---
title: "Nothing Was Hidden. The Subaward Has No Number That Follows It."
dek: "Five months, four public systems, one $599,000 payment. We rebuilt it to within $18,300 of the agency's own total, printed the gap as a finding in June, and spent two more months discovering the agency was right."
kind: analysis (Analysis)
desk: The Body
author: Loretta Sung and Nadia Ferreira-Ost (fictional)
published: 2026-08-19T06:00:00Z
url: https://theyburied.com/article/subaward-with-no-number
tags: subawards, research funding, metadata, transliteration, indirect costs, corrections
genre: satire
status: fiction — nothing in this story happened; every person, source and document is invented
---

# Nothing Was Hidden. The Subaward Has No Number That Follows It.

*Five months, four public systems, one $599,000 payment. We rebuilt it to within $18,300 of the agency's own total, printed the gap as a finding in June, and spent two more months discovering the agency was right.*

A single subaward of $599,000, paid to one institution through one intermediary over four years, is disclosed in four public systems. We spent five months trying to follow it through them and could not do it from a standing start.

Nothing was concealed. Every element was already public. The problem is that a subaward has no number.

## What the claim was

The account we set out to test is that money for enhanced-pathogen work was routed through an intermediary organisation in order to keep it off the books. It is a claim about a channel: not about what the work was, but about whether the payment was meant to be findable.

So we tested findability. We picked one subaward, gave ourselves only public sources, and wrote down everything.

## Four systems, four different truths

The federal award database has it. The subaward appears as a subrecipient row — recipient name, period, dollar figure, updated quarterly. The row carries the prime's award number, because there is no field for a subaward number. There is no such field because subawards do not have numbers.

The prime awardee's audited accounts have it, as one line in a schedule of subrecipient payments, rounded to the nearest thousand dollars. Over four lines, that rounding alone can move a total by $2,000 in either direction.

The agency's progress-report archive has it, under a project title that changes between the third and fourth report — from a phrase naming a virus family to a phrase naming a surveillance activity. Search for one and you lose two years. Search for the other and you lose the other two.

The recipient country's foreign-grant register has it twice, under two transliterations of the same institution's name, in the same register year, neither cross-referenced to the other.

> **Figure:** One subaward, four public systems, and no identifier that survives the crossing — see https://theyburied.com/article/subaward-with-no-number

## The eleven dead ends

We publish all of them. Four were search strings that were simply wrong. Three were a wrong assumption about the fiscal year boundary. Two were a genuinely different organisation with a near-identical name in a neighbouring city. One was a downloaded extract we had truncated ourselves at 10,000 rows without noticing.

The eleventh took six weeks: we had been reading a currency conversion applied at the register's filing date rather than at payment, which is correct practice and which nobody had told us about because nobody imagined we did not know.

The closest thing in our own vault to this exercise, [[PX-2131]], is six subject-access responses covering one week of one reporter's card payments. It took four months and produced a smaller ledger than this one.

## The $18,300

Our reconstruction closed at $580,700. The agency's published total is $599,000.

In June we drafted that difference as a finding. We wrote that a gap of $18,300 in a $599,000 subaward could not be reconciled from public sources and that the agency had not explained it. That draft is published with this story.

The agency had explained it. Adaeze Kilbride-Onuoha, its deputy director for grants policy, told us on 19 March, in two sentences, that the difference was indirect-cost recovery falling in a different fiscal period. We did not accept it, because we could not check it — the four systems we were working from do not carry fiscal-period detail at that resolution.

It took two further months and a ledger extract to establish that she was right, to the dollar.

> I would rather be checked than believed. Neither is currently available.
>
> — Adaeze Kilbride-Onuoha

> **Caution.** **Updated 19 August 2026.** This story supersedes the June draft circulated to sources, which carried the $18,300 as an unexplained gap. That draft was wrong and is published alongside this piece rather than withdrawn.

## What the desk takes from it

There is a shape of story in which five months of work and a stubborn $18,300 becomes a channel, a screen, a thing kept off the books. Every ingredient for it was on our desk in June, in our own handwriting.

The duller reading is the correct one. Four public systems, built by different people in different decades for different purposes, were never asked to speak to each other, and a payment that all four disclose honestly is traceable in none of them without five months and $11,150.

That is not a cover-up. It is what happens when a thing is not given a name, and it is fixable for roughly what this investigation cost.

## Sources (invented)

- Federal award database, prime award and subrecipient reporting, Public award and subaward reporting for the prime, financial years 2019–2024 — Downloaded in full and reconciled line by line; every extract published with this story (March–July 2026)
- Audited financial statements of the prime awardee, Five years of audited accounts and the single-audit schedule of subrecipient payments — Obtained from the public filing portal; the 2022 schedule read against the 2023 restatement (April 2026)
- Adaeze Kilbride-Onuoha, Deputy director for grants policy at the funding agency — Interviewed by video call on 19 March and 4 August 2026; sent the June draft, the working spreadsheet and every intermediate figure (March–August 2026)
- Ekaterina Voskoboynikova-Hale, Research-administration consultant; twenty years reconciling subaward ledgers for universities — Paid to reconstruct the same subaward independently from the same four public systems, without sight of our working (May–June 2026)
- The recipient country's foreign-grant register, Statutory register of foreign funding received by domestic research institutions — Read in the original with a paid translator, then re-read after the translator flagged a second spelling of the institution's name (May and July 2026)

## Documents cited (invented)

- PX-2131: https://theyburied.com/vault/PX-2131

## What we could not confirm

- What the money bought. This story is about whether $599,000 can be traced, not about what was done with it. We reconstructed the payments; we did not audit the science, and nothing here should be read as a finding about the work itself.
- Whether the year-four retitling was deliberate. The project's title changes between the third and fourth progress report, from a phrase naming one virus family to a phrase naming a surveillance activity. The agency says titles are supplied by the awardee and change routinely. We found eleven other awards in the same portfolio where a title changed at renewal, which is consistent with routine, and is not proof of it.
- Whether our $18,300 explanation is the whole explanation. Indirect-cost recovery falling in a different fiscal period accounts for $18,300 of the gap to the dollar, and the agency's ledger shows it. We cannot rule out that two errors of similar size cancelled, and neither can the agency, and we asked.

## Right of reply

**Adaeze Kilbride-Onuoha, deputy director for grants policy** (Replied in writing on 11 August 2026 and asked that it run without cuts. Printed unedited.):

> On 19 March I told you, in a video call your colleague recorded with my consent, that the difference you were seeing was indirect-cost recovery landing in a different fiscal period. It took two sentences. I offered to walk you through the ledger line that shows it. You thanked me and went away for twelve weeks and printed a draft saying our number could not be reconciled.
> 
> I am not writing to complain about being doubted. Doubt me. That is your job, and I would think less of you if you took a policy official's word for an accounting question. I am writing because of what happened in between, and because I think it is the more interesting story and you have buried it under an apology.
> 
> You could not check what I told you. That is the actual finding. Not that a number was hard to trace — that when the person who holds the answer offers it to you, you have no independent way to test the offer, because the four systems you were working from do not contain the fiscal-period detail that resolves it. So you were left choosing between believing me and not believing me. You chose not to, honourably and at length, and you were wrong, and you would have been equally wrong if you had chosen to believe me and I had been mistaken.
> 
> That is what a metadata failure costs. Not concealment. The loss of the ability to check.
> 
> Now let me push back on your framing, because I think you have been slightly too kind to yourselves in the other direction. You write that the subaward 'has no number'. It has our award number, a subrecipient name, a period of performance and a dollar figure, in a public database, updated quarterly, going back to 2010. That is not nothing. It is more than almost any other country publishes and more than we published fifteen years ago. What you have discovered is that it is not enough to follow one payment through four systems that were built by different people, in different decades, for different purposes, and were never asked to speak to each other. Nobody designed that. It accumulated.
> 
> I have been asking for a persistent subaward identifier for nine years. I have written it into three policy consultations. It has been costed twice and deferred twice, and the reason both times was the burden on recipient institutions, which is a real reason and not a pretext — the burden falls hardest on the smallest ones, and I have never had a good answer to that.
> 
> So: print my figure, print your wrong one, and print this. If you want to help, the useful sentence is not that a payment was hard to trace. It is that a nine-year-old proposal to give subawards a number of their own is sitting in a consultation file, that it costs about what your newspaper spent on this investigation, and that no one has ever written about it because there is no scandal in it.
> 
> I would rather be checked than believed. Neither is currently available.

---
They Buried is a satirical newspaper. Every story, source, document and person in it is invented. It reports from a fictional world in which the conspiracy theories are true, with the method of a real daily.