Sweeter Carrots and Harder Sticks: Rethinking U.S. Security Assistance


Daily news coverage of burned-out Russian armor, as well as Russia’s eastward retreat, serves as tangible evidence that U.S. weapons and other military supplies being delivered to Ukraine — about $3.5 billion authorized so far this year — are having real effects. Less clear, however, are the results of the roughly $20 billion in military aid which the United States provides to other partners around the world annually. When evaluated in terms of U.S. national interests and partner-nation outcomes, a great deal of this security assistance fails to meet the mark.

History is marred with episodes such as the initial failure of U.S.-trained and equipped Iraqi forces against ISIL in 2014, the collapse of Afghan forces following coalition withdrawal in 2021, and others in the more distant past. As E. John Teichert recently argued, the U.S. security assistance enterprise is slow-moving, outmoded, and requires a major overhaul if the United States wants to remain the partner of choice vis-à-vis strategic competitors such as Russia and China. He is correct, but only addresses half of the problem. Currently, the United States uses an approach which prioritizes what are known as “capacity-building efforts” — mainly the unconditional provision of equipment and training — but fails to address underlying institutional weaknesses. Instead, the United States should use aid as leverage to achieve both reform in partner nations’ defense establishments and broader U.S. policy goals.

The What, Why, and How of Security Assistance

“Security assistance” refers to a specific set of programs authorized by the Foreign Assistance Act and Arms Export Control Act. These programs are overseen by the State Department in cooperation with the Department of Defense. “Security cooperation” describes separately authorized Defense Department-led activities such as “global train and equip” programs. The Ukraine Security Assistance Initiative, which has provided military aid to that country since 2016, is one such example of these programs run by the Pentagon. I will use the more common term here, security assistance, in reference to either type. When security assistance works well, it gives partner nations the tools to address internal instability and deter and defend against external adversaries, reducing the likelihood that direct U.S. intervention will be called for in the future. It also helps to ensure that the United States maintains access, basing, and overflight privileges, strengthens interoperability, and accrues the less tangible benefit of military-to-military personal relationships.





Though Congress authorizes security assistance programs each year, it provides little official direction as to how they ought to be carried out. The National Defense Authorization Act and other laws, though not perfectly enforced, impose some constraints. These include prohibitions on assisting human rights violators and various reporting requirements. Still, agencies are largely left alone to figure out the details of executing security assistance. In practice, security assistance programs are designed through a give-and-take process between the host nation, the U.S. embassy country team, combatant commands, and the Departments of State and Defense in Washington. On the leading edge are military officers assigned to embassy security cooperation offices around the world who interact with host-nation counterparts every day. Program implementation falls mainly to the Defense Security Cooperation Agency and a bevy of stateside offices that carry out the minutiae of program management, logistics, and contracting.

Security Assistance as Incentive

Security assistance generates a classic principal-agent problem. In such relationships between a principal and its agent or proxy, there exists an information asymmetry where the agent inherently has more information about their own capabilities, activities, and interests than the principal. Agents, being rational actors, place their own interests above those of the principal and, in the absence of other incentives, behave accordingly. The greater the degree of misalignment, the more problematic this becomes. U.S. frustration during its 20-year war in Afghanistan over Pakistan’s perceived failure to adequately confront extremist groups, despite significant military aid, is an example of principal-agent problems at work.

Fortunately, these are not insurmountable obstacles. Security assistance in the form of arms, equipment, or other support can be an effective incentive when it is used as a carrot to reward desired behavior, such as achieving a specific reform or taking some other action aligned with the interests of the principal. Alternately, such aid becomes a stick when it is withheld. An example of this carrot-and-stick approach being used effectively is described in Walter Ladwig’s study of U.S. assistance to the Philippines to combat the Hukbalahap rebellion in the decade following World War II. Heavy-handed tactics used by the local constabulary forces were a major factor driving support for the rebels. Recognizing this problem, U.S. advisors insisted on a series of reforms within the security forces as a precondition for further aid. These measures, along with the appointment of the reform-minded Ramon Magsaysay as defense chief, contributed to a far more effective counterinsurgency campaign and the eventual defeat of the rebel movement.

This approach of using aid as an incentive stands in sharp contrast to Ladwig’s other case study, Vietnam, where U.S. aid was used as an inducement in the hope that it would eventually change partner behavior. In this case, U.S. advisors tried in vain year after year to cajole prime minister Ngo Dinh Diem and successor regimes to implement reforms that would have broadened domestic political support and made the South Vietnamese forces more effective. All the while, massive U.S. assistance continued to flow in with the expectation that it would eventually bring about a change of heart in the host-nation government. Such change never came, and the United States and its South Vietnamese proxies lost the war.

The more recent U.S. experience in Iraq from 2003 until the withdrawal in 2011 contains echoes of Vietnam. David Lake’s study of that conflict suggests that provision of unconditional security assistance may still be America’s default setting. Despite massive volumes of military aid provided to the Iraqi government during this period, U.S. urging to reform the security forces and national government repeatedly failed. Nouri al-Maliki, prime minister during much of this period, resisted efforts to make the armed forces more broadly inclusive and apolitical, instead continuing to employ them as his own sectarian instrument. Again, the United States used an approach focused on training and equipping, known in the current lexicon as building partner capacity. The Iraqi government received assistance regardless of its effort, or lack thereof, to implement much-needed reform while the United States failed to use the leverage which hundreds of millions of dollars of assistance could buy. Several years later, the Iraqi military in which the United States had invested so much fell apart virtually upon first contact with the Islamic State’s irregular militia.

The “building partner capacity” approach has two main flaws. First, it ignores fundamental problems, such as corruption, coup-proofing, and otherwise weak defense institutions which additional military capacity cannot overcome and may even exacerbate. Paradoxically, a focus primarily on capacity-building often fails to build meaningful, long-term capacity because it neglects underlying institutional problems. Second, it ignores principal-agent problems by assuming that once in possession of highly capable security forces, the partner will wield them in a manner aligned with U.S. interests. U.S. advisors and diplomats may recognize the pressing need for security sector reform in the partner. Unfortunately, the ingrained culture of U.S. security assistance seemingly relies on the false premise that with enough U.S. training and relationship-building, other countries’ militaries will voluntarily reform even though it is often not in their personal interests to do so.

Hard Is Not Impossible

These failings notwithstanding, there exists no shortage of arguments in favor of the status quo and against shifting America’s focus in security assistance to security-sector reform and conditional aid. In an era of great-power competition, Russia, China, or regional competitors may simply step in to fill the gap with the partner nation if the United States insists upon unwelcome reforms in exchange for aid. I know this is a valid concern because I have sat in meetings where partners made thinly veiled reference to other nations which may provide assistance if the United States is unable or unwilling. Still, the quality and technological edge of U.S. kit and the total-package approach which includes training and follow-on support do provide the United States a competitive advantage. But the bureaucratic apparatus responsible for delivering aid is far too slow and, as a result, is a liability. To outcompete, the security assistance enterprise itself requires major reform to shorten typical lead times from years to months. The Pentagon and State Department need the agility to turn aid on and off like a switch or rheostat.

Another argument against placing preconditions on security assistance is that the stakes are too high to deny a partner a critical capability while waiting for reform to happen. This is the “we can’t let them fail” mentality. It is also valid — but only in certain crisis situations where aid is needed immediately, and reform truly cannot wait. Ukraine today is a good example. Rather than ceasing aid writ large until a condition is met, place conditions on discrete packages of aid. The intent should be to provide sufficient incentive to the host-nation government to modify behavior, not threaten national survival. During the years I spent in Afghanistan, the security threats faced by the government never appeared to afford the breathing space to withhold aid while institutional reform took place. In hindsight, doing so may have required coalition forces to carry a greater operational burden while the Afghan forces sorted themselves out, but it also would have been worth it. As the events of August 2021 demonstrated, 20 years of unconditional capacity-building built shockingly little capacity. Implementing this change requires a recurring review of Defense Department-led train and equip programs to determine which truly qualify as crisis situations. Those which do not should be moved under State Department control for more deliberate, reform-focused approach.

There is also the simple argument that instituting meaningful reform in a recalcitrant partner nation is hard. This is true. But hard is not impossible. The security assistance enterprise should approach reform not in broad, sweeping terms but incremental steps. Rather than requiring the partner to eliminate corruption wholesale across the security forces, perhaps require that a particularly corrupt commander be removed before providing assistance to that unit. In the case of Iraq in the late 2000s, the United States might have required Prime Minister Nouri al-Maliki to fire certain commanders guilty of sectarian excesses or modify the personnel policy to make the armed forces more broadly inclusive. In other cases, reform measures may be as mundane as switching to an electronic pay system to reduce the opportunity for graft. Such a change in approach would also require a change in how the United States prepares officers for security assistance duties. The Defense Security Cooperation Agency has made great strides in its educational program to professionalize the security assistance workforce in recent years. It could further improve this program by embedding security sector reform more deeply into the curriculum.

Better Rewards, Tougher Inducements

The United States is not getting the most for its security assistance dollars. Programs intended to build partner capacity tend to underperform because they fail to address underlying weakness in defense institutions or take principal-agent problems into account. To build meaningful, long-lasting capacity in partners and better align security assistance with policy goals, the United States should shift from a “building partner capacity” approach to a reform focus using conditional aid as leverage. Doing so in an environment where security assistance is yet another arena for strategic competition with Russia, China, and others will require internal reform of the U.S. security assistance apparatus. Rather than a slow-moving bureaucracy, the United States needs an agile and streamlined security assistance enterprise able to deliver aid rapidly when needed and be paused or turned off just as easily when conditions require. In short, the United States needs sweeter carrots and harder sticks.

While the current U.S. effort to arm and equip Ukrainian forces is the most visible example today, it is not representative of the day-to-day work of security assistance which the United States carries out around the world. In most cases, the greater threat facing these partners is not an invading army but rather defense institutions for which reform is needed but not properly incentivized. Modifying the U.S. approach to security assistance is one way to create that incentive and better align aid with policy goals.





Jeremy Gwinn is a U.S. Army infantry officer currently in the Office of Security Cooperation – Iraq. His previous experiences include multiple tours in Afghanistan and Iraq conducting counter-insurgency and security force assistance as well as teaching economics in the Department of Social Sciences at West Point. Gwinn has a Ph.D. in international relations from the Fletcher School of Law and Diplomacy where his doctoral research focused on U.S. unconventional warfare post-9/11. He has published on a variety of security affairs topics. The views expressed herein are his own and do not reflect those of the Army, the Department of Defense, or the U.S. government.

Image: Army Sgt. Jennifer Shick