Eliot L. Sherman, Raina Brands, and Gillian Ku. "Dropping Anchor: A Field Experiment Assessing a Salary History Ban With Archival Replication." Revise and resubmit at Management Science.

Could a salary history ban (SHB) reduce the gender wage gap? Support for this idea continues to build among policy-makers and practitioners—to the extent that it is now illegal in fourteen states in the U.S. to ask a job applicant to disclose their salary history. Proponents believe SHB can address earnings disparities by eliminating the anchor of an applicant’s current salary, which is often lower for women than for comparable men. Alternatively, related research regarding obscured criminal histories suggests that SHB could exacerbate disparity, insofar as hiring managers resort to statistical discrimination in the absence of disconfirming salary information. But SHB could also have no effect—due to hiring managers’ non-compliance and prior knowledge of internal applicant salaries, or because prior salary is less of an anchor than extant theory would suggest. In order to adjudicate between these perspectives, we assessed the impact of SHB in two different ways. Study 1 examined the effect of SHB via a 16-month field experiment at a mid-sized educational institution. Our analysis, which assessed 230 hires deriving from both inside and outside the organization, indicated no main effect of SHB on salary offer, nor interaction effects regarding gender or mobility channel. Study 2 leveraged placement data from a recruitment agency that voluntarily adopted SHB. This analysis, which compared 687 hires in the post-treatment period with 3,001 hires in the pre-treatment period, shows that salaries were lower under SHB; however, this effect did not achieve statistical significance in models that also specified gender. Taken together, our analysis strongly suggests that SHB fails to consistently achieve its aims.