Inventory accountability is not bookkeeping trivia for a flagship weapons program; it is the control system that protects tens of billions in government-owned assets, sustains fleet readiness, and underwrites credible financial reporting. The Government Accountability Office’s recent work on F-35 parts shows that this control system remains immature — and that the program’s logistics rhetoric and dashboards have not yet closed the gap between what exists and what can be reliably traced.
At a Glance
- GAO identified roughly 26 million F-35 parts valued at about $31 billion spread across 700+ facilities; the Department of Defense lacks the procedures to reliably account for them.
- As of March 2026, more than 17,000 parts worth about $1.3 billion were absent from the inventory system — a concrete manifestation of the control failure, not a theoretical risk.
- Earlier GAO work found the Joint Program Office could not provide basic cost, quantity, and location data for the global spares pool and relied on contractor records.
- Program statements about analytics, ODIN, and supply-chain reforms are improvements in progress, not evidence that the accountability problem has been solved.
What GAO actually found: scale, mechanism, and the gap between ownership and records
Start with the scope: the F-35 program manages roughly 26 million government-owned parts worth around $31 billion, distributed across more than 700 government- and contractor-managed locations worldwide. Those numbers are not claims about loss; they describe exposure — the inventory at risk if the accountability system is weak. GAO’s central finding is exactly that: DOD has not established procedures sufficient to ensure its own records can support reliable accountability and financial reporting for those parts. In plain terms, the accountable property system of record — the official ledger of what the government owns, where it is, and what it is worth — is not reliable for this inventory.
That weakness is not abstract. GAO’s work and contemporaneous reporting cite more than 17,000 parts, with an associated value of about $1.3 billion, missing from the program’s inventory system as of March 2026; the valuation could be higher once missing cost data and ancillary acquisition expenses are included. Separately, over one million parts worth more than $85 million flagged in 2023 still lacked a resolved status in 2026 — a continuity failure that signals weak exception clearing and deficient follow-through. While some portion of these records gaps may reflect items in repair, at contractor sites, or trapped in incompatible systems rather than physically lost, the program has not produced the reconciliation needed to distinguish administrative absence from actual disappearance. That is the problem an accountable property system is supposed to solve — automatically, repetitively, and with audit evidence.
Why F-35 parts are hard to govern: distributed custody and fragmented systems
The F-35 is not a single warehouse with a single barcode scanner; it is a global enterprise moving serialized and non-serialized material through prime and sub-tier contractors, service depots, and international partners. Accountability routinely breaks where government ownership meets contractor custody, particularly when multiple logistics systems must synchronize. GAO’s 2023 report found the Joint Program Office could not provide total cost, quantity, and location information for the global spares pool at non–prime-contractor facilities and continued to rely on contractor records as its de facto source of truth. That reliance is not rare in defense programs; it is exactly where recurring Department of Defense inventory findings have arisen for decades. The persistence here matters because, absent a mature JPO property-accountability program — procedures, trained accountable officers, periodic physical inventories, discrepancy resolution — errors propagate faster than the organization can correct them.
When the accountable record lags the physical world, readiness and financial credibility both take a hit. Units chase parts that the system shows “available” but are actually awaiting repair; program financial statements absorb adjustments and write-offs the ledger cannot support with documentation; and investigations into potential losses start late and end ambiguously. GAO explicitly ties the F-35 parts-accountability weaknesses to DOD’s broader struggle to produce auditable financial statements — not as a rhetorical flourish but because the same control disciplines govern both outcomes.
The counter-case: promising tools, incomplete evidence of closure
The program and its prime contractor have put forward a modernized sustainment narrative: ODIN to replace ALIS, cloud-native data environments, 360-degree logistics dashboards, anomaly detection, and automated cycle counts. They describe “real-time visibility” into inventory and readiness and a multiyear procurement framework to align aircraft production with spare-parts buys and modifications. On their face, those are the right design moves — stable demand signals reduce bullwhip effects; predictive analytics can flag consumption spikes; cloud platforms can rationalize data silos. But those capabilities are means, not ends. GAO’s 2026 findings post-date many of these promises and still concluded that the JPO lacked a mature, fully functioning property-accountability program with program-specific procedures and training for accountable-property officers. Until the government’s own accountable property system reconciles to physical inventories across those 700-plus facilities, claims of real-time visibility remain, at best, partial.
Consider the most concrete test: can the program produce a part-level reconciliation — by part and serial number where applicable, with custodian, location, condition, cost, and last verified date — for the 26 million items it says it owns? GAO’s answer so far is no. Can it trace the more than 17,000 items absent from the inventory system in March 2026 to either a recovered location or a documented loss report, with supporting evidence? Again, no public reconciliation exists. Program dashboards may reduce future exceptions; they do not retroactively substantiate past balances.
Separate the headline numbers: inventory exposure versus confirmed loss
One reason this topic breeds confusion is numerical conflation. The $31 billion figure is the valuation of the parts population at issue — the universe the government must be able to track — not a quantified loss. The known problem set is smaller but still material: tens of thousands of parts off the books in 2026, with valuation in the low billions on some measures; more than one million parts worth over $85 million flagged in 2023 with unresolved status years later. An honest appraisal does two things at once: it resists sensationalizing the full $31 billion as “missing,” and it refuses to downplay the operational and financial risk inherent in an unreliable ledger governing a $31 billion asset class. Both are true.
What closure looks like: the accountability work the JPO must finish
The remedy is not mysterious; it is disciplined property management at enterprise scale. First, stand up and staff a program-specific property-accountability function with trained accountable officers and authority to drive corrective actions across government and contractor facilities. Second, complete a population-to-book reconciliation: for each part, establish ownership status, condition, location, and cost; for discrepancies, execute searches, document recoveries, and record validated losses through the proper financial and investigative channels. Third, enforce system integration rather than deference: where contractor systems are the only place a record exists, require feeds that populate the government’s accountable property system of record and validate them through periodic cycle counts and independent physical inventories.
Finally, link accountability outcomes to readiness metrics and contract incentives that matter. If analytics and ODIN produce fewer stockouts and faster repairs, those operational gains should correlate with a shrinking exception file in the accountable system. Incentive fees should be conditioned on both: on-wing availability and auditable property records. GAO’s record suggests the F-35 enterprise has repeatedly improved pieces of the sustainment machine without fully aligning them to the government’s fiduciary and audit obligations. Closing that last mile is not optional; it is how a mature program operates.
Bottom line
GAO’s evidence is decisive: the F-35 program controls a vast, valuable parts universe with an accountability regime that has not kept pace. The result is a measurable population of untracked items, a reliance on contractor data for core government assertions, and financial records that cannot withstand audit scrutiny. The program’s digital modernization — ODIN, dashboards, anomaly detection — is the right trajectory, but credibility arrives only when the government’s own accountable records reconcile to reality across hundreds of sites. Until then, inventory exposure remains a risk vector measured in tens of billions, and readiness will continue to pay for gaps the ledger cannot see.
Sources:
insidedefense.com, gao.gov, stripes.com, breakingdefense.com, media.defense.gov, f35.com
