Equitable Downsizing at AutoNow? (A) and (B)

Zahra Nasser, strategy director at AutoNow—a  fictional, 30,600-employee auto parts manufacturer in the Detroit area—faces a $200-million labor cost reduction mandate following the company’s leveraged buyout by private equity firm Tramonto Capital Partners. Seven downsizing criteria have been proposed: (1) strategic realignment: eliminate legacy combustion-engine manufacturing positions; (2) cost center streamlining: target Administrative, HR, and Supply Chain departments; (3) last in, first out: cut least tenured employees company-wide; (4) AI occupational exposure: eliminate roles with highest AIOE scores; (5) geographic optimization: reduce the company’s facility in Texas; (6) merit-based rightsizing: cut lowest-rated performers; and (7) reliability: target employees with highest absenteeism.

Nasser must choose one—or design a defensible alternative—and apply it to a real workforce dataset containing 14 observable employee variables. Then, after making her selection using identity-neutral criteria, she receives a memo from Evelyn Choi, general counsel of Tramonto Capital Partners (Case B), requiring a formal adverse impact analysis by both racialized identity and gender, and a revised recommendation if disparate impacts are found.

The case is designed so that every available downsizing option produces a measurable disparate impact on at least one protected group. There is no safe answer. Each option harms a different group through a different structural mechanism. The central insight is that identity-neutral decisions applied to structurally unequal workforces produce unequal outcomes, because prior differential treatment and disparate allocation are already encoded in the workforce’s structure. Students discover this empirically, in real time, through hands-on data analysis.

• Calculate adverse impact for each protected group (racialized identity, gender) by comparing the demographic composition of a downsized pool to the overall workforce, and interpret the direction, magnitude, and statistical significance of observed disparities.
• Distinguish among the mechanisms that produce disparate impact—differential evaluation (bias in performance ratings), disparate allocation (sorting of groups across roles, departments, and sites), measurement bias (criterion variables that encode structural inequality), and tenure cohort effects—and identify which mechanism is operative for each downsizing option.
• Evaluate the ethical, legal, and strategic tradeoffs of each downsizing criterion using both deontological and consequentialist frameworks.
• Design a revised downsizing plan that minimizes adverse impact while meeting a financial target, and articulate the justification for that design in terms a PE investor, a labor attorney, and a DEI officer could each accept.
• Assess the limits of “objectivity” in data-driven HR decisions—specifically, explain why an identity-neutral criterion can still produce discriminatory outcomes, and what organizational conditions produce that result.