EV-0016
Fairness works on performance by running through trust
Colquitt, J. A., Scott, B. A., Rodell, J. B., Long, D. M., Zapata, C. P., Conlon, D. E., & Wesson, M. J. (2013). Journal of Applied Psychology, 98(2), 199–236 · Source
What this entitles us to say
A meta-analysis of 493 independent samples finding that the relationships between justice and both task performance and citizenship behavior are mediated — justice does not act on performance directly, it acts through social-exchange indicators such as trust.
What it found
493 independent samples, testing direct, mediating and moderating hypotheses drawn from social exchange theory and from affect-based perspectives. The headline result for us: the significant relationships between justice and both task performance and citizenship behavior were mediated rather than direct — running through social-exchange indicators.
Why it matters here
This is the mechanism entry, and it puts a third trait into the same chain the other Foundational dossiers keep landing on. Faithfulness research said broken agreements cost you trust. Trust research said trust predicts performance and effort. Now justice research says fairness reaches performance through the exchange relationship rather than around it.
Read together, the Foundational region is not four adjacent virtues that each happen to help. It is one mechanism with four inputs, and trust is the thing they all pay into. That is a claim the framework already made structurally; three separate literatures have now arrived at it independently.
It also disciplines the sales argument. Fairness is not a productivity lever you pull. It is a deposit into a relationship that then shows up in the work — which is slower, less dramatic, and true.
The limits
Correlational and largely cross-sectional, so mediation here is a statistical pattern rather than a demonstrated causal chain. Same-source measurement runs throughout. Verified to abstract level, so the mediation finding may be described but no coefficient quoted. It also tests two theoretical accounts against each other rather than settling on one.
What we do not say
"Fairness causes performance," or any claim that a fairness intervention produces a measurable output gain. The design cannot support either.