“Money as a Weapons System”: The Promises and Pitfalls of Foreign Defense Contracting

This paper argues that this strategic spending comes with risks that could erase benefits, create new problems, and complicate achieving desired goals. The first risk is that injecting money into local economies could harm recipients: too much spending can skew markets and create unsustainable dependencies. The second risk is to U.S. interests: the U.S. government is often unable to control where the money goes and with whom it does business, which can lead to political blowback and security risks. The third risk lies in hidden costs: contract spending is not necessarily cheaper than other types of inducements (e.g., arms or aid), and it is sensitive to bidding wars with competitors such as China, which also wields government-directed spending as a policy tool.

The U.S. government needs to better assess the costs of doing this sort of business. The Department of Defense should improve data collection and reporting on the economic and political effects of preferential procurement policies. For its part, Congress should require impact evaluations as a condition for continued authorization of these policies and improve oversight on vendor vetting requirements. Finally, the U.S. government must examine the foreign policy goals driving the use of preferential procurement. No foreign policy tool is perfect, and in some cases spending might be the best option, but doubling down on spending in pursuit of flawed goals is a sure way to throw good money after bad.

In 2017, Congress authorized the U.S. Department of Defense (DOD) to bypass open competition in awarding contracts for goods and services to firms in any country in Africa that “has signed a long-term agreement with the United States related to the basing or operational needs of the United States Armed Forces.”¹ While this represents, to our knowledge, the most broadly scoped preferential procurement policy in U.S. history, it was not the first. Over the past 20 years, government procurement has become an increasingly popular tool for obtaining basing rights and military access, fighting insurgencies, and shoring up weak states and fragile economies. The U.S. government—mostly the DOD—spends billions every year purchasing goods and services from foreign countries.

Over $1 billion a year in overseas defense spending is what we call “strategic,” meaning that it is designed to both support global military operations and target specific firms in pursuit of particular foreign policy objectives. This strategic spending typically takes the form of contracts that are awarded with limited competition through a congressional statute or an international agreement between the United States and a recipient country. The 2017 preferential procurement policy for Africa illustrates limited-competition procurement: “(1) competition is limited to products or services from the host nation; (2) a preference is provided for products or services from the host nation; or (3) a preference is provided for products or services from a covered African country, other than the host nation.”²

Since 2000, Congress has authorized preferential defense procurement policies on four different occasions: once to support the counterinsurgency efforts in Iraq and Afghanistan (2008) and three other times to secure access in Central Asia (2010), Djibouti (2015), and all of Africa (2017).³ The argument underlying these policies is that defense procurement for the goods and services that support a deployed military presence—from coffee deliveries to construction materials—creates jobs and opportunities for local firms, promoting market growth. In theory, this growth yields a host of positive effects, such as winning hearts and minds in counterinsurgency campaigns, rebuilding states through economic development, and increasing incentives for communities to host a U.S. military presence. Moreover, by contracting directly with foreign firms, the United States hopes to procure goods and services for military operations while avoiding entanglement with foreign governments, limiting the risk of human rights abuses, and ensuring that host communities, not regimes, reap economic benefits.

We have found no evidence to support these claims. Moreover, Washington has made little effort to collect data on the economic or political effects of such policies, making it difficult to evaluate whether they work as expected. This is troubling, because procurement spending carries serious risks of unintended and unwanted consequences. First, it is far from certain that strategic spending will always yield economic growth. The amount of money needed to achieve tangible effects can skew local markets and exacerbate economic dependencies, resulting in a host of negative social effects. Second, when engaging in this strategic spending the United States struggles to control who ultimately gets its business and how commercial firms fulfill contracts. The limited transparency of these exchanges has sometimes resulted in the DOD inadvertently funding the enemy. In some cases, attempts to source locally have led to contracting “fratricide,” in which multiple U.S. agencies inadvertently compete for the same goods and services, spiking prices and creating local shortages.⁴ Both occurred repeatedly in Iraq and Afghanistan. Third, using spending to influence recipients can draw the United States into bidding wars with competitors such as China, whose control over state industries and unscrupulous global economic policies allow it to manipulate firms’ hiring and purchasing practices.

This paper presents data on overall trends in U.S. defense spending by country, region, and year between 2000 and 2015. It then discusses the major foreign policy objectives that the United States uses spending to pursue, comparing spending to other inducements such as foreign aid and arms sales. The paper also outlines the potential risks of using preferential procurement as a foreign policy tool. It concludes with policy recommendations for how Washington can exercise greater caution in its use of overseas procurement spending.

Government procurement is usually the largest single market of any economy, making it an attractive tool for states to use strategically both at home and abroad.⁵ For its part, the DOD is one of the largest buyers in the world. In 2015, for example, it spent upward of $15 billion for goods, services, and construction in foreign countries.

To get a sense of how much, where, and with whom the U.S. military spends overseas, we collected data on all U.S. overseas government contracts between 2000 and 2015. This gave us detailed information on nearly 2.4 million contract actions performed in 252 foreign countries and territories between 2000 and 2015.⁶ The data let us see which countries and regions have received the most spending as well as how country-level, regional, and global trends have changed over time. In each year between 2000 and 2015, the United States spent more than $9 billion overseas on defense-related procurement. In most of those years, spending surpassed $15 billion annually—particularly between 2004 and 2013 because of the wars in Iraq and Afghanistan, when the United States spent an average of over $30 billion each year on contracts for goods and services overseas.⁷ Trends in contract spending track broadly with overall U.S. defense spending, which rose each year between 2002 and 2010 and then fell between 2010 and 2015.⁸

Where is the U.S. military spending its money? Figure 1 shows regional trends in U.S. global defense spending, including a notable shift from Europe and toward the Middle East, Asia, and Africa between 2000 and 2015. The regional trends are consistent with America’s shifting overseas military commitments, while the general decline in funding from 2010 onward is consistent with lower deployments following the end of the surges in Iraq and Afghanistan. Following the 2004 global defense posture review, the United States began moving away from reliance on Cold War–style main operating bases in Europe and East Asia in favor of smaller forward operating sites and cooperative security locations (so-called lilypads) in Southwest Asia, Central Asia, and Africa. This shift resulted from both a diminished emphasis on deterring peer or near-peer competitors such as Russia and China, which made large, concentrated deployments less necessary, and the emerging threat posed by nonstate actors such as al Qaeda, which required more flexible deployments capable of rapidly moving into conflict areas.⁹ In light of the reorientation in U.S. foreign policy toward “great power competition,” this trend may reverse itself. Indeed, the Pentagon recently weighed cuts in the U.S. presence in Africa as part of this shift.¹⁰

Figure 2 shows the variation in total U.S. defense spending by country between 2000 and 2015. Unsurprisingly, the two largest recipients were Iraq ($86 billion) and Afghanistan ($81.6 billion). In addition, the United States spent a great deal in NATO member countries as well as in countries surrounding Iraq in the Persian Gulf and Afghanistan in Central Asia. Base host countries Djibouti and Greenland also stand out as major recipients of U.S. spending despite their relatively small populations.

Instead of the total amount of spending by country—which is largely a function of the size of recipient economies—it is useful to consider spending as a percentage of recipients’ GDP in order to gain a sense of which countries received spending that was disproportionate to their market size. Figure 3 repeats the map in Figure 2 but displays the annual average of U.S. spending as a percentage of recipients’ GDP over the 2000–2015 period. The trends here are similar, with the exception that the largest recipients are more clearly concentrated around the Persian Gulf, Central Asia, Africa, and southeastern Europe, as opposed to areas that contain mainly long-standing formal U.S. allies such as Western Europe and East Asia (with the exception of South Korea). As before, there are a few countries whose totals dwarf those of the others, particularly the 11 for which U.S. defense spending accounted for more than 1 percent of GDP. Most of these countries hosted a U.S. military presence: Kiribati (74.7 percent), Afghanistan (51.1 percent), Djibouti (14.3 percent), Kyrgyzstan (8.9 percent), Marshall Islands (8.7 percent), Iraq (5.3 percent), Greenland (5.3 percent), Kuwait (2.6 percent), Bahrain (2.5 percent), Aruba (1.8 percent), and the Bahamas (1.4 percent). Most countries came nowhere near those totals: the median country received U.S. defense spending in the amount of 0.017 percent of its GDP, and more than three-quarters of countries had totals of less than 0.1 percent of their GDP.

Much of this spending reflects the logistical requirements of a global U.S. military presence, but a nontrivial portion is also designed to achieve strategic foreign policy objectives. The primary way to do this is to manipulate who receives a contract, but that is difficult to do by design. According to the Federal Acquisition Regulation (FAR), U.S. laws “require, with certain limited exceptions … that contracting officers shall promote and provide for full and open competition in soliciting offers and awarding Government contracts.”¹¹ Moreover, while the DOD can encourage prime contractors to subcontract with local firms or hire local labor, it cannot legally stipulate how they fulfill the terms of an awarded contract. There are several exceptions, however, that allow the DOD to award contracts with limited or no competition.¹² These exceptions are either codified in the FAR or enacted as ad hoc preferential procurement policies.

The data allow us to examine which contracts were awarded noncompetitively and why. The most common exception is “one responsible source,” meaning that there was only one viable provider.¹³ We are most interested, however, in contracts that are awarded preferentially as a result of congressional statute or international agreement, such as the legislation that mandated favoring Iraqi and Afghan vendors in 2008, Central Asian vendors in 2010, Djiboutian vendors in 2015, and African vendors in 2017, or the 1951 treaty between the United States and Denmark to hire Greenlandic labor to support Thule Air Base.¹⁴ By focusing on these contracts, our measure is in reality almost certainly an undercount of the phenomenon of strategically awarded contracts.

Strategically noncompetitive contracts account for around $1 billion of overall DOD spending per year and follow a similar pattern as total defense spending. However, their distribution is skewed. Ninety percent of countries received amounts of strategic spending that totaled less than 0.01 percent of their GDP while a handful of countries had totals in excess of 0.1 percent of GDP: Greenland (3.7 percent), Kyrgyzstan (1.3 percent), Afghanistan (0.7 percent), the Marshall Islands (0.2 percent), and Saudi Arabia (0.1 percent).¹⁵ Figure 4 shows the distribution of the annual average of these noncompetitively awarded contracts, again as a percentage of recipients’ GDP over the 2000–2015 period.

What do policymakers hope to accomplish with this spending? In a nutshell, strategic spending is a tool designed to achieve foreign policy outcomes by structuring recipient incentives so that it is in their best interest to support U.S. military objectives. The U.S. military contracts for goods, services, and construction where it operates overseas. By strategically selecting which firms receive the contracts, it can simultaneously fulfill both operational and policy purposes.¹⁶ Our research shows three main objectives for which the United States uses its procurement spending overseas: promoting economic development and political stability after war, securing foreign military access, and “winning hearts and minds” in counterinsurgency.¹⁷

The underlying logic is straightforward: defense procurement creates local jobs and economic opportunities. Market growth and economic development are presumed to contribute to post-conflict stabilization. Washington’s earliest experiments with procurement as a major foreign policy instrument began after World War II, when it used spending to revitalize the economies of allies in Europe and Asia. Japan, for example, benefited from U.S. procurement spending as the United States sought to bolster anticommunist forces within the country and ensure that Japan remained aligned with Washington. U.S. officials instituted a policy of preferential procurement for Japan during the early 1950s, which resulted in $754 million in Japanese goods and services being purchased for use in the Korean War. This policy later made its way into the 1954 U.S.–Japan Mutual Defense Treaty.¹⁸

Spending also seeks to support foreign policy aims by changing local public opinion about a U.S. presence. Providing economic opportunities for host communities should in theory make a U.S. military presence more palatable over the long term. Spending represents what Daniel Nexon and Thomas Wright call a “pivoting” strategy: by making itself attractive to host communities, the United States minimizes the risk of being ousted when the host regime in power changes.¹⁹ In exchange for access to Thule Air Base in Greenland and Naval Base Subic Bay in the Philippines, for example, the United States explicitly incorporated preferential procurement arrangements into its basing agreements with both countries.²⁰

More recently, the United States has relied heavily on procurement spending to expand its military footprint into Central Asia and Africa. Because it lacked both an existing presence and formal allies in these regions, Washington used preferential procurement spending—in addition to large amounts of foreign aid and arms sales—to entice states to host U.S. personnel and equipment. In the early 2010s, Congress authorized a Central Asia First policy of preferentially procuring goods and services from states in the “Northern Distribution Network”—Georgia, Kyrgyzstan, Pakistan, Armenia, Azerbaijan, Kazakhstan, Tajikistan, Uzbekistan, and Turkmenistan—to maintain its supply route to Afghanistan.²¹ In 2015, Washington approved similar procurement authorities for Djibouti, which hosts the only enduring U.S. base on the continent at Camp Lemonnier. Examples of contracts awarded under the Djibouti First program include food and produce, construction services, and telecommunications and internet services on Camp Lemonnier.²² The authority was expanded to all of Africa in 2017.²³

Finally, jobs promote security by reducing the economic grievances that might otherwise motivate participation in insurgencies. Spending was a cornerstone of U.S. counterinsurgency strategies in Afghanistan and Iraq, where early efforts by commanders on the ground to use money as a tool for pacification became part of official counterinsurgency policy. As then U.S. Army Maj. Gen. David Petraeus put it, “Money is my most important ammunition in this war.”²⁴ Military commanders in Iraq believed that “procuring Iraqi supplies and services strengthens the Iraqi economy and helps eliminate one of the root causes of the insurgency in Iraq—poverty and a lack of economic opportunity.”²⁵ Similarly, the Afghan First Initiative was premised on the assumption that offering viable jobs would drain the supply of local insurgent recruits dubbed “$10-a-day Talibs.”²⁶ An early foray into spending was the Commander’s Emergency Response Program, which provided funds for small-scale infrastructure projects in Iraq and Afghanistan and encouraged use of local labor and goods.²⁷ Soon, larger-scale contracting with local firms for goods and services became a mainstay of U.S. counterinsurgency efforts, with a formal name—Money as a Weapons System—and policy guidance promulgated for field officers.²⁸ A Center for Army Lessons Learned handbook for field officers spells out the logic:

After initial demand filtered up from commanders in the field, preferential procurement policies in Iraq and Afghanistan were formally authorized in the National Defense Authorization Act (NDAA) for Fiscal Year 2008. The Iraqi First policy explicitly set out to create a “stable source of jobs” for Iraqis, an objective that would require “a dramatic transformation of procurement practices to maximize Iraqi awards.”³⁰ Contracting officers were instructed: “Whenever possible, all acquisition efforts should be sourced under the Iraqi First program.”³¹ For its part, the Afghan First policy directed U.S. personnel to “hire Afghans first, buy Afghan products, and build Afghan capacity.”³² Contracting officers in Afghanistan were told to “look beyond cost, schedule, and performance. Evaluate the success of a contract by the degree to which it supports the Afghan people and economy and our campaign objectives.”³³ U.S. officials regarded spending in the local economies as a means not only to build goodwill toward the United States and the central governments in these countries but also to reduce the population’s incentives to take up arms by fostering economic growth and creating jobs.³⁴

Spending is thus increasingly treated like an inducement similar to foreign aid or arms transfers, two major tools with which the United States and other great powers attempt to shape the behavior of recipient states. Foreign aid and arms transfers, unlike strategic spending, have received extensive academic and policy attention. Studies have shown, for example, that the United States prefers to provide aid to states that have a record of voting with it in the United Nations General Assembly.³⁵ Other research provides evidence that the United States tends to give more aid to its partners and to friendly states that are geographically close to its adversaries.³⁶ Similarly, research into security assistance suggests that the United States provides arms both to buy influence over partners’ foreign policy decisions and to strengthen friendly states.³⁷

Spending has several flexible features that make it look like an attractive complement or substitute for arms and aid.³⁸ For one thing, it might help to circumvent the problem of interest misalignment with partners that the United States routinely runs up against. The literature on United Nations voting and foreign aid, for one, finds mixed evidence that aid actually succeeds in buying votes.³⁹ Attempts to make aid conditional on recipient behavior often fail because of recipients’ overriding domestic political incentives and the presence of rival donors.⁴⁰ Aid can be siphoned off by ruling regimes for their own purposes, leading to corruption and even increased state-sanctioned violence.⁴¹ This problem is particularly acute in counterinsurgencies. Foreign aid often does not reach the intended participants, and political leaders frequently ignore demands for top-down economic reforms. As Stephen Biddle notes, “Local governments which actively promote a fair distribution of resources rarely face major insurgencies in the first place—hence, Americans are rarely asked to wage counterinsurgency at all in such places.”⁴² In contrast to aid, creating jobs and directly hiring local firms through strategic spending may help to ensure that the right people get the money.

For another thing, spending might reduce the risk of partners behaving badly. Arms transfers may embolden partners to behave more aggressively or enable them to act in ways that run counter to American interests.⁴³ Similarly, military aid designed to help partners tackle security threats can create perverse incentives. If U.S. support is designed to bolster the recipient government against a perceived threat to U.S. interests—whether it is from an insurgency such as the Taliban, a terrorist network like al Qaeda, or an external enemy like the Soviet Union—then recipients have an incentive to play up those threats to maximize the amount of support they can attract.⁴⁴ Andrew Boutton, for example, found that receiving foreign aid does not lead to a reduction of terrorism in autocratic regimes because these regimes divert significant portions of the aid toward buying off domestic political supporters to stay in power and have little reason to squelch the terrorist threat since doing so would cause U.S. aid to dry up.⁴⁵ Commercial activity does not raise the same risk of involvement. Moreover, arms and aid are subject to scrutiny in order to avoid enabling human rights abuses (e.g., the Leahy Amendment)—though the effectiveness of such scrutiny is questionable.⁴⁶

Strategic spending comes with risks and potential downsides, like all foreign policy tools. If these risks materialize, it could erase any benefits and undermine broader goals. There are three major risk categories.

The first risk is that the economic effect on local markets will be harmful. For this effect to be a net positive, spending must reach the right people and lead to sustainable growth. Research suggests that domestic military spending has at best a limited effect on economic growth, which raises questions about foreign military spending’s efficacy.⁴⁷ Even more perniciously, there is evidence that injecting large amounts of money into fragile states with weak economies can stymie development. As a bipartisan congressional committee put it, “Massive inflows of cash into a largely agricultural society with an underdeveloped financial infrastructure”⁴⁸ created market distortions in Afghanistan, exacerbating economic problems and hindering development. Instead of market growth driving development and stability, economic gains in Afghanistan proved unsustainable.⁴⁹ Between 2014 and 2016, the Afghan economy lost the same number of jobs as it had gained between 2009 and 2012.⁵⁰

Even where capacity was created, the fledgling firms that received jobs because of limited competition could not stay afloat once contracts were handed over to the host government. In 2011, the Commission on Wartime Contracting concluded presciently:

A boot manufacturer in Afghanistan, for example, that had undertaken a massive investment to create army and police boots to U.S. standards, abruptly lost its contract when the Afghan government opted for cheaper (substandard) boots from China and Pakistan. Employees laid off from the factory considered turning to the Taliban for work.⁵¹

Additionally, it is far from clear that the right people reap the economic benefits, particularly in areas where corruption is rampant. In such cases, not only is spending not reaching the intended recipients, but it might also increase economic inequality and the negative distributive consequences of rent-seeking behavior. Corruption in Afghanistan, for example, led analysts to question “how much of the money spent on U.S. government contracting is having a positive impact on the Afghan economy and how much of the funds is actually staying in country.”⁵²

The second major category of risk concerns the effects on U.S. interests. Preferential procurement policies are not immune to political and security blowback. For one thing, noncompetitive procurement does not automatically guarantee more government control over who gets the business. On several occasions, the United States has experienced undesired deviations from its diplomatic arrangements when the American private sector’s economic interests diverged from the government’s national security objectives. Between 2013 and 2015, the U.S. military faced crises over two key military bases: Thule Air Base in Greenland and Camp Lemonnier in Djibouti. In the case of Thule, the U.S. Air Force inadvertently awarded the base support contract to an American-owned firm in 2015 despite a treaty dating back to 1951 that obligated the United States to favor Danish and Greenlandic firms.⁵³ Challenges to the contract award dragged through U.S. courts for several years, with the U.S. Court of Appeals for the Federal Circuit finally ruling that the contract was legally binding, U.S. treaty obligations notwithstanding.⁵⁴ This diplomatic row had ripple effects: China attempted to exploit the rift to pursue deepening investments in Greenland.⁵⁵ Similarly, the American company in charge of base support for Camp Lemonnier in 2013 created a crisis in U.S.–Djiboutian relations when it fired the local Djiboutian labor in favor of cheaper labor from Southeast Asia. The United States subsequently renegotiated its base agreement terms to be more favorable to Djibouti, including nearly doubling the annual base rent to $63 million and passing the Djibouti First legislation.⁵⁶

More broadly, wherever large amounts of money are at stake, there are risks of political blowback. In Kyrgyzstan, for example, DOD fuel contracts with privately owned firms linked to Russia ultimately created vulnerabilities both for DOD operations in Afghanistan and for the Kyrgyz government. In late 2010, a U.S. House of Representatives report found the following:

Controversy arose after a subsidiary of the Defense Logistics Agency awarded multibillion-dollar fuel contracts to two companies—Mina and Red Star Enterprises—and then directed them to subcontract with companies owned by family members of Kyrgyz President Askar Akayev. After popular protests over corruption and repression forced Akayev from office in 2005, his successor’s son took over the companies, leading to new allegations of corruption that contributed to President Kurmanbek Bakiyev’s ouster as well.⁵⁸

These policies also can create security risks, particularly in active counterinsurgency environments. Both Iraq and Afghanistan were beset with problems with vendor vetting. In the early years of each program, the U.S. military struggled to establish accountability for its business partners. By 2011, a Congressional Research Service report found that “significant amounts of contracting funds flow to criminal networks and insurgents.”⁵⁹ That same year, the Commission on Wartime Contracting reported that “extortion of funds from U.S. construction projects and transportation contracts is the insurgent’s second-largest funding source” in Afghanistan after drug trafficking.⁶⁰ In total, officials estimated “that a minimum of 10 percent of the Pentagon’s logistics contracts—hundreds of millions of dollars—consists of payments to insurgents.”⁶¹ Neither Congress nor the DOD were blind to these problems: the Pentagon created Task Force 2010 to improve visibility over contractors and recover stolen equipment, and the fiscal year 2012 NDAA authorized DOD contracting officers to terminate contracts with vendors determined to be connected to insurgencies in Iraq or Afghanistan. The U.S. military also created a vendor vetting cell to vet companies in Afghanistan.⁶² But the basic problem of understanding who the U.S. military is doing business with persists in new and highly diverse environments (e.g., 54 different states, each with unique political and security contexts in Africa).⁶³

The final problem is the cost of procurement as a foreign policy tool. The implicit assumption is that spending is a materially and politically cheap complement to or substitute for alternative measures such as security guarantees, arms transfers, or infusions of foreign aid. Spending is seen as materially cheap because it is used to buy things that the military already needs for operations. Because the military procures goods and services for itself, it can satisfy operational and policy objectives simultaneously—two birds with one stone. Spending is considered politically cheap relative to other measures: it reduces the risk of entrapment inherent in alliances or arms transfers, and it distances the United States from regimes or militaries with unsavory human rights records.

However, local contracting can become expensive in markets characterized by sparse infrastructure and multiple government agencies bidding on contracts for the same goods or services. In Iraq, Afghanistan, and various countries in Africa, when the DOD and other U.S. government agencies fail to coordinate, they often end up bidding for goods and services provided by the same vendor and artificially inflating demand (what is known informally as contract fratricide).⁶⁴ Second, even if spending avoids political entanglement, it cannot avoid competition and ratcheting pressures. In other research, we found that the amount of spending that Washington provides for access in countries throughout Africa depends on its ability to find alternative hosts, how badly it needs access in a particular region, and the presence of alternative sources of support for the host.⁶⁵

For example, even assuming that Djibouti First is a good policy with local benefits that create goodwill for the U.S. military presence, the amount of money is still a drop in the bucket when compared with China’s financial investment there. As in other strategically important countries, China has pursued “debtbook diplomacy” in Djibouti, investing billions, much of it in loans that are unlikely to be repaid.⁶⁶ Djibouti’s external debt doubled from around 35 percent of its GDP in 2014 to over 70 percent by 2019;⁶⁷ the United States believes that China holds most of this debt. In 2018, Djibouti terminated its contract with DP World, the Dubai-based company that had operated Djibouti’s sole container port for nearly a decade, and the government nationalized its stake in the Doraleh Container Terminal.⁶⁸ American officials expect that China will eventually use its leverage to compel Djibouti to hand over the port in a debt-for-equity swap, much as it did in Sri Lanka to acquire its Hambantota Port.⁶⁹ Local goodwill among a disenfranchised population is unlikely to make much of a difference in such a situation.

Finally, we suspect there are threshold effects when it comes to spending: small amounts of spending will have no effect while large amounts may have outsized effects. This means that achieving benefits requires spending lots of money. And the more money pumped into local economies, the greater the likelihood of negative outcomes.

To date there has been little empirical analysis of how the U.S. military uses overseas spending or what its spending achieves. Policymakers simply do not systematically scrutinize what the DOD spends abroad. Several congressionally mandated investigations into U.S. government activities in Iraq and Afghanistan—e.g., the Commission on Wartime Contracting and the audit efforts of the Special Inspector General for Iraq Reconstruction and the Special Inspector General for Afghanistan Reconstruction (SIGAR)—have examined problems in how the DOD contracted in those environments. But the investigations were scoped exclusively to wartime efforts and ultimately did not generate enough data to enable a comprehensive evaluation. More recent preferential procurement policies, outside of Iraq and Afghanistan, have largely avoided scrutiny.

Several constraints contribute to the lack of transparency. For one thing, defense spending is hard to track across the board. The military services use financial systems that are not interoperable; coupled with stovepipes between policy, program managers, and finance managers, the DOD itself rarely knows precisely what it is spending.⁷⁰ Hence the DOD’s financial management has been on the Government Accountability Office’s “High Risk” list for 25 years.⁷¹ An apocryphal story from Iraq tells of a vendor walking across base from one service’s contracting office to another with a paper bill of services, getting paid repeatedly for the same bill. Other constraints are external. In Iraq and Afghanistan, severely under-resourced contracting shops had few people to dedicate to oversight, and it was dangerous to go out and evaluate; multiple contracting officer representatives are believed to have been killed in efforts to monitor projects.⁷² Insurgents singled out the firms; vetting and registering vendors proved difficult in part because firms were afraid of association with U.S. forces.

Another problem is the apparent inertia around clarifying and improving reporting requirements. Internally, the DOD has advanced few clear policies; externally, Congress has not demanded adequate data collection or reporting on the results of its legislation. In 2012, for example, the SIGAR found that the data collected in Afghanistan did not provide “a suitable basis for measuring the effect U.S. procurement efforts have had on generating short- and long-term employment gains in Afghanistan in support of U.S. goals.”⁷³ Instead, it found that “U.S. agencies collected employment data from their [Afghan First Initiative] awards for different purposes and in various ways, but not with the intention of measuring increases in Afghan employment” (emphasis added).⁷⁴ The U.S. Army Corps of Engineers, the entity responsible for administering most of the contracts, did not concur with the SIGAR’s recommendation to improve data collection, citing “challenges in measuring local employment in Afghanistan.”⁷⁵ The International Security Assistance Force eventually tried to improve data collection: an executive steering committee was set up to measure “what percentage of contracts are awarded to Afghan firms and what percentage of contract employees are Afghan.”⁷⁶ But it is unclear how long this effort persisted and what data, if any, were ultimately collected.

In light of these challenges, we offer the following recommendations as a starting point:

The U.S. government should require collecting and reporting data on the economic effects of its policies. How many jobs were created? With whom? What were the downstream effects? Research needs to determine whether strategic spending actually achieves its purported objectives and at what cost; there has been no systematic evaluation, for example, as to whether spending outperforms tools such as foreign aid and arms sales in making U.S. military presence palatable to host countries.

Instructions for data collection should be explicit and consistent, with an emphasis on standard rules. The onus to improve data collection begins at the top but should incorporate the operational planners and leaders who have driven demand for the policies. Asking implementers in the field to collect data for data’s sake is futile. As one contracting officer said about the Djibouti First initiative, “I didn’t hear anything about it when I first arrived, then I was told I needed to send reports, now the policy has shifted to Africa First and we’re still in the dark as to what is required of us.”⁷⁷ The types of data needed to assess political and economic effects—for example, jobs created—extend beyond the contracting data collected by the current federal procurement database. Most calls to improve data reporting focus on the Federal Procurement Data System—Next Generation, which does need improvement. However, that system is designed to track contract awards, not contract effects.

Better data will help to address another important question: Do the national security gains from giving contracts to foreign firms outweigh the economic costs of diverting business away from American firms? There is an underlying tension between the desire to use defense spending to advance foreign policy interests (“Host Nation First”) and the desire to use government spending to advance domestic economic interests (“Buy American”). The contradiction occasionally rears its head, as it did in 2014 when blind workers in Louisiana lost their jobs after the military moved to procure mess trays overseas under the Central Asia First legislation.⁷⁸ Subsequent legislation specifically protects the blind: the Africa First legislation stipulates that the authority cannot be used if goods “can be produced and delivered by a qualified nonprofit agency for the blind or a nonprofit agency for other severely disabled in a timely fashion to support mission requirements.” But this is a “whack-a-mole” approach that leaves bigger questions about tradeoffs between U.S. domestic and foreign interests unresolved—and unanswerable in the absence of better data.

As an immediate step, Congress could make ongoing authorization of preferential procurement policies contingent on impact evaluations to ensure that money is being spent in ways that advance U.S. national interests. Multilateral organizations such as the World Bank—and to a lesser degree the U.S. government—have long used impact evaluations on foreign aid programs to understand whether the programs succeed as well as the conditions that encourage favorable outcomes.⁷⁹ Evidence suggests that institutionalizing impact evaluations increases the effectiveness of foreign aid.⁸⁰ The effectiveness of procurement spending needs similar scrutiny. Congress has already moved in this direction when it comes to security cooperation and assistance. The 2017 NDAA contains language calling for institutionalized assessment, monitoring, and evaluation of security cooperation activities.⁸¹ The 2021 NDAA, for example, could incorporate a requirement to evaluate the impact of preferential procurement policies.

Congress should also ensure that the DOD has adequate vendor vetting requirements in place for all preferential procurement policies. The Africa First legislation, for example, includes no explicit direction on how to vet or select vendors.⁸² One possible approach is to create an “expeditionary” FAR, similar to the Committee on Wartime Contracting’s recommendation to create provisions for a “wartime” FAR. These provisions were designed to improve government visibility and monitoring of prime contractors and subcontractors as well as to increase contractor accountability. An expeditionary FAR would extend these provisions to countries where the U.S. government is pursuing similar policies but in a peacetime context. Finally, if impact evaluations return negative results, Congress should be prepared to modify or suspend the policies.

Beyond empirical questions about preferential procurement’s microlevel effectiveness in individual cases (bolstering employment, shoring up economic growth, generating goodwill), there remain good reasons to question its ability to consistently achieve broader foreign policy objectives. The United States has very little control over the subcontracting practices of the firms it does business with, which creates an enormous principal-agent problem. By relying on private firms, the United States effectively delegates its foreign policy to actors whose preferences may not always align with its own. In some cases, this can lead to the threat of major diplomatic ruptures, as occurred with Djibouti in 2013 and Greenland in 2015.

Questions around strategic spending’s efficacy are in many ways a symptom of the challenges that are inherent in using material incentives to buy the goodwill of a foreign population and government or to remake their society and economy. There are good reasons to think that spending, for all its flaws, may sometimes be the least bad option when compared with instruments such as foreign aid and arms sales, since it attempts to minimize corruption by bypassing governments and directly targeting local economies. But being in the position of having to choose from flawed options such as procurement spending (and material incentives more generally) is the byproduct of an environment in which the United States has sought military access in countries where there is not necessarily a broad base of support for its presence and in which its relationship with hosts is transactional rather than based on common interests.⁸³ The root of the issue, then, is more what the United States has tried to accomplish (e.g., counterinsurgency and local economic development in areas in underdeveloped and conflict-ridden states) rather than how it has tried to do it. So long as the United States continues to pursue similar objectives, it will inevitably have to choose between imperfect policy tools.

Defense spending is an understudied tool of U.S. statecraft compared with foreign aid and arms sales, but many of its costs and risks look familiar. Like aid, spending can skew local economies, generate dependencies, and foster corruption. Like arms sales, spending can produce unintended consequences and perverse incentives.⁸⁴ Consider the cases in which the DOD inadvertently funded insurgents in Iraq and Afghanistan. Similarly, the presence of alternative sources of material support may drive the United States into bidding wars, as Washington found in Kyrgyzstan and Djibouti during the 2010s.

Defense procurement spending also poses unique problems and deserves greater and more consistent oversight. For one, because spending relies on commercial actors to distribute the benefits, the United States is effectively delegating its economic statecraft to private firms whose subcontracting practices it cannot legally control (with very few exceptions).⁸⁵ The United States thus faces a “democratic disadvantage” when compared with states such as China, which can more easily command its state-owned enterprises to do business in ways that further state interests.⁸⁶ Moreover, a lack of oversight and accountability in contracting practices has in many cases resulted in accidentally funding adversaries—such as insurgents in Iraq and Afghanistan—or in contract fratricide.

That said, there are circumstances in which procurement spending may represent a more suitable—if still flawed—instrument. In particular, spending can be useful for injecting money directly into the hands of local actors, thus bypassing government officials. This may have value in the context of counterinsurgency campaigns, in which the incentives of the central government often diverge from those of the United States, such that the former may prefer to distribute resources to a small number of elite supporters, punish opponents, and undermine potential rivals rather than provide public goods—though even here, bypassing the government may do little to build its legitimacy in the long term.⁸⁷ Similarly, providing benefits for local populations can decrease public resentment of U.S. military presence—whether inside or outside the context of active conflict—and allow Washington to avoid directly supporting autocratic ruling governments through transfers of aid and arms.⁸⁸ Moreover, some of the alternatives to using economic tools to secure military access—such as offering security guarantees—are not ones to deploy lightly, as they carry the potential to entrap the United States in its partners’ conflicts.⁸⁹

Ultimately, the question of spending’s efficacy is one not only of means but also of ends. Costs are relative and must be assessed in light of the value of the foreign policy objectives. Defense procurement spending is a useful tool only when the United States is already engaged in costly and potentially fraught scenarios: those in which its forces are stationed on the soil of foreign countries. Counterinsurgency, for example, is always lengthy and costly and rarely occurs in ideal circumstances; as Biddle points out, the fact that a government faces an insurgency indicates that it is likely corrupt and ineffective in governing.⁹⁰ Meanwhile, other studies suggest that states hosting U.S. military presence may become more willing to take risks such as underinvesting in military readiness, diverting resources to counter internal threats (i.e., “coup-proofing”), and implementing risky fiscal policies in the expectation of being bailed out.⁹¹

It is thus difficult to avoid broader questions of grand strategy: What are core U.S. interests? How large does Washington’s overseas footprint need to be to pursue these interests? What role should the U.S. military play in advancing those interests? We cannot hope to answer such questions here; however, it should be recognized that assessing the use of spending presumes that the sort of goals for which it can be effective are desirable in themselves. What is unequivocal is that overseas defense spending deserves greater scrutiny, from both scholars and policymakers. There is still little evidence that spending has positive effects on recipient economies, is more effective than aid or arms in securing and holding onto military access, or contributes to desired counterinsurgency outcomes.⁹² In the absence of such evidence, spending is a tool best used sparingly and with caution.

Joyce, Renanah Miles, and Brian Blankenship. “‘Money as a Weapons System’: The Promises and Pitfalls of Foreign Defense Contracting.” Policy Analysis No. 892, Cato Institute, Washington, DC, June 3, 2020. https://​doi​.org/​1​0​.​3​6​0​0​9​/​P​A.892.