A Misleading Metaphor: The Nuclear “Arms Race”
There is an emerging threat to the United States that will “endanger everyone,” one that can cause “escalation and misunderstandings” and even increase “the risk of a crisis or conflict that might turn nuclear.” No, these descriptors are not about Russia’s latest doomsday nuclear weapon or China’s provocative military behavior toward its neighbors. Instead, these are the purported consequences of a three-way nuclear “arms race” that some analysts believe the United States is about to ignite.
The Joseph R. Biden administration is currently considering whether and how to adapt the U.S. nuclear posture to China’s rapid nuclear buildup and Russia’s steady increase in its regional nonstrategic nuclear weapons. There is an emerging bipartisan consensus among longtime U.S. nuclear policy officials that the United States will need to adjust and potentially increase its nuclear forces in response to growing threats. Critics believe that such recommendations will increase nuclear dangers, and they frequently employ a metaphor to illustrate how: an “arms race.”
This “arms race” metaphor, however, does far more harm than good in explaining a poorly understood dynamic. It is a simplistic disfigurement of a complicated reality. If policymakers are led to believe that the United States is creating an endless action-reaction loop between itself and China and Russia, not only will members of Congress be unable to see the threat environment as it actually is, but they also may be afraid to make any necessary adjustments to U.S. forces to reinforce deterrence.
We believe that the debate about an arms race can be improved by recognizing four key realities. First, scholars cannot agree to a common definition of an “arms race,” and they acknowledge the metaphor is a poor descriptor. Second, the most informed Cold War studies on the U.S.-Soviet arms competition failed to find a tight action-reaction linkage. Third, there is abundant evidence today that the United States is not the primary driver of China’s and Russia’s nuclear buildups. Fourth, U.S. officials should do a better job of explaining the internal drivers for China’s and Russia’s nuclear procurement decisions.
Specifically, the U.S. government should commission a major study of Cold War and post–Cold War Chinese and Russian defense procurement to assess whether the United States — as it did with the Soviet Union — may be severely misjudging the intent and underestimating the capabilities of its adversaries. The United States can reduce the risk of making the same mistake again by reversing the Biden administration’s decision to remove the “hedging” role for nuclear weapons and preparing for the possibility that China’s leaders, much like their Soviet Cold War predecessors, will not be satisfied with nuclear parity.
The “Arms Race” — A Meaningless Metaphor
For as many times as analysts invoke the specter of an “arms race,” one would think there is a commonly understood definition; there is not. Scholars have written whole books without defining the term because it is so troublesome to differentiate from the regular retirement and modernization dynamic that every military undergoes. Indeed, a respected scholar of strategy, Colin Gray, looked back on 25 years of study by others in the field, including his own widely used 1971 definition of an “arms race,” and concluded that the term cannot be “defined usefully” or “shown to ‘work’ in ways and to ends distinct from other conditions of strategic rivalry.” In short, if one cannot clearly define how an “arms race” looks substantially different from how a state would act during just another day in the anarchic international environment, then there is little basis for establishing a causal link to political tensions, escalation, and war.
Of course, this is not to say there is no “action-reaction” between states. Proponents of the “action-reaction” theory only err when they claim this is the dominant factor for defense decisions. The theory posits that states will seek to improve their security, but even their precautionary investments in military capabilities will appear aggressive to their opponents, who in turn make their own precautionary investments, thus beginning a dangerous cycle. But defense procurement is not an endless tit-for-tat process. States arm themselves in manners and for reasons unique to their strategic culture. The most obvious example is China’s decades-long commitment to a “minimum deterrence” nuclear force posture in spite of the far larger U.S. and Soviet nuclear arsenals throughout the Cold War. Similarly, the size and makeup of the Soviet nuclear arsenal was largely driven by forces unique to its strategic culture — mainly, a weak political leadership exerting little control over its most powerful defense procurement organization, the Military Industrial Commission. As post–Cold War interviews with Soviet officials revealed, the “primary cause of the USSR’s arms buildup” was the commission’s efforts “to ensure stable weapons development and production processes.”
Beyond the definitional and conceptual difficulties, the metaphor of a “race” has particular connotations that have little relation to the real world, but that are easily exploited by activists: a sense of single-mindedness in a sprint against an enemy, no regard for cost, endless, and wasteful. This metaphor is pernicious because it assumes any adjustment in the U.S. nuclear force posture could be the spark for a sprint to extinction, but as U.S. nuclear procurement plans make clear, “sprint” is not the term that should come to mind. For example, the U.S. Air Force will spend over a decade simply developing, not producing, the new nuclear-armed cruise missile, while the U.S. Navy will have spent 25 years developing and producing the planned number of Columbia-class ballistic missile submarines when the final one deploys in the 2040s. The National Nuclear Security Administration will not be able to produce 80 plutonium pits — the cores of nuclear weapons — per year until the 2030s, a far cry from the thousands the United States could produce annually during the Cold War. A “race” this is not.
Studying the Arms Race — Lessons Going Unlearned
Setting aside the difficulty in defining an “arms race,” eminent scholars and practitioners have investigated the analytic utility of the “arms race” metaphor in describing procurement decisions. Indeed, as a point of departure, analysts should draw from three studies conducted during the Cold War — two highly classified and the other from open source material. All were motivated by growing concerns that the action-reaction model had contributed to deeply flawed U.S. strategic arms policy.
In the late 1960s, Fritz Ermarth and Thomas W. Wolfe conducted a classified RAND study of Soviet-American interactions. Completed in 1973, their careful assessment of historical trends revealed an episodic and sluggish relationship. Moreover, when direct reactions did materialize, the authors concluded that “the range and character of responses over time have been so broad and varied that no general theory or model of interaction covers them satisfactorily.”
The following year, Secretary of Defense James Schlesinger commissioned a history of the strategic arms competition, an exhaustive effort that dragged on until 1981. After slogging through a vast and highly classified database, the authors discovered that the interaction process was neither imitative nor tightly coupled. Responses were, in fact, more selective and far slower, shaped by peculiar strategic cultures, organizational preferences, and irreducible budgetary and technological constraints.
These studies confirmed Albert Wohlstetter’s contention, arrived at from publicly available data in 1974, that greater-than-expected estimates had not compelled American planners to dramatically overreact to Soviet initiatives. With Soviet missile building having blown past U.S. force levels, which had leveled off in part to encourage Soviet reciprocal restraint, Wohlstetter categorically rejected “talk of a ‘race’ between parties moving in quite different directions.”
The upshot: the Cold War rivals built distinctive nuclear arsenals for reasons that transcended the external threat environment. A suite of inputs, most notably clashing national deterrence practices, generated asymmetric strategic forces. In the 1970s, the Soviet Union augmented its dense strategic air defense with an array of massive land-based missiles — a comprehensive posture for warfighting and regime survival. The United States, on the other hand, had drawn down its defenses and depended on markedly smaller land and seaborne missiles.
Though Soviet planners eventually accepted nuclear deterrence, its form departed sharply from American theories of strategic stability and mutual vulnerability. The Soviets considered stability a one-way street; that is, a function of Soviet strategic superiority that flowed from enormous war-survival and war-fighting programs. Reality, then, conformed to neither the image of superpowers “jogging in tandem on a treadmill to nowhere,” nor arms controllers’ enduring hope that moderation could induce Soviet cooperation.
Concerns that the Soviets would exploit their missiles to enter the global power-projection business motivated the above analytic histories. Unwarranted fear of arms racing, these analysts contended, would tranquilize American efforts to stabilize a shifting nuclear balance. As Defense Secretary Harold Brown, who confronted the ramifications of unrequited restraint in the late 1970s, famously quipped: “When we build, they build. When we cut, they build.” And Schlesinger, himself a harsh critic of the action-reaction model, detested that an unwillingness to adequately hedge and modernize forces had saddled the United States with “a counterforce plan without counterforce weapons.” Both defense secretaries believed an accurate diagnosis of the interaction process was the first step toward resetting arms policy for long-term competition.
Indeed, both Schlesinger and Brown oversaw the U.S. policy of détente with the Soviet Union, a gradual improvement in bilateral political relations, which refutes another often-claimed but rarely demonstrated assertion: that arms races will lead to worsened political relations, and by extension, nuclear reductions will lead to improved relations. This is a simplistic framing. For instance, during the period of détente, the Soviet nuclear arsenal reportedly grew significantly, but bilateral relations still improved. Today, U.S. political relations with Russia and China are very poor, despite vastly reduced U.S. and Russian nuclear arsenals.
Improved political relations, a near-prerequisite for arms control agreements, can thus develop independently of a perceived “arms race.” Going one step further, even when political relations are poor and there are no negotiated agreements, the “arms race” does not proceed unchecked, at least among some states. As Cold War scholar William Van Cleave noted, “We should remind ourselves that in the democratic states of the West there is always [emphasis original] arms control, even without negotiated agreements. Arms are controlled and limited by the West’s traditional values, by its political and budgeting processes, and by the influence of the media and of public opinion.”
The enduring lessons of the Cold War and immediate post–Cold War therefore risk going unlearned: the composition and size of a state’s nuclear arsenal are generally not tightly coupled or imitative of its adversary’s, nor is the health of political relations determined by the changes in nuclear stockpile sizes. Nor is the presence or absence of binding arms control agreements a reliable gauge of danger.
Today Is No Different
Critics may contend that the Cold War dynamic is different than the one facing the United States today, but contemporary events still show little evidence of a tightly coupled action-reaction process — and certainly not one driven by the United States.
At face value, there is some degree of symmetry between the American and Russian strategic arsenals. Both have around 1,550 strategic nuclear weapons deployed on a triad of bombers, intercontinental-range missiles, and ballistic missile submarines. However, a closer look reveals distinct differences that deviate from what an “action-reaction” theory might predict.
Within the intercontinental-range nuclear forces, Russia deploys most of its weapons on intercontinental ballistic missiles, while the United States favors its ballistic missile submarine force as the “heaviest” leg of the triad. There are yet more stark differences between the nonstrategic nuclear arsenals of the two countries. While the United States reportedly has around 200 nuclear gravity bombs, Russia maintains around 2,000 nonstrategic nuclear weapons across dozens of different delivery systems. These nonstrategic weapons include more exotic systems such as nuclear mines, nuclear surface-to-air missiles, and even nuclear torpedoes.
While some of the diversification of the Russian nonstrategic nuclear arsenal has taken place in recent years, the asymmetry with the size of the American arsenal has persisted for decades. Indeed, the United States has been content letting the gap in size and diversification between the two arsenals persist since at least the end of the Cold War — an action leading to inaction, as it were. The United States has not followed Russia’s lead in its drive for nuclear-powered cruise missiles, nuclear-armed torpedoes of intercontinental range, and a host of other specific weapon types. The difference between the two states could not be more stark as demonstrated by the raging debate over a potential modest expansion of U.S. nonstrategic nuclear capabilities in the form of a nuclear-armed sea-launched cruise missile.
Looking to another Eurasian major power, China was long satisfied with maintaining a minimal nuclear deterrent for over half a century — not following the size or diversity of the U.S. or Soviet nuclear arsenals. Beijing’s long-standing minimal nuclear deterrent alone should be a powerful argument against “action-reaction” as the defining determinant of nuclear behavior.
In recent years, however, U.S. intelligence revealed that China is engaged upon a breathtaking nuclear expansion. This raises the obvious question: Why, after more than half a century of being satisfied with a minimal nuclear deterrent — even during the darkest periods of the Cold War — is China growing its arsenal? And why is it doing so during a time when the Biden administration seeks nuclear reductions with Russia and a multilateral arms control treaty with China?
Action-reaction theory would posit that China is building up from its relatively low nuclear force levels to at least parity with the United States, but the theory cannot explain why it is only doing so now. Since the fall of the Berlin Wall, the U.S. nuclear stockpile has shrunk in number by over 80 percent. The U.S. nuclear modernization program of record is mostly a one-for-one replacement of capabilities through the 2040s, and the Biden administration has shown no appetite for expanding U.S. homeland missile defenses. And yet, until recently, China maintained its relatively small nuclear force size — against the prediction of action-reaction theory. This argues for factors internal to Beijing. A sudden change in China’s nuclear requirements as directed by leader Xi Jinping explains the rapid and massive expansion of its nuclear forces far better than any incremental action — or, better stated, inaction — of the United States. At the very least, U.S. nuclear restraint has not led to Chinese restraint — in fact, it is a case of inaction-action.
If Not “Arms Race” — Then What?
As shown, “arms race” is not an accurate description of how states decide on the proper size and composition of their nuclear arsenals. While other actors’ arsenals are a factor in their decision calculus, they are far from being deterministic. Policymakers and activists should therefore stop using the term “arms race” and recognize that states make defense procurement decisions based on dozens of different factors. That said, the worry about increased numbers of nuclear arms leading to a greater likelihood of war will remain, for a variety of reasons, which must be examined and understood.
Think tanks can make some important open source contributions, but policymakers and defense officials would benefit from high-quality studies that include both classified and unclassified assessments, incorporating both qualitative and quantitative methods. The better U.S. officials and the American public understand why China and Russia act like they do, the better they can formulate U.S. defense policy.
To this end, we recommend the Office of Net Assessment, in conjunction with the Office of the Secretary of Defense Historical Office, commission a follow-on report to the 1981 “History of the Strategic Arms Competition” study referenced above. This new analysis should begin with the year 1973 (where the previous study of that same name stopped its analysis) and proceed to the present, utilizing the full range of classified and unclassified information. Like its Cold War forerunners, the study would diagnose long-term trends — perhaps revealing glaring asymmetries in what the great powers value most in their respective strategic arsenals. Ideally, the study would reveal behavioral predispositions that the United States can exploit for a better understanding of what China’s and Russia’s enduring strengths and weaknesses are.
It would also shed light on China’s and Russia’s approach to nuclear deterrence, and, most importantly, it would give the United States an idea of the underlying dynamics that shape nuclear competition and force posture. Indeed, given the stresses posed by a world in which the United States faces two peer nuclear actors (as will almost assuredly be the case in the 2030s), it is nearly impossible to have a long-term modernization plan that enables the United States to have a credible deterrent posture, based in part on a larger and more diverse strategic and nonstrategic nuclear arsenal, without having conducted such analysis. Unlike the classified Cold War studies, this study’s main conclusions should be unclassified and available to the public so that the factors that inform U.S. nuclear decision-making can be done in as transparent a manner as feasible.
U.S. officials should also add “hedging” back into the stated roles for U.S. nuclear weapons. The 2022 Nuclear Posture Review removed the “hedge” role from the U.S. nuclear arsenal to signal that the United States was reducing the role of nuclear weapons in its national security architecture. But, when the nuclear threat environment is becoming vastly more complicated, and mistakes in setting nuclear deterrence requirements more costly, “hedging” to overcome adverse events has become more important than ever — especially given the decades-long lead times associated with developing and fielding nuclear capabilities.
Finally, given the inaction-action nature of much of the U.S.-China nuclear relationship, U.S. officials should recalibrate their expectations about China’s nuclear breakout. Given the history of U.S. officials expecting the Soviet Union to stop building its nuclear forces when it reached nuclear parity (and being surprised they did not), it is better to seriously examine the implications for U.S. security if China’s intentions, like the Soviets’, are not mere parity. Given that China is not satisfied with parity in any other area, the United States should not assume that the nuclear dimension is different, given the obvious recent about-face in Chinese nuclear force sizing policy.
The concept of an “arms race” has no common definition and misleads as a metaphor meant to describe state behavior. Those who have studied the U.S.-Soviet nuclear arms competition during the Cold War, what proponents of the action-reaction theory hold up as the ultimate example of a senseless arms race, found little evidence of a tight linkage between the two defense establishments. Evidence post–Cold War, especially among the United States, Russia, and China, demonstrates further the action-reaction model has little predictive or analytical value. Instead of a mindless “arms race,” policymakers and the general public should think of U.S.-Russian-Chinese interactions as battles for advantage, with each side positioning itself according to its own unique values and capabilities — a process the U.S. government should seek to understand better through informed analysis.
The “arms race” is a simplistic metaphor that leads to an ahistorical conclusion: U.S. restraint can stop China’s and Russia’s nuclear expansion. This was true neither during the Cold War nor after. The United States should instead recognize that China and Russia may indeed react to U.S. actions, but not in the mechanical “action-reaction” way generally predicted by those who fear an “arms race.” Instead, each state has its own reasons for building its arsenal, and limiting U.S. programs based on the specter of an “arms race” is unwise at best and dangerous at worst.
Matthew R. Costlow is a senior analyst at the National Institute for Public Policy. He is a Ph.D. candidate at George Mason University and served as a special assistant in the Office of the Secretary of Defense, Nuclear and Missile Defense Policy.
Robert Peters is the Nuclear Deterrence and Missile Defense Fellow at the Heritage Foundation. He served in various positions within the Department of Defense over a 17-year career, to include in the Office of the Secretary of Defense, National Defense University, and the Defense Threat Reduction Agency.
Kyle Balzer is a Jeane Kirkpatrick fellow at the American Enterprise Institute. He was a postdoctoral fellow at the America in the World Consortium and received his Ph.D. in history from Ohio University.
The views expressed in this article are our own and do not represent those of any organization with which we are or have been affiliated.
Image: Photo by Lt. Jennifer Bowman