MCC-IRC — Instrument Review Cycle Policy v1.0-draft

Document control

Field	Value
Document code	MCC-IRC (code assigned by the programme administration for this consultation; entered in the Document Register on adoption)
Title	Instrument Review Cycle Policy
Version	1.0-draft
Status	DRAFT — FOR CONSULTATION
Date of this draft	4 September 2026
Tier	Tier 5 (operational policy). It changes no rule and sets no threshold; it says when each instrument is looked at again, by whom, and what the looking produces. Sections 6 and 7 mark the two rows it cannot enact at its own tier.
Effective date	None. This policy is not in force. No review is due under it, no instrument is overdue under it, and the Instrument Review Register it constitutes does not exist.
Document owner	Programme Administrator (transitional, until a Secretariat is staffed — FD-2026-10-01 Schedule 1, C3)
Approved by	Not approved. Prepared for public consultation.
Decision basis	MCC-001 v2.0 cll. 0010, 0038, 0054, 0065, 0078–0086 (five-tier hierarchy and amendment authority), 0088–0097; MCC-001 v3.0-draft cll. 9.1.1, 9.1.2, 10.9, 12.1, 12.3, 12.6, 12.9; MCC-100 v2.1 cll. 0249, 0255, 0256; MCC-200 v2.0 cl. 0010; MCC-400 v1.1 art. 90; MCC-700 v3.1 §§3.3, 14.1, 14.4, 14.5, 17.1, 18.1 (which the document register records as in force, FD-2026-08-24 item D11; see cl. 3.4); MCC-700 v4.0-draft cll. 22.1, 22.3; MCC-800 v1.0 arts. 101, 111; MCC-850 v1.1 art. 45; Transparency Policy v1.1 §9.1 and art. 4; MCC-HCA v1.0 cll. 0010, 0185–0187; MCC-HCA v2.0-draft cll. 9.5, 10.3; MCC-SDCF v2.0-draft cl. 12.4; MIR-2026-001 rev 3 §§0, 3.6, 3.8, 6. Read with MCC-REST (Restatement and Correction Policy), a consultation draft prepared the same day, §§4, 6, 9, 12, and the Rule Change and Public Consultation Procedure v1.0, whose consultation closed on 8 September 2026, §§2, 6, 7, 9.
Supersedes	None (new document). It repeals nothing: clause 14.1 states expressly that every existing review clause continues in force on its own terms.
Language convention	British English; RFC-style normative verbs (SHALL/MUST = mandatory; SHOULD = recommended, deviation to be documented; MAY = optional). One requirement per numbered clause.
Classification	Public — consultation draft
Consultation period	30 days (FD-2026-10-01 Schedule 1, C4), from the opening recorded on the consultation register (FD-2026-09-04 item 2; FD-2026-10-01 item 5)

NOTE (non-normative). What was actually found, and why it justifies a document.

MIR-2026-001 rev 3 §3.6 lists an Instrument Review Cycle Policy as "referenced but never written". When this Policy was drafted on 4 September 2026, no instrument referenced it. The phrase appeared in three places in the entire record — revisions 1, 2 and 3 of MIR-2026-001 — and in no instrument, decision, board pack, Drive record or platform surface. That is the first finding, and it is a correction to the premise, not a quibble: this Policy is not the implementation of a promise some instrument made. It is a proposal.

What the corpus contains instead is thirty-five separate review provisions, scattered across fifteen instruments, with no two of them written by anyone who had read the others. Schedule 1 reproduces every one. The periods they state are semi-annual, annual, biennial, three-yearly and five-yearly; four instruments state two or more different periods about themselves; and the instruments that state no review period at all include MCC-100 and MCC-700 — the Programme Standard and the governance framework — together with the four counterparty and positioning texts, the Definitions Register, the PDD Template, the F01–F05 guidance and the Programme Bible.

No review has ever been performed under any of those clauses. Not one. The decision_register held six entries on 4 September 2026 and holds seven on 1 October 2026, and none is a review outcome; document_register has review columns, added after this draft was prepared, and no review is recorded in any of its rows; the platform's review_findings and technical_reviews tables hold zero rows each; and the served document manifest records a SHA-256 for every published standard and a review date for none. Four reviews have been done — the rulebook audit of 24 August, the programme integrity review accepted by FD-2026-09-03, the SDCF conformity review CR-2026-001 and MIR-2026-001 itself — and not one of them was conducted under an instrument's own review clause, on its schedule, or by the body that clause names. Every review this Programme has ever done was unscheduled.

Five of the bodies those clauses give the work to do not exist and have never existed: the Technical Committee and the AI Governance Committee (MCC-850 art. 45), MCC-HCA's Review Committee (cl. 0186, which MCC-HCA v2.0-draft cl. 9.5 proposes to abolish as non-canonical), MCC-SDCF's Expert Panel (cl. 0171) and MCC-400's VVB Advisory Panel (art. 90). Eight of the thirty-five provisions are assigned to a body that is constituted in nothing. Section 8 assigns no duty to any of them.

And the deadline is real. Seven adopted instruments carry the printed line "Next Review: September 23, 2026" — the Transparency Policy, the Claims Policy, the AI and Human Oversight Policy, the Applicant, Registry and VVB Terms, and the Legal and Regulatory Positioning paper, all dated 23 March 2026 and all six months old on that date. That date has passed, no review of any of them is recorded in the registers, and no record of a Governing Board meeting on 23 September 2026 has been filed. Nothing in the Programme was arranged to notice it.

The argument for this document is not that reviews are good practice. It is that the Programme already promised thirty-five of them, has delivered none, has no register in which the omission would show, and had seven fall due at once on 23 September 2026.

Consultation questions

Each point below is open for comment. A proposal is the administration's proposal, put to consultation under founder direction FD-2026-10-01; it is not a rule in force. The suite is to be adopted as one instrument once its longest consultation period has closed (FD-2026-10-01 Schedule 1, C4). Because it includes the Constitution, whose amendment is within the competence of the constituted Governing Board (FD-2026-08-24 item D11, recorded in the decision register as attested by the founder, with no executed copy filed), each point below is decided by the Governing Board on adoption of the suite; a clause of this text that provides another route for adopting it alone is itself open to comment.

1. Clause 4.5 — the biennial review MCC-400 v2.0-draft drops. Proposed: MCC-400 v2.0 relies on this Policy's Tier 3 interval and does not carry forward art. 90's biennial review, because carrying it forward would create a sixth period. Basis: this draft (cl. 4.5); FD-2026-10-01 item 4. Decided by: the Governing Board on adoption of the suite. Comments are invited on whether MCC-400 v2.0 should instead state its own review period.

2. Clause 6.3 — the Tier 1 review record. Proposed: no proposal is made — a Tier 5 policy cannot bind the review of the Constitution (cl. 6.3), and FDP-2026-001 makes no recommendation on amending MCC-001 v3.0-draft cl. 10.9. Basis: this draft (cl. 6.3). Decided by: the Governing Board on adoption of the suite. Comments are invited on whether cl. 10.9 should state the review record obligations of Sections 7 and 10 as well as the five-year floor.

3. Clause 6.4 — the Tier 2 interval. Proposed: 36 months, on the ground that a Tier 2 instrument that changes every two years is not a stable foundation for the tiers beneath it. The alternative is 24 months, to match Tier 3: MCC-100 and MCC-700 have in fact been rewritten twice in six months, and a 36-month cycle describes a stability the Programme does not have. Basis: this draft (cl. 6.4); FD-2026-10-01 item 4. Decided by: the Governing Board on adoption of the suite; the interval is then enacted in MCC-100 and MCC-700 themselves (cl. 6.4). Comments are invited on whether the interval should be 36 or 24 months.

4. Clause 6.5 — the tier model the Register keys to. Proposed: no choice between the tier models of MCC-001 v2.0 and MCC-001 v3.0-draft is proposed here; until it is made, the shorter interval of the two applies to any instrument they place differently, and the Register records both tier values. Basis: this draft (cl. 6.5); FD-2026-10-01 item 4. Decided by: the Governing Board on adoption of the suite. Comments are invited on which tier model the Register should key to, and on the interim rule.

5. Clause 7.6 — CR-2026-001 as the first recorded review. Proposed: CR-2026-001 is entered as the first recorded review of MCC-SDCF v1.0, as an unscheduled review with outcome class D (replace); it satisfies Sections 9 and 10 in substance. Entering it records a finding and does not decide the status of MCC-SDCF, which is the Board's (cl. 9.4; FD-2026-09-04 item 6). Basis: this draft (cl. 7.6); FD-2026-10-01 item 4. Decided by: the Governing Board on adoption of the suite. Comments are invited on whether CR-2026-001 should be entered, and with which outcome class.

6. Clause 12.7 — flagging an overdue instrument on its own page. Proposed: an instrument overdue beyond a stated period is flagged on its own published page, on the reasoning of cl. 5.3; this draft states no period. The argument against is that a prominent flag reads as a warning about the instrument's soundness when it is a statement about the Programme's diligence. Basis: this draft (cl. 12.7); FD-2026-10-01 item 4. Decided by: the Governing Board on adoption of the suite. Comments are invited on whether the flag should be shown and, if so, after what period.

7. Clause 14.2 — one review cycle or two. Proposed: no proposal is made — either answer is defensible, and this draft takes no position and prepares no amendments; having neither answer is what produced the five periods in Schedule 1. Basis: this draft (cl. 14.2); FD-2026-10-01 item 4. Decided by: the Governing Board on adoption of the suite. Comments are invited on whether, on the adoption of this Policy, each instrument's own review clause should be amended at its next amendment to read "reviewed in accordance with the Instrument Review Cycle Policy", or whether instruments should keep their own periods with this Policy as a backstop.

8. Annex, question 11 — where this Policy sits. Proposed: a standalone Tier 5 policy, under the code MCC-IRC assigned by the programme administration for this consultation; tier and code are entered in the Document Register on adoption. Basis: this draft (document control; cll. 16.1–16.2); FD-2026-10-01 item 5. Decided by: the Governing Board on adoption of the suite. Comments are invited on whether this Policy should instead be a section of the Rule Change and Public Consultation Procedure.

---

1. Purpose and scope

1.1 This Policy SHALL establish a single review cycle for every instrument of the MCC Programme: when each is reviewed, who reviews it, what the review must consider, what it produces, where the result is recorded, and what happens when a review is not done on time.

1.2 This Policy applies to every normative instrument of the Programme and to every annex, schedule, appendix, control table, form and template inside one — the Charter, the Programme Standard and the governance framework, the domain standards, the methodologies and tools, the operational policies and procedures, the counterparty terms, the applicant guidance and the report templates.

1.3 This Policy applies to an instrument that is in force. Clause 6.9 states what it does to a consultation draft, and clause 6.10 to an instrument that is stayed or has never commenced.

1.4 This Policy is about the scheduled examination of a text that may be entirely correct. It is not a correction mechanism and not a change mechanism. Section 10 draws both boundaries, and the reader who has arrived here looking for one of those is directed there first.

1.5 This Policy creates no governance body, confers no decision authority, alters no threshold, and changes no rule in any instrument. It fixes a maximum interval, a register and a record.

---

2. Authority

2.1 The Programme's standing commitment to review its own texts is in the Charter, and it is a Tier 1 commitment. MCC-001 v2.0 clause 0065 provides:

"Adaptive Management Commitment: MCC commits to reviewing and updating standards, methodologies, and operational procedures at least every five years or upon material scientific or policy change, and shall engage stakeholders in all major review processes."

2.2 Clause 0010 of the same Charter provides that "the MCC programme shall be reviewed and assessed periodically", and clause 0038 that "MCC shall review, monitor, and adapt its standards and operations in response to scientific advances, operational experience, and changing environmental and social contexts". Neither states a period, a performer or a record. Clause 0065 states a period and nothing else.

2.3 The successor draft carries the commitment forward in the same terms. MCC-001 v3.0-draft clause 10.9 ("Adaptive review") provides that "the Programme SHALL review every standard and methodology at least every five years, and earlier upon material scientific, legal or policy change, with stakeholder engagement in every major review". It too states a period and nothing else.

2.4 "At least every five years" is a floor, not a schedule. It fixes the longest permissible interval for standards and methodologies. It does not say who performs the review, what the review must consider, what it produces, where the result is recorded, whether a review that finds nothing wrong is a result, or what follows if the five years pass and nothing happens. This Policy supplies those six things and does not touch the floor.

2.5 The Charter also requires the register this Policy extends. MCC-001 v2.0 clause 0086: "The MCC Board shall maintain and publish a document registry showing all current documents, versions, effective dates, and amendment history." MCC-001 v3.0-draft clause 12.6 restates it: the Document Register lists every document "with tier, status, owner, version, effective date and dependencies". Neither list includes a review date, and the register as built when this draft was prepared carried no column for one (clause 7.2); review columns were added on 4 September 2026 (PR #244), and none is populated.

2.6 The publication obligation is the Transparency Policy's. Article 4 fixes 5 business days for governance decisions and policy amendments; MCC-700 v3.1 §14.4 requires quarterly transparency reporting including "audit findings and corrective actions". Section 11 of this Policy uses those two and adds no new cadence.

2.7 This Policy takes no authority it does not have. Clauses 6.3 and 6.4 mark the two rows of the cycle table that a Tier 5 instrument cannot enact, and propose them for enactment at the tier that can.

---

3. Definitions

3.1 In this Policy:

Instrument means any document of the Programme with normative effect, and includes an annex, schedule, appendix, control table, form or template inside another instrument. A schedule inside another instrument is an instrument for the purposes of this Policy. That sentence exists because MIR-2026-001 recorded a text as non-existent twice for want of it.

Review means a scheduled examination of an instrument in force against the matters in Section 9, conducted to a conclusion recorded under Section 10.

Review interval means the maximum permitted period between the reference date of one review of an instrument and the reference date of the next.

Due date means the date on which the review interval expires.

Overdue means that the due date has passed and no review outcome has been recorded.

Reference date means the date from which the review interval runs: for a first review, the date the instrument entered into force; afterwards, the date the last review outcome was recorded.

Review outcome means one of the five findings in clause 10.2, recorded and published.

Own-clause period means a review period stated by an instrument about itself.

Register means the Instrument Review Register constituted by Section 7.

Reviewer means the person or persons who conduct a review and sign its outcome.

3.2 Where a term is defined in the MCC-100 Annex — Definitions Register, that definition prevails. This Policy defines only terms the Definitions Register does not.

3.3 "In force" is used in this Policy in its strict sense — an instrument whose commencement provision has been satisfied. It is not a synonym for "published", "adopted", "served" or "listed in the register".

3.4 In particular, this Policy does not treat MCC-700 as in force. MCC-700 v3.1 clause 18.1 provides that the Framework "shall enter into force upon approval by the Governing Board on [Date to be inserted]". The date was never inserted; the text is headed "Status: DRAFT — For Foundational Review". document_register records MCC-700 v3.1 as in force (FD-2026-08-24 item D11), with adopted_on null. This Policy does not assert that any duty in MCC-700 binds anyone, and where it cites MCC-700 it does so as evidence of what the Programme has written, not of what obliges it.

---

4. What the corpus actually says about review

4.1 Schedule 1 reproduces, in full and with citations, all thirty-five provisions in the in-force corpus and in the rewritten suite that provide for the review of an instrument or of something an instrument owns. This Section states the findings that follow from it. No finding in this Section is drafted away in the Sections that follow; each is carried into Section 6, Section 9 or the Annex as an open question or an express rule.

4.2 Finding 1 — five periods are in force at once. The corpus states, of different instruments, review periods of six months, twelve months, twenty-four months, thirty-six months and sixty months. The distribution follows no principle: it does not track tier, risk, subject matter, rate of change or external dependency. MCC-400 (a Tier 3/4 standard governing third-party assurance) is reviewed every two years; the Host Country Authorisation Procedure (Tier 5) annually; the Transparency Policy (Tier 5) every six months; the Environmental and Social Safeguards Policy (Tier 3 in the successor model) every five.

4.3 Finding 2 — four instruments state more than one period about themselves.

(a) MCC-SDCF v1.0 states six review obligations for itself, on three different cycles. Clause 0066 requires the indicator framework "reviewed and updated annually by the methodology expert panel". Clause 0131 requires the label criteria "reviewed annually". Clause 0171 requires the Expert Panel to "review the SD framework annually". Clause 0033 requires the SDG alignment framework updated "every three years". Clause 0181 requires third-party evaluations "every three years". Clause 0177 requires "periodic (5-year) comprehensive reviews". Read together, the same framework is on an annual, a three-yearly and a five-yearly cycle simultaneously, and its successor draft says so: cl. 12.4's drafting note calls it "a calendar of obligations no one will meet".

(b) The Claims Policy v1.1 contradicts itself on its own cover. Clause 11.2 provides: "Policy reviewed annually (March)." The document's header line reads "Next Review: September 23, 2026". The document's footer line reads "Next Review: March 23, 2027". The header and the clause cannot both be right; the footer and the clause agree. The served adopted text carries all three (content/adopted/claims-policy.md lines 12, 632 and 653).

(c) MCC-HCA v1.0 states the same duty twice, in two places, in different words — clause 0010 ("This procedure shall be reviewed and updated annually to reflect evolving international climate frameworks and host country practices") and clause 0185 ("This procedure shall be subject to annual review to ensure alignment with evolving international climate frameworks and host country practices"). Both are annual, so nothing turns on the conflict; it is recorded because it shows how the clauses were written.

(d) MCC-800 v1.0 sets a period for itself and a different one for a schedule it owns — art. 111, the Standard "at least annually"; art. 101, the fee schedule "annually". Consistent, and noted for completeness.

4.4 Finding 3 — the most important instruments have no review clause at all. Neither MCC-100 (Programme Standard) nor MCC-700 (Governance and Appeals) contains any provision for its own review, in any version. MCC-100 v2.1 was searched in full: it has no such clause. MCC-700 v3.1 §17 provides an amendment procedure and §18 an entry-into-force provision, and nothing between them. The two instruments on which every other instrument depends are the two with no review provision. They are not alone: the Applicant, Registry and VVB Terms, the Legal and Regulatory Positioning paper, the MCC-100 Annex — Definitions Register, the PDD Template, the F01–F05 guidance, the Ten Operational Umbrella Methodologies and the Programme Bible provide for no review of themselves either. A majority of the corpus by instrument count has no review clause.

4.5 Finding 4 — the rewritten suite reproduces the defect, and in one place makes it worse. Of the twenty-five consultation drafts examined on 4 September 2026, nine contain no clause providing for their own review, including MCC-100 v3.0, MCC-700 v4.0, MCC-200 v3.0, MCC-800 v2.0, MCC-600, MCC-820, MCC-CADCP v2.0, MCC-BPM v2.0 and MCC-ROI. And MCC-400 v2.0-draft drops the biennial review its predecessor had: MCC-400 v1.1 art. 90 requires review "at least every 2 years"; the successor's §22 is amendment and entry into force only. This is a loss, and it is recorded here because a consultation is the right place to raise it. This draft's position is that MCC-400 v2.0 relies on this Policy's Tier 3 interval, and that carrying art. 90's biennial review forward would create a sixth period. (Consultation question 1.)

4.6 Finding 5 — one successor draft rejects the calendar altogether, and gives a reason worth taking seriously. MCC-SDCF v2.0-draft clause 12.4: "This Framework SHALL be reviewed when the Programme has completed ten SD assessments, or after three years, whichever is sooner." Its drafting note explains: "the predecessor required annual review, five-yearly comprehensive review, three-yearly third-party evaluation, annual stakeholder webinars and an annual standards document. For a programme with no completed assessment, that is a calendar of obligations no one will meet, and an unmet obligation in a standard is worse than none. Review is triggered by evidence instead." Clause 6.8 of this Policy adopts that reasoning without adopting the mechanism, and the Annex puts the question.

4.7 Finding 6 — the named review bodies do not exist. MCC-850 v1.1 art. 45 assigns the annual review of that Standard to "Technical Committee and AI Governance Committee"; the AI and Human Oversight Policy §15.1 gives quarterly amendment proposals to the same AI Governance Committee; MCC-HCA v1.0 cl. 0186 creates a "Review Committee" of "internal policy staff, external legal advisors, and host country representatives" and cl. 0187 requires it to "meet annually"; MCC-SDCF cl. 0066 assigns the annual indicator review to a "methodology expert panel" and cl. 0171 the annual framework review to an "Expert Panel"; MCC-700 v3.1 §7.1.3.2 assigns it to the "SD Expert Panel" and §4.5.3 gives annual reporting to Methodology Working Groups; MCC-400 v1.1 art. 90 requires the input of a "VVB Advisory Panel". Not one of those bodies is constituted, and eight of the thirty-five review provisions in Schedule 1 are assigned to one of them. The platform's governance_bodies table held fourteen body records across two governance epochs on 4 September 2026 and holds seventeen on 1 October 2026; it contains no Technical Committee, no AI Governance Committee, no Review Committee and no VVB Advisory Panel, and the Methodology Review Committee record added on 8 September 2026 is marked NOT CONSTITUTED. Of the bodies that are recorded, appointments exist against one. The Technical Committee appears in seven instruments and is constituted in none (MIR-2026-001 rev 3 §3.8, correction C2). MCC-HCA v2.0-draft clause 9.5 proposes to abolish the Review Committee expressly: "no Programme document may create a governance body other than the seven of MCC-001 clause 9.1.1".

4.8 Finding 7 — no review has ever been performed under any review clause. Verified against all four shelves and the successor drafts on 4 September 2026:

Where a completed review would appear	What is there
decision_register	6 rows (MCC-BDR-2026-002, FD-2026-08-24, GB-2026-001, FD-2026-09-01, FD-2026-09-03, FD-2026-09-04). None records a review outcome.
document_register	37 rows. No column exists in which a review date could be recorded (clause 7.2).
review_findings (platform)	0 rows.
technical_reviews (platform)	0 rows.
Served manifest lib/documents/manifest.ts	16 published standards, each with a SHA-256 and a revision state. No review date on any.
Drive 00_Styring_og_Vedtak	No review outcome record of any instrument under any review clause.
The rewritten suite	No draft recites a completed review of its predecessor performed under that predecessor's review clause.

Re-checked on 1 October 2026: decision_register holds seven rows (FD-2026-09-07 added), none a review outcome; document_register holds 44 rows and carries review columns (review_cycle_months, last_reviewed_on, next_review_on, review_basis; PR #244), none populated; review_findings and technical_reviews hold 0 rows.

4.9 Finding 8 — four reviews have nevertheless been done, and none of them was scheduled. The rulebook audit of 24 August 2026; the independent programme-integrity desk review of 3 September 2026 accepted by FD-2026-09-03; the SDCF Conformity Review CR-2026-001 of 3 September 2026; and MIR-2026-001 in three revisions. Each was thorough. CR-2026-001 in particular is the model this Policy generalises: it reviewed one framework against the suite, the register and the deployed platform, graded fourteen findings, recorded what was sound as well as what was not, and closed with a reasoned recommendation that the Board — not the reviewer — decide the instrument's status. But none of the four was performed under the reviewed instrument's own review clause, on its schedule, or by the body that clause names. MCC-SDCF cl. 0171 gives that review to an Expert Panel that does not exist; the review that actually happened was done by programme administration because someone noticed. That is the difference between a review cycle and good luck.

4.10 Finding 9 — seven instruments fall due on the same day, and nothing is arranged to notice. The Transparency Policy, Claims Policy, AI and Human Oversight Policy, Applicant Terms, Registry Terms, VVB Terms and Legal and Regulatory Positioning paper each print "Next Review: September 23, 2026" on their cover, and the Transparency Policy and AI and Human Oversight Policy also print it in their footers. All seven were dated 23 March 2026 and re-adopted as v1.1 on 24 August 2026 by FD-2026-08-24 (recorded in the decision register as attested by the founder, with no executed copy filed). 23 September 2026 was nineteen days from this draft and was the date set for the next Governing Board meeting. No item on the board pack GB-2026-002 v2.0 was the review of any of them. No record of a Governing Board meeting on 23 September 2026 has been filed, and no review of any of the seven is recorded in the registers. Clause 6.11 and Schedule 2 address this.

4.11 Finding 10 — a review date printed on a cover is not a review clause, and four of these seven have no clause. The Applicant, Registry and VVB Terms and the Legal and Regulatory Positioning paper print "Next Review: September 23, 2026" and contain no clause providing for a review at all. A date on a cover with no clause behind it tells a reader that a review will happen and creates no duty on anyone to do it. No instrument fixing 23 September 2026 as the review date of these four texts is in the programme record: none is cited on any of them, and no decision in the decision register fixes one (the four texts as served in content/adopted; the decision register at /governance/decisions).

---

5. The principle

5.1 A review that produces no record is indistinguishable from no review. This is the whole of the Policy's reasoning and every operative Section follows from it.

5.2 It follows that a finding of no change required is a result, and SHALL be recorded and published exactly as a finding of change required is (clause 10.3). An instrument that was examined carefully and found sound, and an instrument that nobody has opened since it was written, look identical from outside unless the first is recorded.

5.3 It follows that the record SHALL live where a reader looking at the instrument will find it (clause 7.1), and not in a file that only the reviewer knows exists.

5.4 It follows that an overdue review SHALL be visible as overdue (Section 12), because the failure this Policy exists to prevent is silent omission, not late delivery.

5.5 It follows that the count of reviews done, due and overdue SHALL be published whether or not it is flattering (clause 11.4), on the same reasoning MCC-REST clause 12.5 gives for the correction count.

---

6. The review cycle

6.1 Every instrument in force SHALL be reviewed at intervals no longer than those in the table below, and the interval SHALL run from its reference date.

Tier (MCC-001 v3.0-draft cl. 12.1)	Instruments	Maximum interval	Enacted by
1	MCC-001 Charter and Constitution	60 months, and the annual governance-effectiveness assessment of MCC-001 cl. 0054 continues separately	Not by this Policy — clause 6.3
2	MCC-100 Programme Standard; MCC-700 Governance and Appeals	36 months	Not by this Policy — clause 6.4
3	Domain standards: MCC-200, MCC-400, MCC-800, MCC-850, MCC-600, MCC-ESS, MCC-SDCF, MCC-BPM, MCC-CADCP	24 months	This Policy
4	Methodologies, tools and calculators	24 months, and in any event before the first issuance of any unit under a methodology never yet used	This Policy
5	Procedures, forms, templates, guidance, counterparty terms	24 months	This Policy

6.2 The interval in clause 6.1 is a maximum, not a schedule. Where an instrument states its own shorter period, the instrument's own period governs and is unaffected by this Policy. This follows from the precedence rule itself: MCC-001 v2.0 cl. 0084 provides that where texts conflict the higher tier prevails, and a Tier 5 policy cannot lengthen the review cycle a Tier 1 to Tier 4 instrument sets for itself. This Policy shortens no instrument's cycle and lengthens none.

6.3 This Policy does not fix the Tier 1 interval and cannot. A Tier 5 policy may not bind the amendment or review of the Constitution. The 60 months in row 1 is exactly the floor MCC-001 v2.0 cl. 0065 and MCC-001 v3.0-draft cl. 10.9 already state, restated here so that the table is complete and the Register can carry a due date for MCC-001. It takes effect as a Charter obligation or not at all. Whether MCC-001 v3.0-draft cl. 10.9 should also state the review record obligations of Sections 7 and 10, so that Tier 1 is inside the cycle rather than beside it, is not decided. (Consultation question 2.)

6.4 This Policy does not fix the Tier 2 interval and cannot. MCC-100 and MCC-700 are Tier 2 under MCC-001 v3.0-draft cl. 12.1 and are amended by the Independent Governing Board after Standards Committee recommendation and 60 days' consultation (cl. 12.3). Neither states a review period today (clause 4.4). The 36 months in row 2 is a proposal to be enacted in those Standards themselves, and is recorded in the Register as PROPOSED, not DUE, until it is. (Consultation question 3.)

6.5 Two tier models are in force at once, and the table above uses only one of them. MCC-001 v2.0 cll. 0079–0083 place MCC-700 at Tier 2, MCC-100 at Tier 3, MCC-200 and the methodologies at Tier 4, and the operational policies (MCC-ESS, MCC-SDCF, MCC-HCA, MCC-CADCP and others) at Tier 5. MCC-001 v3.0-draft cl. 12.1 places MCC-100 and MCC-700 together at Tier 2, the nine domain standards at Tier 3, methodologies and tools at Tier 4, and procedures, forms, templates and guidance at Tier 5. document_register records tiers on the v2.0 model. The table in clause 6.1 uses the v3.0-draft model, because the whole rewritten suite does. The two models give different answers for MCC-100, MCC-200, MCC-400, MCC-800, MCC-850, MCC-ESS, MCC-SDCF, MCC-BPM and MCC-CADCP. This is a live conflict and it is not this Policy's to resolve. Until it is decided, the shorter interval of the two models applies to any instrument the models place differently, and the Register records both tier values so the divergence is visible rather than averaged away. (Consultation question 4.)

6.6 An instrument SHALL be reviewed on entry into force plus its interval, and thereafter on each review outcome plus its interval. A review conducted early resets the interval from the date the outcome is recorded; it does not bank the remainder.

6.7 Due dates SHALL be staggered. The Register SHALL be maintained so that no more than six instruments fall due in any calendar quarter, and where the intervals in clause 6.1 would produce more, the Programme Secretariat SHALL bring the excess forward — never defer them — and record which were moved and why. Clause 4.10 is the reason: seven instruments falling due on one day, as they did on 23 September 2026, is not a cycle. It is a cliff.

6.8 An event brings a review forward. A review SHALL be conducted, whatever the interval, within 60 days of any of the following:

(a) an amendment, restatement or entry into force of an instrument on which the reviewed instrument depends; (b) a correction at rung R3 or R4 under MCC-REST affecting the reviewed instrument; (c) the constitution of a body to which the reviewed instrument assigns a duty, or the abolition of one; (d) an audit, conformity review or integrity review that makes a finding against the reviewed instrument; (e) an incident, appeal, grievance or non-conformity whose cause is traced to the reviewed instrument's text; (f) a change in applicable law or in an external standard the instrument cites; or (g) the first application of the instrument to a real case, where it has never been applied before.

6.9 A consultation draft is not reviewed under this Policy. It is under consultation, which is a more searching examination than a review. The interval for an instrument adopted out of consultation runs from its entry into force.

6.10 A stayed instrument and an instrument that has never commenced are recorded in the Register and are not due. MCC-400 Addendum 1 and MCC-850 Amendments 1 and 2 (all stayed by FD-2026-09-03 item D8) carry the status NOT IN FORCE and no due date; how MCC-700 is recorded is stated at clause 3.4. They are not thereby exempt: clause 6.8(c) applies on commencement, and the Register SHALL show the reason no date is set, so that "no due date" is never read as "reviewed".

6.11 First reviews under this Policy. Every instrument in force on the day this Policy commences SHALL be reviewed within twelve months of that day, staggered under clause 6.7, whatever its interval would otherwise give. None has ever been reviewed (clause 4.8) and a cycle cannot start from a reference date that does not exist.

---

7. The Instrument Review Register

7.1 The Programme SHALL maintain an Instrument Review Register, publicly accessible, which for every instrument records:

(a) the instrument code, title, version and tier (both tier values while clause 6.5 stands unresolved); (b) its status — IN FORCE / STAYED / NEVER COMMENCED / SUPERSEDED / DRAFT; (c) its reference date, and whether that date is an entry into force or a recorded review outcome; (d) its own-clause period, quoted, or the express entry "none stated"; (e) the interval that applies and which of the two it is (clause 6.1 or clause 6.2); (f) the due date, or the reason there is none; (g) the date of the last review and the outcome class under clause 10.2; (h) the reviewer, and the body to which the outcome was reported; (i) where the review record can be read; and (j) whether the instrument is currently OVERDUE, and by how many days.

7.2 The Register SHOULD be implemented as columns on the existing document_register rather than as a separate table, and this draft proposes that it is. When this draft was prepared, that register carried 37 rows and fifteen columns — code, title, tier, version, owner_role, status, adopting_instrument, adopted_on, supersedes, superseded_by, successor_draft, note, sort_order, updated_at, parent_code — and not one of them could hold a review date. Review columns (review_cycle_months, last_reviewed_on, next_review_on, review_basis) were added on 4 September 2026 (PR #244); on 1 October 2026 the register holds 44 rows, and none of the four review columns is populated in any of them. A separate register would be a second shelf, and the Programme has recorded a document as non-existent five times in two days for searching one shelf.

7.3 An entry stating "none stated" under clause 7.1(d) SHALL be printed as a finding, not left as an empty cell. This follows the discipline the instrument register adopted on 4 September 2026 when parent_code was added so that a missing annex would print as an absence rather than as nothing at all.

7.4 The Register SHALL be reachable from every published instrument and from the transparency documents page, so that a reader holding an instrument can establish when it was last examined without knowing that a register exists.

7.5 A Register entry SHALL NOT be edited to change what it records of a past review. A correction to a Register entry is made by a further entry that references the first, and by a correction note under MCC-REST.

7.6 Reviews performed before this Policy commences MAY be entered retrospectively, and SHOULD be, where an outcome record exists that satisfies Section 10. On the evidence in clause 4.8 the only candidates are the four unscheduled reviews of clause 4.9; whether any qualifies is a question of fact for the Programme Secretariat, and an entry SHALL record that the review was unscheduled and under which authority it was done. This draft's position is that CR-2026-001 is entered as the first recorded review of MCC-SDCF v1.0 — it satisfies Section 9 and Section 10 in substance — with outcome class D (replace), not A, because recording it makes the Register truthful from its first day rather than empty. (Consultation question 5.)

---

8. Who performs a review

8.1 A review SHALL be performed by the Programme Secretariat, and until the Programme Secretariat is constituted, by the person or persons exercising the Programme Administrator function.

8.2 A review outcome SHALL be reported to the Standards Committee, and a review outcome of class C or D under clause 10.2 SHALL be placed before the body that the Delegated Authority Matrix (MCC-700 Annex A) assigns the amendment of that instrument to.

8.3 This Policy constitutes no body and assigns no duty to a body that does not exist. MCC-001 v3.0-draft clause 9.1.2 provides that "no Programme document SHALL create, name or empower any governance body other than those in clause 9.1.1, except by amendment of this Constitution", and this Policy does not attempt it. In particular it assigns nothing to the Technical Committee, the AI Governance Committee, the Review Committee, the VVB Advisory Panel, the Methodology Review Committee, the Methodologies Panel or the SD Expert Panel — seven bodies named across the corpus, none constituted (clause 4.7).

8.4 Where a duty in clause 8.2 falls on a body not yet constituted, it attaches on that body's constitution, and until then the review outcome SHALL be prepared, recorded and published as Section 10 requires, and placed before the body at its first meeting competent to take it. A review is not deferred because the body that would receive it does not exist. Deferring it leaves the instrument unexamined and the omission unrecorded, which is the state this Policy exists to end.

8.5 The Programme Secretariat SHALL NOT record a review as decided when what has happened is that it was prepared.

8.6 The reviewer SHOULD NOT be the sole author of the text under review, and where that is unavoidable the review record SHALL say so. The Programme is small enough that this will often be unavoidable, and saying so is worth more than a rule that would be quietly broken.

8.7 A reviewer SHALL declare any interest bearing on the instrument under review in accordance with the Register of Interests and Conflict of Interest Screening Procedure (MCC-ROI) on its entry into force. Until then, the declaration SHALL be recorded in the review record itself.

8.8 A review MAY draw on external advice, and where it does the review record SHALL state who advised and on what. This follows MCC-SDCF v2.0-draft clause 12.3, and it is stated here so that the rule is general rather than sitting in one framework.

8.9 This Policy states no rule about VVB competence, impartiality, rotation, cooling-off or oversight, and nothing in it is to be read as one. Those requirements are reserved exclusively to MCC-400 (v2.0-draft cl. 1.4). The review of MCC-400 itself under this Policy is an examination of the text, not of any body accredited under it.

---

9. What a review must consider

9.1 A review SHALL consider each of the following and SHALL record a conclusion on each. A review that does not reach the whole list is not a review under this Policy and SHALL be recorded as incomplete.

(a) Authority. Is every clause, decision, instrument and body the text cites in force, recorded, signed and constituted? A citation to a stayed provision, an unsigned decision or a body that does not exist is a defect of authority, and the review SHALL identify it as such rather than as a wording problem. This limb exists because the Programme published a page "under MCC-400 art. 95.1", a provision that has never been in force.

(b) Internal consistency. Does the text contradict itself — in a clause, in a cover line, in a footer, in a table, in an annex? The Claims Policy states three different next-review dates in one document (clause 4.3(b)); MCC-SDCF states five review periods for itself (clause 4.3(a)). A review that reads only the numbered clauses will find neither.

(c) External consistency. Does the text agree with every instrument above it in the hierarchy, with every instrument it cites, and with every instrument that cites it? Where it does not, which prevails under MCC-001's precedence rule, and is that stated anywhere a reader would find it?

(d) Correspondence with the world. Does the text describe things that exist — bodies, systems, registers, tables, endpoints, processes, counterparties? CR-2026-001 found a Section of MCC-SDCF specifying four database entities of which two were built, five groups of API endpoints of which none existed, and a public register surface that does not exist. A standard is in breach of itself from the day it is approved if it requires what was never built.

(e) Operational experience. What has actually happened under the text since it was last reviewed — decisions taken, findings made, appeals brought, grievances raised, corrections published, questions asked by applicants or counterparties, clauses that no one could apply? Where nothing has happened under it, the review SHALL say so; an instrument that has never been applied is a different kind of risk from one that has.

(f) Implementability. Can the text be complied with by the people it binds, with the resources and bodies that exist? MCC-SDCF v2.0-draft clause 12.4's drafting note is the reasoning: "an unmet obligation in a standard is worse than none".

(g) Whether anything in it has been overtaken — by a successor draft, by an amendment to another instrument, by a decision, by a change in law, or by an external standard it cites.

(h) Whether it should exist at all. A review MAY conclude that the instrument should be withdrawn, or merged into another, and clause 10.2(e) provides the outcome for it.

9.2 A review SHALL check the instrument against all four shelves and the successor draft: the review corpus, the served corpus (content/adopted and lib/documents/manifest.ts), document_register, and Drive. A review that concludes something is absent SHALL NOT record that conclusion until it has done so, and SHALL record that it did. This clause is drafted from MIR-2026-001 rev 3 §0, where three successive revisions each corrected a "does not exist" finding produced by searching one incomplete shelf.

9.3 A review SHALL record what is sound as well as what is not. CR-2026-001 has a section headed "What is sound and should be carried forward", and it is the reason that review is usable as a basis for a rewrite rather than only as a complaint. A review that lists only defects gives the drafter of the successor nothing to preserve.

9.4 A review SHALL NOT decide the status of the instrument it reviews. Status is the Board's. A review records findings and recommends; clause 10.5 states the consequence.

---

10. What a review produces

10.1 Every review SHALL produce a written outcome, recorded in the Register and published.

10.2 The outcome SHALL be classified as exactly one of:

Class	Outcome	What follows
A	No change required	Recorded, published, interval resets. Nothing else happens, and that is the point of the class
B	Correction required — the text says something that is wrong, and putting it right changes no rule	Hands over to MCC-REST, at the rung MCC-REST determines. This Policy takes no view on the rung
C	Amendment required — the text should say something different	Hands over to the Rule Change and Public Consultation Procedure, at the tier and notice period MCC-001 cl. 12.3 fixes
D	Replacement required — the defects run through the text and amendment would touch most of it	Rewrite and consult as a new version, on the reasoning CR-2026-001 gives for MCC-SDCF: "Amendment would require touching Sections 1, 2, 6, 7, 8 and 9 — that is most of the document"
E	Review not completed — impediment recorded	The review could not be concluded. The impediment is named, the instrument stays due, and the entry stays open

10.3 Class A is a result and SHALL be recorded and published in the same way and with the same prominence as any other class. An instrument recorded as reviewed with no change required, and an instrument nobody has opened, are indistinguishable to a reader unless the first is recorded. This clause is the operative core of the Policy and it is short on purpose.

10.4 Class A SHALL NOT be recorded by default, by silence, or by the expiry of a period. It is recorded only where a reviewer has done the work in Section 9 and signed the conclusion. A review outcome that records class A without a conclusion on each limb of clause 9.1 is void and the instrument remains due.

10.5 Class E is not a failure of the reviewer and SHALL NOT be avoided by recording class A. An instrument may be genuinely un-reviewable: it may assign its duties to a body that does not exist, depend on an instrument that has never commenced, or specify a system that was never built. MCC-850 art. 45 is the standing example — its annual review is assigned to two committees, neither of which exists. Recording "the review could not be performed, and here is why" is a true statement about the Programme; recording "no change required" would not be.

10.6 A review outcome SHALL record, at minimum:

(a) the instrument, version and the exact text reviewed, identified by SHA-256 where the served corpus carries one; (b) the date of the review and the reviewer; (c) the conclusion on each limb of clause 9.1; (d) the outcome class, with reasons; (e) every defect found, its grade, and what is recommended for it; (f) what was found sound (clause 9.3); (g) any interest declared under clause 8.7 and any external advice taken under clause 8.8; (h) where the outcome was reported and on what date; and (i) the next due date.

10.7 A review outcome SHALL be published within 5 business days of being signed. That is the Transparency Policy art. 4 period for governance decisions and policy amendments, and a review outcome is not entitled to a slower cadence than the amendment it may recommend.

10.8 A review outcome SHALL NOT be withheld because it is unwelcome, because a governance body has not met, because a body competent to act on it has not been constituted, or because the remedy is not yet drafted. Where the outcome requires a decision that cannot yet be taken, it is published with what remains undecided identified (clause 8.4).

10.9 A review outcome of class B, C or D does not of itself change the instrument. The instrument stays in force in the text it has until it is corrected under MCC-REST or amended under the Rule Change Procedure. A review is a finding, not a decision.

---

11. Reporting

11.1 The Programme Secretariat SHALL report every review outcome to the Standards Committee, and to the Independent Governing Board where the outcome is class C, D or E.

11.2 A review outcome that changes a premise of a paper a governance body is due to consider SHALL be reported before that paper is taken.

11.3 The Programme SHALL publish, quarterly, a review summary stating, for the period: the number of reviews completed by outcome class; the number due and the number completed on time; the number overdue and the age of the oldest; the number recorded as class E and the impediment in each case; and the number of instruments in force that have never been reviewed.

11.4 The summary SHALL be published on the same page and cadence as the quarterly transparency reporting MCC-700 §14.4 contemplates, and alongside the correction summary MCC-REST clause 12.3 requires. It is a count, not a narrative. A high overdue count is the finding, and SHALL NOT be managed down by recording class A without the work.

11.5 The summary SHALL state, until it is no longer true, the number of instruments in force whose review was never provided for by any instrument — which on 4 September 2026 is a majority of the corpus (Schedule 1).

---

12. Overdue reviews

12.1 An instrument is overdue on the day after its due date if no outcome has been recorded, and the Register SHALL show it as overdue from that day, with the elapsed days.

12.2 An overdue review does not affect the force, validity or effect of the instrument. An instrument does not lapse, weaken, or fall out of force because it was not reviewed on time, and no decision taken under it is impeached by the omission. This clause is stated expressly because the opposite reading would be catastrophic for a Programme with a corpus this overdue, and because a reader is entitled to know which it is.

12.3 Where an instrument is overdue by more than 30 days, the Programme Secretariat SHALL record in the Register the reason and the date by which the review will be completed.

12.4 Where an instrument is overdue by more than 90 days, the matter SHALL be placed before the Independent Governing Board at its next meeting as an item in its own right, and the Register entry SHALL name the date it was placed.

12.5 An overdue review SHALL NOT be closed by recording class A without the work in Section 9 (clause 10.4), and SHALL NOT be closed by deferring the due date. A due date is moved only by a recorded decision that says it was moved and why, and moving it does not stop the elapsed-days count from the original date, which the Register SHALL continue to display.

12.6 Where an instrument is overdue because the review is impeded, the outcome is class E (clause 10.5) and the instrument stays due.

12.7 This draft's position is that an instrument overdue beyond a stated period should be flagged on the instrument's own published page, so that a reader of the text sees it, and not only a reader of the Register, on the reasoning of clause 5.3. (Consultation question 6.)

---

13. Boundaries — what this Policy is not

13.1 This Policy and MCC-REST do different work, and the boundary is the state of the text.

	MCC-IRC (this Policy)	MCC-REST
Trigger	The calendar, or an event under clause 6.8	Knowledge that something published is wrong
Premise	The text may be perfectly correct	The text is not correct
Question	Is this still the right text?	How is this untruth put right?
Output	A finding, possibly "no change required"	A correction, at a rung
Timing	The interval in clause 6.1	Immediately to 5 business days from knowledge (MCC-REST cl. 8.3)

13.2 A review that finds an error hands over to MCC-REST at the moment the error is known, and does not wait for the review to be finished. The MCC-REST clause 8.3 periods run from knowledge, not from the review's completion, and a reviewer who holds a known error until the review report is drafted has breached MCC-REST, not complied with this Policy.

13.3 A review that concludes a change is needed hands over to the Rule Change and Public Consultation Procedure, which governs the proposal, the consultation, the decision, the version bump, the publication and the archive (§§3 to 6 of that Procedure). This Policy states no threshold, no notice period, no decision body and no adoption step, and SHALL NOT be read as varying any. The Procedure's consultation closed on 8 September 2026; its closing protocol of 24 September 2026 recommends adoption, and it is not entered in the document register as adopted; clause 15.3 states what follows if it is not.

13.4 This Policy does not vary what a higher tier fixes. It sets no amendment authority, no voting threshold, no notice period, no precedence rule and no tier. Where it names an interval for Tier 1 or Tier 2 it says expressly that it is not enacting it (clauses 6.3 and 6.4). Where an instrument's own clause is shorter, that clause governs (clause 6.2).

13.5 This Policy repeals nothing. Section 9's list of matters and Section 10's outcome classes apply to a review conducted under this Policy; a review conducted under an instrument's own clause continues to be governed by that clause. Clause 14.2 proposes how the two are reconciled and does not do it here.

13.6 This Policy states no rule about the review of a project, a monitoring report, a validation, a verification, an issuance, a unit, an account or a VVB. Those are governance decisions with their own procedures and their own appeal routes. The word "review" carries at least four distinct meanings across this corpus, and this Policy uses it in only one: the periodic examination of an instrument's own text.

---

14. The existing review clauses

14.1 Every review clause listed in Schedule 1 continues in force on its own terms. This Policy adds a floor, a register and a record; it does not displace, shorten, lengthen or repeal any of them.

14.2 The consequence is that some instruments will carry two review obligations — their own and this Policy's. Clause 6.2 resolves the interval (the shorter governs). It does not resolve the duplication, and the duplication is untidy. This draft takes no position on it and prepares no amendments. (Consultation question 7.)

14.3 Nothing in this Policy revives the review duty of a clause whose named body does not exist. MCC-850 art. 45's annual review remains assigned to a Technical Committee and an AI Governance Committee that are not constituted, and this Policy neither constitutes them nor reassigns their work; it records the impediment as class E and reviews MCC-850 under Section 8 in the ordinary way.

---

15. What is and is not in force

15.1 This Policy is not in force. It is a consultation draft. No review is due under it, no instrument is overdue under it, the Instrument Review Register does not exist, and no person owes any duty stated in it.

15.2 No instrument's review status is changed by this document. Every instrument's own review clause stands exactly as it did before this draft was written, and every one of them is as unperformed as it was.

15.3 The two instruments this Policy interlocks with are also drafts. The Rule Change and Public Consultation Procedure v1.0 closed its consultation on 8 September 2026 and is not entered in the document register as adopted (its closing protocol of 24 September 2026 recommends adoption); MCC-REST v1.0, drafted on 4 September 2026, is put to consultation with this suite. If neither is adopted, the hand-overs in clauses 13.2 and 13.3 hand over to nothing, and this Policy would produce findings with no route out of them. That is a reason to sequence adoption, and it is recorded rather than drafted around.

15.4 This Policy does not treat MCC-700 v3.1 as in force (clause 3.4), although the document register records it as in force (FD-2026-08-24 item D11). Nothing in this Policy asserts that a duty in it binds anyone, including the transparency-reporting duty clause 11.4 refers to, which is cited as the intended home for a report and not as an obligation in force.

15.5 MCC-400 Addendum 1 and MCC-850 Amendments 1 and 2 are stayed and have never been in force (FD-2026-09-03 item D8). They appear in the Register under clause 6.10 with no due date and the reason recorded.

15.6 No separate legal entity has yet been incorporated for the Programme (FD-2026-10-01 item 3). This Policy names no controller, no jurisdiction and no governing law, and no obligation in it is stated as a legal obligation of any person to any other.

15.7 Every project row in the platform is a pilot or demonstration. Two projects exist, both SUBMITTED and neither registered; zero credit units have ever been issued; zero SD assessments have been performed; and no VVB is accredited. Nothing in this Policy should be read as describing a pipeline or a registered portfolio.

15.8 What this Policy does not do. It does not review anything. It does not change any instrument. It does not create a body, confer authority, or alter a threshold, a notice period or a tier. It does not decide the tier-model conflict at clause 6.5, the Tier 1 and Tier 2 intervals at clauses 6.3 and 6.4, or the duplication at clause 14.2. It does not correct a single one of the conflicts in Section 4 — it records them, so that the Board decides them once instead of nine drafters deciding them separately again.

15.9 A finding made in preparing this draft. The served document manifest recorded against the Sustainable Development Contribution Framework: "Conformity review against the revised suite is outstanding. No sustainable-development label is awarded until it is completed." That review was completed on 3 September 2026 as CR-2026-001, which found the framework NON-CONFORMING on five blocking grounds and recommended a rewrite. The served statement was untrue from that date until the manifest was corrected on 4 September 2026 (PR #241). It was a class C3 error in a served surface under MCC-REST clause 4.1, material under its clause 5.3 (a status), and requiring R1 within one business day.

---

16. Amendment, review and entry into force

16.1 This Policy is amended by the procedure for a Tier 5 instrument (MCC-001 v3.0-draft cl. 12.3), through the Rule Change and Public Consultation Procedure on its adoption.

16.2 This Policy is subject to itself. It is a Tier 5 instrument and its interval is 24 months, and it SHALL be reviewed additionally on: the constitution of the Standards Committee or the Programme Secretariat; the entry into force of MCC-REST or of the Rule Change and Public Consultation Procedure; the resolution of the tier-model conflict at clause 6.5; and any quarter in which the review summary records more than three instruments overdue by 90 days.

16.3 This Policy SHALL enter into force upon approval under MCC-001 clause 12.3 for a Tier 5 instrument — during the transitional period, upon a Founder Decision recorded and tabled for ratification — on a stated date, following the public consultation on this draft.

16.4 Every point listed under Consultation questions SHALL be resolved, and its resolution recorded, before this Policy may carry any status other than DRAFT — FOR CONSULTATION.

---

Schedule 1 — Every review clause in the corpus (normative reference, non-normative commentary)

Compiled 4 September 2026 by extraction over the 34 in-force texts of the review corpus, the served corpus at content/adopted, the 40 files in content/proposals, and Drive. Quotations are exact. The right-hand column is commentary.

Part A — Instruments in force

Instrument	Clause	What it says	Period	Performed by	Commentary
MCC-001 v2.0	0010	"The MCC programme shall be reviewed and assessed periodically to ensure ongoing alignment with international best practices and evolving scientific understanding."	none stated	not stated	"Periodically" is not a period
MCC-001 v2.0	0038	"MCC shall review, monitor, and adapt its standards and operations in response to scientific advances, operational experience, and changing environmental and social contexts."	none stated	MCC	A principle, not a cycle
MCC-001 v2.0	0054	"The Board shall annually assess the effectiveness of the governance structure and shall make amendments as necessary."	12 months	Board	Never done
MCC-001 v2.0	0065	"MCC commits to reviewing and updating standards, methodologies, and operational procedures at least every five years or upon material scientific or policy change."	60 months	MCC	The only Tier 1 period. The floor in clause 6.1 row 1
MCC-001 v2.0	0075	"The Board shall annually review all external claims for consistency with this section and shall correct any non-compliant claims."	12 months	Board	Never done
MCC-100 v2.1	—	No clause provides for the review of MCC-100.	—	—	Finding 3
MCC-100 v2.1	0249	"Programme-wide lessons learned shall be documented and incorporated into periodic revisions to MCC-100 and related standards."	none stated	not stated	Assumes a cycle that no clause creates
MCC-100 v2.1	0255	"Programme amendments identified as necessary during the Track Record period shall be implemented promptly without waiting for formal revision cycles."	—	—	Refers to revision cycles that do not exist
MCC-100 v2.1	0135	"Corresponding adjustment records shall be reviewed and updated annually."	12 months	not stated	Records, not the instrument
MCC-100 v2.1	0273	"AI governance decisions shall be made by the Governance Board and shall be reviewed annually."	12 months	Governance Board	Decisions, not the instrument
MCC-200 v2.0	0010	"This Framework shall be reviewed and updated no less than every five years or upon material scientific or policy change."	60 months	not stated	Mirrors MCC-001 cl. 0065
MCC-200 v2.0	0123	"The locked portfolio shall be reviewed annually by the Board."	12 months	Board	Portfolio, not the Framework
MCC-400 v1.1	art. 90	"This Standard shall be reviewed at least every 2 years by the Programme Administrator, with input from accredited VVBs and the VVB Advisory Panel, and recommendations for amendment submitted to the Governance Board."	24 months	Programme Administrator + VVB Advisory Panel	The panel is not constituted; no VVB is accredited
MCC-700 v3.1	—	No clause provides for the review of MCC-700. §17 is amendment; §18 is entry into force.	—	—	Finding 3. The document register records the instrument as in force (FD-2026-08-24 item D11); this Policy does not treat it as in force (cl. 3.4)
MCC-700 v3.1	3.3	"MCC shall conduct annual independent audits of role separation compliance. Audit reports shall be published and filed with the Governing Board."	12 months	independent auditor	Never done
MCC-700 v3.1	4.5.3	"Working Groups shall report annually to the Standards Committee with recommendations for methodology improvement."	12 months	Methodology Working Groups	No working group exists
MCC-700 v3.1	7.1.3.2	"The SD Expert Panel shall conduct annual reviews of SD framework effectiveness and recommend indicator updates to the Standards Committee."	12 months	SD Expert Panel	Body not constituted
MCC-800 v1.0	art. 111	"This Standard shall be reviewed at least annually by the Programme Administrator and the Registry Manager, with recommendations for amendment submitted to the Governance Board."	12 months	PA + Registry Manager	Effective 1 April 2026 (art. 107); first review due 1 April 2027
MCC-800 v1.0	art. 101	"The fee schedule shall be reviewed annually and any changes shall take effect no earlier than 90 days after publication."	12 months	not stated	No fee schedule exists; MCC-FEES is a draft
MCC-800 v1.0	art. 89	Requires the Registry Manager to "conduct an annual data retention review".	12 months	Registry Manager	Recorded as outstanding at MCC-PRIV cl. 8.4
MCC-850 v1.1	art. 45	"This Standard shall be subject to scheduled review: Annual review: by Technical Committee and AI Governance Committee; Next formal review: March 2027."	12 months	Technical Committee + AI Governance Committee	Neither body is constituted. The Technical Committee appears in seven instruments and is created by none
MCC-850 v1.1	art. 40	"Data residency policy shall be reviewed annually and updated as laws evolve."	12 months	not stated	No data residency policy exists to review
MCC-BPM v1.0	0028	"Periodic review of buffer rate benchmarks shall occur biennially, comparing MCC rates against Verra AFOLU Non-Permanence Risk Tool and Gold Standard buffer approaches."	24 months	not stated	Benchmarks, not the Policy. The Policy itself has no review clause
MCC-CADCP v1.0	0010	"This Policy shall be reviewed and updated annually to incorporate emerging market practices and regulatory developments."	12 months	not stated	Never done
MCC-ESS v1.0	0010	"This Policy shall be reviewed and updated every five years or more frequently if significant gaps are identified through monitoring."	60 months	not stated	The longest period in the corpus for a Tier 5/3 policy
MCC-HCA v1.0	0010	"This procedure shall be reviewed and updated annually."	12 months	not stated	Duplicated at 0185
MCC-HCA v1.0	0185	"This procedure shall be subject to annual review."	12 months	not stated	Duplicate of 0010
MCC-HCA v1.0	0186–0187	"The MCC shall establish a Review Committee comprised of internal policy staff, external legal advisors, and host country representatives… The Review Committee shall meet annually."	12 months	Review Committee	Body never constituted. MCC-HCA v2.0-draft cl. 9.5 proposes to abolish it as non-canonical
MCC-SDCF v1.0	0033	"MCC shall update the SDG alignment framework every three years."	36 months	MCC	One of six
MCC-SDCF v1.0	0066	"The SD indicator framework shall be reviewed and updated annually by the methodology expert panel."	12 months	methodology expert panel	Body not constituted
MCC-SDCF v1.0	0131	"SD label criteria shall be reviewed annually."	12 months	not stated	One of six
MCC-SDCF v1.0	0171–0172	"The Expert Panel shall review the SD framework annually… Annual reviews shall assess: (a) appropriateness and relevance of selected indicators, (b) effectiveness of assessment methodology, (c) consistency of verification procedures, (d) portfolio-level SD impact, and (e) stakeholder feedback."	12 months	Expert Panel	Body not constituted. Clause 0172 is the best "what a review must consider" list in the corpus and Section 9 is built partly from it
MCC-SDCF v1.0	0177	"MCC shall conduct periodic (5-year) comprehensive reviews of the SD framework comparing performance against Gold Standard and Verra SD VISta benchmarks."	60 months	MCC	CR-2026-001 §S3 records the benchmarking obligation as a defect
MCC-SDCF v1.0	0181	"Third-party evaluations shall be commissioned every three years."	36 months	third party	One of six
Transparency Policy v1.1	§9.1	"Policy reviewed semi-annually (March & September). Amendments proposed quarterly; Board approval required."	6 months	Board	Cover and footer: "Next Review: September 23, 2026". Fell due on 23 September 2026; no review recorded
Claims Policy v1.1	§11.2	"Policy reviewed annually (March). Amendments proposed quarterly. Board approval required for material changes."	12 months	Board	Cover says "Next Review: September 23, 2026"; footer says "March 23, 2027". Three dates, one document
AI and Human Oversight Policy v1.1	§15.1	"This Policy reviewed semi-annually (March & September). Amendments: Proposed quarterly by AI Governance Committee."	6 months	AI Governance Committee	Body not constituted. Cover and footer: "Next Review: September 23, 2026". Fell due on 23 September 2026; no review recorded
Applicant Terms v1.1	—	No review clause. Art. 13.1 is amendment on 90 days' notice.	—	—	Cover and footer: "Next Review: September 23, 2026" — a date with no clause behind it
Registry Terms v1.1	—	No review clause.	—	—	As above
VVB Terms v1.1	—	No review clause. §7.1 "Annual Monitoring Review" is project verification, not instrument review	—	—	As above
Legal and Regulatory Positioning v1.1	—	No review clause.	—	—	As above
MCC-100 Annex — Definitions Register v1.0	—	No review clause.	—	—	In force since 24 August 2026 (FD-2026-08-24 D12)
F01–F05 Guidance, PDD Template, Ten Operational Umbrella Methodologies, Programme Bible v2.0	—	No review clause in any of them.	—	—	The Ten Umbrella Methodologies cl. 14.3.2.1 subjects geographic scope restrictions to "periodic review", which is not a review of the instrument

Part B — The rewritten suite (consultation drafts, none in force)

Draft	Clause	Period	Commentary
MCC-001 v3.0	10.9	60 months for every standard and methodology	Carries cl. 0065 forward; adds "legal" to the early-trigger list
MCC-200 v3.0	12.12	60 months per methodology	Cites MCC-001 cl. 10.9. No review clause for the Framework itself
MCC-400 v2.0	—	none	Drops art. 90's biennial review. Clause 4.5
MCC-100 v3.0	—	none
MCC-700 v4.0	—	none
MCC-800 v2.0	—	none for itself; 21.1 continuity plan annually; 24.2 fees annually; 25.1 monitoring rules annually	Drops art. 111's annual self-review
MCC-850 v2.0	24.3	12 months	"against monitoring results, incident history, audit findings and regulatory developments" — assigns it to nobody, which is an improvement on naming a body that does not exist
MCC-ESS v2.0	25.1	36 months	Shortens the predecessor's 60 months
MCC-HCA v2.0	10.3	12 months	Programme Secretariat, reported to Standards Committee — the only clause in the corpus that assigns instrument review to a body that is provided for. Section 8 follows it
MCC-SDCF v2.0	12.4	ten SD assessments or 36 months, whichever sooner	The evidence-triggered model. Clause 4.6
MCC-600 v1.0	12.2	12 months, public claims	Claims, not the Code
MCC-810 v1.0	4.6	12 months + on change of law
MCC-820 v1.0	3.3, 5.1, 6.1	annual/quarterly per control	No review clause for the Standard
MCC-PRIV v1.0	16.1	12 months + seven named events	The best event list in the suite; clause 6.8 is built from it
MCC-DRBC v1.0	12.2	12 months + four named events
MCC-FEES v1.0	5.3	12 months	Follows MCC-800 art. 101
MCC-REST v1.0	13.3	Transparency Policy cadence (6 months)
Non-Permanence Risk Tool v1.0	11.2	24 months
MCC-BPM v2.0, MCC-CADCP v2.0, MCC-ROI v1.0, Durability Classification, Methodology Template, MCC-FORMS, the report templates	—	none

What Schedule 1 shows. Thirty-five review provisions across fifteen instruments — thirty-two of them in instruments that are actually in force, the other three in MCC-700, which is not. Five distinct periods (6, 12, 24, 36 and 60 months). Four instruments state more than one period about themselves. Eight provisions are assigned to a body that does not exist. MCC-100 and MCC-700 have no review provision at all, and neither do the four counterparty and positioning texts, the Definitions Register, the PDD Template, the F01–F05 guidance, the Ten Operational Umbrella Methodologies or the Programme Bible. In the rewritten suite, nine drafts have no self-review clause and one instrument loses the review clause it had. And no review has ever been performed under any of them.

---

Schedule 2 — The first tranche

Non-normative. What clause 6.11 and clause 6.7 would produce if this Policy commenced today. It is a worked illustration and not a schedule anyone is bound to.

Seven instruments print "Next Review: 23 September 2026" (clause 4.10), and MCC-SDCF v1.0 has already been reviewed out of cycle. Under clause 6.7 no more than six may fall in one quarter, so the seven would be split, and the split would be made on exposure, not alphabetically:

Priority	Instrument	Why first
1	Claims Policy v1.1	Its Article 4 is the legal basis of the Programme's claims prohibition and is enforced in the registry by a database trigger. It is also the instrument whose own cover, clause and footer give three different review dates
2	Transparency Policy v1.1	Its Article 6 is the authority MCC-REST implements, and its §2.9 worked example is to be marked as an illustration under FD-2026-09-04 item 4 (recorded in the decision register as attested by the founder, with no executed copy filed); the served text does not yet mark it
3	Applicant Terms v1.1, Registry Terms v1.1, VVB Terms v1.1	Counterparty-facing, and the three texts with a printed review date and no review clause
4	AI and Human Oversight Policy v1.1	Its own review clause assigns the work to a body that does not exist
5	Legal and Regulatory Positioning v1.1	Non-normative, and the one of the seven whose contents are most exposed to the absence of an incorporated entity
—	MCC-SDCF v1.0	Not in the tranche. Already reviewed as CR-2026-001; the review is entered retrospectively under clause 7.6 as outcome class D, and the instrument does not need reviewing again — it needs replacing

What this schedule really shows. Six of the seven cannot produce outcome class A, because the review would find in each of them something already known to be wrong: three different review dates in the Claims Policy; a review clause pointing at a body that does not exist in the AI and Human Oversight Policy; a printed review date with no clause behind it in three of the counterparty texts. That is not an argument against reviewing them. It is the argument for the Register: each of those defects has been known for days or weeks, and there is no surface anywhere in the Programme on which the fact that they are unreviewed appears.

---

Annex — What consultation is asked to settle

#	Question	This draft's position
1	Should there be one cycle at all, or should each instrument keep its own period?	One cycle as a backstop, with each instrument's shorter period preserved (cll. 6.1, 6.2, 14.1). The alternative is Schedule 1
2	Are the intervals right — 60/36/24/24/24 months by tier?	Defensible but not obvious. The Tier 5 figure of 24 months is the one most likely to be argued: of the periods actually stated in force, twelve months is the commonest and two instruments say six
3	Which tier model does the Register key to — MCC-001 v2.0's or v3.0-draft's?	Not taken (cl. 6.5). Interim rule: the shorter of the two, with both tier values recorded
4	Can a Tier 5 policy set the Tier 1 and Tier 2 intervals?	No. Clauses 6.3 and 6.4 mark them as proposals for enactment at their own tier
5	Calendar review, or MCC-SDCF v2.0's evidence-triggered model ("ten assessments or three years")?	Calendar, with the event triggers of cl. 6.8. But cl. 4.6's reasoning is right for an instrument that has never been applied, and the Board may prefer to extend it
6	Should a review that finds nothing wrong be recorded and published?	Yes (cl. 10.3). This is the clause the Policy exists for. If consultation removes it, the rest is decoration
7	Should each instrument's own review clause be amended to point here at its next amendment?	Not taken (cl. 14.2). Either answer is defensible; having neither is what produced five periods
8	Should an overdue instrument be flagged on its own published page?	Not taken (cl. 12.7)
9	Should CR-2026-001 be entered retrospectively as the first recorded review in the Register?	Yes, as outcome class D (cl. 7.6)
10	Should MCC-400 v2.0 carry forward art. 90's biennial review, which the draft drops?	Not taken (cl. 4.5). This draft's position is that the cycle suffices
11	Where should this Policy sit — standalone Tier 5, or as a section of the Rule Change and Public Consultation Procedure?	A standalone Tier 5 policy, as its document control states; the tier is entered in the Document Register on adoption. The code MCC-IRC is assigned by the programme administration for this consultation; the document register carries it as a draft, with the code recorded as proposed and not yet confirmed by decision, and confirms it on adoption. (Consultation question 8.)
12	Should the Register be columns on document_register, or a separate register?	Columns (cl. 7.2); review columns were added to document_register on 4 September 2026 (PR #244), and none is populated. A second shelf is how a text gets recorded as non-existent

Twelve questions. Question 6 decides whether the Policy does anything; questions 3, 4 and 7 decide whether it can be applied without deciding something else first.

---

Prepared 4 September 2026 for public consultation. Not adopted, not in force. It reviews nothing by itself, and on the day it commences every instrument in the corpus will still never have been reviewed.

Evidence — what was checked, 4 September 2026. Four shelves and the successor drafts, for every negative finding in this text. (1) The review corpus ~/mcc-review/txt — all 34 in-force texts searched for "review cycle", "periodic review", "annual review", "shall be reviewed", "reviewed annually", "reviewed every", "review every", "revision cycle", "scheduled review", "review schedule", "reviewed at least", "biennial", "semi-annual", "every three years", "every two years", "next review" and "document date"; MCC-001 v2.0 cll. 0010, 0038, 0044–0048, 0054, 0065, 0075, 0078–0097 read in full; MCC-100 v2.1 cll. 0135, 0145, 0158, 0193, 0244, 0247–0260, 0273 and the whole of its amendment and track-record material; MCC-200 cll. 0010, 0123; MCC-400 v1.1 arts. 87–92 and Addendum 1 art. 113; MCC-800 v1.0 arts. 89, 99–101, 107, 111–113; MCC-850 v1.1 arts. 41, 44, 45 and the consolidated cover; MCC-BPM cll. 0028, 0170; MCC-CADCP cl. 0010; MCC-ESS cl. 0010; MCC-HCA cll. 0010, 0185–0192; MCC-SDCF cll. 0033, 0066, 0131, 0170–0181; the Transparency Policy §9.1, the Claims Policy §11.2, the AI and Human Oversight Policy §15.1, the Applicant, Registry and VVB Terms and the Legal and Regulatory Positioning paper (covers, footers and amendment articles); F01–F05, the PDD Template, the Ten Operational Umbrella Methodologies and the Programme Bible searched and found to contain none. (2) The served corpus apps/web/content/adopted/ — all 15 files searched; claims-policy.md lines 12, 632 and 653 read for the three-date conflict; transparency-policy.md §9.1 and its v1.1 change log; apps/web/src/lib/documents/manifest.ts read in full (16 published standards, each with SHA-256 and revision state, none with a review date, and the stale SDCF revision text at clause 15.9). (3) The proposals directory apps/web/content/proposals/ — all 40 files searched for self-review clauses; MCC-001 v3.0-draft §§9.1.1, 9.1.2, 10.9, 12.1–12.9; MCC-100 v3.0-draft §§16.6, 24; MCC-400 v2.0-draft §§1.4, 22; MCC-700 v4.0-draft §22 and Annex A row 31; MCC-850 v2.0-draft §24; MCC-HCA v2.0-draft §§9.5, 10.1–10.7 and its change log; MCC-SDCF v2.0-draft §§12.3–12.5 and drafting note; MCC-ESS v2.0-draft §25.1; MCC-PRIV §§8.4, 16.1; MCC-810 §4.6; MCC-DRBC §12.2; MCC-FEES §5.3; the Non-Permanence Risk Tool §11.2; rule-change-consultation-procedure-v1-0.md in full; mcc-rest-restatement-and-correction-v1-0-draft.md and mcc-roi-register-of-interests-v1-0-draft.md in full as the house-style references. (4) Supabase project ifflkkpowmzjoochwavf, read-only: document_register — 37 rows, all fifteen column names enumerated, no review column exists; decision_register — 6 rows, none a review outcome; governance_bodies — 14 rows across epochs E1 and E2, containing no Technical Committee, AI Governance Committee, Review Committee, VVB Advisory Panel or Methodology Review Committee; governance_appointments — 2; review_findings — 0; technical_reviews — 0; vvb_performance_reviews — 0; credit_units — 0; projects — 2, both SUBMITTED; sd_contributions — 0; stakeholder_consultations — 13 (8 OPEN, 5 STAYED), the Rule Change Procedure closing 8 September 2026. Nothing was written. The decision_register, document_register, review_findings and technical_reviews counts were re-checked on 1 October 2026 (clause 4.8), and governance_bodies (clause 4.7); the other counts above are those of 4 September 2026. (5) Drive 00_Styring_og_Vedtak (id 1X7rHCCmVcBCs1v5aC5Je-jOaYvE4-DvG): MIR-2026-001 revisions 1, 2 and 3 read in full; EN-2026-001 read in full; CR-2026-001 read in full; MCC-700 v3.1 read in full (§§3.3, 4.5.3, 7.1.3.2, 13.2, 13.4, 14.1–14.5, 17.1–17.2, 18.1–18.3); the GB-2026-002 board pack, its four addenda and FDP-2026-001 located and their subjects checked. A full-text search of Drive for "Instrument Review Cycle Policy" returns the three MIR revisions and nothing else, which is the basis for the first sentence of the NOTE. No clause number, quotation, figure, date or body in this draft is stated without one of those sources behind it, and every statement that something does not exist was checked against all five before it was written.