What salary history bans actually changed
Hansen and McNichols studied the early effects of statewide salary history bans. They found the gender earnings ratio increased by about 1% in states with bans, with the gains concentrated in several subgroups. Their design estimates the net policy effect; it does not show that every employer switched cleanly from past pay to market pricing.1
Disclosure can help and hurt
A separate field experiment used hundreds of recruiters and more than 2,000 fictional applications. Employers made negative inferences about candidates who did not disclose. For men and other higher-paid candidates, disclosure produced higher salary offers but also fewer callbacks after accounting for salary. Those results do not identify one choice that improves both outcomes for everyone.2
What candidates can control
Know the local rule, ask for the employer's range, and prepare your own walk-away point before the conversation. Never invent a current salary. If you choose not to disclose, be ready to redirect calmly; if you do disclose, understand what signal the number may send in that market.
Why the first number matters
Across three general negotiation experiments, the party making the first offer obtained a better outcome, and first offers strongly predicted final settlements. Those experiments were not salary-history-ban studies, so use them for the narrower lesson: prepare the reference point you want to discuss and the evidence behind it.3
Limitations
- •Legal requirements vary by jurisdiction. This is not legal advice.
- •The two salary-history studies are NBER working papers, not guarantees about an individual negotiation.
- •The negotiation experiments tested first offers generally, not salary-history disclosure specifically.