The Pentagon's stockpile crisis is not a budget problem. It is an industrial one.

A PHONE CALL from the president to the Pentagon's second-in-command, reported to have been an angry one, prompted a hastily convened meeting of military and acquisition officials on Friday. The subject: how to fix a weapons shortfall that has grown so acute the military is reportedly letting some Iranian projectiles aimed at empty stretches of its own bases go unintercepted. That is the point at which munitions economics stops being a Pentagon spreadsheet and becomes a national-security risk.

The numbers are not ambiguous. According to a late-July report from the Centre for Strategic and CSIS Studies (CSIS), a Washington think tank, the United States has fewer than 827 Patriot interceptors in its stockpile and fewer than 278 THAAD (Terminal High Altitude Area Defence) interceptors. Before the war with Iran, which resumed in earnest in late February, the estimated inventory was around 2,200 Patriots of its two most modern variants and 452 THAAD missiles. The Pentagon has fired more than 1,200 Patriots and between 50% and 80% of its THAAD stockpile in one regional conflict. Each Patriot interceptor costs more than $4 million. At more than half of the total pre-war stock, the rate of consumption is not sustainable.

Yet the trouble is not money. The United States spends close to $1 trillion a year on its military-more than the next nine highest-spending nations combined. President Trump has described the stockpile as being "in very good shape", even as General Dan Caine, the chairman of the Joint Chiefs of Staff, told senators last week that without additional funding the Pentagon will "not get the munitions" it needs. The dissonance between rhetoric and reality reflects a deeper structural mismatch: the American defence industrial base was optimised for peacetime efficiency, not wartime surge. The crisis is not that America cannot afford its weapons. It is that America cannot build them fast enough.

The reason is not hard to see. Modern defence munitions are exquisitely engineered, deeply integrated and assembled from supply chains that stretch across hundreds of thousands of suppliers. The Pentagon itself acknowledges that it relies on more than 200,000 contractors, many of whose domestic sourcing status is unknown. Tomahawk cruise missiles take four to five years to replenish a pre-war stockpile. Patriot and THAAD production cycles run 18 to 24 months from component to final assembly. An analysis published in May in War on the Rocks found that munitions whose production surged during the Ukraine conflict did so not because of emergency funding alone but because their production lines had been kept warm by steady pre-crisis procurement, proactive supply-chain management and capital investment in new facilities. Systems whose lines had been allowed to go idle-Stinger missiles, for example, not bought by the Army in 18 years-proved agonisingly slow to restart.

The Pentagon has at least begun to respond. On July 30th the Army awarded Lockheed Martin a contract worth up to $58.6 billion for Patriot interceptors, converting a previous one-year deal into a seven-year procurement plan running to 2032. Lockheed has pledged to triple PAC-3 MSE production capacity and quadruple THAAD interceptor output by the end of 2030 and to invest $8-9 billion through that date in modernising more than 20 facilities across the country. A separate framework agreement with Raytheon's parent company, RTX, aims to boost Tomahawk production from the current rate of about 60 a year to eventually 1,000. Another deal with L3Harris targets component production for Patriot and THAAD systems.

To be sure, the direction of travel is correct. CSIS's Centre for the Industrial Base noted in mid-July that roughly 10,000 new firms have entered the defence market in the past two years, non-traditional companies received over $120 billion in contract obligations in fiscal 2025, and munitions contract obligations have risen 330% since 2010. The Pentagon's 2027 budget request shifts munitions spending toward lower-cost systems-defined as costing under $600,000 each-from 49% this year to 70% by 2031. These are measures worth taking.

But there are reasons for scepticism. The framework agreements with Lockheed, RTX and L3Harris are not yet firm contracts. Five defence-industry executives, speaking to Military Times, confirmed that companies cannot invest heavily in new production capacity until Congress appropriates the required funding. Investing before receiving government payments would weigh on free cash flow and could hurt second-half earnings, they said. In short, the Pentagon is asking contractors to build capacity before it has committed the money to pay for it. That was the pattern earlier this year, too: NBC News reported in May that despite widespread alarm the Pentagon had yet to ink any new multi-year contracts to purchase munitions since Mr Trump took office.

The incentive problem runs both ways. The defence contractors, as publicly traded companies, face pressure from shareholders to return capital. Mr Trump signed an executive order in January targeting firms deemed to be underperforming on government contracts while continuing to distribute profits, but the political threat has not yet translated into altered behaviour. Meanwhile, the Pentagon as monopsonist buyer-there is only one customer-has historically favoured low-cost, low-volume procurement that minimises fiscal exposure from year to year. That approach works when no one is shooting missiles at your bases. It does not work when one regional adversary can consume more than half your air-defence inventory in weeks.

The second-order consequences are already visible. The United States has been pulling munitions from stockpiles in Europe and Asia to feed Middle-East demand. Delays in weapons shipments, including Stinger missiles and Paladin howitzers, have reached Taiwan, where arms supplied since the 1950s are meant to deter China. Ukraine, despite Mr Trump's earlier pause and subsequent resumption of aid, continues to draw on the same production pipeline. If Iran hostilities intensify again-and the fragile ceasefire that held from April to February's escalation has clearly not endured-the mathematics grow uglier.

A CSIS defence analyst and retired Marine Corps colonel, Mark Cancian, warned earlier this month that continued fighting could deplete stockpiles so far that it would impair the military's ability to respond to China or North Korea. The analysis from the Conversation in April put the point more broadly: a $1-trillion-a-year military that cannot sustain its own stockpile in a single conflict is exposing a structural vulnerability that Russia and China are well placed to observe and, eventually, to exploit.

The better answer is not simply more money, though money is necessary. It is different procurement. The Pentagon needs to treat surge production as a core competency, not a crisis response. That means sustained, multi-year buy rates for the munitions that will be needed in the next war, not just the last one. It means keeping production lines running even when there is no active shooting, so that restarting does not take two years. It means investing in component suppliers-rocket motors, guidance chips, propellant-rather than merely asking prime contractors to make magic happen.

The first task is to turn those framework agreements into binding contracts with predictable funding. The Senate Armed Services Committee approved a $1.15 trillion National Defence Authorisation Act in June, backed by multi-year procurement authority for munitions, but the bill is not expected to become law until autumn. A separate supplemental is likely. The question is whether Congress will appropriate at the pace the military now says it needs-Capitol Hill officials and experts have estimated an additional $20 billion just to begin replenishing stockpiles to pre-2022 levels.

The Pentagon's crisis meeting on Friday was a symptom, not a cure. The stockpile will not be rebuilt in a fortnight. The real question is whether the American defence establishment can organise itself so that the next time a president makes an angry phone call, there is a stockpile on the other end of the line to answer it. That bargain is breaking. The task now is to mend it.