Cherry v. Amoco Oil Co., 490 F. Supp. 1026 (N.D. Ga. 1980)

*1027 Lawrence L. Aiken, Atlanta, Ga., for plaintiff.

Earle B. May, Jr., Judson Graves, Jones, Bird & Howell, Atlanta, Ga., for defendant.

ORINDA DALE EVANS, District Judge.

The above-captioned case was tried to the Court sitting without a jury on April 29-30, 1980. Having considered the evidence adduced at the trial, together with the briefs and arguments of counsel, the Court finds in favor of Defendant Amoco Oil Company. Specifically, the Court finds and concludes as follows:^[1]

Plaintiff is a white woman who resides in a predominately non-white residential area in Atlanta, Georgia. She applied for but was denied a gasoline credit card by Defendant Amoco Oil Company. She seeks damages^[2] under the Equal Credit Opportunity Act,^[3] which provides in part that it is unlawful for a creditor to discriminate against a credit applicant on the basis of race. She contends the reason for rejection of her application was race-related and that as such, said denial is proscribed by the Act.

Defendant Amoco Oil utilizes a complex computerized system to evaluate credit card applications. The system takes into account 38 objective factors in scoring each application, including level of income, occupation, and Amoco's prior credit experience *1028 in the U. S. Postal Service zip code area where the applicant resides. The information provided with respect to each of the 38 factors is scored by the computer. Credit is granted to those who receive a passing aggregate score; it is automatically denied to others.

Amoco's testimony indicated that its scoring system assigns a low rating to those zip code areas in which it has had unfavorable delinquency experience. Of those Atlanta zip code areas bearing the prefix 303,^[4] low ratings were assigned to all predominantly non-white zip code areas^[5] as well as to many predominantly white areas. Plaintiff's zip code area, 30310, was assigned a rating of 1, the least desirable rating on a scale of 1 to 5.^[6] The evidence showed without dispute that if Plaintiff lived in a zip code area rated 3 or above, and all other information including level of income shown on her application had remained constant, she would have been granted a credit card by Amoco.

Plaintiff asserts that Amoco's utilization of zip code ratings is racially discriminatory. She contends she has standing^[7] to complain of such discrimination because she was adversely affected by it; to wit, she lives in a predominantly non-white zip code area which has been low-rated by Amoco's system.

Plaintiff produced two witnesses, herself and a sociologist. Plaintiff's testimony established that she received a letter from Amoco indicating her application had been turned down; the letter specified that one of the reasons was "our previous credit experience in your immediate geographical area."^[8] Plaintiff testified she was humiliated and embarrassed by rejection of her credit card application. She further testified she subsequently has had to reveal this denial on other credit applications, but there was no testimony that she has been denied other credit. In short, Plaintiff presented no testimony of actual damages, unless the statutory term "actual damages" is to be interpreted by the Court as permitting an award of damages for mental anguish.

Plaintiff's other witness testified he had received from Plaintiff's counsel a list of those Atlanta zip code areas bearing a 303 prefix which were rated 1 or 2 by Amoco, along with information as to credit application acceptance rate in each area. He then compiled, based on estimates derived from 1970 census data, percentage estimates of non-white population in each of these areas. These two sets of percentages were then correlated on a scattergram he prepared (Plaintiff's Exhibit 5, Figure 4). As may be seen from an examination of the scattergram, it shows a significant correlation between acceptance rate/percentage of white population or conversely, rejection rate/percentage of non-white population.

Defendant Amoco produced witnesses who opined that the methodology used by Plaintiff was faulty for various reasons. *1029 Amoco claimed the scattergram was based on incorrect demographic data and that even assuming the correctness of the data reflected in the scattergram, it did not prove anything about relationship between use of zip code ratings and rejection of black credit applicants, because (1) it made the unsupported assumption that the reason for rejection was the applicant's residence in his particular zip code area and (2) it made the unsupported assumption that Amoco's applicant pool for each zip code area reflected the racial composition of the population as a whole.

The Equal Credit Opportunity Act, 15 U.S.C. § 1691, was originally passed in 1974 to prohibit credit discrimination based on sex and marital status. It was amended in 1976 to, among other things, prohibit race discrimination in credit transactions. The Act provides that an aggrieved applicant may sue for actual damages, punitive damages and equitable and declaratory relief. 15 U.S.C. § 1691e(a)-(c). A successful litigant is entitled to recover costs and a reasonable attorney's fee. 15 U.S.C. § 1691(d). There has been relatively little judicial interpretation of the Act, thus leaving unresolved many provocative issues, including some presented in this case. In the first place, must a plaintiff show actual damages in order to be entitled to any relief under the Act? Secondly, what must a plaintiff show in order to make out a prima facie case? Can discrimination be inferred by using an "effects test" concept, per Griggs v. Duke Power Company, 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971)? If so, what is the proper methodology to be utilized?

Dealing with the damages question first, the Court holds that it is not necessary for plaintiff to prove actual damages to be entitled to relief. In this case, Plaintiff did state she was humiliated by Amoco's failure to grant her credit and that she felt she had been damaged by having to reveal the refusal of credit on future credit applications. This does not constitute proof of actual damage. However, it is clear to the Court that the Act did not envision the necessity of proof of actual damages as a prerequisite to entitlement to punitive damages, equitable or declaratory relief, or attorney's fees. Section 1691e(d) provides that "in the case of any successful action under subsection (a), (b), or (c) of this section, the cost of the action, together with a reasonable attorney's fee as determined by the Court, shall be added to any damages awarded by the court under such subsection." Subsection (a) is the subsection dealing with actual damages; subsection (b) pertains to punitive damages; and subsection (c) relates to equitable and declaratory relief. Obviously, § 1691e(d) envisions that there could be a recovery of punitive damages or equitable or declaratory relief in a situation where no actual damages are found. Therefore, the Court determines that should a violation be found, it has the power to award Mrs. Cherry punitive damages, even where no actual damages have been shown.

The question of what a plaintiff must show under the Equal Credit Opportunity Act to make out a prima facie case is a more subtle one. The Court notes that the drafters of the regulations accompanying the Act, see 12 C.F.R. § 202.6, footnote 7 as well as the legislative history of the Act, see Senate Report to accompany H.R. 6516, No. 94-589, pp. 4-5; House Report to accompany H.R. 6516, No. 94-210, p. 5, U.S. Code Cong. & Admin.News 1976, p. 403, assume that the "effects test" will be utilized in cases under the Equal Credit Opportunity Act. See also Hsia, Credit Scoring and the Equal Credit Opportunity Act, 30 Hastings L.J. 371 (1978).

The so-called "effects test" is derived from a decision of the U.S. Supreme Court, Griggs v. Duke Power Company, 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971), a Title VII case under 42 U.S.C. § 2000e, et seq. There, the employer required persons hired in the labor department of its power generating facility to have a high school diploma. The effect of this requirement was to exclude a significantly higher percentage *1030 of black applicants from qualifying for employment than white applicants. In reversing the Fourth Circuit Court of Appeals' determination that Title VII required proof of intent to discriminate, and that such intent could not be inferred from mere disparate impact of the diploma requirement on black persons, the Supreme Court held that Title VII not only proscribes intentionally discriminatory conduct but also looks to the consequences of conduct, whether intent to discriminate is present or not. It held that plaintiff's evidence of disparate impact on a protected class was sufficient to make out a prima facie case. At that point, it became the employer's burden to show that the job requirement was related to job performance.

Although there are significant differences between the scope and purpose of Title VII and the Equal Credit Opportunity Act, and between the employment and credit settings, the court nonetheless concludes that use of an effects test concept is an available method for a plaintiff to make out a prima facie case. This conclusion is based upon the Court's assumption that otherwise, the Act will provide a remedy only in those rare cases where a company deciding on credit expressly states it is denied for a prohibited reason. Also, the Court is cognizant of the fact that discrimination in credit transactions is more likely to be of the unintentional, rather than the intentional, variety. The employment setting necessarily involves day-to-day dealings and contact between employer and employees; the relationship between a creditor and its customers tends to be more attenuated. Presumably, the creditor will act in whatever manner it deems best calculated to assure the creditworthiness of its customers and the maximum sales of its products. More personal considerations are not likely to be a factor. However, a given creditor could operate under the notion that a particular class of persons protected under the Act, are, for whatever reason, less reliable or creditworthy than others and may consciously or subsconsciously select criteria which will have a tendency to "screen out" applicants in that class. If so, such criteria should be subjected to scrutiny to see if they are really necessary to meet legitimate business objectives, namely, accurately predicting creditworthiness.

The Court notes, however, that utilization of the effects test based on statistical methodology is apt to be quite difficult for a plaintiff. In the first place, the Act specifically proscribes inquiry by the creditor into the race, sex or marital status of a credit applicant, except in loans secured by residential real estate. See 12 C.F.R. § 202.5(c) (5); 12 C.F.R. § 202.13. Therefore, a creditor will not have direct information indicating the racial or other profile of its applicants or of the class of persons whose credit applications were granted. The conventional statistical methodology for showing disparate effect of a facially neutral test or practice is to compare representation of the protected class in the applicant pool with representation in the group actually accepted from the pool. If the statistical disparity is significant, then plaintiff is deemed to have made out a prima facie case.

Here, Plaintiff did not seek to make a statistical comparison based on the actual applicant pool. Presumably this was not done because it could not be, based on the specific proscription in the Act. Instead, she attempted to show that the zip code criterion has a disparate impact on black people by showing that the percentage of applicants rejected in various low-rated zip code areas correlated to the percentage of black population in each of said zip code areas. This evidence does not make out a prima facie case.

The Court agrees with Amoco's contention that Plaintiff's proof fails to show that the zip code ratings tend to adversely affect black applicants disproportionately. Rather, it shows only that the computerized grading system taken as a whole tends to reject a disproportionate number of persons living in predominantly black areas. In other words, it is deficient for two reasons: *1031 (1) It does not test the effect of zip code as a criterion involved but rather tests the effect of the overall 38-criteria grading scheme and (2) it does not deal with either an actual applicant pool or with one which could reasonably be assumed to possess the approximate characteristics of the applicant pool.^[9]

Having rejected the only theory advanced by Plaintiff herein, the Court nonetheless looks to see if other theories exist by which Plaintiff might prevail, based on the evidence introduced at the trial.

The Court does not think that proof of disparate impact need be shown by statistics in every case nor need it be shown by proof of actual disproportionate exclusion from the applicant pool. Because of the particular nature of the criterion complained of here, it may be possible to gain some enlightenment as to its effect from examination of the zip code map and the demographic data before the Court.

If the housing pattern in Atlanta is such that virtually all white persons live in neutrally-rated or high-rated zip code areas, but virtually all black persons live in lowrated areas, then the zip code criterion becomes suspect. Indeed, if the zip code/race correlation is high enough, the criterion itself takes on racial aspects so that it may be considered as a mere substitute for consideration of the applicant's race. The Court's examination of the evidence here indicates that no distinct racial pattern is evident, however. The low-rated zip code areas bearing prefix 303on which Plaintiff's proof is based are not all predominantly black. In fact, the majority of them are predominantly white. Further, looking at the aggregate population of all such low-rated areas, one finds that 60% of such population is white and 40% is black.

A possible problem with zip code ratings in an area with a segregated housing pattern, such as Atlanta, is that as a result of actual or perceived lack of housing opportunities, black persons of all economic strata may tend to be grouped in certain zip code areas. If so, and if the economic housing pattern for white persons differs, it could mean there is a greater likelihood that an otherwise qualified black applicant would live in a low-rated zip code area than a similarly qualified white credit applicant. In order to test this hypothesis, it would be necessary to construct a hypothetical applicant pool on some basis reasonably gauged to approximate the actual applicant pool for example, an income qualified group.^[10] By looking at both the economic and racial composition of various areas of the city, one might determine whether the zip code ratings adversely affect income-qualified black persons more than income-qualified whites. Thus, if the segregated housing pattern in Atlanta means that the zip code ratings penalize^[11] a high percentage of income-qualified black persons but only a low percentage of income-qualified white persons, then Plaintiff will have made out a prima facie case. At that point, it becomes Defendant's burden to show that the zip code ratings make Amoco's credit evaluation system more predictive than it would be otherwise.

The problem here, however, is that the Court does not have before it the data necessary to make such an analysis. Although the Court has some personal familiarity with the racial and economic makeup of various areas of the city, this general knowledge is an insufficient basis for such an analysis. Further, the Court notes that certain facts run counter to a theory that zip code ratings penalize income-qualified blacks but not income-qualified whites in Atlanta. For example, zip code areas 30305 *1032 (Buckhead) and 30309 (Ansley Park, Brookwood Hills) both of which are rated 2 by Amoco, are well-known as predominantly moderate to high income white areas. Therefore, even if the Court were to take liberal judicial notice of facts not presented to it by counsel, there would nonetheless be insufficient evidence from which to draw a conclusion as to whether Amoco's use of zip code ratings treats otherwise qualified white applicants and otherwise qualified black applicants in a significantly different manner.

For the foregoing reasons, the Clerk is DIRECTED to enter judgment in favor of Defendant Amoco Oil Company, with costs cast upon Plaintiff.

SO ORDERED.

[1] Both at the conclusion of Plaintiff's evidence and at the conclusion of all the evidence, Defendant made motions for involuntary dismissal, arguing that Plaintiff had failed to make out a prima facie case. These motions were retained under advisement by the Court at the conclusion of the trial. The Court chooses to enter judgment based on all the evidence and said motions are hereby DENIED.

[2] Plaintiff sought to amend her pleadings on the eve of trial to add a claim for injunctive relief. The request was denied for the reasons stated in the Court's Order of April 25, 1980.

[3] 15 U.S.C. § 1691, also known as Title VII of the Consumer Credit Protection Act, 15 U.S.C. § 1691, et seq.

[4] The 303 zip code areas appear to encompass all of Fulton County except the northernmost part (Alpharetta and Roswell); north DeKalb County is included. Downtown Atlanta lies in the central part of Fulton County. The low-rated areas basically include the central urban area, plus all of south Fulton County below the downtown area. The north Fulton County and north DeKalb areas tend not to lie in the low-rated category, although certain zip codes in those areas (30305, 30324, 30341, 30340) are low-rated.

[5] The majority of Atlanta's ("Atlanta" as referred to in this Order being defined as the area composed of Fulton, DeKalb, Cobb, Clayton, Gwinett, Douglas and Rockdale counties) black population lives in these zip code areas. The testimony of Plaintiff's expert indicated this percentage to be 73%. Defendant's expert testimony attacked this percentage as being too high, but the Court remains convinced that the percentage of Atlanta's black population living in these zip code areas is close to 70%.

[6] In fact, no Atlanta zip code areas are rated 5; only one is rated 4. All the rest are rated 1, 2, or 3. A rating of 1 or 2 has a negative impact on an application. 3 is a neutral rating which has no effect on acceptance or rejection of the application.

[7] The Court's Order of November 28, 1979, dealt with Defendant's challenge to Plaintiff's standing here.

[8] Amoco admits this is a reference to zip code area.

[9] The Court recognizes that some Title VII cases have permitted reference to the general population on the assumption that the applicant pool would possess approximately the same characteristics. But this assumption is not valid in every case, and Plaintiff did not show the Court why it would be valid here.

[10] I. e., a group meeting the median income standard of the applicant pool. Presumably, this information is available from Amoco's computer.

[11] By "penalize" the Court means the negative weight or impact assigned to low-rated zip code areas. See footnote 6, supra.