Stockham Valves and Fittings, Inc. v. Howard Petition for Writ of Certiorari

    IN THE

^uprrrnr (Enurt of tlj? Mntteii States
October T erm, 1977

No.

Stockham V alves and F ittings, I nc.,
Petitioner,

vs.
P atrick James, H oward H arville, L ouis W inston, 

On Their Own Behalf And On 
Behalf Of Those Similarly Situated,

U nited Steelworkers of A merica, AFL-CIO, 
And Local 3036, U nited Steelworkers of 

A merica, AFL-CIO,
Respondents.

PETITION FOR A WRIT OF CERTIORARI TO THE 
UNITED STATES COURT OF APPEALS FOR THE 

FIFTH CIRCUIT

D ouglas Arant 
John J. Coleman, Jr.
James P. Alexander 
W m . M ichael W arren, Jr.

1500 Brown-Marx Building 
Birmingham, Alabama 35203 
(205) 252-4500

Attorneys for Petitioner 
Stockham Valves and Fittings, Inc.

Of Counsel:
B radley, A rant, R ose & W hite 
L inda A. B unsey 
R obert K. Spotswood 

1500 Brown-Marx Building 
Birmingham, Alabama 35203 
(205) 252-4500

P r£3 3  or Byron S. Aham3 P rinting, Inc., W ashington, D. C.

\

*<-  ,X,J ■

j o  G  - ( 'M ’

/* '



I

TABLE OF CONTENTS
Page

I. O pin io n s  B elow  ..................................

I I . J urisdiction  .......................... ^

III. Q uestions P resented  .........................  g

IV. S ta tu tes  I nvolved ..............................  3

V. S ta tem en t  of th e  C a s e .................  ^

Tile District Court Decision.......  7
2. The Court of Appeals Opinion .................. 9

VI. R easons for G rantin g  t h e  W r i t ....... ...............  10

1. The . Decision of the Fifth Circuit Court of
Appeals Conflicts in Principle With Recent 
Decisions of the United States Supreme Court 
and Directly Conflicts With Decisions of the 
h ourth Circuit Court of Appeals on an Im- 
portant Recurring Issue Regarding the Proper 
Use of Statistical Evidence in Litigation Under 
Iitle VII of the Civil Rights Act of 1964___ 10

2. The Court of Appeals So Far Departed From
tlie Accepted Scope of Judicial Review By 
Reevaluating Findings of Fact Below, By Re- 
weighmg the Evidence, By Substituting Its 
Judgment as to Facts For That of the Trial 
Court, and By Rigidly Confining the Trial 
Court s Discretion Regarding Appropriate 
Relief, as to Call For an Exercise of this 
Courts lowers of Supervision .............. . lg

3. Evidence of Quantitative Differences in Edu­
cational Attainment, Universally Recognized 
lo  Influence Earnings, Cannot Be Rejected 
When Comparing the Earnings of White and
Black Employees..................“ .........  25

VII. C on clusion  .............................................



ii TABLE OF AUTHORITIES
Page

Cases:
Alabama Power Co. v. Ickes, 302 U.S. 4G4 (1938) .. 22, 23 
Albemarle Paper Co. v. bloody, 422 U.S. 405 (1975) 1G, 23 
Brown v. Board of Education, 347 U.S. 483 (1954) .. 25
Busey v. District of Columbia, 319 U.S. 579 (1943) .. 22
Commissioner v. Duberstein, 363 U.S. 278 (19G0) . . . .  17
Croker v. Boeing Co., -----  F. Supp. ----- , 15 F.E.P.

Cases 1G5 (E.D. Pa. June 20, 1977) .................... 12
Dayton Board of Education v. Brinkman, -----  U.S.

----- , 53 L. Ed. 2d 851 (June 27, 1977) . . . .  16, 23, 24
Dobbins v. Local 212, 1BEW, 292 F. Supp. 413 (S.D.

Ohio 19G8) ...........................................................  12
EEOC v. Eagle Iron Works, 424 F. Supp. 240 (S.D.

Iowa 1976) ...........................................................  12
EEOC v. United Virginia Bank, -----  F.2d ----- , 15

F.E.P. Cases 1257 (4th Cir. May 10, 1977) .......  14
Great Atlantic & Pacific Tea Co. v. Grosjean, 301

U.S. 412 (1937) .......................................................  22
Griggs v. Duke Power Co., 401 U.S. 424 (1971) . . . .  27, 29
Hazelwood School District v. United States,----- U.S.

----- , 53 L. Ed. 2d 768 (June 27, 1977) .. 10,11,14,16,
22, 26

Hester v. Southern Ry. Co., 497 F.2d 1374 (5th Cir.
1974) .....................................................................  14

Hill v. Western Electric, 12 F.E.P. Cases 1175 (E.D.
Va. 1976), appeal docketed, (4th Cir. No. 76-2439) 15 

International Salt Co. v. United States, 332 U.S. 392
(1947) ....................................................................  23

Mayor v. Educational Equality League, 415 U.S. 605
(1974)    12,16

Patterson v. American Tobacco Co., 535 F.2d 257 (4th
Cir.), cert, denied, 429 U.S. 920 (1976) ..............  13,14

Roman v. ESB, Inc., 550 F.2d 1343 (4th Cir. 1976). 12,13
Swint v. Pullman-Standard, -----  F. Supp. ------, 15

F.E.P. Cases 145 (N.D. Ala. July 5, 1977) ...........  12
Teamsters v. United States,----- U.S.-------, 52 L. Ed.

2d 396 (May 31, 1977) ........................................ 10,14
Uebersee Finanz-Korporation v. McGrath, 343 U.S.

205 (1952) ............................................................. 22
United States v. Greater Buffalo Press, Inc., 402 U.S.

549 (1971) ...........    23

Table of Authorities Continued iii

United States v. National Ass’n of Real Fstate 
Boards, 339 U.S. 485 (1950) ...... [  .. . * 1 -m

United States v. Yellow Cab Co., 338 U.S. 338 (1949) 18,

Washington v. Shell Research d> Development Co

17821 <SD- T« -

Z ‘ niv ‘s Em a %&V- nazeltinê
....................17,19, 22

S ta tu tes  :

28 U.S.C. § 1254(1) ..................
42 U.S.C. U0S1 ........................................................... 3 2

“  Eights J M o f i m '  Secti0”  ?03(a) of the Civil ^°  ts Act or 1964, as amended.......................passim
R u l e s :

Rule 52(a) of the Federal Rules of Civil Procedure 3,17, 24 
O th e r  A u th o rities  :

w' L Hr w "^ ^ ? l6o ? ,? oj?Iingand, E‘ ™îview 409 (1970)’ ^ n c a n  Economic Re-

F. C. Morris, Jr., Current Trends In The Use ( And 
f r \ K ° f  S.iatlstlcs In Employment Discrimina-
Counn J f X ° n . . (EqUal E" Pl0yment*************••••••», 15 16

B. R. Schiller, The Economics of Poverty and Die crimination (1973) ___ wcriy ana Dis-

L. C. Tliurow, Poverty and Discrimination (1969) .. 28

& “(f 9? 3)Human The ^

Action Programs W yants for Affirmative
...................................................................5 ,1 9



IN THE
&ttpr*m* (Erntrt o f  %  Itttfrft States

O ctober T erm, 1977

No.

S tockham  V alves and F ittings, I nc.,
Petitioner,vs.

P atrick J ames  ̂ H oward H arville, L ouis W inston 
On Their Own Behalf And On ’

Trmfr? eiQlf Ttose Similarly Situated
UAnrLocALLW K nS °F AJierica- AFL-CIO, And L ocal 3036, U nited Steelworkers of

A merica, AFL-CIO,
Respondents.

PETITION FOR A WRIT OF CERTIORARI TO tttp 
UNITED STATES COURT OF APPEALS FOR THE 

FIFTH CIRCUIT

,.«Tlle*?enlti0Iler Stockham Valves alKi Fittings Ine
view y -P7 yS ‘ !lat 3 WrH of 4 u e  to

An™ l /  o f the Uniteci States Court of
Appeals for the Fifth Circuit entered in this action 
on September 19, 1977.

I. OPINIONS BELOW

forT tte °F m nc°f  “ •! Y " “ ed States Court »f  Appeals toi the Fifth Circuit dated September 19, 1977 is re­
ported at 559 F.2d 310 and appears in the separate



2

Appendix to the petition at pages 2-99. The Findings 
of Fact and Conclusions of Law of the United States 
District Court for the Northern District of Alabama 
dated March 19, 1975 are reported at 394 F. Supp. 434 
and appear in the separate Appendix at pages 100-22 .

II. JURISDICTION

The judgment of the Court of Appeals was entered 
September 19, 1977. (App. 1). On October 13, 1977, 
the Court of Appeals stayed its mandate to and includ­
ing November 12,1977. (App. 224-25). Jurisdiction of 
this Court is invoked under 28 U.S.C. § 1254(1).

III. QUESTIONS PRESENTED

The questions presented for review are:
1. Whether an employer of blacks in numbers 

which far exceed their representation in the relevant 
labor market is obligated 'as a matter of law to insure 
that blacks are distributed among jobs at all skill 
levels at a percentage equal to black participation in 
the employer’s total work force without regard for the 
qualifications required for various jobs.

2. Whether the Court of Appeals invaded the prov­
ince of the trial court by substituting its judgment as 
to facts for the trial court’s findings of fact and by 
•precluding the trial court on remand from making its 
own findings of fact and exercising its own discretion 
with respect to appropriate relief.

3 Whether an inference of employment discrimina­
tion may be premised on earnings disparities between 
blacks and whites without regard for quantitative 
differences in educational attainment, a productivi y 
factor universally recognized to affect employee earn­
ings.

3

IV. STATUTES INVOLVED

1. 42 U.S.C. §1981:

All persons within the jurisdiction of the United 
States shall have the same right in every State 
and Territory to make and enforce contracts, to 
sue, be parties, give evidence, and to the full and 
equal  ̂benefit of all laws and proceedings for the 
security of persons and property as is enjoyed by 
white citizens, and shall be subject to like punish­
ment, pains, penalties, taxes, licenses, and exac­
tions of every kind, and to no other.

2. Section 703(a) of the Civil Rights Act of 1964, 
as amended, 42 U.S.C. § 2000e-2(a):

It shall be an unlawful employment practice for 
any employer—

( 1 ) to fail or refuse to hire or to discharge any 
individual, or otherwise to discriminate against 
any individual with respect to his compensation, 
terms, conditions, or privileges of employment, 
because of such individual’s race, color, religion, 
sex, or national origin; or

( 2) to limit, segregate, or classify his employees 
or applicants for employment in any way which 
would deprive or tend to deprive any individual 
of employment opportunities or otherwise ad­
versely affect his status as an employee, because 
of such individual’s race, color, religion, sex, or 
national origin.

3. Rule 52(a) of the Federal Rules of Civil Pro­
cedure :

. . . Findings of fact shall not be set aside unless 
clearly erroneous, and due regard shall be given 
to the opportunity of the trial court to judge of 
the credibility of the witnesses. . . .



!

V. STATEMENT OF THE CASE

This petition raises ^uestionso^natiom  
tance as to the proper use of statistical data as evi- 
dence of racial discrimination in job assignments and 
as to the appropriate role of a court of appeals in re­
viewing district court Endings of fact and in fashion- 
ing relief at the appellate level.1

The individual respondents (plaintiffs below), three 
long-service black production and maintenance em­
ployees of Stockham Yalves and Fittings, Inc., com­
menced this action individually and on behalf of other 
black production and maintenance employees similarly 
situated for claimed violations of Title Y II  of the 
Civil Rights Act of 1964, as amended, and 42 U.S.C. 
§ 1981. (App. 3).

Stockham employs more than 1,800 production and 
maintenace employees at its Birmingham valves and 
fittings manufacturing facilities.2 Down through the 
years at least two-thirds of these employees have been 
black (App. 105), although the Birmingham Stand­
ard Metropolitan Statistical Area ( “ SMSA” ) in

4

1 The Union respondents, defendants below, at all times material 
to this action were the bargaining representatives for the individual 
plaintiffs and Stockham’s hourly production and maintenance em­
ployees. (App. 103).

! Jobs at Stockham are divided according to the complexity of the 
job and skills required into Job Classes 2 through 13. (App. 1G6). 
High skilled production and craft jobs fall within Job Classes 10-13. 
Although the base pay rate of a job increases as the job class 
increases, actual earnings of workers are not determined by job 
class because virtually all of Stockham’s production employees 
receive incentive pay (which averages approximately 25 percent of 
base pay). (App. 10G). High skilled and craft jobs are non-incentivo 
jobs. (App. 116).

5

which they reside is only 25 percent black. (App. 
161). Each of the plaintiffs left previous employment 
with other Birmingham employers because of the su­
perior work opportunities available to him at Stock­
ham. (App. 199, 201). Many of the black employees 
at Stockham are relatively unskilled; but for their 
positions with Stockham they would otherwise be con­
sidered “ hard core unemployables” .’ (App. 129).

Plaintiffs claimed in their complaint and at the trial 
that plaintiffs as a class were the subject of discrimi­
nation by Stockham, inter alia, in the allocation of 
jobs, but plaintiffs proffered no^evidence that Stock­
ham had, in fact, discriminated in initial assignments 
o,.]:i.P.I-2IJl0ti°ns against any black employee vis-a-vis any 
white employee with equal or poorer qualifications.4

* Stockham’s black work force was so characterized by Dr Rich- 
ard Barrett a witness for plaintiffs on the testing issue, when he 
visited Stockham s Birmingham plant in 19G8. In 1974 Stockham 
received an award from the Birmingham Urban League referable 
to the fact that it had hired more minority referrals durirm the 
preceding year than any other employer. (App. 129).

‘ Although Stockham employed Wonderlic tests as a factor in the 
evaluation of its employees for hiring and promotion purposes 
betueen 1965 and 1971, plaintiffs proffered no proof whatever to 
reflect that the black employees at Stockham scored worse than the 
white employees on the tests; instead plaintiffs relied upon a publi- 
cation entitled Negro Norms: A Study of 38,452 Job Applicants for 
Affirmative Action Programs, prepared by E. P. Wonderlic & Asso-
and dcfIn d f q 1?' Expert testing Psychologists for plaintiffs and defendant Stockham were in total agreement that Negro Norms 
offered no reliable information with respect to test scores referable 
to tho black popu ation at Stockham. (App. 184). While it is true 
that one of Stockham s expert witnesses recalled some incomplete 

ata with respect to Wonderlic test scores at Stockham, the same 
ltness further expressed her uncontradicted judgment that such 

data was so inadequate that no reliable conclusion could be reached



6

Furthermore, plaintiffs introduced no “ pattern”  evi­
dence as to the relative qualifications of blacks and 
whites at Stoekham and chose to rely instead upon 
statistical evidence to support their claim that because 
of race, blacks, as a class, were allocated the poorer 
paying, hot and dusty jobs and whites the better pay­
ing jobs with less onerous working conditions.5 Ac­
cording to plaintiffs, Stoekham discriminated against 
blacks in job assignments and promotions because (i) 
only 5 percent of the employees in the relatively few 
high skilled and journeyman craft jobs at Stoekham 
are black; and (ii) average hourly earnings of blacks 
at Stoekham are 91 percent of average hourly earn-

with respect to the racial impact of the tests. (App. 185). The Dis­
trict Court found, predicated upon these facts, that there was no 
proof that the Wonderlic test had a racially disproportionate im­
pact on blacks at Stoekham in violation of Title VII. (App. 186).

• Statistical references to the high proportion of black employees 
at Stoekham whose jobs involve hot and dirty working conditions 
seemingly create a “ straw”  issue. Working conditions are immut­
able, and working conditions in foundries are by nature hot and 
dusty. (App. 114). The conception that blacks in substantial num­
bers gravitate to foundry jobs because of their qualifications for 
such jobs is surely more acceptable evidence of nondiseriminatory 
employment policies, in the absence of contradictory evidence, than 
the conception that an employer contrary to his economic interests 
deliberately assigns blacks to hot and dusty jobs merely because 
they are black. Furthermore, plaintiffs’ claims must be considered 
in light of the following facts of record: (i) a great majority of all 
production and maintenance employees at Stoekham are black 
(App. 105); (ii) hot and dusty working conditions are a character­
istic of many of the high skilled jobs held predominately by white 
employees at Stoekham (App. 130); and (iii) there was no evidence 
that any black employee at Stoekham was exposed to more onerous 
conditions on his job than any white employee at Stoekham in the 
same job, or, indeed, more onerous conditions than those of any 
white employee at Stoekham in any job for which a black employee 
was qualified. (App. 115).

7

ings of whites (without adjustments to take account 
of productivity factors such as absenteeism, job skills 
education, etc.). ’ J ’

1* * The District Court Decision

The District Court, viewing plaintiffs’ statistical 
evidence in the context of other salient facts of record, 
concluded that it did not support the claims of dis­
crimination by Stoekham against blacks in either ini­
tial job assignments6 or promotions.7

T h eD isln d  Court, after hearing the witnesses, judging their 
credibility and reviewing the documentary evidence of record, 
expressly found, inter alia, that since the effective date of Title VII ■
(i) there was no evidence that any black applicant at Stoekham had
S I S  and / ■•wJiemcA 3 j °b in favor of a white applicant at Stoekham; (n) there were virtually no black applicants for em-
ployment in the greater Birmingham area who possessed the high 
skills needed for craft jobs at Stoekham (App. 128) • (iii) the 5 
percent representation of blacks in high skilled jobs at Stoekham 
compared favorably with the representation of blacks possessing 
such skills who live in the Birmingham SMSA, according to U.S° 
Census statistics (App. 128); (iv) without exception every black 
employee at Stoekham possessing skills of a journeyman or the 
equna ent was employed by Stoekham in a job commensurate with

( PP‘ 128,} :  (V) the earuil>gs of black applicants 
lured by Stoekham since the effective date of Title VII are sta­
tistically identical to those of post-Act white hirees (App 163) ■

A T 7  0f, St0ckhara’3 black employees are relatively un­
skilled; but for the work opportunities offered at Stoekham to 
blacks, a substantial portion of its black employees would be “ hard 
core unemployables.”  (App. 129).

. T The tD'strict Court’s ultimate conclusion on this issue, adverse 
o p aintiffs, was predicated upon, inter alia, the following facts:

(i) plaintiffs failed to produce any evidence that Stoekham had 
discriminated against any black employee by failing to pro,noS him 
to a job for which he was qualified or for which he possessed quali-

c m p lo jit  S ,  lhM'  by WhUc inCUmb“ t



8

The District Court, on the basis of these facts, 
reached the judgment that “ [t]he relatively small 
number of blacks in certain high skilled and craft jobs 
at Stockham is due not to any discriminatory prac­
tices at Stockham but due instead to the absence of 
qualified blacks.”  The Court added:

An employer is entitled to insist that his workers 
be qualified and as long as the qualifications, as in 
this case, are not artificial or established with an 
intent to discriminate, the employer is not required 
to place individuals of any race who lack such 
qualifications on the job. Plaintiffs have failed to 
establish racial stratification, through either ini-

(ii) there were no lines of progression at Stockham and no on-the- 
job training programs pursuant to which an employee automatically 
became qualified for a higher rated job by virtue of performing the 
duties of his current position (App. 123);

(iii) any Stockham employee may file a “ timely application”  
for a desired job at any time whether or not the job is vacant, and 
mav maintain applications on file for several jobs at once (App. 
123);

(iv) during the relevant period nearly 1,200 timely applications 
were filed, 609 of which were filed by black employees; 27 percent 
of the timely applications filed by blacks, as compared with a 
statistically commensurate 31 percent of timely applications filed 
by whites, were granted (App. 124);

(v) there are equal earnings opportunities in almost all of the 
departmental seniority units (i.e., in ten of the twenty-two seniority 
units unadjusted black gross earnings exceeded unadjusted white 
gross earnings, and in nine of the seniority units the unadjusted 
black hourly earnings exceeded unadjusted white hourly earnings) 
and there was no showing that black employees were denied jobs 
either within departmental seniority units or across departmental 
lines (App. 125, 166) ; and

(vi) plaintiffs’ average hourly earnings as a class in proportion 
to white average hourly earnings at Stockham had increased from 
85 percent to 91 percent since the effective date of Title VII. (App. 
162).

9

tial job assignments or promotion and transfer 
decisions, in the job classification system at Stock­
ham. (App. 213).

2. The Court of Appeals Opinion

The Court of Appeals reversed. The Court concluded 
statistics alone in the case sub judice “ establish 

a clear prima facie case of purposeful discrimination.”  
(App. 35). According to the Court, it was not material 
that black representation in high skilled and craft 
jobs at Stockham compares favorably with local and 
national labor markets. “ The relevant work force for 
comparison purposes is Stockham where 66 percent of 
all maintenance and production workers are black”  
(App. 59) and “ [bjlacks earn, on the average, $0.37 
less per hour than whites.”  (App. 35). The Court did 
not consider other highly relevant statistical evidence 
introduced by Stockham; indeed, it even disregarded 
evidence as to the earnings of Stockham employees 
hiied after the effective date of the Act stating that 
“ relative changes”  in earnings of such employees “ re­
cently hired”  (i.e., hired since the effective date of Title 
V II ) are “ irrelevant to the question of discrimination 
at Stockham.”  (App. 39). The Court also rejected 
Stockham s multiple regression analysis which an­
alyzed black earnings relative to white earnings in 
terms of various productivity factors. The Court chal­
lenged the use of education as a productivity factor 
to explain earnings differences between whites and 
blacks. White employees at Stockham have more edu­
cation than blacks and education “ is not a job require­
ment at Stockham hence, the Court declared, educa­
tion is irrelevant to adequate job performance.”  
(App. 41). In so concluding, the Court ignored Dis­
trict Court findings predicated upon expert testimony



10

that an individual’s educational level, regardless of 
race, impacts earnings.8 (App. 41).

VI. REASONS FOR GRANTING THE WRIT

There are three special and important reasons why 
the writ should be granted:
1. The Decision of lhe Fifth Circuit Court of Appeals Conflicts In 

Principle With Recent Decisions of the United Stales Supreme 
Court and Directly Conflicts With Decisions of the Fourth 
Circuit Court of Appeals On An Important Recurring Issue 
Regarding the Proper Use of Statistical Evidence In Litigation 
Under Title VII of the Civil Rights Act of 1964.

The Court of Appeals failed to apply the princi­
ples recently mandated by this Court on the^projDe^ 
use of statistical evidence in Title V II  cases, in
Teamsters v. United State s ,------ U .S .------- , 52 L. Ed.
2d S9(>, 4iT (T lay  31, 1977), this Court acknowledged 
that statistics play an important role in employment 
discrimination cases but admonished that “ statistics 
are not irrefutable . . . their usefulness depends on all 
of the surrounding facts and circumstances.” Writing 
for the majority in Hazehvood^School v.
United States,------U .S .------- , 53 U  F U 2cl fm t vVV-78
n.12 (June 27, 1977), Mr. Justice Stewart recognized 
that “ when special qualifications are required to fill 
particular jobs, comparisons to the general population 
(rather than to the smaller group of individuals who 
possess the necessary qualifications) may have little 
probative value.” The Fifth Circuit failed to consider

*  8 The superficiality of the Fifth Circuit’s statistical analysis 
may explain why the Court thereafter (App. 35, 37) attempted to 
premise an inference of racial discrimination in job assignments 
upon the fact that until early 1974 some Stockham facilities were 
segregated.

■ 14 7 7

i i

the “ necessary qualifications” for the effective per­
formance of the various jobs at Stockham. Instead, 
that Court relied on undifferentiated statistics reflect­
ing the racial composition of Stockham’s entire pro­
duction and maintenance work force as the basis for 
its erroneous conclusion that Stockham discriminates 
against blacks by placing them in less desirable jobs. 
In effect, the Fifth Circuit’s decision is premised on 
the irrational assumption that each Stockham em­
ployee is presumptively . qualified for eveiy job at 
Stockham regardless of the skills required for success­
ful performance of that job.

The Fifth Circuit held that Stockham discriminated 
against blacks because blacks did not occupy craft and 
high skilled jobs in percentages comparable to the 
percentage of blacks in its entire production and 
maintenance work force. Conversely, the District 
Court concluded that the five percent representation 
of blacks in craft jobs at Stockham was not an under­
representation of blacks compared to the percentage 
of blacks with the necessary qualifications for those 
jobs in the local and national labor markets. (App. 
128). The Court of Appeals reversed this finding as 
‘ 1 clearly erroneous” :

The relevant work force for comparison purposes 
is Stockham where 66 percefit of all maintenance 
and production workers are black. When com­
pared with that figure, 5 percent looks paltry in­
deed. (Footnote omitted; App. 59).

The Fifth Circuit’s analysis fails to conform to the 
mandates of this Court. First, contrary to Hazelwood, 
it ignores the unrebutted evidence credit.erj hv thp. Pis- 
trict Court that special skills are muiired- for the ef­
fective performance -of many jobs held by Stockham 
employees. (App. 134-56). Due to job requirements at



12*

Stockham, “ this is not a case fn which it can be as­
sumed that all [employees] are fungible for purposes 
of determining whether members of a particular class 
have been unlawfully excluded.” Mayor v. Educational 
Equality League, 415 U.S. 605, 620 (1974). See Sivint
v. Pullman-Standard,------  F. Supp. ------ , 15 F.E.P.
Cases 145, 150 (X.D. Ala. July 5, 1977); Croker v.
Boeing C o .,------F. S upp .--------, 15 F.E.P. Cases 165,
203 (E.D. Pa. June 20, 1977); Washington v. Shell
Research & Development C o .,------F. S u pp .------- , 14
EPD H 7821 (S.D. Tex. March 17, 1977); EEOC v. 
Eagle Iron Works, 424 F. Supp. 240, 14 F.E.P. Cases 
536, 543 (S.D. Iowa 1976); Dobbins v. Local 212, 
IB E W , 292 F. Supp. 413, 445-46 (S.D. Ohio 1968). 
Second, the credited and unrebutted evidence is, and 
the District Court so found, that every black employee 
at Stockham who is qualified for a high skilled job 
already works in a position commensurate with his 
qualifications. (App. 128, 134). Absent evidence that 
each employee in a low skilled job could obtain the 
necessary qualifications for a high skilled position 
through on-the-job experience, the only valid bench­
mark for assessing whether an employer’s policies 
and practices have a discriminatory impact on blacks 
seeking high skilled jobs is the percentage of blacks in 
the relevant labor market possessing the qualifications 
needed for the effective performance of the high 
skilled jobs.

The Fifth Circuit Court of Appeal’s reliance on an 
undifferentiated comparison between the racial com­
position of the entire production and maintenance 
work force and that of the high skilled jobs as proof 
of racial discrimination conflicts_with several decisions 

ytu* by the Fourth Circuit Court o f Appeals. In Roman v. 
V - ESB, Inc.. 550 F.2d 1343 (4th Cir. 1976) (en banc),

13

plaintiffs charged their employer with, among other 
mgs, racial discrimination in job assignments in 

violation of Title V II. In concluding plaintiffs failed 
to establish that blacks were underrepresented in 
craft, managerial, and clerical positions, the Fourth 
Circuit repeatedly compared the percentage of blacks 
in those positions to the pmuai£nge of blacks in the 
relevant labor mnrVAt “ gualifiedfSTsiieh wnrlr » JT** 
at 1354-55. By failing to follow "the statistical meth- 
odology used by the Fourth Circuit in Roman and 
thereby comparing the percentage o f blacks in craft 
and high skilled positions at Stockham to the percent­
age of blacks in the relevant labor market qualified 
for such work, the Fifth Circuit ignored the proscrip­
tions of this Court concerning the use of statistics 
m the context of self-evident facts of industrial life. 
The conflict therefore, between the Fifth Circuit and 

e Fourth Circuit is fundamental and irreconcilable.
This conflict between the Fifth Circuit and the 

Fourth Circuit on the use o f comparative work force 
statistical evidence m employment discrimination 
cases is again apparent in Patterson v. Amp.n'mv To-

' P f  denied, 429
. ; 920 (1976)- Patterson involved the Fourth Cir- 

cmt s review of a trial court’s order mandating pref­
erential promotions for women and blacks until their 
percentages in the employer’s supervisory work force 
equaled their percentages in the general Richmond 
SMSA work force. The Fourth Circuit determined 
that the proper comparison was between the percent­
ages o f women and blacks in the Richmond SMSA 
supervisory work force and the percentages of women 

and blacks promoted since 1965 by the employer since 
[t]hose percentages furnish a more realistic mcas-



14

ure of the company’s conduct.”  535 F.2d at 275. 
is difficult to imagine a less realistic benchmark o 
Stockham’s treatment of its minority _ employees than 
the Fifth Circuit’s simplistic statistical work force 
comparison utilized in this case.

The issue presented on the proper use of compara­
tive work force statistics arises in jO illiaill evei7  Ŝ' 
t.inn under Title V II . involving claims of racial dis- 
/.riTninnt.inn in job  assignments and iiupromotions. The 
recent opinions'in Teamsters and Hazelwood provide 
considerable direction on the proper use of statistical 
work force comparisons as evidence of racial discrim­
ination in hiring; this Court’s review of the decision 
of the Fifth Circuit Court of Appeals in this case 
would clarify the role of statistical evidence m Title

•Accord EEOC v. United Virginia Bank, ------  P-2d , 15
F E P Cases 1257, 1259 n.7 (4th Cir. May 10, 1977) (“ Where the 
work requires special qualifications, it is proper to. consider the ratio 
of qualified blacks and whites in the appropriate work force rather 
than the ratio of the gross percentage of blacks and women in the 
whole work force, including, unskilled labor.” )

>» See also Hester v. Southern By. Co., 497 F.2d 1374 (5th Cir. 
1974) where the Fifth Circuit refused to find pi’oof of racial 
discrimination in hiring for a data typist job based on a comparison 
between the percentage of blacks in the general SMSA and the 
employer’s work force:

\C]omparison with general population statistics is of question­
able value when we are considering positions for which, as here, 
the general population is not presumptively qualified Data 
Typist applicants were required to prove their ability to type 
at a minimum speed of sixty corrected words per minute as a 
prerequisite to consideration by Southern for employment . . . 
A more significant comparison might perhaps be between tho 
percentage of blacks in the population consisting of those able 
to type GO wpm or better and, the percentage hired into the 
Data Typist position by Southern. (Italics supplied; Id. at 
1379 n.6.)

15

Y II  cases involving claims of discrimination in ini­
tial job assignments and in promotions.

The statistical methodology mandated by the Fifth 
Circuit in this case requires an inference of discrimi­
nation when blacks are not equally represented at all 
levels of an employer’s work force, without regard for 
qualifications. The anomalous result of the Fifth Cir­
cuit’s rule is to penalize an employer, such as Stock- 
ham, who has attracted to its work force a large num­
ber of blacks, including many of the “ hard core un­
employed.”  (App. 129). See Hill v. Western Electric, 
12 F.E.P. Cases 1175, 1179 n.4 (E.D. Va. 1976), ap­
peal docketed, (4th Cir. No. 76-2439). It is more likely 
that such an employer will be found guilty of unlaw­
ful discrimination in job assignments than the em­
ployer whose job opportunities for blacks are limited 
to the percentage of blacks in the relevant labor 
market."

The decision of the Fifth Circuit in this case illus­
trates the uncertainty presently surrounding the 
proper manner to prove racial discrimination in job 
assignments thrnmrh ;------' ri:
forces. This uncertainty obviously impedes the effec­
tive implementation of the equal employment oppor­
tunity laws and the attempts of employers to identify 
and comply with the mandates of such laws. See, e.g., 
F. C. Morris, Jr., Current Trends In The Use (And 
Misuse) Of Statistics In Employment Discrimination

11 As the District Court found, based upon unrebutted expert 
testimony, the proper inference to be drawn from statistics showing 
a significant over-representation of blacks is that Stockham offers 
superior job opportunities to blacks and that blacks migrate to 
Stockham jobs. (App. 161-G2).



16

Litigation (Equal Employment Advisory Council 
1977). Proper resolution of the issue by this Court is 
necessary to bring order to the chaotic state of em­
ployment discrimination case law.

2. The Court of Appeals So Far Departed From the Accepted 
Scope of Judicial Review By Reevaluating Findings of Fact 
Below, By Reweighing the Evidence, By Substituting Its Judg­
ment As To Facts For That Of The Trial Court, and By Rigidly 
Confining the Trial Court's Discretion Regarding Appropriate 
Relief, As To Call For An Exercise Of This Court's Powers Of 
Supervision.

This Court has become increasingly concerned with 
encroachment of appellate courts upon district courts, 
particularly in cases that may turn upon statistical evi­
dence and the credibility of witnesses or that may re­
quire injunctions of broad impact affecting opportuni­
ties and expectations of a large group of people. In­
deed, this Court has said that the proper allocation of 
functions between district courts and courts of appeals 
raises issues “ every bit as important”  as the issues 
raised by an appeal on the merits, including issues 
raised in a racial discrimination case. Dayton Board,
of Education v. Brinkman, ------U .S .--------, 53 L. Ed.
2d 851, 857, 862 (June 27, 1977); Hazelwood School 
District v. United States, supra, 53 L. Ed. 2d at 780; 
Albemarle Paper Co. v. Moody, 422 U.S. 405 (1975) ; 
Mayor v. Educational Equality League, 415 U.S. 605, 
621 (1973).

In the case at hand the Court of Appeals engaged 
in a de novo fact-finding process, disregarding the 
weight the trial court accorded the evidence. There­
after, it substituted its judgment second hand for 
that of the trial court and fashioned its own remedies.

17

Such an invasion of the trial court’s functions war- 
rants the exercise bv this Court of its supervisory 

not merely because it permeates the entire 
appellate opinion in this case, thus vitiating its legiti­
macy as an appellate decision, but more significantly 
because it impugns the integrity of trial courts gen­
erally and undermines the established, efficient, and 
oiderly allocation of functions in the federal judiciary 
Pair employment statutes have resulted in a prolifera­
tion of complex and lengthy lawsuits, won or lost by 
means of documentary evidence, statistical proof, and 
conflicting testimony. I f  courts of appeals are per­
mitted to reevaluate evidence, determine anew the 
weight and credibility of witnesses’ testimony, and 
adjudicate relief, the value of a trial court’s intimacy 
with the facts over a long period will be forfeited and 
replaced by the dangers attendant to the necessarily 
circumscribed vision of an appellate court, straining 
to see the evidence itself from afar.

+. ^  acknowledging the clearly.ermncnn. r u ]^  
the rifth Circuit reweighed items of evidence bearing 
on critical issues and, after overturning factual find- 
mgs below, utilized its own substituted factual findings 
to pyramid and overturn still other trial court findings 
By reference to its own findings, it thus boot-strapped 
reversals of factual findings by the trial court. For 
example, the Court of Appeals extracted testimony of 
one witness regarding two medical dispensary rest-

12 Fed. R. Civ. P. 52(a). The rule forbids setting aside factual 
findings on appea! unless they are clearly emmeou. and, as inter- 
p eted by this Court, prohibits de novo determination of factual

95 USnia0P0PT 3 n t i Rao i i0 Cr • V- JIazMne K s e a lc tZ  • . V " ?  ( M * ) -  The rule applies to all factual finding 
eluding factual inferences from undisputed facts or documents’ 

Commissioner v. Duberstein, 3G3 U.S. 278, 289-91 (19G0).



18

rooms, construed that witness’ testimony directly con­
trary to the trial court’s own construction and credi­
bility judgment, and thereby changed the facts con­
cluding that the two restrooms were segregated at the 
time of trial. This independent appraisal of oral testi­
mony contravenes the rule requiring that due regard 
be given to the trial court’s opportunity to assess testi­
mony of witnesses. United States v. 1 elloxo Cab Co., 
338 U.S. 338, 341-342 (1949).

The Court of Appeals next overturned the trial 
court’s finding that a dispute over alleged segregation 
of facilities had been resolved by a 1974 conciliation 
agreement between Stockham and the EEOC, predi­
cating its reversal on the proximity of that agreement 
to trial time. According to the Fifth Circuit, the evi­
dence”  of the racially separate dispensary restrooms 
and the company’s “ intransigent resistance”  to de­
segregation of plant facilities indicate that the Dis­
trict Court ‘ 1 over-relied on the conciliation agreement 
(App. 19), despite the explicit, unchallenged findings 
below of Stockham’s efforts over a period of several 
years to resolve the facilities issues with the EEO 
( App. 180-81) ; efforts which were directly frustrated 
by that agency’s failure to perform its statutorily di­
rected conciliation role. The finding of Stockham s 
“ intransigent resistance” , repeated throughout the 
Fifth Circuit’s opinion, was totally without eviden­
tiary support and directly contrary to the District 
Court’s finding of Stockham’s good faith attempts to 
comply with the Act which was neither noted nor over­
turned by the Court of Appeals. The Fifth Circuit 
thus ignored the prohibition against de novo fact-find­
ing on review. It also breached the rule that a finding 
is not clearly erroneous merely because the reviewing

19

court gives the facts another construction, resolves am­
biguities differently, or attributes a more sinister cast 
to actions deemed innocent by the trial court. Zenith 
Radio Corp. v. IlazeUine Research, Inc., 395 U.S. 100, 
123 (19G9) ; United States v. National Ass’n of Real 
Estate Boards, 339 U.S. 485, 495-96 (1950).

Moreover, these invasions of the trial court’s func­
tions were aggravated when the Fifth Circuit utilized 
its own findings regarding the dispensary restrooms 
and the “ intransigent resistance”  to desegregation to 
fashion, at least in part, its independent findings of 
discrimination in initial job assignments (App. 43) 
and in promotions (App. 68), despite the total absence 
of evidence that any purported segregation of em­
ployee facilities ever affected job assignments. The con­
clusions of job discrimination are thus fatally flawed 
because they are premised on invalid (as well as irrele­
vant) findings made by the Court of Appeals in dero­
gation of the District Court.

Invasions of the trial court’s traditional province 
so permeate the Fifth Circuit’s opinion that it reads 
like a district court’s findings of fact. Indeed, the 
District Court and Court of Appeals opinions appear 
to address entirely different lawsuits. The disparity is 
not due to clearly erroneous findings by the trial court 
but, in large part, to independent findings by the Fifth 
Circuit conjured from non-existent evidence and in dis­
regard of unrefuted evidence below," or premised on

"  The Pifth^Circuit made findings contrary to the., undisputed 
evidence when it :

(i) based its conclusion that the Wonderlic test adversely impacts 
black Stockham employees in part upon a nationwide study, (Negro 
Norms), not involving Stockham. (App. 47, 51). The only testimony 
regarding this study was'that of experts for both parties who agreed



20

credibility judgments contrary to those made by the 
trial court,”  or developed de novo after reversals o

that no inference of adverse impact at Stockham may be predicated 
uron such evidence. The irrelevance of this study to Stockham is 
confirmed by the EEOC’s “ Guidelines on Employee Selection
Procedures” , 29 C.F.R. §§ 1607.0, et seq.;

(ii) concluded that the trial court “ gave too much we,0ht to 
plaintiffs’ failure to offer evidence of actual scores blacks and 
whites achieved on the Wonderlic test because accumidatmn of the 
evidence would have been too burdensome on plaintiffs. (App.
& n 391 The Court’s finding of burdensomeness, which has its 
^  in plaintiffs’ appellate'brief, teas made without any si,«w »* 
that a survey of at least a representative sample of scores at Stoe 
ham was impossible or unrealistic (App. 185 H (e )) and is unsup-

P°(iii) based^itTconclusion of discriminatory job allocations in part 
upon its independent finding that the “ vast majority of blacks 
work in undesirable (particularly hot and dusty) working condi- 
tions (App. 35, 75). The Court ignored but left in tact findings 
below, based on unrebutted evidence and made on a.J°b-hy-Job 
basis, that many high skilled jobs occupied Prf  ° “ n ly ^  
whites have onerous working conditions (App. 130 (HI 6 & 7, 134-50, 
especially ^ 29, 31, 32, 34, 35, 37) and that some of the jobs in. 
which blacks predominate and which are associated with hot and 
dusty conditions are “ more desirable”  skilled jobs and, ironica v, 
historically have been considered “ white jobs”  at other employers 
in the southeastern United States. (App. 114, 127 ). i oreovcr,
the Court ignored the obvious reality that since blacks constituted 
two-thirds of all production and maintenance employees at Stoclc- 
ham even if blacks were evenly distributed through all job classes 
as the Court suggests, blacks would continue to predominate in all 
jobs with “ undesirable”  working conditions; and

(iv) remanded the issue on whether the seniority system is bona 
fide (App. 85, 91), despite the District Court’s express, undisturbed 
factual finding that the system was “ developed because of func­
tional, nonracial reasons.”  (App. 127 (16). The District Court s 
finding, which constitutes a conclusion that the system is bona fide, 
was entirely ignored by the Fifth Circuit despite the fact that the 
Court of Appeals seemingly accepted Stockham’s departmental 
seniority system as bona fide to the extent that it was utilized by 
plaintiffs as a productivity factor favorable to blacks in plaintiffs 
statistical presentations.

11 The Court of Appeals accorded no deference to the trial court s

21

legal standards utilized by the District Court, which 
thus never had the opportunity to make factual find­
ings in accordance with the legal standards enunciated 
by the Court of Appeals."

Such appellate fact-finding represents a drastic de­
parture from fundamental principles of appellate re­
view. These principles dictate that a reviewing court * (i)

opportunity to observe the demeanor of witnesses and to evaluate 
their testimony when it :

(i) believed Jack H. Adamson’s testimony that Stockham does 
not utilize the Tabaka tests in making employee selection decisions 
(App. 54), whereas the District Court believed the same witness’ 
testimony, not referred to in the Fifth Circuit's opinion, that the 
tests were utilized (App. 191);

(ii) credited the testimony of plaintiffs’ statistician, Martin 
Mador, who made a poor impression at trial, admitted to numerous 
errors on cross-examination, and whose mathematics the district 
court considered “ simple”  and questionable “  ‘ statistics.’ ”  (App. 
156). The trial court attached greater weight to the testimony of 
defendant’s expert Dr. Gwartney and to his unrefuted testimony 
regarding the need to adjust an earnings comparison to take into 
account productivity factors;

(iij) reversed the District Court’s finding that Stockham had not 
discriminated against the named plaintiffs, although the record 
lacked substantive evidence of discriminatory acts, or of their quali­
fications for any jobs they were purportedly denied. In particular 
the District Court attached little weight to plaintiff James’ testi­
mony, partly because of his audacious exaggeration of his educa­
tional accomplishments which were impeached by a sealed tran­
script, showing a majority of failing grades and many fewer course 
hours than he represented, which was included in the record on 
appeal; and

(iv) extracted, out of context, Dr. Haworth’s testimony in foot­
note 39. See note 13 ( i ), supra.

"  For example, the Fifth Circuit rejected as “ legally irrelevant”  
the trial court’s finding of no discrimination in craft positions, 
premised on absence of evidence that any qualified black had been 
rejected for a craft job. (App. 58). After five.pages of appellate 
fact finding, the Court concluded that Stockham had discriminated 
against blacks in selection and training of craftsmen.



22

must not substitute its judgment as to facts for that of 
the trial court by deciding whether it would have 
found otherwise, but must confine its review to deter­
mining whether the trial court could permissibly find 
as it did. Zenith Radio Corp. v. Hazel tine Research, 
Inc., supra. Accordingly, “ where the evidence would 
support a conclusion either way but where the trial 
coui’t has decided it to weigh more heavily for the de­
fendants . . . [s]uch a choice between two permissible 
views of the weight of the evidence is not ‘ clearly er­
roneous’.”  United States v. Yellow Cab Co., supra, 338 
U.S. at 342. Findings which have substantial support 
in the evidence will be accepted by the reviewing court 
as unassailable. Alabama Poiver Co. v. Iclces, 302 U.S. 
464, 477 (1938); Great Atlantic & Pacific Tea Co. v. 
Grosjean, 301 U.S. 412, 420 (1937). The Fifth Circuit 
abrogated these established principles by weighing 
testimony and other evidence anew.

Moreover, the Court of Appeals’ reversal of the 
legal standards employed below without a remand to 
enable the trial court to reevaluate the evidence 
against the revised legal standards exhibited disregard 
for proper distribution of judicial functions in the 
federal judicial system. That distribution represents 
a deliberate judgment that trial courts are better 
equipped than appellate courts to evaluate evidence 
and to make factual findings. Hazelwood School Dis­
trict v. United States, supra, 53 L. Ed. 2d at 781. See, 
e.g., Uebersee Finanz-Korporation v. McGrath, 343 
U.S. 205, 212-13 (1952); Busey v. District of Columbia, 
319 U.S. 579 (1943).

The Fifth Circuit invaded the trial court’s domain 
not only by making de novo factual findings but also 
by restricting the trial court’s discretion to fashion

23

appropriate relief. Once a plaintiff has established a 
violation of Title V II, the selection of remedies and 
framing of decrees become the duty of the trial court 
which is vested with discretion to model its judg­
ment to fit the exigencies of the particular case. Dayton
Board of Education v. Brinkman, ____ U.S ___ 1 53
L. Ed. 2d 851, 862 (June 27, 1977); Albemarle Paper 
Co. v Moody, 422 U.S. 405, 421-22 (1975); Interna­
tional Salt Co. v. United States, 332 U.S. 392 (1947) 
See also United States v. Greater Buffalo Press Inc 
402 U.S. 549, 556-57 (1971).

Contrary to this fundamental principle, the Fifth 
Circuit on two separate occasions directed that plain­
tiffs are entitled”  to equitable relief,,<l on two sep­
arate occasions stated the District Court “ must”  issue 
particular injunctions or that an injunction is “ neces­
sary” ,17 on seven occasions directed that the District 
Court “ should”  issue particular injunctions,18 and on 
four separate occasions so strongly suggested, in in­
creasingly imperative terms, that the District Court 
grant certain relief as to afford the trial court no re­
alistic choice or room for discretion.19 Moreover, on one 
occasion the Fifth Circuit in effect issued its own in­
junction when it directly ordered Stockham to as-

16 App. 51 (Wonderlic test); App. 67 (age requirements).

. ”  App- 88 ( s e d a t e d  restrooms) ; App. 89 (selection and train- 
ing lor craft and supervisory positions).

18 App 89 (high school and age requirements for apprenticeship 
training); App. 90 (written guidelines for use by supervisors in 
selecting apprenticeship candidates; development of apprentice se­
lection procedures; development of guidelines for supervisor selec- 
tion; restructure of supervisor selection process; recruitment at 
black schools; posting job vacancies and qualifications).

19 App. 88-90 (segregation of facilities); App. 91-93 (type of 
seniority relief); App. 93-96 (backpay); App. 96-97 (front pay).



24

certain and publicize objective qualifications for ap­
prenticeships. (App. 90).

The task of a court of appeals is limited:
I f  [the Court of Appeals] concludes that the find­
ings of the District Court are clearly erroneous, it 
may reverse them under Fed. Rules Civ. Proc. 
52(a). I f  it decides that the District Court has 
misapprehended the law, it may accept that Court’s 
findings of fact hut reverse its judgment because 
of le°-al errors. Dayton Board of Education v.
Brinkman,------U .S .------- , 53 L. Ed. 2d 851, 862
(June 27, 1977).

The Fifth Circuit neither undertook nor understood 
either function. While giving lip service to the 
clearly erroneous rule, it reevaluated evidence de novo 
and substituted its intuitive judgment for that of the 
trial court. After overturning legal standards, it de­
veloped its own findings of fact to which it applied 
the revised legal standards. Though it remanded the 
case to the District Court for “ framing” of relief, that 
task seemingly would be ministerial rather than dis­
cretionary because of the Court of Appeals ligid di­
rections.

In the context of unequivocal usurpation of trial 
court functions, the decision herein of the United 
States Court of Appeals for the Fifth Circuit pre­
sents the clearest kind of case for the application of 
the supervisory powers of this Court in the interest of 
preserving orderly and efficient operation of the fed­
eral judiciary. By granting this petition, this Court 
can eliminate needless anneals with, duplicitous re­
views of fact, minimize incorrect appellate disposi­
tions, and definitively clarify the role of trial courts 
in evaluating statistics and other evidence.

25

3' m^n?nS ? - ° f Q “ an,“ aUve Differences In Educalional Attain- 
ment. u nXverSdJy Recognized To Influence Earnings, Cannot

Employees ComP ^ 9  Earnings of White and Black

Job applicants of whatever race bring with them to 
their employment cognitive skills and other 
mtics over which t ~ '  employer has no control and 
which may have nothing whatever to do with nn 
pjover s job i B M f c  for selection but neverthe-
• j— iT  ‘  dj ect an<* mwirnlilf imijS Z t E T r
cTuTiT^ 1 g, \ mamifacturiug plant such pro­
ductivity characteristics may wholly explain differ­
ences m earnings between black employees and white 
employees winch might otherwise erroneously be at- 
Fonllfed raCiaIly discriminatory policies. This 
taneeVf1 othergettings, has acknowledged the irnnor- 
tance o f one of these factors, education, as an influ- 
ence in all aspects of life. In Brown v. Board of Edi

luffed  7 U‘S- 483’ 493 (1954)’ this Court c o t

h Vi i  \S a p,rinciPal instrument in awaken- ng the child to cultural values, and in preparing 
him for later professional training and in helm 
mg him to adjust normally to his’ environment 
In these days, it is doubtfiil that a i ^ X d ^ y  
reasonably be expected to succeed in fife if  he is 
denied the opportunity of an education

Using a multiple regression statistical technique 
similar to that utilized by plaintiffs in Wade v Mis-

m x  Cir S T l T  ? Xtension 528 F.2d 508
• f C ^ 191 6 ’̂ Dr* James Gwartney, a labor econo- 

mist on the faculty of Florida State University, iden-
V L  i10SC f,actors suscePtible to measurement which 

affected employee earnings at Stockham and reached



26

the conclusion that there was no statistically signifi­
cant difference between the compensation received by 
white and black employees with similar productivity 
characteristics. The trial court relied on Dr. Gwart- 
ney’s study which was not challenged by opposing ex­
pert testimony and concluded that Stoclcham provided 
equal earnings opportunities to both races in its work 
force.

The Fifth Circuit, while not rejecting the validity 
of the analytical technique, dismissed Dr. Gwartney’s 
study as “ factually inadequate”  and criticized his 
choice of productivity determinants. Among the fac­
tors rejected, the Court of Appeals concluded that Dr. 
Gwartney improperly considered educational attain­
ment in assessing productivity. In cavalierly dismiss­
ing the Gwartney study demonstrating the impact of 
productivity factors on earnings at Stockham, the 
Fifth Circuit Court of Appeals committed the same 
error as did the Eighth Circuit in Hazelwood School 
District v. United States, supra, by totally disregard­
ing the possibility that prirna facie statistical proof 
can be rebutted at the trial level.

In making its intuitive determination, the Court of 
Appeals casually rejected a widely-observed and doc­
umented phenomenon that, on average, a higher edu­
cational level will yield higher earnings regardless of 
race. The Fifth Circuit discarded educational attain­
ment for two reasons:

The fallacy in this conclusion [i.e., that education 
impacts earnings] stems from two facts: ( 1 ) 
as the defendant concedes, education is not a job 
requirement at Stoclcham, and (2) white employ­
ees at Stockham have more education than blacks. 
Thus, adjusting for education in a regression an­

27

alysis of earnings where education is not related 
to job performance and where one race is more 
educationally disadvantaged than another, masks 
racial differences in earnings that may be ex­
plainable on the basis of discrimination. Certainly 
such differences cannot fairly be explained on the 
basis of a factor, such as education, concededly ir­
relevant to adequate job performance. (App. 41).

The Fifth Circuit’s error evolves from the sophistry 
of its argument: education is not a requirement for 
employment (or placement on a job) and, therefore, 
it is irrelevant” to adequate job performance. Logi­
cally, tlm conclusion does not necessarily follow from 
the premise. For example, an employer may not have 
a policy that dictates an “ acceptable” maximum level 
of employee absenteeism. Yet excessive absenteeism 
adversely affects productivity on the job without ref­
erence to the employer’s general policy requirements 
on absenteeism.

Embracing a “ warm body”  hypothesis, the Court of 
Appeals categorizes employees as either “ qualified”  
or “ not qualified.”  Within each group, in the Court’s 
view, employees are fungible. The Court’s analysis 
does not acknowledge that employees typically exhibit 
a spectrum of competencies, or the statistically-sup­
ported observation that education favorably influences 
earnings.

Although not articulated, the Court confused the 
mandate of Griggs v. Duke Poiver Co., 401 U.S. 424 
(1971), that employee selection devices have a demon­
strable relationship to job performance with widely- 
held economic principles that productivity factors 
have a statistically significant impact upon employee



28

earnings. The two concepts are logically independent 
and not inconsistent.

The economic literature, without important excep­
tion, acknowledges the premise that on average edu­
cational attainment favorably influences earnings. One 
authority observes:

The conviction that more education leads to 
higher income finds extensive support in statis­
tical data. The simple correlation between educa­
tional attainment is very strong and consistent: 
more years of education do lead to higher income. 
B. R. Schiller, The Economics of Poverty and 
Discrimination 98-99 (1973).

See also, e.g., L. C. Thurow, Poverty and Discrimina­
tion 70-71 (1969); W . L. Hansen, et al., “ Schooling 
and Earnings of Low Achievers” , 60 The American 
Economic Review 409 (1970). This is so largely be­
cause level of education serves as a proxy for a measure 
of cognitive skills and motivation attributes. See B. A. 
Weisbrod, “ Investing in Human Capital” , The Daily 
Economist 149 (1973). The fact that a specified educa­
tion level is not a job entrance requirement is not rele­
vant to productivity.80

The common-sense recognition that educational at­
tainment increases earnings in the workplace escaped 
the Court of Appeals, despite uncontradicted expert 
testimony in the record. I f  the Fifth Circuit s rejec­
tion of education as a matter of law as a factor in­
fluencing earnings is permitted to stand, the conse-

i° The Fifth Circuit’s opinion suggests that seniority, which 
“ favors”  blacks in Dr. Gwartney’s study, is a permissible produc­
tivity factor despite the fact that there are no seniority or tenure 
requirements for jobs at Stockham. (App. 40-42).

29

quences for future employment discrimination litiga­
tion will be dramatic. The Court, without anv regard 
for the community of professional knowledge, rejects 
weighing a factor which demonstrably impacts earn­
ings. By precedent, the Court of Appeals forecloses 
future litigants from exploring the productivity char­
acteristics of their work forces.

The Fifth Circuit’s approach is justified in a voter 
registration or jury selection case where voters, or 
jurors, are fungible as citizens. However, the ana­
lytical framework is illogical and inappropriate in the 
employment context. Economists conclude that many 
characteristics influence employment productivity; 
some characteristics may not be distributed in a ra­
cially balanced fashion in a particular employment 
environment. The Court of Appeals precluded, as a 
matter of law, consideration of productivity factors 
in general, and education in particular, as a possible 
explanation of earnings differentials.

The importance of this issue cannot be overstated. 
I f  the inflexible approach of the Court of Appeals 
survives, then employers will be required to compen­
sate employees according to race rather than produc­
tivity- That result has national implications which 
undermine this Court’s rationale in Griggs v. Duke 
Power Co., 401 U.S. 424 (1971), that employee quali­
fications must serve as the determinants of job success. 
The intervention of this Court is necessary to assure 
that differences evolving from objective productivity 
characteristics will not, in and of themselves, consti­
tute evidence of a violation of Title Y II  of the Civil 
Rights Act of 1964.



30

vn. CONCLUSION

For the foregoing reasons, the petition for a writ 
of certiorari should be granted.

Respectfully submitted,

D ouglas A rant 
John J. Coleman, Jr.
James P. Alexander 
W m . M ichael W arren, Jr.

1500 Brown-Marx Building 
Birmingham, Alabama 35203 
(205) 252-4500

Attorneys for Petitioner 
Stockham Valves and Fittings, Inc.

Of Counsel:
B radley, A rant, R ose & W hite 
L inda A. B unsey 
Robert K . Spotswood 

1500 Brown-Marx Building 
Birmingham, Alabama 35203 
(205) 252-4500