Public Attitudes toward Private Military Companies: Insights from Principal

Abstract

This article examines the distribution and correlates of mass attitudes toward the privatization of US military operations. Relying on insights from principal–agent theory, we form predictions about what beliefs are associated with a willingness to grant authority over military operations to private military companies (PMCs). The model predicts that citizens’ beliefs about actor motives, accountability, and costs are associated with attitudes toward PMCs. Using a nationally representative survey, we find that beliefs about the profit-oriented motives of private firms and perceptions of their lack of accountability reduce support for the use of PMCs, particularly in combat operations. By contrast, belief in private firms’ superior fiscal efficiency increases support for utilizing PMCs in both combat and noncombat operations. The results illustrate the usefulness of principle-agent theory for understanding mass attitudes and help to improve the field’s understanding of the contours of public attitudes toward US defense policy.

Keywords

military contractors principal–agent theory public opinion private military companies

[M]ercenaries are a dangerous way to go to fight a war. These are men without morals, to kill for a living other than being a soldier is despicable.

—Thomas McMahon (Millis, MA)

We hire mercenaries to mask the actual fiscal and human costs of the war(s), because those making the wars know the American public isn’t committed enough to support a draft…If Congress can’t agree that a war is necessary, and the people can’t support it enough to consent to a draft, we shouldn’t go to war.

—Dave (United States)¹ On September 16, 2007, military contractors employed by Blackwater Security Company murdered seventeen Iraqi civilians and wounded twenty more, while escorting a US embassy convoy. The Blackwater incident revealed to the American public that private contractors, and the private military companies (PMCs) that employ them, were not only carrying out vital national security operations but also performing these duties in a manner that violated widely shared American values. Five years later, private security contractors employed by Blue Mountain Group abandoned their posts when Ansar al-Sharia militants attacked the US diplomatic compound in Benghazi, Libya. This attack, which resulted in the deaths of US Ambassador J. Christopher Stevens and three other Americans, again directed public attention to the US government’s reliance on private security contractors in active conflict zones. Indeed, then Secretary of State Hillary Clinton was heavily criticized for using private contractors to guard the embassy and its personnel. These examples, as well as the quotes preceding the introduction, further demonstrate that the American public is often critical of its government’s use of PMCs in conflict situations. Given the potential salience of the issue, further exploring how individuals form their opinions toward the use of PMCs and their comparisons of these actors to traditional military forces deserves additional theoretical and empirical scrutiny.

The question of who should fight wars on behalf of the citizens of the United States is both new and substantively important (see Brunstetter and Holeindre 2018). Despite long-standing international norms against mercenaries and international laws prohibiting their use, many democratic Western states increasingly rely on private corporations to provide key services in combat zones (e.g., Krahmann 2010; Singer 2008). Indeed, private contractors currently outnumber US military personnel in conflict zones around the world (Zenko 2016). The deepening reliance on PMCs and their increasing presence in theaters of active conflict arguably represents one of the most significant shifts in US defense policy in recent years. As such, numerous recent studies have investigated states’ motives for employing PMCs, the potential implications of their use, and international and domestic efforts to control and regulate them (e.g., Avant 2005; Chesterman and Lehnardt 2007; Krahmann 2010; Singer 2008).

Despite the attention PMCs have received in recent years, scholars know seemingly little about public perceptions of these actors. We believe that understanding citizens’ opinions toward PMCs is important for several reasons. First, normative and positivist interpretations of democratic theory rest on an assumption that public policy reflects the preferences of the state’s citizens (Achen and Bartels 2016; Pitkin 1967). Without an examination of the distribution of public attitudes toward PMCs, we have no way of knowing the extent to which the current use of private contractors deviates from the preferences of the public. Second, public opinion has the potential to influence the use of PMCs. Scholars of international relations have argued that public opinion can influence foreign policy decisions (e.g., Holsti 1992), particularly in democracies where audience costs have the greatest impact on leader survival (Fearon 1994). Public opinion may also influence the continuation and success of military operations (Baum and Potter 2015; Sobel 2001). Thus, to the extent that citizens’ preferences shape policy decisions in democratic states, public opinion toward the use of PMCs potentially constrains or facilitates their use even if we do not directly observe citizen input in the policy-making process.² Third, by examining the question of who should fight on behalf of the United States, we offer new insights into existing literature on the public’s foreign policy-related opinions (e.g., Gartner and Segura 1998; Gelpi, Feaver, and Reifler 2009; Hurwitz and Peffley 1987; Milner and Tingley 2015).

In this article, we investigate public perceptions of PMCs and the factors that inform public opinion regarding their deployment. We argue that public support for deploying PMCs in noncombat and combat operations is partly a function of pragmatic concerns over the use of profit-seeking third-party agents engaging in actions that have meaningful implications for national security. We rely on principal–agent theory to derive a set of predictions about what concerns citizens might have when the US government relies on private companies to engage in military operations. These include concerns about adverse selection, the moral hazard problem, and agency costs—concerns that critics of the use of for-profit military forces have raised for centuries (see Percy 2007). Although citizens do not necessarily think in these abstract terms, they often possess sufficient experience with privately held companies to be able to reason about the likelihood that private firms are held accountable for their actions, what motivates private firms’ behaviors, and the comparative efficiency of private firms (Ramirez and Lewis 2017; Smith 2000). As we discuss in greater detail below, this extends to the use of PMCs in armed conflicts.

In the pages that follow, we first discuss the history of PMCs and highlight some important differences between mercenary armies and PMCs. We then review previous literature on the principal–agent problem applied to the armed actors, which we believe has particular relevance to understanding both how PMCs operate and how they are perceived by outside observers. We use insights from this literature to develop a set of arguments about the ways in which US citizens reason about the US government’s use of PMCs to achieve its foreign policy goals. We then describe the research design and present results from a nationally representative survey examining contemporary opinions toward PMCs. We conclude with a discussion of the relevance of our findings for policy-making as well as research related to the use of PMCs in theaters of active conflict.

Old Wine, New Bottles?

The use of private military forces and other freelance soldiers has a lengthy historical precedent. Mercenaries were once a common feature of armed conflict, and for centuries, many Western states relied heavily on various types of for-profit soldiers to fight their foreign wars and defend their colonial holdings (Avant 2005, 26-29; Percy 2007). However, norms against employing mercenaries emerged in Europe at the beginning of the nineteenth century and diffused globally through the twentieth century (Cockayne 2007, 200; Percy 2007). The increasingly negative attitudes toward mercenaries held by Western leaders stemmed from both ethical and strategic factors. In the late eighteenth century, particularly following the British defeat in the American Revolution, European leaders increasingly questioned the effectiveness of mercenaries and raised concerns about the negative externalities associated with their use. Chief among these was the belief that mercenaries were driven by interests that were inconsistent with those of the states that hired them (Percy 2007). Mercenaries were seen as prone to engaging in self-serving, potentially counterproductive predatory behaviors such as plundering, looting, perpetrating wonton violence, and evading challenging combat situations—concerns that reflect the core principal–agent problems that inform the arguments we advance below. Additionally, many leaders perceived that the use of mercenaries signaled a country’s lack of commitment or lack of capability to achieve its war aims. On a normative level, critics highlighted the similarities between slavery and hiring mercenaries, asserting that paying foreign leaders for their troops represented a similarly unethical “trade in human flesh.” Consequently, British leaders came to view mercenaries as a moral disgrace that undermined British greatness (Percy 2007).

This combination of strategic and normative factors ultimately led to a precipitous decline in the use of mercenaries. By the mid-nineteenth century, major European powers largely eschewed the use of mercenaries and instead favored citizen armies to fight their wars (Avant 2000; Krahmann 2010; Percy 2007).³ These norms were codified into international law in various ways during the late twentieth century (Krahmann 2010; Percy 2007).⁴ While international laws prohibiting the use mercenaries are often regarded as weak and ineffective (Percy 2007; Salzman 2008), the formalization of international norms through the ratification of international laws reflects the international community’s general opposition their use.

Despite the existence of such norms and the seemingly pervasive recognition of the potential negative externalities associated with the use of for-profit military personnel, many countries view PMCs as integral components of their military operations (Krahmann 2010; Singer 2008). The apparent rise in states’ use of private security forces might suggest that norms against mercenaries have weakened in recent decades or were perhaps only minimally internalized by many states. Indeed, critics have often asserted that PMCs are simply a twenty-first century repackaging of mercenaries and soldiers of fortune and carry with them many of the same potential political, strategic, and normative liabilities (see Avant 2009; Salzman 2008; Singer 2008, 213-15). Yet, recent scholarship highlights the difficulties in comparing contemporary PMCs with historical conceptions of mercenary armies. Krahmann (2010, 5-7), for instance, contends that the corporate nature of PMCs and the legal status they possess reflect a clear contrast with mercenary forces, which have traditionally operated outside of the law and without any form of formal regulation. Avant (2005, 22) offers a more nuanced assessment of private security, asserting that terms like mercenary and the distinction between public and private forces do not travel well across time and space.

Close scrutiny of the history of the use of contract and for-profit forces by sovereign states reveals a variety of different types of actors, each with a distinct relationship to its employer, set of responsibilities, and organizational structure (Avant 2005, 27-30; O’Brien 2007, 37-38). Moreover, existing international law has continued to permit the use of various forms of commercial military activity (Cockayne 2007, 200-201). Despite international laws explicitly outlawing their use, mercenaries (in the conventional sense also) continued to operate in some conflicts into the later decades of the twentieth century (O’Brien 2007, 35-36). Viewed in this broader historical perspective, the use of PMCs represents less of a dramatic shift in policy and practice by Western democratic governments and more of a continuation and evolution of a long-standing practice of outsourcing security. Consequently, while the existing academic literature largely rejects the simplistic assertion that PMCs are the modern equivalents of mercenaries, it nonetheless suggests that Western states have simply continued the long-established practice of augmenting their official security forces with for-profit organizations or individuals.

As this discussion implies, clearly defining PMCs and differentiating them from mercenaries and other for-profit armed organizations represents both a challenge and a central debate in the recent literature on their use, regulation, and behavior. While acknowledging these challenges, we follow other recent studies (e.g., O’Brien 2007, 38; Singer 2008) and define PMCs as private, profit-motivated companies operating under a contract with the government of a state (or its military or intelligence agencies) to provide a range of security and military services, including tactical combat operations, intelligence gathering, military training, and logistical and technical support for that government’s formal military forces. As a general rule, these organizations are legal businesses, possess a corporate structure, are subject to contract law (in the contracting state), and submit to some (often vague) regulation of their activities (see Krahmann 2010, 9-10). The latter characterization is not central to our definition of PMCs, but it helps to further differentiate the actors in which we are primarily interested from more traditional mercenary groups. PMCs also differ from the professional military forces of the contracting state in several key ways: their services are contracted on an as-needed basis, the specific services they are intended to provide are stipulated by these contracts, they operate outside of the formal military command structure of the state that employs them, and the individual contractors they employ are not (formal) members of that state’s military.

US Reliance on PMCs

While the contemporary use of PMCs to provide services in war zones arguably represents a continuation of long-established practices, the depths of US reliance on such organizations appear to reflect an important change in the way it fights its foreign wars. The American government has employed private firms and civilian contractors to provide construction, maintenance, and transportation services in and around conflict zones since at least the Second World War. The use of PMCs to provide logistical and other forms of support to the US military during armed conflicts increased during the Vietnam War (Kidwell 2005, 13-16), and following the Cold War, the role of PMCs expanded to include military support operations in regional conflicts such as Kosovo, Somalia, Rwanda, and Haiti (Avant 2005).

US reliance on PMCs to support its combat operations deepened significantly as a result of its simultaneous involvement in Iraq and Afghanistan. Conducting concurrent counterinsurgency campaigns in both countries (without implementing the draft) was arguably only possible because the US government was able to shift much of the burden of security operations from the military to PMCs. Prior to early 2000s, private contractors most frequently engaged in noncombat operations such as the construction of infrastructure and logistical support; yet, during the conflicts in Iraq and Afghanistan, the role of PMCs increasingly included combat-related operations such as combat training, interrogation, psychological warfare, hostage negotiations, intelligence, and security details (Schwartz 2011). In these roles, armed private contractors were deployed in situations in which they sometimes became active participants in hostilities, thus largely erasing the lines between military personnel and private civilians. Reliance on private soldiers is now higher than at any point since the eighteenth century as sovereign states and international organizations around the globe routinely contract with PMCs to fill at least some of their logistical and security needs (Singer 2008; Zenko 2016).

The high-profile activities in which many PMCs participate and the (often negative) publicity they have received in the news media has raised numerous questions regarding the potential negative externalities associated with their use, efforts to regulate and control them, and their efficiency and cost relative to conventional state forces. Many of these concerns stem from core principal–agent problems that are common features of armed groups and have been noted by critics of mercenaries for centuries. We discuss these below and then apply insights from the principal–agent problem to develop a set of arguments about the ways in which citizens form opinions about the use of PMCs.

The Principal–agent Problem and PMCs

In its simplest form, principal–agent theory explores the implications of informational asymmetries between a principal, who has the formal authority, and an agent, who is enlisted by the principal to carry out a specific set of tasks. Analyses utilizing principal–agent theory generally assume the following: (1) an agent enlisted by the principal chooses some action among a set of available alternatives, (2) this action affects the welfare of both principal and agent, and (3) the principal rewards (or sanctions) the agent based on observing the results of the agent’s actions. The challenge for the principal is to ensure that the agent carries out the assigned task in an acceptable manner when it cannot necessarily observe the agent’s actions and where the agent therefore possess an informational advantage with respect to its actions (and preferences).

Previous studies highlight several problems detrimental to the principal that arise from these information asymmetries. The first of these problems, adverse selection, occurs at the point of agent recruit when the principal unintentionally chooses agents who are more likely to pursue their own interests. The second problem, moral hazard, focuses on the conflicting goals between the principal and the agent and results when the self-interested agent undertakes actions by which it profits but that ultimately prove detrimental to the principal or its goals. The third problem is agency cost. This represents the costs paid by the principal as a result of the actions of the agent. These include direct costs for the agent to act on the principal’s behalf, costs the principal must pay resulting from the agent’s actions, and any regulatory costs to monitor the agent. In each case, the principal’s information disadvantage adversely impacts its ability to control the agent and increases the odds that the agent engages in self-interested and costly behaviors.

Previous research in politics and policy has made extensive use of principal–agent theory to understand the conditions that allow principals to mitigate or minimize these problems and to investigate their implications (see Brehm and Gates 1999; Miller 2005). Most relevant to our study, existing studies commonly note that informational asymmetries between the principal and its agents can lead to a variety of adverse security or human rights outcomes. For instance, moral hazard and/or adverse selection increases the likelihood that soldiers shirk their costly combat duties and instead engage in more self-serving activities such as smuggling, looting, perpetrating sexual violence, or abusing local civilians (Butler, Gluch, and Mitchell 2007; Gates 2002; Mitchell 2004; Salehyan, Siroky, and Wood 2014; Weinstein 2006). In addition to the obvious human costs, these actions may adversely impact the principal’s broader goals. Principally, where soldiers devote more time to selfish activities and less to combating the enemy, they potentially undermine the principal’s ability to achieve its objectives.

As noted above, military and political leaders have long been aware of the potential negative externalities associated with employing for-profit military forces (e.g., mercenaries; Percy 2007). Many of these problems stem from the informational asymmetries at the core of the principal–agent relationship, including concerns over costs, limitations on accountability, and the divergence between the interests of agents and principals (Avant and Segilman 2010; Cockayne 2007; Stranger and Williams 2006). Agency costs, for instance, represent a persistent problem for states employing PMCs. Although the idea that free market competition leads to more efficient operations represents a primary motive for hiring PMCs (Landler, Schmitt, and Gordon 2017), PMCs can exploit their informational advantage by overcharging the government for supplies and services (Whyte 2006). Moreover, the need for contractors with expertise and security clearances means contract bidding often goes uncontested and market mechanisms fail to drive costs down (Avant 2009; Tkach 2017). The greater expertise of PMCs over newly enlisted US military personnel also means private firms can demand higher wages for their employees, leading to morale problems among military troops performing the same tasks for lower pay.⁵

The inability to hold PMCs accountable for their actions, which exacerbates the moral hazard problem, also presents serious challenges for states. The US military possesses a clearly defined chain of command, defined codes of conduct, a well-developed internal disciplinary system, and effective internal sanctioning mechanisms. Moreover, the military is subject to oversight by elected civilian authorities, which permits the public to exert some (albeit limited) influence over the armed forces. At the minimum, if American citizens strongly disapprove of actions undertaken by the military, they have the option of expressing this disapproval at the ballot box and punishing the elected officials who create and shape policy. These features are absent or comparatively weak in the case of PMCs. For example, PMCs contracts often forbid firms from publicly disclosing their contents, and the Department of State and Department of Defense often avoid discussing the exact nature of PMCs duties and compensation.⁶ Contractors are also not subject to the same chain of command and military codes of conduct as traditional US military forces (Avant 2005, 128-30), impeding the ability of civilian and military authorities to scrutinize, police, and sanction the behaviors of private contractor personnel deployed overseas. Journalists and security analysts also face comparatively greater difficulties in gathering and aggregating information on PMCs operating overseas, further reducing public and congressional knowledge about their activities (Avant and Sigelman 2010).

These conditions make adverse selection a substantial concern for the US government and the public. According to the principal–agent framework, the lack of accountability generates substantial opportunities for carelessness, abuses, shirking, and other activities that are ultimately detrimental to the principal’s security goals. High-profile cases of abuse involving contactors such as those in Nisour Sqaure and Abu Ghraib, or human trafficking by peacekeepers employed at DynCorp during the Kosovo War, illustrate these concerns. In these and similar cases, the actions of PMCs have adversely influence the strategic interests and reputation of the US government and its citizens. Policy makers and military leaders therefore rely on various aspects of national laws to address these issues and assert nominal control over these agents. The most direct mechanisms include screening, institutional checks, monitoring, and sanctions (Cockayne 2007, 202). Such efforts sometimes prevent the most egregious negative externalities associated with PMCs. Nonetheless, a variety of factors, including the distance between principals and agents, the relative bargaining power of the agents (which is a function of the principal’s reliance on the agent), and the presence of multiple principals often minimize the effectiveness of these mechanism in managing principal–agent problems (Cockayne 2007, 202-205). As the examples above demonstrate, despite the efforts of the governments that employ them, PMCs often engage in actions that are contrary to the principal’s goals.

These observations have implications for the US government’s reliance on PMCs to fight its foreign wars. We also suspect that these factors influence the way that citizens evaluate the use of PMCs. In our argument and in the subsequent empirical analysis, we focus on the comparison between for-profit PMCs and professional, largely voluntary state military forces (as opposed to conscript armies). We do this primarily because we are specifically asking US citizens to compare PMCs and the US military. The US military has been a wholly voluntary force since the termination of the draft in 1973, and this characteristic almost certainly influences how respondents think about the potential negative externalities associated with the two types of security actors we consider.⁷ As we discuss in the next section, citizens’ concerns about the potential negative externalities that stem from the principal–agent problem should help shape their opinions about the use of PMCs.

Citizen Reasoning about PMCs

Using these insights, we seek to better understand how US citizens construct opinions regarding the use of PMCs. We suggest the public possesses the ability to recognize core potential principal–agent problems and uses this knowledge in forming opinion about the use of PMCs. Although citizens are not direct principals overseeing private contractors, citizens tend to subscribe to a version of democratic folk theory where government bureaucracies and agents, funded by taxpayer dollars, have a formal obligation to serve the public (e.g., Alvarez and Brehm 1998). When asked to make a judgment regarding the use of PMCs, citizens rely on the same considerations they would use to evaluate any other potential agent: concerns about costs, accountability, and the potential for the agent’s motives to diverge from their own. At first glance, this might seem like a great burden on citizens, whose thinking is frequently described as dominated by partisanship (e.g., Achen and Bartels 2016). Yet, citizens already have experience in evaluating potential agents, such as when they evaluate car mechanics, hire a plumber, or vote for an elected official. Thus, even though citizens are not necessarily very knowledgeable about PMCs, they should nonetheless be able to form coherent opinions regarding their use.

First, citizens should be able to make a judgment about whether PMCs can operate in a more cost-efficient manner than US military forces—a judgment that hinges on whether people view private companies, in a general sense, as benefiting from market forces. There is a common belief among the public that private firms have the ability to exploit competition in the marketplace, which results in more cost-efficient operations. According to this argument, businesses simply do things better, especially when it comes to keeping costs down. In contrast, the government is seen as bloated and wasteful with stories of the US military paying exuberant prices for inexpensive tools and parts. A perception that private firms operate in a more cost-efficient manner than the government should be associated with support for the privatization of US military operations. In contrast, opponents of privatization argue that the financial gains from privatization are never guaranteed. Instead, the operating costs of private firms can often meet or exceed the operating costs of the government, and empirical data often back up such claims (Avant 2009; Bel and Warner 2008; Calaguas 2006; Lundahl et al. 2009). Private firms often have an informational advantage regarding their operating costs that allows them to increase costs and maximize profits (Tkach 2017). A perception that private firms fail to operate in a more cost-efficient manner than the government should be associated with opposition for the privatization of US military operations.

Second, a plausible implication of the principal–agent framework applied in this context is that citizens should be concerned about the accountability of PMCs in instances where PMCs engage in actions that are contrary to the interests of the state or its citizens. The ability to hold public officials and agents working on the behalf of the government accountable is important to the public, particularly in domains where citizens lack specific knowledge (Hutchings 2003). This is because accountability can motivate agents to behave in a manner consistent with the preferences of the principal thereby helping to minimize the moral hazard problem. Accountability also provides a mechanism to recoup losses in the event the agent acts in a manner contrary to the principal’s interests, thereby allowing the principal to avoid the costs of frequent monitoring. Citizens who view private firms as accountable for their actions should be more likely to approve of private firms engaging in military operations.

Third, citizens should be able to form an opinion on whether PMCs act in the interests of the nation. When asked to support or oppose the use of PMCs, people are likely to evaluate whether private contractors possess goals that are consistent with the interests of the state and its citizens. As illustrated by this article’s opening quote, the profit-driven goals of private contractors should be readily visible to most citizens. Citizens might anticipate that private contractors could deviate from the interests of the state in order to increase profits. Indeed, the very act of profiting from conflict could be seen as contrary to American values and ideals. Conversely, some citizens appear to adhere to a belief that private companies, contractors included, operate because of a desire to contribute to the overall well-being of society (Smith 2000). Avant (2009) points to human rights groups who view PMCs as mechanisms to hasten interventions in conflict areas that state actors wish to avoid. For instance, PMCs were vocal about their ability to end genocides in places like Rwanda and Dafur (Isenberg 2008). Although they expect to be paid for these services, some people might perceive this behavior as indicating PMCs do have a desire to provide a societal benefit. Citizen who perceive PMCs as acting in the best interests of the country and its citizens, as opposed to being self-interested actors maximizing personal profits, should be more approving of PMCs.

Data and Measurement

We test these arguments using data from 1,000 respondents to the 2014 Cooperative Congressional Election Study (CCES). The CCES is conducted by YouGov on behalf of a consortium of scholars and uses a sample matched to the US population. All questions except for the measure of deployment in a real-world conflict were collected during in the preelection wave during October 2014. The real-world deployment question was part of the postelection wave in November 2014. We apply sample weights to adjust for any deviation between the US population and the sample (details in the Online Supplemental Material).

Measurement of Support for the Use of PMCs

We conceptualize support for privatization as a positive or negative evaluation about the appropriateness or inappropriateness of contracting government services to private, nongovernmental organizations. First, we asked respondents to express their generic support for PMCs. The question avoided the word “privatization” to minimize confusion among respondents unfamiliar with the term. Instead, it explained the issue in the following simple terms:

Consider some federal government services or functions that might be contracted out to private companies, or to other non-governmental organizations. For each service, please rate how appropriate it is for that service to be provided by private organizations. Very appropriate/should always be provided by private organizations, sometimes appropriate, neither appropriate or inappropriate, sometimes inappropriate, and always inappropriate/should never be provided by private organizations.

Responses to the item “military and security operations” are used to form the variable generalized support. ⁸ Figure 1 shows the distribution of responses to this question. When asked about PMCs without any context of the operation to be performed, very few citizens favor using PMCs. A minority (28 percent) of respondents view using PMCs as “always” or “sometimes” appropriate. A majority (55 percent) of respondents believe using PMCs is always or sometimes inappropriate. There also appears to be room to increase or decrease support for PMCs with 17 percent of respondents indicating that they find the use of PMCs “neither appropriate or inappropriate.”⁵

Figure 1.

Distribution of public support for PMCs.

Second, we asked respondents to express their support for six noncombat and combat functions commonly performed by US military forces. We expected the public might prefer to limit PMCs to some, mostly noncombat, type of operations. The wording of the question also explained the issue in simple terms by asking respondents,

How acceptable do you think it is for private companies hired by the U.S. government to engage in the following activities in order to achieve U.S. military and foreign policy goals. Very acceptable, sometimes acceptable, or never acceptable. [Providing food and supplies to U.S. military troops; Rebuilding of a foreign country after a war; Non-combat peacekeeping missions; Interrogation of enemy fighters; Attacking enemy targets; Attacking enemy targets when civilians are at risk]?

Responses to these items form the logistics, infrastructure, peacekeeping, interrogation, low-risk combat, and high-risk combat variables. In order to compare how citizen support for PMCs compares against traditional US military troops performing these same operations, we asked the same question replacing PMCs with US military forces.

Figure 2 shows responses to these questions for each type of operation. The bars represent the percentage of respondents who found it either “very acceptable” or “sometimes acceptable” for PMCs or the US military to engage in each type of operation.⁹ Almost all respondents support the use of PMCs in noncombat operations (logistics, postconflict reconstruction, and noncombat peacekeeping operations). Support for PMCs wanes when we examine support for specific types of combat operations, but a slight majority of respondents approve of PMCs engaging in interrogations and low-risk combat situations. The only instance in which a majority of respondents oppose PMCs is when asked if PMCs should engage in combat operations where civilians are at risk. However, this is partly because a large number of respondents oppose the type of operation (42 percent oppose the US military engaging in the same type of operation).

Figure 2.

Public support for using PMCs and the US military in specific operations.

We do find a strong correlation between opposition to PMCs more generally and opposition to PMCs engaging in combat operations, suggesting some respondents answered the generic support question as if it implies using PMCs in combat operations. We also find a small percentage of respondents (17 percent) exhibit weakly held opinions regarding the generic use of PMCs. It is these respondents with weakly held opinions that are responsible for much of the difference between support for using PMCs in specific combat situations and support for the use of PMCs more generally.¹⁰

Although the public prefers the military to PMCs in almost all types of operations, the preference for the military is substantively large when the public considers combat-related operations. A test of proportions across groups shows the difference in support for the military engaging in combat operations relative to PMCs is large and statistically significant (interrogation gap = 32 percent, p < .001; low-risk combat gap = 29 percent, p < .001; high-risk combat gap = 20 percent, p < .001). Because the public appears to distinguish between noncombat and combat operations, we subjected the data to an exploratory factor analysis using an oblique promax rotation since the latent factors are likely correlated. We uncover two distinct factors. The first factor is associated with opinions toward combat-related operations (interrogation, low-risk combat, and high-risk combat), while the second factor is associated with opinions toward noncombat operations (logistic support, infrastructure, and peacekeeping missions). Given that the combat and noncombat items are manifestations of two distinct beliefs, we combine the logistic support, infrastructure, and peacekeeping variables into a single additive index representing public support for PMCs in noncombat operations (PMCs noncombat, α = .65). We also combine the interrogation, low-risk combat, and high-risk combat variables into a separate additive index representing public support for PMCs in combat operations (PMCs combat, α = .78). Both variables are standardized to range from 0 to 1 on a near-continuous scale. We use these as the dependent variables in the analyses.

Similarly, respondent opinions toward the US military show a similar structure with opinions toward combat operations loading on one factor, while opinions toward noncombat operations load on a separate factor. Therefore, we combine the variables asking respondents whether they support the US military engaging in logistic support, infrastructure, and peacekeeping into an additive index representing opinions toward the US military in noncombat operations (military role [noncombat], α = .66). We also combine the variables asking respondents whether they support the US military engaging interrogations, low-risk combat, and high-risk combat into an additive index representing opinions toward the US military in combat operations (military role [combat], α = .65). Both variables are standardized to range from 0 to 1 on a near-continuous scale. We use the military role (noncombat) variable in the PMCs noncombat model and the military role (combat) variable in the PMCs combat model in order to control for each respondent’s general support or opposition for the different types of operations as well as beliefs about the US military.

Measurement of Beliefs about Accountability, Motivation, and Cost-efficiency

Principal–agent theory suggests that citizens should be concerned with the accountability of PMCs. Although respondents are unlikely to have specific opinions regarding the accountability of PMCs, they should have some sense of the ability to hold private firms accountable for their actions (Ramirez and Lewis 2017; Smith 2000). We measure each respondent’s belief about the accountability of private firms by asking, “Occasionally, private companies break the law or are involved in public scandals. In such cases, how often do you think private companies are held responsible for their actions?” Responses range on a 101-point scale from 0 = “never held responsible” to 100 = “usually held responsible.”

Principal–agent concerns over costs should also factor into how people form their opinions toward PMCs. We measure whether respondents believe that private companies can operate at a lower cost than the government by asking, “In general, do you feel that hiring private companies to provide government functions and services is likely to save the government money or cost the government money?” The question gets to the heart of arguments that privatization can provide financial savings to taxpayers by operating more efficiently than the government. Responses to this question form the agent efficiency variable. Higher values indicate a belief that private companies can operate more efficiently than the government.

Principal–agent theory also suggests that citizens should be concerned about selecting agents with goals or motives contrary to their own. We measure each respondent’s belief about the motivation of potential agents by asking them the following: “Please indicate if you think members of each of the following organizations are motivated more by serving the public’s interests or by money and personal benefits…[Private companies/the US military]?” We form the agent motivation variable by taking the difference between the belief that PMCs are motivated by profits and the belief that the US military is motivated by profits.¹¹

Measurement of Control Variables

Given that PMCs might be seen as a substitute for US military forces, we expect that the same variables associated with support for the use of military force to be associated with support for PMCs. We build on recent work by Gelpi, Feaver, and Reifler (2009) to construct a set of control variables related to the use of military force. The core values of universalism and need for security were measured using questions from the Human Values Scale designed to capture these specific value orientations (Schwartz 2007). Respondents were asked to rate the importance of two dimensions of each value. The question preface asked respondents, “Please rate how important each of the following goals is to you as a guiding principle in your life?” The universalism variable (α = .85) consists of each respondent’s combined rating of “equality (equal opportunity for all)” and “being broadminded (tolerant of different ideas).” The need for security variable (α = .88) consists of each respondent’s combined rating of “national security (protection of my nation’s enemies)” and “family security (safety of loved ones).”

We also include variables for respondent gender (1 = women), race (1 = white), education (no high school, high school graduate, some college, two-year degree, four-year degree, and postgraduate degree; but treated as continuous), age in decades, and if they live in a military family (1 = military family). We also include a set of religious variables that previous research suggests are correlated with opposition to the use of force. These include variables indicating whether the respondent is Atheist (1 = atheist or agnostic) and Catholic (1 = Catholic). It also includes a variable indicating how important religion is to the respondent, religious (1 = religion is important). We control for political ideology (7 = very conservative) using the standard seven-point National Election Study variable. We control for partisanship using a binary variable for respondents that identify with the Democratic Party (Democrat = 1) and a binary variable for respondents that identify with the Republican Party (Republican = 1). We use these binary measures instead of the seven-point partisanship scale since the latter scale would exclude respondents that identify with third parties (e.g., Tea or Green Party supporters) from the analyses. All variables are standardized for the analyses to assess their relative importance.

Estimation and Results

We estimate public opinion toward PMCs in noncombat and combat roles (the PMCs noncombat and PMCs combat variables) using a seemingly unrelated regression (SUR). We choose this approach given each dependent variable ranges from 0 to 1 on a near-continuous scale and the likelihood that any unmeasured factors that shape support for PMCs in noncombat and combat roles are correlated in their error structure. In other words, the unobserved variables related to PMC support in each model are likely the same.¹² SUR estimates do not provide estimates of casual relationships. Instead, the model reasonably assumes that respondent’s preferences toward PMCs do not shape respondent’s demographic characteristics, core principles, and beliefs about private firms. The latter is consistent with Smith’s (2000) observation that beliefs about private firms are “devoid of policy content.”¹³

The complete SUR estimates are shown in Table 1. Figure 3 plots the standardized coefficients along with 95 percent confidence intervals to show the relative substantive effect of the main variables of interest. Each estimate in the plot shows the change in support for PMCs from a one-unit shift in each variable. Overall, the principal–agent model does a good job of explaining public opinion toward PMCs. A belief that corporations are accountable for their actions shows a positive and statistically significant relationship with support for allowing PMCs in combat operations (β_sur = 0.07, 95 percent CI [0.02, 0.13]). A standard deviation increase in perceptions of corporate accountability (27 points on the scale) is associated with a 0.07 increase in support for PMCs in combat operations. This translates to about one-third of a standard deviation shift in support for PMCs in combat operations or about the same effect size that ideology has on support for PMCs. In regard to PMCs engaging in noncombat operations, the accountability variable is positive but statistically insignificant (β_sur = 0.02, 95 percent CI [−0.03, 0.07]). The desire to hold PMCs accountable for combat-related behaviors, but not non-combat-related behaviors, might simply reflect the public finding combat-related abuses more egregious than noncombat-related abuses.

Table 1.

SUR Estimates of Public Support of Private Military Companies.

	Noncombat		Combat
Sex	0.01	(0.02)	0.02	(0.03)
Race	−0.00	(0.03)	−0.08*	(0.03)
Education	0.02	(0.03)	−0.06*	(0.03)
Age	0.01	(0.03)	0.00	(0.03)
Catholic	0.00	(0.03)	0.06*	(0.03)
Religious	−0.04	(0.03)	0.05	(0.03)
Atheist	0.03	(0.03)	0.04	(0.03)
Democrat	−0.01	(0.03)	0.01	(0.03)
Republican	−0.03	(0.03)	0.04	(0.03)
Ideology	0.13*	(0.03)	0.09*	(0.04)
Military	−0.09*	(0.03)	−0.00	(0.03)
Patriotism	0.15*	(0.03)	0.01	(0.03)
Universalism	0.03	(0.04)	−0.13*	(0.04)
Need for security	−0.01	(0.04)	0.18*	(0.04)
Agent efficiency	0.10*	(0.03)	0.06*	(0.03)
Agent motivation	−0.04	(0.03)	−0.21*	(0.03)
Agent accountability	0.02	(0.02)	0.07*	(0.03)
Military role (noncombat)	0.46*	(0.03)
Military role (combat)			0.28*	(0.03)
Constant	−0.04	(0.03)	−0.04*	(0.03)
N	939		939
R ²	0.26		0.26

Note: Coefficients are estimates from SUR models. Standard errors are in parentheses. The correlation of residuals between the combat and noncombat models = .32. Variables are standardized. SUR = seemingly unrelated regression.

*p < .05.

Figure 3.

SUR estimates of respondents’ opinions toward PMCs in combat and noncombat roles.

The agent motivation variable shows a negative and statistically significant relationship with approving of PMCs in combat operations (β_sur = −0.21, 95 percent CI [−0.27, −0.16]). This means respondents who believe private firms are motivated by self-interest to a greater extent than the US military are less likely to support PMCs in combat-related roles. A standard deviation increase in the belief that private firms are motivated by self-interest (a one-point categorical shift in the variable) is associated with a 0.21 decrease in support for PMCs in combat operations—almost an entire standard deviation shift in the dependent variable. Thus, perceptions of agent motivation have one of the strongest substantive effects of any variable in our models. In contrast, the agent motivation variable shows no relationship with approving of PMCs in noncombat operations (β_sur = −0.04, 95 percent CI [−0.10, 0.01]). This suggests that respondents are more concerned with the possibility that the for-profit motive of private firms could lead to abuses and shirking of combat goals than noncombat goals. An equality of coefficients test partly confirms this claim, showing a clear difference in how agent motivation is related to attitudes toward PMCs engaging in combat versus noncombat roles (χ² = 25.12, p < .001).

Citizens are also concerned about the ability of PMCs to operate more efficiently than the government. The agent efficiency variable is positive and statistically significant in both the combat (β_sur = 0.06, 95 percent CI [0.00, 0.12]) and noncombat models (β_sur = 0.10, 95 percent CI [0.04, 0.16]). Respondents who believe private firms operate more efficiently (cheaply) than the government are more likely to approve of PMCs in both types of operations. A standard deviation shift in the agent efficiency variable (a one-point categorical shift in the variable) is associated with a 0.06 increase in support for PMCs in combat operations and a 0.04 increase in support for PMCs in noncombat operations. This effect is substantively small equating to about one-fifth of a standard deviation shift in combat operations and a one-seventh of a standard deviation shift in noncombat operations. Thus, efficiency doesn’t appear to be a major concern among the public when thinking about privatizing military operations. This is likely because basic concerns about well-being and security trump concerns about costs.

The results also suggest that approval of PMCs is associated with factors that are correlated with other foreign policy–related opinions such as the use of force. Respondents who express a higher need for security are more supportive of PMCs in combat roles (β_sur = 0.18, 95 percent CI [0.10, 0.26]). A standard deviation increase in someone’s need for security (a one-point categorical shift in the variable) is associated with a 0.18 increase in support for PMCs in combat roles, which is over half of a standard deviation shift in the dependent variable. The need for security variable shows a negative and statistically insignificant relationship with support for PMCs in noncombat roles (β_sur = −0.01, 95 percent CI [−0.09, 0.06]).

Similarly, the universalism variable shows a negative and statistically significant relationship with support for PMCs engaging in combat operations (β_sur = −0.13, 95 percent CI [−0.22, −0.05]) and a positive and statistically insignificant relationship with support for PMCs engaging in noncombat roles (β_sur = 0.03, 95 percent CI [−0.04, 0.11]). A standard deviation shift in the universalism variable (a one-point categorical shift in the variable) is associated with a 0.13 decrease in support for PMCs in combat operations. This is slightly less than a half of a standard deviation shift in the variable. An equality of coefficients test shows these differences are statistically significant (χ² = 14.24, p < .001). This suggests that respondents who strongly value equality and tolerance are less likely to support PMCs in roles that could potentially involve physical violence; however, they are generally unconcerned about the use of PMCs in noncombat roles.

The results also show that conservatives are more likely to support PMCs in combat (β_sur = 0.09, 95 percent CI [0.01, 0.16]) and noncombat operations (β_sur = 0.13, 95 percent CI [0.05, 0.20]). Since we control for each respondent’s general preference for having the military engage in the same types of operation and beliefs about the relative efficiency of private firms, we suspect the relationship between ideology and support for PMCs is due to either general philosophical beliefs toward limited government rather than the more hawkish views of conservatives. A standard deviation shift in ideology (a one-category shift in self-identification) is associated with a 0.09 increase in support for PMCs in combat operations and a 0.13 increase in support for PMCs in noncombat operations. The latter is almost half of a standard deviation shift in the dependent variable and equivalent in size as the effect of universalism. Neither variables representing identification with the Republican or Democratic Party show a statistically significant relationship with support for PMCs. This suggests that partisanship has yet to serve as a cue to structure how citizens should think about PMCs.

Testing the Deployment of PMCs in a Real-world Conflict

It is possible that respondents’ opinions toward PMCs in hypothetical situations differ from their opinions toward the deployment of PMCs in a real-world conflict. Much of the literature on foreign policy opinions relates to real conflicts where casualties are a concern (e.g., Vietnam, Iraq). Therefore, the postelection wave of the CCES asked a small subset of respondents whether they were willing to deploy PMCs in an ongoing conflict that is likely to result in PMC casualties.¹⁴ Specifically, a random third of respondents were asked

A recent proposal to stabilize the situation in the Middle East calls for the deployment of troops from a private security firm, which is expected to result in the death of an estimated 200 private security personnel. All things considered, how good of an idea is it for the United States to send troops to the Middle East to stabilize the situation there?

Response options range on a 101-point scale from 100 = “a very good idea” to 0 = “a very bad idea.” These responses are used to create the deployment variable. We intentionally constructed the question to be generic rather than to refer to any specific event in a specific location (“stabilize the situation in the Middle East”) since we did not know how events would unfold from the time we submitted the survey protocols to the time when the survey was administered. This could introduce some random measurement error into the responses. However, by November 2014, the United States was largely out of Iraq with the situation in Syria and the Islamic State of Iraq and Syria (ISIS) having grown into the major problem in the Middle East. Since the latter dominated media coverage of the area, we suspect most respondents had this conflict in mind when answering the question. The United States had approved airstrikes against ISIS in October 2014 with no commitment to sending ground troops at the time of the administration of the postelection wave of the survey—although the potential for ground troops was openly discussed in the media. The mean support for the deployment of private security forces suggests respondents were hesitant to deploy PMCs in a more realistic conflict situation (mean = 34.00, 95 percent CI [31.29, 37.72]).

Regression results using the deployment variable are shown in Table 2. Despite the low statistical power resulting from the reduced sample size, regressing the deployment variable on the model shows continued support for two of the three variables representing principal–agent concerns. Figure 4 displays the coefficients of interest. The variable accounting for beliefs about agent efficiency shows a positive and statistically significant relationship with support for deploying PMCs in a Middle East conflict zone with potential casualties (β_ols = 6.48, 95 percent CI [2.64, 10.31]). This translates to an approximately one-fourth of a standard deviation increase in support for the deployment of PMCs in combat zones from a categorical increase in the belief that corporations operate more efficiently than government. This is slightly higher than the effect agency efficiency had on support for combat operations in a hypothetical combat situation.

Table 2.

Support for Deployment of PMCs in Conflict Zones.

Sex	−0.56	(1.81)
Race	−4.61*	(2.32)
Education	0.53	(1.62)
Age	−0.04	(1.70)
Catholic	1.64	(1.69)
Religious	−2.08	(2.07)
Atheist	0.14	(2.11)
Democrat	3.65	(1.91)
Republican	−1.69	(1.81)
Ideology	2.54	(2.62)
Military	−0.74	(1.86)
Patriotism	1.66	(3.17)
Universalism	−3.66	(2.74)
Need for security	6.48*	(2.72)
Agent efficiency	6.48*	(1.95)
Agent motivation	−8.46*	(1.80)
Agent accountability	−0.31	(1.69)
Military role (combat)	−0.61	(1.90)
Constant	33.50*	(1.63)
N	249
R ²	0.25

Note: Coefficients are estimates from an ordinary least squares (OLS) regression. Standard errors are in parentheses. Variables are standardized. PMCs = private military companies.

*p < .05.

Figure 4.

OLS estimates of support for deploying PMCs in the Middle East.

By contrast, the agent motivation variable shows a negative and statistically significant relationship to support for the deployment of PMCs in a conflict zone (β_ols = −8.46, 95 percent CI [−12.00, −4.9]). A categorical shift in the agent motivation variable is associated with an approximate one-third of a standard deviation shift in support for the deployment of PMCs. This means respondents who believe private firms are motivated by self-interest to a greater extent than the US military are less likely to support the deployment of PMCs to a conflict zone. This occurs despite the possibility that the negative views of PMCs held by respondents might make them perceive PMCs as more expendable than regular US troops. In other words, respondents with a negative view of PMCs might be more inclined to support sending them into a conflict zone when casualties are expected. Instead, the results further support the notion that citizens are more worried about the conflict of interest in deploying PMCs working for profit.

Finally, the agent accountability variable is negative but statistically insignificant in this model (β_ols = −0.31, 95 percent CI [−3.6, 3.00]). It is therefore possible that accountability matters in an abstract sense, but that it is less important to respondents when they are asked to think about sending troops into an actual combat situation.

Conclusion

In this article, we presented an argument and empirical evidence regarding public support for PMCs. First, we provide the first empirical evidence regarding the distribution of opinions toward the use of PMCs showing stark differences in how the public evaluates PMCs when asked more generally versus in more specific terms. We also demonstrate the public distinguishes between using PMCs in noncombat versus combat situations. Second, our study presents a novel application of principal–agent theory to the study of public opinion in order to make predictions about the correlates of public opinion toward PMCs. We found that when asked to form an opinion about PMCs, citizens rely on the same considerations they use to evaluate other potential agents. These consist of beliefs related to the agent’s cost-effectiveness, the potential for the agent to act contrary to the interests of principal, and the ability to hold agents accountable for failures and abuses. These beliefs are closely related to both the expected benefits that motivate the decision to outsource tasks to agents (efficiency) and the concerns principals have about the effectiveness of agents and the negative externalities associated with their use (i.e., adverse selection and moral hazard). Thus, we contribute to the growing evidence that citizens can make principled foreign policy judgments (Kertzer and Zeitzoff 2017).

Beliefs about agent efficiency, accountability, and motivation were all correlated with public approval of PMCs in theoretically predictable ways. Concerns about adverse selection and moral hazard problems appear to be more relevant to the public’s preferences regarding PMCs in combat roles. Based on this evidence, we suspect that the public would be more opposed to PMCs if they were aware of the full extent that PMCs can avoid accountability for behaviors that are counter to American values and US interests abroad. In the absence of clear partisan or elite divisions on the issue, this finding points to a generally pragmatic public.

Somewhat surprisingly, we did not to find overwhelming opposition to the use of PMCs in specific combat operations. The results initially might lead to the conclusion that current defense policies, which are reliant on PMCs to engage in the same types of roles described in this study, are not too far out-of-step with public preferences. Despite the public’s acceptance of PMCs in a wide range of military roles, elected officials should not get too comfortable supporting their use. The public might levy more blame for abuses by PMCs than US military personnel, which could extend to elected officials. Moreover, we have not examined how the public would react to the revelation that many private contractors are foreign nationals. We suspect support for PMCs would decline from the levels shown here if the public was made aware that US diplomatic security or its military mess halls abroad were staffed with third-country nationals from places ranging from Peru to Pakistan. This calls for more in-depth measures of public preferences for PMCs that further match the real-world use of contractors.

The result also speaks to a common claim that the public would prefer to put private contractors, rather than the military, in harm’s way. Proponents of private contractors often argue that contractors are more expendable than military personnel allowing the United States to stay engaged in conflict zones in the face of high casualties. We find that only a small proportion of respondents support deploying PMCs in a real conflict zone when faced with the prospect of casualties. Of course, during this period, the public was weary of the continued conflict in the Middle East (e.g., Iraq) and the mission objectives of additional troops were unclear by the question wording. The public might be more inclined to favor sending PMCs in other contexts with clearer objectives.

Supplemental Material

Supplemental Material, JCR-18-0098.R1 - Public Attitudes toward Private Military Companies: Insights from Principal–agent Theory

Supplemental Material, JCR-18-0098.R1 for Public Attitudes toward Private Military Companies: Insights from Principal–agent Theory by Mark D. Ramirez and Reed M. Wood in Journal of Conflict Resolution

Footnotes

Acknowledgments

The authors thank Ben E. Goldsmith, Michael Koch, and Scott Gates for helpful comments on earlier drafts of this article. The authors would like to acknowledge the work of Paul G. Lewis in helping construct some of the survey items used in this analysis.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the National Science Foundation, Award #1430505.

Supplemental Material

Supplemental material is available for this article online.

Notes

References

Achen

Christopher H.

Bartels

Larry M.

. 2016. Democracy for Realists: Why Elections Do Not Produce Responsive Government. Princeton, NJ: Princeton University Press.

Alvarez

R. Michael

Brehm

John

. 1998. “Speaking in Two Voices: American Equivocation about the Internal Revenue Service.” American Journal of Political Science 42 (2): 418–52.

Avant

Deborah

. 2000. “From Mercenaries to Citizen Armies: Explaining Change in the Practice of War.” International Organization 54 (1): 41–72.

Avant

Deborah

. 2005. The Market for Force: The Consequences of Privatizing Security Cambridge, UK: Cambridge University Press.

Avant

Deborah

. 2009, Summer. “Private Security Contracting Undermines Democratic Control of Foreign Policy.” In The Mercenary Debate: Three Views, The American Interest, vol. 4, edited by Friesendor

Cornelius

Friedrichs

Jorg

. Accessed May 10, 2017. https://www.the-american-interest.com/2009/05/01/the-mercenary-debate/.

Avant

Deborah

Sigelman

Lee

. 2010. “Private Security and Democracy: Lessons from the US and Iraq.” Security Studies 19:230–65.

Baum

Matthew A.

Potter

Philip B. K.

. 2015. War and Democratic Constraint: How the Public Influences Foreign Policy. Princeton, NJ: Princeton University Press.

Bel

Germa

Warner

Mildred

. 2008. “Does Privatization of Solid Waste and Water Services Reduce Costs? A Review of Empirical Studies.” Resources, Conservation, and Recycling 52 (12): 1137–348.

Berinsky

Adam J.

2009. In a Time of War: Understanding American Public Opinion from World War II to Iraq. Chicago, IL: University of Chicago Press.

10.

Brehm

John O.

Gates

Scott

. 1999. Working, Shirking, and Sabotage: Bureaucratic Response to a Democratic Public. Ann Arbor: University of Michigan Press.

11.

Brunstetter

Daniel R.

Holeindre

Jean-Vincent

. 2018. The Ethics of War and Peace Revisited. Washington, DC: Georgetown University Press.

12.

Burns

John F

. 2009. “John Burns Q. and A.: Private Military Contractors.” The New York Times, October 22, 2009. Accessed December 8, 2017. https://atwar.blogs.nytimes.com/2009/10/22/john-burns-q-and-a-private-military-contractors/.

13.

Butler

Christopher

Gluch

Tali

Mitchell

Neil

. 2007. “Security Forces and Sexual Violence: A Cross-national Analysis of a Principal-agent Argument.” Journal of Peace Research 44 (6): 669–87.

14.

Calaguas

Mark

. 2006. “Military Privatization: Efficiency or Anarchy? Chicago-Kent Journal of International and Comparative Law 6 (1): 58–252.

15.

Chesterman

Simon

Lehnardt

Chia

. 2007. From Mercenaries to Markets: The Rise and Regulation of Private Military Companies. New York, NY: Oxford University Press.

16.

Cockayne

James

. 2007. “Make or Buy? Principal-agent Theory and the Regulation of Private Military Companies.” In From Mercenaries to Market: the Rise and Regulation of Private Military Companies, edited by Chesterman

Simon

Lehnardt

Chia

, 196–216. Oxford, UK: Oxford University Press.

17.

Fearon

James D.

1994. “Domestic Political Audiences and the Escalation of International Disputes.” American Political Science Review 88 (3): 577–92.

18.

Gartner

Scott Sigmund

Segura

Gary M.

. 1998. “War, Casualties, and Public Opinion.” Journal of Conflict Resolution 42 (3): 278–300.

19.

Gates

Scott

. 2002. “Recruitment and Allegiance: The Microfoundations of Rebellion.” Journal of Conflict Resolution 46 (1): 111–30.

20.

Gelpi

Christopher

Feaver

Peter D.

Reifler

Jason

. 2009. Paying the Human Costs of War: American Public Opinion and Casualties in Military Conflicts. Princeton, NJ: Princeton University Press.

21.

Hainmueller

Jens

. 2012. “Entropy Balancing for Causal Effects: A Multivariate Reweighting Method to Produce Balanced Samples in Observational Studies.” Political Analysis 20:25–46.

22.

Holsti

Ole R.

1992. “Public Opinion and Foreign Policy: Challenges to the Almond-Lippmann Consensus.” International Studies Quarterly 36 (4): 439–66.

23.

Hurwitz

Jon

Peffley

Mark

. 1987. “How Are Foreign Policy Attitudes Structured? A Hierarchical Model.” American Political Science Review 81 (4): 1099–120.

24.

Hutchings

Vincent L.

2003. Public Opinion and Democratic Accountability: How Citizens Learn about Politics. Princeton, NJ: Princeton University Press.

25.

Isenberg

David

. 2008. “Contractors vs. Genocide?” Middle East Times, August, 8. Accessed January 23, 2016. https://www.cato.org/publications/commentary/contractors-vs-genocide.

26.

Kertzer

Joshua D.

Zeitzoff

Thomas

. 2017. “A Bottom-up Theory of Public Opinion about Foreign Policy.” American Journal of Political Science 61 (3): 543–58.

27.

Kidwell

Deborah

. 2005. Public War, Private Fight? The United States and Private Military Companies. Global War on Terrorism Occasional Paper 12. Ft. Leavenworth, KS: Combat Studies Institute Press.

28.

Kingdon

John W.

1973. Congressmen’s Voting Decisions. New York: Harper and Row.

29.

Krahmann

Elke

. 2010. States, Citizens and the Privatization of Security. Cambridge, UK: Cambridge University Press.

30.

Landler

Mark

Schmitt

Eric

Gordon

Michael R.

. 2017. “Businessmen Get a Say on Afghan War Strategy.” New York Times, July 11, 2017, p. A1.

31.

Lundahl

Brad W.

Kunz

Chelsea

Brownell

Cyndi

Harris

Norma

Vleet

Russ Van

. 2009. “Prison Privatization: A Meta-analysis of Cost and Quality of Confinement Indicators.” Research on Social Work Practice 19 (4): 383–94.

32.

Miller

Gary

. 2005. “The Political Evolution of Principal-agent Models.” Annual Review of Political Science 8:203–25.

33.

Milner

Helen V.

Tingley

Dustin

. 2015. Sailing the Water’s Edge: The Domestic Politics of American Foreign Policy. Princeton, NJ: Princeton University Press.

34.

Mitchell

Neil

. 2004. Agents of Atrocity: Leaders, Followers, and the Violation of Human Rights in Civil War. New York: Palgrave MacMillan.

35.

O’Brien

Kevin A.

2007. “What Future, Privatized Military and Security Activities? The Need for Pragmatic Engagement.” The RUSI Journal 152: 254–61.

36.

Pitkin

Hanna Fenichel

. 1967. The Concept of Representation. Berkeley: University of California Press.

37.

Percy

Sarah

. 2007. Mercenaries: The History of a Norm in International Relations. Oxford, UK: Oxford University Press.

38.

Ramirez

Mark D.

Lewis

Paul G.

. 2017. “Beliefs about Corporate America and the Structure of Opinions toward Privatization.” Political Behavior. Accessed December 11, 2017. https://link-springer-com-443.web.bisu.edu.cn/article/10.1007/s11109-017-9434-4.

39.

Salehyan

Idean

Siroky

David

Wood

Reed M.

. 2014. “External Rebel Sponsorship and Civilian Abuse: A Principal-agent Analysis of Wartime Atrocities.” International Organization 68 (3): 633–61.

40.

Salzman

Zoe

. 2008. “Private Military Contractors and the Taint of a Mercenary Reputation.” New York University Journal of International Law and Politics 3 (Spring): 853–92.

41.

Singer

Peter W.

2008. Corporate Warriors: The Rise of the Privatized Military Industry. Ithaca, NY: Cornell University Press.

42.

Schwartz

Moshe

. 2011. The Department of Defense’s Use of Private Security Contractors in Afghanistan and Iraq. Congressional Research Service Report. Washington, DC, 1–22.

43.

Schwartz

Shalom H.

2007. “Value Orientations: Measurement, Antecedents and Consequences across Nations.” In Measuring Attitudes Cross-nationally: Lessons from the European Social Survey, edited by Jowell

Roger

Roberts

Caroline

Fitzgerald

Rory

Eva

Gillian

, 169–204. Thousand Oaks, CA: Sage.

44.

Smith

Mark

. 2000. American Business and Political Power. Chicago, IL: The University of Chicago Press.

45.

Sobel

Richard

. 2001. The Impact of Public Opinion on US Foreign Policy since Vietnam: Constraining the Colossus. New York: Oxford University Press.

46.

Stranger

Allison

Williams

Mark Eric

. 2006. “Private Military Corporations: Benefits and Costs of Outsourcing Security.” Yale Journal of International Affairs 2 (Fall/Winter): 4–19.

47.

Tkach

Benjamin

. 2017. “Private Military and Security Companies, Contract Structure, Market Competition, and Violence in Iraq.” Conflict Management and Peace Science. doi: 10.1177/0738894217702516. Accessed October 13, 2017. https://journals-sagepub-com.web.bisu.edu.cn/doi/abs/10.1177/0738894217702516.

48.

Weinstein

Jeremy

. 2006. Inside Rebellion: The Politics of Insurgent Violence. Cambridge, UK: Cambridge University Press.

49.

Whyte

Dave

. 2006. “The Crimes of Neo-liberal Rule in Occupied Iraq.” The British Journal of Criminology 47 (2): 177–95.

50.

Zenko

Micah

. 2016, 5/6. “Mercenaries Are the Silent Majority of Obama’s Military.” Foreign Policy. Issue 218. Accessed January 7, 2016. https://foreignpolicy.com/2016/05/18/private-contractors-are-the-silent-majority-of-obamas-military-mercenaries-iraq-afghanistan/.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.21 MB

0.00 MB