Fairness in employment gets discussed as if it were a mood, something conveyed by tone and townhalls. It is more usefully treated as a set of contract clauses. Ask what a rational adult would need to see in writing before committing years of finite life to an organization, and the answer converges on three clauses: hours that are what they say they are, numbers that can be checked, and an exit that does not destroy the relationship. Each now has evidence attached.
Clause one: hours that mean something
The evidence on time arrangements has quietly become some of the best in management research. A randomized controlled trial at Trip.com with 1,612 employees found that hybrid work (two home days a week) cut voluntary quits by a third, from 7.2 to 4.8 percent, with no detectable damage to performance or promotion across two years (Bloom, Han and Liang, Nature, 2024). On duration, a six-month four-day-week trial covering 2,896 employees in 141 organizations across six countries found reduced burnout and improved mental and physical health, with bigger individual reductions in hours producing bigger gains (Fan, Schor and colleagues, Nature Human Behaviour, 2025; quasi-experimental, self-selected firms).
The pattern across both: giving workers real control of time costs less than employers fear and returns retention and health. A fair 2026 contract therefore states the actual hours, the actual flexibility, and the actual after-hours expectations, and treats deviations as what they are, an unpriced wage cut.
Clause two: numbers that can be checked
Pay fairness is not a single lever; the research shows the design details carry the effect. Zoë Cullen's review in the Journal of Economic Perspectives (2024) distinguishes horizontal transparency (peers), vertical (up the ladder) and cross-firm, with opposite effects: peer comparisons can demotivate, while knowing what the boss makes tends to increase effort by revealing the return to advancement. A field experiment in Indian manufacturing (Breza, Kaur and Shamdasani, QJE 2018) found that pay inequality workers could not justify cut output by 0.45 standard deviations and attendance by 18 percentage points, but when the better-paid colleague's higher productivity was observable, the same inequality did no damage. The variable is not equality; it is justifiability.
The contract clause that follows: publish the structure, ranges, the basis for differences, and enough of the firm's numbers that claims about affordability can be audited. Ricardo Semler has described the mechanism at Semco: a fixed share of each unit's profit went to the people who produced it, with the books open so anyone could check the math; the side effect, as he tells it, was people turning off lights because the money had become theirs. Open numbers do not just legitimate pay. They convert expenses into everyone's expenses.
Clause three: exits without punishment
The old deal treated departure as betrayal, and its instruments survive: counteroffers made only under resignation, references held hostage, knowledge hoarded against the day someone leaves. In a labor market where median tenure is 3.9 years (U.S. BLS, January 2024), designing employment around the fiction of permanence guarantees that every real departure is handled as an emergency and a wound.
A fair contract prices the exit in advance: clear notice terms both ways, no forfeiture traps in variable pay, references as a routine right, and where possible a bridge rather than a cliff. Semco built the extreme version, as Semler tells it: employees who wanted to leave could take the equipment, become suppliers, and sell back to Semco and to its competitors; half the company eventually went, across a bridge built plank by plank with the company holding the other end. Few firms will go that far. Every firm can stop punishing the door.
The meta-clause
Behind all three runs a single principle: anything the organization claims about itself should exist in checkable form. Stated hours checkable against messages sent at midnight; stated pay philosophy checkable against ranges and books; stated respect checkable against what happened to the last person who resigned well. The generational renegotiation of work is, at bottom, a demand to move the deal from the unfalsifiable (culture, family, passion) to the falsifiable. That demand will read as impertinence to institutions accustomed to being taken on faith. It is better read as the return of ordinary contractual hygiene to the one contract most adults sign without it.
Sources
- Bloom, Han & Liang (2024). Hybrid working from home improves retention without damaging performance. Nature. RCT, 1,612 employees: hybrid work cut quits by one third (7.2%→4.8%) with no performance or promotion penalty. https://www.nature.com/articles/s41586-024-07500-2
- Fan, Schor, Kelly & Gu (2025). Work time reduction via a four-day workweek. Nature Human Behaviour. 2,896 employees, 141 orgs, 6 countries: reduced burnout, better mental and physical health; dose-response in individual hours reduced. https://www.nature.com/articles/s41562-025-02259-6
- Cullen (2024). Is Pay Transparency Good? Journal of Economic Perspectives. Effects depend on design: horizontal, vertical and cross-firm transparency have opposite effects. https://doi.org/10.1257/jep.38.1.153
- Breza, Kaur & Shamdasani (2018). The Morale Effects of Pay Inequality. QJE. Unjustifiable pay inequality cut output ~0.45 SD and attendance 18pp; observable productivity differences eliminated the harm. https://academic.oup.com/qje/article-abstract/133/2/611/4430649
- U.S. Bureau of Labor Statistics (2024). Employee Tenure in 2024. Median tenure 3.9 years, lowest since 2002. https://www.bls.gov/news.release/archives/tenure_09262024.htm