Analysis

The Art of the Arms Race

To avoid disaster, the United States must relearn crucial Cold War lessons.

arms race-senor-salme-illustration-lead
arms race-senor-salme-illustration-lead
Señor Salme illustration for Foreign Policy
By , a professor of global affairs at the Johns Hopkins School of Advanced International Studies.

Arms control is dying, and arms races are roaring back to life. Over the past two decades, key pillars of the superpower arms control regime erected during the Cold War have collapsed, one by one: the Anti-Ballistic Missile Treaty, the Conventional Armed Forces in Europe Treaty, the Intermediate-Range Nuclear Forces Treaty, and the Open Skies Treaty. The most important U.S.-Russian agreement that remains, New START, may become a casualty of Russian President Vladimir Putin’s war in Ukraine. China, meanwhile, is rapidly building up its conventional and nuclear forces as part of a push for dominance in the Pacific and beyond. Around the globe, emerging technologies are promising dramatic advances in military power.

Welcome to a world primed for arms races—a world in which tensions are sharp, the military balance is hotly contested, and there are ever fewer constraints on which kinds and what quantity of weapons great powers can wield. This new world will, in fact, be replete with challenges reminiscent of an earlier era of rivalry. To avoid disaster, the United States must relearn what it knew during the Cold War: how to arms-race well. 

THE BACK TO THE FUTURE ISSUE: This article appears in the Summer 2022 print issue. Subscribe now for unlimited access to FP, plus the print magazine.

Arms control is dying, and arms races are roaring back to life. Over the past two decades, key pillars of the superpower arms control regime erected during the Cold War have collapsed, one by one: the Anti-Ballistic Missile Treaty, the Conventional Armed Forces in Europe Treaty, the Intermediate-Range Nuclear Forces Treaty, and the Open Skies Treaty. The most important U.S.-Russian agreement that remains, New START, may become a casualty of Russian President Vladimir Putin’s war in Ukraine. China, meanwhile, is rapidly building up its conventional and nuclear forces as part of a push for dominance in the Pacific and beyond. Around the globe, emerging technologies are promising dramatic advances in military power.

Welcome to a world primed for arms races—a world in which tensions are sharp, the military balance is hotly contested, and there are ever fewer constraints on which kinds and what quantity of weapons great powers can wield. This new world will, in fact, be replete with challenges reminiscent of an earlier era of rivalry. To avoid disaster, the United States must relearn what it knew during the Cold War: how to arms-race well. 

To be sure, arms races—in which two or more rivals compete to secure a favorable military balance—have an awful reputation. At best, they are viewed as a mindless accumulation of weapons or the product of a sinister military-industrial complex and, at worst, as a principal cause of spiraling tensions and cataclysmic war. “[T]he United States is piling up armaments which it well knows will never provide for its ultimate safety,” U.S. President Dwight D. Eisenhower told his National Security Council in 1956, according to a memo of the meeting. “We are piling up these armaments because we do not know what else to do.”

But arms-racing has unfairly gotten a bad name. As the geopolitical environment becomes nastier, it helps to take a more objective look.

As the United States’ sharpest Cold War thinkers understood, arms-racing is hardly mindless. Preserving a favorable balance of power against an aggressive adversary is the best means of deterring war, not an incitement to it. An arms race, moreover, is a deeply strategic interaction that can be shaped through smart investments and tilted in one’s favor over time. Arms control, finally, is properly seen not as an alternative to arms-racing but as a vital component of a strategy for attaining a competitive edge. Today, the United States has a chance to thrive amid intensifying military rivalries—but doing so will require Washington to reacquaint itself with the art of the arms race.


Arms races are timeless, but the term became commonplace only in the early 20th century. New technologies, such as the dreadnought battleship and the airplane, were creating the potential for rapid shifts in the military balance. Intensifying great-power tensions made the search for military superiority more urgent. In the decades before World War I, for instance, the competition between Britain and a rising Germany played out in a feverish contest to build the most and best battleships.

Yet it was during the Cold War, with the advent of nuclear weapons and the rise of strategic studies as an academic discipline, that our understanding of arms races really matured. Scholars such as Samuel P. Huntington and Colin S. Gray sharpened the definition of an arms race—essentially, an open-ended, back-and-forth contest in which rivals sought to dominate the military balance and reap the strategic rewards that followed. In government and academia, analysts studied the development of the U.S. and Soviet military arsenals and the degree to which moves by one side influenced moves by the other. Amid a long bipolar struggle for supremacy, the intricacies of the superpower arms race became a veritable obsession for intellectuals and policymakers alike.

The U.S.-Soviet military competition, of course, soon took on terrifying proportions. And as Moscow and Washington each acquired the ability to destroy human civilization with nuclear weapons, “arms race” became a term of opprobrium. The nuclear arms race was often seen as an exercise in absurdity—a reminder of how the search for security could cause existential insecurity instead.

The arms control pacts of the 1970s and later were, in part, an effort to reduce this insecurity by capping the superpowers’ nuclear arsenals and constraining those capabilities that were considered destabilizing, such as missile defense systems. (The theory was that one side might more readily consider launching a nuclear first strike if it had the ability to shoot down the other side’s retaliatory barrage of missiles.) The language of mutual assured destruction—the idea that no one could win a nuclear arms race and that it was dangerous to try—became pervasive. “We do not want a nuclear arms race with the Soviet Union,” U.S. Defense Secretary Robert McNamara declared in a 1967 speech. “The action-reaction phenomenon makes it foolish and futile.”


But in reality, things weren’t so simple. “The armaments race,” Eisenhower acknowledged to U.S. Secretary of State John Foster Dulles in 1957, “was a result rather than a cause”: The superpowers armed themselves because they were enemies, not vice versa. Winning the arms race—or at least not losing it—was imperative: The threat of war, or simply of a Western geopolitical collapse, would surely increase if an expansionist rival attained a decisive military edge. The shrewdest observers realized, moreover, that arms-racing was not a foolish, robotic endeavor. It was a discipline that rewarded creative thinking and strategic insight.

Arms-racing has unfairly gotten a bad name. As the geopolitical environment becomes nastier, it helps to take a more objective look.

This more sophisticated U.S. approach to arms-racing was epitomized by Andrew Marshall, a longtime defense intellectual who became the first director of the Office of Net Assessment, the U.S. Defense Department’s in-house think tank that rigorously assessed the military balance. Marshall argued that McNamara’s “action-reaction” model was too simplistic: Soviet and U.S. arms programs reflected historical legacies and bureaucratic biases as much as any tit-for-tat process. More importantly, since Washington could not responsibly avoid a military competition with Moscow, it needed to shape that interaction to its advantage. “[T]he United States will have to outthink the Soviets,” Marshall wrote in 1972, since it could not “continue to outspend them substantially.” The key was making Soviet costs rise and difficulties multiply by identifying “areas of U.S. comparative advantage” and steering “the strategic arms competition into these areas.”

Case in point was the U.S. strategic bomber program. Moscow, Marshall pointed out, had an exaggerated fear of aerial attack, because Adolf Hitler’s Luftwaffe had destroyed much of the Soviet air force on the ground in 1941. By building even a modest bomber fleet, Washington could—and, indeed, did—goad the Kremlin to invest heavily in air defenses, diverting resources from offensive capabilities more threatening to the West. And during the decisive final decade of the Cold War, Marshall’s logic was pervasive: An array of targeted U.S. military investments put great strain on the Soviet Union by negating plans and capabilities that Moscow had assembled at enormous cost.

The development of precision-guided munitions, low-flying cruise missiles, and stealth aircraft upended the Soviet concept of operations in Europe by giving the Pentagon the ability to wreak havoc deep in the enemy’s rear. The deployment of highly accurate intercontinental ballistic missiles (ICBMs), along with improved targeting capabilities, threatened Moscow’s plan to keep its leaders alive during a nuclear war by sheltering them in a fantastically expensive bunker complex. U.S. President Ronald Reagan’s Strategic Defense Initiative—a plan for a space-based missile shield—posed a potentially dire, if distant, peril to the efficacy of the land-based missile force Moscow had spent decades developing. U.S. defense programs, a 1982 Pentagon planning document stated, should “impose disproportionate costs, open up new areas of major military competition and obsolesce previous Soviet investment.”

Contrary to most predictions, aggressive arms-racing actually enabled historic arms control: Reagan’s strategic buildup gave Moscow an incentive to make deep, disproportionate cuts in its arsenal of intermediate-range ballistic missiles and heavy ICBMs. It also put an economically and technologically declining Soviet Union at such a steep competitive deficit that its leaders eventually opted to sue for peace. “If we won’t budge from the positions we’ve held for a long time, we will lose in the end,” Soviet leader Mikhail Gorbachev conceded in 1986. “We will be drawn into an arms race that we cannot manage.” For the United States, winning the superpower military competition was a prerequisite to winning the larger Cold War.


Washington’s aptitude for arms-racing declined after the Cold War ended: The United States possessed such military dominance that it seemingly had less need for a creative, ruthless strategy. Yet that generous margin of safety is now gone, which means that Washington must master an old discipline anew.

Russia’s invasion of Ukraine marked the culmination of a two-decade buildup of conventional and nuclear forces meant to allow Moscow to batter its neighbors while using the threat of nuclear escalation to hold Washington at bay. Russia’s military may have performed abominably in Ukraine, but its conventional and nuclear capabilities—paired with Putin’s increasingly aggressive behavior—will threaten NATO for years to come. China is following a similar playbook by developing power-projection capabilities to coerce its neighbors, anti-access and area denial capabilities to keep U.S. forces at a distance, and a growing nuclear arsenal to deter U.S. policymakers from intervening in the first place. Russia and China have been arming themselves to support their determined programs of geopolitical revisionism—and they have absorbed many lessons about arms-racing the United States has forgotten.

For years, Beijing did not try to match Washington’s military platform-for-platform. It invested in specific capabilities—anti-ship missiles, air defenses, and anti-satellite weapons, to name a few—that threaten the aircraft carriers, communications satellites, and regional bases the United States uses to project power worldwide. Beijing, in other words, has taken Marshall’s advice to heart: It is forging a Chinese way of war that could make the American way of war obsolete, much as Washington made Moscow’s way of war obsolete in the 1980s.

The United States has a chance to thrive amid intensifying military rivalries—but doing so will require Washington to reacquaint itself with the art of the arms race.

There isn’t much relief in sight. If current trends continue, Washington will confront not one but two nuclear peer challengers by decade’s end. Notwithstanding Russia’s losses in Ukraine, the balance of conventional forces along the Eurasian peripheries of the U.S. alliance system will be fraught—if not unfavorable. As during the Cold War, a dangerous military imbalance could tempt U.S. rivals to forcibly contest the status quo, or it could simply eat away at the foundation of confidence on which the U.S. alliance network rests. Preserving U.S. interests will once again require running an arms race—and winning it.

Victory will be partially a matter of money. Even the most brilliant brains cannot forever compensate for a dearth of dollars. The Pentagon will require greater defense spending to preserve a conventional edge vis-à-vis China and Russia simultaneously. It will also need a larger nuclear arsenal to deter two nuclear peers rather than one. Major outlays may be necessary to turn tantalizing technologies—artificial intelligence, quantum computing, synthetic biology—into real capabilities that can be fielded at scale. Military outlays equivalent to at least 5 percent of GDP, as compared with the less than 3.5 percent that the United States currently spends, will probably be the minimum price of peace through this decade and beyond.

But even if the money flows, Washington must also outthink its rivals in order to outperform them.


Outthinking one’s rivals first requires not deceiving oneself. Arms control advocates sometimes argue that Washington should unilaterally limit its own capabilities—whether the development of thermonuclear weapons in 1950 or military applications of AI today—in hopes that adversaries will do likewise. This almost never works.

We now know that a U.S. decision to defer building the hydrogen bomb in the early 1950s would simply have allowed the Soviet Union to build it first. When McNamara halted the U.S. strategic buildup during the 1960s, Moscow raced forward to claim a position of parity. “When we build, they build. When we cut, they build,” U.S. Defense Secretary Harold Brown quipped in 1979. The particular technologies change, but the hard truth doesn’t: Securing restraint from an autocratic adversary typically requires demonstrating that it can’t run an arms race unopposed.

Second, arms-racing effectively requires knowing the enemy intimately. One of Marshall’s insights was that understanding what made the Soviets tick was vital to throwing them off balance. Similarly, there is no good way to make decisions about present-day U.S. military programs without grasping what Russia and China want, what they fear, and how they intend to operate. There are, alas, no shortcuts: During the Cold War, it took a generational investment in Sovietology to get inside the enemy’s head.

This knowledge is so important because an arms race neither requires nor rewards competing equally everywhere. The United States doesn’t need to emulate every Chinese breakthrough in hypersonic weapons. These weapons can’t provide, at a reasonable cost, the volume of firepower Washington would need in the Western Pacific. The Pentagon also shouldn’t match the Kremlin’s vast short-range nuclear arsenal: The United States simply needs enough limited nuclear options to keep adversaries from feeling emboldened to probe the space between U.S. conventional forces and the strategic nuclear arsenal.

The better approach is to think asymmetrically—to use distinct U.S. advantages to disrupt the enemy’s theory of victory and drive up its costs. The way to devalue China’s military buildup vis-à-vis Taiwan, for example, is for Washington and its allies to exploit a key advantage: Defending a rugged island surrounded by rough seas is far easier than conquering it. They should do so by fielding overwhelming numbers of anti-ship missiles, sea mines, unmanned aerial and underwater vehicles, and other cheap capabilities that can turn a cross-strait invasion into a bloody nightmare for Chinese forces. Likewise, if Beijing wants to run an intermediate-range missile race, Washington can use its network of allies to turn a present Chinese advantage into a future liability. After all, U.S. intermediate-range conventional missiles based on allied territory can easily reach the Chinese mainland, whereas Chinese intermediate-range missiles cannot reach the United States. And as China pours more money into aircraft carriers and other large vessels, Washington can hold a generation’s worth of naval modernization at risk by maintaining its edge in undersea warfare. By consistently challenging Beijing’s plans and depreciating its capabilities, Washington can eventually force Chinese leaders to question what an arms race will achieve.

Here, a related rule is helpful: Don’t forget the defensive side of the arms race. Today, as in the past, arms control advocates often claim that ballistic missile defenses are destabilizing, or simply useless, because they can be beaten by cheap countermeasures. Yet U.S. missile defenses are improving rapidly, while the use of directed energy weapons (such as lasers) and other new technologies may soon mitigate problems such as the high cost and limited quantity of interceptors. Fielding limited ballistic missile defenses against Russia and China—not just rogue states like North Korea—can complicate Moscow’s and Beijing’s doctrines of nuclear coercion, which envision using a small number of nuclear strikes to disrupt or deter U.S. intervention in a regional conflict. It can also push Russian and Chinese costs skyward by forcing them to invest more in expensive, novel nuclear delivery vehicles—such as a nuclear-armed submarine drone and other doomsday device-like weapons Putin has brandished—that can defeat missile defenses only at a very high price.

Arms races, of course, have both qualitative and quantitative dimensions. That reminds us of another principle: Numbers are not the only things that matter. The key U.S. achievement in the 1980s was to leap ahead even in a situation of numerical parity. Revolutionary improvements in the accuracy of U.S. ICBMs made Soviet officials fear for the survival of their nuclear forces. While it’s clear that Washington’s nuclear arsenal will need to grow in the coming years, maintaining a favorable balance will equally require exploiting U.S. advantages in missile accuracy, ISR (the intelligence, surveillance, and reconnaissance capabilities that provide unparalleled global awareness), and other qualitative factors.

Deterrence, though, is a state of mind: It hinges entirely on what one side thinks the other can and will do. So U.S. policymakers should remember that perception is as important as reality. During the 1980s, the Pentagon used crafty informational strategies—dribbling out news about stealth technology, advertising the ability to sink Soviet nuclear missile submarines, dramatically revealing (and sometimes exaggerating) the effects of precision-guided munitions—to manipulate Moscow’s perceptions of the military balance. This time around, the United States may try to instill caution in Russia or China by demonstrating some sophisticated new capability—or lure them into unrewarding areas by making them fear some technological breakthrough that has not in fact occurred. New technology creates new possibilities, with the cyber field particularly ripe for deception because the true balance of capabilities is so difficult to know.

All this implies a sharper, tenser competition. Yet a final lesson from the past is that arms-racing can go hand in hand with arms control. Sometimes, the latter abets the former: During the 1970s, Washington used the Anti-Ballistic Missile Treaty to slow the defensive arms race until the United States had recovered from the Vietnam War and was better prepared to sprint ahead. And the former can also lead to the latter, as Reagan’s experience in the 1980s showed.

Arms control is still a good idea: The extension of New START in 2021 made sense from an arms-racing perspective because Moscow is better positioned to build up its strategic nuclear forces in the near term, even if it will struggle to outpace an economically superior United States over time. And building up to build down is still the right formula. Trilateral agreements to limit intermediate-range missiles, strategic nuclear forces, or the potentially destabilizing applications of AI and other new technologies may eventually become possible—but it will very likely require the United States to demonstrate first that an unconstrained arms race will leave its rivals poorer and more vulnerable in the end.


“The term arms race,” Gray wrote in Foreign Policy’s Winter 1972-73 issue, “suggests hostility, danger and high taxes.” Yet running an arms race may be necessary to avoid uglier outcomes, such as defeat in war or the gradual loss of influence that results from military inferiority. And the rewards of arms-racing can be substantial if an intelligent strategy forces a revisionist adversary to adjust its approach—and perhaps even reconsider its long-term objectives. High-stakes military competitions are already raging today, and the United States badly needs to shape them. An arms race is only futile if you lose.

Hal Brands is the Henry A. Kissinger distinguished professor of global affairs at the Johns Hopkins School of Advanced International Studies, a member of the U.S. State Department’s Foreign Affairs Policy Board, and the co-author, with Michael Beckley, of Danger Zone: The Coming Conflict with China. Twitter: @HalBrands

Join the Conversation

Commenting on this and other recent articles is just one benefit of a Foreign Policy subscription.

Already a subscriber? .

Join the Conversation

Join the conversation on this and other recent Foreign Policy articles when you subscribe now.

Not your account?

Join the Conversation

Please follow our comment guidelines, stay on topic, and be civil, courteous, and respectful of others’ beliefs. Comments are closed automatically seven days after articles are published.

You are commenting as .

More from Foreign Policy

A propaganda poster from the 1960s shows Chinese leader Mao Zedong.
A propaganda poster from the 1960s shows Chinese leader Mao Zedong.

Xi’s Great Leap Backward

Beijing is running out of recipes for its looming jobs crisis—and reviving Mao-era policies.

A textile worker at the Maxport factory in Hanoi on Sept. 21, 2021.
A textile worker at the Maxport factory in Hanoi on Sept. 21, 2021.

Companies Are Fleeing China for Friendlier Shores

“Friendshoring” is the new trend as geopolitics bites.

German children stand atop building rubble in Berlin in 1948.
German children stand atop building rubble in Berlin in 1948.

Why Superpower Crises Are a Good Thing

A new era of tensions will focus minds and break logjams, as Cold War history shows.

Vacationers sit on a beach in Greece.
Vacationers sit on a beach in Greece.

The Mediterranean as We Know It Is Vanishing

From Saint-Tropez to Amalfi, the region’s most attractive tourist destinations are also its most vulnerable.