A New Framework for Assessing the Risks from U.S. Arms Sales


In the past two years, Congress has tried (and failed) twice to halt American arms sales to Saudi Arabia in response to that country’s intervention in Yemen’s civil war. This level of concern is historically unusual. Arms sales rarely spur much debate in Washington, where they are viewed as a critical tool of American foreign policy. The traditional refrain holds that arms sales promise leverage over recipient countries, help the United States support allies and manage regional balances of power, and generate economic benefits to boot. With some exceptions, few have challenged the wisdom of American arms sales practices.

In a recent study for the Cato Institute, however, we argue that the government’s approach to arms sales is misguided. The United States accepts as given the potential benefits of selling weapons while underestimating or simply ignoring the potential risks. The result has been too many arms sales to too many countries where the risks are likely to outweigh the benefits. Between 2002 and 2016, America delivered $197 billion worth of major conventional weapons, equipment, and training through its Foreign Military Sales program to 167 states worldwide. It is difficult to imagine what sort of process would rate so many of the world’s roughly 200 countries as safe bets to receive American weapons. Indeed, using a “risk index” we created to assess U.S. arms sales, we found that in this time period, the average dollar value of U.S. arms sales per nation to the riskiest states was higher than to the least risky states. Even more disturbing was our finding that 32 of the 167 recipients had risk index scores higher than the average score of the 16 nations currently banned from purchasing American weapons.

For the United States to make more responsible use of arms sales, the approval process needs to change. And though our initial study focused on arms sales, the logic is the same for arms transfers (where the United States provides weapons to states or groups at no cost). There are often compelling reasons to consider providing weapons even (and sometimes especially) to risky clients, but the United States should account more carefully for both the benefits and the costs. The easiest place to start is cases of sales and transfers to nations engaged in conflict, fragile states, or states with poor human rights records, as well as in cases that do not directly enhance American national security. In these cases, the approval process should be more transparent, the bar for approval should be higher, and the government should do more to monitor weapons after they are sold to better understand unintended consequences that may blunt the benefits of arms sales and undermine U.S. security.

The Arms Sales Approval Process

On paper, the United States appears to have a robust procedure in place to assess arms sales. Since 1976, the Arms Export Control Act has required that the executive branch conduct a risk assessment for large government-to-government sales of major conventional weapons (commercial sales of many small and light weapons are covered by different rules). Once a foreign government decides it would like to buy an American system, it submits a letter of request. The request kicks off an extensive process that runs through a variety of offices at the Departments of Defense and State, as well as other agencies. The Country Team Assessment is the American government’s official risk assessment and is intended to determine:

how the [weapon] will be used, how it contributes to the defense and security goals of the partner nation and of the US, how it will change the partner country’s military capabilities, how the partner country will protect and safeguard sensitive technology, and the partner nation’s human rights record.

Once approved, the matter is turned over to Congress, which serves as an emergency brake for this process. Absent sufficient congressional opposition in the form of a veto-proof resolution of disapproval, the sale is made.

In reality, the outcome of this process is almost inevitably the same: approval. Though the United States won’t sell its latest technology to everyone, it will sell most things to just about anyone. Although a full explanation of this is beyond the scope of our work, three possible reasons are worth noting. First, the benefits of arms sales are obvious and immediate, while the negative consequences are often less obvious, tend to emerge much later, and often receive little media coverage. Second, there is no constituency in Washington opposing arms sales. Presidents see them as a foreign policy tool, Congress sees them as economically beneficial benefiting its constituents economically, and the defense industry provides financial encouragement all around through campaign donations. Finally, the United States has been the world’s leading arms exporter for so long that the presumption that arms sales work seems to have become ingrained in the national security bureaucracy.

The Risks of Arms Sales                       

Arms sales are attractive for many reasons. They offer a low-cost, flexible way to help allies and partners wage wars, deter adversaries, and fight terrorism. Moreover, arms sales and transfers certainly pose less political danger for presidents than sending American troops into harm’s way. But just because the United States has good reasons to sell weapons in a specific instance does not mean the benefits are a sure thing – nor does it ensure that the benefits will outweigh the costs.

The negative side effects of arms sales take many forms. One extreme example is blowback — when Americans weapons end up being used against American interests. After the Iranian Revolution in 1979, the revolutionary government took possession of billions of dollars’ worth of American fighter jets and other weapons, an arsenal that Iran has used to exert itself ever since. A more common example is when American troops end up fighting other forces armed with American-made weapons that the United States had willingly provided, as happened in Somalia in 1991 with weapons provided during the Cold War.

Arms sales and transfers can also harm the regions into which American weapons flow. Another danger is dispersion — when weapons sold to a foreign government end up in the hands of criminal groups or adversaries. This risk is highest with sales or transfers to fragile states that are unprepared, unwilling, or too corrupt to protect their stockpiles adequately. For instance, despite America’s efforts to train and equip the Iraqi army, ISIL fighters in 2014 captured three Iraqi army divisions’ worth of American equipment, including tanks, armored vehicles, and infantry weapons.

American arms sales can also prolong and intensify interstate conflicts. Although the goal might be to alter the military balance of a conflict to facilitate a speedy end, sending weapons can also encourage the recipients to continue fighting even with no chance of success, leading to more casualties.

Finally, U.S. weapons sales in the name of battling terrorism and insurgency undermine U.S. national security when they are made to corrupt regimes and to nations with a history of human right violations. American firepower can enhance regime security and enable oppressive governments to mistreat minority groups and wage inhumane actions against insurgents or terrorist groups. Currently, Saudi Arabia is waging war in Yemen using primarily American weapons, which the United States has continued to provide even though the Saudis have been cited repeatedly for human rights violations and targeting civilian populations. In countries where serious corruption is endemic, American weapons can be diverted from their intended recipients and wind up in the wrong hands. For example, as a result of military and police corruption, the small arms and light weapons that the United States sends to Mexico and to several other Latin American countries in support of the war on drugs often facilitate the very crimes they were meant to stop.

Assessing the Risk                                

To provide a better accounting of the risks of arms sales and transfers, we developed a risk index based on five indicators that previous research suggests correlate strongly with negative outcomes. We know of no previous efforts to categorize recipients with respect to the risks of negative outcomes. Indeed, there is very little historical data available on the outcomes, positive or negative, of arms sales: There is no database to tell us how frequent various negative outcomes are, which makes it difficult to identify the causes of those outcomes with much precision. As a result, the risk index remains a work in progress.

For each of the 167 recipients of American weapons between 2002 and 2016, we compiled the scores from five existing metrics:

  • The Fragile State Index. Fragile governments are at greater risk for having weapons stolen or sold to third parties and are also more likely to collapse, raising the possibility that a friendly customer could turn into a national security threat.
  • The Freedom House Index and the S. State Department’s Political Terror Scale. Measures of oppression and political terrorism are a proxy for the risk that a recipient will misuse the weapons against its own people.
  • The Global Terrorism Index and UCDP/PRIO Armed Conflict Dataset. Civil and interstate conflicts create the conditions for all sorts of risks, including weapons dispersion, the misuse of weapons against a state’s own population, the amplification and intensification of conflict, and blowback.

To create a single risk index from these sources, all of which use different scales, we sorted the nations into three categories for each risk factor: 1 (low risk), 2 (medium risk), or 3 (high risk). This was straightforward for two of the measures. The Freedom House Index codes nations as “Free,” “Partly Free,” and “Not Free,” while the UCDP/Prio dataset codes nations as being involved in “High Intensity” conflict, “Low Intensity” conflict, or not involved in conflict. In each case, our categorization corresponded directly with the three existing categories.

The State Department’s political terror scale has five levels, from the first level where state terror and torture are rare or extremely rare, to levels two and three, in which police brutality and political imprisonment occurs but is limited, to levels four and five, in which gross rights violations, including torture and murder, are common. The states at levels one and two, which are the least likely to use force against their own citizens, received a score of 1 – least risky. Those states at levels four and five are the most likely to use force against their own citizens, so we coded them as 3 – most risky. The group of states in the middle received a score of 2 – medium risk.

The Global Terrorism Index scores states on a zero to 10 scale. We coded states that scored between 0 and 2 (no to low impact from terrorism) as least risky. States that scored between 2 and 6 we rated as medium risk, and states that scored between 6 and 10 we ranked as highest risk. Finally, the Fragile State Index runs from roughly 18 to 113. We divided the index into thirds, rating the most fragile third as most risky and the least fragile states the least risky.

To get each nation’s risk index score, we simply added its scores for each of the five risk factors. As Table 1 shows, the resulting index ranges between 5 and 15. If a country scored “low risk” on all five individual measures its overall risk index score was 5, while a country that scored “high risk” on all five earned a score of 15. The average score across the 167 nations was 8.2; the standard deviation was 2.9.

The result is an index that passes the common-sense test and provides a starting point for policymakers to rethink the risk assessment process, although it unavoidably sacrifices the nuance of the individual components. Because our approach measures only significant differences in state freedom, stability, and so forth rather than attempting to claim undue precision, there is good reason to believe that nations scoring higher on this index are indeed riskier customers even though we cannot be certain about the precise weighting of different components (an area for future research).

Figure 1. Cato Arms Sales Risk Index 2018

Table 1. Arms Sales Risk Index

Risk Category Risk Index Score Average U.S. Arms Sales to Recipients 2002-2016 ($ billions) # of Nations Examples
Highest Risk 15 1.8 5


Iraq, Yemen, Sudan
Very Risky 9-14 14 64 Afghanistan, Egypt, Philippines, Saudi Arabia
Somewhat Risky 6-8 1.1 60


Brazil, Cambodia, UAE
Lowest Risk 5 1.0 38


Netherlands, New Zealand


Three critical observations emerge from this exercise. First, there are a lot of risky customers in the world, and the United States sells weapons to most of them. Thirty-seven nations (21.8 percent) scored in the highest risk category on at least two metrics and 77 (45.3 percent) were in the highest risk category on at least one of the five measures. There are simply not many safe bets when it comes to the arms trade.

Second, as simple as it is, our risk index does correlate with negative consequences. The five countries (Libya, Iraq, Yemen, the Democratic Republic of Congo, and Sudan) that scored as high-risk on all five measures are places where the negative consequences have been worst. American weapons have been stolen, misused, and used against American interests in each of these places since 2002. The very risky category also illustrates the full range of unintended consequences from arms sales. Afghanistan, Egypt, Saudi Arabia, Somalia, and Ukraine fall into this category.

Finally, the analysis provides compelling evidence that despite the warning flags, the United States does not discriminate between high- and low-risk customers. The average sales to the riskiest nations are higher than those to the least risky nations. The 24 countries labeled “highest risk” on the Global Terrorism Index, for example, bought an average of $2.2 billion worth of American weapons during the time period we analyzed. The 31 countries in active, high-level conflicts bought an average of $3.1 billion. This strongly suggests that the most obvious risk factors play little or no role at all in limiting U.S. arms sales.

Applying the risk assessment framework to the list of embargoed nations only sharpens this point. The 16 nations currently banned from buying American weapons had an average score of 11.6. Many of the nations on the list are adversaries (Iran, North Korea) but others, like Sudan and Eritrea, were only banned after egregious levels of insurgency, terrorism, and human rights abuses emerged. Although it is a sign of progress that these countries were eventually placed on an embargo list, the United States still sold weapons to them when most of the risks were already apparent. Moreover, America’s current customer list includes 32 countries like Egypt, Nigeria, and Pakistan, with a risk index score above the average of those who are banned.

Toward More Responsible Arms Sales

To be clear, our index is not intended as a standalone tool for deciding where and when to provide weapons to other nations. For each individual sale, the potential risks should be considered in light of the potential benefits. And in some cases, arms sales advocates can rightly argue that our risk index is a measure of which states need the most help. For example, fragile states like Iraq and Afghanistan score high on our index, but no one doubts they need assistance to ensure their security. However, a debate can and should be had about whether providing such states with weapons will achieve the intended goals. Our goal is not to suggest that there aren’t good reasons for providing weapons, but merely to help decision-makers do a better job considering the risks involved so that when the United States does sell weapons abroad, it has done a full accounting of the potential outcomes.

Our research suggests the United States could improve its arms sales policy in three important ways. First, the approval process must become more transparent. Neither Congress nor the public has much idea what benefits and risks were considered before an approval was made. All the public gets are stock phrases like “This proposed sale will contribute to the foreign policy and national security objectives of the United States,” and “The proposed sales of equipment and support will not alter the basic military balance in the region.” The executive branch should make the proposed benefits and potential risks clear, so Congress and the public can judge the justification of each sale and its outcomes accordingly.

Second, the government should raise the bar for approving arms sales. The United States enjoys a highly favorable strategic environment and an unmatched (though not unlimited) ability to influence events abroad. Focusing more on avoiding the negative consequence of arms sales than in the past would not significantly impair that global influence. One simple step in the right direction would be to stop selling weapons to states with clear warning signs: the most fragile states, those embroiled in civil wars, and those with the worst human rights records.

Third, the United States should implement a more comprehensive system of end use monitoring that tracks not only the physical location of American weapons (as is done today), but also attempts to account for how those weapons are used over the several decades during which they will be in service. As much of this data as possible should be made publicly available. Much of the difficulty in weighing risks and benefits is a product of the simple lack of data on what happens over the long-term when America sends weapons abroad. Having more information about both the positive and negative outcomes will help policymakers make more informed decisions about arms sales in the future.

The Saudi air campaign in Yemen, enabled by American weapons and American support, makes it clear that selling weapons is an inherently dangerous business. Even when the United States has important reasons for doing so the risks are real. A more prudent approach that denies sales to the riskiest clients would help the United States minimize its unintended consequences. Limiting arms sales, especially to countries engaged in conflict, would also give the United States greater diplomatic flexibility. It is difficult to be a credible mediator while arming one or both sides of a conflict. By taking this step, the United States would send a powerful signal to the international community about the dangers of unrestricted arms sales and America’s intention to be part of the solution to violence rather than an enabler of it.


A. Trevor Thrall is an associate professor at George Mason University’s Schar School of Policy and Government and a senior fellow at the Cato Institute’s Defense and Foreign Policy Department. Caroline Dorminey is a policy analyst at the Cato Institute’s Defense and Foreign Policy Department.

Image: marines.mil via Wikimedia Commons