Composite scenario: the resume that didn't get a callback

EDITORIAL NOTE: This is a composite scenario, not a record of a specific person's case. It draws on Bertrand & Mullainathan's landmark 2004 audit study and its replications.

An applicant with strong credentials submits resumes to employment listings using the name they were given at birth — a name more common among Black Americans than white Americans. Callbacks come back at one rate.

Same applicant, same resume, same week, the experimental version uses a name more common among white Americans. Callbacks come back at a noticeably higher rate.

Bertrand and Mullainathan's 2004 paper found a 50% higher callback rate for the white-name resumes. A 2017 meta-analysis by Quillian, Pager, Hexel, and Midtbøen covering studies from 1989 to 2015 found no decline in hiring discrimination against Black Americans over that quarter-century.

Title VII of the 1964 Civil Rights Act, enforced by the EEOC, covers this category of discrimination. Filing instructions appear on the Pathways page under 'Employment / EEOC'.

The Bertrand and Mullainathan resume-audit study, published in the American Economic Review in 2004, is one of the most widely cited empirical studies in labor economics. The researchers responded to approximately 1,300 help-wanted advertisements in Boston and Chicago newspapers in 2001 and 2002 using approximately 5,000 fictitious resumes. Each resume was assigned either a first name distinctively common among Black Americans (such as Lakisha, Tamika, Jamal, or Tyrone) or a first name distinctively common among white Americans (such as Emily, Anne, Brendan, or Greg). Resumes were also varied along credentials — some resumes were uniformly higher-quality in skills, work history, and education, others were uniformly lower-quality. The fictitious resumes were otherwise carefully matched. Each employer received pairs of resumes, allowing within-employer comparison. The researchers measured callback rates by name category and by resume quality.

The principal findings were three. First, callback rates were roughly fifty percent higher for the white-name resumes than for the Black-name resumes (9.65 percent vs. 6.45 percent). Second, the callback differential was statistically indistinguishable between higher-quality and lower-quality resumes; the discrimination did not diminish with stronger credentials. Third, the differential was substantially uniform across employer industry, employer size, advertised wage, and advertised skill requirement. The implication was that name-based screening was a general feature of the entry-level labor market rather than a localized phenomenon in particular industries.

The Bertrand and Mullainathan study has been replicated and extended many times. Quillian, Pager, Hexel, and Midtboen (PNAS, 2017) conducted a meta-analysis of audit studies of hiring discrimination against Black Americans covering the period 1989 to 2015. The meta-analysis found no statistically significant decline in hiring discrimination over that quarter-century interval. The white-Black callback ratio in the meta-analytic average was approximately 1.36 in 1989, approximately 1.30 in 2015, and substantially flat across intervening years. The persistence of the differential despite the increasing prominence of equal-employment-opportunity legislation, employer diversity programs, and human-resources professionalization is one of the principal puzzles in the modern labor-economics literature on discrimination.

The legal framework for addressing employment discrimination is Title VII of the Civil Rights Act of 1964, enforced by the Equal Employment Opportunity Commission. Title VII prohibits discrimination on the basis of race, color, religion, sex, or national origin in hiring, firing, compensation, promotion, and other terms and conditions of employment. The EEOC's complaint mechanism is the principal first-step administrative remedy. Complaints must be filed within 180 days of the alleged discriminatory act in non-deferral states, or 300 days in deferral states that have their own employment-discrimination agencies (most states). The EEOC's intake process produces a right-to-sue letter after investigation; the complainant may then proceed to federal-court litigation under Title VII.

The practical difficulty of bringing individual hiring-discrimination cases is the asymmetry of information. An individual job applicant who is not called back does not know, in the typical case, why. The applicant does not know how many other applicants applied, what their qualifications were, or what the employer's actual decision criteria were. The discoverable evidence in an individual Title VII case is the employer's records of the hiring decision, which the employer may not have systematically created. Title VII's disparate-impact framework (Griggs v. Duke Power Company, 1971) provides an alternative to the intent-based disparate-treatment claim where a facially-neutral employer practice produces a statistically disparate outcome by protected class; the disparate-impact framework requires a smaller showing on intent but a larger showing on the statistical pattern. Pattern-or-practice litigation brought by the EEOC or by class counsel is the principal effective vehicle for systemic employment-discrimination remedies.

The ban-the-box movement and the parallel efforts on salary-history-inquiry restrictions, blind-screening of applications, and structured-interview protocols represent the principal current policy responses to the documented persistence of name-based screening. The empirical evidence on these reforms is mixed. Several jurisdictions that have prohibited employer use of criminal-history information at the application stage have produced documented improvements in callback rates for applicants with criminal records, but smaller and less consistent effects on race-conditional hiring outcomes. The platform's pathways pages cover the principal EEOC and state-agency intake routes for individual employment-discrimination complaints.

The contemporary EEOC enforcement framework has been the subject of substantial subsequent guidance and rule-making activity addressing the kinds of hiring-discrimination patterns documented in this composite scenario. The EEOC's 2023 technical assistance document on the use of artificial-intelligence-driven hiring tools, the 2024 settlement with iTutorGroup (the first federal AI-hiring-discrimination settlement), and successive subsequent guidance documents have established the operational framework for contemporary employment-discrimination cases involving algorithmic decision-making tools. The EEOC's intake process produces a charge of discrimination, an investigation, and either a right-to-sue letter or a settlement.

The state-level employment-discrimination frameworks provide additional substantive and procedural protections in most jurisdictions. State human-rights agencies handle complaints under cooperative agreements with the EEOC in the deferral-state framework. Several states have enacted protected-class additions beyond the federal framework (sexual orientation, gender identity, marital status, lawful occupation, ban-the-box restrictions on criminal-history inquiries) and additional procedural protections (longer filing windows, expanded damages availability). The platform's pathways pages cover the principal EEOC, state-agency, and civil-society resources for individual employment-discrimination complaints.