PRICE WATERHOUSE v. HOPKINS, 109 S.Ct. 1775 (1989) (p. 38) - MIXED MOTIVE CASE.  π worked for Δ for 5 years when she was proposed for partner.  Denied partnership b/c not feminine enough.  Good Motive: Interpersonal skills lacking, gruff, demanding.  Bad Motive: Treated her different because she was a woman; conduct tolerated in men was penalized in women.  Ct. of Appeals held that Δ could prevail by proving by clear and convincing evidence that it would have made the same decision without regard to gender.

Issue:  Is Title VII violated where both a discriminatory and legitimate motive lies behind an ENT decision?  YES.

Rule:  "Once a π in a Title VII case shows that a factor (gender) played a motivating part in an ENT decision, the Δ may avoid a finding of liability only by proving by a preponderance of the evidence that it would have made the same decision even if it had not allowed the factor (gender) to play such a role." 

O'Connor's concurrence wanted π to show that an impermissible category was a substantial motivating factor - this change abrogates that difference.  This decision was a plurality, so the answer is not entirely clear: a factor or a substantial factor.

Note:   However, the Civil Rights Act of 1991 addresses the issue:  new section 703(m) says that the statute is violated if race etc "was a motivating factor for any ENT practice, even though other factors also motivated the practice."  If so, a violation is established.

            HOWEVER, if the ER shows that "it would have taken the same action in the absence of the impermissible motivating factor" then the π is not entitled to damages or reinstatement; the court may only order Declaratory or Injunctive relief and attorneys fees directly related to the 703(m) claim. (see 706(g)(2)(B)).

            Basically, "a" motive violation may allow judicial relief rather than allowing ER to escape from liability.  Burden of proof is on ER to prove affirmative defense.  How?  Strong evidence of consistent behavior with respect to this conduct, e.g., we take affirmative steps to catch any cashier stealing money, not just this particular cashier; we always fire those discovered.

            CRA of 1991 makes punitive damages available - previously had not been.

Holding Continued:

!  Evidence in Title VII action that partners in accounting firm placed female partnership candidate "on hold" based on evaluations suggesting that she be required to take "a course at charm school," and dress more femininely and wear makeup, together with testimony of social psychologist indicating that decision was based on sexual stereotyping, was sufficient to establish that sexual stereotyping played part in decision to place candidacy on hold.

!  When plaintiff in Title VII case proves that her gender played motivating part in employment decision, defendant may avoid finding of liability only by proving by preponderance of the evidence that it would have made same decision even if it had not taken plaintiff's gender into account.  (Would - Not might have been)

!  ER Cannot avoid liability by alleging that LNR played a part in motivating ENT decision.  Must show would have so acted standing alone on that LNR.  On the other hand, if discrimination was a factor - not the sole factor - then ENT decision is illegal.

!  ER cannot avoid liability by alleging a LNR that did not motivate ER at the time of the ENT decision.

BOTTOM LINE:  where mixed motive is shown by π, burden shifts to Δ to PROVE that the same action would have been taken for the VALID motive.

How to advise your client:

1.  Don't make decisions based on stereotypes--mixed motive may subject you to declaratory judgment that you did wrong and possibly expenses.  (Less said the better)

2.  Remember that this change is a response to risk of underenforcement; lower threshold shifts burden to Δ to show by preponderance of the evidence that the same decision would have been made.  π gets more chips to bargain with in settlement process, making it more attractive to sue.

DISCUSS REMEDIES:  (nutshell p. 220 et seq.)

Curious things to note: 

Why does Title VII apply to a decision not to promote someone to partnership?:  Denial of partnership without more is not enough to constitute an "ENT decision."  But, associates are EEs.  If condition of being an associate is consideration for partnership, then it is a covered ENT decision. 

What is the appropriate remedy--could they award partnership?  YES -- award of partnership is within the equitable powers of the courts.

Is the Price Waterhouse case still entirely good law?  NO, it was overruled in part by the CRA of 1991.

What if  Δ said it fired male for failing to be macho--would that have worked?  NO--men would be required to be macho and women required not to be.

!          Note on Relief Available to Individual Victims of Title VII (p. 48) - governed by Section 706(g) (1).  Purposes are to make whole the victim of intentional discrimination.

            availability of backpay as damages:  backpay is generally available and should not be denied for reasons which, if applied generally, would not frustrate the intent to eradicate discrimination and make people whole.

                                    what π could have earned w/ no discrim.

                             -   what π did earn in other ENT             

                                    Measure of damages

Effect of ER good faith on availability of backpay:  ER good faith is not a sufficient reason to deny backpay because Title VII has a compensatory purpose.  Under Ford Motor Co. v. EEOC, 102 S.Ct. 3057 (1982), ER's unconditional offer of reinstatement terminates the back pay period (even if the offer does not include backpay or retroactive seniority).

            But if employer relies in good faith on a published opinion of the EEOC and turns out to be wrong, the CRA of 1991 says no liability.

Duty to mitigate ' 706(g):  (1) backpay damages should be reduced by earnings or amount one could have earned with reasonable diligence.  (2) ER has burden of showing π could have obtained comparable work.

Time limit on backpay:  2 years before the filing of the EEOC charge.

Optional award if job reinstatement is not a possibility:  front pay.

Types of damages available under Title VII:  equitable + compensatory + punitive.

Appropriate Remedies in Disparate Treatment:  (this is the page 48 stuff)

1.         Backpay is almost universal remedy . . . must mitigate damages.

2.         Prevailing  π, presumptively, gets atty's fees.

3.         Prevailing π ordinarily entitled to job and status; where impossible, additional monetary payments, "front pay," may be awarded.

4.         ER may be enjoined to take affirmative steps to cure pattern & practice.  ADDED by CRA of 1991:

5.         Entitled to jury trial.

6.         Award of compensatory damages and in some cases punitive damages.

            The CRA of 1991 allows compensatory damages for intentional discrimination, pecuniary damages for pain and suffering, mental anguish, etc., and punitive damages if intentional discrimination is with malice or reckless indifference, but these are only allowed under Title VII if cannot recover under ' 1981 and are capped relative to the number of employees.

C.  Proving Systemic Disparate Treatment - likely in P's prima facie case

            Ordinarily, an ER's policy of deliberate discrimination will  be systemic and it usually leaves a very persuasive statistical trail.  If judge finds a pattern or practice of DT there is an inference that the ER complied with the pattern or practice in every individual instance, and the burden shifts to the ER at the remedy stage.

            Since 1972, the EEOC has been authorized to bring suit against private ERs alleging pattern or practice discrimination.  Under President Carter, the procedure was changed so that the AG brings such suits against state and local governments.

!          Section 707(a) governs:

            "Whenever [there is] reasonable cause to believe that any person or group of persons is engaged in a pattern or practice of resistance to the full enjoyment of any of the rights secured by this subchapter, and that the pattern or practice is of such a nature and is intended to deny the full exercise of the rights herein described. . . . "

Cases:

TEAMSTERS v. U.S., 97 S.Ct. 1843 (1977) (p. 50)

The plaintiff demonstrated that in some geographical areas where blacks made up over 30% of the population, the ER had virtually no minority workers.  The D. Ct. and court of appeals each found that govt had proven a prima facie case.   Δ argued that statistics alone are not enough to show a practice of discrimination and that this was a result of pre-1964 discrimination.

Issue:  Had govt. proven a prima facie case?  (Raise inference of discrimination by stats):

Held:  yes.

1.  In order to prove a claim of systemic DT, the government has to prove more than sporadic acts of discrimination--it has to establish by a preponderance that discrimination was the company's standard operating procedure--rather than just the unusual practice.

2.  Statistics are generally admissible (and adequate) to prove systemic discrimination, but their usefulness depends on the surrounding circumstances.  Statistics can be rebutted.

Rationale:  (1) statistics here were backed up with evidence of actual discrimination.

(2) Δ's argument on the statistics would have merit except for the fact that Δ did a lot of hiring (hundreds over 4 years)--even though this total # of workers declined.  Most of new hires were predominantly white.

Why is this case here?  (1) to show that practices of systemic discrimination will leave a statistical trail.   (2) evidence met the standard in this case because it was statistics + individual instances of discrimination. (40 witnesses).  (3) Disparate Treatment (Question = Motive); Disparate Impact (Question = Consequence)   Teamsters was a DT case, so the ultimate question of fact was the motive of the ER in taking the adverse employment action.

How does an individual EE get anything in a systemic DT case:

(1) once the government proves a pattern or practice of discrimination, there is a presumption that π EE was affected by it.

(2)  burden then shifts to  Δ to prove that this EE was not affected by the pattern or practice.

(3)  How do we remedy is such a way as to make the victim whole w/o interfering with ER's business operation?

HAZELWOOD SCHOOL DISTRICT v. U.S., 97 S.Ct. 2736 (1977) (p. 55)  Statistical evidence case--used to try to prove systemic Disparate Treatment. (Suburban St. Louis)  Rule:  The relevance of statistics depends on the circumstances.  The smaller the percentage of minorities /protected class, the greater the inference of discrimination.  At what point does statistical evidence give rise to inference of disparate impact or treatment?  When the percent hired is significantly less than the percent of qualified applicants.  Here they looked at the entire labor market, not percent that applied.

            what was the Trial Ct.'s main error in this case:  using irrelevant statistical evidence of the comparison between the racial composition of students vs. teachers; the relevant comparison is to the relevant labor market.  what is the relevant labor market?  The group of people in the area in which people would travel to Δ's workplace.

            S. Ct:  compare applicant rate of Blacks to hiring rate (most probative) - but even that may be tainted if word of mouth discouragement.  Black teachers in County: 15%;  Black teachers in suburban market: 5.4%; Black teachers in Hazelwood: 3.7%.

            what was the ultimate question in both Teamsters and Hazelwood?: because they are both DT cases--the ultimate question was the defendant-ER's motive.

Ct of App. erred in determining the govt had proved PFC.

!  Evidence of nondiscriminatory hiring post act will not result in finding of liability.

!  Which statistics should the percentage of blacks hired be compared to, the percen­tage working in the County only, or the whole County including the City.  To deter­mine this court looked at several factors:

            A.        were the illegal practices in effect in the year the census figures were gathered

            B.        did those policies change the makeup of the staff from what it would otherwise have been

            C.        what, if any, impact did the hiring practices of the city have on the availability of black teacher applicants for the county.

            D.        to what extent did desire of teachers to work in city or suburbs effect applications

            E.        What were the experiences of the other county school districts and how does this relate to Hazelwood.

!  Ct remanded for consideration of these factors.

McKENNON v. NASHVILLE BANNER, 115 S.Ct. 879 (1995)  After-Acquired Evidence Case.  EE afraid she was going to be fired in violation of ADEA; collected evidence of this wrong doing.  EE was fired and she sued; during the subsequent litigation, the ER found out about this evidence and said they would have fired her anyway.  6th Cir. said that EE's misconduct was grounds for her termination and any Title VII violation was irrelevant:  no relief to McKennon.  SCt. reversed Unanimous Decision:  She was not barred from all relief.

Took case under 2 assumptions:  (1) McKennon fired for age discrimination; (2) Banner would have fired McKennon if it knew about after-acquired evidence at that time.

Rules:  1.  ADEA violation cannot be altogether disregarded b/c that would frustrate the statute's goals. 

2.  Proper calculation of back pay is from the date of the unlawful discharge to the date the after-acquired information [of EE wrongdoing] was discovered.  Also, to limit the award of backpay, ER must show that it would have──not could have── terminated EE for such misconduct had it been aware of it.

3.  Public policy advanced every time P wins in a discrimination case (system must induce meritorious lawsuits).  (Pat says public policy advanced every time justice is served - not just if P wins).

Comment:  The ADEA claim has 2 key aspects:  (1) π pursues own case; (2) Every π in a status discrim. action is acting as a private AG vindicating private and public policy behind the statute.

!          Outcome:  ERs will undertake to scrutinize ENT history of EEs who make discrim. claims ag. them (investigate for misconduct).  It is a violation of '703/704 to ask if EE has even filed claim ag. other ERs during the interview process.  

!          Conflict between Impact and Treatment is that Title VII requires the employer to look at each applicant or employee minus any stereotypes.  Impact may be viewed as past wrongs.  The conflict is not completely resolved.

D.  Proving Unjustified Disparate Impact (DI)

Basic Theory and Method of Proof:  Whereas in a DT case, the main Q is the ER's motive, in a  DI case, the main Q is the consequence of a facially neutral ENT practice.

GENERAL RULE:  Title VII prohibits ENT practices which have a DI on protected classes, unless the ER can prove business necessity of the ENT practice.

GENERAL RULES--shifting burdens in a DI case:

1.  π need only show that a "particular ENT practice" produces a significantly DI on the basis of race, color, religion, sex or national origin.  Ward's Cove; CRA 1991 ' 105.

2.  Δ must then show that the practice has a manifest relationship to job performance.  CRA 1991 ' 105.

3.  if Δ carries that burden, π may still win if he proves that:

            a. the ER could achieve its interests with an alternative practice which produced less disparity, and,

            b. the ER refused to adopt such alternative ENT practice.  CRA 1991 ' 105.

Cases:

GRIGGS v. DUKE POWER, 91 S.Ct. 849 (1971) (p. 74)  Δ-Duke Power employed people in 5 job categories, the lowest being Laborer, the next to lowest being Coal Handler. Before 1965, (the effective date of the CRA), blacks were employed only in the laborer category, and the highest paid job in this category paid less than the white categories.  In 1955, Δ began to require a high school diploma for hiring or transfer to any dept. above labor.  In July 1965, blacks were eligible to be hired or transferred to any of the depts.  Also in 1965, Δ required applicants for the 4 higher categories to pass 2 aptitude tests.  These requirements were applied non-discriminatorily to blacks and whites alike.

!          Trial Court found no evidence of discriminatory purpose (conceded by P) and held that the requirements were lawful under Title VII.

Held:  Reversed:  Unanimous decision to reject the ENT practice because the EFFECT of the tests was to discriminate.  MAJOR DEVELOPMENT:  Title VII prohibits (facially neutral), unintentional discrimination.

Disparate Treatment analysis - There were levels within each job category and promotion was from within each category.  Job progression lines went from the top of one category to the bottom of the next. This might require a pay cut and loss of seniority.  Disparate Treatment requires that there be bad motive.  This system was OK because there was no Bad motive.

Disparate Impact Analysis  -  The critical criteria in administering tests are (1) are they relevant to the job to be performed, (2) do they discriminate in effect even if not in intent.  "Built-in headwinds" for minority groups.

!          evidence that non-high school graduates held the higher category jobs and worked just fine:  no need for high school graduation requirement.  In fact a significant number of the non-graduate employees in these jobs were promoted at about the same rate as those who had completed the diploma.

!          "Any requirement must have a manifest relationship to the ENT in question."  (Compare with Ward's Cove:  "significant service to legitimate goals.")

!          "If an ENT practice which operates (1) to exclude [a protected group disproportionately] cannot be shown to (2) [business] relate to job performance, the practice is prohibited."  Test that has the effect of classifying EEs by race, gender etc. must be job-related.  Screening device must be job-related.

!          Once ER shows job relatedness, the π may then show the availability of a less discrim. means of serving the ER's legitimate interest.

What do we gain:  expanded employment opportunities for minorities, costs not very high.      What do we lose:  costs of changing the system.

CONNECTICUT v. TEAL, 102 S.Ct. 2525 (1982) (p. 78)  (EEOC 80% Rule Case)  In order to determine who would be considered for promotion to supervisor, Δ ER required each EE to pass a written test. (Apparently, a score of 80% was required for passing the test).  65% of the total passed the test.  54.17% of blacks passed the test, while 79.54% of whites passed.  This correlated to a 68% passing rates for each black per each passing white.  ER argued that although a particular test or requirement cut-off may have a discrim. effect, the ER compensated for that effect by hiring or promoting EEs in the protected class in number sufficient to offset the Disparate Impact (DI).  SCt. rejects this argument:

Held:  Unlawful discrimination can occur without discrim. intent.

1.  Title VII is aimed at protecting every individual's ENT opportunities--not just the group in abstract.  In the final selection, a significant number of blacks were promoted.  Court observes that the statute was designed to protect the individual from discrimination; in other words, the fact that other blacks have not been discriminated against has nothing to do with the fact that, in a specific individual case, a person has been denied an ENT opportunity because of race.

            It was a limitation or classification which deprived individuals of the Opportunity.

            The statute is designed to protect the individual not the group as a whole.  Here the group did alright but the individuals were discriminated against. Title VII says no discrimination against individual, the test did.  The negative disparate impact was not relevant.

2.  A non-job-related test that has a DI, and is used to limit or classify EEs, is used to discriminate within the meaning of Title VII whether or not the ER so intended and whether or not the ER later compensated for the discriminating effect.

            The bottom line in percentages promoted is not the dispositive factor: the bottom line does not prevent Ps from establishing a pfc, nor does it provide D with a defense.  Those who would assert the bottom line provides a defense are confusing unlawful discrimination with discriminatory intent.  A racially balanced workforce cannot immunize the ER from liability for specific acts of discrimination.  The qualifying test remained discriminatory.

Counter Argument:  This rule means that statistical info works only one way: a π can use it to show discrimination, but a Δ/ER cannot use it in defense.  That would be unfair.  This case answers that intent is not necessary for discrimination.

            Implies defensive statistical info is relevant in a DI case; quotes Furnco Construction Corp. v. Waters (1978):  "Proof that [a] workforce was racially balanced or that it contained a disproportionately high % of minority EEs is not wholly irrelevant on the issue of intent when that issue is yet to be decided."

Dissent:  argues that a DI claim must be judged by the "impact of the entire process on the group as a whole."  Worries that employers (1) will hire by the numbers (quotas), and (2) will continue to use the test but hide it and not use a magic cutoff number.

FURNCO CONSTR. CORP. v. WATERS, 98 S.Ct. 2943 (1978):  ER wants only brick layers who have experience or whom he knows will do quality work.  (Whites had worked for him before).  U.S. Supreme Court reversed lower court finding ag. ER and remanded case for determination of ER's motive.  (Implies ER not motivated by race). 

Held:  "Proof that [a] workforce was racially balanced or that it contained a disproportionately high % of minority EEs is not wholly irrelevant on the issue of intent when that issue is yet to be decided."

Evidence of racial balance in the workforce is not sufficient to disprove discrimination, but it may be considered by the trial court as evidence of the employer's motive.

COMMENT:  Can this practice be justified as a business necessity?  If not, the ER has incentive to abandon this practice and pay attention to the racial composition of its workforce.  Powerful anomaly:  existence of DI liability implies the ER should be race conscious (he has a monetary incentive to be).

DOTHARD v. RAWLINSON, 97 S.Ct. 2720 (1977) (p.  85)  ER (State of Alabama) adopted requirement that had a substantial foundation:  height (5'2") and weight (120 lbs) minimums for prison guards.  Problem was that this requirement would screen out people who would be good prison guards.  (1/3 of women and 2-3% of men screened - disproportionate impact on women).  Court allowed Ps to use statistics on national averages - soft numbers (not skewed to Alabama labor market).  Nature of rule would have limited applicant pool - so court excused Ps for not going to applicant files for evidence.

Issue:  did π have a valid DI claim?  YES

Rules:   Procedure for proof as follows: THIS IS THE GRIGGS TEST

1. To make a pfc of DI, π need only show that the facially neutral standards have a discriminatory effect = result in disparate hiring.

2. Employer must then meet the burden of showing that the challenged requirement bears a "manifest relationship" to job performance

3. If so, the π must then show that there are other non-discriminatory ways to achieve same end.  (Other selection devices without a similar discriminatory effect would also serve "the ERs legitimate interest in efficient and trustworthy workmanship.")

Rationale:  1. Δ offered no evidence that size corresponded to needed strength.

2. Δ should have had a strength test if this was its real concern; the height/weight criteria was too blunt an instrument to achieve the objective.  3. Strength may not be the most important attribute of a corrections officer -- probably can't be justified in a business sense.  Congress never meant that the "business necessity" be an essential job function.  Just that it be reasonable, served a purpose and was justifiable.

NYC TRANSIT AUTHORITY v. BEAZER, 99 S.Ct. 1355 (1979) (p. 89) (Limits on Stats.  P did not have quite enough evidence to prove Disparate Impact (DI))  Regs forbidding drug users from working in dangerous positions upheld even though the users were on methadone.Case was not argued with enough evidence to show that the rule had Disparate Impact (DI) on racial group as opposed to methadone users.  They had the statistics on both sides but did not establish causation.  Also there were significant safety concerns.  Stats failed to establish a PFC for DI.

π attacked use of ENT test used by Δ for ALL applicants regarding methadone use.  Pass/Fail threshold test.  π uses statistics to show that approx. 65% of methadone treatment patients whose race was known were black or hispanic.  THUS, black/ Hispanic were enrolled in these programs at approx. 2x their make-up in general population.

Issue:  Had π proved a Title VII claim?    Held:   NO.

Rules:  1. Stats that show that an ENT practice has the effect of denying the members of one race equal access to ENT opportunities may establish a prima facie violation of Title VII.

2. Even if a pfc is established, Δ prevails if the practice proves to be job related.

Rationale:  1. πs here charge that the practice of weeding out methadone users but presented limited statistics on the racial composition of methadone users who seek private treatment.

2. π presented no evidence of racial composition of methadone-using applicants (only 5% of applicants were weeded out by this test).

3. the lack of evidence on privately treated methadone users leaves open the possibility that the total methadone using population is not disproportionately black and Hispanic.

4.  With safety concerns, Δ also proved that the requirement was job related.

A π need not point to a particular criteria when alleging DI if the hiring or evaluation process as a whole has a DI.

WATSON v. FT. WORTH BANK and TRUST, 108 S.Ct. 2777 (1988) (p.94) Black woman bank teller passed over for promotions four times in favor of whites.  A few things had been said to π, such as "this job involves what would be a whole lot of counting for a Black."  The promotion evaluation practice was purely a subjective evaluation.  Disparate Treatment Claim dismissed.  But appeals court agrees to apply Disparate Impact Analysis.

Issue:  Does DI analysis apply to discretionary and subjective ENT practices?  YES.

Rules/policies:  1. The necessary premise of the DI approach is that some ENT practices, adopted without a deliberately discriminatory motive, may in operation be functionally equivalent to intentional discrimination. 

2. Griggs could largely be nullified if DI analysis were applied only to standardized selection practices.   

3. Thus, subjective or discretionary ENT practices may be analyzed under the DI approach in appropriate cases.   Burden on the Defendant/ER in DI case to rebut pfc, but ER can rebut simply by producing SOME evidence of a legitimate reason.  Then EE must prove by preponderance of evidence that proffered reason is pretext.

4. Delegation of promotion responsibilities to supervisors does not relieve the employer of responsibility.   

Can a π lose the Disparate Treatment (DT) case (= e.g., the race factor is causative) and still make out a pfc of Disparate Impact (DI)?:  YES, because even if subjective factors were not race, they still could have been factors which would have had a DI on a certain race or gender.

Possible Anomaly:  In the pre-Watson era, if the trier of fact was in equipoise on both the DI and DT claims, then P would lose on the DT case and D would lose on the DI case.  O'Connor Plurality Opinion in Watson:  "Ultimate burden of proving discrimin­ation against protected group caused by specific ENT practice remains at all times with the Title VII Plaintiff."

How does Title VII mandate a sacrifice on the part of the ER in order to effectuate its policy?   ERs may be forced to give up their discretionary systems in favor of a more structured, but less discriminatory, ENT practice.

What would be needed for π to have enough for a pfc of DI?  Evidence that the same thing happened to other Black EEs.

WARDS COVE PACKING CO., INC. v. ATONIO, 109 S.Ct. 2115 (1989) (p. 98) (began in 1974, decided 1989 with remand)  Salmon packing plant in Alaska.  Δ/ER had 2 kinds of seasonal jobs:  plant workers/cannery jobs, which were unskilled, held mostly by Indians (low pay); and, administrative/noncannery jobs, which were mostly skilled (higher pay), and filled mostly by Whites.  Cannery and noncannery job workers had separate eating and sleeping quarters.  There were 5 different practices which kept the cannery jobs for the Indians:  1) nepotism;  2) rehiring preferences, 3) separate hiring channels, one union for the cannery, another for the non-cannery, 4) no in-house promotion, and 5) lack of objective hiring practices.  9th Cir. en banc found that Ps created a pfc of DI discrimination relying solely on statistics showing a higher % of nonwhite workers in cannery positions and a low % of nonwhite workers in noncannery positions.  Δ employs in these positions only seasonally.

Issue:  Do statistics showing a high percentage of white workers in noncannery jobs and a high percentage of minorities in cannery jobs establish a pfc for DI?  NO - unanimous 9 justices on this point.  NO Disparate Impact.

Rule:   1.  The relevant comparison is between the racial composition of employed v. racial composition of relevant labor market (applicant pool).  Statistics were inadequate - did not look at qualified applicants.  Each job was hired from specific pool, numbers were OK from that view.

2.  Where such numbers are hard to ascertain, the composition of otherwise qualified applicants is relevant.

Rationale:  1. comparison of mostly unskilled cannery workers was not representative of those qualified for generally  skilled noncannery positions.  The 9th Cir. analysis would mean that any ER who had a segment of work force that was imbalanced could be hauled into court to give a "business necessity" for the imbalance.

2. if the disparity is due to lack of qualified nonwhite workers, it is not Δ's fault.

What other things did this case settle for DI cases?:  (1) π will have to show that disparity results from a specific ENT practice:  the EE must identify the device/test and link it to DI.  (2) Δ then will have the burden of production (only must offer some evidence) to show job related reason for the ENT practice.  (3) then π has the burden of persuasion to prove not job related.

How did this case set a different standard than in Griggs:  In Griggs, the court said that the Δ had to establish "business necessity" in order to justify a practice which had a discriminatory impact; Ward's Cove made it sound more like a rational basis test:  that practice was ok if it rationally could be viewed as being job related ("business purpose").  RATIONALE:  get ERs away from hiring race/gender just to avoid DI liability.  ********Key for Pat Hardin ************  (Congress Overruled)

What did Ward's Cove add which is relevant to the coverage of subjective ENT decisions?  Case said that P must identify the particular ENT practice that leads to DI.

How does Ward's Cove square with Watson:  It is very hard to reconcile with Watson. 2 ways to read Ward's Cove:  (1) it wiped out Watson; (2) it left open the possibility that you can just identify the subjective practice as a whole and that would be sufficient.

The main purpose of the CRA of 1991 was to establish the procedures by which a DI claim is judged.  The Act amended ' 703, adding (k)(1)(A):

            An unlawful ENT practice based on DI is established under this title only if--

            (i) a complaining party demonstrates that a respondent uses a particular ENT practice that causes a DI on the basis of race, color, religion, sex or national origin and the respondent fails to demonstrate that the challenged practice is job-related for the position in question and consistent with business necessity OR

            (ii) the complaining party makes the demonstration . . . with respect to an alternative ENT practice and the respondent refuses to adopt such alternative ENT practice.

Also added was ' 701(m):  The term "demonstrates" means meets the burden of production and persuasion.

"The `alternative ENT practice' provision contemplates that even if an ER proves business necessity, the π still has the opportunity to show that the ER's legitimate interest could be served by less discriminatory means.  Not only does the π bear both the burden of production and persuasion on this point, he must show that the ER refused to adopt the alternative measure."  Albemarle

Effect of CRA of 1991 on Ward's Cove:  CRA of '91 expressly overruled Ward's Cove weakened business purpose standard for ENT practice by putting even the burden of persuasion on the ER to show the Griggs standard of business necessity.

Why do we put a heavier burden on the ER in a DI case than in a DT case?:  With DT, it is hard to disprove a state of mind.  But, with DI, it is much easier for an ER to prove the job relatedness of its ENT practice.

Legislative responses to Ward's Cove:  Congress reinstituted pressure on ER to make race-conscious decisions in hiring to avoid DI liability.

1.  Burden shifted (burden of persuasion) back to ER to show business necessity and job related for the position in question.

2.  Business necessity … job related; definition now locked in is Griggs (business necessity) explicitly, and explicitly NOT Ward's Cove.

3.  Legislative history confined to single memorandum - wanted pre-Ward's Cove status quo to prevail.

4.  π shows impact of each particular challenged ENT practice unless he can demonstrate they are not capable of separation for analysis in which case they can be evaluated as one ENT practice. (Back to Watson)  EE must link each practice to DI.

5.  3d step:  P has burden of production and persuasion to show that ER could achieve same effect by less discriminatory means is now codified.  (note:  π who can do this is close to showing that ER DOES intend discriminatory outcome_ boundary of DT.

Political Irony to Ward's Cove and the Congressional reaction:  It is reasonably clear the Ward's Cove (& Hicks and others) were motivated partly by perception by 5 or 6 justices that the DI law  as it existed was an invitation to ERs to engage in taking race/gender motivated material into account in hiring--EDI law was inconsistent with DT law.

!          THUS, these cases show intent to decrease the impact of DI law--not completely nullify it, but certainly diminish it.

!          The outrageous case may be easy, but mostly these cases make it harder for the π to win.

!          STATUTORY law is more favorable to πs as a result (irony is that SCt.'s prescription to help Δs produced a legislative prescription to help πs).  

Read notes up to Note 5 following Ward's Cove regarding major changes.

PROCEDURE for DI Case: 

1.  CRA 1991 ' 105 (p.29 of Supp) P must identify a specific ENT practice which leads to DI (Ward's Cove).  (P must use particularity)

2.  Once there is a pfc, ER bears burden of production and persuasion to show the ENT practice is (1) job related for the position in question, and (2) a business necessity.

            Note:  Congress explicitly rejected the Ward's Cove view of business purpose (i.e., "significant service to legitimate goals") and froze the state of the law back to where it was prior to the Ward's Cove decision (business necessity).

3.  Remedies - Disparate impact π who wins gets injunction cease and desist.  Back pay may be included if individuals can show they were fired or did not get jobs.

Establishing BUSINESS NECESSITY (BN):

!          The key to business necessity is job relatedness.  Any test or criterion must measure the capacity of prospective EE to do the job.

!          Even though an ENT test may not be related directly to job performance, it may sufficiently pass the standard of business necessity if it is linked to ability to perform in a training program. Washington v. Davis

!          The more complex a job is, the more deferential the courts will be in deferring to the ER's judgment as to an ENT test.  Compare Wash. v. Davis and U.S. v. South Carolina with Duke Power v. Griggs.

!          Adopting EEOC guidelines, the SCt. said "discriminatory tests are impermissible unless shown by professionally acceptable methods, to be 'predictive of or significantly correlated with important elements of work behavior which comprise or are relevant to the job or jobs for which candidates are being evaluated.'" Albemarle Paper Co. (quoting 29 CFR 1607.4(c)).

Cases

ALBEMARLE PAPER CO. V. MOODY, 95 S.Ct. 2362 (1975) (p. 111) Like the ER in Griggs, Δ here used 2 tests to gauge nonverbal (Revised Beta Exam) and verbal (Wonderlic Personnel Exam) intelligence for ENT.  Prior to 1965, the paper mill had racially segregated the "lines of progression," reserving the higher paying and more skilled lines for Whites.  The "racial identifiability" of whole lines of progression persisted until 1968, when the lines were reorganized under a new collective-bargaining agreement.  This reorganization left Negro employees "locked' in the lower paying job classifications.' (Job-related lines of progression) The formerly 'Negro' lines of progression had been merely tacked on to the bottom of the formerly 'white' lines, and promotions, demotions, and layoffs continued to be governed--where skills were "relatively equal"--by a system of "job seniority."  Because of the plant's previous history of overt segregation, only whites had seniority in the higher job categories.  When it quit openly discriminating, Δ required a high school diploma and the passing of two tests for the skilled lines of progression.  Δ tried statistically and through experts to validate the job relatedness of these tests.  Of the 105 EEs within the groups, only 4 were black.  To rate the ENT performance in the study, supervisors were asked subjective questions as to which person they thought was doing a better job; conclusion was that there was a correlation between scores on tests and good evaluations by supervisors.

            Albemarle hired an industrial psychologist to study the 'job relatedness' of its testing program.  His study compared the test scores of current employees with supervisorial judgments of their competence in ten job groupings selected from the middle or top of the plant's skilled lines of progression.  The study showed a statistically significant correlation with supervisorial ratings in three job groupings for the Beta Test, in seven job groupings for either Form A or Form B of the Wonderlic Test, and in two job groupings for the required battery of both the Beta and the Wonderlic Tests.

Note:  criterion validation:  finding test that predicts that people will be good at job validated by showing that people doing well at job will do well on test; Test predicts good job performance = Validation.)  If employer can show that the test score has statistical relationship to job performance, then it is OK.  What is in the test is not relevant. 

Pitfalls of this criteria testing:

1.  Supervisorial rankings were very subjective; no specific categories for evaluation.

2.  Those evaluated and tested were EEs who had stayed w/ company and were near the top of the line.  You can't validate by the success of those who already are on the job.  New test takers would not have this experience, so comparison was not relevant. 

3.  Test was validated for jobs other than the ones being tested for. 

4.  Hiring is done at entry level.  These evaluations were not good indicators for the success of an applicant for the lower-level jobs.

Issue:  Are these tests a business necessity (sufficiently job related)?   NO

Rules:  1. The job-relatedness of a test for a given job is not supported by a showing of test's relevance for other jobs unless there is "no significant difference between studied and unstudied jobs."

2.  Criteria used for job evaluations in these studies must represent major or critical work behaviors as revealed by careful job analyses.

3.  ER should only test for higher level job skills if applicants, in a reasonable period of time, applicants will probably be considered for the higher positions.  Otherwise, the testing should relate to entry level jobs.

4. Where feasible, the validating data should include minority and nonminority groups.

5.  ER should have tested whether those with low scores would nevertheless make good entry-level workers.

Rationale:

(1) Only more skilled jobs were tested here.  (2) 3 of 10 validation is not enough anyway.  (3) Vague standard for level of worker's quality is not related enough--probably applied inconsistently.  (4) Test might not be applicable to positions not likely to be leading to quick promotion.  (5) Too few Blacks tested here.

BOTTOM LINE:  Company's attempt to validate its test requirements was insufficient, among other things, because of heavy reliance on the use of subjective supervisor evaluations and the lack of neutral on-site oversight.

WASHINGTON v. DAVIS, 96 S.Ct. 2040 (1976) (p. 118)  Not really a Title VII case, but Court used the same standards.  No such thing as DI under the 14th Amendment = to violate 14th Amendment must show race/gender motive.  Supreme Court upheld a literacy test with a cutoff score administered to police cadets despite discriminatory effect, because the test predicted success in the cadet training program.  DCt. found that π's evidence warranted conclusions that (a) the # of Black officers in the DC PD was disproportionately low as compared to the DC population; (b) a higher percentage of blacks failed Test 21 than whites; and, (c) Test 21 was not well linked to job performance.

            TEST -------E TRAINING PROGRAM --------E JOB PERFORMANCE

!  valid b/c success on test correlates to success in training program (content validity study).  No evidence that success in the training program leads to success on the job.

Issue:  Did use of this test violate 14th Amendment (or Title VII standards)?  NO.

Rule  Although test may not relate to actual job performance, it is valid because it relates to the training program--a minimal level of reading ability is needed to do well in the training program.  It is not necessary to show that the test or the training program itself be connected to success on the job.  Success in the training program is enough.  [Arguably, this case involved some amount of deference to the police dept.  It should not be illegal to require that police cadets be able to read.]

Problem:  Test screens for only one criterion, when we know multiple criteria are important for success on the job.

COMMENT:  This case is more famous for its holding that a claim of discrimination under the 14th Am. requires specific intent to discriminate:  "racial animus."  There is no Disparate Impact violation of the Equal Protection Clause.  Must sue under a Disparate Treatment theory for an Equal Protection violation.

U.S. v. SOUTH CAROLINA, 445 F.Supp. 1094 (1977) (p. 123)  Decision of a 3-judge district court.  South Carolina teachers were required to pass a test after completion of training program to get a job.  Thus, the test looks backwards and asks if the Trainees learned what the training program taught, rather than looking forward to see if they would make good teachers (job skills).  Yet it is an acceptable means to determine teacher qualification.  The link between the training program and the job is assumed ("sufficiently closely related to the curriculum . . . to be an appropriate measure of achievement with respect to that curriculum.")  Again there was a content validity study to show correlation between the test and the success of the training program.  3-judge Ct. (Haynsworth) cited Davis as support for the decision to validate the test against the training program rather than subsequent job performance.  S.Ct. denied cert.

            TRAINING PROGRAM -----E TEST --------> JOB PERFORMANCE

(test is pre-requisite to get out of program)--Pat: rather draconian--make or break

NEA v. SOUTH CAROLINA, 98 S.Ct. 756 (1978) (p. 125)  Byron White's Dissent from a summary affirmance of the 3-judge district court holding in the case above.  πs attack validity of test.  Δs justify it b/c it adequately measures success in the training program.  (SCt. affirmed D.Ct.)  What is wrong with this approach?  Despite efficiency, the test was both over and under inclusive:  some very good teachers are rejected because they don't pass test and some who are bad teachers end up passing.  Does this make it unconstitutional?  S.Ct. said NO:  w/o proof of bad motive, the efficiency of the test is reasonable enough to uphold it.  (FLAW:  fails to draw inference of bad motive).

(Imperfect analogy:  Bar exam - not so draconian, tests actual problem-solving ability)

These cases together say:  We're not going to require an impossible level of technical justification; allows Δs to win if they can show real/reasonable level of justification (implies no bad motive).  South Carolina involved the use of content validation which is based not on the test's ability to predict job performance, but rather on the fact that it approximates the content of the job, by including a representative sample of important areas of job knowledge or required skills. 

Systemic Disparate Treatment requires π to show BAD MOTIVE.  But here that could (should) have been inferred from the numbers.

See Note 5, p. 133.  (discussion of different kinds of test validation methods, pp 128-142.)

Differential Validity:  Std. Test:  Score of 50 (by %) = 100% will be satisfact. EEs

Score of 45 (by &) = 100% will be satisfact. EEs.  But with score of 44, 95% of females are good employees.  Can you set different scores for genders when using this test?  Pre-CRA of 1991 answer:  Yes (you must), relates to performance.

(race/gender)

Norming:  Sets rates such that 75% of Whites and 75% of Blacks pass.  (W=score of 50, B = Score of 48) (adjustment after the fact to ensure that 75% passed)  This was disallowed.  Too related to race and not performance.  Congress doesn't like this and the CRA of 1991 makes it illegal to adjust scores (See p. 31 in Supp.)

1.  Title VII disfavors minimum cutoff scores and drives employers away from "magic number"

2.  Any category (human relations, skills, literacy, aptitude, rules) with DI, must be justified.

3.  Justification becomes easier as the skill level increases for the job.  The more complex the job the more difficult a content test would be.   Assessment becomes more subjective.  More and more away from magic number.

E.  The Regulation of Status Discrimination by Unions:

!          Section 703(c) of Title VII proscribes the same 5 types of status discrimination for unions as section 703(a) does for ERs.

!          This proscription covers both the representation of EEs and in the conduct of internal Union affairs.

Cases

GOODMAN v. LUKENS STEEL CO., 107 S.Ct. 2617 (1987) (p. 144)  Allegation was that U refused to process grievances that the ER was engaging in racially discriminatory practices.  Union defense:  we address most grievances on non-racial grounds, the rest were from probationary EEs.  We don't have enough resources to process all the claims.

Issue:  did the Union violate Title VII and Section 1981 by refusing to file grievances on behalf of EEs when the ER is  known to be discriminating in violation of the Collective Bargaining Agreement (CBA)?

Held:  YES.  Union intentionally classified grievances on racial grounds and avoided processing them, either so as not to antagonize ER and achieve success with other claims or in deference to wishes of White members, violated Title VII.  (See language on p. 145-146).