The independent resource on global security

Is an arms race just a race to the bottom?

Partha Gangopadhyay

An arms race is an interstate competition that motivates states to innovate, design and deploy the most lethal war technology in order to gain the upper hand against their rival states. However, arms races also create the looming danger of mutual destruction as an unintended by-product of both states striving to gain the upper hand in the battlefield.

Primarily, in older economic and political theories, arms races are viewed as an action–reaction process triggered, fuelled and shaped by real or perceived external threat. One state, fearing a second state as a threat, embarks on a military build-up. The rival state, observing the action of the first state, reacts by augmenting its military power, which in turn motivates the first state to increase its military power and thus the arms race starts.

This action–reaction framework is consistent with several prominent international relations models such as the security dilemma, the spiral model, and structural neo-realist theory. It is well-documented in the existing literature that military spending can pose a security dilemma, when a state chooses to retaliate to the military build-up of another state because it is unaware of the rival’s true intentions. The reciprocated increases in arming potentially engender a spiral of hostilities, increasing the chances for the outbreak of armed conflict.

In the latest works on arms races it is usually assumed that a sequence of states, or leaders of these states, each chooses in turn one of two options, A (to arm) or B (not to arm), with each state observing all of its predecessors’ choices. They have common preferences among the two choices but do not know which is better. Each state knows the costs and benefits of its own military build-up, but it is unsure of the costs and benefits of its rivals in the arms race.

In other words, the costs and benefits of arming often depend on the social, moral, political and psychological considerations of the leaders of states, which are often idiosyncratic and country- or leader-specific. In this scenario, leaders have an incentive to learn from social interactions about their rival’s true preferences, actions and intentions. There is an extensive literature in economics on the relevance of social interactions and social learning for economic behavior in a wide range of contexts. As an example, we now know that social interactions in neighborhoods can shape, influence and propel individual choices in important contexts ranging from education to crime. The literature has firmly established that social interactions can have a wide range of effects on the properties of the economic equilibrium: social interactions can lead either to conformity of behavior or to polarized actions. They can also cause multiplicity of equilibrium in cases in which equilibrium would otherwise be simple and unique, which in turn can create indeterminacy and make consequent outcomes inexplicable to modern economics.

The latest vintage of game-theoretic models of arms race has provided a complete formalization of the critical role of information revelation, transmission and pre-play communications to offer new insights into the dynamics of arms race. In a similar vein, the latest cohorts of international relations models highlight the role of social learning, information problems and information acquisition to explain the onset of arms race. From these valuable new works we now learn that there is nothing automatic, instantaneous and sacrosanct about arms race as there is positive probability that the détente equilibrium will prevail to stem costly and self-destructive arms race from occurring. Arms races will not go astray as the desire to arm will be bound by the leash of this détente equilibrium. In other words, apparently there is no ‘economic’ justification for arms races to race to the bottom.

This optimism can be terribly short-lived if one makes a necessary refinement of the recent literature on the informational problems of arms races by exploring rational inference of states in social-learning settings, which is popularly termed information cascades in economic theory. Communications and learning are important when the underlying problem is about incomplete information about rival states’ preferences and capacities for military build-ups. Even in fully rational models of arms race, information cascades can occur once the information contained in a group’s observed military build-ups becomes so great that an individual state’s private information can never ever affect its optimal military build-up.

One can redesign informational cascades as herds that can, in turn, cause mutual imitation by states that, in turn, leads them to a convergence of their decisions on an arms race. Arms races can thus arise in a fully rational world as the end result of a copycat strategy. Herds in arms races fail to eventually aggregate information if, and only if, an information cascade begins before the truth about one’s rival state is revealed. This can occur in the long-run if the environment is significantly ‘coarse’.

The obvious implication is that the military build-up is socially inefficient since, despite a plethora of private signals that reveal the right choice or action, states chase each other on the incorrect and costly choice on arming with positive probability. The détente equilibrium will then fail to materialize and all hell will break loose. One can then establish that the decision to get armed to the teeth and subsequently embroiled in a war and other violent conflicts can derive from incorrect choices based on whims, fads and distrust: every state is individually rational to arm and fight a bloody war, even if all these states have overwhelming (private) information that there is absolutely no need to do so.

In other words, everybody will willingly choose the ‘wrong behaviour’ even when there is enough individual evidence in favour of choosing the ‘right behaviour’ of disarming. This is what one can term the (arms) race to the bottom. A cursory glance at the real world seems to suggest that arms races have started racing towards the bottom in the new millennium.

The annual military spending of the world today is about $1.7 trillion, which has exceeded cold war levels. The major producers and suppliers of conventional weapons are the five permanent members of the UN Security Council—the USA, Russia, France, China and the UK—and Germany. Together, these six states contribute nearly 80 percent of reported global major conventional arms exports. The major buyers of conventional arms exports are a handful of developing countries struggling with their developmental aspirations, deep poverty, internal violence and potentials for cross-border conflicts: ten developing countries absorb 61 percent of the arms exported to the developing world. In 2011, according to the Grimmett Report of 2012, the three major markets for conventional weapons in the developing world were Saudi Arabia (21 per cent), India (13 per cent) and the UAE (6 per cent) while many other states have continually upgraded their military capability. Both China and Pakistan are major absorbers of conventional weapons, yet the data from China and Pakistan on arms spending and their bilateral trade in arms are too unreliable to make a reasonable analysis.

Since the permanent members of the UN Security Council dominate the global arms trade, they have few incentives to introduce and enforce strict regulations and controls in the global arms market and, as a result, the global arms trade and arms race continue unabated in a legal and moral vacuum. The Arms Trade Treaty (ATT) is an interesting recent phenomenon that seeks to regulate cross-border arms trade in order to keep weapons out of the hands of human rights abusers, notorious dictators, known terrorist groups and criminal syndicates. The ATT aims to set standards for all cross-border transfers of conventional weapons ranging from small firearms to tanks and attack helicopters. However, there are two serious features in the treaty that will possibly make it vacuous. First, once the arms shipments reach a foreign destination, the foreign government still has the full control over the distribution and possible abuses of the arms within its borders. Second, for the U.S., the largest player in the global arms market, there is little change in the game as the practical implications of the treaty for U.S. arms manufacturers are expected to be null since, as John Kerry has correctly noted, the United States already has in place the kind of strict export controls for weapons that are outlined in the treaty. Note that the US arms manufacturers control more than 80 per cent of the cross-border trade in weapons that are covered by this specific treaty.

In the absence of meaningful regulation the global arms market, akin to anarchy, suffers from widespread corruption, bribery and kickbacks in the midst of which the top three armament firms usually share immense market spoils, for example, a whopping sale of arms worth almost $98 billion, roughly 10 per cent profits for the largest three firms and possible profits of $50 billion for the entire supply chain in 2012, from producing machines for aggravating human miseries.

By the end of the cold war, from various research works published in the 1990s, we now know how armament firms have regularly spread false rumours about the military and naval programs of various states, engaged in scaremongering, played one country off against another, influenced public opinion on armament through control of media and formed powerful arms cartels to promote a global arms race.

In other words, while money talks, big money talks so much louder that it forces many states to home in on an arms race even when they are fully convinced that there is absolutely no need to do so. It is often argued that the arms race has deepened the cycle of violence, oiled terrorism and increased human rights violation mostly in developing states.


This blog post is published as part of a collaborative partnership between SIPRI and Economists for Peace and Security (EPS).