Regulation as Surveillance

Regulation has become integral in the neoliberal context of the contemporary nation-state, highlighting the shift from a centralized state to a more pluralized form of governance. Various forms of social and economic life are now “governed beyond the state” (Rose & Miller Reference Rose and Miller1992) and monitored by the state through various forms of regulation. This development has also been identified in the field of security, where a range of state and nonstate security providers operate within a “policing web” (Brodeur Reference Brodeur2010) or within “extended policing families” (Johnston Reference Johnston2003) and enact forms of “plural policing” (Jones & Newburn Reference Jones and Newburn2006; Loader Reference Loader2000) and “twilight policing” (Diphoorn Reference Diphoorn2016). In the context of this pluralized security landscape, much debate centers on the role of the state and the need for regulatory mechanisms to ensure that its legitimacy and authority are not undermined. Policing and security are often understood as operating under this “regulatory state” (Braithwaite Reference Braithwaite2000), which oversees and monitors the activities of other agents. Osborne and Gaebler (1993) have conceived of the state’s role in terms of the metaphors of “rowing” and “steering,” whereby the state steers other bodies to row in a particular direction. This creates a process in which governments are “directing rather than doing” (Shearing Reference Shearing, Wood and Dupont2006:24). Regulation, therefore, is a part of the directing, or “steering.”

In the literature on private security, regulation is defined as the “presence of formal, direct mechanisms of control established with the stated intention of preventing or reducing injustice, corruption, negligence, and incompetence” (O’Connor et al. Reference O’Connor2004:140). The most common form of regulation occurs through the implementation of legislation that ensures that an industry operates within particular legal parameters, often including specific labor laws and training requirements. In terms of the private security industry, O’Connor et al. argue that regulation schemes highlight how the state regards it “as a diversely constituted problematic and acts on these problems through multiple regulatory mechanisms” (2008:204). Regulation is believed to instill a higher degree of professionalism and increase accountability (Crawford & Lister Reference Crawford and Lister2006; O’Connor et al. Reference O’Connor2004). Yet state regulation is also a means of legitimizing the role of private security companies in the first place (Crawford Reference Crawford2006). It represents an “ideological claim about the legitimate place of private security in society and the appropriate form and reach of the state’s role in regulating and fencing-in security markets” (Goold et al. Reference Goold, Loader and Thumala2010:16). Companies, in turn, point to the existence of regulation as a way to legitimize their practices (White Reference White2011).

In addition to state regulation, there are also various forms of self-regulation, involving mechanisms enforced or conducted by voluntary employer associations from within the industry that set particular standards (O’Connor et al. Reference O’Connor2004; Rigakos Reference Rigakos2002; Stenning Reference Stenning2000). In general, self-regulation has been less researched than other forms of regulation and also has been criticized for its lack of effectiveness and credibility, and for being nondemocratic and exclusive (Crawford Reference Crawford2006; O’Connor et al. Reference O’Connor2004; Stenning Reference Stenning2000). But there is also “agency-level regulation,” by which companies regulate their employees by setting their own training standards, recruitment policies, and disciplinary measures. It is this form of regulation that is generally considered to have the most impact on security officers, as will be discussed later.

The existence of these various forms of regulation of the private security industry highlights the fact that regulation consists of various layers and should be regarded as “an assemblage of different, even incongruous, parts” (O’Connor et al. Reference O’Connor2008:204)—what Crawford (2006:454) calls “regulatory pluralism” and O’Connor et al. (2008:208) call a “plurality of regulatory regimes.” O’Connor et al. argue further that since the mechanisms of self-regulation are often guided by the state, these two forms of regulation are somewhat indistinguishable and that employer associations represent what might be called “hybrid forms of governance” (2008:208). I agree that regulation of industry is in itself an assemblage and that studies on regulation should include not only state-based efforts, but also the numerous ways in which associations or companies regulate security officers themselves. Regulation, that is, involves not just preventing or reducing corruption and incompetence, but also controlling a particular labor force that is deemed problematic. For this purpose it relies on what Lyon (Reference Lyon2002) refers to as the “phenetic fix,” a trend of collecting data from human bodies in order to influence, manage, and control them, with the ultimate aim of “social sorting” (Lyon Reference Lyon2002:3) and the classification of individuals in terms of risk (see also Monahan Reference Monahan2010; Haggerty & Ericson Reference Haggerty and Ericson2000). ^{Footnote 4}

But I also argue here that regulation needs to be regarded specifically as a form of surveillance, and that it should be analyzed as part of a larger “hierarchy of surveillance” (Haggerty & Ericson Reference Haggerty and Ericson2000:606) or “surveillance assemblage.” Surveillance studies is a field that has expanded over the past decades, particularly since the attack on the World Trade Center in September 2011, which propelled a range of intelligence and security measures and created what Lyon (Reference Lyon2002) refers to as a “surveillance society.” According to Lyon, “surveillance” refers to “purposeful, routine, systematic and focused attention paid to personal details, for the sake of control, entitlement, management, influence or protection” (2008:2). Much of the research in surveillance studies has focused on the technological side of surveillance, such as CCTVs (see Goold, Loader, & Thumala Reference Goold, Loader and Thumala2013; Lippert Reference Lippert2009; Monahan Reference Monahan2010), ID cards, fingerprints, communication records, and biometrics. This is what Wakefield (2003:xxi) refers to as the “security hardware sector” of the private security industry. However, although not unrelated to these technical aspects, the “manned or staffed services” of the private security industry—namely the actual individuals who perform surveillance-type activities, such as access control—have been given less attention in this field. Studies on private security officers have shown how they engage in various forms of surveillance (Diphoorn Reference Diphoorn2016; van Steden Reference Van Steden2007; Wakefield Reference Wakefield2003), but I want to take the analysis a step further. Throughout my research in South Africa and Kenya, it became evident that under the guise of “regulation,” security officers themselves operate under the Foucauldian “panoptic gaze” of their many masters (see Diphoorn Reference Diphoorn2016). Referring back to Lyon’s (Reference Lyon2008) definition of surveillance, I argue that security officers are exposed to many regulatory efforts that are purposeful, routine, and systematic means of controlling them—that is, that there is “surveillance of the surveillers.”

Examining regulation through the conceptual lens of surveillance, then, allows us to understand how security officers are (literally) being watched and observed in a routine manner and on an everyday basis. Regulation, in other words, is not just a version of steering; it consists of observing and controlling. In the following sections, I will demonstrate that the regulatory mechanisms in South Africa and Kenya have the fundamental aim of categorizing and controlling private security officers and ensuring that they fall under the category of “safe,” rather than “risky,” individuals and “potential criminals.”

The Private Security Industry in South Africa and Kenya

South Africa and Kenya both have high rates of criminal violence, a ubiquity of nonstate policing, and a growing private security industry. South Africa is ranked ninth globally in terms of its homicide rate and it has the third highest murder rate in the African continent (UNODC 2013). It has the largest private security sector in the world in terms of GDP (Abrahamsen & Williams Reference Abrahamsen and Williams2011; de Waard Reference De Waard1999; Singh, Reference Singh2008). In 2014 there were 8,144 registered private security companies and 487,058 active registered security officers (PSIRA 2013–14). ^{Footnote 5} The Private Security Industry Regulatory Authority (PSIRA), a quasi-state body, divides the industry into twenty different categories of security services. Due to this diversity, the types of security officers are also varied, including armed response officers, security guards with varying levels of training, cash-in-transit officers, and so forth.

Kenya is also known for its high rates of crime, particularly the capital city, Nairobi, often ascribed the nickname of “Nairobbery.” In 2003, with 1,395 murders and 9,916 cases of “general stealing” (Ngugi et al. Reference Ngugi2004:18), Nairobi was designated by the United Nations as one of the most dangerous capital cities. ^{Footnote 6} Kenya is also regarded as a country under constant terrorist threat, with the attack on the Westgate shopping center in September 2013 as the most well known. The private security industry has operated in Kenya since the 1960s and has experienced an exponential boom in the last two decades, particularly since the Westgate attack. It is estimated that over two thousand private security companies operate in Kenya, of which only nine hundred are registered, and it is estimated to have an annual turnover of KSh32.2.billion (U.S.$43 million) (Wairagu et al. Reference Wairagu, Kamenju and Singo2004). In 2007 the industry accounted for more than three hundred thousand employees (Mkutu & Sabala Reference Mkutu and Sabala2007:411), and during fieldwork in 2015 industry employees repeatedly stated that this figure had increased to four hundred thousand security officers. Security companies offer services such as electronic monitoring as well as providing bodyguards and security guards, with the latter constituting 47 percent of the industry (Wairagu et al. Reference Wairagu, Kamenju and Singo2004:29).

In South Africa and other African countries private security officers are often armed, but they are unarmed in Kenya. ^{Footnote 7} The question of arming private security companies has been debated for the past decade, and one interviewee described the issue as a “minefield” (Nairobi, April 9, 2015). According to Mkutu and Sabala (Reference Mkutu and Sabala2007) and Abrahamsen and Williams (2005:16), most company owners are against the arming of their guards. I also heard about such concerns. A company manager once stated to me, “When they start arming guys here, I will leave Kenya” (interview, Nairobi, June 10, 2014). However, other companies support the move to arm security officers, especially those working in particular sectors, such as cash-in-transit and alarm response. These owners emphasized the high amount of risk that officers face and the need for additional protection, often referring to the casualties at Westgate and Garissa as examples of the lives that could have been saved if the security officers had been armed.

Yet despite the differences between South Africa and Kenya, there are various similarities that are also characteristic of the industry worldwide. The first is the fierce competition among companies for acquiring new clients and contracts, as stressed by industry employees in both countries. This competition not only shapes marketing strategies, but also how private security owners communicate with clients and how security officers are expected to operate. The second is the presence of unregistered and illegally operating companies. In South Africa these are referred to as “fly-by-nights”: unregistered companies that appear and quickly disappear and are most commonly found in the guarding sector. Between 2011 and 2012, 122 security service providers were labeled “untraceable” by PSIRA. ^{Footnote 8} In Kenya illegally operating companies are referred to as “briefcase companies” and “Juakali.” ^{Footnote 9}

The third commonality concerns the nature of the labor force and the poor conditions of the occupation, as has been highlighted by other studies (e.g., Button Reference Button2002; Diphoorn Reference Diphoorn2016; Rigakos Reference Rigakos2002; Wakefield Reference Wakefield2003). Underpayment, long working hours, and poor training levels are prominent in both countries. And this element is closely related to the fourth, and perhaps most important, similarity, namely the suspicion directed toward private security officers. In South Africa guards are often referred to as “criminals in uniform,” and similar claims are made in Kenya. When certain crimes take place—particularly those related to a failure of private security, such as theft—most people’s initial suspicion is that it was an “inside job.” ^{Footnote 10} Although all employees are considered potential insiders, security officers are the first to be suspected, since their low wages are perceived to make them more susceptible to bribes and corruption. In addition, since most of the security officers live in lower socioeconomic neighborhoods, areas widely regarded to be home to criminals, collaborative efforts are seen as eminently feasible.

In both countries this suspicion is not entirely unfounded—security officers are regularly convicted of engaging in criminal activity, both on and off duty. Throughout my fieldwork in South Africa, I learned of numerous cases in which security officers were involved in criminal incidents, and similar reports have been documented in Kenya (Abrahamsen & Williams Reference Abrahamsen and Williams2005; Nguyi et al. Reference Ngugi2004). Regulation schemes, or “surveillance of the surveillers”—both those led by the state and those within the industry—thus have the ultimate aim of preventing this phenomenon and controlling the “criminals in uniform” through routine and everyday mechanisms.

This article is based on qualitative methodology and data gathered in South Africa and Kenya. Between 2007 and 2012 I spent twenty months in Durban, South Africa, analyzing the role of armed private security officers. My main research method was participant observation, interviews, and focus group discussions, supplemented by secondary-level data analysis. The data in Kenya was collected over a period of four months (2014–2015) in Nairobi, where I also used a range of qualitative research methods, with interviews (ranging from open to semi-structured) as the most prominent.

State Regulation

In South Africa state regulation occurs on several levels and is carried out by various state bodies. The primary regulatory authority is the Private Security Industry Regulatory Authority (PSIRA), a body that is monitored by the Ministry of Police but funded by the industry itself, which operates through the corresponding Private Security Industry Regulation Act (PSIRA). ^{Footnote 11} According to this arrangement, private security companies must be registered with PSIRA as companies providing “security services” and must pay either monthly or annual fees. ^{Footnote 12} In turn, the act specifies how the private security industry must operate and determines forms of (judicial) punishment in the event of misconduct. In addition, the Department of Labour determines the wages and employment standards for security companies, and the Safety and Security Sector Education and Training Authority (SASSETA) monitors security training.

This comprehensive system of state regulation can be understood only in the context of the industry’s history. Although private security companies operated in South Africa’s urban centers before the 1980s, the industry grew primarily throughout this decade during the height of the political conflict. As resistance intensified, the state armed forces needed extra manpower and they outsourced former state functions to the industry through various forms of legislation. The main change in legislation was the establishment of the National Key Points Act (NKPA) 102 of 1980, which stipulated that responsibility for security provision (predominantly guarding) at strategic sites deemed crucial for national security should be transferred to the management/owners of these sites, who in turn hired private security firms (Singh Reference Singh2008).

A collaborative relationship developed between the private security industry and the apartheid state, and this was further strengthened by the creation of the Security Officers Act (SOA) of 1987 and the accompanying Security Officers Board (SOB), which was the birth of state regulation in South Africa. The SOA was “a framework for the extension of the network of a state–corporate ‘partnership’ policing further into civil society” (Brogden & Shearing Reference Brogden and Shearing1993:72). After a period of exponential growth in the industry during the 1980s, there were increasing demands for a formal regulation system. This was primarily for the purposes of monitoring and controlling security officers, who were predominantly black males and seen as potential members of the African National Congress (ANC). To overcome this “political problem,” company owners used their collective contacts within the South African Police (SAP) to set up an informal screening system whereby the SAP would assess potential employees in order to determine their viability for employment by a private security company. As time passed, this informal system was formalized, and the main goal of the SOA was to regulate the employees in the industry through the oversight of the SOB. The SOA entailed compulsory registration with the board and laid down rules regarding disqualification and withdrawal of registration. The SOA was thus the first step toward state regulation of the industry. In this period, however, regulation symbolized a partnership between the public and the private—a unified effort to achieve the same goal (Berg Reference Berg2003).

Yet after the political transition of 1994, the relationship between the South African state and the private security industry was transformed. The postapartheid government regarded the industry as part of the old order and this was particularly true for the SOA, which was seen as a partnership between the industry and the former state (Singh Reference Singh2008). To gain further control over the industry, amendments were implemented to expand the scope of the industry, and this resulted in the birth of the Private Security Industry Regulation Act No. 56 of 2011. Whereas the SOA was regarded as a partnership between the two bodies, PSIRA was conceived as an industry watchdog.

In South Africa, state regulation of the security industry implies that all personnel in the industry must be registered with PSIRA. If a service provider is not registered or does not operate in accordance with PSIRA’s legislation, a charge of misconduct is opened, with the penalties differing according to the case. PSIRA has a broad scope of regulation, which is exemplified in its definition of a security service provider as “a person who renders a security service to another for a remuneration, reward, fee or benefit and includes such a person who is not registered as required in terms of this Act.” Particular parts of the act also attest to its wide scope. The first is PSIRA’s zero-tolerance policy, which dictates that any form of malpractice leads to a charge. The second is the consumer liability clause, which states that any person who knowingly or without the exercise of reasonable care contracts security services and provision that are contrary to the act is guilty of an offense. Consumers are “legally obliged to ensure that the companies they are using are registered” and operate according to the act (Berg Reference Berg2003:187). In 2013 PSIRA was amended by the Private Security Industry Regulation Amendment Bill as a means of further improving the quality standards of the industry. ^{Footnote 13}

However, despite international acclaim for South Africa’s regulation scheme, members of the industry, police officers, and even PSIRA employees are heavily critical of PSIRA. Among my informants, PSIRA—both the organization and the legislation—was regularly described as a “toothless bulldog” that was incapable of enforcement. PSIRA is criticized for being understaffed and inefficient, and allegations of corruption and favoritism among PSIRA inspectors are rife. The criticism directed at PSIRA from across the policing field suggests that the legislation is not always enacted and illegal practices persist. Furthermore, though security officers are registered with PSIRA and pay their monthly fees, many do not feel represented by the organization.

Unlike South Africa, Kenya has no formal state regulation of the private security industry. A private bill was drafted in 2004 to establish a system that would resemble the South African state regulation system, with the National Security Intelligence responsible for vetting the staff. The Regulation Bill was initially to be incorporated in the drafting and implementation of the new Constitution of 2010, which included several acts regarding police reform. Yet although some acts have been passed, such as the establishment of the Independent Police Oversight Authority (IPOA), no real action has been taken regarding the Regulation Bill.

The reasons given for this are numerous. Several interviewees mentioned that since the ratification of the new constitution, other new legislative amendments and bills have been given priority. As one informant stated, “Out of all of the security issues in this country, private security is not the priority” (interview, Nairobi, June 6, 2014). Several other industry employees added that it is not in the interest of the government to implement an efficient regulation system. Several interviewees assert that tighter state regulation would primarily target the “Jua kali” companies, which, they claim, are either owned by or strongly tied to politicians and Members of Parliament. They argue that most of these smaller companies win the lucrative government tenders and that state regulation would infringe on their practices. State regulation would also entail the implementation of better labor conditions for security officers, resulting in less profit for the companies. And it would inherently require a national framework outlining interactions between the industry and particular state bodies, such as the state police—an issue that is particularly problematic in the ongoing process of national police reform in Kenya. And finally, regulation would have to deal with the “minefield” of whether or not to arm security officers, which, as mentioned, is controversial.

Yet although there is no state regulation that specifically concerns the industry, all businesses in Kenya are registered under the Companies Act of Kenya and fall under the Ministry of Trade and Industry (Mkutu & Sabala Reference Mkutu and Sabala2007). Furthermore, all wage issues are reviewed by the Ministry of Labour and the Protective Security Industry Order, which stipulates the regulations that govern people who work in the security industry. ^{Footnote 14} Private security firms are thus registered as “businesses” and are governed by general business laws that apply to all companies. However, this also means that any individual is capable of registering, and thus establishing, a company. There is no control over whether this individual possesses the expertise and skills to actually carry out the tasks. Additionally, there are no standards to “control the quality of any security product or service” (Mkutu & Sabala Reference Mkutu and Sabala2007:402). Due to a lack of formalized partnerships with state law enforcement agencies, there is no standardized procedure to conduct background checks on security officers. This leaves companies, as I discuss later, to perform these criminal checks themselves.

Self-Regulation

Due to an absence of state regulation, regulation in Kenya occurs primarily through self- and agency-based regulation. Self-regulation is carried out by two (rival) employers’ associations: the Kenya Security Industry Association (KSIA) and the Protective Services Industry Association (PSIA).

The members of the KSIA generally include the foreign-owned and larger companies, such as KK Security and G4S. One of the members of the KSIA highlighted that the main goals are to act as a representative body of the industry and to ensure that the companies are “legitimate and legal” (interview, Nairobi, June 9, 2014). The KSIA website (www.ksia.co.ke) mentions thirty members, while interviewees mentioned numbers ranging from twenty to thirty. Aspiring members of the KSIA must submit an official application and pay an initial fee of KSh50,000. The company will then be vetted by two other members (representatives or employees from other companies) and the new member, if accepted, will pay an annual membership fee between KSh40,000 and 160,000, depending on the size of the company (interview with KSIA member, Nairobi, April 10, 2015).

KSIA members who were interviewed all stressed that their association consists of “good companies,” while the rival association, the Protective Services Industry Association (PSIA), consists of illegal companies—the “Jua kali.” In fact, several KSIA members described the PSIA as an “offshoot” of the KSIA, consisting of KSIA applicants who had not been accepted (interviews with KSIA members, Nairobi, April 10, 2015). But while some interviewees emphasized the rivalry between the two associations, others claimed that they regularly collaborated on issues (which I did not observe during my fieldwork), and others stated that they communicated only about matters that concerned the entire industry, such as wage disputes.

According to Abrahamsen and Williams (Reference Abrahamsen and Williams2005, Reference Abrahamsen and Williams2011), the rivalry between these two companies emerged due to different stances on a variety of issues, but mostly in regard to the Legal Notice 53 of 2003, which concerned the implementation of the new minimum wage of private security officers. These authors state that the KSIA endorsed the new minimum wage, while the PSIA rejected it, claiming that increasing the wages of the security officers would make private security inaccessible to smaller businesses or less fortunate households. This issue of wages was also evident in interviews with members from both associations and at their meetings.

The end result is that self-regulation of the industry occurs in two divergent ways, whereby companies adhere to different standards. One example of this difference is what some KSIA members referred to as the “Blacklist.” ^{Footnote 15} Apparently the KSIA website contains a portal wherein all members can access a “list” through a user name and password. According to one owner, this list contains a register of “bad employees”: employees who were fired for misconduct, of which theft is the most common. When a company that is a member of the KSIA fires a security guard on the basis of “foul play,” the former employee’s name is placed on the list. One company owner mentioned that he himself managed this list for his company, while another company manager mentioned that someone from the human relations department controlled this list (interviews, Nairobi, June 9, 2014; June 10, 2014). The idea is that all KSIA members can check it for the presence of potential future recruits. As one company manager stated, “Once you’re on this list, you’re never going to work in the industry again” (interview, Nairobi, June 3, 2015). Some informants mentioned that individuals who are merely suspected of misconduct, but not necessarily convicted, are placed on the list, while others highlighted that it comprises only convicted individuals. If the former is true, it suggests that even blameless individuals might be banished from the industry. ^{Footnote 16}

All of the interviewees who had access to this list spoke about it with pride. One manager described it as a “clever way of keeping our own tracks” (interview, Nairobi, June 10, 2014). However, the opinions about its effectiveness were debated: some argued that it is very well maintained, while one owner mentioned that it is simply an idea that was initially successful, but that the list has not been updated over the last few years (interview, Nairobi, May 15, 2015). But regardless of its efficiency, this list is clearly not just a way of preventing negligence, but also a systematic means of “sorting” out employees and controlling them. None of the security officers whom I interviewed informally seemed to be aware of the existence of the “Blacklist,” which according to one company manager is “precisely the point” (interview, Nairobi, June 10, 2014). Furthermore, companies that are not members of the KSIA (such as PSIA members or companies that are in neither organization) are excluded from this service. The blacklist is thus an exclusionary form of surveillance that is not applicable to all companies. This points toward a clear “hierarchy of surveillance” (Haggerty & Ericson Reference Haggerty and Ericson2000:606) and highlights how individuals are unequally exposed to various surveillance measures.

In South Africa, attempts have been made to implement such a system, but at the time of writing (2015) no such formal system existed despite the abundance of self-regulation efforts. In South Africa, there are at least thirty-eight employers’ associations (Shearing & Berg Reference Shearing, Berg, Jones and Newburn2006), and the oldest one, the Security Association of South Africa (SASA), was established in 1965, when there was no form of state regulation (Berg Reference Berg2003). There have been numerous attempts to merge these groups into one overarching organization, although these umbrella organizations are widely regarded as inefficient. The first attempt was made in 1986, when the Minister of Law and Order encouraged the establishment of the South African Security Federation (SASFED). By 1989 fifteen different associations were represented under this umbrella organization (Grant Reference Grant1989). In August 2003 another group regulatory body was established, the Security Industry Alliance (SIA). Although SIA has established a memorandum of understanding (MOU) with government departments and structures such as PSIRA, it is not regarded as representative by other employer associations.

The large number of associations in South Africa is an indication of the competition in the industry. The rivalry between associations and companies, often initiated and prolonged by personal vendettas, hampers coordination within the industry. Some associations simply exist on paper, some act more as “social clubs” where various company owners and managers regularly meet, and others are quite prominent in determining industry standards. An example of a rather efficient and well-respected employers association is the South African Intruder Detection Services Association (SAIDSA), the leading employers association in the armed reaction sector, which it regulates through its various by-laws and directives. ^{Footnote 17} Although these regulations are not legally enforceable, members who do not comply are “kicked out” of the association and any form of misconduct is reported to PSIRA (interview with SAIDSA administrator, Johannesburg, Aug. 18, 2010). In fact, many employees in the sector ascribe more authority to SAIDSA than to PSIRA.

Yet the main reason for the abundance of self-regulation lies with the perception that state regulation, conducted by PSIRA, is a form of punishment rather than a form of industry representation. As discussed, the Security Officers Act of 1987 (SOA) was the result of an alliance between the state and members of the industry. Back then, the Security Officers Board (SOB) had six representatives from the private security industry (out of the total of ten members), while the current council lacks any representation from the industry. ^{Footnote 18} Many members of the industry feel that regulation has been “hijacked” by the state, that it does not represent their needs, and that it damages the industry rather than protecting it; as the owner of a company said to me, “PSIRA is taxation, not representation” (interview, Durban, April 6, 2009).

Furthermore, during my interviews with numerous company owners, they expressed distrust toward the criminal record checks conducted by PSIRA. When individuals register with PSIRA, a criminal record check is conducted, but this does not catch “unrecorded” criminal activity. Many security officers did indeed tell me about engagement in criminal activity such as drug dealing, both before and during employment, for which they were not convicted and which therefore went unrecorded. Furthermore, checks after registration are conducted only when the registrations are updated, which occurs after further training is completed or if alleged reports of criminality surface. Many members of the industry want to impose regular criminal record checks after registration and want changes to the legislation (Draft Bill) to include such regular checks every five years after registration.

In 2010 the SIA outlined a plan called “Project Sanitize,” which is very similar to the Kenyan “Blacklist.” The aim of the project was to create a shared “criminal database” by regularly screening employees and establishing a database with information regarding the criminal behavior of workers, especially those who were not reported to the police or for whom there was insufficient evidence to charge them. Although numerous reasons were given for the failure of implementing this project, the most common reason was the lack of cooperation from the government, which deemed it “illegal” and a “human rights violation” (interview with SIA chief executive officer, Johannesburg, Aug. 17, 2010). Nevertheless, all of these issues—preference for an employee association over PSIRA, a lack of confidence in PSIRA, and a perceived need to establish a database beyond state control—highlight the fact that state regulation is not necessarily at the top of the “hierarchy of surveillance” in South Africa, and that nonstate forms are often more pervasive, adhered to, and preferred. This refutes some of the claims discussed earlier that self-regulation efforts are not credible and/or effective.

What we thus see in both Kenya and South Africa is rivalry among employee associations that results in a lack of coordination to amalgamate or steer various self-regulation efforts. In Kenya, due to a lack of state regulation, companies have more freedom to impose particular mechanisms, such as the “Blacklist,” despite its potentially illegal nature and its infringement on the livelihoods of security officers. In South Africa self-regulation mechanisms are abundant and sometimes favored, but these are monitored by the state, which prohibits particular actions, such as “Project Sanitize.” These issues highlight the various forms of regulation that overlap, contradict one another, and inhabit different positions within the “hierarchy of surveillance.”

Agency-Level Regulation

In addition to state regulation and self-regulation, private security companies also instill a range of surveillance measures to monitor private security officers, defined by O’Connor et al. (Reference O’Connor2004) as “agency-level regulation” efforts. These mechanisms are rather understudied (particularly in comparison to state regulation), and I argue that they are the most influential for security officers in their line of work. The instilling of control through surveillance is a recurrent feature of the private security industry that occurs during training, registration, and recruitment, and while on duty. As stated by Rigakos (2002:101), these initiatives are “an obsession for security firms.” Due to the limited scope of this article, I will only discuss the main forms of surveillance imposed by companies in South Africa and Kenya.

It starts at the training facilities, where security officers are ideally molded into obedient and utilizable employees (Singh Reference Singh2005). In South Africa, SASSETA regulates the training and the standards of the facilities. In Kenya, there is no uniform training system and companies are left to train their employees themselves. Most companies provide basic training services, while some companies provide no training at all. Although training facilities, such as the Private Security Training Academy (PSTA), are emerging and are trying to establish uniform standards, many companies prefer to have their own training sessions to create their own means to “control” their staff (interview with PSTA trainer, Nairobi, May 13, 2015). In June 2014 the training manager of one of the largest companies gave me a small tour of its main training center in Nairobi. While he was showing me the different facilities, he highlighted that the security training is intended not only to provide skills, but primarily to ensure that security officers know “not to mess around.” He referred to the frequent occurrence of malpractice among guards and stated that the security training is the first step toward “cleaning them out.”

The recruitment procedure is the next phase of surveillance. In addition to satisfying the formal requirements of the job, all security officers undergo some form of psychological testing, for which companies use a range of techniques, including panel interviews, aptitude tests, integrity checks, polygraph tests, and psychometric evaluation. These are used to assess officers’ capabilities, to understand their distinctiveness, to compare them to “the norm,” and to focus on their “moral habits” (Singh Reference Singh2008:54). The range of these checks is dependent on the financial capacities of the company involved.

Yet for all companies the main concern upon recruitment is the criminal record check. In South Africa, approval from PSIRA upon registration requires that the officer does not have a criminal background, yet numerous companies perform their own police checks, often through their own contacts with the state police. Similarly, some company owners check applicants’ financial backgrounds for evidence of previous irregularities, which might suggest corrupt practices. In Kenya companies are responsible for conducting the first criminal record check, although the company owners I interviewed all stated that these checks are not very reliable. They referred specifically to the corrupt nature of the Kenyan Police, who apparently hand out “clean records” in exchange for “payment.” Some companies have therefore implemented additional means to ensure these checks are reliable, such as using their own contacts within the state police. Another company manager mentioned that any potential recruit is accompanied by another higher-ranked company employee when collecting the statement (interview, Nairobi, June 13, 2014). Yet despite these efforts, interviewees complained about the persistent faults of the system.

Hierarchy among security officers is another means of internal surveillance. Hierarchical structures and ranking are used to cultivate obedience and discipline (Diphoorn Reference Diphoorn2016; Singh Reference Singh2008; Wakefield Reference Wakefield2003). Company managers in Kenya and South Africa argue that hierarchy is a normal part of policing and that it determines the quality of the officers’ performance. According to Rigakos, hierarchical structures establish “a role model system” and a “distribution system in which officers are ranked by skills, aptitudes, and experience” (2002:104). Creating a hierarchy displays to the “lower levels” what is needed to reach the “higher levels,” providing the former with goals and incentives. Each company has a different hierarchical system, yet they are generally based on dividing up geographical areas that are managed by area supervisors, who are responsible for monitoring the behavior of their subordinates, often by means of unplanned “spot-checks.”

In addition to supervision and hierarchy, which are informal types of surveillance, private security officers in Kenya and South Africa are exposed to numerous forms of electronic surveillance while on duty. Many companies have installed satellite-tracking systems in their vehicles and a few have “live tracking” facilities, which allow them to monitor the location and movement of vehicles in real time. In addition to cameras, various types of equipment are used to monitor the productivity of security officers. Discussing Performer Guard Patrol Systems, an electronic surveillance system, Singh argues that such mechanisms aim to control the guards’ behavior and exert a “constant pressure to conform” (2005:169).

Managers and owners claim that such surveillance methods are intended to help the officers, particularly when clients accuse them of misconduct. They contend that surveillance processes exist to ensure accountability and transparency, to deliver maximum performance to customers, and to protect the security officers from clients that make unreasonable claims. For most private security officers, however, these various forms of surveillance instill a sense that they are untrustworthy and in need of constant monitoring. A security guard working at a shopping center in Durban described this feeling as follows:

Security guards working at residential homes in Nairobi also voiced their dislike of electronic forms of surveillance that monitor when and where they conduct their patrols. Many said that they felt “watched” and “controlled.” ^{Footnote 19}

I argue that these experiences—of feeling watched and controlled—are not encapsulated within the current approach to regulation in the literature. Rather, I argue that they are better understood when analyzed as forms of surveillance. Furthermore, I argue that the various forms of surveillance implemented by companies during security officer training, the recruitment phase, and on the job have the most impact on the daily lives of the officers. In South Africa many security officers complained about the measures implemented by the companies; very few ever referred to PSIRA and other forms of state regulation, which were more often discussed by company managers. Therefore, analyzing regulation within the context of a “surveillant assemblage” (Haggerty & Ericson Reference Haggerty and Ericson2000) not only allows us to incorporate the various levels of control that are exerted on security officers, but also provides room to examine which regulation efforts target particular levels within the industry. And if we look at the security officers, we see that the informal and everyday means implemented by companies are the most influential, and a conceptual lens of surveillance allows us to analyze that further.

Concluding Remarks

This article presents the results of a comparative analysis of private security regulation in South Africa and Kenya. In South Africa, we see an encompassing state regulatory framework, a vast and diverse amount of self-regulation efforts that are poorly coordinated, and numerous agency-level regulatory mechanisms. In Kenya, a state regulatory framework has been drafted, yet not implemented, leaving the industry to perform its own form of regulation, either through two (rivaling) employers’ associations or by companies themselves.

It has not been my aim to evaluate which forms of regulation are more effective, but rather to provide a comparative analysis of these two countries. One could argue that the South African context is more efficient because of the encompassing state regulatory framework, which is perhaps due to the fact that the private security industry in South Africa is in itself more organized. Yet one could also argue that the self-regulatory mechanisms in Kenya are more efficient, and that the “Blacklist,” for example, is a more direct way of “sorting out” the potentially “risky” individuals. Therefore, the answer to which system is more effective will differ according to who is being asked.

This lack of consensus underlines the myriad of motivations behind regulatory schemas and the need to analyze regulation as an assemblage consisting of numerous layers. Such an approach also highlights the different “hierarchies of surveillance” and the extent to which they are not clear-cut: although state regulation is comprehensive in South Africa, measures enforced by companies are more pervasive for security officers. And in Kenya self-regulation efforts are the most prominent forms, yet they are guided by state initiatives. I should also point out that this article has excluded a range of regulatory mechanisms, such as initiatives undertaken by local police officers, other forms of legislation, such as the Firearms Control Act 60 of 2000 in South Africa (Berg Reference Berg2003), and the role of clients. This suggests that the assemblage is in fact much larger and requires further research, both in country-specific case studies and through comparative approaches. This is particularly the case for Africa, a region that is often overlooked in analyses of regulation and surveillance in general.

Furthermore, in this call for further research, I also encourage more symbiosis between scholars engaged in regulation and surveillance studies, which generally operate as two distinct bodies of work. In this article I have attempted to bring some of these ideas together by analyzing regulation as a form of surveillance. Through this conceptual lens, I have shown how a strong foundation of suspicion toward private security officers exists in both African countries, in which the primary goal of regulation efforts is to exercise control over the private security officers: to ensure that recruited security officers are not “risky” and “criminals in uniform,” but are responsible and disciplined. On the one hand, this suspicion is not strange: as the raison d’être of the industry is to fight crime, criminal activity among employees is seen as a serious problem. The cases of guards working as criminal “insiders” and the continuing operation of illegal and unregistered companies in both South Africa and Kenya highlight the degree to which this suspicion is not entirely unfounded. On the other hand, as argued by Lyon (Reference Lyon2002), suspicion breeds suspicion: the numerous measures installed by the state, the industry, and the companies consolidate and exacerbate levels of suspicion. I am not implying that security officers are obedient entities that lack any form of agency and cannot enact any form of resistance to these measures. Rather, I am arguing that regulatory measures largely influence their work and that regulation should thereby be regarded as a form of surveillance. This approach entails analyzing regulation as a systematic means of controlling security officers, rather than as a set of direct mechanisms that prevent negligence and ensure that companies adhere to certain standards. By examining the “surveillance of the surveillers,” we gain insight into the more informal, systematic, and everyday means of control and thereby gain a more holistic perspective of how the private security industry operates.