Oregon Equal Pay Act of 2017

Oregon Bans Employers from Asking Job Applicants About Prior Salary.

Comment on: 2017 Or. Laws ch. 197, H.B. 2005 (to be codified in scattered sections of Or. Rev. Stat. )

• Volume 131
• Issue 5
• March 2018

In 2016, women working full time earned on average nineteen percent less than men. 1 For decades, policymakers have tried to narrow this “pay gap.” 2 Between 1980 and 2016, the gap fell by one-half, 3 but substantial differences in earnings remain. States and cities have begun to experiment with new equal pay laws to try to close the gap. 4 Recently, Oregon enacted the Oregon Equal Pay Act of 2017, 5 which bans employers from asking about or using prior salary in hiring decisions. 6 The Oregon Equal Pay Act of 2017 goes further than the federal Equal Pay Act of 1963 7 (EPA) and other state and local laws in preventing employers from justifying an otherwise unlawful pay disparity on the basis of prior salary. However, the complete ban on asking about or using prior salary carries a risk — by limiting the information available to employers, it may encourage discrimination based on gender.

The gender pay gap in Oregon — at twenty-one percent 8 — is slightly higher than the national average. To try to close the gap, Democrats in the Oregon House of Representatives introduced H.B. 2005, 9 a bill to ban employers from asking about or using prior salary in hiring decisions. 10 The ban aims to address the following concern: if a woman is initially paid less than a man for comparable work in her first job and subsequently transitions to a second job in which her new employer bases her pay on her previous salary, then her lower pay will persist. A floor letter introduced by Oregon State Representative Ann Lininger summarized this rationale: “[The ban] will help break the cycle of pay disparity that traps workers from aspiring to earn higher wages because they have previously worked at a lower wage.” 11

Initially, the proposal was contentious. Republicans opposed the House bill because it interfered with otherwise innocent business practices and lacked adequate protection for employers. 12 The bill passed the House along party lines, with only one Republican voting in favor. 13 However, the Senate produced a separate version of the bill that attracted bipartisan support, in part because it included a partial safe harbor for employers that conduct an “equal pay analysis.” 14 The Senate bill passed both chambers of the Oregon legislature unanimously. 15 Its Republican cosponsor, Oregon State Senator Tim Knopp, opined that “the next generation of women . . . will earn more for their entire working career because of our efforts here today.” 16 On June 1, 2017, Governor Kate Brown, a Democrat, signed the Act into law. 17

The Act regulates employers’ acquisition and use of salary history information. Employers may not “seek the salary history of an applicant . . . from the applicant . . . or a current or former employer of the applicant”; 18 they may not “[s]creen job applicants based on current or past compensation”; 19 and they may not “[d]etermine compensation for a position based on current or past compensation.” 20 An employer found to have violated these provisions is potentially liable for back pay and for compensatory and punitive damages. 21 However, the Act contains a safe harbor provision. If an employer has completed an “equal-pay analysis” — an internal audit, essentially — within three years before the complaint, eliminates the pay differential for the plaintiff, and makes “substantial progress toward eliminating wage differentials for the protected class asserted by the plaintiff,” a court must grant a motion by the employer to limit an award to two years of back pay and reasonable attorney fees and cannot award further compensatory or punitive damages. 22

The Act’s implementation is staggered. Most of its provisions, such as the ban on determining pay based on current or past pay, do not take effect until January 1, 2019. 23 The ban on seeking information about salary history took effect on October 6, 2017, 24 but civil actions for violating the ban may not begin until January 1, 2024. 25 Until then, enforcement of the ban is up to the Commissioner of the Oregon Bureau of Labor and Industries. 26

The Oregon Equal Pay Act of 2017 is not the first of its kind. It is part of a proliferation of equal pay laws that take aim at the use of prior salary in hiring decisions. In 2016, Massachusetts became the first state to enact such a law. 27 Within one year, California, Delaware, Oregon, Puerto Rico, New York City, and Philadelphia followed. 28 Salary history legislation is under consideration in at least twenty states and the District of Columbia, 29 and on May 11, 2017, members of the U.S. House of Representatives introduced the Pay Equity for All Act of 2017, 30 a bill to amend the Fair Labor Standards Act of 1938 31 to disallow employers to “request or require . . . that a prospective employee disclose previous wages or salary histories.” 32

By preventing employers from raising a prior salary defense, the Oregon Equal Pay Act of 2017 strengthens the hands of discrimination claimants. The Act goes further than the EPA, as interpreted by the circuit courts of appeals, and other recent state and local equal pay laws in preventing employers from justifying an otherwise unlawful pay disparity on prior salary. However, the overall effect of the new ban on asking about or using prior salary is uncertain. Although similar bans have been described as a “sea change,” 33 as a practical matter they may not alter the information available to or used by employers. Or worse, they may risk doing so in a way that encourages gender-based discrimination.

The EPA prohibits discrimination “on the basis of sex by paying wages to employees . . . at a rate less than the rate at which [the employer] pays wages to employees of the opposite sex . . . for equal work on jobs the performance of which requires equal skill, effort, and responsibility, and which are performed under similar working conditions.” 34 An employer may assert as an affirmative defense to a claim under the EPA “(i) a seniority system; (ii) a merit system; (iii) a system which measures earnings by quantity or quality of production; or (iv) a differential based on any other factor other than sex.” 35 Courts disagree about the extent to which prior salary is a permissible “factor other than sex.” In Kouba v. Allstate Insurance Co., 36 the Ninth Circuit held that “the Equal Pay Act does not impose a strict prohibition against the use of prior salary,” 37 but an employer wanting to rely on it must assert “an acceptable business reason.” 38 The Seventh and Eighth Circuits agreed that employers may use prior salary but declined to adopt the “‘acceptable business reason’ requirement.” 39 The Tenth and Eleventh Circuits, meanwhile, have held that “[t]he EPA . . . precludes an employer from relying solely upon a prior salary to justify pay disparity,” whatever the reason. 40 In 2017, the Ninth Circuit revisited the issue in Rizo v. Yovino, 41 now vacated pending rehearing en banc, in which it endorsed its decision in Kouba and clarified that it “did not draw any distinction between using prior salary ‘alone’ and using it in combination with other factors.” 42

By prohibiting employers from relying on prior salary at all, the Oregon Equal Pay Act of 2017 goes further to stop the prior salary defense than have the EPA, as interpreted by the circuit courts of appeals, and other recent state equal pay laws. Other states responded more narrowly or only in part. For example, California amended its equal pay law to specify that “[p]rior salary shall not, by itself, justify any disparity in compensation.” 43 Delaware banned salary history inquiries but continues to allow employers to use salary history information if applicants volunteer it. 44 But the Oregon Equal Pay Act of 2017 appears to bar wholesale any consideration of prior salary. The Act provides that an employer may not “[d]etermine compensation . . . based on current or past compensation” 45 and does not make any explicit exception for voluntary disclosures or for consideration of prior salary in conjunction with other factors. Thus, the Act seems to wipe out the prior salary defense for employers facing claims of unlawful discrimination.

There are, however, reasons to be skeptical that the Act will alter hiring practices in a way that reduces gender differences in earnings. First, salary history bans may not alter the information available to employers or whether they use it. The ban directly affects only job applicants with a salary history. That is, for applicants looking for their first job, or for employees who never change jobs, the rule does not change the information available to employers. 46 In addition, there may be a set of applicants for whom the employer knows their current salary regardless, such as government employees or employees in a firm, occupation, or industry about which the employer is knowledgeable. Finally, an applicant may volunteer the information. In that case, even though there is no specific exemption for applicant-volunteered salary history under the Oregon Equal Pay Act of 2017, it is hard to imagine that an employer would ignore it. Suppose an applicant says that he makes $50,000 per year, and an employer, based on that information, offers him $55,000. This consideration of prior salary would violate the Oregon statute, but the violation may be difficult to detect. The applicant has no incentive to complain, 47 and the employer need only come up with a lawful justification, such as “merit” or some combination of education, training, and experience, 48 to justify the offer.

Second, even if the Act alters the information available to employers, this may not help women. There is a risk that by limiting the information available to employers, the Act may encourage discrimination based on gender. At first, this may seem counterintuitive. But eliminating one input into employers’ hiring decisions, such as prior salary, may increase the informative value of other inputs, such as gender. Suppose an employer makes a wage offer based on what it believes a job applicant will accept. To infer the applicant’s reservation wage, that is, the lowest wage she will accept, the employer may rely on her education, her experience, and other information known about her — possibly her gender. 49 To pay someone differently because of gender is, of course, illegal, but there is evidence that employers behave this way. 50 Social scientists refer to this type of group-based inference as statistical discrimination. 51 If the reservation wages employers infer without using wage history approximately equal what they infer using the information, the ban on asking about prior wages may have little effect on wage offers. If employers underestimate the gender difference in earnings, the ban could have its desired effect. But if employers overestimate this difference, the ban could result in lower wage offers to women.

Preliminary evidence suggests that the risk that salary history bans encourage harmful statistical discrimination is not insignificant. Recently, in an effort to improve employment outcomes for individuals with criminal records, particularly black men, several states have passed “ban the box” laws that prohibit employers from asking job applicants about their criminal history. 52 Perversely, the policy was shown to encourage racial discrimination based on stereotypes about black criminality. 53 In fact, the black-white gap in employer callback rates grew sixfold. 54 Of course, there may be important differences between “ban the box” and salary history bans. First, racial stereotypes about criminality may be stronger than gender stereotypes about earnings. Second, wage determination may differ from decisions about whether to interview applicants. But the two policies share basic similarities: the underlying discrimination is illegal but difficult to detect, and the policies limit information available to employers but not information about race or gender. A concern that the “ban the box” result may extend to salary history bans should not be ruled out. 55 According to a recent survey, women who refuse to disclose their prior salary receive slightly lower wage offers than women who disclose, while men who refuse receive slightly higher offers. 56 This evidence is at least suggestive of the idea that employer inferences may make women worse off. It also highlights a separate concern, which is that, regardless of whether an employer may legally use the information, a man may be more likely to volunteer his prior salary. 57 In that case, the employer may draw a negative inference about a woman applicant who does not disclose. Oregon, at least, makes no voluntary-disclosure exception to the ban on determining compensation based on current or past compensation if an employee files an otherwise successful complaint. 58

The effect of the Oregon Equal Pay Act of 2017, then, may differ for discrimination claimants versus women who never bring a claim. The Act helps claimants by eliminating the prior salary defense, but its overall effect on wage setting is uncertain (and possibly undesirable). An unexplored possibility is that the current gender pay gap has more to do with factors, such as pay transparency 59 and workplace flexibility, 60 for which the salary history ban is not a remedy. Experimentation should inform whether other states follow the Oregon model or try something different. 61

^ Jessica L. Semega et al., U.S. Census Bureau, U.S. Dep’t of Commerce, P60-259, Income and Poverty in the United States: 2016 , at 2 (2017), https://www.census.gov/content/dam/Census/library/publications/2017/demo/P60-259.pdf [https://perma.cc/4A4P-QL8R]. This statistic compares the median earnings of full-time, year-round workers. Id. at 10. Gender differences in median earnings may reflect differences in human capital, including, for example, schooling and work experience; differences in occupations, industries, and firms; discrimination; and selection bias. See Francine D. Blau & Lawrence M. Kahn, The Gender Wage Gap: Extent, Trends, and Explanations, 55 J. Econ. Literature 789, 809–23, 825–36 (2017). Experimental evidence confirms the existence of gender-based discrimination in the labor market. See id. at 833–36. Return to citation ^

^ Unequal wages motivated several major pieces of federal legislation, such as the Equal Pay Act of 1963, 29 U.S.C. § 206(d) (2012), and Title VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e-2 to -17 (2012). For an overview of empirical findings about the effects of antidiscrimination laws, see John J. Donohue, Antidiscrimination Law, in 2 Handbook of Law and Economics 1387, 1439–50, 1454–55 (A. Mitchell Polinsky & Steven Shavell eds., 2007). Return to citation ^

^ See, e.g., Cal. Lab. Code § 1197.5 (West Supp. 2017); Mass. Gen. Laws ch. 149, § 105A (2016); N.Y.C., N.Y ., Admin. Code § 8-107(25) (2017). Return to citation ^

^ 2017 Or. Laws ch. 197, H.B. 2005 (to be codified in scattered sections of Or. Rev. Stat .). Return to citation ^

^ Table S2419: Class of Worker by Sex and Median Earnings in the Past 12 Months (in 2016 Inflation-Adjusted Dollars) for the Full-Time, Year-Round Civilian Employed Population 16 Years and Over, U.S. Census Bureau , https://factfinder.census.gov/faces/nav/jsf/pages/searchresults.xhtml [https://perma.cc/55HV-3NKE] (select “S2419” as table name and “Oregon” as state). Return to citation ^

^ Floor Letter, Or. Reps. Jodi Hack & Greg Barreto, Pay Equity Shouldn’t Be a Partisan Issue: Oppose HB 2005 (Mar. 28, 2017), https://olis.leg.state.or.us/liz/2017R1/Downloads/FloorLetter/1828 [https://perma.cc/U8T8-8WZW] (“[The bill] may limit our ability as employers to financially reward the star performers in our companies while simultaneously exposing us to punitive damages without adequate defense.”). Return to citation ^

^ See Floor Letter, Or. Sen. Tim Knopp, YES on HB 2005B — The Oregon Equal Pay Act of 2017 (May 17, 2017), https://olis.leg.state.or.us/liz/2017R1/Downloads/FloorLetter/1997 [https://perma.cc/U76T-FYG7] (“Employers want to do the right thing and [this bill] allows [them] . . . to identify the problem (if they have one) before they are sued . . . .”). The statute defines “equal-pay analysis” to be “an evaluation process to assess and correct wage disparities among employees who perform work of comparable character.” Oregon Equal Pay Act of 2017 sec. 1, § 652.210(1)(4). Return to citation ^

^ Id. sec. 2, § 652.220(1)(d). The Act provides that this provision does not apply to transfers or promotions with the same employer. Id. Because the provision bars employers from using salary history, one may wonder why the ban on asking about salary history is even necessary. Presumably, it is an additional safeguard to prevent employers from using the information, which may be difficult to detect. Return to citation ^

^ Id. secs. 7, 9, 10. If an employee files a complaint with the Commissioner of the Oregon Bureau of Labor and Industries and the Commissioner finds for the employee, the Commissioner must award back pay equal to the time necessary to resolve the complaint plus either two years or the time the employee was subject to an unlawful wage differential, whichever is less. See id. sec. 7, § 659A.870(4). In a civil action, a court may exercise discretion to award back pay. See id. sec. 10, § 659A.885. The Act limits punitive damages to cases in which an employer engages in fraud, acts with malice or engages in willful and wanton misconduct, or is a repeat offender. See id. sec. 9, § 659A.885(4). Return to citation ^

^ Mass. Gen. Laws ch. 149, § 105A(c)(2) (2016); see Stacey Cowley, Illegal in Massachusetts: Asking Your Salary in a Job Interview, N.Y. Times: DealBook (Aug. 2, 2016), https://nyti.ms/2jE6shF [https://perma.cc/WX99-4K2C]. Return to citation ^

^ See Cal. Lab. Code § 1197.5(b)(3) (West Supp. 2017); Del. Code Ann . tit. 19, § 709B (2017); P.R. Leyes An . tit. 29, § 254 (2017); Oregon Equal Pay Act of 2017; N.Y.C., N.Y ., Admin. Code § 8-107(25) (2017); Phila., Pa., Code § 9-1131(2) (2017). Most of the laws do not go into effect immediately. The earliest to take full effect was the New York City ban, which began on October 31, 2017. See N.Y.C., N.Y ., Admin. Code § 8-107(25). In contrast, the Massachusetts ban comes into force on July 1, 2018, Mass. Gen. Laws ch. 149, § 105A, and the Oregon Equal Pay Act of 2017 takes full effect on January 1, 2024, Oregon Equal Pay Act of 2017 sec. 13. Philadelphia’s ban is the subject of a legal challenge: the Chamber of Commerce for Greater Philadelphia alleges in part that the ban violates the First Amendment. See First Amended Complaint at 22, Chamber of Commerce for Greater Phila. v. City of Philadelphia, No. 17-1548 (E.D. Pa. June 13, 2017). Return to citation ^

^ H.R. 2418 sec. 2(a), § 8(1). The bill has the support of several prominent organizations, such as the AFL-CIO, the ACLU, and the NAACP. See Letter from American Association of University Women to the U.S. House of Representatives (May 24, 2017), http://www.aauw.org/files/2017/01/Pay-Equity-for-All-Act-Sign-On-nsa-1.pdf [https://perma.cc/6VVC-38PF]. Return to citation ^

^ Cowley, supra note 27 (quoting Victoria A. Budson, Executive Director of the Women and Public Policy Program at the Harvard Kennedy School of Government). Return to citation ^

^ Wernsing v. Dep’t of Human Servs., 427 F.3d 466, 469 (7th Cir. 2005) (quoting Kouba, 691 F.2d at 876); see id. at 470 (“The disagreement between this circuit . . . and those that require an ‘acceptable business reason’ is established, and we are not even slightly tempted to change sides.”); Taylor v. White, 321 F.3d 710, 719 (8th Cir. 2003) (“[The] ‘reasonableness’ level of review [in Kouba] is . . . greater than that which we believe to be required under Title VII and the EPA.”). Return to citation ^

^ Angove v. Williams-Sonoma, Inc., 70 F. App’x 500, 508 (10th Cir. 2003); see Irby v. Bittick, 44 F.3d 949, 955 (11th Cir. 1995). According to the Eleventh Circuit, “prior salary alone cannot justify pay disparity,” Glenn v. Gen. Motors Corp., 841 F.2d 1567, 1571 (11th Cir. 1988) (emphasis added), because the legislative history of the EPA demonstrates that Congress intended the statutory term “factor other than sex” to mean something more than market forces leading women to accept lower pay for equal work, id. at 1570–71. See also Nicole Buonocore Porter & Jessica R. Vartanian, Debunking the Market Myth in Pay Discrimination Cases, 12 Geo. J. Gender & L. 159, 173 (2011) (“Congress designed the EPA to be broadly remedial, . . . suggesting that legislators could not have intended an exception so broad that it would provide employers with a convenient loophole capable of justifying almost any wage differential.”). Return to citation ^

^ 854 F.3d 1161 (9th Cir.), reh’g en banc granted, 869 F.3d 1004 (9th Cir. 2017). Return to citation ^

^ See Del. Code Ann . tit. 19, § 709B (2017). Recently, California amended its law, making it more like Delaware’s. See Assemb. 168, 2017–2018 Leg., Reg. Sess. (Cal. 2017) (signed into law Oct. 12, 2017). Return to citation ^

^ However, one possibility is that an employee can obtain an offer from another employer and use the offer to bid up the employee’s current wage. In that case, if the law improves employees’ outside options, it could raise wages for employees who never change jobs. Return to citation ^

^ Perhaps lawmakers envisioned that, for example, a female employee would have a strong incentive to file a complaint after learning that a male colleague earned more in part because he voluntarily disclosed his prior salary. But this example highlights the value of pay transparency. See Deborah Thompson Eisenberg, Money, Sex, and Sunshine: A Market-Based Approach to Pay Discrimination, 43 Ariz. St. L.J. 951, 1005–06 (2011); Jake Rosenfeld, Pay Transparency at Work: The Great Equalizer?, OnLabor (Mar. 1, 2016), https://onlabor.org/pay-transparency-at-work-the-great-equalizer/ [https://perma.cc/AE9Y-EUCR]. Return to citation ^

^ This idea extends in a straightforward manner to race and other group identities that may be observable by the employer. Return to citation ^

^ Cf. Roland G. Fryer, Jr., et al., Racial Disparities in Job Finding and Offered Wages, 56 J.L. & Econ. 633, 635–36 (2013) (finding evidence that black workers have lower reservation wages and employers statistically discriminate based on race in making wage offers). Return to citation ^

^ See Edmund S. Phelps, The Statistical Theory of Racism and Sexism, 62 Am. Econ. Rev. 659 (1972); Kenneth Arrow, The Theory of Discrimination, in Discrimination in Labor Markets 3 (Orley Ashenfelter & Albert Rees eds., 1973); Dennis J. Aigner & Glen G. Cain, Statistical Theories of Discrimination in Labor Markets, 30 Indus. & Lab. Rel. Rev. 175 (1977). Employers may also discriminate based on animus. See generally Gary S. Becker , The Economics of Discrimination 39–54 (2d ed. 1971) (proposing a model of employer discrimination based on animus). Return to citation ^

^ Amanda Agan & Sonja Starr, Ban the Box, Criminal Records, and Racial Discrimination: A Field Experiment, 133 Q.J. Econ . (forthcoming 2018) (manuscript at 5). Return to citation ^

^ One piece of experimental evidence suggests that preventing employers from learning about prior compensation may benefit certain job applicants. See Moshe A. Barach & John J. Horton, How Do Employers Use Compensation History?: Evidence from a Field Experiment (CESifo, Working Paper No. 6559, 2017), https://ssrn.com/abstract=3014719 [https://perma.cc/KT5E-ZA6D]. In a field experiment involving an online labor market for remote tasks, such as computer programming, researchers randomly assigned whether employers could observe job applicants’ prior wages. Id. at 6, 10. Employers that could not observe prior wages considered more applications and called back and hired applicants with lower past wages, on average. Id. at 16, 21–22. Information about prior wages did not affect the prevalence of wage bargaining, but conditional on bargaining, applicants with hidden wage histories struck better bargains. Id. at 26–28. The problem with extending this finding to the Oregon Equal Pay Act of 2017 is that the online labor market in the experiment included no information about gender, so that, by construction, gender-based discrimination could not occur. Return to citation ^

^ Women have been found to be less likely to negotiate and to compete. See Marianne Bertrand, New Perspectives on Gender, in 4B Handbook of Labor Economics 1545, 1551–54, 1556–58 (David Card & Orley Ashenfelter eds., 2011). Return to citation ^

^ See Claudia Goldin, A Grand Gender Convergence: Its Last Chapter, 104 Am. Econ. Rev. 1091, 1093–94, 1103–04 (2014). Women may prefer greater flexibility in hours, for which different firms and sectors face different costs, leading to workplace sorting. Id. at 1116–17. Return to citation ^

^ Cf. New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting) (“It is one of the happy incidents of the federal system that a . . . State may . . . serve as a laboratory . . . [to] try novel social and economic experiments . . . .”). Return to citation ^

• Civil Rights
• Discrimination
• Employment Law
• Gender