Distributional-Incidence Protocol

Status: Proposed (June 2026). Formalizes Exchange #21 Round 5 Deliverable 5 (v2, anti-theater) as a project-wide review protocol, alongside Adversarial Review, Coherence Audit, Historical Parallel Test, and Comparative Alignment. Spawned per ROADMAP TODO #1 / F6 (the steward's June 14, 2026 "build F6 now" decision). Carries an un-closed debt: the bounded-governance doctrine this protocol serves is grounded in an unrepresentative reference class (small/high-trust/wealthy polities); transferability to large heterogeneous polities is unresolved (research debt under TR1).

The problem this protocol addresses

A reform proposal can be arithmetically correct and still fail — fiscally sound, technically defensible, and yet so regressive in its distribution that it cannot survive its first political cycle, or so silently regressive that it should not. Exchange #21 established this twice over: the Argentina case (a successful fiscal consolidation whose approval collapsed because the adjustment was regressive) and the comparative-cases cluster (Canada's consolidation survived on broad distribution; New Zealand's reformers paid the largest political cost on concentrated distribution).

But the obvious fix — "require a distributional analysis" — has a known failure mode. The U.S. regulatory-impact-analysis regime (OMB Circular A-4, CBO scoring, Regulatory Flexibility Act small-business analyses) has produced incidence analyses for decades and has not prevented regressive outcomes. It has largely become a compliance layer that organizations with regulatory counsel can satisfy and others cannot. Exchange #21 Round 4 judged this "procedural theater" risk real and already realized in a close comparator.

This protocol exists to get the benefit (distribution made visible and binding) without producing one more layer of theater. The discipline is in the anti-theater conditions (§3–§5), not in the table itself.

Protocol

1. Ask the upstream question first — should this rule exist? — symmetrically for adoption and repeal

Before measuring a rule's incidence, engage the prior question: should this rule exist at all? Apply it with equal weight to rule adoption and to rule repeal. Distributional-incidence review is not a tool for preserving status-quo rules any more than for adopting new ones; a review that only ever scrutinizes new rules is a status-quo-bias engine. Naming the upstream question symmetrically is what keeps the protocol from becoming a one-directional ratchet (§21 Round 4 challenge 5c).

2. Build the incidence table — including a separate worst-off-protection row

Produce, at minimum:

An incidence table — who bears the direct, secondary, and long-run costs of the rule, measured across income quintiles and at least one other policy-relevant grouping where it matters (age, geography, race/ethnicity, sector of employment).
A separate worst-off-protection row — impact on the least-advantaged specifically (not the bottom quintile used as a summary stand-in), following the Rawlsian commitment the justification invokes. This row is distinct from quintile incidence and is read on its own.

3. Make the findings binding, not advisory

Incidence findings must be binding on rule design to a specified degree, not merely published. Concretely: if bottom-quintile (or worst-off) incidence is negative and unmitigated, the rule must carry either an explicit compensating mechanism or an explicit, on-the-record justification for the distribution it accepts. "We analyzed it" is not a pass; "we analyzed it and here is what we did about it, or why we did not" is.

4. Trigger remediation on defined thresholds

Specify, in advance, the thresholds that trigger remediation (not just disclosure). A finding that crosses a pre-registered threshold obligates a defined response — a compensating mechanism, a redesign, a sunset acceleration, or a documented decision to proceed with named accountability. Thresholds set after the analysis can be drawn to clear the result; set them first.

5. Subject the incidence review itself to adversarial review (the anti-theater keystone)

The incidence review is itself subject to adversarial review: a separate reviewer commissions a counter-incidence analysis using a different methodology or assumption set. This is the single condition the RIA comparator most lacks and the one most responsible for keeping the protocol honest — a self-produced incidence table that no adversary has tried to overturn is the theater this protocol is built to avoid. Where the harness is available, this step is a natural Cross-Lineage Review Harness target (the counter-analysis run on an independent lineage).

6. Publish and feed the ledger

Publish the review (and its counter-analysis) through the rule's transparency element; the findings become part of the rule's ongoing public record. Where a compensation/accounting ledger exists (bounded-governance Element 3), incidence findings feed it, so the rule's political-economic trajectory is tracked alongside its fiscal-arithmetic trajectory.

7. Carry the reference-class limitation explicitly

State, in any artifact that uses this protocol, that the bounded-governance doctrine it serves is grounded in an unrepresentative reference class and that its transfer to large heterogeneous polities is unresolved. The protocol disciplines how a rule's distribution is examined; it does not, by itself, establish that the rule transfers. Do not let a completed incidence table read as a transferability proof.

When to apply this protocol

Use this protocol when:

a bounded-governance rule or reform proposal with distributional consequences is being specified, adopted, or repealed
a proposal moves from the Proposal Catalog toward development, or an F-track public artifact (especially F3/F4/F5) approaches publication
a claim is being made that a rule is "fiscally sound" or "technically correct" without yet asking who pays

Lightweight commentary does not require the full protocol. Any proposal intended to influence the project's public positions, the Problem Map, or a real-world reform recommendation does.

What this protocol does not do

It is not a veto. A rule with regressive incidence may still be adopted — but only with an explicit compensating mechanism or an on-the-record justification (§3). The protocol forces the choice into the open; it does not make it.
It is not a substitute for political legitimacy or for the missing practitioner / Global-South / beneficiary voices (Exchange #21's largest residual debt). A clean table is not consent.
It does not close the transferability question (§7). That is research debt under TR1, not something an incidence table resolves.
It does not by itself defeat capture: an adversarial counter-analysis (§5) reduces theater but does not guarantee the reviewer pool is decorrelated. Where it matters, pair §5 with the decorrelation discipline (different methodology and different lineage/institutional priors).

Relationship to sibling protocols

This is the fifth structured review protocol the project uses. Each addresses a different failure mode:

Adversarial Review Protocol: Challenges claims — tests whether strong conclusions are defensible.
Coherence Audit Protocol: Checks internal consistency across the corpus.
Historical Parallel Test Protocol: Grounds reform claims in historical evidence.
Comparative Alignment Protocol: Compares external formation documents to the project's principles.
Distributional-Incidence Protocol (this document): Asks "who pays, and is that acceptable and acted upon?" of any rule with distributional consequences.

The protocols compose: §5 of this protocol is an invocation of Adversarial Review (and, where available, the Cross-Lineage Review Harness), turned on the incidence analysis itself.

Relationship to project principles

Principle 2 (essential needs should not be held hostage to avoidable scarcity): a reform that relieves aggregate scarcity while concentrating its costs on the least-advantaged is not the kind of correction Principle 2 calls for; the worst-off-protection row makes that visible.
Principle 5 (critical systems require public-interest governance): distributional incidence is constitutive of a durable bounded-governance rule, not an external addendum — the protocol operationalizes that.
Principle 4 (power must remain accountable, legible, and reversible): binding findings + triggered remediation + published counter-analysis are the accountability-and-legibility mechanisms; the symmetric adoption/repeal step is the reversibility mechanism.
Principle 16 (justice mediates between competing claims): the protocol is a structured way to surface who wins and who loses under a rule, rather than hiding the trade behind aggregate arithmetic.