Sections

Commentary

Do low munitions inventories invite aggression?

U.S. military personnel take away Joint Direct Attack Munitions, removed from a U.S. Air Force B-1 Lancer bomber at RAF Fairford in southwest England on March 15, 2026.
U.S. military personnel take away Joint Direct Attack Munitions, removed from a U.S. Air Force B-1 Lancer bomber at RAF Fairford in southwest England on March 15, 2026. (Henry Nicholls/AFP via Getty Images)

According to Pentagon officials, U.S. precision weapons inventories have taken a big hit due to the wars in Ukraine and the Middle East. About half of some offensive weapons, like Tomahawk cruise missiles and Joint Air-to-Surface Standoff Missiles, have been expended in attack missions. These can be used to attack land targets or surface ships. Even higher fractions of many advanced defensive weapons, like certain types of the Navy’s Standard Missiles as well as Army Patriot and Terminal High Altitude Area Defense interceptors, have been used up by American forces (or transferred to Ukraine over the last four years). Those are all important to defend American bases, ships, and allied interests. Although the Pentagon has been trying to reprioritize the Pacific theater since roughly 2010, assets previously assigned to that region have been shifted from South Korea and Japan to the Middle East in light of recent demands.

This is all troubling, to be sure. There is no doubt that, as concerns precision weapons, more is better—and lots more is lots better. Deputy Secretary of Defense Stephen Feinberg and other top military officials are right to seek new deals with industry to spark expanded production of all of the above-mentioned weapons, as well as several other types. But we need not despair over the present situation. Deterrence of Xi Jinping, Kim Jong Un, and Vladimir Putin is unlikely to fail due to depleted U.S. weapons inventories. Here’s why.

Low stockpiles do not equal failed deterrence

To begin, while weapons inventories are indeed too low—largely because the military services rarely prioritize the procurement of munitions—they have not fallen below some mystical minimum. There is no such threshold. Calculations about the need for offensive and defensive weapons are always approximate at best. (The great political scientist Richard Betts put it well in his book about such matters, entitling one chapter “Lies, Damn Lies, and Readiness Statistics.” The “probabilities of kill” for any given type of weapon can easily vary by 50% up or down due to things like both sides’ electronic warfare and ability to conduct barrage attacks.

Even more to the point, we do not know how long any war would last or how great its ordnance consumption rates would turn out to be. An attempted Chinese invasion of Taiwan, if defended correctly, might be defeated in a week; a Chinese blockade of Taiwan, by contrast, might involve less intense shooting but last weeks or months. Either way, we’d be prudent to have larger stocks of munitions than we now possess, as would U.S. allies in key regions—but there is no clearly demonstrable minimum. What the Pentagon calls munitions “requirements” are really just rough estimates of expected need.

In addition, according to my own calculations as well as wargames run by various think tanks and Pentagon contractors, there are at least four additional fundamental uncertainties beyond inventories of key weapons that would determine the outcome of intensive wars against China, Russia, or North Korea—heaven forbid that any of those ever happen. Any would-be aggressor would have to weigh all of them before launching a war.

Four uncertainties that matter more than inventories

First is the question of who lands the first blow in any war and how effective it turns out to be. Because of the speed and accuracy of modern missiles, as well as the huge numbers of drones now on the market, forward bases are vulnerable. Even bases within the deep interiors of the United States, its allies, and its potential adversaries could suffer strikes. Such physical attacks would surely be complemented by cyberattacks, including against key civilian infrastructure needed to support the deployment and resupply of a country’s military forces. 

Second is the survivability of communications and reconnaissance systems in a future war (what the military calls “C4ISR,” or command, control, communications, computer, intelligence, surveillance, and reconnaissance systems). Physical attacks against satellites, undersea cables, radars, and transmission antennas, as well as cyberattacks against key communications systems, should all be expected. Any country suffering interruptions to its sensor-shooter connectivity could see smart weapons turned into dumb ones very quickly. 

Third is the potential for geographic expansion and/or nuclear escalation with any of these potential conflicts. Notably, whoever is losing a major war will have major incentives to conduct nuclear brinkmanship. A country probably would not immediately resort to mutually suicidal all-out nuclear attack, but it could detonate a nuclear weapon or two—as a demonstration strike or a counterforce strike against a key enemy military base—hoping that a limited attack could seriously disrupt enemy military operations as well as its will to keep fighting. Even if it were not true, a country might believe its own crucial nuclear forces to be under attack because of the commingling of C4ISR assets used for both conventional and nuclear purposes, and retaliate accordingly, with dangerous possible results.

Fourth, as we are witnessing now, even in regard to relatively weak Iran, is the way economic considerations factor into military conflict—shaping its trajectory and, even more so, influencing each side’s war termination strategies. In a future war over the Taiwan Strait, for example, the United States might physically block much of China’s access to oil; China would surely embargo shipments of crucial commodities like rare earth minerals and active pharmaceutical ingredients in life-saving medicines to America and its allies. Who would blink first as a result of such considerations?

What past wars tell us about uncertainty 

If one attempts to quantify these uncertainties, the results are sobering. Indeed, in past and generally simpler wars—Iraq, Kosovo, Afghanistan, Iraq again, and now Iran—initial expert estimates about a conflict’s duration and casualties were often off by 100% or more compared with actual results. Uncertainties would be even greater in any conflict against nuclear-armed China, Russia, or even North Korea.

The good news is that any of these dynamics could work to the American advantage as well as its disadvantage. To be sure, we as a nation should continue to mitigate vulnerabilities, make the country more resilient, and yes, expand our magazine depths as well as surge production capacities for crucial munitions. But any adversary thinking that depleted U.S. inventories of several key weapons create a major window of opportunity for aggression against American interests would be making a huge and quite probably foolish roll of the dice.

The Brookings Institution is committed to quality, independence, and impact.
We are supported by a diverse array of funders. In line with our values and policies, each Brookings publication represents the sole views of its author(s).