The Market for Force: Exploring the Privatization of Military Services

by Deborah Avant

The Market for Force: Exploring the Privatization of Military Services

by Deborah Avant

Prepared for discussion at the Council on Foreign Relations Study Group on the Arms Trade and the Transnationalization of the Defense Industry: Economic versus Security Drivers.

Colleagues: This paper is a very rough draft. Its basis is a proposal for a project on which I am just beginning to do serious research. I have included enough of the proposal (I hope) to give you an idea of the broader project - because I would love to have your feedback. I have also included some case material, but it is pretty speculative. I'm not sure it fits together all that well, but it should give us something to talk about. I welcome your comments and advice, but please do not cite this paper without asking me first.

Max Weber defined the state by its monopoly over the legitimate use of violence. In concert with that definition, we have grown used to thinking about war as an activity carried out by military branches of the state. However, it is not only state armies that provide security in today's world. Increasingly, private military companies provide security, either through supplying and training personnel or by operating right alongside regular military forces. When Sierra Leone faced an insurgency, Valentine Strasser's government hired Executive Outcomes, a private South African security firm, to protect it.1 When the Dayton Accords required that the Bosnian government receive training for its military, the U.S. and its European allies orchestrated a bidding process and Military Profession Resources Incorporated (MPRI) was hired to build and train an effective military force in Bosnia. It is not only states that hire these firms to protect their security. In 1997 World Wildlife Fund received a bid from Saracen, a private security firm, to help prevent poaching that threatened the Northern White Rhino in the Congo. Humanitarian relief organizations such as CARE and Refugees International have had similar brushes with private security.

What are the implications of privatizing security? Analyses of privatization typically focus on the potential for increased cost efficiency. My interest, however, lies in how accountable these firms are likely to be to civilian (and democratic) goals.2 Is the move toward private forces simply a different vehicle for harnessing state resources (whether more efficient or not) that promises little change in the ultimate control of coercive power? Will established systems for holding militaries accountable work just as easily for private firms? Will new systems develop that reflect prevailing norms? Or does the devolution of security tasks to private firms threaten to empower new groups and transform authority over who decides when, how, and over what to fight? Will this undermine systems of public accountability? Should we worry about the privatization of security?

In this project I am trying to answer these questions by charting the general dimensions of the phenomenon and looking at case studies to examine whether (or under what conditions) accountability issues arise. Looking at broad patterns of spending, numbers of firms, and other market indicators, my first goal is to determine the degree, timing, and regional patterns of change in the private provision of force. The second is to focus on the relationship between exporter governments and military companies, importer governments and military companies, and non-governmental organizations and military companies to see whether and how the private provision of force matters for democratic accountability in specific instances. In this essay I will lay out the broad outlines of the project and then examine a couple of illustrative cases.

Mapping Change: What is New? Any analysis of the privatization of security must begin with a documentation of what has changed. This is important for determining the scope of the implications. After all, during the Cold War, the rise of the military industrial complex (composed of mostly private firms) in advanced countries was well documented. The supply of weapons and logistics support by private companies has long been a fact of life in the United States. Similarly, paying foreign forces to fight is hardly new. Mercenaries were a common and generally accepted phenomenon in Europe before the French Revolution.3 They were common but frowned upon in Africa during the Cold War.4 What is different about recent trends? Do they represent a change in the scale of privatization? A shift in the states sanctioning or allowing private security activities? Are new practices being privatized? Has there been a change in the nature or behavior of the firms providing the services? Are there new actors involved in the employment of security instruments?5

The anecdotal evidence suggests a variety of changes. Shearer argues that there has been an increase in the scale of private military activities.6 Private firms also appear to be reaching into new areas. While the most overtly military activities of private security firms still occur in Africa, and we continue to see privatization of non-core missions in the U.S. and Britain, increasingly private firms from advanced countries are carrying out training missions in other states. Also, while the Cold War saw some instances of private companies hiring private security, the post-Cold War has introduced the phenomenon of non-governmental organizations (NGOs) hiring private security firms. Finally, the private military firms of the post-Cold War look quite different from the mercenaries of the Cold War. They are corporate, many of them have web sites and they grant interviews to journalists. These companies want to be considered legitimate. This leads them to operate with attention to international law, and claim to work only for established governments. Their behavior complicates an analysis that simply classifies these firms as new era mercenaries.

To get a better handle on this evidence, it is useful to think about whom pays for security (the horizontal axis) and who performs it (the vertical axis). See Table 1. This matrix draws from privatization studies that examine the choice between public and private financing and performance.7 However, I add a distinction based on whether the public entity that pays for the service is the national government, a foreign government, or an international organization. (It is national payment if the forces reside in the same country that pays them; it is foreign payments if they reside in a different country) There is also a category for private payment. Similarly, I distinguish between whether those supplying the service are national forces, foreign forces, multilateral forces, or private security forces. 8

As we move from national delivery down the vertical axis to private delivery, we move from typical national militaries, men and women in service to their country, to mercenaries, foreigners or private companies providing military service. This trajectory risks problems with accountability. Moving across the horizontal axis, we move from tax dollars to foreign, multilateral or private dollars, risking a change in who holds militaries accountable. Using this matrix as an organizing tool, we can begin to chart the general pattern of the privatization of military services.

[This work has yet to be done, so let me just elaborate a little on what I plan to do. David Shearer usefully divides the tasks typically provided by military firms into the following five general areas (listed here from most to least overlap with military organizations): military operational support, military advice and training, logistical support, security services, and crime prevention services.9 It is my intention to look at how has the pattern changed between the Cold War and post-Cold War. Using data from a variety of sources (ACDA, The SIPRI Yearbook, The Military Balance, etc.) I will construct as best I can, the general spending patterns in each category during each decade going back to 1950.10 I will also examine changes in the number of military firms, the services they provide, and the clients they work for during the same time period. Finally, in the case studies elaborated on below, I will examine the nature and behavior of prominent firms. (There are some instances where firms such as Vinnell operated to support US operations in Vietnam and elsewhere during the Cold War.) Any suggestions about how to add to this empirical analysis would be much appreciated!]

Why Privatization of Military Services Now? A related set of questions focuses on what is driving privatization in the post-Cold War. In brief, there seem to be both supply and demand factors driving the increase in military firms. The supply factors derive from both local (the end of apartheid in South Africa) and global (the end of the Cold War) phenomenon, and are often complicated by issues at the regional level. The downsizing of militaries whether due to the end of the Cold War or internal changes has resulted in a flood of experienced military personnel into the private sector. Looking to history, it is often the case that the downsizing of militaries (for whatever reason) leads to a rise in private military activity.11

Concomitant with the increase in supply has been an increase in the demand for military skills on the private market. Countries downsizing their militaries have seen hiring private security firms as a cost-effective way to maintain defense. The idea that "private" is more efficient has surfaced as a prominent movement in many western states.12 Defense departments have not been immune from the effect of the privatization movement. Interest in privatization and outsourcing in the US led the founders of MPRI to see such efforts as a business opportunity. MPRI began as an attempt by retired American military officers to capitalize on what they saw as increasing demand in the United States for private companies to supply non-core military services. Similar dynamics have encouraged firms in other western countries as well.

Once these companies are established (for whatever reason), they have been pulled into international work. The demand has been partially from other countries (as I discuss below). It has also, however, provided another tool for the conduct of foreign policy. In the US, for instance, advice and training missions have never been a center piece mission for military organizations. The US's history of using troops for these purposes has caused politicians to often be skeptical of training missions.13 Having private firms filled with recently retired officers allow US officials with a more politically palatable choice with which to affect change in unstable countries. The US can maintain neutrality, US troops are not in danger, and yet the policies being implemented can be very similar to what US forces would do.14

Another layer of demand is in countries in Central and Eastern Europe who have sought to upgrade and westernize their militaries as a way to demonstrating their ability to cooperate with the West and (particularly) enter NATO. Though the demand for western style military advice is not new and there were military-military exchanges even before the end of the Cold War, the demand for advice has increased at the same time that it has become increasingly clear that these services are a drain on national militaries. Private companies based in the US or Britain present attractive alternatives to these countries.15

Meanwhile, rulers of weak states with miserable or disloyal armies have seen private security firms as offering a mechanism for defense that does not risk political stability. This dynamic is most prevalent in Africa, though there have also been a few instances in Asia. If a ruler cannot trust his own military or is facing immediate security problems and does not have the time to build up indigenous military skills, private firms offer a useful alternative.

Finally, of course, it is not only states that worry about security in these situations. Private firms, NGOs, groups of citizens, etc. are also in the market for security.16 NGOs are in particularly difficult straits. As crises proliferate and weak states cannot be trusted to refrain from pilfering aid, governments in the west funnel more and more aid through NGOs.17 These organizations, though, (particularly conservation and relief organizations) frequently find themselves operating in areas where the individual security of staff as well as the security of their mission is threatened. Private security firms offer advice in dealing with personal security. They can also help train staff with security skills to better pursue their mission goals.

I should add one caveat to the demand side of this argument. As we move from advanced industrial countries outsourcing non-core functions to less developed countries seeking military advice, we find increasing issues about who will pay. Entering NATO states (or NATO want-to-be's) are eager for advice, but not always eager to pay. This has been one of the big issues of negotiation. But for NATO expansion to muster political support in the US, it had to be inexpensive. US estimates for the first round of NATO expansion were considered to be unreasonably low by most sources. This is because Hungary, Poland, and the Czech Republic do not have the resources necessary to upgrade their militaries to NATO standards. Companies like MPRI have made significant sales pitches to provide the advice about how to upgrade their militaries (in fact it was during one of these sales pitches that Croatia purportedly learned of MRPI), but have not landed a contract because it is not clear who will pay.18

The problems are even more severe in many African States. With the end of the Cold War has also come a lessening desire on the part of either the US or the Russians to use military advice or arms for diplomacy. Countries in Africa that depended on this influx of military expertise have found themselves suddenly footing the bill for their own security, unable to pay off powerful forces in the military and facing domestic unrest on top of it all. These governments may prefer to hire "legitimate" private security firms like MPRI (or even Executive Outcomes). But only those countries with significant resources in oil, minerals and gems can afford to pay. Others may be limited to hiring less savory (and less expensive) elements from their own demobilizing security forces or ex-patriots from Serbia, former-Soviet States, or other African countries without the same commitment to international rules and norms.

This variety of domestic and international circumstances seems to have made today's world ripe for the privatization of security. There is an abundance of international supply of well-trained military personnel, and domestic, regional, and global dynamics have increased demand for their services. Though there are market issues having to do with the sometimes mismatch between concerns with security and ability to pay, the features driving the demand side of this market are unlikely to change. The supply is not as likely to continue (and the question of what will happen when MPRI no longer has an overabundance of retired military personnel is one that should be given due consideration), but this is unlikely to affect the market in the short run. So what are the implications of this development?

Assessing the Implications

In thinking about the consequences of privatization for accountability, two relationships are particularly important. First is that between those firms providing security to foreign entities and their home governments. Do firms operate in a way that supports foreign policy or does their role in the provision of security allow firms to affect national goals? If they support foreign policy, do they do so in such a way as to allow governments to skirt public debate or otherwise evade mechanisms for democratic accountability? Though I will look at a variety of exporters, the prominence of US companies abroad makes this a particularly important policy issue for the United States.

Second, what is the relationship between the governments or private entities paying for security and the companies that provide security to them? Do firms provide security in a way that enhances the political control of the principal or do they use their expertise to affect political outcomes according to other interests? Also, embedded in this relationship is the association between the entity employing private security and the population it represents. Does the employment of private security circumvent the accountability of the entity to its electorate or other responsible party? These relationships, of course, are all connected - particularly when the interests of the firm's home government (or the population it represents) and its employer clash.

According to some analysts, these relationships are more or less complicated according to the nature of the task. In a widely acclaimed analysis of privatization issues in general, John Donahue argues that the nature of the job we want done can help us decide whether private contractors or civil servants will be best. There are basic differences between profit seeking contractors and civil servants: profit seekers, in exchange for a price, deliver a product; while civil servants, in exchange for a wage, agree to accept instructions. If your goals are clear, you care more about ends than means, and it is easy to replace unfaithful agents (i.e., there are competing agents and there is little cost to replacement), contracting makes sense. If tasks are uncertain, it's hard to measure the value of production, you care about means and it is disruptive to switch agents, civil servants make more sense. Privatization of the wrong tasks can lead to deleterious effects on accountability. According to Donahue, then, the privatization of defense tasks should be most successful when the tasks have clear and measurable goals, and when the goals are more important than the means by which they are achieved.

Moving through the categories listed above, one can imagine tasks that are favorable for privatization. Logistics support may be, particularly if one is talking about more mundane tasks like laundry and food services.19 Other categories, however, particularly operational support, and advice and training entail significant concerns about means that should complicate the success of privatization initiatives.20 According to Donahue, privatizing these tasks runs the risk of allowing private firms discretion over policy or an impact on what our public goals are. For example, as I will discuss in more detail below, the Croatian government hired MPRI to help it create western civil-military institutions. The U.S. licensed the company, believing that it was in U.S. interests to educate Croatia on western norms. MPRI's training mission, however, took place while Croatia was involved in a war. What was to stop the trainers, all former U.S. military personnel from giving Croatia operational advice? Indeed, the Croatian army made vast gains in the wake of MPRI's efforts. While the company has denied any involvement in Croatian plans, speculation abounds that in the implementation of an ambiguous goal, MPRI may have affected a political outcome - the Croatian offensive to retake the Krajina from Serbia. Using military firms for this kind of task, Donahue would argue, is likely to lead away from democratic accountability.21 Others argue that this may lead to a change in whose interest the state serves.22

Others focus not the task, per se, but the normative framework in which the actors operate. David Shearer argues that military firms have become popular precisely because they are able to take on the messy tasks of intervention that western militaries are eager to avoid. He suggests that instead of turning away from military companies we should engage them. Shearer's logic focuses on the importance of actors' identities and the norms within which they act. If a firm is well integrated with the identity of its government, it is likely to represent its government's interests abroad. Also, if we create a structure of international organizations and law with which we can effectively negotiate and regulate military companies, we will be able to obtain the benefits these companies provide without sacrificing accountability. This strategy is likely to be successful if the new military companies see themselves as legitimate purveyors of expertise rather than mercenaries. Thus an international system of engagement with these firms would not only create processes of accountability, but would also build on and reinforce an identity that would bias these firms toward abiding by international rules.23

These two analyses suggest different levels of concern about privatization in security affairs and different courses of action. If Donahue is right, we should worry about many of the tasks being privatized. If the agency relationship is inherent in the task, we should try to limit the private use of force and move ambiguous tasks back into the realm of national militaries.24 If Shearer is right, though, it may very well be that the vehicle delivering the services is not as important as the norms surrounding the process. Private security firms may just fit right in as one more arm of the state. If the behavior of military firms is a product of the normative structure within which they operate, we should embrace attempts by private firms to work within international norms. We should also create international structures that reinforce the identity of these firms as legitimate international actors.

Case Method Tracing the accountability issues that arise in different normative and task environments is best accomplished by looking at cases. A task oriented approach would expect that the firms would be most subject to control when the task is the clearest, and most measurable. A norms oriented approach, however, would expect greater control when there are norms in place that define actors and regulate behavior appropriately. By looking at whether accountability is more likely when the task is clear or when the firm identifies with the state (or some combination), we can begin to think about how to best regulate this phenomenon.25

Obviously there is not a great deal of variation in the broad set of international norms that regulate these companies behavior today. There may be variation, however, both historically. How have countries, particularly western, democratic countries such as the US and Britain as well as countries that never quite gave up mercenaries in Africa viewed military firms over time? Does the behavior of firms providing military services vary according to their international legitimacy? Were they more likely to violate human rights, work on the wrong side of international law, etc. during the 1960s and 1970s when they were viewed as less legitimate?

Second, there is variation to be found in the norms between governments and firms in different countries. Generally, we would expect that the closer the identity between the firm and the government as well as the tie between the firm's identity and the government's goals, the more likely we are to see satisfactory outcomes. For instance, one would expect that a firm composed of retired, high level military personnel in states with a strong tradition of civilian control (such as MPRI in the U.S.) should be more likely to operate to the satisfaction of its home government. Firms in states with a rockier history of civil-military relations or those who hire personnel from a variety of different countries and backgrounds (such as EO in South Africa) would be less likely to be accountable to their government.26

With respect to the second relationship, the ability of employers to hold military firms accountable, several different issues arise. The first is whether the employment of these firms may affect politics in the country by empowering (or reinforcing the power) of particular actors. Do private security firms elevate some issues at the expense of others? In accordance with the task oriented arguments above, some argue that this is true in Africa. 27 Sierra Leone's mining interests, always strong, have become stronger still since EO's intervention. Despite the oft-cited argument that EO enhances stability, then, it may have also enhanced the possibility that national interests will be tied to particular economic interests in a way that promises problems for democratization. MPRI, however, asserts that its training programs, if allowed to move forward in Angola and Equatorial Guinea would enhance democratization by spreading norms of good civil-military relations.28 If true, this would support the normative arguments above, that the identity of the players and their interest in abiding by prevailing norms is more important.

Also, what is the outcome when a firm's home government's interests and its employer's clash? Do the firms run the show in the conflicts they are hired for, operating in their own interests? This would suggest that the more military firms operate in the international arena, the more we will see a separate set of interests represented, supporting Donahue's worries and dovetailing with arguments about globalization that suggest this phenomenon will have significant implications for international outcomes. Do firms operate according to the interest of their home government? This would suggest the least international change; perhaps military firms are just new tools for conducting foreign policy.29 Or do firms operate according to the interests of their employer?30 This may suggest a different kind of international change where military power is dispersed to entities according to their ability to pay.31

The issue is further complicated when the employer is a NGO. What is the relationship between state interests and NGO interests? Are issue oriented NGOs that face security problems in a good position to insure that a military firm will give them appropriate services? Who is the NGO ultimately accountable to? Is it their supporters, their donors, the country in which they are citizens, the country in which they operate, or is it some combination? To what degree are NGOs using private security? And are they using these firms to help them secure the safety of their staff or to help them pursue their substantive goals?32

MPRI and EO: Do Tasks or Norms matter for the Relationship between Exporters and Military Firms? Military advice and training missions, because of their tendency to be more ambiguous, are the most likely to present a critical test for task oriented analysis.33 Can a close identity with one's home government ameliorate the accountability issues for these complicated and ambiguous tasks? The obvious cases in point for this question are MPRI and EO. While both are composed of former military personnel (from the US and South Africa, respectively), they have different relationships with their governments. MPRI advertises its connection to the US government and sees itself as promoting the US's engagement vision by "assisting foreign governments in converting their military into western models that support democratic institutions." EO's relationship with the ANC government is quite different. EO's personnel are drawn from the same South African Defense Forces (SADF) that targeted ANC leaders during the apartheid regime. It is networked with a number of other companies, some offshore, and is not seen to be working in close connection with the South African government. I am not willing to come to any hard and fast conclusions about these relationships at this early stage in my research. However, as I suggest in the discussion below published accounts of the US experience with MPRI in Croatia and South Africa's experience with EO in Angola and Sierra Leone, suggest some interesting similarities between these seemingly different cases.

EO was formed in 1989 as an intelligence training team for the SADF Special Forces.34 Its founder, Eeban Barlow, was second in command of the reconnaissance wing of the 32nd Battalion in the mid-1980s and then later moved to work with the SADF's Civilian Cooperation Bureau (CCB).35 The CCB (disbanded in 1990) was famous for its efforts to circumvent sanctions against South Africa. It was also purported to be involved in efforts to assassinate ANC members.36 EO's first major operation was in support of the Angolan government in 1993.

The Angolan government hired EO to regain the oil town of Soya, which had fallen under Uniao Nacional para a Indepedencia Total de Angola (UNITA) control.37 Backed by two Angolan battalions, about 80 EO employees retook the town.38 The government then hired EO to train 5,000 troops, 30 pilots and direct its operations against UNITA.39 Overcoming limited equipment and an incompetent, poor force with outdated training, and quite a bit of suspicion40, less than 200 EO employees turned the Forcas Armada Angolanos (FAA - Angolan Armed Forces) around.41 They drove UNITA forces out of the Northwest strategic centers, retook Soya and reduced UNITA controlled territory by one-third. In November 1994 a new peace was signed, though Savimbi was not present at the signing and urged that all foreign forces be removed from Angola. Eventually the US and the UN agreed and Clinton urged his Angolan counterpart, Jose Eduardo dos Santos, to stop using the EO forces.42 The company was removed from Angola shortly thereafter.

In 1995, Valentine Strasser's government in Sierra Leone followed Angola's lead in hiring EO.43 Strasser had come to power in a 1992 coup and inherited an insurgency by the Revolutionary United Front (RUF). The RUF continued to operate in Sierra Leone and in 1995 had worked with rouge elements of the country's army to take control of the titanium oxide mine in Sierra Rutile, a Swiss-owned bauxite mine, and a large area of diamond mining in the Kono district. By April, the RUF was threatening the capital, Freetown. EO deployed some 285 advisors to assist in the reorganization of the Sierra Leone Military Forces (SLMF). Within weeks, EO had scoured Freetown of rebels and in the next 6 months, EO worked with the (SLMF) to clear the RUF from the Kono diamond mines, and eventually the other mining areas. They also trained self-defense teams to maintain local resistance to the RUF. The RUF agreed to participate in peace talks in January 1996, the first in five years of fighting. Elections were held in March that led to a civilian government led by Ahmed Tejan Kabbah.44 Kabbah's background as country representative for the UN Development Program made him somewhat more suspicious of EO, but he continued to use their help and even authorized an offensive after the RUF reneged on its pledge to sign a peace accord.45 In November 1996 agreement was finally reached.46

The South African government's role in all of this is difficult to discern. Many have pointed to the dilemmas EO poses for Mandela's government. On the negative side, the personnel EO hires were notorious for their deeds against the ANC during apartheid. Some officials have been outspoken in their criticism of EO. Rusty Evans, the Director-General of the South African Foreign Affairs Department in 1996 called EO a "dangerous criminal and destructive force in Africa."47 Also South African officials purportedly searched for a way to end EO's participation in Angola and tried to persuade its partners in the Organization for African Unity (OAU) to take a position against EO and other such firms.48 In May 1998 South Africa did pass the Regulation of Foreign Military Assistance Act that requires authorization from the National Conventional Arms Control Committee (NCACC) for any South African, permanent resident or company registered or incorporated in South Africa to both offer and supply services to a foreign client.49

However, there are many ways in which EO has furthered the interests of the ANC government. It has been argued that having former employees of the SADF employed outside South Africa is a security benefit for the ANC government.50 There is no love lost between some of the EO employees and their new government. It is best to have such highly trained and less than loyal personnel well paid for work outside South Africa than hanging around inside. Indeed, some have argued that South Africa could have done more and sooner to restrict EO's activities had it wanted to. In fact, they go on, EO has contributed to the goals of South African defense by aiding its long time allies in Angola and Sierra Leone.51 Some have gone so far as to argue that EO has carried out unofficial South African policy and a British intelligence report links the ANC government with EO and concludes that the government sanctioned EO's foreign contracts.52 Both EO and the ANC government deny this link. Despite the denials and the new legislation, South African officials admit to "frequent meetings and reciprocal sharing of information between the SADF and EO."53

The legislation has yet to be tested although EO's former CEO argued that, "the crux of the new bill [is] to regulate foreign military assistance to parties involved in armed conflict. So far as I'm concerned, if I'm not going to support any party involved in armed conflict then I don't need to apply for permission or authorisation. If a country like the United States wants to contract me to provide a specific service for them, why would I need to ask the government's permission? Because this is not regulated by the bill."54 In what is argued to be an unrelated development, in the wake of the new bill, EO has undergone significant changes resulting in the reconfiguration of the set of interconnected enterprises and the closing of EO as of January 1999.55 Many of its subsidiary and related firms are still in business and few employees have lost work. Despite denials, this development may be a move to reinstate the company offshore so as to avoid accountability under the new law.

Though MPRI and EO are frequently the two cases compared in the writings on private military firms. However, they are several feature differentiate the companies. MPRI has closer ties to the US policy community, it has a policy against participation in conflict, and it has, up until now, had little experience in Africa. In general, MPRI has a better reputation than EO.

MPRI was founded in 1987 by top-ranked retired US generals.56 It boasts a database of over 7,000 potential employees all having significant experience in the US armed forces, and has managed to bring in over $12 million in annual revenue.57 The company began by contracting with the US government to train (and provide other support for) the US armed forces and has gradually moved overseas.58 Part of what makes MPRI advice so sought after is that this is the same advice that is sold to the US armed forces.59 MPRI has worked for Sweden, Taiwan, Croatia, and the Bosnian Federation. It is seeking work with former Eastern-bloc countries, and entering NATO states, and has contracts pending with Angola, Equatorial Guinea, and Nigeria.

Most noted of MPRI's endeavors has been its experience in the Balkans. The train and equip program in the Bosnian Federation was a corner stone of the Dayton Accords and has received much attention. Because that contract was so closely and clearly negotiated with the US government, however, it is less interesting for examining accountability than the contract with Croatia.

The genesis of MPRI's contact with the Croatian government is disputed. Some have argued that the US sent the Croatians to MPRI. MPRI claims that the Croatians had seen a presentation they made to states eager to get into NATO. For whatever reason, the Croatian Defense Minister Gojko Susak sent a letter in March 1994 requesting permission from the Pentagon to negotiate with MPRI to obtain training in civil-military relations, program and budget.60 Croatia and MPRI signed a contract in September and the State Department licensed the project in December. MPRI then teamed its experts with English speaking Croatians to set up courses in budgeting, leadership development, military ethics, and relations with civilian society. According to a Pentagon official, MPRI personnel were briefed for days before they went to Croatia and closely monitored once they were there.61

However, many have argued that the coincidence between MPRI's presence and the turn around in the Croatian Army's performance and tactics is suspicious. Shortly after MRPI began working with the Croatian Army, the Croats began to experience success in their quest to regain their territory (Serb forces occupied about 30% of Croatian territory when MPRI arrived). In May 1995, the Croats retook the areas to the Southwest of Zagreb, then they recaptured parts of western Slovenia, and finally, in August 1995 the Croatian Forces launched Operation Storm to retake the Krajina region.62 The offensive in Krajina looked (to observers in the region) like a textbook NATO operation and has led to endless speculation about whether MPRI was instrumental in Croatia's operational success.63 "The Lightning five-pronged offensive, integrating air power, artillery and rapid infantry movements, and relying on intense maneuvers to unhinge Serbian command and control networks bore many hallmarks of US Army doctrine."64 There has also been speculation about whether the US shared satellite information with the Croats.65

According to General Soyster in 1995, the instruction "has no correlation to anything happening on the battlefield today."66 MPRI, according to Soyster had far too much to lose to violate the terms of its license by advising the Croats on battlefield strategy, weapons, or tactics.67 Of course, this assumes that MPRI would have lost its license. More likely is the speculation that if MPRI did violate its license, it was with the knowledge of officials in the Pentagon and elsewhere.68 As Zarate argues, MPRI's training (even as it was touted as a long-term venture) was meant to have a positive effect on the Croatian military - otherwise, why would they have landed the job?69

The strongest evidence supporting MPRI's claims that it was not responsible for the Croatian success is simply the fact that there were only 14 MPRI personnel in the field during this time. It seems far-fetched to attribute the Croatian Army's success to those 14 people. Whether MPRI did violate the terms of the arms embargo is something that is hard to know for sure, as is the impact of having a retired US general advising them on Croatian morale.

As it was, the Croatian successes changed events on the ground such that strategic bombing by NATO could push the Serbs over the edge and to the negotiating table - the results of which were the Dayton Accords (and another contract for MRPI). By licensing MPRI, the US retained its neutral status while influencing events on the ground. The assessment of the costs and benefits of the policy is complicated by the Croatian human rights abuses in the Krajina region in the wake of the Operation Storm. It was, up until the latest events in Kosovo, the largest incidence of ethnic cleansing since the end of the Cold War. It is estimated that Croatian forces uprooted between 150,000 and 170,000 Serbs from their homes.70

The rhetoric surrounding government relations between South Africa and EO and the US and MPRI is different. No US officials are calling MPRI criminal. Though some have argued that civil-military relations have been more fractious than normal during the post-Cold War, the US military has not had anywhere near the kind of involvement in domestic politics that the SADF had during apartheid.71 The identification of the retired US officers that work for MPRI with the goals and interests of the United States is high and there is reason to expect that these people are highly integrated and socialized into the norms of patriotism. Their statements lead one to believe they are willing and interested in abiding by US (and international) law. Additionally, the history and culture of the US military biases toward civilian control and a clear separation between what is political and what is military that one would expect would lead these personnel away from improper involvement in other countries. Finally, the US had processes and regulations in place with which to regulate the export of military services.

The South African cases suggest a different relationship. It is unclear if former SADF personnel identify with the ANC government. Some clearly do not. EO's statements in the wake of the bill regulating the export of military services suggest less interest in abiding by the spirit of the law. And, of course, the history and culture of the South African military has been much more politically charged. On top of all, the South Africans had no regulations in place and only passed such regulation last year. There are clearly different normative environments at work - both in the identification of the government and the armed forces and the normative structure in which the government operates. All of these would lead one to expect greater potential for improper actions in the South African cases and more potential problems with accountability.

What is interesting about both cases, though, is the degree to which the firms seem to have operated in the strategic interest of the countries from which they hail. EO has worked for countries traditionally allied with South Africa - even when it meant working against Savimbi in Angola. In both countries reports have speculated that the companies are clandestinely carrying out the foreign policy of the state. More research needs to be done to get a better handle on the relationship between the governments and the firms, but in neither case does the firm appeared to have violated the interests of the state.

In both cases, however, a potential accountability issue has arisen over the degree to which having these firms may allow the state to evade some processes of policy making (or implementation). In other words, to the degree that there seems to be an accountability issue, it appears to arise around the state skirting public debate rather than the firm skirting state control. The availability of these firms and the services they offer may allow governments to avoid traditional processes in the creation of foreign policy. It is simply more cumbersome and opens up more public debate when countries send national troops abroad. Sending a for-profit company is easier.

Some may decry this circumvention of the democratic process and argue that bad policy will come of it.72 Others will no doubt argue that this is a good thing-that avoiding public debate over every foreign excursion will allow more "rational" foreign policy.73 Looking at this development as a policy good or bad, however, may not provide the most useful insight. US policy, for example, may be more "rational" without public debate, but it may also lead to embarrassing and counterproductive associations (like the ethnic cleansing in the Krajina).

Instead, I believe it suggests that Donahue and other privatization analysts may be right in suggesting that the privatization of jobs that are ambiguous are likely to lead to serious political ramifications. That is, we will get different kinds of policy, the process will work differently, and new actors will be empowered as we move down this path. The processes of globalization should only exacerbate this tendency. As we may be witnessing in South Africa, globalization complicates the ability of governments to regulate its firms. The US has had an easier time with MPRI because part of what it is selling is access to the US government. It will be interesting to see what happens as MPRI moves into Africa (if it does - the politics of licensing in Africa have frustrated its efforts so far).

As I began the analysis of potential implications, I contrasted a task-based approach and a normative approach suggesting that these analyses led to different prescriptions (as well as different empirical predictions) about what we should do - engage military companies or restrict their use. Looking at the brief cases above, I believe, provides evidence that norms and tasks matter. These are only two cases, though, the two that are most frequently analyzed (and thus have the most published work available). Understanding how lesser known firms operate in the US and South African context. Exploring how Britain has dealt with these issues (Britain, France and Israel are also heavy exporters of military services) will also be important. Finally, any analysis of the relationship between exporters and their firms will have to also uncover the dynamics between the firms and those that purchase their services. So, I will end by reasserting my hesitancy to come to a firm conclusion!

Conclusion Private security firms offer both promise and peril in the post-Cold War world. On the one hand, they present an additional tool for achieving security. There are an increasing number of instances when sending in the US military (or the British or the French) is simply not a good option. Private firms that offer advice and training to the local population can both deal with security needs and potentially provide education that may enhance movement toward democracy in insecure countries. Private firms may also be able to provide tools to NGOs that allow them to remain in dangerous parts of the world fulfilling a stabilizing role. This promises great benefits to the US and its multilateral view of the future. On the other hand, the accountability issues raised by private security firms (inherent even in states relations with their own militaries) loom large in these new relationships. Privatization in a global economy offers many avenues for avoiding state controls. Only by examining the way in which these firms interact with their governments, foreign governments and NGOs can we understand how best to maximize the promise and minimize the peril. This project offers a step in that direction by seeking to explore the extent of the privatization phenomenon and its implications in the areas where it is most likely to have a harmful effect. By both mapping the general dimensions of the trend toward the private provision of security and examining specific cases to determine its implications, I hope to generate understanding with which we will be able to better evaluate and manage the market for force.

End Notes 1 Some argue that the international community should do the same in Sierra Leone today. See Rubin, 1999. 2 Which is not to say that efficiency is unimportant. It is very important to understand whether these processes create greater efficiency, especially since many analyses of privatization demonstrate that particular market characteristics must obtain in order for efficiency to be increased and maintained. It is also important to know who wins and loses because of these new processes. Some of these issues may bear on my concerns with accountability. 3 See Thomson, 1994. 4 See Mockler, 1969, 1985. 5 The most thorough analysis thus far is Shearer, 1998. On individual cases, see also Metz, 1987 and Reno, 1997. 6 Shearer, 1998, Ch.2. 7 See Donahue, 1989. 8 Note that the foreign payment/national delivery and national payment/foreign delivery cells refer to the same cases from different vantagepoints. 9 The general areas are articulated in Shearer 1998. Note that for purposes of illustration in the matrix above, I have drawn from both military operational support and military advice and training. 10 I do not expect to have complete data on this. I can find better data on the US than other countries, but even for the US the data seems to be incomplete for some services. 11 See Lock, 1999. He mentions the increase of European officers in Latin American states in the wake of the Congress of Vienna agreement and the influx of German military personnel to advise Chiang Kai-Shek following the Treaty of Versailles. Though exact comparisons are probably weak, it is useful to remember that the general phenomenon is not new. 12 Feigenbaum, Henig, and Hamnett, 1998. 13 Training is sometimes a slippery slope to more involvement. Other times, the US is uncomfortably associated with a government of whose policies US citizens do not approve. 14 This is the case even for non-military missions. Many of the "observers" sent to the Balkans in the last several years have been civilian employees of military firms - MPRI sent observers to Bosnia, Dyncorp sent observers to Kosovo. Because of their military background, these observers are better equipped to deal with dangerous situations and may be more attuned to signs of security problems. 15 Hungary, for instance, has hired Dyncorp. 16 Van Creveld, 1991 makes the argument that this trend is occurring everywhere. Indeed, we do see gated communities in the US and private security firms specializing in personal security are also on the rise. These are much more serious issues in states that are weak and/or failing. 17 See Clapham, 1996, p. 256. 18 Dyncorp does have a contract with Hungary. 19 Even in these cases, however, some have argued that the fact that these services are provided in war like situations make private firms unsuitable. See Marano, 1998. 20 Indeed, in a long chapter on the defense industry, Donahue argues that many of the problems with defense procurement might be solved if we moved more of the process back within the public realm. 21 If assertions about doctrinal support are true, MPRI may have worked to the satisfaction its Croatian employer, but not the specifications of the U.S. license. Going outside the bounds of the license may or may not have dissatisfied U.S. government officials (who were generally pleased with the Croatian victory, if not with its means). By evading the process, however, the alleged action would still be considered a change in the process because having a private company (rather than US troops) provide the service avoided public debate about the wisdom of US intervention. Some may view this as a good outcome; others may see it less positively. 22 Feigenbaum and Henig, 1994. 23 The focus on norms accords with an intellectual tradition based in sociology that has received renewed attention by organizational theorists in recent years. Known broadly as the "new institutionalism" a variety of arguments have been made about the way in which social institutions affect behavior contrary to what we would expect from an analysis based on material conditions alone. Most prominent in this line of analysis are works by John Meyer. See, for instance, Meyer, 1980 and 1996. For a good review of the new institutional perspective(s), see Powell and DiMaggio, 1991 and March and Olsen, 1998. See Katzenstein, 1996 for a recent and prominent set of analyses on national security issues. There are actually two different traditions wrapped up in Shearer's logic - one based on liberalism and regulatory norms, the other based on what has come to be known as constructivism that focuses on constitutive norms. For a discussion of the difference between these types of norms, see Kratochwil, 1989. See also the discussion in Katzenstein, 1996, Ch.1. 24 Though this may not be possible. 25 Obviously, it is difficult to know exactly how to evaluate accountability in these cases. I will look at a number of indicators, including whether people in the government satisfied with the outcome and whether the process cuts out players that would be prominent if private firms were not used in order to make an assessment. In some instances, it may be that private firms are simply accountable to different players (rather than being more of less accountable), but this is still important to know. The value of case studies is that it allows one to make these judgements. 26 EO is hardly the low end of the market for force. There are many other actors in Africa with shadier reputations. My focus in this project is on what is new about this phenomenon, though, so I want to look at firms that at least paint themselves as legitimate. 27 See Reno, 1997 for an analysis that support this argument. 28 Interview with Ed Soyster at MPRI, 1 December 98. 29 This is true unless, of course, this new tool allows different kinds of foreign policy. Perhaps part of the reason these firms are flourishing is that they offer mechanisms for governments to do things that they would not be able to do with government troops for political reasons. 30 This issue comes up whether we are talking about another government or an NGO hiring a private security firm. 31 I will select the country cases as best I can, but will be dependent on access to information. There are indications that I can get access to government officials in Croatia and probably Bosnia. Speaking with people in places like Angola and Sierra Leone will be more complicated. For these cases, I may have to rely on area experts. William Reno, for instance, has done a lot of work on this issue in these African states. 32 Preliminary indications suggest that there is much more use of private security firms to deal with staff security than to pursue programmatic goals. At least two NGOs (one relief organization and one environmental organization) have seriously considered using private security in pursuit of more programmatic goals. The book will examine both of these cases in detail. 33 Logistics support may be a better test for the norms argument. Even in a situation where the goals are clear, do the norms under which actors operate matter as well? 34 Pech, 1999, p. 84. 35 Shearer, 1998, p. 41. 36 Lafras Luitingh was another high level EO employee with similar ties. 37 A cease-fire between Movimento Popular para a Libertacao de Angola (MPLA) and UNITA was agreed in 1992 and elections, deemed free and fair by the UN, were held. UNITA lost to MPLA and its leader, Jonas Savimbi refused to accept the results. As fighting resumed UNITA retook up to 80% of the country. Reports suggest that the contract was initiated by oil companies - either Heritage Oil & Gas and Sonangol (Angola's state-owned oil company) or Gulf-Chevron and Pentrangol) See Shearer, 1998, p. 46; Goulet, 1997, p. 427. 38 UNITA retook Soya when EO pulled out. 39 The contract stipulated that EO provide weaponry. The troops trained were from the FAA's 16th Regiment, and the contract was for $40 million. 40 In their previous employment as SADF soldiers, many had worked with UNITA and Savimbi. 41 Hooper, 1997. 42 Goulet, 1997, p. 427; Shearer, 1998, p. 48; Pech, 1999, p. 91. 43 It appears as if the Sierra Leone contract was for mining concessions rather any particular dollar figure. 44 Strasser had been removed in January 1996 in a palace coup and Julius Maada Bio came to power. See Reno, 1997. 45 Shearer, 1998, p. 51. 46 The conflict has since broken out again. See Rubin, 1998 for an argument that EO should be brought in again. 47 Ashworth, 1996, p. 1. 48 Duke, 1996. 49 Pech, 1999, pp. 94-5. 50 Shearer, 1998, pp. 54-55; Rubin, 1997, p. 54. 51 Duke, 1996; Zarate, 1998, p. 102. 52 Duke, 1996; Zarate, 1998, p. 102. 53 Zarate, 1998, p. 102, ftnt 184. 54 Pech, 1999, pp. 96-7. 55 Block, 1998. 56 General Frederick Kroesen (Chairman of the Board), Major General Vernon Lewis, Jr. (President, CEO), General Robert Kingston, General Robert Sennewald, and Lieutenant General Richard West. Eqully impressive are those who have since joined: Lieutenant General Harry E. Soyster (Vice President for International Operations) and General Carl E. Vuono (Vice President and General Manager of the International Group). 57 Cillers and Douglas, 1999; Thompson, 1996; Zarate, 1998. 58 Its overseas ventures are governed by the Arms Export Control Act of 1976 (and the ITAR processes which flow from it) which regulates all foreign sales of defense goods and services. 59 Interview with Ed Soyster at MPRI, 1 December 98 60 Graham, 1995. 61 Personal interview, October 1997. 62 Zarate, 1998, pp. 106-7. 63 Fox, 1995. 64 Cohen, 1995. 65 Eagar, 1995. 66 Ibid. 67 Ibid. 68 Cohen, 1995 reports senior State Department officials admitting that Croatia became our de facto ally - that arms flowed in despite the embargo and top retired American generals were allowed to advise the Croatian Army. 69 Zarate, 1998, p. 108. 70 Cohen, 1995, Danner, 1999. 71 For analyses of the "crisis" in American civil-military relations, see Kohn, 1994; Ricks, 1997. 72 See Marcus Raskin's analysis of the Kosovo crisis. Washington Post, May 7, 1999, editorial page for this logic. 73 This, of course, is the logic behind Theodore Lowi's argument in, "Making Democracy Safe for the World."

Bibliography "The Mercenary As Corporate Executive" African Business December, 1997.

Ashworth, Michael, "Privatising war, by 'The Executives'," The Independent (16 September 1996) 1.

Bazargan, Darius, ""High Risk Business," The Guardian, Monday, September 8, 1997.

Block, Robert, "African Mercenary Supplier To Close; Marks End of Era," Wall Street Journal (10 December 1998) newswires.

Cillers, Jakkie and Mason, Peggy, eds., Peace, Profit or Plunder?: the Privatisation of Security in War-torn African Societies. (South Africa: Institute for Security Studies, 1999).

Cillers, Jakkie and Ian Douglas, "The Military as Business: Military Professional Resources, Incorporated," in Jakkie Cillers and Mason, Peggy, eds., Peace, Profit or Plunder?: the Privatisation of Security in War-torn African Societies. (South Africa: Institute for Security Studies, 1999).

Clapham, Christopher, Africa and the International System: the Politics of State Survival (Cambridge: Cambridge University Press, 1996).

Cohen, Roger, "US Cooling ties to Croatia After Winking at Its Buildup," New York Times, October 28, 1995, p. 1.

Danner, Mark, "Endgame in Kosovo," New York Review of Books Vol. XLVI, No.8, May 6, 1999, p. 8.

Dogan, Rhys and Michael Pugh, "From Military to Market Imperatives: Peacekeeping and the New Public Policy," paper presented at the British International Studies Association Conference, University of Durham, 17-18 December 1996.

Donahue, John, The Privatization Decision: Public Ends, Private Means (Basic Books, 1989).

Duke, Lynne, "South Africa's Ex-Soldiers Becoming 'Dogs of War': Apartheid Commandos Peddling Skills," Houston Chronicle (24 March 1996) Lexis.

Eagar, Charlotte, "Invisible US Army Defeats Serbs," The Observer November 5, 1995, p. 25.

Feigenbaum, Harvey, Jeffrey Henig, "The Political Underpinnings of privatization: a Typology," World Politics Vo. 46, No. 2 (January 1994).

Feigenbaum, Harvey, Jeffrey Henig and Chris Hamnett, Shrinking the State: the Political Underpinnings of Privatization (Cambridge: Cambridge University Press, 1998).

Fox, Robert, "Fresh War Clouds Threaten Ceasefire: Secret US Military Advice Helps Cocky Croats Push Toward Eastern Slovenia," Sunday Telegraph, October 15, 1995, p. 26.

Gormley, William, ed., Privatization and Its Alternatives (Madison: University of Wisconsin Press, 1991).

Goulet, Yves, "Mixing Business With Bullets," Jane's Intelligence Review (September, 1997).

Graham, Bradley, "US Firm Exports Military Expertise: Role In Training Croatian Army Brings Publicity and Suspicions," Washington Post (11 August 1995) A1.

Hartung, William, "Mercenaries, Inc." The Progressive 60 (April, 1996).

Herbst, Jeffrey, "The Regulation of Private Security Forces" Paper prepared for South Africa Institute of International Affairs conference (10 December 1998).

Hooper, Jim, "Executive Outcomes," World Air Power Journal 28 (Spring 1997) 38-49.

Isenberg, David, Soldiers of Fortune, Ltd. Center for Defense Information Monograph, November, 1997.

Kammerman, Sheila and Alfred Kahn, eds., Privatization and the Welfare State (Princeton: Princeton University Press, 1989).

Katzenstein, Peter ed., The Culture of National Security (New York: Columbia University Press, 1996).

Richard Kohn, "Out of Control: The Crisis in Civil-Military Relations," The National Interest, Vol. 35 (Spring 1994).

Kratochwil, Frederic, Rules, Norms and Decisions: On the Conditions of Practical and Legal Reasoning in International Relations and Domestic Affairs (New York: Cambridge University Press, 1989).

Mabry, Marcus, "Soldiers of Misfortune," Newsweek (February 24, 1997).

Marano, Lou, "The Perils of Privatization," Washington Post, January 1998.

March, James G. and Johan P. Olsen, "The Institutional Dynamics of International Political Orders," International Organization 52:4 (Autumn, 1998).

Metz, Steven, "Pretoria's 'Total Strategy' and Low-Intensity Warfare in Southern Africa," Comparative Strategy 6:4 (1987).

Meyer, John W., "The Changing Cultural Content of the Nation State: A World Society Perspective," in George Steinmetz, ed., New Approaches to the State in the Social Sciences (Ithaca: Cornell University Press, 1996).

Meyer.John W. "The World Polity and the Nation State," in Albert Bergesen, ed., Studies of the Modern World System (New York: Academic Press, 1980)

Millar, Stuart, "Soldiers of Misfortune," The Guardian, Monday March 27, 1997.

Mockler, Anthony, Mercenaries (London: Macdonald, 1969).

Mockler, Anthony, The New Mercenaries (London: Garden City Press, 1985).

Pech, Khareen, "Executive Outcomes: A Corporate Conquest," in Jakkie Cillers and Peggy Mason, eds., Peace, Profit of Plunder? The Privatization of Security in War-Torn African Societies (South Africa: Institute for Security Studies, 1999.

Powell, Walter W. and Paul J. DiMaggio, eds, The New Institutionalism in Organizational Analysis (Chicago: University of Chicago Press, 1991).

Reno, William, "Privatizing War in Sierra Leone," Current History (May, 1997).

Thomas Ricks, "The Widening Gap between the Military and Society," Atlantic Monthly, July 1997, pp. 67-78.

Rubin, Elizabeth, "Saving Sierra Leone, at a Price," New York Times 4 February 99, p. A27.

Rubin, Elizabeth, "An Army of One's Own," Harpers February, 1997.

Shearer, David Private Armies and Military Interventions, (Adelphi Paper 316, Oxford University Press, 1998).

Shearer, David, "Outsourcing War," Foreign Policy Fall, 1998.

Sheppard, Simon, "Foot Soldiers of the New World Order: The Rise of the Corporate Military," New Left Review 228 (March/April 1998).

Stephen, Chris, "Doing it for Love," New Statesman and Society 5:184 (January 10, 1992).

Thompson, Mark, "Generals For Hire," Time (15 January 1996) 34-36.

Thomson, Janice, Mercenaries, Pirates and Sovereigns: State Building and Extraterritorial Violence in Early Modern Europe (Princeton: Princeton University Press, 1994).

Van Creveld, Martin, The Transformation of War (New York: Free Press, 1991).

Zarate, Juan Carlos, "The Emergence of a New Dog of War: Private International Security Companies, International Law, and the New World Disorder." Stanford Journal of International Law 34 (1998) 75-162.