Exceptional Proff 602

| About LexisNexis | Privacy Policy | Terms & Conditions | Copyright © 2019 LexisNexis

Date and Time: Monday, March 25, 2019 9:23:00 PM CDT

Job Number: 85607755

Document (1)

1. ARTICLE: Stopping a Vicious Cycle: The Problems with Credit Checks in Employment and Strategies to Limit

Their Use, 18 Geo. J. Poverty Law & Pol'y 45

Client/Matter: -None-

ARTICLE: Stopping a Vicious Cycle: The Problems with Credit Checks in Employment and Strategies to Limit Their Use

Fall, 2010

Reporter 18 Geo. J. Poverty Law & Pol'y 45 *

Length: 19465 words

Author: Sharon Goott Nissim *

* J.D. 2010, Northwestern University School of Law; B.A. 2006, Yale University. Many thanks to Professor Kimberly Yuracko for all her advice, wisdom, and patience. Thanks to those I worked with at the Lawyers' Committee for Civil Rights, especially Sarah Crawford and Audrey Wiggins, whose passion inspired this article. And thank you to my dad, Eugene Goott, for reading many drafts with his careful eye. (c) 2010, Sharon Goott Nissim.

Highlight

This article explores a new and increasingly common phenomenon: the use of credit checks by employers to evaluate potential and current employees. This practice has profound implications in this current weak economy, as those who most need jobs often are the ones turned away due to bad credit. The use of credit checks also has a disproportionate effect on racial minorities, as statistically they tend to have worse credit than non-minorities. Employers often assert that credit checks are necessary, despite the lack of hard data proving a link between poor credit and poor job performance.

This article examines two ways to combat these troublesome new policies--through litigation and legislation. While litigation has rarely been used to challenge credit check policies directly, a case can be made that these policies violate Title VII disparate impact provisions. However, there are many obstacles to a successful credit check lawsuit, including the lack of specific statistics that courts increasingly find necessary to prove a disparate impact claim. Therefore, more success may be had through legislation and the public policy arena in general. Currently, four states have passed some form of anti-credit check legislation and many others are debating the issue in their legislatures. In this article, I offer my proposal for model credit check legislation and then evaluate the legislative responses that have been enacted or are under consideration at the state and federal level and how they measure up to this model.

Text

[*45] I. INTRODUCTION

Kevin Palmer, who was fifty years old last August, lived in homeless shelters for several months in California before finally landing an interview with a property management company. 1 The job he interviewed for was a "glorified

1 Jonathan D. Glater, Another Hurdle for the Jobless: Credit Inquiries, N.Y TIMES, Aug. 7, 2009, at A1.

Page 2 of 27

[*46] clerk's job, taking homeowners' complaints," Mr. Palmer explained. 2 Yet after a great interview during which a manager walked him around the office, introduced him to people, and even showed him an empty desk, Mr. Palmer was denied the job when a credit check revealed that he had filed for bankruptcy. 3 The job would have helped him get back on his feet after the bank foreclosed on his condominium, the event which led to his bankruptcy filing. 4

Mr. Palmer eventually found a similar job at a company that did not run a credit check on him, but his story is being repeated all over the country. Employers are increasingly using credit checks as a strategy to screen job applicants and occasionally to monitor current employees. During an economic downturn, the use of credit checks as a screening tool only exacerbates the difficulty of finding work. This leads to a vicious cycle: the people who most need a job (generally those already having credit problems) are not able to get or keep a job, resulting in more credit problems and no way to remedy them. Employers often justify these policies in vague terms, and do not cite concrete evidence of any connection between bad credit and poor job performance. Because minority populations tend to have poorer credit scores, these credit-check policies have a greater impact on minority job applicants than on non-minority applicants.

Despite the troubling impact of these policies on minorities and the lack of any convincing business justifications, few lawsuits on this issue have been filed. 5 There are very few cases that speak directly to the legality of employer credit check usage and whether their use may potentially violate any laws, such as the disparate impact provision of Title VII. There are older court cases that may at first glance seem analogous to the issue at hand (e.g., employer use of arrest or conviction records), but courts have now largely abandoned the expansive approach to disparate impact claims that they took during that time period. It is very difficult to predict how courts would come out today on a claim that credit checks violate the disparate impact provision of Title VII.

Because of the uncertainties associated with litigation, change should be [*47] approached through other avenues. The only arena in which the problem is being discussed currently is in legislatures across the country, including at the federal level. Several states have passed laws that are moving in the right direction, although they do not constitute perfect solutions to the problem. 6

In this article, I begin by outlining the problem posed by increased use of employer credit checks, highlighting the disparate impact on minorities and the lack of research proving any connection between bad credit and bad job performance. I then move on to explore whether this problem could be addressed within the framework of Title VII disparate impact litigation, and ultimately conclude that while older precedents seem helpful, there are reasons to be very cautious in seeking to curb the use of credit checks through litigation. The next section examines the

2 Id.

3 Id.

4 Id.

5 Recently, however, the EEOC has at least filed one of the few, if only, credit-check lawsuits in Equal Employment Opportunity Commission v. Freeman in district court in Maryland. This case involves "a nationwide lawsuit alleging that a company had unlawfully discriminated against African-American, Hispanic and male job applicants by using credit history and criminal background checks as a selection criteria." Background Checks Under Attack: EEOC Takes Aim at Hiring Practices, GENTRY, LOCKE, RAKES & MOORE LLP (Oct. 14, 2009), http://www.gentrylocke.com/showarticle.aspx?Show=1330 (discussing Equal Emp't Opportunity Comm'n v. Freeman, No. 90cv02573 (D. Md., filed Sept. 30, 2009)). Also, in January 2010, an administrative law judge with the Department of Labor found that Bank of America discriminated against minorities in its hiring practices, and pointed to the use of credit-checks as part of the problem. The judge evaluated the case under Title VII's disparate treatment framework, finding the statistical evidence to be strong enough to constitute proof of discriminatory intent. The judge acknowledged that the lack of specific criteria used by the Bank in evaluating a person's credit made the process "highly subjective."OFCCP v. Bank of America, 1997-OFC-16, ALJ's Recommended Decision and Order at 221 tbl.1 (Dep't of Labor Jan. 21, 2010).

6 See infra Part IV.

18 Geo. J. Poverty Law & Pol'y 45, *45

Page 3 of 27

political response to credit checks and how that type of pressure may be more effective than litigation. I propose a model bill that would generally ban employer use of credit checks With some specific exceptions. Lastly, I consider the legislative responses that have been enacted or are under consideration at the state and federal level and evaluate how those provisions measure up to this model.

II. THE PROBLEMS WITH CREDIT CHECKS FROM A SOCIAL/POLICY PERSPECTIVE

In the past few years, employers have increasingly used credit checks as a screening tool for prospective employees, often to the detriment of job applicants. 7 Sixty percent of employers report checking a job applicant's credit (sometimes including late payments on mortgages, rents, and student loans), representing an increase from thirty-six percent in 2004. 8 Certain industries, such as the retail industry, rely very heavily on this practice. 9 Although it is common to run credit checks on potential employees for positions that deal with sensitive financial data or those with national security concerns, 10 employers are not required to prove any connection between the position they are hiring for and a person's credit history. 11

In fact, there are hardly any regulations in place as to what employers can do with credit checks they obtain. Rather, most of the regulations in place deal with [*48] procedural requirements, and do not actually prevent employers from using credit reports in hiring decisions. 12 The Federal Credit Reporting Act (FCRA) clearly allows the procurement of credit reports for employment purposes. 13 As a restriction on this procurement, the FCRA merely mandates that employers comply with all federal and state equal employment laws in the use of credit information. 14 An employer also must make a "clear and conspicuous" disclosure that it has requested a credit report from an agency and an employee must give his/her approval. 15 Unfortunately, this notification requirement does not provide much protection for employees, as most will assume that if they do not give permission for an employer to view their credit history, then they will not get the job. 16 Employers must also notify applicants if their credit report is then used to make an adverse employment decision. 17 In practice, however, job applicants will rarely challenge an employer under this provision because the applicant has no way of knowing why he or she has

7 See Thomas Frank, Job Credit Checks Called Unfair, USA TODAY, Feb. 13, 2009, at 1A.

8 Id.; SOC'Y FOR HUMAN RES. MGMT., BACKGROUND CHECKING: CONDUCTING CREDIT BACKGROUND CHECKS (2010), available at http://www.shrm.org/Research/SurveyFindings/Articles/Pages/BackgroundChecking.aspx.

9 ADAM T. KLEIN, RENIKA MOORE & SCOTT A. MOSS, EMPLOYER CREDIT-HISTORY CHECKS AND CRIMINAL RECORD CHECKS OF JOB APPLICANTS FOR HIRING DECISIONS: THE ILLEGALITY UNDER TITLE VII DISPARATE IMPACT DOCTRINE 1-2 (2007), available at http://www.outtengolden.com/files/Klein_EEOC_Testimony_final.pdf.

10 See Frank, supra note 7 ("The TSA bars people with $ 5,000 in overdue debt or any federal or state tax lien from working as airport screeners. That ruled out 22% of applicants from 2005 to 2007, a report found.").

11 See id.

12 See, e.g., 15 U.S.C. § 1681 (2006).

13 15 U.S.C. § 1681b(a)(3)(B) (2006).

14 15 U.S.C. § 1681b(b)(1)(A)(ii) (2006).

15 15 U.S.C. § 1681b(b)(2)(A) (2006).

16 Eileen Ambrose, Employers Frequently Use Credit Checks to Screen Job Applicants, BALTIMORE SUN, Oct. 24, 2010, available at http://articles.baltimoresun.com/2010-10-24/business/bs-bz-ambrose-credit-report-20101024_1_credit-checks- credit-reports-time-employers.

17 15 U.S.C. § 1681b(b)(3) (2006).

18 Geo. J. Poverty Law & Pol'y 45, *47

Page 4 of 27

not been hired. Finally, the FCRA is weak because it allows inaccurate information to go out in credit reports to employers. 18 Additionally, the Bankruptcy Code separately states that when an applicant has filed for bankruptcy, an employer cannot deny employment on that basis alone, but it would be difficult to show that the decision was made only because of the applicant's bankruptcy. 19

Statistics clearly show that minorities are disproportionately impacted by these credit check policies, as they tend to have worse credit than non-minorities. The Federal Reserve reported to Congress that African-American credit scores are half those of white non-Hispanics. 20 A Brookings Institution Study conducted in [*49] 2006 found correlations between areas with high minority populations and lower average credit scores, 21 and a 2004 Missouri study found that residents in zip codes with high minority populations tend to have much lower credit scores. 22 Several other studies find similar statistics. 23

Not only do credit checks disproportionately impact minority job applicants, but employers' arguments for using credit checks are weak because credit checks do not accurately predict job performance. Employers who use credit checks generally believe that employees with good credit will be more responsible employees, will be less likely to steal from the employer, and will be less likely to commit crimes in general. 24 Employers are also undoubtedly influenced by lawsuits finding employers liable for negligent hiring and hiring of dangerous employees. 25 Yet, studies have not found support for any such correlation: in fact, one study conducted by professors at Eastern Kentucky University reviewed the credit reports of 200 employees and found no link between credit score and job performance. 26 Even a TransUnion Credit Bureau official admitted under oath [*50] that, "At this point we don't have any research to show any statistical correlation between what's in somebody's credit report and their job performance or their likelihood to commit fraud." 27

Additionally, credit reports are often rife with errors, thereby undermining even the purported connection to job performance. 28 Specifically, twenty-five percent of credit reports in 2005 contained errors serious enough to cause that person to be denied a loan or employment. 29 One survey found that seventy-nine percent of consumer

18 Kelly Gallagher, Note, Rethinking the Fair Credit Reporting Act: When Requesting Credit Reports for "Employment Purposes" Goes Too Far, 91 IOWA L. REV 1593, 1603 (2006) ("The FCRA requires that credit reporting agencies use 'reasonable procedures' to ensure accuracy. However, the 'technical accuracy defense' allows reporting agencies to void liability for factually correct information that is incomplete or misleading.").

19 11 U.S.C. § 525(b) (2006). Under this section, added in 1984 to the Bankruptcy Code, private employers cannot terminate an employee or otherwise discriminate against an employee because the employee has been a "debtor under this title or bankrupt under the bankruptcy act." This provision has been interpreted somewhat narrowly, having been found not to apply to situations where someone has merely filed for bankruptcy. This stands in contrast to § 525(a), which applies to the government and is written and interpreted more broadly. See Hope A. Comisky & Christopher P. Zubowicz, Labor and Employment Law Update: Employers May Consider Applicant's Bankruptcy History in Making Hiring Decisions, PEPPER HAMILTON LLP (July 1, 2004), http://www.pepperlaw.com/publications_update.aspx? ArticleKey= 321.

20 FEDERAL RESERVE, REPORT TO THE CONGRESS ON CREDIT SCORING AND ITS EFFECTS ON THE AVAILABILITY AND AFFORDABILITY OF CREDIT 80 (2007), available at http://www.federalreserve.gov/boarddocs/RptCongress/creditscore/creditscore.pdf (reporting scores of 25.6 out of 100 for African-Americans, 54.0 for non-Hispanic whites, and 38.2 for Hispanics).

21 MATT FELLOWS, THE BROOKINGS INSTITUTION, CREDIT SCORES, REPORTS, AND GETTING AHEAD IN AMERICA 9 (2006), available at http://www.brookings.edu//media/Files/rc/reports/2006/05childrenfamilies_fellowes/20060501_creditsco res.pdf (finding that counties with credit scores ranging from 850-720 (very low risk) had a 1% African-American population and a 4% Hispanic population while counties with credit scores from 500-559 (very high risk) had a 26% African-American population and a 23% Hispanic population).

22 Gallagher, supra note 18, at 1607 n.97 (citing BRENT KABLER, STATE OF MO. DEP'T OF INS., INSURANCE BASED CREDIT SCORES: IMPACT ON MINORITY AND LOW INCOME POPULATIONS IN MISSOURI 1 (2004), available at http://insurance.mo.gov/reports/credscore.pdf).

18 Geo. J. Poverty Law & Pol'y 45, *48

Page 5 of 27

credit reports contained mistaken information, including errors such as false delinquencies or accounts involving someone other than the consumer. 30 And, of course, a person may have bad credit for reasons completely beyond that person's control, such as disability, divorce, death in the family, illness, or identity theft, just to name a few--all circumstances that would have little to no bearing on how a person would perform in his or her job. 31

The Equal Employment Opportunity Commission (EEOC) has not weighed in on this issue in a comprehensive way but has indicated that it recognizes employer credit checks as a problem. The EEOC Guide to Pre-Employment Inquiries ("Guide") suggests employers avoid certain inquiries of job applicants, including credit history questions, as they could potentially expose the employer to liability for disparate impact discrimination. 32 The Guide explains that employment decisions made on the basis of information elicited from such questions violates Title VII unless the "information is needed to judge an applicant's competence or qualification for the job in question." 33 The EEOC has, in the past, issued comprehensive guidelines to employers on how to deal with checking applicants' criminal histories, which could also apply to the use of [*51] credit checks, as there are similarities between the two policies. 34 The EEOC also issued two decisions in the 1970s indicating that the use of credit checks violates Title VII disparate impact provisions if done without a legitimate business necessity. 35 The agency stated in 2005 in testimony before the Hawaii state legislature that credit checks could be "challenged under the disparate impact framework." 36 Then, on May 17, 2007, the EEOC held a public commission meeting in which one of the topics discussed was the discriminatory impact of credit checks. Finally, one of the goals of EEOC's E-Race initiative of November 2007 was for the Commission to issue further guidance on this matter. 37 Although EEOC policies are not generally given much deference by courts, 38 it is surprising that the EEOC has not done more to combat the potentially discriminatory use of credit checks by employers.

In short, credit checks by employers are troubling. Their connection to job performance is uncertain, yet the detrimental impact of their use on some of society's most vulnerable groups--racial minorities and the poor--is clear. As Hawaii State Rep. Marcus Oshiro put it, for those caught in a vicious cycle of bad credit and unemployment, "[i]t's almost like being forever sentenced to debtors' prison." 39 In the next section, I consider the extent to which protection from credit checks may be found in already existing antidiscrimination laws.

23 A study by the Texas Department of Insurance in 2004 found that "in general, Blacks have an average credit score that is roughly 10% to 35% worse than the credit scores for Whites. Hispanics have an average credit score that is roughly 5% to 25% worse than those for Whites. Asians have average credit scores that are about the same or slightly worse than those for Whites." TEX. DEP'T OF INS., REPORT TO THE 79TH LEGISLATURE: USE OF CREDIT INFORMATION BY INSURERS IN TEXAS 13 (2004), available at http://www.tdi.state.tx.us/reports/documents/creditrpt04.pdf. A 1996 study in the Boston area found that minority applicants for home mortgages generally have weaker credit histories than white applicants. Gallagher,supra note 18, at 1606 (citing Alicia H. Munnell et al., Mortgage Lending in Boston: Interpreting HMDA Data, AM. ECON. REV. 25, 26 (1996)).

24 See Andrew Martin, As a Hiring Filter Credit Checks Draw Questions, N.Y. TIMES, Apr. 9, 2010, at Al (citing Eric Rosenberg of TransUnion Credit Bureau quoting statistics that retailers lose 30 billion a year from employee theft, workplace violence costs employers 55 million a year in lost wages, and a third of employees put false information on their resumes); Gallagher, supra note 18, at 1599.

25 Gallagher, supra note 18, at 1599.

26 JERRY K. PALMER & LAURA L. KOPPES, AM. PSYCHOLOGICAL SOC'Y, FURTHER INVESTIGATION OF CREDIT HISTORY AS A PREDICTOR OF EMPLOYEE TURNOVER 1 (2003), available at http://www.house.mi.gov/SessionDocs/2009- 2010/Testimony/Committee15-3-17-2010-7.pdf; see also Martin, supra note 24 ("Jerry K. Palmer, a psychology professor at Eastern Kentucky University, said his studies, though relatively small, found no correlation between the quality of an employee's credit report and that worker's job performance or likelihood to quit. He said he was not aware of any studies that showed a correlation between poor credit and employee fraud or violence."); Gallagher, supra note 18, at 1595 n.3 ("The fact that … articles fail to articulate why employers should perform credit checks and claim no empirical support for this screening technique itself suggests a lack of foundation for believing credit scores reliably predicts job performance and propensity to steal from an employer.").

27 Martin, supra note 24.

18 Geo. J. Poverty Law & Pol'y 45, *50

Page 6 of 27

III. THE LEGAL/DOCTRINAL RESPONSE AND WHY IT IS INADEQUATE

I will first give a brief overview of the purpose and goal of the disparate impact cause of action and describe in general what a disparate impact claim looks like. I will then describe in more depth each part of a disparate impact claim--the plaintiff's prima facie case and the defendant's business necessity rebuttal. In [*52] each of those sections I will explore whether the legal doctrine and precedents justify hopefulness or caution about the potential for success of a credit check lawsuit under Title VII.

A. Overview of Disparate Impact Framework

Until 1971, the only legally cognizable employment discrimination claim based on race was a disparate treatment claim. In order to make out a successful disparate treatment claim, a plaintiff has to prove that an employer intentionally discriminated against an employee based on the employee's race or other protected characteristic. 40 But the limits of disparate treatment as an effective bulwark against employment discrimination increasingly became clear as overt racism faded, and more complex and often unintentional forms of discrimination resulted from seemingly benign employer hiring policies. 41 The landmark Supreme Court case Griggs v. Duke Power Co. recognized this problem in 1971, and established the disparate impact cause of action under Title VII. 42 The Griggs Court recognized that Congress' intention in passing the Civil Rights Act of 1964 was broad:

The objective of Congress in the enactment of Title VII is plain from the language of the statute. It was to achieve equality of employment opportunities and remove barriers that have operated in the past to favor an identifiable group of white employees over other employees. Under the Act, practices, procedures, or tests neutral on their face, and even neutral in terms of intent, cannot be maintained if they operate to "freeze" the status quo of prior discriminatory employment practices. 43

The issue in Griggs was whether the Duke Power Company's policy of requiring a high school diploma or passage of an IQ test as a condition of employment was legal under Title VII. This case found, for the first time, that plaintiffs no longer needed to prove discriminatory intent in order to have a cognizable employment discrimination claim

28 House Bill 31 -- Relating to Employment Practices; Credit History Check Prohibition: Hearing on H.B. 31 Before the S. Comm. on Labor (Haw. 2009) [hereinafter EEOC Testimony] (testimony of Stuart J. Ishimaru, Acting Chairman of the EEOC) (citing CR Investigates. Credit Scores: What You Don't Know Can be Held Against You, CONSUMER REPORTS, Aug. 1, 2005); see also Gallagher, supra note 18, at 1598 (citing CONSUMER FED'N OF AM., NAT'L CREDIT REPORTING ASS'N, CREDIT SCORE ACCURACY AND IMPLICATIONS OF CONSUMERS (2002) (finding 29% of credit reports contained errors that could end up with denial of credit); see also U.S. Gov'T ACCOUNTABILITY OFFICE, CONSUMER CREDITR: LIMITED INFORMATION EXISTS ON EXTENT OF CREDIT REPORT ERRORS AND THEIR IMPLICATIONS FOR CONSUMERS (2003) (statement of Richard J. Hillman, Director, Financial Markets and Community Investment), available at http://www.gao.gov/new.items/d031036t.pdf. Similar concern over accuracy arises in the arena of criminal history checks; apparently commercial criminal databases and services have also been found to be full of error. Rebecca Oyama, Note,Do Not (Re)Enter: The Rise of Criminal Background Tenant Screening as a Violation of the Fair Housing Act, 15 MICH. J. RACE & L. 181, 181 (2009).

29 EEOC Testimony, supra note 28.

30 U.S. PIRG, MISTAKES DO HAPPEN: A LOOK AT ERRORS IN CONSUMER CREDIT REPORTS 2004, available at http://www.uspirg.org/home/reports/report-archives/financial-privacy--security/financial privacy--security/mistakes-do-happen-a- look-at-errors-in-consumer-credit-reports.

31 Gallagher, supra note 18, at 1599.

32 See generally FAIR EMPL. PRAC. MAN., EEOC GUIDE TO PRE-EMPLOYMENT INQUIRIES 8A (1992) [hereinafter EEOC GUIDE]; EEOC EMPLOYMENT TESTS AND SELECTION PROCEDURES, http://www.eeoc.gov/policy/docs/factemployment_procedures.html (last visited Nov. 12, 2010).

33 EEOC GUIDE, supra note 32.

18 Geo. J. Poverty Law & Pol'y 45, *51

Page 7 of 27

under Title VII, but rather that a discriminatory impact or effect was potentially sufficient for a successful claim. 44 Griggs proscribed practices that are "fair in form, but discriminatory in operation," with an exception only for those practices that a defendant could [*53] show bore a "demonstrable relationship" to successful job performance. 45

Courts at the time were fairly candid about their expansive interpretation of the relatively new disparate impact cause of action under Title VII. Post-Griggs, these courts seemed eager to give full measure to Congress' intentions in creating Title VII. 46 "Indeed," the Court wrote in a case decided soon after Griggs, "the entire thrust of the [Griggs] opinion is toward a liberal construction of Title VII so as to fully effectuate the congressional mandate to insure members of minority groups equal employment opportunities and eliminate employment practices which act as 'built-in headwinds' for minority groups." 47 In recent years, courts have backed away from this expansive approach, but that was how courts originally viewed the spirit and congressional intent of the disparate impact cause of action.

Disparate impact cases generally involve a two-step process. 48 First, the plaintiff must make a statistical showing that the employer's policy does indeed have a disparate impact on a protected group. 49 This often consists of statistics involving either the actual applicants for a job or the qualified labor pool in a relevant geographical area. 50 Additionally, the statistical impact the plaintiff shows must be legally cognizable--that is, the court must not consider the trait at issue voluntary or mutable. 51 If the plaintiff is able to make out a prima facie case, the employer must then show that the policy is job related or consistent with business necessity, and if the employer satisfies that burden, then the employee can rebut by showing that there is an alternative policy that would have a less discriminatory impact. 52

34 EQUAL EMP'T OPPORTUNITY COMM'N, EEOC COMPLIANCE MANUAL § 604 APP.A, POLICY STATEMENT ON THE ISSUE OF CONVICTION RECORDS UNDER TITLE VII OF THE CIVIL RIGHTS ACT OF 1964, AS AMENDED, 42 U.S.C. § 2000E (1982) (1987) (stating that when considering conviction records, employers are to consider "(1) the nature and gravity of the offense; (2) the time that has passed since the conviction and/or completion of the sentence; and (3) the nature of the job held or sought"); EQUAL EMP'T OPPORTUNITY COMMON, EEOC GUIDANCE: POLICY GUIDANCE ON THE CONSIDERATION OF ARREST RECORDS IN EMPLOYMENT DECISIONS UNDER TITLE VII OF THE CIVIL RIGHTS ACT OF 1964, AS AMENDED, 42 U.S.C. § 2000E (1982) 1 (1990) (stating that exclusion from employment on the basis of arrest is only justified when it "appears that the applicant or employee engaged in the conduct for which he was arrested and that the conduct is job-related and relatively recent" and blanket policies are not allowed).

35 EEOC Decision No. 72-1176, 5 Fair Empl. Prac. Cas. (BNA) 960 (1972) (holding that a bank policy of using credit information to evaluate potential employees was unlawful without business justification); EEOC Decision No. 74-2, 6 Fair Empl. Prac. Cas. (BNA) 830 (1973) (holding that a manufacturing company's policy of using applicant's financial status was unlawful without business necessity).

36 EEOC Testimony, supra note 28, at 4.

37 Id.

38 See El v. Se. Pa. Transp. Auth., 471 F.3d 232, 244-45 (3d Cir. 2007) (finding these guidelines not entitled to deference and instead adopting a broader standard).

18 Geo. J. Poverty Law & Pol'y 45, *52

Page 8 of 27

B. Plaintiff's Prima Facie Case

The plaintiff's prima facie case begins with a legally cognizable statistical showing of a disparate impact on a protected group that resulted from the policy at issue. I start by outlining helpful precedents, suggesting plaintiffs may be able to succeed in making a prima facie case that credit checks cause a disparate impact. I then show that this optimism should be tempered, as more recent cases provide a more complicated picture and indicate that it is not at all clear how a [*54] suit under Title VII against credit checks would fare in today's courts.

1. Helpful Precedents

The first step in making a prima facie case is to demonstrate that the policy at issue is causing a statistically significant disparate impact on minority job applicants. 53 In United States v. Chicago, one of the rare cases actually addressing the credit-check issue directly, the Northern District of Illinois found that a police department's practice of conducting a general investigation of job applicants' character, including credit history, had a disproportionate impact on African-Americans, and the court granted a preliminary injunction. 54 The court used very broad language, concluding that using socio-economic criteria constituted discrimination because "blacks are more likely than whites to possess negative socio-economic attributes." 55 The court cited to EEOC decisions containing census data indicating that a disproportionate percentage of nonwhites are below the poverty level and more likely to have financial difficulties. 56 This case contained no actual discussion of statistics, referring only to the general EEOC study on census data, and the outcome arguably seems to implicitly rely on stereotypes about race and economic conditions. 57

This expansive approach to finding a prima facie case of disparate impact is evident in other decisions and types of cases that are analogous to credit check cases. Cases involving wage garnishments raise similar issues to those arising from the use of credit checks 58 because they involve employees who are having financial trouble and employers' concerns over how that will affect their performance on the job. 59 Courts in these wage garnishment

39 Frank, supra note 7.

40 Civil Rights Act of 1964, 42 U.S.C. § 2000e-2(k)(1)(A) (1964) (amended 1991).

41 See Griggs v. Duke Power Co., 401 U.S. 424, 432 (1971) ("We do not suggest that either the District Court or the Court of Appeals erred in examining the employer's intent; but good intent or absence of discriminatory intent does not redeem employment procedures or testing mechanisms that operate as 'built-in headwinds' for minority groups and are unrelated to measuring job capability.").

42 See generally id. at 424.

43 Id. at 429-30.

44 Id. at 429-30 (noting that the policy had the effect of eliminating large numbers of African-Americans from those positions).

45 Id. at 431.

46 See, e.g., Johnson v. Pike Corp. of Am., 332 F. Supp. 490, 494 (C.D. Cal. 1971) ("[T]he entire thrust of [Griggs] is toward a liberal construction of Title VII so as to fully effectuate the congressional mandate to insure members of minority groups equal employment opportunities and eliminate employment practices which act as 'built-in headwinds' for minority groups."); Wallace v. DeBron Corp., 494 F.2d 674, 676 (8th Cir. 1974) (citing Johnson court language above).

47 Johnson, 332 F. Supp. at 494.

48 Civil Rights Act of 1964, 42 U.S.C. § 2000e-2(k)(1)(A) (1964) (amended 1991).

49 Id.

50 See Wards Cove Packing Co. v. Antonio, 490 U.S. 642, 650-51 (1989).

18 Geo. J. Poverty Law & Pol'y 45, *53

Page 9 of 27

cases also do not delve into the minutiae of statistics, preferring instead to make broad connections between the fact that African-Americans as a group tend to be worse off financially and the fact that they therefore would be more likely to have their wages garnished, resulting in a disparate impact. 60 For example, in Johnson v. Pike Corp. of America, the court cites studies on wage garnishment, finding "that minority group members suffer wage garnishments substantially more often than others, i.e., the proportion of racial minorities among the group of people who have had their wages garnished is significantly higher than the proportion of [*55] racial minorities in the general population." 61 With that first step established, the court then goes on to assert that:

the fact that blacks and other racial minorities are so often subject to garnishment action is related to the fact that they are to a disproportionate extent from the lower social and economic segments of our society. Approximately three times the proportion of blacks as compared to whites are in the lower end of the economic scale . . . [and] minority groups members are more often in debt, are more frequently subject to questionable credit practices and harassment, and have less capacity to defend themselves. 62

On the other hand, cases challenging the use of arrest and/or conviction records in hiring have delved into statistics on a somewhat deeper level than have the wage garnishment cases. Employer policies that exclude applicants who have been arrested or convicted are similar to credit check policies because they involve criteria that may not be in the person's control (or could be the result of error) and employer fears of workplace theft and violence. 63 In Gregory v. Litton Systems, the court cited statistics on national arrest levels showing that African-Americans are arrested at higher rates than whites. 64 Yet these data were not specific in any way to the area in which Litton Systems was located or the pool of potential employees.

In Green v. Missouri Pacific Railroad Co., a seminal case on employer use of conviction records, the court held that a plaintiff could make out a prima facie case of disparate impact under Title VII in three different ways: 1) using statistics showing that blacks as a class (or in a specific geographic area) are "excluded by the employment practice in question at a substantially higher rate than whites;" 2) comparing the actual percentages of African- American to Caucasian applicants excluded by the specific employment policy at issue at a particular company; or 3) analyzing the percentage of qualified African-Americans employed by the defendant company as compared to

51 See, e.g., Garcia v. Spun Steak, 998 F.2d 1480, 1487-89 (9th Cir. 1993); Rogers v. Am. Airlines, Inc., 527 F. Supp. 229, 231-33 (S.D.N.Y. 1981).

52 Civil Rights Act of 1964, 42 U.S.C. § 2000e-3(k)(1)(A) (amended 1991).

53 42 U.S.C. § 2000e-2(k)(l)(A) (1964) (amended 1991).

54 United States v. Chicago, 385 F. Supp. 543 (N.D. Ill. 1974).

55 Id. at 557.

56 Id. (citing to an EEOC study that, perhaps not surprisingly, used census bureau statistics to make their case, and did not address the percentage of blacks compared to whites in the area from which the defendant draws its workforce have relatively poor credit or financial records).

57 Id.

58 In these cases, the plaintiffs claimed that an employer's policy of firing individuals who have had wages garnished more than once disproportionately impacted African-Americans.

59 See Johnson v. Pike Corp. of Am., 332 F. Supp. 490, 494-95 (C.D. Cal. 1971); Wallace v. DeBron Corp., 494 F.2d 674, 675 (8th Cir. 1974).

60 See Johnson, 332 F. Supp. at 494-95; Wallace, 494 F.2d at 675.

61 Johnson, 332 F. Supp. at 494.

62 Id.

18 Geo. J. Poverty Law & Pol'y 45, *54

Page 10 of 27

the percentage of African-Americans in the "relevant geographical area." 65 The court in Green concluded that the plaintiffs made out a prima facie case because "the statistics established that MoPac's employment practice under consideration disqualifies black applicants or potential black applicants for employment at a substantially higher rate than whites." 66 The court applied the second method (comparing the actual percentage of African-American to Caucasian applicants excluded by the specific employment policy at issue at a particular company) to come to this conclusion, analyzing [*56] separately the number of whites rejected due to conviction records divided by the total number of white applicants (1.4%) and the number of blacks rejected because of prior conviction divided by the total number of black applicants (2.05%), ultimately deeming that ratio significant. 67

In Carter v. Gallagher, a 1971 case dealing with a fire department's hiring policies including the use of arrest records and conviction records, the Eighth Circuit affirmed the lower court's finding that the plaintiffs had made out their prima facie case successfully using statistics. 68 Evidence in the record showed that none of the 535 men in the fire department at issue were Black, Indian, or Mexican-American, while Blacks made up 6.44% of the city's (Minneapolis) population in 1970. 69 The court cited these statistics before concluding simply that, "statistical evidence can make a prime facie case of discrimination … the [lower] court found there was no substantial evidence to rebut the inference of discrimination based on the statistics." 70 Similarly, in Dozier v. Chupka, the court relied on the one statistic that 2.31% of the Fire Department's work force was black, compared to 9.9-18.5% of the population in the surrounding metropolitan community of Columbus. 71

These general statistics seem to be sufficient in many of the cases that share analogous features to a potential credit check lawsuit. These precedents suggest that plaintiffs bringing credit check cases will have a relatively easy time making a statistical showing. By presenting studies indicating that African Americans are more likely to have bad credit, plaintiffs can show that minorities are more likely to be harmed by hiring practices that disfavor those with bad credit. 72

63 See, e.g., Gregory v. Litton Sys. Inc., 316 F. Supp 401 (C.D. Cal. 1970), aff'd 472 F.2d 631 (9th Cir. 1972); Green v. Mo. Pac. R.R. Co., 523 F.2d 1290 (8th Cir. 1975).

64 Gregory, 316 F. Supp. at 403.

65 Green, 523 F.2d at 1293-94.

66 Id. at 1295.

67 Id.

68 Carter v. Gallagher, 452 F.2d 315, 323 (8th Cir. 1971).

69 Id.

70 Id.

71 Dozier v. Chupka, 395 F. Supp. 836, 849 (S.D. Ohio 1975).

72 See Oyama, supra note 28, at 188-89. These cases are analogous to credit check cases in that the effect of policies that exclude from employment those with criminal backgrounds is compounded on racial minorities because they are also more likely to have been arrested or convicted of a crime, just like they are more likely to have had bad credit. Also, some state legislatures have begun to provide protection to individuals with criminal records. The Johnson court addresses one possible defense argument--that wage garnishments affect the poor, and don't really primarily discriminate against racial minorities. The court responds to that potential argument by saying that while it may be true that among people on the same economic level wage garnishments have no disproportionate on African-Americans, the defense did not provide any statistics on this issue and "the question is immaterial . . . the fact that [the policy] may also discriminate against the poor white is irrelevant to our consideration under Title VII." Johnson v. Pike Corp. of Am., 332 F. Supp. 495 (C.D. Cal. 1971).

18 Geo. J. Poverty Law & Pol'y 45, *55

Page 11 of 27

But, beyond a statistical showing, a legally cognizable disparate impact claim must address whether the characteristic leading to the harm is immutable. 73 This [*57] might present a challenge to plaintiffs in credit check cases. Unlike race or gender, poor credit might be seen as a characteristic that a person created and therefore can potentially change, albeit with some difficulty. Courts tend to be less sympathetic to claims of disparate impact discrimination based on a trait that is within a person's power to change or is brought on by some action the person took. 74 However, as mentioned earlier, one's credit score and history is not entirely mutable: many credit histories contain serious errors, it is very difficult to raise a low credit score, and many bad credit scores are due to situations entirely outside of a person's control. 75 Fortunately, there are legal precedents for successful Title VII disparate impact suits being brought on the basis of mutable

characteristics. Precedents from the 1970s, dealing with employers' use of conviction and arrest records to screen job applicants (and fire employees), provide useful guidance for linking a mutable characteristic to race in a disparate impact case. Several cases, starting with Gregory v. Litton Systems, 76 have found that eliminating employees from consideration based on an arrest record (and often conviction record as well) has a disparate impact on African-Americans and is illegal under disparate impact provisions of Title VII or analogous statutes, 77 provided it does not meet the business necessity test. 78 In these cases, courts took it as a given that if use of arrests and convictions "favors white … men over black … men … defendants have used a criterion which has a discriminatory effect," without regard for the fact that getting arrested or convicted could be seen as somewhat "voluntary." 79

[*58] While the arrest/conviction cases suggest that even plaintiffs with mutable traits can state valid disparate impact claims, another category of cases, those dealing with wage garnishment, suggest that perhaps poor credit should not be viewed as a voluntary or mutable trait in the first place. In these wage garnishment cases, courts take a very narrow view of what can be considered voluntary or mutable, which could support an interpretation of poor credit as involuntary as well. 80 It is illegal under federal law to fire someone for a one-time wage garnishment. 81 In these cases, courts suggest that having one's wages garnished is not the result of voluntary action, and they

73 Kimerly Yuracko, Trait Discrimination as Race Discrimination: An Argument About Assimilation, 74 GEO. WASH. L. REV. 365, 374 (2006) ("Current antidiscrimination doctrine allows employers to engage in job irrational trait discrimination unless two independent requirements are satisfied. The trait at issue must be immutable, and it must have a disparate impact on protected group members. Only when immutability and a disparate impact are both present do courts prohibit employers from imposing trait requirements that are not job relevant.").

74 See, e.g., Garcia v. Spun Steak, 998 F.2d 1480, 1487-89 (9th Cir. 1993) (upholding employer's English-only policy as applied to bilingual employees, because the trait was one they could choose to adopt, making any disparate impact not legally cognizable); Rogers v. Am. Airlines, Inc., 527 F. Supp. 229, 231-33 (S.D.N.Y. 1981) (holding that refusing to hire someone based on a braided hairstyle is not impermissible because a hairstyle is an "easily changed characteristic"); Garcia v. Gloor, 618 F.2d 264, 270 (5th Cir. 1980) (holding that employer's English-only rule for bilingual employees does not constitute disparate impact or disparate treatment under Title VII because "[t]hereee is no disparate impact if the rule is one that the affected employee can readily observe and non observance is a matter of individual preference").

75 See supra notes 28-31 and accompanying text.

76 316 F. Supp 401 (C.D. Cal. 1970), aff'd 472 F.2d 631 (9th Cir. 1972).

77 See, e.g., Carter v. Gallagher, 452 F.2d 315, 323 (involving a claim under 42 U.S.C. §§ 1981 and 1983, which the court acknowledges contains provisions "analogous" to those of Title 7 and does not require discrimination to be "willful or intentional").

78 Gregory, 472 F.2d at 632; see also Green v. Mo. Pac. R.R. Co., 523 F.2d 1290, 1298-99 (8th Cir. 1975) (holding that refusing employment to anyone convicted of a crime other than minor traffic offense results in a disparate impact); Reynolds v. Sheet Metal Workers Local 102, 498 F. Supp 952, 971-72 (D.D.C. 1980), aff'd 702 F.2d 221 (D.C. Cir. 1981) (holding that using arrest records is not an acceptable policy when no attempt is made to validate such use); Carter, 452 F.2d at 320 (holding that

18 Geo. J. Poverty Law & Pol'y 45, *56

Page 12 of 27

therefore are sympathetic to the plight of those who had their wages garnished. 82 In Wallace v. DeBron Corp., the Eighth Circuit, in concluding that the lower court had erred in granting summary judgment for the defense, went so far as to note that:

[T]here is no evidence in the record that garnishments generally are the result of voluntary conduct undertaken with the knowledge of the consequences, and we are unwilling to take judicial notice that such is the case in the face of studies indicating that poverty is the root cause of many garnishments. 83

The Wallace court took a narrow view of the word "voluntary," refusing to recognize judgments of the court (wage garnishments) that have some connection to past voluntary action (whatever action put the person in court in the first place) as truly being "voluntary." Further, the court recognized the link between poverty and wage garnishments--neither of which are voluntary--which arguably resembles the link between poverty and a bad credit score. 84

The court in Johnson v. Pike Corp. of America also reacted sympathetically to the wage garnishment plaintiffs, finding that "discharging an employee solely because his wages have been garnished once or several times benefits no one; the employer loses an otherwise capable employee . . . the employee loses his source [*59] of income and may become dependent upon unemployment compensation or welfare; and the creditor is less likely to recover his claim." 85 The Johnson court also viewed voluntariness narrowly, and affirmed the congressional view that wage garnishments are often "saddled on a poor ignorant person who is trapped in an easy credit nightmare." 86 This type of rhetoric could lead a court to conclude that a person's poor credit is not a voluntary or mutable characteristic, which could assist a plaintiff in a credit check lawsuit. 87

Cumulatively, these precedents give hope and direction to a party seeking to make a successful prima facie challenge to employer credit check practices under Title VII's disparate impact provision. It seems that, by meeting

employers cannot utilize arrest records in hiring decisions but can use conviction records as long as they have some relation to the job); Dozier v. Chupka, 395 F. Supp. 836, 850 (S.D. Ohio 1975) (holding that use of arrest records is discriminatory).

79 Dozier, 395 F. Supp. at 850. While some cases do allow employers to use conviction records adversely against employees, it is generally not because of mutability issues but rather because courts find that a conviction has a much greater link to the job itself, rendering it more of a "business necessity." See also, e.g., Carter, 452 F.2d at 326.

80 See, e.g., Wallace v. DeBron Corp., 494 F.2d 674, 676 (8th Cir. 1974); Johnson v. Pike Corp. of Am., 332 F. Supp. 490, 494 (C.D. Cal. 1971) ("The fact that blacks and other racial minorities are so often subject to garnishment action is related to the fact that they are to a disproportionate extent from the lower social and economic segments of our society. Approximately three times the proportion of blacks as compared to whites are in the lower end of the economic scale, due in large measure to racial discrimination. Minority group members are more often in debt, are more frequently subject to questionable credit practices and harassment, and have less capacity to defend themselves."); EEOC Testimony, supra note 28 ("We noted that similar to credit check policies, people of color have also used the disparate impact framework to challenge employer policies of discharging persons whose wages have been garnished to satisfy creditors' judgments.").

81 15 U.S.C. § 1674 (2006).

82 See, e.g., Johnson, 332 F. Supp. at 496.

83 Wallace, 494 F.2d at 676.

84 Of course these are not quite the same, as one's wages are only garnished when he or she is unable to pay a court judgment, whereas one can have a bad credit score and not technically be impoverished However, the two are analogous for the purposes of analyzing voluntariness.

85 Johnson, 332 F. Supp. at 496.

86 Id. (quoting 114 CONG. REC. H688 (daily ed. Feb. 1, 1968) (statement of Rep. Sullivan)).

87 See Frank, supra note 7, at 4.

18 Geo. J. Poverty Law & Pol'y 45, *58

Page 13 of 27

the historically low threshold for evidentiary statistics, and by drawing on case law to argue that credit history is not a mutable characteristic, a plaintiff may succeed in litigation challenging employer credit checks. Unfortunately, this is not the end of the story.

2. Reasons for Caution

The recent jurisprudential picture is more complicated. As time went on, courts began to view disparate impact causes of action more narrowly, making it increasingly difficult for plaintiffs to make out their prima facie case. The seminal case of Wards Cove Packing Co. v. Antonio, 88 decided in 1989, established the modern state of the law on this issue:

The "proper comparison [is] between the racial composition of [the at-issue jobs] and the racial composition of the qualified … population in the relevant labor market." It is such a comparison-between the racial composition of the qualified persons in the labor market and the persons holding at-issue jobs-that generally forms the proper basis for the initial inquiry in a disparate-impact case. Alternatively, in cases where such labor market statistics will be difficult if not impossible to ascertain, we have recognized that certain other statistics-such as measures indicating the racial composition of "otherwise-qualified applicants" for at-issue jobs-are equally probative for this purpose. 89

In recent disparate impact cases, much of the fighting is over whether the plaintiff has introduced the right kind of statistics, or enough statistics, to prove that the policy at issue actually results in a disparate impact on minorities. 90 [*60] Courts do not have a unified approach to this question. 91 Many use the four-fifths rule, which allows a prima facie case to be made when one group's "pass" rate (for an employment test) or overall hiring rate is less than four-fifths (80%) of another group's rate. 92 On the other hand, some courts go further to look for "statistical significance"--generally at the 95% level--to ensure that the disparity is not due to random chance. 93 Of course, parties will generally advocate for the method which makes it easier for them to prove their respective cases. 94 Regardless of the method used by a court in any particular case, the analysis will involve a much more in-depth examination of statistics than the earlier cases discussed in this paper, in which courts would rule on a showing of general connections between race and poverty, or simple statistics on the number of minorities excluded by an employment policy. 95

Beyond the debate over what level of statistical significance courts should require, there are often case-specific fights over which numbers should be used for the statistical calculations in the first place. These arguments often grapple with the composition of the relevant applicant pool: for example, whether to look at everyone in the entire geographical area, or just those who meet a set of threshold qualifications. 96

88 490 U.S. 642 (1989).

89 Id. at 650-51 (quoting Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 308 (1977) (internal citations omitted).

90 Jennifer L. Peresie, Toward a Coherent Test for Disparate Impact Discrimination, 84 IND. L.J. 773, 773-74 (2009).

91 Id.

92 Id. at 773-74. Additionally, the EEOC, DOJ, and other federal governmental organizations have adopted the four-fifths rule. Id. at 781.

93 Id. at 774.

94 Id.

95 See generally supra notes 52-72 and accompanying text.

96 See Reynolds v. Sheet Metal Workers Local 102, 498 F. Supp 952, 965 (D.C. 1980) ("Plaintiffs rely on actual applicant flow data and relevant labor force figures to show that black representation in the proper pool of applicants was between 35.5-45.5%,

18 Geo. J. Poverty Law & Pol'y 45, *59

Page 14 of 27

One particular early employment case illustrates this challenge. In Reynolds v. Sheet Metal Workers Local 102, the plaintiffs utilized all three of the methods set out in Green to prove their statistical prima facie case. 97 The plaintiffs asked the court to apply a rule from a previous case that found disproportionate impact based on general population demographic data alone, but the court declined to do so because the general population's figures were not borne out by the composition of the defendant's work force. 98 The court required the additional statistical step of finding labor market data, wherein the plaintiffs presented statistics showing that the percentage of blacks in the applicant pool was higher [*61] than the percentage of blacks chosen for the apprenticeship program at issue. 99 The plaintiffs also presented actual applicant data, which is generally favored by courts. 100 After a lengthy analysis of why the plaintiffs' preferred labor pool statistics should be accepted, the court ultimately combined the labor pool data with applicant data to find that the plaintiffs had made out a prima facie case. 101

Courts also frequently require more specific statistics in housing discrimination cases, which follow a similar framework as employment discrimination cases. 102 In fact, courts "frequently borrow" legal standards and precedents from Title VII cases to apply in Title VIII cases, and vice versa. 103 In Bronson v. Crestwood, decided in 1989, the court found that an apartment complex's requirement that renters earn income of three times the rent had a disproportion-ate adverse impact on minorities and therefore violated fair housing and state laws. 104 The court allowed the plaintiffs to use the general pool of all Yonkers renters in making statistical calculations, but subjected their use of 1980 census data (from nine years earlier) to heavy scrutiny. 105

The problem with credit check cases lies not only in proving the racial composition of the qualified labor pool or the applicant pool, but also in proving that any disparity is caused by the credit checks. Even if the plaintiffs can prove a disparity between blacks who are hired and either blacks in the relevant labor pool or the applicant pool, they still have to show that the main reason for the disparity is the credit check. 106 Therefore, the increased reliance on statistics to [*62] prove the disparity will also inform the type of credit-check data plaintiffs would need to prove

and therefore in excess of the percentage of blacks chosen for the apprentice program … The defendants claim that in this case the applicant pool is either 26.67% or 25.6% black, based on the census figures for the percentage of black males, aged 18-34, in the civilian labor force or general population in the Washington SMSA. When the percentage of blacks selected for the apprenticeship program from 1971-79 (29.0%) or for 1979 (25%) is compared to these SMSA figures, the discriminatory impact on blacks appears minimal."), aff'd 702 F.2d 221 (D.C. Cir. 1981).

97 Id. at 963.

98 Id. at 964. In other words, the number of blacks in the apprenticeship program at issue was about equal to the percentage of blacks in the general population.

99 Id. at 965.

100 Id. (showing a 12.5% difference between percentage of blacks hired and whites hired, violating 80% standard set by EEOC; it was "almost significant" under the 5% standard, but the court concluded that applicant flow data didn't even show all the discrimination that was taking place).

101 Id. at 968-70 ("Other courts have recognized that the congruence of general population percentages and racial representation in the defendant's work force should not bar consideration of more relevant evidence of discrimination … [T]his refusal to be tied to general population percentages is consistent with the mandate of Title VII … Title VII protects qualified individuals from discrimination in employment. It does not require employers to hire minority groups in proportion to their representation in the population. Consequently, general population statistics lose their significance when they are no longer reasonable proxies for the applicant pool.").

102 Bronson v. Crestwood Lake Section 1 Holding Corp., 724 F. Supp. 148, 154 (S.D.N.Y. 1989).

103 See Oyama, supra note 28, at 199 (citing Trafficante v. Metropolitan Life Ins. Co., 409 U.S. 205, 209 (1972)) (relying on congressional intent in passage of Title VII to find similar congressional intent under Title VIII); Huntington Branch, NAACP v. Town of Huntington, 844 F.2d 926, 934-35 (2d Cir. 1988); Metro. Hous. Dev. Corp. v. Vill. of Arlington Heights, 558 F.2d 1283, 1288-89 (7th Cir. 1977)).

104 Bronson v. Crestwood Lake Section 1 Holding Corp., 724 F. Supp. 148, 154 (S.D.N.Y 1989).

18 Geo. J. Poverty Law & Pol'y 45, *60

Page 15 of 27

causation. In other words, if plaintiffs show a disparity between blacks hired and blacks in the applicant pool, they also will need credit information on all the applicants. This may not be so hard to get if the employer has already done the credit checks, but if plaintiffs point to a disparity between blacks in the general qualified labor pool versus those hired, then the plaintiffs would also need credit information from that qualified labor pool. The difficulty here is that most studies that have been done on credit checks have not been done on a local level. 107

Additionally, if a credit check is only one of multiple hiring requirements or "tests," it may be even more difficult for plaintiffs to tease out the effect of the credit check itself. 108 Under Title VII, a plaintiff must prove that the hiring factor or policy individually had a disparate impact, unless all the factors are "not capable of separation for analysis"--for example, if the employer has bad record keeping. 109 Given that most employers probably do not use a person's credit history as the sole hiring criteria, this may be a significant obstacle to a successful disparate impact lawsuit. Credit scores most likely correlate with other factors that employers may care about, such as class, and it will be difficult to disentangle them. Without very specific credit statistics it will be extremely difficult to prove that the credit score is the reason for any employer discrimination.

And to make the prospect of successful credit check litigation more unlikely, courts could still refuse to treat this as a cognizable disparate impact if they view credit scores as a mutable characteristic. In several recent disparate impact cases, courts have refused to recognize that a policy disproportionately burdens a group, not because the plaintiffs did not make out a sufficient statistical case, but rather because the courts felt that the policy at issue did not disproportionately burden the plaintiffs because they could choose whether or not to comply with the policy. 110 This gloss on disparate impact theory is nowhere to be found in the text of the Griggs decision, but courts are using it with increasing frequency and it could pose problems for the success of credit-check lawsuits. 111

For example, in Garcia v. Spun Steak, the Ninth Circuit ruled that an English-only rule did not disproportionately burden bilingual Spanish speaking employees because they could choose to comply with the rule of their own volition. 112 This case was different than a credit-check lawsuit in that it dealt with the "conditions, terms, or privileges" of employment, which are different than [*63] hiring criteria and a bit more nebulous. 113 The court focused on the requirement that the policy have "a significantly adverse impact" on the plaintiffs 114 (which would

105 Id. at 155 (explaining that because a significant racial migration occurred in and out of Yonkers, or between income groups, that the 1980 census data may not be accurate, but it is the most recent, and requiring plaintiffs to produce "specific estimates of the current demographics of Yonkers residents drawn from varying income groups" in order to demonstrate the disparate impact).

106 See Civil Rights Act of 1964, 42 U.S.C. § 2000e-2(k)(1)(A) (2000). In many cases involving testing procedures the causation question is clearer. See also Peresie, supra note 90, at 780 ("Causation is important because employers are only liable for disparities where plaintiffs establish that the challenged employment practice 'causes a disparate impact. It is not enough for a plaintiff to show that an observed disparity exists. The plaintiff must establish with some certainty that the challenged practice, not chance or some other factor, is the cause of the disparity.").

107 See supra notes 20-23 and accompanying text.

108 42 U.S.C. § 2000e-2(k)(1)(A) (2000).

109 42 U.S.C. § 2000e-2(k)(1)(B)(i).

110 Yuracko, supra note 73, at 374-75.

111 Id. at 374 (explaining that a trait must be immutable for a court to find disparate impact).

112 Garcia v. Spun Steak, 998 F.2d 1480, 1487-89 (9th Cir. 1993).

113 Id. at 1486 ("When the alleged disparate impact is on the conditions, terms, or privileges of employment, however, determining whether the protected group has been adversely affected may depend on subjective factors not easily quantified.").

114 Id.

18 Geo. J. Poverty Law & Pol'y 45, *62

Page 16 of 27

be more obvious in a case where someone was not hired because of a particular policy), and concluded that it was not that difficult for Hispanic employees to comply with the English-speaking rule at issue. 115 Similarly, in Rogers v. American Airlines, the court held that Rogers could not make a disparate impact claim for being forced to change her corn-row hairstyle because hairstyle is an "'easily changed characteristic,' and, even if socio-culturally associated with a particular race or nationality, it is not an impermissible basis for distinctions in the application of employment practices by an employer." 116 Yet it is not entirely clear that trait mutability is what is behind courts decisions in these cases; in fact, the decisions may have more to do with how courts view the value of the trait at issue. 117

Defendants in a credit-check lawsuit may succeed at arguing that a plaintiff's bad credit history is the plaintiff's own fault and is a mutable characteristic, unlike race or gender. However, plaintiffs may be able to successfully argue that one's credit score is not really mutable. Bad credit is very difficult to improve, especially if that bad credit is preventing a person from getting hired. While of course employers will argue it is the person's fault that they have bad credit, it is still very difficult to change once it has already happened. Additionally, not being hired because of bad credit history is different than not being allowed to speak Spanish at one's job, or being told one cannot wear a certain hairstyle. It is much easier to change a hairstyle, or even to speak English instead of Spanish if one is bilingual, than it is to change a bad credit score. And once a person is not hired for a job because of her credit history, it is unlikely that she will get that job later even if her credit improves.

[*64] In light of this, it is likely to be very difficult at this time for plaintiffs to establish a prima facie case of disparate impact. If a court follows earlier precedents, plaintiffs might be able to show the required statistical disparity because they could use more general nationwide credit check and race statistics outlined earlier in this paper, but with courts becoming increasingly rigid regarding plaintiffs statistical proof burdens, those figures now will most likely not suffice. 118 Courts simply have not been consistent in their interpretation of statistical requirements and what traits are considered mutable and it is difficult to know whether a plaintiff would even get past the first stage of such a lawsuit. This uncertainty serves to further highlight the need to pursue other avenues to curb the use of credit checks, as well as the need for more detailed social science studies to help plaintiffs make out a prima facie case if a lawsuit is brought in the future.

C. Defendant's Rebuttal--Business Necessity

Even if a plaintiff is able to make out a prima facie case, he or she can still lose if the employer can show that the policy at issue (e.g., credit checks), while resulting in a disparate impact, serves some business necessity. The Court first articulated this potential defense in Griggs, finding that a discriminatory employment test may pass

115 Id. at 1488 ("Title VII is not meant to protect against rules that merely inconvenience some employees, even if the inconvenience falls regularly on a protected class. Rather, Title VII protects against only those policies that have a significant impact. The fact that an employee may have to catch himself or herself from occasionally slipping into Spanish does not impose a burden significant enough to amount to the denial of equal opportunity. This is not a case in which the employees have alleged that the company is enforcing the policy in such a way as to impose penalties for minor slips of the tongue. The fact that a bilingual employee may, on occasion, unconsciously substitute a Spanish word in the place of an English one does not override our conclusion that the bilingual employee can easily comply with the rule. In short, we conclude that a bilingual employee is not denied a privilege of employment by the English-only policy.") (emphasis added).

116 Rogers v. Am. Airlines, 527 F. Supp. 229, 232 (S.D.N.Y. 1981).

117 Yuracko, supra note 73, at 380 ("As in Rogers, however, trait mutability does not seem to really explain or drive the court's decision … The court's effort to distinguish a no-smoking requirement from a no-beard requirement and equate it to a no- cornrows requirement seems to have far more to do with judgments about the social meaning and value of the trait being required than with concerns over trait mutability or impact.").

118 See generally Peresie, supra note 90.

18 Geo. J. Poverty Law & Pol'y 45, *63

Page 17 of 27

muster if it bears a "demonstrable relationship to successful performance of the jobs for which it was used." 119 As in the previous section, I will first outline some cases that seem to suggest that courts might not be very sympathetic to a business necessity defense regarding credit checks. I will then argue that any optimism should be tempered, as there is just as much evidence indicating that courts may potentially embrace a business necessity defense regarding credit checks.

1. Helpful Precedents

Several early disparate impact cases interpreted the business necessity defense quite narrowly. 120 In Johnson, a wage garnishment case, the defense put forth several explanations as to why the company needed to fire those who have had their wages garnished multiple times. 121 Several of the proffered reasons involved the expense and time needed to deal with the bureaucracy of wage garnishments and the "annoyance" involved in handling employees' creditors. The court had a very clear response to these justifications, holding that "the sole permissible reason for discriminating against actual or prospective employees [*56] involves the individual's capability to perform the job effectively." 122 This approach leaves no room for arguments regarding "inconvenience, annoyance, or even expense to the employer." 123 The court explained further that with the passage of Title VII, Congress intended to put a "burden" on employers in order to help end discrimination. 124

The defendant in Johnson offered one more excuse for its wage garnishment policy--that individuals who are not making money will not perform efficiently--an excuse the court found unduly speculative. 125 The defendants in Wallace attempted to make the same argument, and also failed, as they did not "provide the District Court with basic supportive facts which establish a business necessity justification as a matter of law." 126

In Gregory v. Littin Systems, Inc., an arrest records case, the court treated the defense's purported reason for its policy with disdain, finding:

[t]here is no evidence to support a claim that persons who have suffered no criminal convictions but have been arrested on a number of occasions can be expected, when employed, to perform less efficiently or less honestly than other employees. In fact, the evidence in the case was overwhelmingly to the contrary. 127

The court further noted that the County of Los Angeles had stopped asking for arrest information in employment applications, indicating its "irrelevance." 128 In Carter v. Gallagher, the court considered, inter alia, a policy that

119 401 U.S. 424, 431 (1971).

120 See Dothard v. Rawlins, 433 U.S. 321, 331 (1977); Wallace v. Debron Co., 494 F.2d 674, 677 (8th Cir. 1974); Johnson v. Pike Corp. of Am., 332 F. Supp. 490, 496 (C.D. Cal. 1971).

121 Johnson, 332 F. Supp. at 495.

122 Id.

123 Id.

124 Id. at 496.

125 Id. at 495.

126 Wallace v. Debron Co., 494 F.2d 674, 677 (8th Cir. 1974).

127 Gregory v. Littin Sys., Inc., 316 F. Supp 401, 402-3 (C.D. Cal. 1970).

128 Id. at 403.

18 Geo. J. Poverty Law & Pol'y 45, *64

Page 18 of 27

examined job applicants' conviction records in addition to arrest records. 129 The lower court had ordered the fire department to stop using arrest records and conviction records as part of its hiring process, with limited exceptions for conviction records. 130 The appellate court agreed with the lower court that arrest records have no place in the hiring process, but took a more nuanced view of convictions. The court found that a per se bar from employment due to a felony or misdemeanor conviction was too broad, but that a conviction may have some bearing on the suitability of an applicant for the job of firefighter and could therefore sometimes be considered in hiring. 131 Even though the defendants were [*66] able to convince the court of their argument on the point of convictions, this case still shows that the court is capable of making nuanced judgments about the connection, or lack of one, between a job qualification and the job's requirements and that the court is willing to narrowly tailor employers' hiring criteria.

In Green, the court continued in this same vein but went even further, concluding that the railroad's policy of denying employment to anyone convicted of a crime other than a minor traffic offense was not justified by business necessity (reversing the lower court). 132 The court was quite firm in its rejection of the business necessity defense, ultimately finding that:

We cannot conceive of any business necessity that would automatically place every individual convicted of any offense, except a minor traffic offense, in the permanent ranks of the unemployed. This is particularly true for blacks who have suffered and still suffer from the burdens of discrimination in our society. To deny job opportunities to these individuals because of some conduct which may be remote in time or does not significantly bear upon the particular job requirements is an unnecessarily harsh and unjust burden. 133

The defense had offered justifications such as fear of cargo theft, employees handling money, liability for hiring persons with known violent tendencies, and more. 134 The appellate court reversed the lower court's refusal to follow EEOC non-binding guidelines requiring empirical validation of the arrest/conviction record policies. 135 Moreover, the court was unsympathetic to the lower court's argument that it is too difficult to get data on job success of ex-convicts. 136

The business necessity defense also exists in disparate impact housing cases. Courts seem to take the same narrow view of business necessity in Title VI housing cases as they do in Title VII arrest record cases. For example,

129 Carter v. Gallagher, 452 F.3d 315, 320 (8th Cir. 1971).

130 Id.

131 Id. at 326 ("We are persuaded by defendants' argument that applicants' conviction records, at least in cases of aggravated offenses and multiple convictions, may have a bearing on the suitability of an applicant for a fire department position both from the standpoint of protecting fellow firemen and the public. The trial court in its discretion may require the defendants to submit to it for approval a rule with respect to the consideration to be given to an applicant's conviction record, which at a minimum should not treat conviction as an absolute bar to employment. We would not consider any rule giving fair consideration to the bearing of the conviction upon applicant's fitness for the fire fighter job to be inappropriate.").

132 Green v. Mo. Pac. R.R. Co., 523 F.2d 1290, 1298 (8th Cir. 1975).

133 Id.

134 Id.

135 Green v. Mo. Pac. R.R. Co., 381 F. Supp. 992, 996 (E.D. Mo. 1974).

136 Id. at 997 (citing EEOC Guidelines on Employee Selection Procedures, 29 C.F.R. Part 1607). Similarly, in Reynolds, the Court refused to entertain a business necessity defense for the use of arrest records in hiring because defendants made no "attempt had been made to validate the arrest inquiry as job related." Reynolds v. Sheet Metal Workers Local 102, 498 F. Supp. 952, 973 (D.C. 1980). Nor, the court in Green points out, had the defendants shown that there was not any "less-restrictive" alternative with a "lesser racial impact" that could not serve their business necessity just as well. Green, 523 F.2d at 1298 (citing Green, 381 F. Supp at 996).

18 Geo. J. Poverty Law & Pol'y 45, *56

Page 19 of 27

in Bronson, discussed earlier, the defendants attempted to justify their policy of turning down those renters who had Section 8 vouchers or did not meet their triple income [*67] policy. 137 They argued that these renters may not set aside enough of their Section 8 payments to pay the rent, that the Section 8 lease forces the defendant to enter into undesirable clauses, and that generally there is no law preventing the defendant from seeking to ensure that tenants will be able to pay their rent. 138 Again, the court rejected these justifications because they were merely "general conclusory assertions" and the defense offered no evidence to show that the policies were "reasonably necessary" to insure payment of rent, or that in the past the defendants had encountered losses as a result of accepting Section 8 tenants or tenants who did not meet the triple income test. 139

Similarly, in Commission on Human Rights and Opportunities v. Sullivan, the court found that Connecticut's statutory exception allowing landlords to deny applicants based on insufficient income applied only with "respect to income requirements that bear a reasonable relationship to a prospective tenant's ability to meet . . . rental obligations." 140 These housing cases are analogous to credit check cases, and suggest that courts should allow the use of credit checks only if an employer has demonstrated a reasonable connection between the policy and a person's ability to do the job for which they are applying.

These precedents should give plaintiffs challenging employers' credit check policies reason to be optimistic. The cases suggest that business necessity really means necessity, not merely avoiding an inconvenience, and that an employer must prove the connection between the policy and an employee's job performance, not merely speculate that it is so. Therefore, if a plaintiff is able to make a successful prima facie case, the defendant may not be able to show a business necessity, since employer justifications for using credit checks are often framed in very vague terms without any studies to back them up. 141

2. Reasons for Caution

In recent times however, courts simply seem unwilling to go too far in invalidating seemingly neutral employer policies, perhaps out of worry that all [*68] manner of work policies often taken for granted could come under attack. 142 Courts' discomfort with tying the hands of employers may play out in a refusal to find a disparate impact at all, or courts may use the business necessity defense as their excuse instead. Even as far back as Washington v. Davis in 1976, courts were expressing skepticism towards plaintiffs' disparate impact claims and seizing on employers' business necessity defenses to stem some of the tide of changes brought on by disparate impact

137 Section 8 is a HUD housing choice voucher program, providing low-income people with vouchers for a rent subsidy that can be applied at any apartment building where they are accepted.

138 Bronson v. Crestwood Lake Section 1 Holding Corp., 724 F. Supp. 148, 155-56 (S.D.N.Y. 1989).

139 Id. at 156.

140 Comm'n on Human Rights & Opportunities v. Sullivan, 939 A.2d 541, 551 (Conn. 2008) (explaining that the statutory exception in Connecticut allowing realtors to refuse to rent to those with "insufficient income" can only be used when the income requirement bars a "reasonable relationship" to tenants' ability to meet rental obligations).

141 See also Dothard v. Rawlins, 433 U.S. 321, 343-44 (1977) (rejecting employers' "common sense" argument that prison guards must be strong to justify hiring criteria that roughly measured strength); El v Se. Pa. Transp. Auth., 471 F.3d 232, 244-45 (3d Cir. 2007) ("The lesson is that employers cannot rely on rough-cut measures of employment-related qualities; rather they must tailor their criteria to measure those qualities accurately and directly for each applicant.").

142 Michael Selmi, Was the Disparate Impact Theory a Mistake?, 53 UCLA L. REV. 701,751 (2006) (noting, regarding the case of pregnant women, "if the disparate impact theory were applied with rigor to policies that adversely affect pregnant women or women with childrearing responsibilities, it could conceivably invalidate many central, and common, employment policies, including routine work hours, most leave policies, and mandatory overtime").

18 Geo. J. Poverty Law & Pol'y 45, *66

Page 20 of 27

lawsuits. 143 After Wards Cove Packing Co. v. Antonio in 1991, when the Court tried to replace the business necessity standard with something obviously less stringent ("business justification") 144 and shift the burden of proof to the employee, Congress responded by passing the Civil Rights Act of 1991. 145 The Civil Rights Act rejected parts of Wards Cove, placing the burden of proof back on the employer and stating that for selection procedures, the practice must bear a "significant relationship to successful performance of the job" and "demonstrable evidence is required." 146 In essence, Congress returned business necessity jurisprudence to what it was before Wards Cove, but courts have still found ways to lessen the burden on defendants.

Indeed, in recent arrest records cases, courts have broadly interpreted the business necessity defense. For example, in El v. Southeastern Pennsylvania Transportation Authority, a case from 2007, the appellate court upheld summary judgment for the defendant, holding that it was acceptable policy to terminate an African-American para- transit driver for a forty-year-old murder conviction. 147 The court reiterated that an employer must present real evidence that the challenged criteria "measure[s] the person for the job and not the person in the abstract" but also held that a policy need not be "perfectly tailored." 148 The court "require[s] that employers show that a discriminatory hiring policy accurately--but not perfectly--ascertains an applicant's ability to perform successfully the job in question." 149 Additionally, the court held that "Title VII allows the employer to hire the applicant most likely to perform the job successfully over [*69] others less likely to do so." 150 The court then framed the question as one of an employer's need to "manage[] [the] . . . risk" that an employee will "endanger the employer's patrons," and eventually held that even a bright-line policy against hiring people with convictions is permissible if it can "distinguish between applicants that do and do not pose an unacceptable level of risk." 151 In granting the defendant's motion for summary judgment, the court almost blamed the plaintiffs (and the EEOC) for not providing even a "scintilla" of support for their position. 152 What is frustrating about this decision is that, while it purports not to be putting the burden on the employee, it does exactly that by focusing on what evidence the plaintiff should have shown to rebut the defendant's assertion of business necessity. 153

Additionally, the court opened the door to more successful business necessity defenses by holding that all a policy need do is distinguish between different levels of risk. In El, the court distinguished between employer policies that can be justified on the basis of potential harm to fellow employees or customers (para-transit users) versus policies

143 Washington v. Davis, 426 US 229, 248 (1976) ("A rule that a statute designed to serve neutral ends is nevertheless invalid, absent compelling justification, if in practice it benefits or burdens one race more than another would be far reaching and would raise serious questions about, and perhaps invalidate, a whole range of tax, welfare, public service, regulatory, and licensing statutes that may be more burdensome to the poor and to the average black than to the more affluent white.").

144 Wards Cove Packing Co. v. Antonio, 490 U.S. 642, 659 (1989).

145 42 U.S.C.A. §§ 1981-2000e (1991).

146 42 U.S.C.A. § 2000e-2.

147 El v Se. Pa. Transp. Auth., 471 F.3d 232, 235 (3d Cir. 2007).

148 Id. at 240 (citing Dothard v. Rawlins, 433 U.S. 321, 332 (1977) (citing Griggs v. Duke Power Company, 401 U.S. 424, 436 (1971))).

149 El, 479 F.3d at 242.

150 Id.

151 Id. at 244-45.

152 Id. at 248 (granting the EEOC policies "Skidmore" deference and finding them too "terse" to be helpful).

153 Id.

18 Geo. J. Poverty Law & Pol'y 45, *68

Page 21 of 27

that can be justified only by other employer needs. 154 The court noted that previous business necessity cases were not particularly instructive since the consequences in this case involved potential harm to other employees as opposed to hiring someone with a lower IQ or someone who did not pass a test. 155 The court then tried to draw the two types of cases together by using the "management of risk" standard. 156 In other words, it does not to matter how much, or whether, the policy at issue actually has any connection to an individual's ability to do the job, but rather it matters whether the policy has levels or gradations built in to better manage risk. This analysis implicitly accepts managing risk as a valid business necessity--no matter what the purported risk maybe. It is unclear which category credit checks would fall into--employers could argue that employees with bad credit who end up stealing could be a risk to customers, but certainly they are not a violent risk the way that someone with a criminal past might be.

In a similar vein, some other cases seem to suggest that the magic ingredient needed to prevail on a business necessity defense is having a policy with sufficient objective criteria. 157 In Lanning v. Southeastern Pennsylvania Transportation Authority, an earlier companion case to El, the district court originally found that the police department at issue had not shown that the fitness test that resulted in a disparate impact on women was a business necessity because the [*70] department did not demonstrate that the "cutoff measures the minimum qualifications necessary for successful performance of the job in question." 158 This decision, however, was later reversed after the court found that the minimum standards set by Southeastern Pennsylvania Transportation Authority (SEPTA) were appropriate given the public safety issues at stake. 159 This result could potentially both help and hurt credit-check lawsuits. It could help because it might be hard for employers to explain why a credit score one point lower than the cutoff makes an employee unfit for a the job, making it harder for employers to justify their cutoffs under this reasoning. But it also seems that when safety becomes a concern, courts give defendant employers more leeway. Even though Choicepoint, a background-checking firm, stated that "credit has not turned out to be a good predictor of workplace theft," 160 courts may very well accept the argument that credit is, in fact, a good predictor of such behavior.

The state of disparate impact jurisprudence, particularly with regard to business necessity, is not settled. It is difficult to predict how courts might view an argument that it is necessary for an employer to carry out credit checks on potential employees because the outcome may depend on the exact fact pattern that comes before the court, as well as the precedents in that jurisdiction. Yet these precedents do suggest that any credit check litigation may face a formidable obstacle in the form of the employer's business necessity defense. In the media and in the non-legal arenas of argument over this, employers most common line of defense has been a de facto business necessity defense, with employers arguing that people with poor credit are untrustworthy workers, or more specifically, that

154 Oyama, supra note 28, at 209-10.

155 El, 479 F.3d at 243.

156 Id. at 244.

157 See, e.g., Lanning v. Se. Pa. Transp. Auth., 181 F.3d 478 (3d Cir. 1999).

158 Id. at 489.

159 Lanning v. Se. Pa. Transp. Auth., 308 F.3d 286 (3d Cir. 2002) ("It would clearly be unreasonable to require SEPTA applicants to score so highly on the run test that their predicted rate of success be 100%. It is perfectly reasonable, however, to demand a chance of success that is better than 5% to 20%. In sum, SEPTA transit police officers and the public they serve should not be required to engage in high-stakes gambling when it comes to public safety and law enforcement. SEPTA has demonstrated that the cutoff score it established measures the minimum qualifications necessary for successful performance as a SEPTA officer.").

160 Why Credit History Employment Inquiries Matter, CONSUMERIST.COM (June 15, 2009), http://consumerist.com/2009/06/why-credit-history-employment-inquiries-matter.html.

18 Geo. J. Poverty Law & Pol'y 45, *69

Page 22 of 27

they cannot be trusted around money, retail merchandise, or matters of national security (because they can more easily be bribed). 161 Given the courts' approach to risk management and its deference to public safety concerns, these are all defenses that could potentially be validated by a court following the precedents outlined above.

IV. POTENTIAL LEGISLATIVE SOLUTIONS

While there are reasons to think that a suit against employer use of credit checks could succeed, there also are a number of reasons to believe that success [*71] would be quite difficult. This confusion and uncertainty does not provide an ideal atmosphere for litigation aimed at reducing the use of credit checks. Given the complicated state of disparate impact jurisprudence, and given courts' increasing skepticism about claims analogous to those involving credit checks, a political response may be more successful. I will begin by outlining my proposal for ideal credit check legislation. I will then evaluate the legislative responses that have been enacted or are under consideration at the state and federal level and will analyze how they measure up to this ideal.

In determining what an ideal credit check bill would look like, I do take political feasibility into account to some extent; yet clearly the details of the type of bill that could actually be passed would vary widely from state to state. In general, the ideal bill would prohibit employers from procuring credit reports for the purpose of screening job applicants or deciding whether to fire a current employee (any "adverse employment decision"). 162 The bill also would inevitably contain several exceptions to the general rule, for both political and public policy reasons. It is clear from the survey of bills that have passed or are being debated that no bill could survive without at least some exceptions and, beyond politics, there are situations in which a person's credit may have a connection to the person's ability to perform a certain job.

There are two categories of exceptions that are necessary in this type of legislation. One exception is for positions in which the employee would have access to significant cash or other financial or business assets. The other important exception is for positions that have access to confidential or secure information. The rationale for these exceptions is that a person with bad credit cannot be trusted around money or sensitive information that could be exchanged for money. 163 Though this argument is somewhat of a slippery slope, narrowly crafted exceptions could protect employers from some of the more serious potential consequences of hiring someone with bad credit, while still protecting employees with bad credit from discrimination in most jobs.

These exceptions must be worded precisely in order to ensure that they do not become giant loopholes. The bill would refrain from using phrases such as "job related," "business necessity," or "bona fide occupational qualification" without defining them, as those phrases can be interpreted in wildly different ways by the courts and can allow for too much discretion. 164 An ideal bill would define the minimum amount of cash or assets that an employee would have to control in order for an employer to be allowed to run a credit check for an employee in that position. The bill would also in some way define "control." As for the exception for employees with access to confidential and secure information, the bill also should clearly define the scope of this exception and restrict it to information that [*72] requires a security clearance or information that has been entrusted only to certain high-level employees. An exception covering anyone exposed to confidential information could include any retail employee exposed to consumer information and would be too broad.

Ideal legislation would not contain an exception for all managerial positions because such an exception, depending on how courts interpret it, could become a very large loophole and leave many employees unprotected. The link between one's credit and one's ability to perform as a manager seems tenuous. Additionally, the bill would not

161 Martin, supra note 24.

162 Language from the proposed federal "Equal Employment for All Act." Equal Employment for All Act, H.R. 3149, 111th Cong. (2010).

163 See supra note 24 and accompanying text.

164 See supra Part III.C.

18 Geo. J. Poverty Law & Pol'y 45, *70

Page 23 of 27

exempt any category of employer from the definition of employer, even banks. 165 There is no reason to exempt banks entirely from the definition of employer under the bill, since the exception for access to financial and confidential information would cover the appropriate bank employees.

While narrow exceptions are ideal, legislators should take care not to make them unduly narrow. The policy problem with crafting exceptions that are too narrow is two-fold: first, a bill with such exceptions will not be able to pass in any legislature; second, unduly narrow exceptions may not address real potential threats resulting from hiring people with bad credit for certain types of positions. For example, the potential risk in giving employees access to money or confidential information without a credit check is a serious one and should not be limited only to a certain class of such employees, such as managers, as some of the bills in state legislatures propose to do. 166

Additionally, the ideal bill would not limit the liability of employers who violate the credit check law. 167 However, it may be possible to assuage employers' fears about being held liable for employee actions that arguably could have been prevented if the employers had performed credit checks. This fear does not seem valid because it is unlikely that an employer would be found liable for not performing an illegal credit check, but perhaps legislators may find it useful to explicitly address liability in credit check legislation in order to minimize these types of fears. Ideally, a model bill would explicitly provide for a civil damages remedy and attorney fees for prevailing plaintiffs, as this would encourage those whose rights have been violated to bring a suit where they otherwise might not do so. 168 While it is true that such a bill would increase employers' risk of litigation, that risk is not so great as to outweigh the need for our government to protect the vulnerable.

Several states recently have considered, or are still considering, various legislative proposals to limit employers' ability to use credit histories for hiring or firing purposes. Washington, Hawaii, Illinois, and Oregon have passed credit [*73] check laws, with Washington passing its law on April 18, 2007, Hawaii on July 15, 2009, Oregon on July 1, 2010, and Illinois on August 10, 2010. 169 Legislators in many other states, including Connecticut, Georgia, Illinois, Indiana, Maine, Maryland, Michigan, Missouri, New Jersey, New York, Ohio, Oklahoma, Pennsylvania, South Carolina, Vermont and Wisconsin, have considered or are considering credit-check legislation. 170 California Governor Arnold Schwarzenegger vetoed credit check legislation the California assembly passed in 2009, and it seems unlikely that the legislature will be able to override that veto. 171 On the federal level, Representative Scott Cohen introduced a bill on July 10, 2009, called the "Equal Employment for All Act" that would amend the Fair Credit Reporting Act to prohibit the use of consumer credit checks for certain employment

165 For example, the Illinois bill does this. See infra notes 179-82 and accompanying text.

166 See A.B. 943, 2009 Leg., Reg. Sess. (Ca. 2009); Equal Employment for All Act, H.R. 3149, 111th Cong. (2010), infra notes 174-85 and accompanying text.

167 See infra notes 192-93 and accompanying text.

168 See Employee Credit Privacy Act, H.B. 4658, 96th Gen. Assemb., Reg. Sess. (Ill. 2010).

169 S.B. 5827, 60th Leg., Reg. Sess. (Wash. 2007); H.B. 31, 25th Leg., Reg. Sess. (Haw. 2009); Employee Credit Privacy Act, H.B. 4658, 96th Gen. Assemb., Reg. Sess. (Ill. 2010); O.R.S. § 1 (2010).

170 H.R. 5521, 2009 Leg., Jan. Sess. (Conn. 2009); A.2067, 2009 Leg., Reg. Sess. (N.Y. 2009); A.B. 367, 2009-10 Leg., Reg. Sess. (Wis. 2009); see also SOCIETY FOR HUMAN RESOURCE MANAGEMENT, NEW OREGON CREDIT CHECK LAW WARRANTS NATIONWIDE REVIEW OF EMPLOYER PRACTICES (2010), available at http://www.shrm.org/LegalIssues/StateandLocalResources/Pages/NewOregonCreditCheckLaw.aspx; Martin,supra note 24 (noting that bills introduced in California, Maryland, and Connecticut "have been stalled amid opposition from credit bureaus and other businesses"); Christine Lagorio, States Propose Limiting Credit Checks by Employers, INC. MAGAZINE (Mar. 1, 2010), http://www.inc.com/news/articles/2010/03/credit-check-legislation.html.

171 Bill No. 943 passed on Sept. 8, 2009 and Governor Schwarzenegger vetoed it. California Governor Veto Message, Bill AB 2918, Sept. 30, 2008.

18 Geo. J. Poverty Law & Pol'y 45, *72

Page 24 of 27

purposes. 172 The bill has 34 co-sponsors and was referred to the House Committee on Financial Services in July 2009. 173

Illinois' credit check legislation is similar in many ways to the ideal legislation described above. The Illinois Employee Credit Privacy Act states that the prohibition against use of credit checks does not apply if credit history is an established "bona fide occupational requirement" of a particular position, but, most importantly, the bill states unequivocally that "a satisfactory credit history is not a bona fide occupational requirement unless at least one of the following circumstances is present." 174 These circumstances include positions involving: "custody" or "unsupervised access" to cash or assets of $ 2500 or more, "signatory power" over business assets of $ 100 or more per transaction, or "access to personal or confidential information, financial information, trade secrets, or state or national security information." 175 These exceptions are very specific and thorough, covering many possible situations where credit checks indeed might be necessary, while attempting to limit their application as much as possible, and making the Illinois bill very close to the ideal bill in terms of how narrowly it defines its exceptions.

[*74] This Illinois bill also is ideal in its treatment of enforcement. It specifically addresses enforcement and is again very protective of employees, allowing any person injured by a violation of the bill to pursue civil damages as well as injunctive relief. 176 The bill also provides for reasonable costs and attorney's fees to be paid to a successful plaintiff. 177 Additionally, the Illinois bill prohibits any retaliation or discrimination against an employee who files a complaint or participates in an investigation under the bill. 178

However, there are some aspects of the Illinois bill that are not perfect. First, the Illinois bill exempts all bank and financial institutions from the definition of employers, meaning that they do not have to comply with the bill's provisions for any of their employees. 179 This is overly broad and undercuts the protections provided in the specific descriptions of the type of positions for which a credit check can be considered a "bona fide occupational qualification." 180 Second, the definition of "personal and confidential information" is very broad. 181 Lastly, any "managerial" position is exempted from the credit check prohibition, and that term is defined broadly and vaguely as a position that involves "setting the direction or control of the business." 182

The vetoed California bill is similar to the Illinois bill, but it has fewer and narrower exceptions, which may be why it was never enacted. The proposed legislation defined "job-related" as involving "access to money, other assets, or

172 Equal Employment for All Act, H.R. 3149, 111th Cong. (2010).

173 H.R. 3149: Equal Employment for All Act, GOVTRACK.US (last visited Nov. 15, 2010) http://www.govtrack.us/congress/bill.xpd?bill=h111-3149.

174 Employee Credit Privacy Act, H.B. 4658, 96th Gen. Assemb., Reg. Sess. (Ill. 2010) [emphasis added].

175 Id.

176 Id.

177 Id.

178 Id.

179 Id.

180 Id.

181 Id. ("Sensitive information that a customer or client of the employing organization gives explicit authorization for the organization to obtain, process, and keep; that the employer entrusts only to managers and a select few employees; or that is stored in secure repositories not accessible by the public or low-level employees.").

182 Id.

18 Geo. J. Poverty Law & Pol'y 45, *73

Page 25 of 27

confidential information." 183 By narrowing the precise circumstances under which the California legislature concluded that a credit history could possibly be job-related, the proposed bill removed the discretion left in other state bills, discussed below, that allowed judges to interpret "job-related" however they saw fit. By defining that term in the statute, the California legislature would potentially have made it much harder for employers to assert a "job- related" need for prospective or current employee credit checks, as employers would have had to prove that the position for which they were hiring involved access to money, other assets, or confidential information. This exception for positions involving access to assets or confidential information addressed one of the concerns of the bill's opponents--that this bill would lead to money theft or identity theft or [*75] would cause security concerns in general. 184

However, in order for a credit check to factor into an employer's hiring decision under the California legislation, the position in question would have had to involve not only access to assets or confidential information, but also would have to have been managerial in nature or a position in city/county government or law enforcement. 185 Thus, the exceptions in the California bill would have been significantly narrower than those in the Illinois bill, which has separate, independent exceptions for managerial positions and positions having access to financial information (thereby allowing more leeway for employers), and this may explain why Governor Schwarzenegger twice vetoed the California bill. 186

The proposed federal legislation is also similar to the Illinois bill and the proposed California bill. The proposed federal bill allows employers to use credit information only when the potential job requires national security or FDIC clearance, is a position with a state or local government agency that currently requires a credit check of its applicants, or involves a "supervisory, managerial, professional or executive position at a financial institution." 187 These exceptions are quite specific like those in the Illinois bill; however, because this bill will need to pass through the House and the Senate, it will be susceptible to significant weakening of its provisions if it is ever passed. At the same time, the exception for financial institutions is narrower than Illinois' exception and similar to the California exception: it exempts only managerial positions at financial institutions. This exception strikes an ideal balance between reducing harmful effects of the credit check and maintaining financial security of institutions. It acknowledges that not every employee at a financial institution, nor every managerial employee, has a job in which the employee's performance can be judged by his or her credit record; yet, because the risk of theft of financial and confidential information is at its highest where someone is a manager at a financial institution, such an exception is appropriate.

Other laws have exceptions that are not ideal because they are too broad. In Washington, for example, the employer must show that credit checks are "substantially job related," while in Hawaii, employer credit checks must "directly relate to a bona fide occupational qualification." 188 "Substantially job related" is extremely vague, and ties the bill to changing judicial interpretations [*76] of that term. 189 "Bona fide occupational qualification," the term used in Hawaii's bill, refers to situations in which an employer is permitted to pursue an admittedly discriminatory hiring policy because it is intrinsically related to the qualifications for the job (for example, hiring only women to

183 A.B. 943, 2009 Leg., Reg. Sess. (Cal. 2009).

184 See Floor/Committee Analysis of California AB 943, Senate Committee on Labor and Industrial Relations, AROUNDTHECAPITOL.COM (June 24, 2009), http://www.aroundthecapitol.com/billtrack/analysis.html?aid=30756.

185 A.B. 943, 2009 Leg., Reg. Sess. (Cal. 2009). This is particularly interesting since Hawaii's bill actually exempts hiring of managerial/supervisory employees from the entire bill-clearly there is some thought that the job of a managerial employee has some greater relation to credit history than lower level jobs do. Perhaps the thought is that managerial employees can do more harm if they are dishonest.

186 California Governor Veto Message, Bill A.B. 2918, Sept. 30, 2008.

187 Equal Employment for All Act, H.R. 3149 (2010).

188 S.B. 5827, 60th Leg., Reg. Sess. (Wash. 2007); H.B. 31, 25th Leg., Reg. Sess. (Haw. 2009).

189 See supra Part III.C.

18 Geo. J. Poverty Law & Pol'y 45, *74

Page 26 of 27

serve as Hooters waitresses). 190 This formulation of the exception still ties it to changing interpretations of that term by courts. And without any decisions interpreting Hawaii's or Washington's laws, it is difficult to know exactly how state courts will interpret the language of the bills, and the exceptions could potentially swallow the protections in the bills. 191

These other laws and pending legislation are also weak in their treatment of the enforcement issue. Many of them don't specifically address enforcement, most likely because the remedy for a violation of the bill would have to come in the form of a lawsuit. 192 Connecticut's bill makes it especially difficult to hold an employer liable, as it provides that a person shall not be held liable for a violation of the bill if "he or she showed, by a preponderance of the evidence, that, at the time of the alleged violation, he or she maintained reasonable procedure to ensure compliance with this section." 193 This gives the employer an out in the case of a mistake, but presumably if the employer intended to check an applicant's credit and use a credit report in a situation prohibited by the bill, then the employer would still be liable. The problem is that an employer might be able to argue successfully that he or she thought that the credit history was job-related even if a court might not find it to be job-related, thereby creating a significant loophole.

Most of the bills also have various types of disclosure requirements detailing when an employer must disclose to potential employees that a credit history has been requested or used. 194 Yet these requirements offer no new protection, as the same disclosure requirements exist now in the Fair Credit Reporting Act and, as explained earlier, are impossible to enforce because individuals to whom the required disclosure is not made will never learn that fact that a credit check has been conducted. 195 Wisconsin's proposed bill further weakens the disclosure requirements by allowing the requirement to be fulfilled by a written statement [*77] contained in employment application materials--materials that many people do not read carefully. 196

The Illinois bill, the vetoed California bill, and the pending federal legislation come closest to meeting the standards of an ideal credit check statute, although none of them meet all of the criteria. The legislation enacted and under consideration in a number of other states falls even more seriously short of the ideal.

V. CONCLUSION

The increased use of credit checks by employers to screen job applicants (and current employees) is a serious public policy problem that needs to be addressed. In light of the current economic climate, any measure that makes it more difficult for people to get jobs should be scrutinized carefully and should be used only if absolutely necessary. Credit checks are especially harmful, as people often get trapped in a vicious cycle in which they cannot get work and then cannot pay their bills, causing their credit to worsen, which then further hurts their job prospects,

190 See Latuga v. Hooters, Inc., No. 93 C 7709, 1994 WL 113079 (N.D. Ill. Apr. 1, 1994).

191 See supra Part III.C.

192 See S.B. 5827, 60th Leg., Reg. Sess. (Wash. 2007); H.B. 31, 25th Leg., Reg. Sess. (Haw. 2009).

193 H.R. 5521, 2009 Leg., Jan. Sess. (Conn. 2009).

194 H.R. 5521, 2009 Leg., Jan. Sess. (Conn. 2009) ("Whenever employment . . . is denied . . . because of information contained in a consumer credit report . . . the use of the . . . report shall so advise the consumer against whom the adverse action has been taken"); A.2067, 2009 Leg., Reg. Sess. (N.Y. 2009) (falling short of having a disclosure requirement, but requiring that if employee gives permission for credit check to be looked at, the employee has to sign a form); A.B. 367, 2009-10 Leg., Reg. Sess. (Wis. 2009) (containing nothing on the subject); S.B. 5827, 60th Leg., Reg. Sess. (Wash. 2007) ("A clear and conspicuous disclosure has been made in writing to the consumer before the report is procured . . . that a consumer report may be obtained for purposes of considering the consumer for employment . . . written statement contained in employment application materials"); Hawaii and California have nothing.

195 See supra notes 12-19 and accompanying text.

196 A.B. 367, 2009-10 Leg., Reg. Sess. (Wis. 2009).

18 Geo. J. Poverty Law & Pol'y 45, *76

Page 27 of 27

and so on. And given the current housing mortgage crisis, this vicious cycle is even more prevalent. Then, of course, there is the racial bias inherent in the use of credit checks--these checks have a larger impact on minority groups, who tend to have worse credit (and more difficulty finding jobs), and therefore are more likely to get caught in that vicious cycle.

Because the link between credit checks and job performance is so tenuous, employers in general do not seem to present very persuasive arguments in support of their use (with some specific exceptions, of course). Bad credit can occur for many reasons, including some that are beyond a person's control, and most have no connection to job performance. There are some circumstances in which an employer might legitimately need to look at the credit history of an applicant but, as discussed above, those situations can be dealt with by exceptions in a bill that otherwise prohibits credit checks. 197

Given their disparate impact on minorities and the unconvincing justifications for their use, credit check policies may violate Title VII disparate impact provisions. While courts have not yet held that they do so, and may not ever so hold, the litigation approach is still something that should be considered in the public policy debate. We should be worried that these credit check policies violate the spirit of disparate impact provisions, even if courts may not find an actual violation. The entire purpose of bringing disparate impact claims, as the Court laid out in Griggs and elaborated and expanded upon in future decisions in the 1970s, was to remove barriers to minority advancement, which is why the courts in such cases now require only that a policy have a discriminatory effect.

[*78] As explained in this paper, the policy/legislative arena presents a greater chance for success in lessening the use of credit checks than litigation. There are too many obstacles to a successful Title VII disparate impact suit regarding credit checks, including statistical issues of proof, courts' attitude towards mutability, and relaxed standards for defendants to prove business necessity. On the other hand, there has been moderate success in the legislative arena that could potentially be replicated across the country. Obviously, state-by-state legislation is more time-consuming in some ways than litigation (depending on whether it is at the state or federal level), and federal legislation would be ideal. Yet state and federal legislatures may be open to arguments that courts may not be able to consider under the strictures of Title VII as it has recently been interpreted. A state-by-state approach also allows each state to craft slightly different laws, instead of having a credit check ban imposed by judicial fiat. Additionally, legislation can address the entire credit check problem at once, whereas a court decision would be limited to the specific facts of the case. Again, this is not to say that litigation should not continue, but rather that I believe that litigation could be more effective in conjunction with legislation.

In addition to legislation, it is important for administrative agencies such as the EEOC to weigh in on this debate. This might even help goad legislatures into action. Increased social science research is needed to help with both litigation and legislation by providing more locally specific statistics to bolster arguments by litigants and by proponents arguing in favor of credit-check legislation. Public awareness is just beginning to be raised on this issue, but the more the problem becomes publicized and personal, the more likely legislatures will be pressured to pass laws to fix it. If it is presented in the right light, public sympathy may very well be on the side of passing such laws.

Georgetown Journal on Poverty Law & Policy Copyright (c) 2010 Georgetown Journal on Poverty Law & Policy Georgetown Journal on Poverty Law & Policy

End of Document

197 See supra Part IV.

18 Geo. J. Poverty Law & Pol'y 45, *77