
rebuffed, because otherwise all differences in discipline could
be justified by the circular reasoning that Blacks must have de-
58/
served more discipline because they received more discipline.—
III. THE DISTRICT COURT IMPROPERLY DISMISSED PLAINTIFFS'
DISPARATE IMPACT CLAIMS
Griggs v. Duke Power C o ., 401 U.S. 424 (1971), articulated
the disparate impact model of discrimination, under which proof
of aiscriminatory intent is not necessary. The Court held that
Title VII "proscribes not only overt discrimination but also prac-
5 9/tices that are fair in form but discriminatory in operation."— ■'
"'what is required is the removal of artificial, arbitrary, and
unnecessary barriers to employment" that "operate as 'built-in
headwinds' for minority groups." I_d. at 431, 432.
58/ The district court also held that in order to find "systematic"
discrimination in discipline it would have to be satisfied that
"every individual member of the class has been subjected to racial
prejudice" (RE 440) citing Teamsters v. United States, 431 U.S.
at 361-62, for this remarkable proposition. Teamsters, of course,
does not so hold; the cited discussion deals with the presumption
that operates at Stage II of a Title VII case when individual claims
are adjudicated after a finding of systematic discrimination.
59/ Under the Griggs model, the plaintiff establishes a prima
facie case by showing that the employment practices at issue
impact more harshly on the members of plaintiff's class. 401
U.S. at 431. The burden then shifts to the defendant to prove
that its practices have a "manifest relationship to the employ
ment at issue." Id., at 432.
55
Plaintiffs alleged and were prepared to prove at trial a sig
nificantly smaller percentage of Blacks than Whites in the avail
able internal workforce were promoted. (RE 25; PX 1). To deal
with the possibility that the court might rule that the disparate
impact model could not be applied to the final results of a multi
component decisionmaking process, plaintiffs alleged and were
prepared to demonstrate the adverse impact of several component
parts of the process, including all-White review boards, details,
awards, and discipline. (RE 25-26; PX 1). Except for written
tests, which had previously been dismissed from the case, none of
the component subparts challenged by plaintiffs were capable of
objective scoring.
A week prior to trial, the court dismissed all of plaintiffs
disparate impact claims, relying on the Fifth Circuit's decision
in Pouncy v. Prudential Ins. Co. of America, 668 F.2d 795 (5th
Cir. 1982). The court concluded that the disparate impact model
of proof applies only to objective employment criteria, and then
only to isolated, specific components of the selection process.
I d . QOS Pouncy represents the rejection of a consistent line
of pre-1981 Fifth Circuit cases decided prior to the division of
60/ The Fifth Circuit in Pouncy held that the disparate impact
model may not be used to challenge the cumulative results of an
employer's selection process, where that process includes two or
more components or stages. Rather, the plaintiff must demonstrate
the adverse impact of a specific component of the selection pro
cess, such as an intelligence test or a minimum height requirement.
668 F .2d at 800. Pouncy also held that the disparate impact
model is inapplicable to any subjective component of a selection
process, ^d. at 801.
56
. Of course, those
the Fifth and Eleventh Circuits in 1981.
earlier cases are binding in this Circuit and cannot be overruled
6 2/
except by the Court acting en banc. In fact, this Court
recently recognized that these former Fifth Circuit precedents,
rather than Pouncy, apply in the Eleventh Circuit. Eastland v . _
Tennessee Valiev Authority, 704 F.2d 613, 619-20 (11th Cir.
63/
1983), cert, denied, 52 U.S.L.W. 3631 ( 1984).—
Johnson v. Uncle Ben's, Inc., 628 F.2d 419 (5th Cir. 1980)
applied the disparate impact theory to a promotion system that is
strikingly similar to that used by defendant. There, plaintiffs
showed that the proportion of Blacks in each job grade did not
reflect the proportion of Blacks available for promotion from
lower job levels. 628 F.2d at 423. The promotion system involved
several different components, most of which were subjective. 628
F .2d at 426-27. Nonetheless, the court held that plaintiffs
61/
61/ S e e , e.g6±/ ___ _____
(5tn Cir. 1980), remanded
902 (1981), reaffirmed in
cert, denied, 51 U.S.L.W.
Johnson v. Uncle Bens, Inc., 628 F .2d 419, 426-27
for further consideration, 451 U.S.
relevant part~ ̂ 657 F . 2d 750 ( 1981 ),
3339 (Nov. 1, 1982); Crawford v.
1300, 1316-18 (5th Cir. 1980);Western Elec. Co., Inc., 614 F .2d 1300,
Rowe v. General Motors Corporation, 457 F .2d 348, 354-59 (5th
Cir. 1972)
62/ Bonner
1981 ) .
v. City of Pritchard, 661 F.2d 1206, 1207 (11th Cir,
63/ Even in the new Fifth Circuit, panels have reached con
flicting decisions with regard to the issues raised in Pouncy.
Compare Harrell v. Northern Elec. C o . , 672 F.2d 444 (5th Cir.
1982), reaffirmed in relevant part,~679 F.2d 31 (5th Cir.
1982) , cert, denied, 51 U.S.L.W. 3419 (Nov. 29, 1982), with _
Carroll v. Sears, Roebuck & C o . , 708 F.2d 183, 188-89 (5th Cir.
1983) . s p a also Pace v. U . S . Industries, Inc., 726 F . 2d 1038 ,
1045-46 (5th Cir. 1984). The Sixth Circuit has not followed
Pouncy, see Rowe v. Cleveland Pneumatic Co., Numerical Control,
690 F .2d 88, 93-95 (6th Cir. 1982), while the Fourth and Tenth
Circuits have reached results that are consistent with Pouncy,
see Pope v . City of Hickory, N.C. , 679 F .2d 20 , 22 ( 4th Cir.
1982); Mortensen v. Callaway, 672 F.2d 822, 824 (10th Cir. 1982)
57
could establish a prima facie case under the disparate impact
model by showing that the results of the system as a whole had an
64/
adverse impact on blacks. I_d. at 426-27.
The decisions in Johnson v. Uncle Ben's and similar cases
are consistent with the relevant Supreme Court authority. The
overriding concern of the Court in Griggs was the use of barriers
to employment that were not related to ability to do the job in
question. The Court did not differentiate between objective and
subjective barriers, but rather concluded that "Congress has made
[job] qualifications the controlling factor, so that race, reli
gion, nationality and sex become irrelevant." 401 U.S. at 436.
"If an employment practice which operates to exclude Negroes
cannot be shown to be related to job performance, the practice is
prohibited." I d . at 431. In fact, the Griggs opinion is replete
with references to "practices" and "procedures, terms that
clearly encompass more than isolated, objective components of the
overall process. Nowhere in the Griggs opinion does the Court
6 6/
even suggest limits such as those adopted in Pouncy.
64/ Accord Rule v. I .A .B .S .0.I ., Local Union No. 396, 568
F .2d 558, 566 (8th Cir. 1977).
65/ E.g., 401 U.S. at 430 ("practices, procedures, or tests")?
id. at 431 ("practices")? id. at 432 ("employment procedures
or testing mechanisms")? _id. ("any given requirement").
66/ The Fifth Circuit apparently construed a reference in Griggs
to facially neutral practices as limiting the disparate impact
model to devices or criteria that are capable of objective
scoring. The phrase was used in connection with the Court's
rejection of the argument that Title VII proscribes only inten
tional discrimination. The passage reads as follows.
[footnote continues on next page]
58
The broad reach of the disparate model of proof is confirmed
by the Court's decision in Connecticut v. T e a l , 457 U.S. 440 (1982)
The Court repeatedly emphasized that any "barrier to employment
opportunities," 457 U.S. at 447, 448, 449, 450, 451, 453, can be
challenged under the disparate impact model. Moreover, the dissent
ing Justices in Teal agreed that the process is subject to the
disparate impact model. "[0]ur disparate impact cases consistently
have considered whether the results of the employeer's total
selection process have an adverse impact upon the proctected
group." 457 U.S. at 458 (Powell, Burger, Rehnquist, Rehnquist,
O'Connor, JJ., dissenting.)— ^
66/ continued
The objective of Congress in the enactment of Title VII ...
was to achieve equal employment opportunities and to remove
barriers that have operated in the past to favor an identifi
able group of white employees over other employees. Under
the Act, practices, procedures, or tests neutral on their
face, and even neutral in terms of intent, cannot be main
tained if they operate to "freeze" the status quo of prior
discriminatory employment practices.
401 U.S. at 430. When viewed in context it is clear that the
Court used the phrase "neutral on their face" to refer to poli
cies or practices that are not discriminatory on their face.
For example, a policy that Blacks need not apply is facially
discriminatory, while a policy of using a review panel to make
selections is facially neutral.
Finally, the reference to practices and procedures "neutral
on their face" occurs only once in the Griggs opinion. If the
Court had intended this reference to limit the disparate impact
model to objectively-scored criteria, it surely would have
included "facially neutral" as a modifier in the numerous
instances in which it stated that the disparate impact rule
applies to employment practices, procedures and requirements,
see note 63, supra.
67/ The dissenters noted that employers could simply integrate
the test results into one overall hiring decision that considered
a multitude of criteria that could not be challenged "unless the
actual hiring decisions had a disparate impact on the minority
group." 457 U.S. at 463, n.8.
59
Subjective practices that are not job-related, such as inter
views and supervisory recommendations, are as capable as written
tests of operating as "barriers" or "built-in headwinds" to
minority advancement. A supervisor may give a good faith evalua
tion of an employee's performance of a particular task. However,
it is possible that the ability to perform the task evaluated is
not related to performance of the job for which the candidate
is applying. Similarly, an interviewer may attempt to select
the best applicant, but be incapable of making a valid decision.
Such practices serve as "artificial, arbitrary, and unnecessary
barriers to employment," condemned in Griggs.
Moreover, exclusion of subjective practices from the reach
of the disparate impact model of proof is likely to encourage
employers to use subjective, rather than objective, selection
criteria. Rather than validate education and other objective
criteria, employers can simply take such criteria into account
in subjective interviews or review panel decisions. It cannot
have been the intent of Congress to provide employers with an
incentive to use such devices rather than validated objective
criteria.
Limiting the disparate impact rule to isolated components
of a selection process also is inconsistent with Supreme Court
authority. The Court in Griggs and Teal repeatedly described
the disparate impact model as applying to "practices and proce
dures," which clearly encompass the entire selection process or
60
system.— ' Moreover, the legislative history of the 1972
amendments to Title VII leaves no doubt as to Congress intent
on this issue. In 1972 Title VII's coverage was extended to
federal employers, including defendant in this case, and to
state and local governments. Congress explicitly approved of
the Griggs d e c i s i o n ^ / and the Senate Report noted:
Employment discrimination ... today is a ... complex
and pervasive phenomenon. Experts familiar with the
subject now generally describe the problem in terms
of 'systems' and 'effects' rather than simply inten
tional wrongs."
S. Rep. No. 92-415, 92nd Cong., 1st Sess. 5 (1971) (emphasis
added.)
Finally, limiting the disparate impact model to situations
in which a single component of the process results in an adverse
impact completely exempts the situation in which an adverse
„ 70,
impact is caused by the interaction of two or more components.
The adverse impact caused by the interaction of two or more
68/ See also Pullman-Standard Co. v. Sw in t, 456 U.S. 273, 276-
77 (1982) which holds that a seniority system would be subject
to the disparate impact test but for § 703(h) of Title VII.
69/ S. Rep. No. 92-415, 92nd Cong., 1st Sess. 5 (1971); H.R.
Rep. No. 92-238, 92nd Cong., 1st Sess. 8 (1971).
70/ This possibility was recognized by the Court of Appeals for
"the Eighth Circuit in Gilbert v. City of Little Rock, A r k . ,
722 F . 2d 1390, 1397-98 (1983). The court stated:
The district court neglected to adequately consider
the interrelationship of the component factors and
... whether the oral interview and performance ap
praisal factors ... had a disparate impact on black
officers' scores. . . .
61
non-job-related subparts of the selection process is no less a
"barrier" to employment opportunities than that of a single
71 /component of the proces.— '
The rulings in Johnson v. Uncle Ben's and similar cases
are also supported by the the Uniform Guidelines on Employee
Selection Procedures, 29 C.F.R. § 1607. The four federal
agencies charged with enforcing Title VII have interpreted the
disparate impact model to apply to the results of a multi—com
ponent selection process and to all selection procedures,
72/
whether objective or subjective.— The Uniform Guidelines
are explicitly binding on all federal agencies. 29 C.F.R. §
1607.02. Yet, Pouncy would preclude enforcement of the agencies'
obligation to comply with these Guidelines. For this reason the
7 1 / The Court in Pouncy justified its conclusion that plain-
tTffs must isolate the adverse impact of specific subparts of
the selection process on the ground that it would be unfair to
require the employer to defend the validity of its entire system
when some elements might have no adverse impact. This reasoning
also is faulty. Under the rationale of Griggs, the employer
could eliminate a particular component from controversy by
demonstrating that that component did not contribute to the
overall disparity. If such a component exists, the employer
should be able to meet its burden easily, since under Title VII
regulations the employer is required to keep records on the
impact of each subpart of the selection process where the
process as a whole results in an adverse impact. 29 C.F.R.
§ 1607.4.
72/ Any measure, combination of measures, or procedure used
as a basis for any employment decision. Selection pro
cedures include the full range of assessment techniques
from traditional paper and pencil tests, performance
tests, training programs, or probationary periods and
physical, educational, and work experience requirements
through informal or casual interviews and unscored appli
cation forms.
29 C.F.R. § 1607.76 (emphasis added).
62
court in Harrison v. Lewis, 559 F. Supp. 943 (D.D.C. 1983),
Memorandum Opinion, June, 7,1982, aff'd, ___ F .2d --- (D.C.
1984), explicitly rejected the Pouncy decision.
IV.
THE DISTRICT COURT'S FINDINGS WITH REGARD TO
CLASS MEMBERS SHOULD BE REVERSED AND REMANDED
At trial 24 Black class members give examples of how
the discrimination demonstrated by the statistical evidence
affected them. The district court rejected the testimony of
all of the class members in toto, holding that no class members
were ever discriminated against in any one of their nearly 100
claims encompassing the period 1969-1981. In many instances,
.... 73/
the trial court found that their testimony was not credible.
On the other hand, the court uniformly credited the testimony of
74/
defendant's witnesses.—
73/ See, e_̂ g_. , RE 278; 291; 307; 314 , 326; 330 ; 349; 391; 377 .
74/ The court also concluded that plaintiffs' expert witness,
Dr. James Outtz, a reputable industrial psychologist, lacked
credibility. (RE 411, 413-14.) Thus, the court adopted defen
dant's efforts to malign Dr. Outtz's professional reputation
by citing only one case in which he participated and in which
his testimony was rejected. (RE 398) As Dr. Outtz testified,
that case is on appeal. On the other hand, the trial court
disregarded the acceptance of Dr. Outtz's expert testimony has
been accepted and credited in numerous other cases. S e e , 6.g ..
Seqar v. Civiletti, 508 F. Supp. 690 (D.D.C. 1981); as his
resume reflects he has testified in numerous Title VII actions
involving the federal government as well as private employees
including Harrison v. Lewis (D.D.C.); Lewis v ♦ NLRB (S.D. Tex.);
Ennis v. The Home Loan Mortgage Corporation, Gilbert v. City of
Little Rock, and Pegues v. Mississippi State Employment Service
(TR 1549-50; PX 95). Further, Dr. Outtz is an industrial
psychologist in private practice and an instructor at Howard
University (TR 1547; PX 95).
63
Plaintiffs recognize that with regard to findings of fact
based on credibility, their burden is a heavy one to obtain
reversal. Our basic position is that a determination of the
individual claims of class members should be remanded to the
district court for reconsideration in light of the classwide
discrimination that has been shown in Argument II. See
Donaldson v. Pillsbury C o ., 554 F.2d 825 (8th Cir. 1977).
We also, however, wish to point out the fact that in a
number of instances the district court's holdings were flatly
contradicted by the record. For example, class member Leroy
Robinson testified with regard to a number of disciplinary
actions taken against him which he contended were either dis
criminatory or taken for the purpose of retaliation. In one
instance he testified that he was one of a group of persons who
were seen at a soft drink machine; he was written up while his
white companions were not. (TR 1409; 1412.) Testimony to the
contrary was not introduced, and yet the district court simply
rejected this uncontradicted testimony out of hand.
Completely undisputed were the circumstances surround
ing the defendant's attempt to fire Mr. Robinson in 1981,
after he appeared on a local radio talk show and criticized the
post office for discriminatory practices. He was immediately
served with a notice of proposed removal; one of the three
charges was specifically his appearance on the radio show. (TR
1431; PX 87a.) The Merit Systems Protection Board threw out the
charge and reversed his removal, holding that the attempt to
discipline him for his appearance on the show violated his
64
rights under the First Amendment. (PX 87d.) As was argued to
the district court, the incident was also a clear violation of
Title VII's prohibition against retaliation for opposing dis
crimination. See, e.g., Payne v. McLemore's Wholesale & Retail
Stores, 654 F.2d 1130 (5th Cir. 1981). Nevertheless the district
court found that the charges against Robinson were "supported by
some evidence," and that the defendant's "inability to prove
them does not warrant the conclusion that they were the product
of racial discrimination." (RE 378) Another example is the case
of Mrs. Juanita Bacon who testified concerning an incident in
1973 when she and her husband took the examination for placement
on the supervisory register. (TR 756). Following the examination
neither she nor her husband were given their scores. Believing
they had done well, they inquired and were finally told by the
then head of personnel that it was believed that they had cheated.
(TR 758-61). They denied the accusation and repeatedly asked for
written confirmation of the reason why they did not receive the
scores. They never received such confirmation, and they never
did receive the test scores. (TR 761-62; 2049; RE 291). Defen
dants introduced no testimony whatsoever in contradiction of Mrs.
Bacon's testimony. Despite the complete lack of any supporting
evidence, the court held that it was "reasonable for the post
office to believe" that the Bacons had cheated. (RE 292.)
Finally, it is clear that the determination by the court below
that there was no classwide discrimination in no way constitutes
a determination that there was no aiscrimination against indi-
65
vidual class members.— Thus, the appropriate action was to
inform all class members that they were entitled to pursue their
individual claims or, alternatively, await the outcome of this
appeal. If plaintiffs are successful on appeal, class members
can present individual claims in the Stage II proceedings of this
action. If, on the other hand, this Court affirms the finding
of no discrimination, it should require that a notice be sent to
all class members informing them of their right to pursue their
76/
individual claims.
75/
V.
THE COURT'S AWARD OF COSTS TO THE
DEFENDANT WAS I M P R O P E R ^
The District Court's Judgment of November 23, 1983, ordered
that the "defendant William F. Bolger, Postmaster General,
75/ Connecticut v. Te al, 457 U.S. 440 (1980).
76/ It is now firmly established that the filing of a class
action in a Title VII case tolls the time for individual class
members to pursue individual claims or another class action.
Crown, Cork & Seal Co. v. Parker, ___ U.S. ___ , 76 L.Ed.2d 628
(1983). Thus, class members are entitled to rely on the
pendency of the class action as tolling the time in which they
would have to pursue their individual claims. Plaintiffs note
that a determination as to the res judicata effect of a finding
of no classwide discrimination on the right to pursue an
individual claim is pending before the United States Supreme
Court in Cooper v. Federal Reserve Bank of Richmond, 83-185.
77/ This issue is pending before the Court of Appeals for
the Fifth Circuit. Lewis v. National Labor Relations Board,
No. 83-2055. Further, a judgment from the District Court for the
District of Columbia which awarded costs to the federal government
as the prevailing party was summarily affirmed in a non-preceden-
tial order. Clark v. Lewis, Civil Action No. 80-1636 (D.D.C.
August 18, 1982), aff'd, sub nom., Clark v. Do l e , 713 F.2d 864
(D.C. Cir. 1983), cert. den. , ___ U.S. 79 L. Ed. 2d 164
( 1 984) .
66
recover of the plaintiffs ... his costs of the action."
The Federal Rules of Civil Procedure allow for the taxation of
costs to the prevailing party unless the court directs otherwise.
Rule 54, F.R. Civ. Proc. It is well established that the
decision whether to award costs rests within the sound discre
tion of the court. Delta Airlines, Inc, v. August, 450 U.S. 346
(1981). Yet, the court's discretion to impose extraordinary
costs should be used sparingly so as not to discourage litiga-
tion. Farmer v. Arabian American Oil C o . , 379 U.S. 227, 275
(1964). In accordance with this principle there are signifi
cant legal, statutory, and administrative impediments to the
government's receiving costs.
Plaintiffs-appellants submit that the facts of this case
necessitate a reversal of the lower court's award of costs to
the defendant for two primary reasons. First, it was inequitable
to tax costs since this case was brought in good faith in order
to challenge employment practices which plaintiffs reasonably
believed were violative of Title VII of the Civil Rights Act of
1972, as amended. Second, the defendant, as a federal government
agency, is precluded from seeking costs because of an official
directive issued by the United States Department of Justice in
1978 and in effect at the time the costs at issue here were
incurred.
78/
78/ Plaintiffs have opposed defendants' bill of costs in the
court below. This matter was pending at the time this brief
was written.
67
A. ?j,?Si!i^ g t fnAjeigSL^§n
A determination of whether to award costs should be decided
according to the justice of the cause and the facts and circum
stances of the case. See In re Northern Indiana Oil C o ., 192
F .2d 139 (7th Cir. 1951). Accordingly, in Lichter Foundation,
Inc, v. W e l c h , 269 F.2d 142 (6th Cir. 1959), the Court held that
Rule 54(d) was intended to allow for the taxation of costs in
favor of the prevailing party except when under all the circum
stances in the case it would be inequitable to place the burden
of costs upon the unsuccessful party.
A major factor to be considered was stated by the Seventh
Circuit in Chicago Sugar Co. v. American Sugar Refinery, 173
F . 2d 1, 11 (7th Cir. 1949) :
... where it is clear that the action was
brought in good faith, involving issues as
to which the law is in doubt the court may
in its discretion require each party to bear
its own costs although the decision is ad
verse to the plaintiffs.
See also Andersen v. Clear Ridge Aviation, 9 F.R.D. 50,
79/
54-55 (D.C. Neb. 1949).—
79/ "The court considers that the plaintiffs instituted and
prosecuted this action, generally to enjoin the continued
operation of the defendant's airport in their immediate neigh
borhood, in good faith and in the assertion of contentions,
which, or at least some of which, they believed to be well
founded, and not for the purpose of harassment or oppression.
While the court found against them upon the merits of their
demand, its ruling involved no finding of unworthiness or malice
in the tender of their claim."
68
Here, defendant did not allege that the plaintiffs insti
tuted the litigation for the purpose of harassment and, indeed,
it is difficult to imagine how this suit could have the effect
of oppression when the federal government is the defendant.
Nor can it reasonably be said that this action was vexatiously
initiated or that plaintiffs' claims were frivolous. Plaintiffs
patiently pursued their administrative remedies pursuant to
Title VII, and when unsuccessful sought relief in the federal
courts.
Further, in deciding whether an award of costs would be
equitable, the Fifth Circuit has held that economic considera
tions should play a role in the assessing of costs in Title VII
actions. In the case of Miller v. International Paper Company,
408 F .2d 283, 293-94 (5th Cir. 1969), although involving
slightly different circumstances than those presented here,
the issue of assessing costs against unaffluent plaintiffs was
considered. The court held that the imposition of a penalty on
plaintiffs would not only cause undue hardship, but it would
also tend to undermine the Congressional policy of financially
assisting complainants in Title VII suits because it "would lead
to a harsh result when plaintiff is an individual litigant and
defendant is a large corporation." Further, "to place such an
undue burden on plaintiffs "could only chill individual litigants
of modest means seeking to vindicate their individual and class
69
rights under the civil rights laws."
The courts have repeatedly recognized that Congress in
tended to encourage private litigation under Title VII. Thus,
attorneys' fees are provided because plaintiffs act in the
capacity of "private attorneys general," enforcing a Congres
sional policy of great importance. Newman v. Piggie Park
Enterprises, 390 U.S. 400 , 402 ( 1 968).— / Similarly, the
standard for granting attorneys' fees under Title VII to pre
vailing defendants is far stricter than when the plaintiff
prevails, being available only where the plaintiffs' action was
frivolous, unreasonable or without foundation. Christiansburg
Garment Co. v. Equal Employment Opportunity Commission, 434 U.S.
412 ( 1978) .
Potential plaintiffs who are aware that costs may be
taxed against them will hesitate to bring actions regardless of
the merits of their claims. Thus, costs would have a "chilling
effect" and frustrate a sound Congressional policy of uprooting
discrimination. This is particularly true where the defendant is
the federal government since neither the EEOC nor the Attorney
General can enforce Title VII against federal agencies, and the
individual charging party is the only person who can bring such
79/
79/ Schaulius v. CTB/McGraw-Hill, Inc., 496 F. Supp. 666 at 680
(N.D. Ca. 1980). See also Martin v. Frontier Federal Savings
and Loan Association, 510 F. Supp. 1062, 1069 (W.D. Okla. 1981);
Eldreage v. Carpenters, 46 California Counties, 83 F.R.D. 136
(N.D. Ca. 1979); County of Suffolk v. Secretary of the Interior,
76 F.R.D 469 (E.D. N.Y. 1977); Maldonado v. Parasole, 66 F.R.D.
388, 390 (S.D. N.Y. 1975) .
80/ See also Bradley v. School Board of Richmond, 416 U.S. 696
7T974); Miller v. Amusement Enterprises, 426 F.2d 534 (5th Cir.
1970); Lea v. Cone Mills Co rp., 438 F.2d 86 (6th Cir. 1971).
70
who can bring such actions. See Parker v. Califano, 561 F.2d
320, 331 (D.C. Cir. 1977). More importantly, the federal
government, which is vested with the authority to initiate Title
VII suits against private employers and state and local govern
ments for discrimination, should not, when it is the defendant,
seek to exact penalties from plaintiffs who have sought to
eliminate discrimination within the government itself.
In the case at bar the defendants have requested a sum
of $62,533.16 as costs in this case. We urge that an award of
costs against the plaintiff here would act as a clear signal
to other plaintiffs in Title VII actions that they may be pen
alized for pursuing their legal rights. Such a result would
contravene the very intent of Title VII, whose purpose is to
eliminate discrimination in employment by encourag
ing plaintiffs to initiate private actions.
B. Defendants Are Precluded From Seeking Costs
Because Of A Justice Department Directive
In 1978, the Department of Justice issued a directive set
ting out guidelines governing the federal government in seeking
81/
costs in Title VII actions.— in these guidelines the
Department of Justice mandated that the standards established
82/
in Christianburg be followed when the government seeks costs.
81/ This directive was published in CCH Fair Employment Practices
11 5083 ( 1 978). See Schlei & Grossman, Employment Discrimination
Law (2nd. Ed. 1983), p. 1215, n. 186.
82/ The guidelines were rescinded in April, 1983, after the
trial in this case. The great majority of the costs here were
incurred between 1980 and 1982 while the 1978 Memorandum was in
effect.
71
Accordingly, costs should not be sought unless (1) the plaintiff
instituted the action in bad faith; (2) the plaintiff pursued
the claim in a harassing or vexatious manner; (3) the plaintiff
litigated a claim which was patently groundless or frivolous.
In that memorandum the government cited with approval the
decision of the District Court in Jaspers v. Alexander, 15
F.E.P. Cases 1238 (D. D.C. 1977), which applied similar standards
in deciding not to award costs.
Indeed, the history of the guidelines demonstrates that
they were developed in settlement of an appeal involving this
very issue. Set out in the Appendix to this Brief is a copy of
the order of the court entered in Thomas v. Department of the
N a v y , No. 77-1965, and the motion of the parties filed therein.
As those documents demonstrate, the memorandum of the Depart
ment of Justice was issued by agreement in Thomas in settling
the question of the appropriate standards to govern the award of
costs to the United States in a Title VII action.
The government is bound by its own regulations and guide
lines and may not disregard them. See United States ex re l.
Accardi v. Shaughnessy, 347 U.S. 260 (1954); Service v. Dulles,
354 U.S. 363 (1957); Vitarelli v. Seaton, 359 U.S. 535 (1959);
United States v. Ni xo n, 418 U.S. 683 (1974). It would be
particularly inappropriate to permit the government to do so
here since the guidelines were issued as part of an agreement
between the government and a civil rights organization seeking
to litigate not only the Thomas case, but to address generally
the appropriate standard to govern the award of costs in Title
72
VII cases. Since it is clear that the present case does not fit
within the guidelines established by the government, the Bill of
Costs should not have been filed to begin with.
The case law governing costs to the defendant in a Title
VII action where the employer is the federal government is
scant. However, where the courts have considered this issue
they have, in most cases, concluded that costs should not be
granted.— 7 As the 1978 Memorandum states, costs "are
intended to deter baseless or unreasonable clams and not to
create an extreme hardship on the plaintiff." Consequently, the
granting of costs in this case, which was instituted and liti
gated in good faith, serves no constructive purpose and will
place a substantial financial burden on plaintiffs.
Plaintiffs' trial attorneys, who are experienced civil
rights litigators, were fully aware of the Department of Justice
policy. The named plaintiffs in this lawsuit pursued the action
based on the advice of counsel. Neither the plaintiffs nor their
counsel contemplated that they would be faced with the assessment
of costs if they lost the case. Indeed, quite to the contrary,
plaintiffs had every reason and right to expect they would not
he so charged in light of the government's April 14, 1978,
directive.
8 3 / in Dual v. Cleland, 79 F.R.D., 696 at 697 (D. D.C. 1978),
the court applied the Christiansburg standards in concluding
that costs should not be granted under Rule 54(d) unless the
action was frivolous, unreasonable or without foundation.
Similarly, in Torres v. Claytor, 25 F.E.P. Cases 998 (S.D. Cal.
1978), the court denied a request for costs by the Department of
the Navy. See also Mizrany v. Texas Rehabilitation Commission,
522 F. Supp. 611 (S.D. Tex. 1981). But see cases cited in n.77,
supra.
- 73 -
Clearly, manifest injustice "would stem from the imposition
of new and unanticipated obligations on a party without notice
or an opportunity to be heard." Ia Power & Light Co. v.
Burlington Northern Inc., 647 F.2d 796, 806 (8th Cir. 1981).
See also Lawrence v. Staats, 665 F .2d 1256, 1258-60 (D.C. Cir.
1981). Having relied on the existence of the policy and having
litigated the case in good faith it would be unjust to
impose on the plaintiffs costs for litigation activities.
Therefore, this Court should adopt the position that costs
should not be awarded to the government unless the Christiansburg
standards have been met.
CONCLUSION
For the foregoing reasons, the decision of the court
below should be reversed and the case should be remanded with
instructions to enter judgment for plaintiffs.
Respectfully submitted,
JACK GREENBERG
GAIL J. WRIGHT
CHARLES STEPHEN RALSTON
PENDA HAIR
/
99 Hudson Street
16th Floor
New York, New York 10013
W. BENJAMIN KYLE
1248 W. Edgewood Avenue
Jacksonvile, Florida 32208
Attorneys for Plaintiffs-Appellants
74
CERTIFICATE OF SERVICE
I hereby certify that on May 1984, copies of the
revised version of Plaintiffs-Appellants' Brief were
served on counsel for defendant listed below, by deposting
them in the United States mail, first class postage
prepaid:
Wyneva Johnson, Esq.
Office of Labor Law
United States Postal Service
475 L'Enfant Plaza West, S.W.
Washington, D.C. 20260-1131
Ernst D. Mueller, Esq.
Assistant U.S. Attorney
P.O. Box 600
Jacksonville, Florida 32201
CHARLES STEPHEN RALSTON
IN THE UNITED STATES COURT OF APPEALS
FOR THE DISTRICT OF COLUMBIA CIRCUIT
CHARLES A. THOMAS,
Plaintiff-Appellant,
v.
SECRETARY OF THE NAVY,
Defendant-Appellee.
)
)
)
)
) No. 77-1965
)
)
)
)
JOINT MOTION TO VACATE ORDER AND DISMISS APPEAL
The plaintiff-appellant, Charles A. Thomas, and the
defendant-appellee, Secretary of the Navy, hereby Jointly
move that the district court’s August 31, 1977 order
taxing costs against the plaintiff be vacated as moot
and this appeal be dismissed.
1. Plaintiff, a civilian employee of the Department of
the Navy, brought this suit under Title VII of the Civil
Rights Act of 1964, as amended, 42 U.S.C. 2000e-l6,
charging that he was the victim of race discrimination.
Plaintiff was unsuccessful, and final Judgment was entered
in favor of the government. The government then moved that
costs in the sum of $409-50 be taxed against the plaintiff.
Plaintiff opposed any taxation of costs, arguing that the
government was statutorily barred from recovering costs in a
Title VII case. While maintaining that there was no stat
utory bar, and that under Rule 54(d), Fed. R. Civ. Pro., the
United States, like any other litigant, is entitled to
costs, the government stated that since plaintiff appeared
to be heavily in debt due to the litigation, there would be
no objection "to a discretionary reduction of costs by the
Court."
2
2. The district court, in an order entered August 31,
1977, agreed that, as a matter of law, the government was
entitled to costs. However, the court, noting the govern
ment's suggestion, and observing that, even though he lost,
the plaintiff had performed a public service in bringing the
suit, reduced the cost award to $248.65. Plaintiff appealed
solely on the cost question.
3. During the pendency of the plaintiff's appeal, the
Department of Justice issued guidelines specifying the
circumstances under which the government would seek costs
1/
when it prevailed as a defendant in a Title VII case. Under
the criteria set forth in the guidelines, this is not the
type of case in which the government would seek costs.
4. The issuance of the guidelines, and the govern
ment's current position, based on the guidelines, that
plaintiff should not be taied for cost has eliminated any
controversy between the parties and has rendered the appeal
moot. Accordingly, the plaintiff and the government Jointly
agree to waive all rights which they may have against each
other for costs or attorney's fee incurred in this action at
the trial or appellate levels. They consequently Jointly
1/ A copy of those guidelines, issued on April 14, 1975,
are attached to this motion.
- 3 -
(
request that this Court vacate as moot the August 31,1977,
order taxing costs against the plaintiff 'and dismiss the
appeal.
Respectfully submitted,
PAUL BLANKENSTEIN, 202-739-3^27
Attorneys for the defendant-
appellee
Appellate Section, Civil Division,
Department of Justice,
Washington, D.C. 20530
ARTHUR F. GREENBAUM, 202-331-^706
Attorney for the plaintiff-
appellant
Hogan & Hartson
815 Connecticut Avenue
Washington, D.C. 20006
l-7t N ew D ev elo p m en ts
participation
as part of a
riveiy with minimal disruption of er.force-
ent effort;
) that they will issue a detailed plan
witflin 90 days of this agreement out-
liningSsteps and procedures on how en
forcement including responsibility for
litigation\will be passed irom DOL to
EEOC;
(4) that aKangrments will be made to
enable £ E O C \o participate in the en
forcement proceW prior to July 1, 1979
and for DOL toNwntinue
subsequent to July 1979
staged, orderly, transits
(5) that they will w o^t together in
providing training of enforcement staff
of both agencies working in >he EPA
and ADEA programs through tSe tran
sition, and that a conference witnsjer-
sonnel from both agencies responsible^
effectuating the transition will be heL'
Washington, D. C. in the latter a^rt of
1973 to assure coordination at both head
quarters and in the field;
(6) a. that they will -^cognize the
President's stated desinp^that organiza
tional changes be made with careful con
cern for the beings whose jobs
and careers arjxiffected. Every effort
will be made^Xo minimize the impact of
such chan|p3upon employees affected by
the reorganization. Personnel changes
will boriaccomplished in accordance with
direptfves issued by the Office of Man-
ient and Budget (OMB) and the
Civil Service Commission (CSC),
parties recognize the need to tulfi)T any
obligations to meet with unions Ufat may
be required by existing collpmive bar
gaining agreements, includin^^ecognition
of the nght of unions to be^resent in the
event that there are meetings with em
ployees;
b. that EEOC iy^nterested in obtain
ing from DOL pv^onnel who are trained
and expenenced/m enforcement of EPA
and ADEA afld that DOL and EEOC
will begin jfmely discussions with each
other ansr OMB and CSC concerning
emplovpfent options of employees from
DOUXprior to July 1, 1979 and that af-
fec^Q employees will be advised of their
ftioos as soon as possible;
(7) that after the effective date of trans
fer, DOL will inform local EEOC offices
of any apparent sex or age discrimination
under EEOC jurisdiction which is noted
the course of DOL investigative ac-
^vity;
a. that during and following^ the
transieicn period, representatives of EEOC
will ha^Vaccess to DOL ADEA/EPA
enforcemeiHvaction files and related ma
terials;
b. that EEOC not reveal the iden
tity of a complainant i^^ny DOL enforce
ment action file to who^confidentiahtv
has been pledged.
Back references.—1211, 212T
[f 5083] GUIDES FOR SEEKING COSTS IN FEDERAL EMPLOYEE SUITS
Guidelines for government attorneys in filing motions for the assessment of court
costs against federal employees bringing bias actions against federal agencies were supplied
in a memorandum issued by the Assistant Attorney General on April 14, 19/3. Text ot
the Memorandum appears below.
In C'nristiansbvrg Garment Co. v. E<}vai
Employment 0 p portimit-; Commission, [IS
EPD 1 8041] 46 U. S.' L. W. 4105, 4107
(U. S. Jan. 23, 1978), the United States
Supreme Court, in declining to allow attor
neys’ fees to be routinely awarded to pre
vailing defendants in Title VII actions,
characterized the Title VII plaintiff-em
ployee as "the chosen instrument of Con
gress to vindicate 'a policy that Congress
considered of the highest priority’" (quot
ing Netvmcn v. Piggie Pork Enterprises. [2
EPD J 9534 ] 390 U. S. 400, 402 (1963)).
Because the Title VII plaintiff-employee
appears before the court "cloaked in a
mantle of public interest," United States
Steel Corp v. United States. [9 EPD * 10,225]
519 F. 2d 359 , 364 ( 3d Cir. 1975), the rule
that prevailing parties generally are entitled
to the costs of the lawsuit should not apply
automatically to actions in which the Fed
eral Government defeats an employee’s Title
Employment Practices
VII charge of discrimination. As one court
has stated: "The awarding of costs to the
Government when it successfuiy defends
itself against a Title VII claim should not
become a mechanical process that will dis
courage potential plaintiffs from pursuing
debatable claims." Jaspers v. Be’-nstr.n, Civil
Action No. 76-1411 (D. D. C, Sept. 19,
1977); cf. Alonso v Union Oil of Calif., 71
F. R. D. 523, 524 (S. D. N. Y, 1976). The
Government must be particularly wary of
creating ar. fn trrrorem effect on potential
plaintiffs with meritorious Title VII claims,
in view of the fact that many of these
plaintiffs can scarcely afford to bear their
own costs of litigation. See Miller v. Inter
national Paper Co., [1 EPD £ 9965] 4C8 F.
2d 283. 293 (5th Cir. 1969); cf. Boas Box
Co v. Proper Polding Box Corp., 55 F. R. D.
79, 81 (E. D. N. Y. 1971) (considering
economic resources of parties).
H 5083
APPENDIX II
3 2 6 4 N ew D ev e lo p m en ts tO i-7*
Accordingly, the Government should not
move for cost* u a prevailing defendant
in a di*crimination *uit brought by a fed
eral employee under Title VII unless any
one of the following three circumstances
is found to exist:
(1) the plaintiff brought the action in
bad faith;
(2) the plaintiff pursued the claim in a
harassing or vexatious manner;
(3) the plaintiff litigated a claim that
was patently • groundless or frivolous.
Application of the first factor (bad faith)
is illustrated by the case of Carrion v.
Yeshiva Unwrrsity, S3S F. 2d 722 (2d Cir.
1976). In allowing the imposition of costs
and attorneys' fees on the plaintiff, the
court found that the plaintiff’s Title VII
claim was substantially the same as charges
she had previously brought and litigated,
that she had perjured herself in pursuing
her claim, and that she and another em
ployee had deliberately attempted to ruin
the reputation of their supervisor. .
Application of the second factor (harass
ing or vexatious conduct) is illustrated by
the- case of Quaker Chair Corp. v. Litton
Business Systems, Inc., 71 F. R. D. 527, 537
(S. D. N. Y. 1976). There the court
awarded costs to the (non-Title VII)
plaintiff in connection with its opposition
to defendant's motion to vacate a deposi
tion. Even though the deposition was
vacated, the court found it appropriate to
assess costs against the defendant, where
the defendant, after refusing to stipulate to -
facts that plaintiff sought to discover and
after forcing plaintiff to incur costs in seek
ing to establish those facts, belatedly con
ceded those facts. See also Reeves Brother: ,'
Inc. v. U. S. Laminating Corf., 417 F. 2d
869, 873 (2d Cir. 1969) (awarded costs to
prevailing party where opposing party “con
sumed needless time and effort in trying
frivolous 'shotgun' claims”).
In applying this second factor, the Gov
ernment should avoid penalizing the client
unfairly for the unreasonable conduct of
his or her attorney. If the lawyer's conduct
is unacceptable, the Government should in
appropriate cases consider alternatives that
do not impute that conduct to the client—
for example, asking the court to find the
lawyer in contempt or initiating disciplinary
action. Where plaintifFs counsel “so mul
tiplies the proceedings in any case as to
increase costs unreasonably and vexa-
tiously," the attorney may be held per
sonally liable for such costs under 28
U. S. C. § 1927 The Government should
move to hold the plaintiff liable for costs
under this factor only where it would not
be unfair to charge the plaintiff with coun
sel’s unreasonable conduct. Generally, if
the client dearly did not know of counsel's
behavior or otherwise took no part in it,
the Government should not move for costs
based upon this factor. Moreover, in assess
ing the unreasonableness of plaintiff's « n -
duet, the Government should consider
whether its own conduct was at all times
reasonable.
In applying the third factor (baselessness
of the claim), the Government should heed
the Supreme Court’s cautionary language
in Christiansbvrg, 46 U. S. L. W. at 4107-08;
[I]t is important that a district court
resist the understandable temptation to
engage in post-hoc reasoning by conclud
ing that, because a plaintiff did not ulti
mately prevail, his action must have been
unreasonable or without foundation. This
kind of hindsight logic could discourage
all but the most airtight claims, for
seldom can a prospective plaintiff be sure
of ultimate success. No matter how
honest one's belief that he has been the
victim of discrimination, no matter how
meritorious one's daim may appear at the
outset, the course of litigation is rarely
predictable. Decisive facts may not
emerge until discovery or trial. The law-
may change or clarify in the midst of
litigation. Even when the law or the facts
appear questionable or unfavorable at the
outset, a party may have an entirely
reasonable ground for bringing suit.
While it is difficult to establish any
hard-and-fast rules for determining whether
a claim is patently frivolous, a few guide
lines may be offered to assist in that
determination. First, a plaintiff is less
likely to have a "frivolous” claim if he or
she is able to establish a prrma facie cast
Second, a claim based on a novel legal
theory should not be deemed “frivolous”
if the theory, though novel, is at least
debatable. Finally, if the plaintiff abandons
a frivolous suit promptly after the discovery
of the facts that demonstrate its baseless
ness, the Government generally should not
seek costs; the abandonment of patently
frivolous claims will thereby be encour
aged.
As a final note, there may be cases in
which the assessment of costs is appropriate
but the plaintiffs financial situation war
rants reduction of costs. Remember that
motions for costs are intended to deter
baseless or unreasonable daims and not to
create an extreme hardship on the plaintiff.
Back reference.—f 1211.
fl 5083 © 1979, Commcce Clearing House, Inc.