Senator Mark Kelly, a Democrat from Arizona, retired Navy aviator and former NASA astronaut, questioned Pete Hegseth at a Senate Armed Services Committee posture hearing in May 2026 about how much money the Department of War, formerly the Department of Defense, was requesting to expand production of the critical munitions the United States had just spent in the war on Iran. The figure on the budget table was 31.8 billion dollars. Hegseth looked down at his notes and said the figure was actually 53 billion, for munitions acceleration over fourteen critical munitions. Kelly pressed. Hegseth cited long-range fires, JASSM and LRASM and Tomahawk, at 238 billion. Then he added 40 billion for hypersonics, bringing the broader total to approximately 330 billion dollars. The exchange took a few minutes. When Kelly asked how long the replacement would take, Hegseth's answer was "months and years."

This piece is about that answer. About what the war on Iran has shown the West cannot do anymore, about the Chinese chokehold on the materials needed to rebuild what was just expended, about why the United States has spent more on Israel's missile defence than Israel did, and about what the Irish foreign-policy posture has been quietly assuming would always be there to lean against. The Irish question lands at the end, because the Irish answer is downstream of the American one. The piece works from the empty magazines back to Dublin.

Hegseth's "months and years" is not a careless answer. It is the operational reality the budget is trying to paper over. The 1.5 trillion dollar Fiscal Year 2027 defence request the Department of War unveiled is mostly a stack of purchase orders for things that cannot actually be procured at the rate they are being used. The structural argument the rest of this piece will make is that this matters not only for Washington and Tel Aviv but for every European capital that has been assuming the American magazine was deeper than it now visibly is. Dublin is one of those capitals. The neutrality framework Ireland has been operating under has always leaned, quietly, on the assumption that allied air defence had enough depth to make Irish defence forces a peacekeeping cohort rather than an existential question. That assumption is empirically failing in real time.

The arithmetic of Epic Fury

The June 2025 war on Iran ran for forty days. The Pentagon and US Central Command call it Epic Fury. The CSIS report by Mark Cancian and Brandon Park, Last Rounds? Status of Key Munitions at the Iran War Ceasefire, published in April 2026, sets out the burn rate across the four most-depleted munition types in the US arsenal.

Tomahawks. US naval assets fired more than 850 Tomahawk cruise missiles in the first month of the war. By the ceasefire the total had passed 1,000. Each missile costs around 2 million dollars, depending on which block and which contract. The Department of War's FY27 budget request includes 785 Tomahawks. The CSIS rebuilding analysis estimates production lead time at roughly 34 months from contract award to first delivery. The 785 missiles will not arrive in inventory before 2030.

Patriots. The PAC-3 MSE interceptors, at approximately 3.9 million dollars per round, are the single workhorse of US ground-based air defence. CSIS estimates US forces fired between 1,060 and 1,430 Patriot interceptors during Epic Fury, expending roughly half of the pre-war inventory in the first 39 days. Lockheed Martin announced in January 2026 a framework agreement with the Department of War to more than triple PAC-3 MSE production from 600 to 2,000 units per year, with the ramp completing by 2030. The expansion is the largest single missile production ramp the US industrial base has attempted since the Cold War. It will not match attrition at current burn rates. Ukraine, by way of comparison, has received around 600 modern Patriot interceptors over four years of war, a figure Volodymyr Zelensky pointed out in mid-2025 was less than the Middle East fired in three days.

SM-3 ship-launched interceptors. CSIS records that nearly half of the SM-3 inventory was expended in the first 39 days of Epic Fury, building on roughly 80 used the year before in support of Israel during the 2024 12-Day War. Standard build time is multi-year. The Navy's FY27 budget asks for 4.3 billion dollars for SM-6 procurement, the lower-end interceptor that pairs with SM-3 in the integrated air and missile defence stack. The full Standard family will not return to pre-war inventory before the late 2020s.

THAADs. The Terminal High Altitude Area Defense system is the high-end ground-based interceptor against ballistic threats. CSIS estimates 190 to 290 interceptors fired during Epic Fury, depleting somewhere between half and four-fifths of pre-war stocks. The United States has eight THAAD batteries worldwide, seven deployed and one in reserve. One battery was moved from South Korea to the Middle East during the war, leaving Northeast Asia visibly thinner. Public statements from Lockheed Martin frame replenishment over multiple years.

These numbers come from one analysis. They are corroborated across Washington Post reporting, Military Times reporting, Foreign Policy Research Institute analysis, Semafor coverage and US Naval Institute News briefings. The Department of War has not contested them. They are the operational floor the rest of this piece is built on.

The 200-to-90 problem

The Washington Post report of 21 May 2026, U.S. bears brunt of Israel's missile defense, Pentagon assessments show, is the cleanest single document on the joint-operation framing. Pentagon assessments record that the United States fired more than 200 THAAD interceptors in defence of Israel, roughly half of the total US inventory of that category. The US also fired more than 100 SM-3 and SM-6 missiles from naval vessels in the eastern Mediterranean. Israel, over the same campaign, fired fewer than 100 Arrow interceptors and approximately 90 David's Sling interceptors, with some of those used against less sophisticated projectiles inbound from Yemen and Lebanon.

The arithmetic is not a 2-to-1 ratio, it is a structural imbalance. The Pentagon and the Israeli embassy have publicly characterised the war as a joint operation in which the burden was shared equitably. The pre-agreed bilateral framework, in operational practice, placed high-end interception on US assets. The result is that Israel left the war with substantial domestic interceptor stocks intact. The United States left the war with half its THAAD inventory expended, half its Patriot stock expended, half its SM-3 stock expended, and queue depth for replacement measured in years.

It is the kind of imbalance that is easier to defend in public than in classified inventories. The Pentagon's framing is that operations were coordinated at the closest possible level for the benefit of both countries. The Israeli framing is the same. The arithmetic underneath those framings is one-way. The American magazine is the magazine that emptied.

For an audience accustomed to thinking of the United States and Israel as roughly comparable contributors to a coalition, the verified numbers are corrective. The coalition was real, and the war is over for now, and Iran's nuclear programme has been substantially degraded by US strikes the Pentagon has not been shy about claiming credit for. None of that changes the inventory floor that Hegseth and his successors will have to work back up from over the next decade. The US is short missiles. Its allies in Europe and the Pacific now know it.

What Switzerland and South Korea figured out

Switzerland ordered five Patriot batteries in 2022. The original delivery schedule was 2026 through 2028. The Ukraine war pushed it back. The Iran war pushed it back further. As of May 2026, the delivery schedule is five to seven years from now, and the price has roughly doubled, from approximately 2.3 billion Swiss francs to approximately 4.6 billion. Reuters and Aviation Week report that Switzerland is looking at IRIS-T, SAMP/T, the Israeli Arrow-3, and the South Korean Cheongung II as alternatives. Cancellation of the Patriot order outright is being discussed.

This is a country that is not at war and is not at risk of being at war and has no domestic political pressure to add air defence in haste. It ordered the systems for prudential reasons, has been waiting almost four years already, and is now being told it will wait at least five more, at twice the price. The supply chain visibly cannot serve countries in the queue who are doing nothing wrong. Israel and Ukraine are at the front of the line because they are at the most-active war. Everyone else, including formal NATO members, is somewhere further back.

South Korea worked out the same arithmetic when its government discovered that one of the THAAD batteries it had been depending on for ballistic-missile defence against North Korea had been moved to the Middle East. Patriot transfers from US stocks in Korea to the Middle East were also under discussion. The Guardian's reporting in mid-2025 documents the rattled tone in Seoul and the parallel discomfort in Tokyo. The American commitment to Northeast Asia, in operational terms, has been thinned by the Iran war. Seoul and Tokyo can read inventory schedules as well as anyone. The South Korean government is now investing roughly 172 million dollars in rare-earth stockpile expansion to reduce its dependence on Chinese supply chains for the materials that go into the magnets in the missiles. Korea is not waiting for someone else to fix the problem. It is starting to fix it itself.

The same conversation is happening in Tokyo. Officials there have been quieter about it. The substance is the same. American protection that was assumed to be available on demand is now visibly rationed. Allies are noticing.

The samarium answer

The single cleanest instance of the Chinese chokehold on Western military rebuilding is samarium. Samarium-cobalt magnets are used in the actuators and seekers of advanced munitions. Unlike neodymium, which is used in everything from electric vehicles to wind turbines and therefore moves through a global civilian market that licensing regimes find difficult to police, samarium is essentially a defence-only material. There is no civilian volume to hide a re-export inside. The licensing chokepoint is the entire bottleneck.

China produces all of the world's samarium. Keith Bradsher reported in the New York Times in June 2025, in the piece titled China's Chokehold on This Obscure Mineral Threatens the West's Militaries, that Lockheed Martin uses approximately 50 pounds of samarium-cobalt magnets per F-35 Lightning II. The figure is corroborated by Adamas Intelligence in its breakdown of rare-earth content in the F-35 airframe. Every F-35 the United States and its allies operate, and every F-35 sold to Switzerland and South Korea and Norway and the United Kingdom and Israel and Japan and Italy and Singapore and Belgium and Denmark and Australia, contains a quantity of a material that exists only because a Chinese refinery sold it to a Chinese-controlled magnet maker. The Tomahawk cruise missiles the United States fired by the thousand at Iran contain samarium-cobalt actuators. So do the THAAD interceptors and the SM-3s.

The Biden administration tried to fix this. In 2022 and 2023 the Defense Department awarded 35 million dollars to MP Materials, the operator of the Mountain Pass mine in California, to build a domestic samarium separation facility. MP added roughly 100 million dollars of its own funds to buy equipment. The equipment arrived, was unpacked, and the company concluded that the market was too small for a second producer to be commercially viable. The samarium processing line was left sealed in warehouses. James Litinsky, MP's chief executive, said publicly that the structural problem was that civilian demand for samarium does not exist at scale; the defence-only market is too narrow to amortise a competitive plant when the existing Chinese supplier could undercut on price.

The Biden administration also funded the Australian rare-earth refiner Lynas, in tranches totalling over 300 million dollars, to build a heavy rare-earth separation plant at Seadrift, Texas. The plant was not built. Lynas concluded the Texas economics did not work and is now expanding its existing facility in Kuantan, Malaysia. The Texas site is dormant.

The American defence industrial base was, by autumn 2025, in the position of having an entire fighter-jet programme and an entire long-range fires inventory dependent on a single Chinese-controlled material with no domestic substitution capacity. The NYT December 2025 follow-on, How U.S. Defense Industry Dodged a Rare-Earth Shortage After China's Curbs, documented the next step. Solvay, a Belgian chemical company, had been sitting on approximately 200 tonnes of rare-earth concentrates at its La Rochelle facility on the French Atlantic coast. The stockpile had been accumulated over the 1970s and 1980s, when the West still processed its own rare earths, and had been quietly held for decades. The American defence industry, working through Arnold Magnetics and Less Common Metals, signed agreements in 2025 to take the entire stockpile and convert it into magnet stock. Aisha Haynes, formerly responsible for supply chain resilience at the Department of Defense and retired from federal service in April 2025, told the Times in characteristic understatement that you never want to give any one nation that much power, because they can turn the source on and off whenever they want.

The 200 tonnes covers approximately one year of US defence-industry samarium demand, which runs at slightly less than 200 tonnes annually. The Solvay stockpile is the stopgap. There is no second stockpile of the same scale elsewhere in the OECD. When the La Rochelle inventory is processed, the United States needs to have built domestic samarium separation or it needs an exporter willing to fill the gap. There are no other exporters.

What China actually did with the export controls

The April 2025 Chinese export-control regime placed seven heavy rare earths, including samarium, dysprosium, and terbium, plus samarium-cobalt permanent-magnet materials and any downstream parts or components containing 0.1% or more Chinese-origin rare-earth input, onto licensed-export status. The regime is administered by China's Ministry of Commerce. Each shipment requires a separate licence. There is no general authorisation for defence-sensitive end-uses. The October 2025 expansion broadened the scope further. In November 2025, as part of a US-China commercial truce, China announced a one-year suspension of the October expansion for US-destined exports, running to 10 November 2026. The April 2025 regime remains in force.

The transcript that prompted this piece characterises the dual-use carve-out logic somewhat differently from the regulatory reality. The Chinese controls do explicitly cover dual-use rare-earth materials and the technologies to process them. Samarium is on the dual-use list. The reason it is harder for Western buyers to evade the controls on samarium than on neodymium is not that the regulatory regime is different; it is that there is no civilian commercial volume of samarium downstream of Chinese exporters into which a defence-bound shipment can be hidden. Neodymium magnets ship into electric-vehicle factories. Samarium magnets ship to missile programmes. The audit trail on a samarium licence is short and unambiguous.

The US Department of War's response, beyond the funded substitutions discussed above, includes a Defense Federal Acquisition Regulation Supplement rule, 252.225-7052, taking effect on 1 January 2027. From that date, no covered defence acquisition can use samarium-cobalt or neodymium-iron-boron magnets containing material originating in China, Russia, Iran, or North Korea. The rule is the forcing function behind the urgency of the substitution programmes. It is also a rule the supply chain cannot currently comply with. There is no path under which the US defence industry can be fully detached from Chinese rare-earth content by January 2027. The rule sets the standard. The industrial base will need waivers, exceptions, or some politically uncomfortable redefinitions of compliance to keep production lines moving.

The Trump administration's response since taking office in January 2025 has been to spend money. The MP Materials package was upgraded to a 400-million-dollar federal commitment, including a 150-million-dollar loan and a 15% federal equity stake. The Office of Strategic Capital, a Department of War financing vehicle, announced a 700-million-dollar package on 21 November 2025, with 620 million going to Vulcan Elements for permanent-magnet production and 80 million to ReElement Technologies in Indiana for rare-earth separation. Ucore Rare Metals, a Canadian company building a Strategic Metals Complex in Alexandria, Louisiana, has received 22.4 million dollars in Pentagon funding for separation of neodymium, praseodymium, samarium, gadolinium, terbium, and dysprosium. None of these facilities is currently producing at commercial scale. Each is years from delivering material that meets defence specifications. The Pentagon and the White House are publicly disputing the merit of the 1.4-billion-dollar package that includes ReElement and Vulcan, with the Office of Strategic Capital expressing scepticism that ReElement's process can scale and Peter Navarro publicly attacking the Office of Strategic Capital's vetting standards as private-equity frameworks unsuited to a national-security emergency. The Cato Institute has documented a Trump family connection to one of the firms in the package. The contracting environment is contested.

The broader project, branded as Project Vault, is a 12-billion-dollar Strategic Critical Minerals Reserve combining a 10-billion-dollar Export-Import Bank loan with roughly 2 billion in private capital. Donald Trump invoked the analogy with the Strategic Petroleum Reserve at the announcement in February 2026. The analogy holds in form. It does not hold in substance, because the Strategic Petroleum Reserve was built across decades when American oil production was secured by domestic capacity, while Project Vault is being built today to insulate American defence production from a foreign supply chain it does not currently control and cannot replace at commercial pace. The Strategic Petroleum Reserve has, separately, been drained at the largest weekly rate on record during May 2026 to manage oil-price pressure following the Strait of Hormuz disruption during Epic Fury. Project Vault is being built while the older reserve is being emptied.

Every other mineral, all the way down

Samarium is the cleanest single instance. It is not the only one. The Tomahawk cruise missile, the Patriot interceptor, the F-35 fighter and the THAAD launcher rely between them on roughly 18 critical minerals beyond samarium and neodymium. The figures below are drawn from the US Geological Survey's Mineral Commodity Summaries 2025 and the International Energy Agency's Global Critical Minerals Outlook 2025, which are the cleanest available primary sources. Where the YouTube briefing that prompted this piece cited rounder figures, the numbers below are the more conservative published ones.

Tungsten, used in control surfaces and high-temperature components. China accounts for roughly 80% of world production and the dominant share of refined chemicals.

Gallium, used in semiconductors that go into guidance and radar systems. China accounts for approximately 98% of world production. The December 2024 Chinese export ban on gallium, germanium, and antimony to the United States is suspended under the November 2025 truce, with conditional licensing in place. Intel's Leixlip campus in County Kildare, Intel's largest non-US fabrication facility, is one of the European semiconductor operations whose exposure to that licensing regime has not been publicly quantified by Irish business journalism.

Cobalt, used in samarium-cobalt magnets and in superalloys. The Democratic Republic of Congo accounts for approximately 70% of world cobalt mining. China accounts for approximately 80% of cobalt refining. The transcript that prompted this piece characterised China as the dominant miner. It is not. It is the dominant refiner. The structural pattern is consistent across the rare-earth and battery-metal sectors. Mining happens in Africa, Australia, and South America. Refining happens in China. The countries that own the deposits are not in control of the supply chain.

Bismuth, used in fuses and electronic components. China accounts for approximately 82% of world production.

Graphite, used in battery anodes and in heat-resistant components. China accounts for roughly two-thirds of mined production and is projected to retain more than 90% of battery-grade refining capacity through 2030.

Molybdenum, used in steel alloys for missile bodies and aerospace structures. Chinese production share is closer to 45% than the 81% figure that has been circulating; the higher number does not match USGS or other primary data.

Neodymium, used in the magnets that drive every part of advanced electronics, every electric vehicle, every wind turbine, and every guided weapon. China holds approximately 85 to 90% of refining and separation capacity, with the gap between 90% and 100% representing the modest non-Chinese capacity at Lynas Malaysia, Solvay, and a handful of smaller operations.

The pattern is that every part of the Tomahawk, every part of the Patriot, every part of the F-35, and every part of the THAAD passes through a refining step that occurs in a Chinese facility under Chinese export-control authority. The materials are not exotic. They are not new. They are the materials the West outsourced when refining was treated as a low-margin commodity step that did not need to remain on Western soil. The American defence industrial base has been buying its inputs from a state that has now declared a regulatory authority over them.

The structural revelation

The reason the Senate exchange between Mark Kelly and Pete Hegseth matters is not because Hegseth got the budget number wrong. It is because the budget itself is operationally incoherent at the rates at which munitions are being expended. The 1.5-trillion-dollar Fiscal Year 2027 request, of which 1.15 trillion is annual appropriations and roughly 350 billion comes through reconciliation, is the largest single-year defence ask in the history of the US Republic. The Department of War characterised it as historic and fiscally responsible. The substance is that it is a stack of purchase orders for missiles, interceptors, magnets, and chemical inputs that cannot be manufactured at the rate they are being used.

This is a different problem from a budget overrun. A budget overrun is a question of whether you can afford to spend the money. The American defence industry can afford to spend the money. The Department of War can afford to spend the money. The Trump administration can afford to spend the money. The constraint is supply: there is no Western capacity to convert that money into the actual physical articles being requested, on any timescale shorter than years, and that timescale assumes the Chinese controls on the input materials hold their current shape rather than tightening further. If Beijing tightens, the timescale lengthens. If the November 2025 truce collapses, the timescale lengthens.

The cumulative argument across the verified data is that the United States has been, since the early 2010s, running an alliance-protection architecture that assumes American interceptor stocks are deep enough to defend Israel, Ukraine, South Korea, Japan, Saudi Arabia, and continental Europe simultaneously. The Iran war demonstrated that the architecture cannot perform that function. It can defend Israel against Iran. It cannot do that and also handle Ukraine and also handle Northeast Asia and also handle a contingency anywhere else. The current operational tempo is the tempo of a single one-theatre commitment that has half-depleted the magazine. There is no extra capacity for a second theatre. There is no excess for any third party that has been planning around the assumption that American protection is available on demand.

This is the answer to the question Hegseth could not answer when Kelly asked how long replacement would take. Months and years. The point is not that it will take months and years. The point is that during the months and years it takes, the commitments the United States has made to other countries are denominated in inventory that does not exist and will not exist on the relevant timescale. The commitments are, in the language of military procurement, fiscally unfunded. They are not technically broken. They are simply not honoured at the depth the recipient countries have been planning around. The empty magazine is the structural commitment.

The Irish backstop

This is the piece's Irish hinge. Ireland's foreign-policy posture has, since the late 1950s, been formally non-aligned and, in operational practice, embedded in the security architecture of the European-North Atlantic alliance through a series of quiet, unwritten, and frequently unacknowledged arrangements. The Royal Air Force operates intercepts of unidentified aircraft within Ireland's Flight Information Region under a memorandum of understanding that has been continuously in place for roughly seventy years and has been quietly refreshed at intervals since. The Irish Air Corps has not operated combat jets since the retirement of the Fouga Magisters of the Light Strike Squadron in 1998. The Air Corps did not operate primary military radar until late 2025, when the Government opened negotiations with France for a 500-million-euro radar acquisition from Thales to be delivered between 2026 and 2028.

The 500-million-euro radar deal is the State formally acknowledging the air-defence gap. The earlier UK MOU was the State quietly relying on someone else to fill it. The combination has been the operational reality of Irish air defence for most of the period since the Cold War. The combination was structurally tenable while the British and American military magazines were assumed to have meaningful depth.

Ireland is now preparing to assume the rotating presidency of the Council of the European Union for the second half of 2026, from 1 July to 31 December. The Government, anticipating Russian hybrid pressure during the presidency, has begun discussions with European partners about deploying an Aster-equipped French Navy frigate to Dublin Bay, with the Aquitaine-class Bretagne named publicly as a possibility. The Government is fast-tracking a counter-drone procurement. These are responses to a strategic environment in which the implicit backstop of US and UK air defence is, for the reasons documented in the first half of this piece, visibly thinner than it was assumed to be when the Irish posture was last formally examined.

The structural-direction observation that the rest of this section is built on is this. The Defence (Amendment) Bill 2025, which removes the United Nations Security Council mandate requirement from the triple-lock authorisation for Defence Forces overseas deployment, was published by the Government on 21 May 2025. The Iran war began in June 2025. The legislative move to dismantle the triple lock predates the war by weeks. The Government is repositioning Ireland inside a NATO-aligned framework at exactly the moment that framework is visibly thinning. The triple-lock dismantling is moving with the strategic current, not against it, and the strategic current is being shaped by stockpile inventories the Government has not publicly acknowledged it is even reading.

Tánaiste Simon Harris frames the change as recovering Irish sovereignty from veto by United Nations Security Council permanent members. The framing has a public-facing logic that does not depend on the empty-magazine question. The substantive direction of travel, in terms of where Irish defence policy is being repositioned, points toward closer alignment with European Union and NATO frameworks that themselves face the structural-capacity problem identified above. Sinn Féin's foreign affairs spokesperson Donnchadh Ó Laoghaire characterised the Iran strikes as a dangerous and reckless act of aggression with no basis and called for Ireland to follow the Swiss example of refusing military overflights connected to the war. Social Democrats Senator Patricia Stephenson introduced a bill in the Seanad to prevent US military use of Shannon Airport and Irish airspace, citing the more than 3,000 US troops who transited Shannon during the relevant month and the unreported overflights documented by the Irish Times. Sinéad Gibney, the Social Democrats' defence spokeswoman, told RTÉ the Government was moving Ireland closer to NATO and abandoning the UN.

The opposition arguments are arguments about direction. The Government's counter-arguments are arguments about sovereignty and Programme for Government commitments. None of those arguments is engaging directly with the question this piece is asking, which is whether the assumptions Ireland has been making about backstop air defence remain operationally true. Stephenson's bill is the closest to engaging the operational question, because Shannon's role in the Iran war is the most concrete instance of the backstop architecture functioning in the way the Irish Government has been quietly relying on. The Irish Times revealed in April 2026 that 248 US military flights over Ireland had gone unreported during the relevant period because of what the Government described as administrative error. Even on the State's own accounting, the actual operational facts of Irish neutrality during the war on Iran were not what the public record claimed they were.

The Defence Forces budget arithmetic completes the picture. The Commission on the Defence Forces in February 2022 recommended three Levels of Ambition. The Government chose Level of Ambition 2, the middle option, committing to lift defence funding to 1.5 billion euros, in 2022 prices, by 2028. Budget 2025 allocated 1.35 billion euros. Budget 2026 allocated 1.5 billion euros nominal, a 15% increase, hitting the Level of Ambition 2 nominal target two years early. In real terms, adjusted for the price-level changes between 2022 and 2026, the LOA2 nominal target is being approached with somewhere between 80 and 85 cents of 2022 purchasing power per nominal euro. The Government has hit the target without hitting the substance.

The Defence Forces have stabilised at approximately 7,500 personnel. The LOA2 commitment is 10,500 by 2028. Recruitment and retention are widely characterised in the Defence Forces community as in crisis. The 47-year United Nations Interim Force in Lebanon mission, the single largest sustained overseas deployment for the Irish Defence Forces and the operational source of significant promotion, allowance, and retention pressure, ends on 31 December 2027. The Defence Forces Chief of Staff has publicly acknowledged the retention impact of the UNIFIL wind-down. Ireland's overseas posture is contracting at the same time the Government is repositioning the legislative framework for overseas deployment to make broader contingencies possible.

The Irish defence budget runs at approximately 0.4% of gross domestic product. The United States runs at approximately 3.4% and the formal NATO debate now centres on a 5% target. No Irish source identified during the research for this piece compares the 1.5-billion-euro 2026 Irish defence allocation to the 1.5-trillion-dollar US Fiscal Year 2027 request. The numerical coincidence is striking. The Irish allocation is one one-thousandth of the American request. The Irish allocation buys the Defence Forces. The American allocation buys a queue for inventory that, on Hegseth's own answer, takes months and years to fill.

What hasn't been said yet

The argument this piece has made is that the structural-direction observation already implicit in the verified data points to a single uncomfortable conclusion that no Irish minister, no senior Irish commentator, no IIEA paper, no Royal Irish Academy briefing, and no published Irish Centre for Defence Policy analysis has yet made explicitly. Four observations sit on the table waiting for a writer.

First, the empty-magazine story and the Irish neutrality story are the same story. The neutrality framework has been operationally sustainable because the United States and United Kingdom held the air-defence depth Ireland's force structure does not. The framework is being held to the same standard now that the supporting depth is visibly thinning. The triple-lock dismantling is moving with the depletion, not in response to it, but it cannot operate as if the depletion did not exist.

Second, Intel Leixlip is the single largest non-US Intel fabrication facility, employs around 5,000 people, and is a significant contributor to Irish gross domestic product and to Ireland's industrial position in the European Union. Its exposure to Chinese export controls on gallium and rare earths has not been publicly reported by Irish business journalism. The Irish industrial base has the same supply-chain exposure as the American defence base, in a different sector, with the same Chinese chokepoints, and the same November 2025 truce sitting underneath whether the controls bite or lift. No Irish-published analysis isolates this risk.

Third, the proportional comparison between the Irish 1.5-billion-euro defence allocation and the US 1.5-trillion-dollar defence request is a comparison nobody has put in print. The numerical coincidence makes the comparison easy to surface. The substantive point is that the Irish allocation, even at the Level of Ambition 2 nominal target, sits below the 0.5% GDP floor that any serious capability-recovery would require. The political conversation about Irish defence runs almost entirely inside the LOA2 envelope. The envelope is not big enough to engage the operational question this piece has documented.

Fourth, the Iran-related Western capacity contraction touches Irish business and Irish citizens in ways that have appeared in Irish journalism in fragments. The 28% rise in diesel prices and 25% rise in petrol prices in early 2026, the related fuel protests, the no-confidence motion launched by the opposition, the cybersecurity warnings from the National Cyber Security Centre, the Strait of Hormuz energy-price shock, and the unreported military overflights have appeared, but as separate stories. The conjunction has not been written. The conjunction is the story.

The publication that lands the conjunction first has the editorial position. The opposition parties have the legislative bills. The Government has the Programme for Government. The IIEA has the capability-recommendation framework. None of them has, as of the date of this piece, made the structural-direction argument that the Western backstop the Irish neutrality framework has been quietly relying on has been visibly debited and cannot be quickly refilled. That argument is the argument this piece has just made.

Months and years

Hegseth's answer to Kelly was an honest one. Months and years. The phrase covers the time it takes to replace what was just used. It also covers the time during which the commitments the United States has made to other countries are denominated in inventory that does not exist. The honest framing of those commitments is that they are deferred, dependent on Chinese export controls remaining at their current level of stringency rather than tightening, dependent on the Trump administration's substitution programmes performing at scale, dependent on the substitution programmes not being captured by political interference of the kind Cato has already documented in the Trump-family-connected rare-earth firm, and dependent on the Solvay stockpile holding the line for the year or so it can.

Each of those dependencies has visible failure modes. The Chinese controls are politically reversible at any moment. The substitution programmes are visibly failing at MP Materials and Lynas and have not yet been built at Ucore and ReElement and Vulcan. The 2027 DFARS rule that is meant to be the forcing function for substitution is not technically achievable on its stated timetable. The Strategic Critical Minerals Reserve is being built while the Strategic Petroleum Reserve is being drained. The Western defence industrial base is in the position of having to build out a refining capacity it allowed to atrophy over four decades, on a timescale of years, while running the most-used air-defence inventory in the modern era against threats that the same industrial base helped its allies provoke. The structural picture is not a picture of incremental difficulty. It is a picture of a system that has been operating on accumulated stocks and now does not have them.

The Irish question downstream of all of this is what the State does once the underlying assumption is named. The Government has, by repositioning the triple lock, made a directional choice that points toward closer alignment with European defence frameworks that themselves face the supply-chain problem documented above. The opposition has made arguments about Shannon and about overflights and about the legal characterisation of the Iran war that engage parts of the question but not the structural whole. The publication ethics of this beat are that the structural whole should be named, by someone, before the next set of decisions is made. The structural whole is that the inventory underwriting Irish neutrality has been substantially expended, that replacing it will take time measured in years rather than in budget cycles, and that the Irish political conversation has not yet caught up with this fact.

The substantive question this piece is leaving in the air is the one that the accountability programme companion to the rules-of-capital pair already poses in different form. Which of the assumptions underwriting Irish neutrality is the Government willing to defend in public, and which is it willing to revise? The honest version of that question, after the war on Iran, is which of those assumptions the Government still believes is operationally true. The answer to that question has not been requested by the opposition in the form this piece has just put it in. The answer has not been offered by the Government in the form this piece has just put it in. The answer is, on the available verified evidence, that several of the assumptions are no longer true and the Government has not said so.

The closer is Hegseth's. Months and years. That is the time during which the Irish foreign-policy debate has to surface what it has been assuming. That is the time during which the conjunction this piece has documented has to be written down somewhere by someone in print rather than left to be inferred from a stack of separate Irish Times reports and a CSIS PDF. The Government's gradual move on the triple lock and its 500-million-euro French radar order and its EU presidency air-defence requests indicate that the Government's professional civil service has been reading the same inventory tables this piece has. The political class has not. The publication ethics of this work are to put the inventory tables in front of the political class until they admit they have read them.

The verified data are above. The conjunction is named. The question of which assumptions the State is willing to defend in public is now somebody's question to answer.

This piece is part of the publication's structural-direction arc. Related: What the surplus hides, The war you can't afford and the crisis they won't show you, The fab and the island, and the wealth-tax diptych What the Finance Minister Already Knew and The Wealth Tax Ireland Already Has.

Source notes. Center for Strategic and International Studies: Mark Cancian and Brandon Park, "Last Rounds? Status of Key Munitions at the Iran War Ceasefire", April 2026. CSIS, "Rebuilding U.S. Missile Inventory: A Multiyear Project". Washington Post, "U.S. bears brunt of Israel's missile defense, Pentagon assessments show", 21 May 2026. New York Times, Keith Bradsher, "China's Chokehold on This Obscure Mineral Threatens the West's Militaries", 9 June 2025; "How U.S. Defense Industry Dodged a Rare-Earth Shortage After China's Curbs", 22 December 2025. Reuters, "Swiss considering rival air defences after Washington delays Patriots over Iran war", 13 May 2026. Guardian, "Hasty redeployment of US missiles from South Korea to Middle East leaves Seoul rattled". Kelly.senate.gov, "WATCH: In SASC Hearing, Kelly Challenges Hegseth on Munitions Burn Rate, 'No Quarter' Comment, and Massive Defense Budget Proposal", May 2026. Bloomberg, "Pentagon doubts over rare earths deal provoke White House clash", 21 May 2026. Mining.com syndication of the same story. Cato Institute, "White House intervention bolsters Trump Jr.-connected rare earths firm". US Geological Survey, Mineral Commodity Summaries 2025. International Energy Agency, Global Critical Minerals Outlook 2025. Department of War, "$1.5 Trillion Budget Request Prioritizes Service Members, Modernization", April 2026. Department of War press release on $35 million MP Materials samarium grant. Tech Startups, "Pentagon Lends $700M to Vulcan and ReElement to Boost Domestic Rare Earth Magnet Production Amid China Tensions", 21 November 2025. Mining Weekly, "Ucore secures $18.4m US DoD funding for Louisiana rare earth processing plant", 20 May 2025. Solvay press release, "Solvay strengthens rare earth supply chain", 2025. Department of Defense Federal Acquisition Regulation Supplement rule 252.225-7052, effective 1 January 2027. Newsweek, "America's Emergency Oil Reserve Is Shrinking Fast". Australian Department of the Prime Minister and Cabinet, Critical Minerals Strategic Reserve. Irish Times, "Ireland to buy €500m military radar system from France", 16 December 2025; "US troop numbers passing through Shannon soar 80%", 24 April 2026; "248 US military flights over Ireland went unreported due to 'administrative error'", 18 April 2026; "Iran war has drained US supplies of critical and costly weapons", 24 April 2026; "Iran conflict sharpens debate about Ireland's role in global security", 2 March 2026; "Drone wars: Irish Defence Forces watching the skies as EU presidency set to begin", 26 May 2026. RTÉ, "Government publishes legislation to remove Triple Lock", 21 May 2025; "End of UNIFIL mission a 'source of regret', says Harris", 31 October 2025; "Including services in OTB 'not implementable' - Taoiseach", 26 May 2026. Sinn Féin party releases on the Iran war. Social Democrats party releases on the Shannon Airport Seanad bill. Gov.ie, "Statement by the Tánaiste on developments in the Middle East", 22 and 23 June 2025. The Commission on the Defence Forces, Report of the Commission on the Defence Forces, February 2022. Institute of International and European Affairs, "A New Level of Ambition: The Capability Recommendations of the Commission on the Defence Forces". US Naval Institute News on Navy munitions budgets. Lockheed Martin, "Lockheed Martin and the U.S. Department of War Expand THAAD Interceptor Production", 2026. Lockheed Martin, "Lockheed Martin and Department of War Advance Landmark Acquisition Transformation to Accelerate PAC-3 MSE Production", 6 January 2026. Bradsher's NYT quote of Aisha Haynes (formerly Department of Defense supply chain resilience, retired April 2025).

Overwatch Report is an independent publication. We have no financial positions in any entity mentioned.