Governance research · Structural fairness · Washington, D.C.
It is not a feeling about a government. It lies in how the arrangements people live under are built, and how people judge those arrangements can be measured. CARO is a governance research organisation: it publishes the theory, the measurement method and its formulae, and studies how governance works in practice.
Political philosophy has long asked when an institution may claim authority over free and equal persons, and has mostly answered from inside institutions: fair procedures, just outcomes, rules no one could reasonably reject. CARO’s work begins a step earlier. An institution can follow fair procedures and still deny people equal standing, because of how it is built: who can see the reasons behind a decision, who can contest them, and who bears the risk of dissent.
CARO calls the right to that standing the Meta-Right to Structural Fairness.
Two are familiar. The third is the subject of CARO’s work.
Fairness judged by what a decision produces: who gains, who loses, and whether the distribution can be defended.
Fairness judged by how a decision was reached: whether the stated rules were applied, consistently and in the open.
Fairness judged by the institution itself: how power is distributed and constrained, and whether everyone subject to it can see, weigh and contest the reasons that bind them.
A firm hires the best-qualified candidate, using the stated criteria. The candidate is also the employer’s friend, and nothing in the process separated the decision from the friendship. The outcome is defensible and the procedure was followed. The structure is not fair.
Structural fairness is prior to the other two. A fair procedure inside an unfair design does not make the result fair.
Three areas of work, all published, all open.
What structural fairness is, stated once and precisely: the Meta-Right to Structural Fairness, Equitism as the political philosophy that carries it, Fairocracy as the institutional logic, and the Unified Fair Democratic System as the constitutional form.
Read the standard →The Fairness Index: twenty-five questions across five domains, nine formulae, and the rules by which one version succeeds another. Published unabridged, so anyone holding the raw responses can reproduce every figure.
The formulae →How governance works in practice. Bangladesh is CARO’s primary case, and Bangladesh — the Architecture is in preparation. The framework applies equally to established democracies. From 2027, AI governance: AI used in governance asks people to accept conclusions whose reasons they cannot evaluate or challenge.
Publications →The method has been applied in full to one cohort. These are the figures, and these are their limits.
Bangladesh, April 2026. All five domains scored below 2.00 — a broad consistent signal rather than unfairness concentrated in one area. The lowest was public services and welfare, at 1.44.
The cohort was 71 per cent aged 18–25, 61 per cent male, 62 per cent urban and 54 per cent undergraduate. It is not nationally representative, and nothing above should be read as a measurement of Bangladesh. It is a test of whether the instrument produces differentiated, interpretable results on a real population. It did.
Every question, every distribution, every standard deviation →
Eighty-five per cent answered completely unfair — the clearest consensus anywhere in the instrument. The question concerns regional differences in the quality of public education. Article 17(a) of the Constitution of Bangladesh commits the state to “a uniform, mass oriented and universal system of education”. Article 17 sits in Part II, and Article 8(2) states that the principles in that Part are fundamental to the governance of Bangladesh but shall not be judicially enforceable. No citizen can claim it in court.
The cohort’s strongest judgment falls on a commitment the Constitution makes and Article 8(2) places beyond the courts. CARO calls this Legal Injustice: a right declared in a form that cannot be claimed.
There is nothing to apply for.
The standard and the method are published for anyone to use, adapt and translate — any population, no fee, no licence, no permission. CARO claims no authority over a body that applies them, holds no seat on it, and does not certify, endorse, fund or audit it. Such bodies are not chapters, branches or affiliates of CARO.
CARO uses its own method on the same terms as everyone else.