Sources — OCHA CBPF research
Every quantitative or status claim in Reserve allocation without a pre-selection freeze. maps to a primary locator below. OIOS 2025/098 and 2020/005 were downloaded as PDFs and read with pdftotext -layout on 24 Sep 2026 (Africa/Cairo). OIOS 2017/014 is noted separately under retrieval. On 11 Oct 2026, the full 2025/098 PDF was downloaded again and the selection paragraphs checked visually; its SHA-256 matched the retained 24 Sep file. Fletcher’s 24 Aug 2026 statement was read directly on OCHA’s publication page on 11 Oct 2026.
Primary sources
Landing page for 2025/098: https://oios.un.org/en/audit-management-and-oversight-country-based-pooled-funds-office-coordination-humanitarian-affairs
2017/014 retrieval. On 24 Sep 2026, curl to the file URL above returned HTTP 404, including with a browser user-agent. The same URL’s report text was retrieved the same day through document fetch (cover: REPORT 2017/014, 17 March 2017, Assignment No. AN2016/590/02). Figures in this table were checked against that text. If the token link fails in a browser, the OIOS audit-report index is the public catalogue: https://oios.un.org/en/audit-reports
Claim → locator table
Framing / method (not OIOS facts)
| Claim | Locator |
|---|---|
| Public OIOS and public Guidelines only; not Trace-li clients; no “would have prevented” | Note method — labelled as such |
| Finished-work pre-signature frame; humans dispose; Trace-li freezes proof | Trace-li Research Desk framing — labelled as such |
| Pattern matrix and value-chain map are synthesis | Chart captions — labelled synthesis |
| Wireframes are illustrative Global Fund-shaped desk UI, not a live OCHA desk run and not the CBPF control set | Captions in the note |
| Ethiopia is vintage 2017 | Labelled in the case section |
| Cross-links to the Global Fund note, the published OIOS 2025/098 case card, and the OIOS 2020/005 fixture | /research/global-fund, /cases/oios-2025-098-cbpf-reserve, /cases/oios-2020-005-iraq-forensic |
CBPF oversight — OIOS 2025/098
| Claim ID | Statement | Locator | Qualification |
|---|---|---|---|
| C-01 | Report 2025/098, 30 December 2025, Assignment AN2024-590-02. Audit period 1 January 2023 to 31 December 2024 | Cover; executive summary | — |
| C-02 | Overall, management and oversight mechanisms for CBPFs were effective, but needed strengthening in private-sector fundraising, transparency in partner selection, and project closure. Gaps included completeness of documentation supporting some project selections and pre-selections of implementing partners under the reserve modality | Executive summary | Audit opinion. Not a fraud finding |
| C-03 | OIOS made six recommendations. OCHA accepted all and has initiated action to implement them | Executive summary | Accepted / initiated ≠ closed |
| C-04 | Standard allocations support the response when less time-sensitive, against an HRP or equivalent plan, with a competitive call for proposals. Reserve allocations respond to rapid-onset and unforeseen circumstances when a standard allocation is considered infeasible. Applications come from a limited number of pre-selected eligible partners, by invitation, identified according to explicit criteria | ¶23 | OIOS’s statement of what the Global Guidelines provide |
| C-05 | Globally, 58 per cent of funds (US$661 million) in 2023 and 56 per cent (US$521 million) in 2024 were awarded under reserve allocations | ¶24 | Share of funds awarded. Not a loss |
| C-06 | That reserve share included five CBPFs in 2023 and three CBPFs in 2024 that used reserve allocations for 75–100 per cent of their awards | ¶24 | — |
| C-07 | Among 15 reserve allocations reviewed: 5 followed a limited competitive selection of pre-selected partners; 10 used a single-sourced partner for each project | ¶25 | Sample of allocations |
| C-08 | The basis for single-source pre-selection was justified for five of those ten, using project scorecards or other tools, and was not documented for the other five | ¶25 | Undocumented justification. Not a finding that the partner was ineligible |
| C-09 | The reserve allocation strategy for three of those five undocumented allocations included some partner selection criteria | ¶25 | Criteria in the strategy ≠ a documented partner-level justification |
| C-10 | Since 2025, PFMB has reviewed HFU allocation-strategy quality before finalization, including partner selection criteria for reserve allocations. It did not review whether HFUs consistently documented justifications for partner pre-selections, especially single-sourced partners. HFUs and PFMB did not always save allocation strategies in the CBPF Data Hub repository, as required | ¶25 | — |
| C-11 | Recommendation 2 accepted: strengthen monitoring of partner pre-selection under reserve allocations, and ensure allocation strategies are documented and stored centrally. OCHA said it would continue that monitoring and introduce a OneGMS feature for regular upload of allocation strategies to the Data Hub | ¶25 and OCHA response | Accepted action ≠ implemented in the report |
| C-12 | The Funds approved 1,287 projects (US$1.14 billion) in 2023 and 1,103 projects (US$938 million) in 2024 | ¶29 | Approvals. Not losses |
| C-13 | OIOS reviewed 60 projects in OneGMS: 35 standard allocations and 25 reserve allocations | ¶31 | Sample |
| C-14 | Project scorecards or alternative project selection documents were missing for 2 standard-allocation projects and 11 reserve-allocation projects | ¶31 | Missing record. OIOS does not say those 13 projects were improper awards |
| C-15 | OCHA indicated that project scorecards were not mandatory under reserve allocations, in which partners are pre-selected on criteria in the reserve allocation strategy | ¶31 | OCHA’s position, reported by OIOS. Read with Annex 4 (G-01, G-02), which scopes scorecards to standard allocations and says they may also be used for reserve allocations with a limited competitive process |
| C-16 | Some mismatches between project scores awarded by review committees and partners recommended for selection, without documented justification | ¶32 | No count of mismatches in the report. Do not invent one |
| C-17 | Recommendation 3 accepted. OCHA said it would strengthen the process through a OneGMS feature that ensures consistent upload of mandatory project scorecards prior to projects being recommended for funding | ¶32 and OCHA response | This is OCHA’s accepted-action wording. It is not a statement that scorecards were already mandatory for every reserve project. Accepted ≠ in force |
| C-18 | Global Guidelines: projects should be closed within 270 days of the project completion date, assuming no delays by the partner or audit firm, after clearance of final reports, project audit, and any refunds | ¶33 | Guideline timing as stated by OIOS. Assumption is in the sentence |
| C-19 | As of November 2025, 3,938 projects launched between 2016 and 2023 were at various stages of closure in OneGMS, 35 per cent of the projects launched in that period | ¶34 and Table 1 | Closure backlog. Not a fraud stock |
| C-20 | The Secretariat, including OCHA, did not have a formal procedure for sanctioning partners, nor a procedure for sharing information on non-performing or non-compliant partners with other United Nations system organizations. In 2024 the Board of Auditors recommended the Secretariat strengthen such information sharing. OCHA, with other Secretariat entities, was working on such arrangements | ¶37 | Information-sharing gap. Do not invent a UN partner-sanctions list. “Working on” ≠ done |
| C-21 | PFMB managed a database of suspected and confirmed fraud cases and updated the Board of Auditors, relevant donors, and the Pooled Fund Working Group | ¶37 | A database of cases is not a sanctions regime |
Current reporting context — OCHA, 24 August 2026
| Claim ID | Statement | Locator | Qualification |
|---|---|---|---|
| F-01 | Fletcher called for unified donor tracking to replace parallel reporting | As-delivered Ukraine briefing, paragraph beginning “Ukraine is a strong model” | Reporting priority, not a mandate for a new pre-award product, a procurement rule or an endorsement |
| F-02 | Reusing a selection record might reduce repeated evidence requests | Trace-li proposal in the new current-context section | Pilot hypothesis. No measured saving or OCHA implementation is claimed |
| F-03 | OIOS report issued 30 Dec 2025; publicly available 30 Jan 2026 | Report cover; OIOS landing page “Date available” | Issue and public availability are distinct dates |
Iraq — OIOS 2020/005
| Claim ID | Statement | Locator | Qualification |
|---|---|---|---|
| I-01 | Report 2020/005, 8 April 2020, Assignment AN2019-590-03. Period 1 January 2018 to 30 September 2019, including management of the Iraq Humanitarian Fund | Cover; executive summary | — |
| I-02 | CBPF standard operating procedures give three options for reported fraud: (a) the implementing partner conducts an internal investigation (mostly self-reporting); (b) a forensic review by an independent audit firm; or (c) an OIOS investigation. OIOS can also launch its own investigation and is not obliged to notify OCHA | ¶33 | Pathway description |
| I-03 | Only 2 of the 35 suspected fraudulent cases under forensic audit from 2015 to 2019 had been completed | ¶35 | Suspected cases referred for forensic audit. Not 35 substantiated frauds |
| I-04 | Those two forensic audits took 382 and 586 days from the partner notification date to completion | ¶35 | Elapsed time for the two completed files |
| I-05 | Both completed files concluded “no evidence of fraud/wrongdoing” | ¶35 | Applies to the two completed audits. Not a clean bill for the other 33 |
| I-06 | Terms of reference for external audits expect them to be ideally completed in six months. There is no timeframe attached to forensic audits | ¶35 | — |
| I-07 | Recommendation 3: review procedures for commissioning forensic audits, and estimate standard timeframes. OCHA accepted. The recommendation remained open pending evidence that procedures had been updated, including estimated standard timeframes. OCHA said timelines were not entirely within its control | ¶36 and response, p. 8 of the PDF text | Open at the date of this 2020 report. This note does not update the recommendation to a later status |
| I-08 | OIOS also recommended guidance on the consortium approach for grants to national NGOs. Separate from the Ethiopia pass-through finding | Executive summary | Do not merge with 2017/014 |
Ethiopia — OIOS 2017/014 (vintage)
| Claim ID | Statement | Locator | Qualification |
|---|---|---|---|
| E-01 | Report 2017/014, 17 March 2017, Assignment AN2016/590/02. Period January 2014 to October 2016. OCHA Ethiopia / Ethiopia Humanitarian Fund | Cover; ¶8–10 | Vintage. Not evidence of the 2023–2024 control state in 2025/098 |
| E-02 | For January 2014–October 2016 the EHF received US$121.4 million in contributions. US$108 million was allocated to 159 projects | ¶5 | Portfolio context for that period |
| E-03 | OIOS made four important recommendations. OCHA accepted them and had initiated action | Executive summary | Initiated ≠ closed |
| E-04 | During the audit period, reserve was the main modality: 65 per cent, about US$76.4 million. Standard allocation: about US$41 million, 35 per cent, as of October 2016 | ¶22 | 2014–2016 EHF only. Do not place on the 2023–2024 global chart |
| E-05 | Pass-through modality: OCHA signed and disbursed to a lead NGO, which signed and disbursed to participating NGOs. Three projects totalling about US$7 million, two lead NGOs and seven participating NGOs. Not formally stated in the EHF operational manual. No evidence it was cleared by the EHF governing bodies, the Funding Coordination Section, and the Executive Officer. Process before signature was not adequately documented | ¶30 | About US$7 million is the report’s wording |
| E-06 | Unlike cases where NGOs select their own sub-implementing partners, OCHA Ethiopia, through the clusters, selected the participating NGOs. OCHA had no legal recourse over funds disbursed to the participating NGOs, and they were not accountable to OCHA. By selecting the sub-partners, OCHA potentially diluted recourse against the lead NGO | ¶30 | Legal/control design gap as stated by OIOS |
| E-07 | Recommendation 1 remained open pending evidence that OCHA Ethiopia had sought Office of Legal Affairs advice and developed policies for consortium projects. OCHA said it would consult OLA. Target in the annex: 31 December 2018 | Rec. 1 and Annex I | Open at report date. Not updated here |
| E-08 | A major donor had audited the head office and global operations of Implementing Partner A. The audit resulted in the donor suspending funding to that partner globally. OCHA Ethiopia assessed bankruptcy risk, fielded a team on the partner’s procurement controls, and with the Administrative Service Branch decided to suspend any further grants to the partner until further notice | ¶34 | Donor suspension and OCHA’s further-grant suspension are stated. Partner is not named. Do not invent the name |
| E-09 | Implementing Partner A was implementing five EHF-funded projects totalling US$11.2 million, of which about US$9 million (80 per cent) had been disbursed | ¶35 | Portfolio still with the partner. Not a measured loss |
| E-10 | OCHA Ethiopia estimated its financial exposure at about US$3.7 million. OIOS believes the exposure could be much higher because the projects were still ongoing or had been completed but not yet audited to determine the propriety of the expenditures | ¶35 | OCHA’s estimate, not an OIOS loss figure. “Could be much higher” is qualitative. Do not chart US$3.7 million as a loss |
| E-11 | Recommendation 2 remained open pending receipt of audit and monitoring reports for Implementing Partner A. OCHA said ongoing projects would be audited as soon as they were completed | Rec. 2 | Open at report date |
Global Guidelines annexes (control surface)
| Claim ID | Statement | Locator | Qualification |
|---|---|---|---|
| G-01 | CBPF scorecards are used by Review Committees to assess project proposals submitted under standard allocations. Scorecards are developed in accordance with the allocation strategy before the strategic review commences | Annex 4 ¶1–2 | Not a statement that every reserve project requires a scorecard |
| G-02 | The scorecard may also be used under reserve allocations involving a limited competitive process | Annex 4 footnote 2 | Permissive, not a universal reserve mandate |
| G-03 | Four scorecard categories and weights: strategic relevance 30%, quality programming 30%, monitoring 15%, cost-effectiveness 25%. The same categories and weightings apply across CBPFs; indicators are reviewed before each allocation. Selection is informed by the scorecard, but not exclusively | Annex 4 ¶4–5 | HC may consider other aspects |
| G-04 | Eligibility process includes due diligence and capacity assessment, after preliminary screening and registration in the UN Partner Portal and the Grants Management System. Declarations in the guidance note include accurate information, recognition of United Nations compliance activities, conflict of interest, non-support to United Nations designated entities, previous or pending legal processes or investigations, and protection from sexual exploitation and abuse | Annex 7 contents and declaration headings | The note’s control surface. Not a Trace-li rule pack |
| G-05 | Annex 8 guides HFUs on non-compliance other than serious misconduct such as fraud and sexual exploitation or abuse, which are dealt with through applicable standard operating procedures. Measures are suggested rather than mandatory | Annex 8 opening | Do not describe Annex 8 as the fraud-sanctions regime |
| G-06 | The IP commits to immediately (within 48 hours) notify OCHA if it becomes aware of a breach or possible breach of any Grant Agreement obligation, whether by the IP or by a subcontractor (Art. IV, para. 8) | Annex 9 §3 | Partner duty in the guidance note. Not evidence of what any fund did in the OIOS samples |
Charts
| Chart | Data used | Sources |
|---|---|---|
charts/01-reserve-share.png | US$661m / 58% (2023); US$521m / 56% (2024) | C-05 |
charts/02-reserve-preselection.png | 5 limited competitive; 5 single-source documented; 5 single-source not documented | C-07, C-08 |
charts/03-selection-documents.png | 2 of 35 standard; 11 of 25 reserve | C-13, C-14 |
charts/04-iraq-forensic-days.png | 382 and 586 days | I-04. The 2-of-35 count is in the caption, from I-03 |
charts/05-pattern-matrix.png | Qualitative synthesis | C-08, C-14, C-20, I-03, E-05, E-09 — labelled synthesis |
charts/06-value-chain-map.png | Intake → rule map → evidence gaps → judgment → freeze, named to annexes | G-01 through G-06 and the OIOS gaps — labelled synthesis |
Dropped / not used
| Item | Reason |
|---|---|
| Private-sector fundraising percentages in 2025/098 (0.2% / 0.4%) | Real, but not the pre-selection story |
| Localization shares (36% / 46%) | Real, but not the pre-selection story |
| US$187,544 refund by a United Nations agency in 2017/014 ¶33 | Different matter from Implementing Partner A. Left out so the two are not merged |
| US$3.7 million drawn as a loss, or as OIOS’s own exposure figure | It is OCHA Ethiopia’s estimate. OIOS says exposure could be much higher and the projects were not yet audited |
| “No evidence of fraud” extended from two completed Iraq audits to all 35 cases | The report limits that conclusion to the two completed files |
| A Trace-li or OCHA partner-sanctions list | ¶37 says the Secretariat had no formal partner-sanctions procedure |
| Any “Trace-li would have prevented” counterfactual | Hard rule |
| Recommendation accepted, or “initiated”, described as done | Reports distinguish acceptance from closure |