Oversight · pilot analysis · program 198 — F-35

The traceability break

Can one major program be followed from the budget request, through obligation, into the account records, and out to the audit findings that judge those records — using only the sources this site publishes? Three of the four links hold. The one that fails is the one that matters, and the audit layer names it independently.

Budget FY2027 P-1 / R-1 · 2026-08-22Execution contracts vintage 2026-08-06Accounts File A · File C 2026-08-06Audit 2026-09-08
Source
USASpending contract award archive; USASpending File C
Grain
Acquisition program code x fiscal year
Vintage
2026-08-06
Rows loaded
4,220

Path data/usaspending/warehouse/contracts + accounts/file_c_contracts · extracted 2026-09-19 04:54 · refresh daily

Limitations FPDS records 'no acquisition program' as code 000 / description NONE, and roughly three quarters of DoD contract dollars carry that code, so this covers only the program-attributable quarter of the file -- dm_program_coverage states how much per year. federal_accounts_funding_this_award lists every account funding an award without apportioning the obligation across them, so no figure here is split by account. A program obligation total is not comparable to that program's appropriation: it draws on prior-year balances this cut cannot age, and Foreign Military Sales trust dollars (011-8242) sit on the same contract actions as Department appropriations.

Every figure on this page is queried live from the loaded data, not transcribed from the analysis memo. The budget layer is the one exception to the dm_load provenance system — see section 01.

The chain, as the data actually supports it

01 · BUDGET$12.5B · FY202510 lines · 70 aircraft02 · EXECUTION$36.3B obligated850 actions · 320 contracts03 · ACCOUNTS$33.5B untraceable7.6% names an account04 · AUDITDisclaimer of opinionMW #14 of 26 — named7.6% names an account — and File C, the file that would settle it, is empty for these appropriationsthe program dimension lives only in the award files; the account dimension only in File A and File B
FY2025 F-35 obligations
$36.3B
850 actions · 320 contracts
Of those dollars, naming no Treasury account
92.4%
$33.5B · was 14.2% in FY2021
Carried by one contract
59.4%
N0001923C0003 · 5 actions
Department-wide File C linkage
2.6%
year closed at FY2025P09

The program dimension exists only in the award files. The account dimension exists only in File A and File B. File C is the sole bridge between them — and for the F-35 appropriations it is, in this warehouse cut, empty.

01 · Budget — what was asked for

The F-35 lines in the FY2027 President's Budget exhibits, read live from the loaded exhibit tables. Procurement totals are Add rows only — weapon system cost, less prior-year advance procurement, plus current-year AP — so the Non-Add advance-procurement detail is not double counted.

FY2025 enacted
$12.5B
70 aircraft · 6 procurement lines, 4 RDT&E
FY2026 total
$12.1B
47 aircraft · $1.0B of it reconciliation funding
FY2027 request
$37.7B
170 aircraft · 210.3% against FY2026
FY2027 mandatory share
58.1%
Discretionary base 41.7% against FY2026

Discretionary and mandatory, by year

$12.5B
FY25
$12.1B
FY26
$37.7B
FY27
  • Discretionary
  • Mandatory / PL 119-21

The headline reverses when the request is split. The total grows; the discretionary base does not. A trend line across F-35 totals without this split describes a different program in FY2027 than in FY2026.

Aircraft quantity

70
FY25
47
FY26
170
FY27
  • Aircraft

Quantity is summed on weapon-system-cost rows only. Modification and spares lines carry dollars but no aircraft, so dividing the totals above by these counts would not give a unit cost — that needs the P-5 / P-21 exhibits, which are not in this cut.

AccountLineTitleFY2025FY2027 disc.FY2027 mand.
1506N0147Joint Strike Fighter CV$2.5B$3.3B$7.9B
1506N0152JSF STOVL$2.1B$280.7M$3.9B
1506N0592F-35 STOVL Series$229.8M$857.8M$162.0M
1506N0593F-35 CV Series$154.2M$271.2M$162.0M
3010FATA000F-35$5.0B$6.3B$4.7B
3010FF03500F-35 Modifications$394.5M$671.9M$324.0M
1319N0604840MF-35 C2D2$465.0M$939.6M$1.6B
1319N0604840NF-35 C2D2$451.1M$857.1M$1.4B
3600F0207142FF-35 Squadrons$45.4M$94.8M$0
3600F0604840FF-35 C2D2$1.1B$2.3B$1.9B

Amounts converted from the exhibits' $ thousands. Joint Strike Missile (3020F JSM000) is a different program and is deliberately excluded.

FY2027 budget exhibits (p1 · r1) · 2026-08-22

What the exhibits do not contain

A scan of the FY2027 O-1 returns no F-35 line at all. Operation and maintenance — where sustainment is funded, and the larger share of the program’s lifecycle cost — carries no program dimension at exhibit level. Yet F-35 contracts draw on O&M accounts, as section 03 shows. Those dollars are real, they are obligated against F-35 contracts, and no “-1” exhibit attributes them to the program.

These figures come from war_budget_line, which predates the dm_load provenance system and carries its own ingest stamp instead of a load row. Stated here rather than left to look like the rest of the site’s measures.

02 · Execution — what was obligated

Contract obligations tagged to the F-35 acquisition program code. Action counts are flat within a narrow band while dollars move by multiples: obligation volume on this program is not an activity measure, it is a definitization calendar.

FY2025 obligations
$36.3B
850 actions across 320 contracts
Carried by five actions
72.7%
$26.4B
Landed in August–September
53.1%
$19.3B — year-end concentration
Largest single contract
59.4%
N0001923C0003 · $21.6B

Obligations, and the share in the five largest actions

$9.8B
FY21
$20.8B
FY22
$36.5B
FY23
$12.4B
FY24
$36.3B
FY25
$10.4B
FY26*
  • Five largest actions
  • Everything else

* fiscal year in progress — period-to-date, not a closed year.

Action count, same years

991
FY21
901
FY22
875
FY23
867
FY24
850
FY25
430
FY26*
  • Actions

* fiscal year in progress — period-to-date, not a closed year.

Read the two together. Dollars swing by multiples while the count barely moves — the years differ by when a production lot was definitized, not by how much work was placed on contract.

Fiscal yearObligatedActionsContractsTop-5 shareAug–Sep shareNames an account
FY2021$9.8B991346107.8%40.3%85.8%
FY2022$20.8B90132950.7%38.3%84.7%
FY2023$36.5B87534260.9%14.2%77.2%
FY2024$12.4B86733627.2%26.2%34.6%
FY2025$36.3B85032072.7%53.1%7.6%
FY2026 *$10.4B43019641.1%0.0%8.7%

The year in one contract

N0001923C0003 carries $21.6B across 5 modifications — 59.4% of all FY2025 F-35 obligations to LOCKHEED MARTIN CORPORATION. Its largest action fell on 2025-09-29. THIS MODIFICATION DEFINITIZES LOT 18 & 19 AIRCRAFT CLINS

FY2025 composition

Recipient

  • LOCKHEED MARTIN CORP$31.2B
    85.9%
  • ROCKWELL COLLINS AUSTRALIA PTY LIMITED$4.8B
    13.3%
  • NORTHROP GRUMMAN CORPORATION$97.5M
    0.3%
  • BOOZ ALLEN HAMILTON HOLDING CORPORATION$28.1M
    0.1%
  • AUGUST SCHELL ENTERPRISES, INC.$27.1M
    0.1%

Extent competed

  • C · Not competed$35.7B
    98.4%
  • B · Not available for competition$447.7M
    1.2%
  • A · Full and open competition$98.5M
    0.3%
  • D · Full and open after exclusion of sources$48.2M
    0.1%
  • G · Not competed under simplified acquisition procedures$1.8M
    0.0%

Contract pricing

  • L · Fixed price incentive$25.2B
    69.4%
  • V · Cost plus incentive fee$7.7B
    21.2%
  • U · Cost plus fixed fee$2.0B
    5.5%
  • J · Firm fixed price$1.4B
    3.8%
  • S · Cost no fee$21.2M
    0.1%

Extent competed and pricing arrive from FPDS as single-letter codes; the code book lives in the ETL, not in page-level string matching, so the label is the decoded value with its code retained. A sole-source program of record awarding incentive contracts through a lead service is the expected shape — stated so the next section’s failure is not mistaken for one of these.

Correction · what this cut does not cover

An earlier draft of this analysis called FPDS program tagging dense, on the strength of a null test that found almost nothing. That was wrong. FPDS records “no acquisition program” as the explicit code 000, description NONE — not as a null. Only 25.3% of FY2025 contract obligations, and 0.51% of actions, carry a program code at all. A sentinel in a code field is not a null — the same failure mode as the single-letter codes above. It is now published as control PROG-06.

Fiscal yearContract obligationsCarrying a program codeShare of $Share of actionsPrograms seen
FY2021$387.0B$71.2B18.4%0.58%252
FY2022$414.4B$79.7B19.2%0.56%286
FY2023$456.8B$113.5B24.8%0.53%317
FY2024$446.0B$88.7B19.9%0.54%390
FY2025$491.7B$124.4B25.3%0.51%418
FY2026 *$282.2B$69.8B24.7%0.43%335

The denominator every program figure needs. The program dimension is precise where present and absent for most of the file; both halves are true and only one is visible from a program page.

03 · The account bridge, and where it breaks

To connect an obligation to an appropriation you need to know which Treasury account it drew on. Two fields could answer that. Neither does.

Share of F-35 obligations naming a Treasury account

0%49%99%FY21FY22FY23FY24FY25FY26*

Hover the line for a fiscal year.

From 85.8% in FY2021 to 7.6% in FY2025. Where this falls the obligations do not become less real — they become unattributable to an appropriation.

Traced and untraced obligations

$9.8B
FY21
$20.8B
FY22
$36.5B
FY23
$12.4B
FY24
$36.3B
FY25
$10.4B
FY26*
  • Names an account
  • Names no account

* fiscal year in progress — period-to-date, not a closed year.

It is not submission lag

1 of the 6 fiscal years here are at a submission period ending P12 — reported closed — and the linkage does not improve in them. FY2025 sits at FY2025P09 with 12 File C rows for this program’s contracts. A closed year that never filled in is a completeness failure, not a late one. The memo in analysis/ tests this further by re-running the same fiscal years at two warehouse vintages a month apart: the figures are identical to the dollar.

The accounts named on FY2025 actions

Accounts namedAccountsObligatedActions
057-3400 + 057-38402$1.4B16
017-1319 + 057-36002$528.7M39
017-1506 + 017-1804 + 057-3010 + 057-34004$462.2M12
097-04001$426.7M22
017-1506 + 057-30102−$73.7M86
011-8242 + 017-1319 + 017-1506 + 057-3010 + 057-3400 + 057-3600 ⚑6−$54.5M2
017-1319 + 017-1506 + 057-3010 + 057-36004$42.7M10
017-15061$42.0M5

Read a row as: this much was obligated on actions that named this combination of accounts. It is not this much from each account — the file states no split, and nothing here apportions one. A joint-service production contract names two services' appropriations and no division between them.

SCOPE-01 has an execution-side blind spot

Accounts outside Department scope appear on these actions: 011-8242. Disclosed rather than dropped, per PROG-04. SCOPE-01 keeps agency 011 out of Department budgetary totals, which is right for budget. On the execution side those dollars — Foreign Military Sales trust money — sit on the same contract actions as Department appropriations and the file does not separate them. That is an independent reason the obligation total is not comparable to an appropriation.

File C, the file that would settle it

Fiscal yearAward-file obligationsFile C obligationsFile C rowsLinkageLatest submission
FY2021$9.8B−$884.6M288-9.02%FY2021P06
FY2022$20.8B−$61210.00%FY2022P06
FY2023$36.5B$86,07360.00%FY2023P12
FY2024$12.4B$1,20910.00%FY2024P06
FY2025$36.3B−$60,42212-0.00%FY2025P09
FY2026 *$10.4B$000.00%FY2026P03

File C is the account-linked contract obligation file — the only artifact tying a contract to a Treasury account with a dollar amount attached. This is what it holds for this program's own contracts.

The same collapse, department-wide

0%3%7%FY21FY22FY23FY24FY25FY26*

Hover the line for a fiscal year.

Fiscal yearAward filesFile CLinkage
FY2021$387.0B$22.8B5.9%
FY2022$414.4B$8.1B2.0%
FY2023$456.8B$18.4B4.0%
FY2024$446.0B$3.9B0.9%
FY2025$491.7B$13.0B2.6%
FY2026 *$282.2B$3.1B1.1%

A fiscal year showing no File C rows matched nothing in this warehouse cut. That is a statement about the extract, not evidence that no such report was made. Nor is File C’s shortfall an error estimate for the award files: they are two reporting chains, and the ratio measures linkage completeness. File A carries the F-35 appropriation accounts but has no program dimension at all, and File B has the same limitation — which is why File C is the only bridge. See reconciliation and funds control.

04 · Audit — the layer that names the same failure

The independent auditor's report on the FY2025 financial statements, from the curated audit register.

Opinion
Disclaimer of opinion
Consecutive disclaimers
8
FY2018–FY2025
Auditor-identified material weaknesses
26
down from 28 in FY2024
Of which, this program
#14
Joint Strike Fighter Program

Material weakness #14 of 26

“Joint Strike Fighter Program”. The program traced on this page is not merely affected by the Department’s audit condition — it is a named material weakness in its own right. Four other weaknesses on the same list bear directly on the break in section 03: the universe of transactions, government property in the possession of contractors, intragovernmental transactions, and budgetary resources.

Share of total assets under scope limitation
43%
Share of budgetary resources
64%
NFRs open at year-end
2,972
2,473 issued · 1,004 closed
Projected FY2027 remediation cost
$1.7B
Joint Task Force Audit Established March 2026
#Category
1Financial Management Systems Modernization
2Configuration Management
3Security Management
4Access Controls
5Segregation of Duties
6Interface Controls
7Universe of Transactions
8Fund Balance with Treasury
9Inventory and Stockpile Materials
10Operating Materials and Supplies
11General Property, Plant, and Equipment
12Real Property
13Government Property in the Possession of Contractors
14Joint Strike Fighter Program
15Accounts Payable
16Environmental and Disposal Liabilities
17Leases
18Unsupported Accounting Adjustments
19Intragovernmental Transactions and Intradepartmental Eliminations
20Gross Costs
21Earned Revenue
22Reconciliation of Net Cost of Operations to Outlays
23Budgetary Resources
24Service Organizations
25Component Entity-Level Controls
26DoD-Wide Oversight and Monitoring

The auditor's own material-weakness categories, in report order. A separate framework from the FMFIA self-assessed counts on the audit page; the two are shown side by side there rather than conflated.

Full audit posture on audit; program-level execution figures, and the other eleven programs carried at this depth, on program. Two F-35 hearings sit in the congressional corpus — see congressional. Their dates are not printed here: the filenames carry the acquisition date, not the hearing date, and the real dates are not in the loaded data.

05 · What this pilot does not support

Stated as prominently as the findings, because each is a comparison the data invites and cannot bear.

The two big numbers cannot be reconciled. $36.3B obligated against $12.5B appropriated: obligations draw on prior-year balances this cut cannot age, an unknown share is FMS and partner money, and 92.4% of the dollars name no account. Publishing a budget-versus-execution variance here would assert a comparison the data cannot bear.

This is a fraction of the contract file. Only 25.3% of FY2025 contract obligations carry any acquisition program code. Nothing here describes the rest of it.

An account named is not an account charged. Where several accounts appear on one action the file states no split between them, and nothing on this page apportions one — the shape of the account table is what stops a reader from trying.

File C is not an error estimate. The award files and File C are two reporting chains. A dollar absent from File C is not a dollar that was not obligated; the ratio measures linkage completeness and nothing else.

No unit cost is claimed. Quantity is summed on weapon-system-cost rows only, and an aircraft’s cost is split across advance-procurement years. A defensible unit cost needs the P-5 / P-21 exhibits, which are not in this cut.

Absence is absence in this cut. An empty result means the filter matched nothing in the vintage named at the top of this page — never that the Department did not report it.

06 · The controls that keep this page honest

These are not proposals. They run inside the load transaction, and a critical failure rolls the whole load back so the previous vintage stays published.

ControlSeverityAssertionResult
PROG-01criticalEvery row in dm_program_fy carries a non-empty program_code that exists in dm_program_dim, and hangs off a dm_load row with a vintage.1/1 pass
PROG-02criticalFor every fiscal year, the sum of dm_program_account.obligation for a program does not exceed that program's fiscal-year obligation.72/72 pass
PROG-03highEvery dm_program_fy row carries traceable_obligation and untraceable_obligation summing to its obligation, with traceable_pct consistent with them.72/72 pass
PROG-04highWhere an account outside agency codes 097/021/017/057 is named on a program's obligations, dm_program_account records it in out_of_scope_accounts rather than omitting the row.1/1 pass
PROG-05moderateA program fiscal year is flagged partial exactly when it is the newest fiscal year in the current contract vintage.6/6 pass
PROG-06highdm_program_coverage carries one row per fiscal year in which attributed and unattributed obligations sum to the fiscal-year total, and attributed_pct is consistent with them.6/6 pass

PROG-03 is the one that matters here. A program obligation total published without the account-traceable share beside it reads as reconciled when it is not — so the data layer returns the pair on one row and the control asserts they foot. Every control on this list was also confirmed to fail against deliberately corrupted rows; a control that only ever passes is not a control. Full suite on controls.

Method and sources: sources · definitions · the full memo and a script that re-derives every figure from the warehouse live in analysis/ in the repository.