Accountability Policy Development for the Private Security Industry
A comparative analysis of the United Kingdom and The Netherlands
Master Thesis – Final Version Leiden University
Faculty of Governance and Global Affairs MSc Crisis and Security Management February 2016 Cohort
Mattijs Dijkstra
Student Number: 1745824
Supervisor: Dr. Joery Matthys Second Reader: Dr. Sanneke Kuipers
Abstract:
Security in the 21st century is more than the absence of threat (Baldwin, 1997). It has grown into an international business where the public provision of security can no longer do without support from the private sector (de Waard & van Steden, 2012). The growth of the private security sector over the past few decades has caused severe implications for governments in matters of oversight, control, and accountability. These issues can be resolved through policy development, however the factors influencing policy development for this specific sector has heretofore been been left relatively unaddressed. This research analysed the development of accountability mechanisms for private security companies, whilst looking through the lens of theories on agenda setting, policy formulation and policy change. Based on the body of knowledge on policy development several factors have been distilled which were then qualitatively tested using a multiple case study design. This led to the conclusion that among others, an advocacy coalition is the strongest factor influencing policy development for the private security sector. The accountability mechanisms for the sector, as well as the industry itself are subject of continuous scrutiny by governments, academics, and other professionals, due to its ever-changing character. Therefore the mechanisms to ensure transparency, accountability and safeguarding public values are permanently under construction.
List of Abbreviations:
ACF = Advocacy Coalition Framework ACPO = Association of Chief Police Officers
ACS = Approved Contractor Scheme
AIV = Adviesraad Internationale Vraagstukken (Advisory Committee
International Questions)
Art. = Article
BSIA = British Security Industry Association CCTV = Closed Circuit Television
CCV = Centrum voor Criminaliteitspreventie en Veiligheid (Centre for Crime
Prevention and Safety)
CEO = Chief Executive Officer
CoESS = Confederation of European Security Services CRC = Criminal Record Certificate
EU = European Union
FOI = Factor of Importance
G4S = Group 4 Securicor
IA = Internal Affairs
IOOV = Inspectie Openbare Orde en Veiligheid (Inspection for Public Order and
Security)
IVP = Integraal Veiligheids Plan (Integral Safety Plan) ISI = Inspectorate for the Security Industry
KNVR = Koninklijke Nederlandse Vereniging van Reders (Royal Dutch Assembly
of Ship Owners)
NGO = Non Governmental Organisation
NLG = Lew Labour Government
NVQ = National Vocational Qualifications PCO = Police and Crime Officers
PSC = Private Security Company
POV = Point of View
PPC = Penal Procedure Code
PvdA = Partij van de Arbeid (Labour Party) SCI = Special Criminal Investigator SIA = Security Industry Authority
SITO = Security Industry Training Organization
UK = United Kingdom
UKAS = United Kingdom Accreditation Services
UN = United Nations
URN = Unique Reference Number
VPD = Vessel Protection Detachment
WPBR = Wet Particuliere Beveiligingsorganisaties en Recherchebureaus (Law on
Private Security Organisations and Detective Agencies)
Table of Contents: Abstract 2 List of Abbreviations 3 1. Introduction 5 1.1. Academic Problem 5 1.2. Objectives Research 9 1.3. Research Question 10 1.4. Societal Relevance 10 1.5 Academic Relevance 12 2. Theoretical Framework 14 2.1 Accountability Mechanisms 12 2.2. Policy Formulation 18 2.3. Policy Change 22 2.4 Factors 24 3. Methodology 25 3.1 Research Design 25 3.2 Case Study 27
3.2.1 Why Multiple Case Study 28
3.2.2 Why UK and The Netherlands 29
3.3 Data Collection 30
3.4 Operationalization 32
4. Analysis 36
4.1. The United Kingdom 36
4.1.1. Historic Overview 36 4.1.2. Moments of Development 40 4.1.3. Factors Applied 43 4.1.4. Conclusion 50 4.2. The Netherlands 52 4.2.1. Historic Overview 52 4.2.2. Moments of Development 55 4.2.3. Factors Applied 57 4.2.4. Conclusion 65 4.3. Comparative Overview 67 5. General Conclusions 70
5.1. Answering the Research Question 71
5.2. Recommendations 72
5.3. Reflection 73
References 74
1. Introduction
1.1.
Academic ProblemFirst of all it is important to clarify the distinction between public and private accountability. In the area of public accountability, or accountability for the public police force, there have been several decades of policy development put into effect already. In general these mechanisms do not directly apply to the private sector. First the mechanisms for the public sector will be discussed, followed by the mechanisms in place for the private sector. This way the differences and similarities between the two will become evident. From there on out the focus will be on accountability for the private security industry. In order to fully understand the functioning of these mechanisms it is important to first establish what or who influences accountability policy development. The origin of various current policies in place will be subject of backwards reasoning in order to understand what urged the development of accountability mechanisms. How policy initiatives come about is influenced by the goals they have to protect and incorporate, therefore this part of the analysis will be of great relevance to the academic debate on policy development.
Foremost, the public sector has several (accountability) mechanisms in place for the policing apparatus. Dubnick (2005) mentions greater transparency, access to impartial arenas, pressure and oversight, and improvements in the quality of government services, as the four objectives of accountability mechanisms. Various oversight organisations, whether or not governmental, have been keeping oversight over the public police for decades already. These have all been involved with their own ‘type’ of accountability. Here the distinction between several opposing types of accountability comes to the fore. Perez (2000) mentions multiple systems of accountability: internal accountability/ external accountability and governmental/non-governmental. External accountability for public police practices is ensured through several local, federal and state actors. The main goal of the collective of accountability typologies and actors is to reduce police misconduct by enforcing the law and increase confidence in the public policing apparatus. For instance public integrity units investigate police misconduct, although they rarely prosecute accordingly due to the lack of authority. Subsequently the different ombudsman offices process and resolve complaints about police practices. One of the problems with external accountability through these mechanisms is the inability to bring about structural change. Although systemic reform is encouraged, the aforementioned integrity units lack the authority to
effectively enforce this in practice. According to Perez (2000) this is the result of what he describes as the ‘accountability paradox’.
“Most law enforcement officials and most of the public welcome the implementation of systems of police accountability. At the same time, however, they tolerate varying degrees of misconduct, making it more difficult to implement reforms (2000).”
In this respect the tolerance towards misconduct entails (minor) violations of individual rights of suspects whilst maintaining overall law and order. It is a public misconception that safeguarding the law cannot be successful without (suspects’) civil rights infringements. The problem here lies with the public acceptance of misconduct; since public denunciation would ‘force’ governments to implement policy changes in turn reforming behaviour. This discussion is where internal and external accountability are going toe-to-toe. The difference between the two has consequences for the development of policies since internal accountability can best be achieved by and within the sector itself, whereas external accountability can best be achieved through nationally developed policies. McLaughlin provides a set of conditions for both internal and external accountability. In the area of internal accountability he highlights the necessity of clearly formulated organisational policies, the presence of reporting systems, codes of ethical standards and cultural ethos, disciplinary regulations, and an inspection regime (McLaughlin, 2005). External accountability on the other hand can be realised through courts of law, police complaints systems, pressure groups and the news media as a public mouthpiece for (im-) proper conduct. These mechanisms are primarily aimed at keeping oversight of public (security) service providers; therefore it is interesting to find out (if, and) in what way the construction of private sector accountability can be compared to public mechanisms. Generally, mechanisms of oversight, or accountability mechanisms for the private sector come about in a different manner and are based on different principles and values, although they are in no way above the existing provisions of accountability already in place for the public police system (Walsh & Conway, 2011). Nowadays private security services are increasingly operating in addition to the public police system (for instance private security for residential areas). The private sector has therefore increased its area of operation from serving the public security services to a more commercial clientele base. However, as Walsh and Conway (2011) argue, the interaction between public-private governance and accountability provides a “major field of research and debate”.
This research will look at policy formation for the (private) security industry in the United Kingdom and the Netherlands. A comparison of the two countries will be built upon two cases each containing country-specific elements to see whether or not there is a difference in current policy standards. If there are significant differences between the two cases, a theoretical framework will be used to explain the discrepancy between both.
Nowadays private security companies are used for an increasing amount of tasks previously the sole responsibility of the public security sector. Even though the monopoly of violence still remains within the mandate of the state, private security providers have taken up the responsibility for tasks such as surveillance, monitoring, intelligence, transport and other facilitating services (Avant, 2004) (Abrahamsen & Williams, 2006). Up until the 1970’s the public police force was perfectly capable of keeping up with the demand for (domestic) security as was expected of the state (White, 2011). The monopoly on the use of violence made the state the sole possessor of adequate tools to provide security. Especially during, but also after the Cold War the demand for security in western civilisations rose faster than the public supply of security could keep up with. On the one hand because the pluralisation of society made combatting crime far more complex than it was before, on the other because augmented attention to feelings of insecurity as a self-fulfilling prophecy, demanded increased forms and levels of security (White, 2011). The private security industry was able to fill the gap between demand and supply. Governments allowed private security companies to supply part of the security services on contractual basis, which can be seen as a partial shift of function from the public to the private sector. Vertically organised state structures made way for complex horizontal networks, which David Osbourne (1992) describes as state function that shifted from ‘steering’ to ‘rowing’ (Abrahamsen & Williams, 2009). The privatisation of security by outsourcing services to the private security industry through market mechanisms ultimately resulted in the commodification of security (Krahmann, 2008).
Due to the increased presence of these services even in everyday life the implications for societies as a whole need to be researched in order to ensure transparency and responsible practices (Stenning, 2000). In the first place there is a notable rise in the outsourcing of security services to the private sector, in particular from the 1980’s onwards employment statistics have shown a steady increase in private security personnel in the United Kingdom, whilst the number of public police staff has remained more or less the same (Krahmann, 2009). Economic figures
confirm the aforementioned increase in the business, since over the last fifteen years the industry has quadrupled its turnover by performing tasks such as patrolling of public-private property (shopping malls, universities, train stations and airports) office buildings, surveillance and numerous other tasks (CoESS, 2013). As a result, the private sector has taken over a large portion of the security market from the public police apparatus through measures of privatization.
The development of accountability mechanisms varies in accordance with the context in which they are constructed. Therefore this section will start by explaining how general accountability is constructed in historical perspective. From there on out a summary is given of the different drivers behind accountability mechanism development in both the UK and the Netherlands. First, accountability has played a key role in the shift from monarchical rule to a representative government system (early democracy). Mechanisms to ensure ministerial responsibility put the centre of attention onto political behaviour making officials accountable for everything that happened under his or her jurisdiction (Dubnick, 2005). However, over the years the demand for accountable behaviour has spread from the political domain to public administrators and organizations as well. Primarily the goal of improved accountability was to increase public-sector performance (Dubnick, 2005). One of the main drivers of accountability development is the public desire for government officials to be held responsible and accountable for their decisions. This form of accountability can be seen as instrumental accountability based on a principal-agent relationship (Dubnick, 2005), in which the public is the principal delegating authority (through elections) to the government as agent to act in their best interest. However, for non-governmental organizations (NGOs) democratic motivations such as re-election are not the number one priority and therefore the question as to whom they should be accountable invokes different drivers behind accountability mechanisms. John Clark (1991) argues whether NGOs should be accountable “to their board of trustees, to their governments, to donors, to their staff or to their project partners” (Najam, 1996). All of the aforementioned actors can influence the development of policy concerning accountability of NGOs and with that policy development for responsible practices of the private sector.
1.2. Objectives of the Research
The objective of this research will be to determine (if and) what factors influenced policy development concerning accountability for the private security industry, in the United Kingdom and the Netherlands, during the period 1990 – 2015. In order to do so the differences between the United Kingdom and the Netherlands in their accountability mechanisms and licensing operations for private security providers will be analysed in light of academic theories by (amongst others) Kingdon, Sabatier, and Jones & Baumgartner. To ensure feasibility multiple cases will be selected, which will be embedded within the two countries earlier mentioned. Due to their differences in governmental organization a comparison of these two might provide some valuable insights into the governance of private security companies. However, since they are both European countries I do expect to find some similarities in overarching transnational principles (for instance, EU directives). Moreover, since the industry has been around for almost a millennium in the UK, there is a vast amount of literature available on the subject. The Netherlands, albeit a younger in industry age and documentation, is more accessible for me as a researcher.
The private security industry has expanded rapidly over the last few decades (Avant, 2004) (de Waard, 1999), which has several consequences for public private partnerships. The expansion has subsequently raised a desire for transparency of the industry resulting in an increased societal demand for governmental attention to safety and security issues (Bacon & Samuel, 2009) (Brinkerhoff & Brinkerhoff, 2011). Therefore accountability mechanisms are not only relevant for governance related matters of responsibility but also for public perceptions of norms and values (Stenning, 2000). In this respect the objective of the research will be to what extent accountability mechanisms or other factors (in the UK and the Netherlands) are in place and what the (societal) consequences are for the development of such a public private partnerships (Bacon & Samuel, 2009) (Sadran, 2004). It is expected that turning events (or critical junctures) where both the public police as well as the private security industry came short in the provision of security will play a crucial role. The conclusion of the research will entail future expectations of policy and industry development and maybe even policy recommendations.
1.3. Research Question
The question was formulated by using the funnel technique (Devroe, Lecture 5 - Case Study Design, 2016). Starting from neo-liberalism as a turbulent time for the shrinking role of states. Increased private involvement in public interests posed several problems for the accountability of non-governmental organisation (NGO) practices (Kamat, 2004). How the different regimes of accountability in both countries are constituted, might work together, complement each other, overlap or contradict one another will have to be clear at the end of this research. In particular the formulation of accountability mechanism policies and the motivation behind this will be a major theme throughout this research. One of the expected results of this research is that the different organisational structures of the UK and the Netherlands will provide an explanation for the differences in policy development. Subsequently theories concerning policy development, public opinion, and the organisation(s) within the industry are expected to be of influence. The analysis will be aimed at answering the research question: What factors in the United Kingdom and the Netherlands influenced policy development concerning accountability for private security companies, during the period 1990 - 2015?
1.4. Societal Relevance
The next section will elaborate on what security exactly is, how it is offered and how we want these companies to be accountable. Is reality like Christopher Pollitt (2003) puts it and ‘can we just not get enough’ security, or is it more in line with Mark Bovens’ description of a ‘garbage can filled with good intentions’ (2005)? In order to decide on the societal relevance for this research it is important to first elaborate on what security is exactly and how the provision of it benefits or harms society, or in short: how is it offered? This will be followed by a brief overview of the international developments after the Cold War, taking the role of NGO’s into account. Subsequently the accountability mechanisms involved for those NGO’s will later be assessed in light of their contribution or erosion of society’s security norms and values.
First, what exactly fits under the umbrella of security? Emma Rothschild (1995) in her article describes different ‘entities’ on multiple levels ranging from local to international involved in defining security, being: individuals, nations and “systems” (or organizations). The wide range of actors and entities involved with constructing a definition of security results in numerous concepts of security. Therefore Rothschild (1995) argues the various concepts of security to be:
“Diffused in all directions from nation states, including upwards to international institutions, downwards to regional or local government, and sideways to nongovernmental organizations, to public opinion and the press, and to the abstract forces of nature or of the market
(Rothschild, 1995).”
However, for the sake of this research the concept of security will be demarcated to the nation state and individual security. On a theoretical level however the concept of security is defined in a different dimension. The COT, Institute for Safety, Security and Crisis Management argues in line with Rothschild (1995) that security is a ‘contested concept lacking a general agreed upon definition’ (COT, 2007). However, the COT (2007) does acknowledge that the aforementioned definition of security has shifted from primarily a nation-state affair towards a multiple-level concept as well. In this respect we just cannot get enough, with new threats and actors joining the field of security. Due to an increase in actors, (types of) conflicts and academic contributions to the debate there has been a both a substantial broadening as well as a deepening of the concept(s) of security. Especially after the Cold-War deterrence politics and (mostly) globalisation have had a strong influence on the definition of security (COT, 2007). New concepts and definitions are the result of academic, economic and societal developments in modern society. An adaption of norms and values results in the re-evaluation of what can be considered to be a “the absence of threats to acquired values (Baldwin, 1997)”, since both the concept of ‘threat’ as well as of ‘acquired values’ are debatable nowadays. New conflicts primarily involve the cross-border perils of for example ecological threats as a result of global warming, transnational terrorism, and numerous international conflicts (for instance in Ukraine and the Middle-East). Finally, the increased number of actors involved spurs redefinition of security since NGOs, transnational advocacy networks and countless other non-state actors all contribute to the norm-database upon which the concept of security is to be built. Subsequently, the involvement of non-state actors necessitates a review of the rules of the game, to ensure the good intentions will not go to waste. Where previously the (national) security focus was on nation-states based on a military-realist paradigm (COT, 2007), mechanisms of control and oversight were based on the military (and state) as primary security services provider accordingly. Nowadays involvement of new actors such as PSC’s results in contested mechanisms of oversight and accountability for these same services were heretofore solely provided by the state.
A major turning point for the security industry was after the Cold War. As Buzan (1993) argues the security agenda after the Cold War consisted of an increased amount of issues, priorities and topics, covered by a variety of non-state actors. Rather than the previously state dominated security agenda where war and military force (or the monopoly of violence for states) were the principal determining factors. The widening of topics covered by the security agenda went hand in hand with a change in security ideology and policy image. Traditionalist viewed security as international conflict and the waging of war, whereas wideners such as the Copenhagen School and the securitization theory advocated the inclusion of several environmental and societal topics, without losing intellectual coherence (Buzan, 1997).
1.5.
Academic RelevanceSeveral points of view will be discussed to explain the different drivers behind policy development, in particular policy development pertaining accountability mechanisms for the private security industry. This in turn is the knowledge gap. There have been multiple studies addressing policy development by amongst others Sabatier, Kingdon, Baumgartner & Jones. But none of the authors describing policy formulation or change has specifically addressed the development of accountability mechanisms for the private security industry. Since the industry is ever growing it is however of significant importance to address how it can be regulated best. Therefore different theories concerning policy development and change will be applied to accountability mechanisms in order to explain the dominant drivers behind the creation and enforcement of these mechanisms. The focus here will not lie with accountability mechanisms themselves; rather they provide context and purport for understanding the regimes for policy development. Authors such as Bovens, Walsh, and Conway describe the significance of accountability development in light of the changing character of security services. Since the theoretical notions of what accountability should look like are thoroughly discussed, this research will focus on how these notions relate to another extensively researched subject: policy development. The link between these two fields of study will consist of a set of factors from the extensive body of knowledge on policy development specifically relevant for the private security sector. These factors will be aimed at clarifying what policy makers concerned with accountability mechanism development could and should take into account when drafting policy
proposals, in order to ensure efficient and effective cooperation between the public and private sector in the 21st century provision of security, also known as plural policing.
At the end of the research it will be clear which factors are of significance for the development of accountability mechanisms. The knowledge gap between policy development theories and sector specific trends will be filled by the results and recommendations of this research. This could fuel the debate for academics to reconsider, deepen, or expand their views and theories on policy development in light of the global changes in the security field.
2. Theoretical Framework
2.1. Accountability Mechanisms
Aaron Wildavsky: ‘If accountability is everything, it may be nothing’
The academic discussion in the literature on accountability is pre-occupied with definitional implications concerning normative matters for accountability standards and the assessment of actual behaviour in light of these norms (Bovens, 2010). However, when looking at accountability as a social mechanism the focus shifts the spotlight from has the actor behaved responsibly to whether or not an actor can be held accountable? The latter is the form present in the UK (and various other European countries), where accountability is seen as a (social) mechanism rather than the American interpretation as a virtue. The two forms of accountability as explained by Bovens (2010) are briefly shown in Table 1.
Table 1 – Two conceptualisations of accountability
As a Virtue As a Mechanism
Actual performance of officials. The way institutions operate and function. Standards for good governance. Focus on relations & social control. Inappropriate behaviour & organisational
deviance are indicators.
Control & feedback deficits are indicators.
Source: Bovens (2010)
These concepts of accountability are two of many different forms. Subsequently they are too vague as of right now to test whether or not they are present in the two cases chosen for this research. However, since a detailed investigation into the effectiveness of both regimes is not the focus these definitions are used here to clarify any definitional ambiguities. Koppell (2005) adds five conditions related to accountability as a virtue, which could provide the basis for formulating hypotheses based on these characteristics; transparency, liability, controllability, responsibility, and responsiveness (Bovens, 2010). These elements could be attributed to good governance practices as well, according to Walsh & Conway (2011), who argue that the interaction between governments and external actors is based on similar principles as accountability mechanisms. Therefore governance and accountability are inherently connected, they argue (2011).
Walsh and Conway (2011) explain the scientific relevance of accountability studies by noting the extreme makeover of policing services, due to increased demands by an “increasingly diverse, technological, urbanized, globalized, mobile, sophisticated, rights-conscious and knowledge-based society” (Walsh & Conway, 2011). Among several changes taking place the
past two decades they highlight the growth of the private sector in policing activities and with that the rise of plural policing on the one hand. Whilst on the other there is a notable growth of civilian oversight mechanisms and augmented (national) centralization of judicial oversight, whereas previously these mechanisms of oversight were a regional or even local responsibility. The former is a perfect example of how external accountability ‘infiltrated’ internal oversight mechanisms, that previously dominated (police) governance practices and control mechanisms.
For instance legal procedures initiated by civilians can now instigate policy alterations and with that revise behavioural guidelines, since judges and courts (in western democracies) can shape police practices through judicial rulings instructing policy change. In short, the complaint handling mechanisms and judicial oversight can both be seen as the previously explained form of passive accountability giving. However, it is even more so desirable to prevent these mechanisms from being invoked in the first place. This can be realized by ensuring proper measures and guidelines for police behaviour are in place. International legislation such as numerous UN charters and (EU) Council directives play an important role here (Walsh & Conway, 2011). The jurisprudence as a result of (international) legal rulings could prove to be an important driver behind accountability mechanism policy development. In order to effectively analyse accountability mechanisms and their development an unambiguous definition of accountability is desired. For public accountability Bovens (2005) provides a well used definition: “Accountability can be defined as a social relationship in which an actor feels an obligation to explain and justify his or her conduct to some significant other”. Therefore this research will look at the role of social relationships between private and public actors in the private security sector. The obligation mentioned can be deduced to the establishment of common norms and values, and abiding by them in accordance with guidelines set out by an authoritarian body of oversight. For the private security sector this body of oversight can either be an institutionalized non-governmental organization, or the respective governments themselves. This will be discussed later in this research. First the developments, size and role of the private security sector will be closely looked at in light of its relation to the theoretical concepts of accountability and policy development.
Over the past two decades private security companies have come to be used for an increasing amount of tasks previously the exclusive responsibility of the public security sector. Even though the monopoly of violence still remains within the mandate of the state, private
security providers have taken up the responsibility for tasks such as surveillance, monitoring, intelligence, transport and other facilitating services (Avant, 2004) (Abrahamsen & Williams, 2006). In the Netherlands the yearly turnover of the private security industry in 2013 was €1.3 billion. In that same year a total of 28.550 private security guards were employed (CoESS, 2013) this means an increase of employment compared to the data from the Heliview (2011) research. The report mentions general guarding (62%), airport security (16%) and monitoring and remote surveillance (12%) as the primary tasks with which the industry is concerned. Due to the increased presence of these services even in everyday life, the implications for societies as a whole need to be researched in order to ensure transparency and responsible practices (Stenning, 2000). In the Netherlands there is little to no research done into the effectiveness of accountability for the private security sector. Although this research will not analyse matters of effectiveness, it does aim to provide a holistic image of the industry and the accountability mechanisms involved, resulting in a consistent analytical framework eligible for future empirical research into accountability practices (Bovens, 2010). Earlier work by Bovens (2005) describes accountability as an ‘evocative political word’ that can be used to sway political debates portraying trustworthiness and justice. When looking at policy proposals in the United States for instance, over fifty proposal-titles included the word ‘accountability’, only to never be mentioned again throughout the actual proposal. This underlines the political power the word itself has, rather than its contribution to the (analytical) debate on good governance (Bovens, 2005)
The lion’s share of the literature has been polarized between research into governance of the police (for the most part) and criticising the lack of sufficient accountability mechanisms for the private sector. The form of plural policing we find in the 21st century has been underexposed in the literature (Stenning, 2009). However, looking at plural policing could provide valuable insights into redesigning accountability mechanisms. One of the features of plural policing is to make policing a ‘problem of the commons’. Instigating and supporting neighbourhood watch initiatives or community advisory boards are fine examples of delegating policing activities to ‘regular’ citizens, although they do create a whole new range of issues to be dealt with in light of accountability mechanisms. For instance the composition of an advisory council, or the legitimacy of behaviour by neighbourhood watch members, complicates governance of the policing apparatus by extending responsibilities to private entities, which generally do not fall under the jurisdiction of (existing) public police accountability mechanisms (Walsh & Conway,
2011). This mainly entails certain types of activities. For instance Hoogenboom (2006) explains in his article that previously public police responsibilities such as supervision, enforcement, upholding order, and criminal intelligence are more and more becoming subject to a hybrid form of policing where public private partnerships are the rule rather than the exception. In this respect involvement of the private industry would fit well into this debate. Subsequently, all the new actors joining ‘the game’ might not fall under the legislation at hand, therefore plural policing could require legal revision of current regimes. Hoogenboom (2006) mentions the ‘hydraulic principle’ where dirty work is outsourced to private security companies in areas of intelligence gathering without an official court order. This is a perfect example of why there is a necessity to create and improve accountability mechanisms. In addition, in an article from 2009 minister Hirsch Ballin (Justice) and Ter Horst (IA) stress the need for improved mechanisms of oversight for the private security sector. Private security companies should be under the direct supervision of the (local) public police. Prior to granting a license to a particular company a (re-) liability check has to ensure whether or not this company is eligible and reliable enough to operate under public police mandate. However especially control efforts during security activities are of insufficient quality, and therefore in need of improvement. Amongst these control efforts actively reporting back to the public police, respecting privacy rules (for civilians), and abiding by public law are the most problematic. Especially the respect for privacy is debatable in private investigative practices (Rechtenmedia, 2009). The latter value has been of great importance in the accountability discussion, as will become clear later in this research. Not because this particular value is discussed, but because it is part of the governments responsibility to safeguard public liberties and protect these values from infringement. The last mentioned issue is of great importance to the public and its opinion of the private sector and eventually could lead to policy formulation through advocacy coalitions as will be elaborated on later as well.
2.2. Policy Formulation
The debate on how policies in general come about is fuelled by various academics that all have their own take on the topic. This section will elaborate on how policy formulation comes about, what factors influence policy formulation (and policy change), and provide some insights into where the competing theories overlap or contradict.
There are different theories on why policy formulation occurs and by which paths. Paul Sabatier (1998) provided a version of the Advocacy Coalition Framework (ACF) in which a multiplicity of actors from various organizations impose influence on the policy formulation process in particular concerning topics that require high technological expertise over a longer period of time (Sabatier, 1998). He describes the interaction between relatively stable parameters such as: the basic attributes of the problem area, distribution of natural resources, (fundamental) socio-cultural values and the constitutional structure. Furthermore he highlights exogenous factors such as the changes in: socio-economic conditions, public opinion, systemic governing coalition, and policy decisions and/or impacts from other subsystems. Sabatier’s ACF model pays special attention to policy learning where fundamental long-term policy beliefs and behaviour, influenced by new experiences or information, changes to form new objectives, thoughts and beliefs. This could eventually lead to new policy formation or more accurately put: policy change. According to Sabatier (1998) five principles have to be taken into account when considering the presence of an ACF. First of all there has to be a modicum1 of rationality addressing the role of technical information on the magnitude, facets, causes and impacts of a certain problem. Second, the timeframe involved has to be at least seven years or a decade. Sabatier argues that this rather specific demarcation is the minimum timeframe to assess policy effectiveness and implications. Third, one of the most complex elements of the ACF is the existence of policy subsystems upon which Baumgartner and Jones (1991) elaborate even more. These may overlap on another and/or vary in formation and structure, since the members of a subsystem may originate from numerous backgrounds other than merely political. Fourth, the different levels of government ranging from local to regional, national and international. These in turn will influence the actors involved within the subsystems. Finally fifth, the similarities between public policy programs and believe systems can be interpreted as ‘the ability to map beliefs and policies on the same canvas’ providing a viable lens through which the influence of
individual actors and technical information can be assessed (Sabatier, 1998). The aforementioned conditions for an ACF will be part of the analysis when deciding what factors influence policy development for accountability in the private security sector. However, Sabatier is not the only author whom developed a policy formulation theory. There are several others, with both quite similar views as well as some fundamental differences.
In this respect Kingdon’s agenda setting theory provides a more abstract view, roughly dividing the policy process into three streams: problems, policies and politics. In the literature this is known as the Multiple Streams Framework. The authors’ definition of ‘problems’ comes down to a situation where a large group of people either in the form of advocacy coalitions or ‘regular’ people in organised form want to change the situation by motivating policymakers to do something about the predicament they are in (Larkin, 2012). This difficult situation may be the result of a sudden event such as a natural disaster or a more deeply rooted long term problem, which could have been suppressed for a long time (Birkland, 1998), but has now reached it’s (social) boiling point causing the current situation to be unsustainable, opening up a window for policy development. In order for a policy window to be taken advantage of two other conditions have to be met apart from the opportunity itself. The ‘policies’ mentioned by the author are the proposals, and the ‘politics’ can be seen as the process for policy development. First, there has to be a problem, moreover; it needs to be recognised as such. Only when a problem is recognised as in need of change it is possible to reach sufficient support for policy development. Secondly, when the problem is clear, policy entrepreneurs whom share the same values or goals will start to develop policy alternatives addressing the issue. The actions of various policy entrepreneurs can be described as surfers waiting for the big wave (Hill, 2013), meaning all conditions have to be exactly right to spring into action and develop that specific improved policy. These so called policy entrepreneurs take into account the intrinsic values and goals of the coalition or institution they are part of. This simultaneously is the sole boundary to policy development since no (in-) formal institution will develop any policy beyond their interest or capabilities (Larkin, 2012). In other words this means that policy advocates for the private security sector are not interested in promoting environmental policies for instance.
Apart from the institutional boundary, the possibilities of policies in general are endless and considered as a ‘policy primeval soup’ where actors pitch ideas, denounce them or combine them with others, all to reach a general consensus (Kingdon, 1995). Usually an advocacy coalition has
already developed a certain idea for action. That idea has to be prepped for policy implementation, a process that might already be in development for years before even coming to the table or onto the (political) agenda. Third, the political stream is more complicated since it depends on the region or country involved in making the decision. It is more than likely that when a policy window occurs various different organisations, institutions or coalitions will suggest multiple solutions. One of the most vital factors in deciding what kind of policy change might fit the problem best is societal support. Depending on the place, time and society in which a certain event occurs different solutions might be suitable. The morals and values dominant in society determine to some extend the severity of the perceived problem at hand and with that the need for attention to, or solving of the problem. Advocacy coalitions have to ‘lobby’ of ‘venue shop’ for (societal) support in order for a certain policy idea to be put on the agenda in the first place and eventually maybe even implemented. Important to note is that the theoretical distinction between parties involved in either stream in practice is not as strict as depicted in theory. Policy actors in this third stream evaluate and debate the various alternatives and options formed by policy entrepreneurs. They consider not only the values and goals of the involved policy entrepreneurs, but various other political features such as elections and public opinion as well (Chow, 2014).
Even though the ACF provides a far better explanatory framework for policy change than it does for policy formation, it can still be applied to the formation process. Weible et. al (2012) distinguish three reasons for change in the current policy environment: events, learning and negotiation/cooperation. For he first event to instigate policy change there has to be a crises or event causing (socio-economical) unrest, either by one or more big events or the accumulation of several smaller ones. For the second path to be of significant influence on policy change there needs to be what Weiss (1977) calls ‘learning as sedimentation over longer periods of time’. The build-up of lessons learned will lead to policy improvements in the long run. The final path for policy change has nine conditions according to the ACF, which have to be met in order to realize policy change2. The factors for policy change show that policy learning is of great influence in forming (new) policies. The way this comes about in practice can be explained using the policy process or cycle.
2 Weiss et al (2012): a “hurting stalemate”, representative composition, leadership, consensus decision rule, adequate funding, commitment to negotiate, focus on empirical issues, trust and the lack of alternative venues.
The policy process can be interpreted from different points of view. One of the possibilities is to look at the process from the single policy perspective focusing on a specific stage in the policy cycle (as shown in Image 1). It has to be noted that the policy cycle is not made up of individual stages fluently flowing from one into another. Rather they overlap, check and check back to one another. Moving forward through the cycle might be the goal for policy advocates, however in
practice it is a two steps forward, one step back process. In addition, to this point of view Kingdon’s Multiple Streams Framework would perfectly apply, in combination with Sabatier’s ACF. When looking at the policy process through this framework it becomes evident policy formulation is the result of conflict, where two or more coalitions compete for their policy proposal over longer periods of time (Weible, Heikkila, deLeon, & Sabatier, 2012). It is important to understand the function of policy subsystems. These are made up of government officials, stakeholders from advocacy groups, and scientists all focused on the same issue or area. As an individual they might not have that much influence on the policy process, but by collaborating both horizontally as well as vertically they understand the limitations of policy possibilities, share information, and are able to form an advocacy coalition capable of instigating change. The identification of a policy subsystem and with that an advocacy coalition is a first factor of importance (FOI) for determining what to test for in the analysis. Since policy subsystems have to function within a political system the governmental organization influences the policy process as well.
An advocacy coalition of any kind is based on a common set of values, beliefs, and interests. The number of actors can vary, just as the frequency of (high-quality) interaction that can be distinguished as an important factor shaping the effectiveness of the coalition. Since interaction helps establishing common ground and desired (policy) outcomes either through conflict or consensus. Over time different means, resources and goals will be discussed amongst the members of the coalition ultimately resulting in a policy proposal. Note that this is not a clear-cut workable policy proposal for politicians ready to implement. Rather it is a consensual idea based upon common interest and values feasible with the available resources within the coalition. In practice, no eventual policy (proposal) will fully encompass the coalitions interests.
Consensus is another important attribute of the advocacy coalition, since common means and values could best be achieved by bartering or through unanimity. On the one hand it can be achieved through horizontal cooperation between the actors, based on transparency and evenness. On the other however, vertical hierarchy would ensure compliancy and with that efficiency of the coalition as a whole. As long as the end result is a positive zero-sum game for the actors within the coalition, they could be found willing to acquiesce, giving up a certain degree of authority and autonomy (Rhodes R. , 2006). For the security industry the sharing of power has created powerful networks over the last decades, especially in the UK, taking over governance responsibilities by self-regulating and controlling the industry. Public private partnerships have prevailed over hierarchy in the process of regulation by marketization (Rhodes R. , 2006). The analysis later in this research will show that the differences between vertical and horizontal cooperation influences the types of actors within, as well as the advocacy coalition as a whole.
2.3. Policy Change
The authors describe the policy image as a perception of certain policies based on values and beliefs within the policy subsystem. This image can change either over time, due to subtle influences, or as the result of a dramatic event. The role of the media is often underexposed in the academic literature on constructing either positive or negative policy images. Subsequently, in the media often one of the two types of image prevails over time, whilst on an individual level a more nuanced policy image could be present, taking both positive and negative externalities of policies into account. The policy image is constructed by on the one hand those who formulate the policy, but on the other those affected by it. Here an issue arises what Baumgartner and Jones (2004) call the venue problem. This problem sees to what entity has jurisdiction over a specific (policy) issue, and with that has the power to influence it. Does the jurisdiction lie with the government, regional authorities, the (private) market, or even with families on micro level? A change in jurisdiction will most likely instigate a change in policy image. Constructing a receptive policy image is therefore vital for policy makers and advocacy coalitions. The search for a receptive venue for a proposal can be seen as a process of trial and error, rather than a rational one. The rationality of a venue to either support or denounce a policy proposal could be a study in itself. However based on the literature it is clear that the media have a strong influence in shaping the receptiveness of said venue.
An American study showed that increased attention, whether negative or positive decreases public support in particular for technical issues (Baumgartner & Jones, 1991). The authors conducted an in depth analysis of media reporting on technically advanced subjects, in comparison to policy debates resulting in policy formulation (and change). In essence they found that ‘no news is good news’ (for policy development), since reporting either way influences the policy image and receptivity of the venue. For example for the nuclear power industry in the United States the initial support in the 1960s was rather large. However, the general (public and governmental) opinion towards the industry changed over a period of twenty years. The way this shift in perception came about is relevant for this research since the policy formation process could show resemblances to the private industry policy process. In the US advocacy coalitions (or iron triangles) primarily formed by the elite promoted the industry, causing favourable policies and economic investments to increase. Only when a discrepancy within the expert field arose, the opponents of nuclear power increase found viable soil for their counter attack. The policy adversaries managed to sway public opinion through media channels by first increasing general attention, and later shifting focus entirely towards negative externalities of the industry. This in time led to increased governmental attention, questioning the desirability of the industry’s growth. When the (financial) market got wind of this development investors re-evaluated their opportunities resulting in withdrawal of funding. While the rest of the utilities market was on the rise, the nuclear industry suffered a reduction in value, (public) favourability, and ultimately a reduced number of supportive policies. Public support for policies in favour of the industry especially declined after a serious accident in the Three Mile Island Nuclear Generating Station. The article however shows a long-term development for the disbandment of the advocacy coalition of which the Three Mile Island incident is merely the moment suprême. Various committee hearings, lawsuits, internal governmental debates and public debates have foregone the incident collaborating to what the authors describe as “the demise of the policy subsystem” (Baumgartner & Jones, 1991). This demise can be seen as the result of changes in the policy image and venues, causing the subsystem to lose influence on the policy process.
Wolfe, Jones and Baumgartner describe the role of the media in agenda setting. Often the influence of media on both policy as well as agenda change is overlooked. Media studies usually focus on the public, while policy studies primarily focus on government agendas. The media here
can be seen as somewhat in the middle, extending their reach to both sides of the spectrum. The question still is whether the agenda setting process influences the media, the effect is reciprocal, or the other way around entirely. To better understand this process it is important to understand the different concepts it involves. Agenda setting, according to Wolfe, Jones & Baumgartner (2013) is “about organizing attention in an environment with competing information, interest, issues and ideas and limited processing capacity”. McCombs and Shaw argue the influence of the media in this process to be significant, since by focusing (public) attention on some aspects of the issue they shape the train of thought of policy developers and the population. On an individual level this means the media can influence the way people think about politicians, programs or issues. Therefore certain media coverage could either cement or deconstruct viable basis for policy proposals influencing policy imagery, frames, and ultimately policy change. Nonetheless, there are opposing views to this matter. John Kingdon’s view on agenda setting to some extent does include the media in the process, although he argues the role of the media to be limited to speeding up the process, rather than changing the outcome. In addition, the media is not to be seen solely as the source or vantage point, rather as a medium passing on specific information from one side of the spectrum to the other. This means that the discussion of media influence is actually a discussion of governmental institution communications and information interaction (Wolfe, Jones, & Baumgartner, 2013).
2.4. Factors
From the theoretical framework several factors can be derived and used for the analyses. The image directly left shows a schematic overview of said factors, further elucidated in the operationalization chapter (p.32). Four main factors influencing policy development have been distilled so far: Advocacy Coalition, Crisis/Disaster, Policy Image, and Policy Learning. Subsequently a Black Box has been created for factors that might emerge during the analysis process.
3.
Methodology
3.1. Research Design
The research question for the analysis will be what factors in the United Kingdom and the Netherlands influence policy development concerning accountability for private security companies, during the period 1990 - 2015? In order to answer this question it is first important to establish what types and pieces of legislation has been developed over the last two decades, and is now in place. In order to answer both of these issues, an operationalization of specific factors will be discussed further on, although various other concepts concerning context (Nation) related variables, and turning events3 will not be specifically elucidated or operationalized in a similar manner. These will become clear in the case respective descriptions. The private security industry will be seen as context rather than subject. However, for the sake of facilitating a knowledgeable discussion, a brief but unambiguous definition will be distilled from the body of knowledge. The types of accountability mechanisms present will be clear after extensive document research. These documents will consist of policy (formulation) documents, laws, directives, codes of conduct and company specific data. For the agenda setting theory the main focus will be on critical junctures, which will be identified by analysing media reports on private security company (PSC) behaviour as well as policy proposals by various ministries. The ACF entails some concepts that will be harder to find in practice, however by analysing academic literature it will become possible to formulate a set of conditions to be met by a group of actors to qualify as such. Then the ACF will be applied to the two cases so see whether or not an ACF was of influence in forming the present accountability mechanisms. The definitions of crises and disasters will not have to be extensively operationalized, since a strict demarcation of when and under what conditions an incident will be a crisis is not the focus of this research. Once again, a definition will be provided, based on academic literature.
This research will be based on a qualitative study in the form of a comparative analysis where the two countries (UK and the Netherlands) will provide context, which in turn will be the reference framework through which the specific cases will be analysed. The relevant elements of each case will be selected on the basis of the theoretical framework provided by various academics. This method will provide internally valid data but the sample will not be generalizable to the entire
‘population’. However, it might provide sufficient insights for the development of a new theory or expectations concerning policy formation, which could then be tested further empirically in future research. The focus will be on explaining the situation in the United Kingdom and Netherlands by taking the (further) developed situation in the United Kingdom as a vantage point, highlighting the differences and similarities. An explorative or descriptive study would be redundant since the private security industry has been subject of many policy debates and public researches already (Abrahamsen & Williams, 2006) (Brinkerhoff & Brinkerhoff, 2011).
The units of observation will be the private security related cases in the UK and the Netherlands. The units of analysis will be the accountability mechanisms that have to be abided by, by PSC’s within the national context, subsequently looked at in light of the theoretical framework. The choice for a comparative analysis has some advantages but some negative externalities as well. By comparing existing data on the industry, to an in depth case and context analysis, and the theoretical framework based on the literature review as well, the matter can be analysed from three different points of view. By using this method of triangulation various forms of validity and the reliability of the research will improve (Neuman, 2013). For instance content validity (do you measure every element of the construct as a whole), internal validity (no errors in the internal design) and external validity (can the results be generalized) will benefit from a multiple points of view approach. That last mentioned form of validity would not be directly relevant for this research. The feasibility of this research has played a dominant role in selecting the methods. It would be unrealistic to conduct survey research or use interviews due to insufficient time and funding4. This will give an accurate description of the accountability mechanisms for PSC’s and the social, cultural and political environment in which they operate. Due to the limited resources and possibilities for conducting research abroad the Netherlands has been selected. For the data on the United Kingdom extensive literature research and reports on the private security industry will show the size of the industry abroad, which is easily accessible through academic channels. Since accountability for the private sector can be a result of societal unrest, media coverage on private sector behaviour and public opinion will be part of the analysis as well. This section will entail newspaper articles, social media (reports), and other media through which the public might express their opinion or (dis-) satisfactions.
4 Though this might be an option for future empirical research to ensure external validity, since this cannot be fully achieved by this (type of) research.
Now that the methods of data collection are clear, it is important to clarify how the collected data will be exploited and assessed. First, extensive desk research will lead to a specific set of conditions or factors pertaining accountability mechanisms and policy formation. By reading up on the different theories and academic perspectives it will become possible to form a set of conditions required for the identification of accountability mechanisms. Secondly, once this set of conditions has been formed further document research into policy documents and the like will show whether or not these conditions can be identified in the previously mentioned countries and their embedded cases. Here policy documents on both countries will be assessed. Thereafter it will thirdly be possible to compare the differences and/or similarities between countries and their respective cases in light of the theoretical framework. Finally circling back to the theory will not only provide an explanation for the differences between the assessed countries, but it might even produce recommendations for policy developments for the private security sector which might help the collaboration between public and private actors.
3.2. Case Study
Like incremental policy development a qualitative case study is extended over a longer period of time. Especially when trying to find an answer to how and why questions; when the subject of study is not influenced by the researcher; when context is important in explaining the unit of analysis; and when it is unclear how context and subject are intertwined (Baxter & Jack, 2008) a case study could provide a detailed depiction of reality, and with that a proper answer. Detailed in this respect means the context is thoroughly accounted for in the analysis5. However, the focus of a case study might as well be to explore or compare a case rather than providing a holistic description. This research will therefore not provide a holistic description as such, describing every element of both cases. Rather it will be aimed at exploring and comparing the Netherlands and the United Kingdom in order to highlight significant different factors influencing policy development and change. First it is important to further specify the unit of analysis, as mentioned earlier. This selection could depend on whether a researcher wants to investigate an individual, a program a process or organizational differences. The focus for this study will be on various programs and processes entailing policy formation. The private security sector will merely
5 Meaning that amongst others the political background of countries, sub-systems, and socio-economic means and values are all accounted for.
provide context. When it is difficult to draw the line as to where exactly the context ends and the case starts, a case study in particular could be a helpful tool according to Yin (Kohlbacher, 2006). The selection of cases severely influences the research question as well, since demarcating what is considered to be of importance, and which factors fall out of scope sets boundaries to the topics to be researched. Such limitations involve the setting of a timeframe, place and conceptual definitions (Baxter & Jack, 2008). When trying to find an explanation for complex social phenomena, a case study allows the researcher to make in in depth analysis of ‘all’ the (relevant) real life elements contributing to an explanation. In order to prevent ambiguities triangulation of data is necessary since a case can be looked at from multiple perspectives (political, social, academic, corporate, etc.) only a multiplicity of angles will achieve a realistic and verified usable image. For this research this means that both academic theories on the matter of policy formation will be analysed, a second point of view will stem from the industry itself, and a third will come from the office of policy makers. However, the result of a case study analysis based on multiple data sources is not as such automatically generalizable to the entire population. According to Yin the results from a case study will however be generalizable to theoretical propositions (Kohlbacher, 2006). These will be derived from the theoretical framework where theories on policy formation by, amongst others; Kingdon, Baumgartner, Jones, and Sabatier are evaluated. A previously determined and well elaborated on theoretical framework is of vital importance when conducting a case study research. Since it will be the lifeline to generalizable results and with that academic value of the research. Cases with similar theoretical traits will be eligible for comparison. Where the cases differ, an explanation for the to be elucidated phenomena might be found. Now that the method of case selection is clear, it is important what kind of study will be conducted upon these cases.
3.2.1. Why Multiple Case Study
There are basically three types of case studies (explanatory, exploratory and descriptive) sub-divided into another three categories (single, holistic and multiple-case). From the former selection the last two types of case study are not relevant for this research, since an exploratory case study might be useful if the consequences of a case are unclear and desire prediction, whilst a descriptive case study is limited to describing a situation within the embedded context. The first however, an explanatory multiple case study design, will provide clear differences and
similarities between the cases, making it relevant for this research. Upon which a comparison can be built in order to foresee either similar situations or predictable differences in light of a certain theoretical framework (Yin, 2003). This is the type of study that will be conducted in this research. The explanatory multiple case study design is used when trying to explain the link between cases and contexts, which cannot be examined by surveys or experiments due to the complexity of the issues. For this research a multiplicity of cases is desired since without it, no comparison can be made in order to provide a viable explanation. A single case study design for instance would not support theoretical notions sufficiently to assume them to be true. Furthermore, multiple case study analysis could result in predicting similarities, differences or desired results based on a theoretical framework (Baxter & Jack, 2008). For this research a holistic case study with embedded research also would have been a possibility, however, since there are multiple contexts involved (the UK and the Netherlands) a multiple case study design is better suited. In general the multiple case study approach is used when the research aims to answer why or how questions; for this research that means the question is how various factors from the theories on policy development influence accountability mechanism development for the private security sector, a process that cannot be steered or otherwise influenced by the researcher since policy formulation is a far to complex system to be influenced by a single study. In addition the (political) context of different countries is of importance since it seriously influences the policy formulation process.
3.2.2. Why UK and The Netherlands
The first reason behind case selection is the feasibility of the research. In the allotted timeframe it would be unrealistic to select more than two cases, therefore this research is limited to the UK and The Netherlands. Secondly as a researcher located in the Netherlands the choice for the country of origin is primarily based on accessibility of the required information. More readily accessible data will increase the feasibility of the research. Another, third, reason for choosing the Netherlands is the relatively underdeveloped legal framework that will be one of the vital elements of this research. When comparing a somewhat ‘young’ legal regime to a more ‘mature’ piece of legislation, useful insights based on the lessons learned by the more experienced regime could positively contribute to the former. Therefore an analysis of the UK, with their years of experience in regulating the private security industry, might lead to helpful recommendations for
