University of Amsterdam Graduate School of Social Science
European Security Politics
In the post 9/11 landscape, how is the London
insurance market making the risk of terrorism
governable?
Submitted by: Maria Doljikova Supervisor: Marieke de Goede Secondary Reader: Daniel Mugge Date: 24.06.2016
Table of Contents
Introduction………..3
Theoretical framework Part 1: Beck, insurance and the risk of terrorism………5
Part 2: Insurance and the catastrophic risk of terrorism……….12
Part 3: Governing terrorism through insurance……….19
Methodology Part 1: London insurance market ………..30
Part 2: Expert interviewing………..33
Part 3: Content analysis………40
Empirical Findings Part 1: Calculating the incalculable………..43
Part 2: Risk mitigation………..55
Part 3: Expertise………59
Conclusion………67
Introduction
Terrorism is a serious and sustained risk which threatens the lives of people across the world. Since the start of the 21st century, there has been a nine-fold increase in the number of reported deaths from terrorism, rising from 3,329 in 2000 to 32,685 in 2014 (Institute for Economics and Peace 2015:2). The increasing death toll calls attention towards the security apparatus which is in place for the purpose of protecting citizens. In today’s fight against terrorism, the scale and complexity of the risk demands that national security responsibilities are not limited to the state. Counter-terrorism initiatives have now been extended to the private sector. Of particular importance is the role of the insurance industry.
The insurance industry plays a critical, yet hidden, role in the fight against terrorism by acting as both a risk transfer and risk mitigation mechanism. The London insurance industry, which is the world’s only specialist insurance market, is of particular importance in the management of terrorism risk. The London market prides itself as being the global hub for insuring complex risks. Terrorism is arguably one of the most complex risks that the London market insures. The ever evolving and increasingly complicated nature of terrorism make it a risk that is difficult to accurately assess and manage. Insuring against terrorism is problematic because insurers have a completely insufficient amount of data to work with. The limited amount of historic incidents relating to terrorism, combined with the difficulty in accurately being able to establish the frequency and severity of a future attack, significantly complicate the insurability process. In addition, the human element of terrorism make it a risk that is majorly unpredictable. Despite all of these barriers, terrorism is still being insured, the purpose of this thesis is to explain how this is being done.
The empirical project of this thesis is to investigate how the London insurance market is making the risk of terrorism governable. This thesis will argue that insurance governs the risk of terrorism through calculation, risk mitigation and expertise. In addition to exploring these elements, this thesis will also provide research that will enable an understanding of how insurers understand the risk of terrorism and their role as security provides. In order to answer the elements set out, the paper will be split into three sections.
The first section will provide a theoretical framework relating to the risk of terrorism and insurance. The paper will defend why 9/11 has been chosen as a starting point for analysis and will draw attention to the scholarly debates surrounding the difficult in insuring terrorism. In addition, this section will also conceptualise insurance as a technique of risk governance. It will be explained why governance has been chosen as a lens through which to study insurance and its relation to risk. Further, this part of the paper will outline how an empirical investigation will be conducted based on the theoretical framework.
The second section will justify why the London market has been chosen as the case for analysis. Further, the methodological approaches of expert interviewing and content analysis will be rationalised.
The third section will present empirical findings. The findings will be split into three sections: calculation, risk mitigation and expertise. Once the findings have been presented their theoretical implications will be discussed.
The thesis will conclude with a discuss of how the research question has been answered. Attention will be drawn to the extent to which empirical findings are in line with theories presented by other scholars. Further, it will be made clear what contributions this thesis has made to literature relating to terrorism insurance. The thesis will finish with concluding remarks relating to suggestions for future research.
Theoretical Framework
Part 1: Beck, insurance and the risk of terrorism
The purpose of Part 1 is to embed terrorism insurance within Beck’s uninsurability thesis in order to show how he conceptualises the risk of terrorism and its insurability. Firstly, this section aims to demonstrate that Ulrich Beck (2002) saw 9/11 as a turning point which marked the distinction between terrorism being an insurable and uninsurable risk. Secondly, this part of the paper presents the view that there are limits to the empirical validity of Beck’s arguments concerning the insurability of terrorism. Although the case of the Terrorism Risk Insurance Act (TRIA) does provide empirical support to some of Beck’s claims, this section aims to show that terrorism continued to be insured following 9/11. Thirdly, this section wants to draw attention to Beck’s conceptualisation of the relationship between risk and its insurability. By focusing on scientific control, Beck’s conceptualisation of risk and its insurability provides a sound theoretical and empirical starting point for explaining how insurers approach risk. Beck’s approach thus provides a comfortable starting point from which to answer the question of how terrorism continued to be insured post 9/11.
Beck’s uninsurability thesis
September the 11th 2001 marks a significant theoretical turning point because the event is used by Beck (2002) to distinguish the transition between terrorism evolving from an insurable to an uninsurable risk. Following the attacks perpetrated by Al Qaeda, Beck famously stated that the atrocities conducted by the terrorist group clearly depicted that terrorism has morphed into an “uncontrollable risk” (Beck 2002: 41). The ‘uncontrollable’ nature of terrorism consequently meant that “the boundaries of private insurability (have) dissolve(d)” (Ibid). Although Beck does seem to suggest that following 9/11 terrorism seized to be an insurable risk, he does not completely disregard the capabilities of insurers. He argues that one of the consequences of 9/11 is that “private insurance is partly being replaced by the principle of state insurance” (Idem: 45). Beck seems to suggest that 9/11 has demonstrated that terrorism has transformed into a risk which is too large for the private insurance industry to manage. In consequence, the responsibility for providing terrorism insurance has shifting from the private market to the government. The uninsurability premise, that 9/11 acted as the interface between terrorism transforming
from a risk which could be insured by the private market to one which could not, forms part of Beck’s broader theoretical reasoning relating to two distinct phases of modernity.
First Modern Society
Prior to September the 11th Beck (2002: 44) argued that terrorism was insurable because it took place within “first modern society”. A distinguishing feature of ‘first modern society’ is that risks can be controlled by “means of scientific rationality” (Beck, Bonss, Lau 2003: 4). Beck, Bonss and Lau (Ibid.) state that, “first modern societies unfold themselves on the basis of a scientifically defined concept of rationality that emphasise instrumental control…this implies a belief that scientization can eventually perfect the control of nature”. Control is necessary because risks are produced as a result of industrialisation which is driven by decision making aimed at producing and distributing wealth in conditions of scarcity (Aradau and Van Munster 2007). Beck thus sees first modern society as being able to demystify the abstract concept of terrorism via controlling it through scientific rationality. The example of the scientific tool used to control terrorism is insurance. Insurance is seen as an apparatus of scientific rationality because it “is based on the fundamental potential for compensation of damages and on the possibility of estimating their probability by means of quantitive risk calculation” (Beck 2002: 41). Insurance is therefore a distinguishing feature because “calculating risk is part of the master narrative of first modernity” (Ibid). The first modern society is thus exceptional because during this phase of modernity the risk of terrorism could be scientifically quantified via risk calculation and hence was seen as insurable.
World Risk Society
Following 9/11, terrorism was perceived by Beck (2002: 44) as uninsurable because modernity had transitioned from first modern society and entered into the much academically debated “world risk society”. The transition from first modern society to ‘world risk society’ produced a “gulf between the world of quantifiable risks in which we think and act, and the world of non-quantifiable insecurities that we are creating” (Idem: 40). The transition between these two phases occurred because new lines of research and technology, which are pursued for the purpose of modernisation, are deepening, widening and multiplying risks (Beck, Bonss, Lau 2003). Modernisation has turned up
scientific rationality and control (Ibid ). An example of a risk which exceeds controllability is terrorism, it has evolved “beyond rational calculation into the realm of unpredictable turbulence” (Beck 2002: 43). Thus, in contrast to first modern society which saw terrorism as a controllable risk, world risk society conceptualises terrorism as an uncontrollable risk because it exceeds the means of scientific calculation. The distinction between first modern society and world risk society demonstrate that Beck has a prescribed way of conceptualising the controllability of risk. He associates control with scientific tools such as mathematical calculation. The calculability of a risk appears to categorise whether a risk is controllable or not. In the same fashion, Beck also affiliates calculability with insurability. Beck seems to suggest that if a risk cannot be calculated, it cannot be insured. Terrorism is thus seen as uninsurable within world risk society because it is considered a non-quantifiable risk.
Terrorism Insurance post 9/11 - United States
The case of the Terrorism Risk Insurance Act (TRIA) issued in the United States provides temporary support to some of Beck’s theoretical assumptions. TRIA demonstrated that immediately post 9/11 the “boundaries of private insurability” did temporarily dissolve and “private insurance was (partly) replaced by state insurance” (Beck 2002: 41- 45) . This phenomenon occurred as a result of the private insurance industry incurring losses of $32.5 billion as a result of the Al Qaeda attacks (Insurance Information Institute: 2011 ). The large losses forced private reinsurance companies to withdraw from the market and resulted in the provision of state insurance which was offered via TRIA.
Reinsurance companies reduce the risk for insurers by accepting a premium in return for a portion of the insurer’s risk, this allows insurers to spread their risk and hence reduce their capital exposure. The reinsurers exit from the market severely limited the amount of terrorism insurance policies availability. This was because insurance companies would have to absorb all of the risk caused by a potential loss, as opposed to spreading it with the reinsurers. In response to the limited availability of terrorism coverage Congress passed TRIA, the purpose of which was to establish a temporary Federal program that provides for a system “of shared public and private compensation for insured losses resulting from acts of terrorism” (US Congress 2002: 101b). The reasoning behind the Act was that the US Government would temporarily replace private companies as the reinsurers. Thus, the state would be responsibility for absorbing terrorism relating loses if
an attack occurred. The implementation of TRIA is likely to have motivated Beck to argue that the boundaries of private insurance dissolved post 9/11 and were replaced by a state alternative. This argument is plausible given that 9/11 caused both TRIA to be enacted and motivated Beck to propose his terrorism uninsurability thesis. Although TRIA does provide empirical support to Beck’s claims concerning terrorism insurance post 9/11, two major flaws can be traced in his argument.
Firstly, Beck’s association with insurability and calculation would lead to the assumption that TRIA was enacted because private reinsurance companies miscalculated the risk posed by terrorism. It could be argued that Beck assumed this miscalculation caused the reinsurers large losses and forced them to exit the market. Although plausible, this assumption is technically problematic as Beck never specified exactly what it was that insurers were calculating. Nevertheless, miscalculation did not cause the reinsurers to exit, the large losses occurred as a result of poor policy wording. Prior to 9/11 “few policies contained express terrorism exclusions” (Allyn 2003: 822). As a result, “insurers covered terrorism in most ‘all risk’ policies” because “the risk was perceived to be so small that is was covered for ‘free’” (Rhee 2013: 2). Terrorism was thus a risk that had not been expressly account for in insurance policies. In relation to Beck, it is not that terrorism had been miscalculated, it is more so that it had not been accounted for. Terrorism was a risk that had not been factored in to any type of calculations made by US insurers.
Secondly, contrary to Beck’s assumption, although TRIA remains implemented as a federal backstop, the availability and affordability of terrorism insurance provided by private companies has improved in the US since 9/11 (President’s Working Group 2006: 4). Firstly, this finding shows that the private market continued to insure terrorism post 9/11. Secondly, it demonstrate that more policies became available at cheaper prices. This evidence empirical challenges Beck’s claim that terrorism seized to be insured by the private market following 9/11.
Terrorism Insurance post 9/11 - Globally
Beck’s uninsurability thesis is also empirically problematic on a global scale. The global take-up rate for insurance coverage against terrorism increased following 9/11 (Statista: 2016). In 2011, the overall terrorism coverage take-up rate reached 64 percent, in 2003 the take-up rate amounted to 27 per cent (Ibid). Take-up rate measures the percentage of eligible people who accept a particular good or service (Economics Help: 2016). A rise in take-up rates for terrorism insurance represents a rise in the amount of insured protected against the risk of terrorism. The below graph demonstrates that there was a gradual increase post 2003. Therefore, in contrast to Beck’s assumptions, terrorism continued to be insured globally. In addition, there was actually an increase in the amount of insured covered against the risk of terrorism. This empirical data empirically and conceptually challenges Beck’s claim that terrorism became an uninsurable risk post 9/11.
Literature Review
The boldness of the statements made in Beck’s uninsurability thesis has created a breeding ground for academic debate which supports the argument that terrorism continued to be insured following 9/11. Aradau and Van Munster (2007: 95) state that the argument that catastrophic terrorism became “incalculable and uninsurable” appears to contradict the “institutional measures and actions that followed the tragic event of 9/11”. The institutional measures addressed here refer to TRIA. Other scholars also acknowledge that terrorism continued to be insured after 9/11 (Collier: 2008, O’Malley: 2003, Bougen: 2003, Ericson and Doyle: 2004). The continuation of terrorism insurance post 9/11 raise important conceptual questions. Ericson and Doyle (2004a) contend that given that insurance companies continued to provide coverage for terrorism, it does not seem plausible to asserts that private insurance serves as the boundary marker of risk society. Aradau and Van Munster (2007: 94) align with this argument as they point out that it is empirically difficult to locate a threshold between first modernity and risk society, “which in turn renders problematic the claim that today we live in a world risk society”. The continuation of terrorism insurance following 9/11 thus raises empirical flaws in Beck’s argument and brings into question his conceptual grounding.
Beck’s conceptualisation of insurance and risk
The way that Beck conceptualises insurer’s relationship to risk, which centres on calculability, can assist in explaining why he argued that terrorism became uninsurable post 9/11. Beck (2002: 44) argues that “in the terrorist risk society the world of individual risk is being challenged by the world of systemic risk, which contradicts the logic of economic risk calculation”. Individual risk can be defined as unsystematic, which refers to a risk that is industry specific (Investopedia: 2016). In contrast, systemic risk is a risk that affects the overall market or industry, it is unpredictable and impossible to completely avoid (Investopedia: 2016). Unlike individual risk, systemic risk cannot be mitigated. Beck thus suggests that individual risks can be controlled via economic risk calculation, however, the same practice is not effective in overcoming systemic risks.
Beck’s conceptualisation of systemic risk can be applied to what Beck terms as the “axes of conflict in world risk society”, these consist of ecological conflicts, global financial
conceptual term Beck applies to what he believes are systemic risks. Unlike individual risks, the axes of conflict are risks which are systemic in nature because they exceed spatial, temporal and social dimensions (Ibid). Within the spatial dimension, systemic risks are those that exceed the limit of national state boundaries. These risks are also temporal in nature as their effects are long term and unpredictable. Lastly, systemic risks exceed the social dimension. When they cause damage, it is difficult to appoint liability to a certain group or individual who can be held responsible. Specific examples of the systemic risks used by Beck (2002) are Chernobyl, global warming and 9/11. Beck’s conceptualisation of systemic risks thus rests on their spatial, temporal and social dimension which makes them incalculable and thus uninsurable.
Part 2: Insurance and the catastrophic risk of terrorism
The purpose of Part 2 is to demonstrate that academic literature perceives terrorism as a man-made catastrophic risk which exceeds the boundaries of calculation. The limited loss history and challenging nature of obtaining current data surrounding terrorism means that the information needed to accurately calculate the frequency and severity of an attack is limited. As a result, terrorism proves to be a particularly difficult risk to insure. In response to the problematic nature of calculating the risk of terrorism, this section presents scholarly arguments which contradict Beck’s uninsurability thesis by presenting alternative solutions by which terrorism continued to be insured post 9/11. These alternative approaches are theoretically important because they demonstrate that insurability is not rigidly centred on calculability.
Catastrophic Risks
Beck’s conceptualisation of systemic risk can be applied to what insurers term catastrophic risk. The connection between these two types of risks provides theoretical and empirical insight into why Beck perceived terrorism to be uninsurable post 9/11. Beck (1992: 24) asserted that, “it may be that risk society is a catastrophic society”. Similarly to systemic risks, catastrophic risks also exceed spatial, temporal and social dimensions. Catastrophic risks are not limited by geography, time or social liability. The similarities between the nature of systemic and catastrophic risks leads to the premise that Beck sees both as exceeding insurability. O’Malley (2003: 277) supports this argument, he states that Beck’s uninsurability thesis is problematic because it relies “upon the image of catastrophes as ungovernable”. For Beck, governability refers to the process of being able to achieve mathematical rationality, which in turn enables insurability. On these grounds, it can be logical to assume that in addition to systemic risks, Beck also views catastrophic risks as uninsurable.
Insurers split catastrophic risks into two categories: natural and man-made. Natural catastrophic risks are those which are naturally occurring and have catastrophic consequences, examples include earthquakes, hurricanes and storms. Man-mad catastrophic risks also carry catastrophic consequences, however, unlike with natural disasters, these types of risks are purposely instigated by humans, the most prominent
The key feature which differentiates catastrophic risks from systemic risks is that the former are low probability and high severity. Although this criteria can be applied to some systemic risks, such as Chernobyl and 9/11, it is not as well suited to longer term events such as global warming or a financial crisis. Thus, systemic risks do differ from catastrophic risks, although terrorism is an example which is applicable to both. The exceptional nature of terrorism as a risk raises important theoretical and empirical concerns. Beck’s assumption concerning the uninsurability of catastrophic risks does follow a sensible logic in that terrorism is a systemic, low probability, high severity risk and thus should technically not be calculable and hence insurable.
The systemic, low probability and high cost nature of catastrophic risks makes them notoriously difficult to insure due to the difficulty in establishing their frequency and severity. “If frequency and severity cannot be reasonably calculated and assessed, the viability of insurance appears doubtful” (Ericson & Doyle 2004b: 29). Frequency refers to how likely an event is to happen. In insurance terms, this translates into predicting how many claims an insurer expects to occur. High frequency equals a high number of claims. Severity refers to the amount of damage an event is likely to inflict. Insurers base severity on how expensive they predict a claim to be. High severity claims are more expensive than low severity claims.
There are two main reasons why knowledge relating to the frequency and severity of a risk is important. Firstly, it allows insurers to calculate their financial exposure to a risk. Insurers base their decisions on how much of a risk they can prudently accept on factors relating to frequency and severity. Secondly, information relating to these two factors allows insurers to forward plan in order to reserve funds for claims they think may occur. The information which drives knowledge relating to frequency and severity is derived from past and current data relating to the risk in question. Past data is termed by insurers as loss history. The richer the loss history and current data set the easier it will be to determine the frequency and severity of a set risk. Catastrophic risks are traditionally seen by the insurance industry as low frequency and high severity. This combination complicates insurers risk decisions as there is a limited amount of information to base judgement from and a very high cost to pay in the case of a claim. The low frequency and high severity nature of catastrophic risks therefore makes them more difficult to calculate. The catastrophic damage caused by Al Qaeda on 9/11 is most probably what led Beck to assume that the risk of terrorism seized to be insurable.
Terrorism as a Catastrophic Risk
Within catastrophic risks, terrorism is a particularly difficult risk to insure due to its extremely limited loss history and human element. Prior to 9/11 the worst man-made disasters had been industrial accidents such as Chernobyl (Rhee 2006). Based on previous loss history, prior to 9/11 insurers in the US considered terrorism to be a “low intensity, high visibility risk” (Rhee 2008: 2) . It was only after the attacks that US insurers saw terrorism as being able to achieve the same catastrophic scale as natural disasters. 9/11 stands apart as “the only manmade disaster that had a deep, albeit temporary, impact on the US economy” (Rhee 2006: 584). Post 9/11, insurers remain in a struggle to overcome the difficulty in predicting the frequency and severity of an attack. Within the US, “there have only been two instances of foreign inspired terrorism” (Thomas 2004: 500). The events consist of the 1993 World Trade Centre bombing and 9/11, “this is a totally insufficient number of data points to be able to do any traditional actuarial work to forecast what future costs might be” (Ibid).
In addition to the insufficient loss history, the human element of terrorism makes the frequency and severity of an attack difficult to estimate. Terrorism is a more volatile risk than natural disasters because unlike with nature, which can be predicted via fractal patters as well as geographic and geological forecasts, terrorist are human beings who “will shift in response to changes in public and private defences” (Lakdawalla & Zanjani 2004: 463). If a terrorist group has reason to believe that their plot is likely to be foiled, they are able to abandon the operation and act rationally based on the new information which they have acquired. Thus, the limited loss history and human element of terrorism problematises the ability of insurers to determine the frequency and severity of future attacks. This complicates the insurability process.
Insuring Catastrophes
The low frequency and high severity nature of catastrophic events has left scholars asking the question of just how risks such as terrorism continue to be insured. Bougen (2003: 258) cites , “just how the catastrophic underwriter arrives at key decisions has proven to be a source of fascination for many years, yet literature on the topic is particularly scarce”. An underwriter is an employee of an insurance company who
the terms and conditions that legally bind the insurance policy. The question of how underwriters reach decisions relating to risk has also fascinated Strange (1996:123) who states, “how insurers exercise power over risk is a fundamental question for contemporary international political economy. For fifteen years I have waited, in vain, for someone to write a definitive account”.
This gap in academic literature, relating to how underwriters make risk decisions, is likely to be present as a result of the difficulty in accessing insurance based information. Insurers confidentiality is due to the sensitivity surrounding client data and the competitive nature of the insurance market. Ericson and Doyle (2004: 17) highlight the fact that insurance companies do not operate in an independent sphere, instead, they function within market based conditions which are “highly competitive, and therefore subject to the same pressures as any other commodity”. Like any other business, insurance companies sell products with the hope of generating a profit. In order to achieve this profit insurers have specific sales targets which are set internally by high management. In the pursuit of meeting these targets it is critical for insurers to have a competitive edge over other firms. Insurers complete to win business from a limited pool of brokers . In the race to compete 1 for clients, it would be unproductive for insurers to openly disclaim their methods of risk calculation and management. This is because it would enable competitors to familiarise themselves with an insurers’ underwriting strategy and hence would make that company lose its competitive edge. In consequence, information relating to insurers underwriting decisions is kept confidential.
Calculation
In line with Beck, other scholars also prescribe scientific rationality, in the form of risk calculation with much importance when addressing the question of how companies insure terrorism. Catastrophic risks, such as terrorism, are seen as governable by insurance via risk calculation (Beck 2002, Bougen 2003, O’Malley 2003, Ericson and Doyle 2004a, Ericson and Doyle, A. 2004b, Ericson 2005, Collier 2008). Ericson and Doyle (2004b: 5) argue that insurers’ “own science of actuarialism converts myriad risks in the world into insurance technologies that spread and compensate losses with fairness and efficiency”. Actuarial science applies the mathematics of probability and statistics to
Brokers are the intermediary between the insurance buyer and the insurance company, their role 1
define, analyse and solve the financial implications of uncertain future events (Investopedia: 2016).
In the case of insurance, the data set which actuaries work with is based on the risk’s loss history, as well as currently available information, which helps to predict frequency and severity. Collier (2008: 235) refers to loss history as “archival statistical knowledge”, it is an “‘archive’ of past events that could be analysed statistically”. This type of data is needed to “make decisions about what to insure, whom to insure and what premiums to charge” (Idem 227). The loss history, or ‘archival statistic knowledge’, combined with up-to-date risk information is therefore the data set which allows insurers to calculate risk and hence insure it. Scholars are in agreement that calculation is a key feature which enable terrorism to be insured. However, given the problematic nature of scientific rationality as a risk management tool, the scholars do vary on how much importance they prescribe to calculation, as a means to insure risk.
The problematic nature of insuring terrorism is evident when conceptualising archival statistical knowledge as the element in driving insurance decisions. This is because when confronted with risks such as terrorism, which is low probably and high severity, the data is very scarce and hence proves inadequate in functioning as a risk assessment tool. Ericson and Doyle (2004b: 5) present this idea as their main thesis: they argue that “a high degree of scientific and technical uncertainty permits the insurance industry, the very business that is charged with transforming uncertainty into risk”. Uncertainty can be defined as “the lack of secure knowledge about an unwanted outcome. Insecure knowledge may result because of unavailable or unreliable data about frequency and severity” (Ibid). Scholars are in agreement that there is a limit to statistical data’s use as a risk assessment tool (Bougen 2003, O’Malley 2003, Ericson and Doyle 2004a, Ericson and Doyle, A. 2004b, Ericson 2005, Collier 2008,). In response to the limit of data, these scholars have argued that insurers respond with creative alternative solutions.
Reinsurance
In his analyses of catastrophic risk, Bougen (2003) stresses the importance of reinsurance as a catastrophe-financing mechanism. He argues that reinsurance is permitting the insurability of traditionally non-insurable risks, such as terrorism (Ibid).
statistical technologies” and hence can only become insurable if “imaginatively assembled risk networks for catastrophe financing can be discovered”. Reinsurance is the method of financing to which Bougen is referring to. It is a “financial instrument developed in conjunction with capital markets design to spread the cost of catastrophes across a wider financing base” (Idem: 253). Bougen (Ibid) refers to this process as the “securitisation of catastrophic risk”. This process involves “packaging the risks associated with a particular catastrophic event and transferring them to capital markets, which assume the obligation for financing the costs of catastrophes in return for some form of payment from the reinsurer” (Ibid). Reinsurance is thus a risk spreading mechanism which allows for the transfer of risk from one party to another, in exchange for a premium. By spreading risk, both insurers and reinsurers are limiting their financial liability in the face of catastrophic risk. The rationality behind this mechanisms is: if a loss does occur, because the risk is spread across so many parties, no one party will incur a loss large enough to cause considerable financial damage. Spreading risk via reinsurance is therefore a method used by insurance companies to allow them to insure traditionally non-insurable risks such as terrorism.
Risk Mitigation
Ericson and Doyle (2004a), Ericson and Doyle (2004b), Ewald (2002) and O’ Malley (2003) take an alternative position to Bougen (2003), they highlight insurers use of risk mitigation techniques as a way to manage complex risks such as terrorism. Insurers play “key but often hidden roles in establishing preventive security and loss prevention infrastructures, whether based on environmental design, electronic surveillance technologies, or private security operatives” (Ericson and Doyle 2004a: 139). When risk management decisions are “beyond calculus of probability and severity. They (insurers) achieve certainty through a different logic, that of precaution, vigilance and pre-emption” (Ericson and Doyle 2004b: 7). Pre-emption was translated through new approaches to security which are based on the precautionary principle (Ewald 2002). The precautionary principle emphasises that low frequency, high severity risks must be addressed through extraordinary measures that reflect ‘zero tolerance’ and aspire to ‘zero risk’ (Ewald 2002). The purpose of risk mitigation strategies is to deter the risk before it flourishes, this type of security technique is best labelled as preemptive.
The preemptive approach to security is most closely associated with the Bush Doctrine which arose in response to the attacks of 9/11 and was used to justify the invasion of Iraq in 2003. Preemption involves enforcing action for the purpose of mitigating a risk before it becomes threatening (Dershowitz 2006). Preemptive action is part of the precautionary principle which is enforced for the purpose of addressing risks which exceed calculability and have the potential to cause catastrophic damage (Ewald 2002). Terrorism demands to be addressed via the precautionary principle because it upsets the very postulates of insurance which is based on accurately calculating frequency and severity (Ibid). Insurers’ approach to terrorism can be compared to Bush’s war on terror which was governed by preemptive action. De Goede (2008: 164) argues that “the war on terror recognises that the sheer uncertainty and randomness of terrorist attacks renders conventional risk assessment techniques inadequate. However, it subsequently moves to incorporate this uncertainty into policy - making a basis for preemptive action”. In the same fashion as the war on terror, the incalculability of terrorism creates uncertainty and demands that insurers take preemptive action, in the form of risk mitigation, for the purpose of avoiding the risk before it materialises.
Human Judgment
Human judgment is an additional element that Ericson and Doyle (2004a) and O’Malley (2003) prescribe attention to when answering the question of how insurers address the risk of terrorism in the face of uncertainty. O’Malley (2003: 277) points out that “many models of future governance rely more on estimation than on calculable risk, and many of these have long and effective histories”. The “model of future governance” to which O’Malley (Idem) is referring to is insurance. Specifically in the case of terrorism, “loss estimate models are heavily depend upon the subjective opinions of selected experts” (Ericson and Doyle 2004a: 150). It is important to acknowledge that “insurers impose meaning on uncertainty through non-scientific forms of knowledge that are intuitive, emotional, aesthetic, moral and speculative” (Idem: 138). Ericson and Doyle (2004b: 10) stress that “knowledge of risk and uncertainty are sources of fear, and these emotions can fundamentally affect how people approach security”. It is important to recognise that insurance companies are run by people and not machines. Thus, it is critical to encompass the role that human judgment plays in risk making decisions because this is what ultimately guides risk insurability.
Part 3: Governing terrorism through insurance
The purpose of Chapter 3 is to introduce governance as theoretical approach to understanding the relation between insurance and terrorism. Insurance is a technique of governance, it is the result of the way in which insurers approach risk. Governance has been chosen as a model to approach terrorism insurance because it offers insight into the broad array of techniques, methods and rationalities that enable a risk to be insured. Firstly, this Chapter will present the argument that insurance is a technique of governance implemented for the purpose of approaching risk. Secondly, the Chapter will outline how this thesis conceptualises governance, the model of Miller & Rose (1990) will be applied. Thirdly, this Chapter will argue that insurers govern risk for the dual purpose of economic gain and societal risk reduction. Finally, the chapter will outline how insurance will be empirically analysed as a model of governance.
Insurance as a technique of governance
Insurance is a technique of governance orchestrated by insurers for the purpose of managing risk. Insurance “has become a central institution of governance beyond the state, a key institution for aspects of governance such as risk management and security provision” (Ericson, Barry and Doyle 2000: 553). Ericson and Doyle (2004) put forward the argument that within a neo-liberal society, although the state is the ultimate risk manager (Moss 2002), the state cannot manage all risks without the insurance industry. This argument is echoed by O’Malley (2004) who argues that neo-liberalism seeks to transform risk from beyond a technique of social security provision to a responsibility to be assumed by self governing entities. Insurance companies are an example of these self governing entities, they are “crucial for governing the everyday world of safety and security” (Ericson and Doyle 2004b: 4). Insurance can thus be seen as a tool of governance implemented for the purpose of risk management. Insurers manage risk by offering policies which transfer the financial liability which may arise from a loss from the insured to the insurer. Thus, insurance governs risk by providing citizens with financial protection against insured losses.
Viewing insurance as a technique of risk governance impacts the way that risk is conceptualised. Actors which play a role in governing risk, such as “government agencies, private organisations and insurance companies, all conceptualise risk in
different ways” (Tobin 2015: 495). This thesis follows the definition of risk prescribed by O’Malley (2004), who holds a “social constructionist view of risk” (Ericson 2005: 670). Risk is treated by O’Malley (2004: 15) as “neither real or unreal”, instead it is a way “ in which the real is imagined to be by specific regimes of government in order that it may be governed”. Ericson (2005: 559) states that O’Malley (2004) defines government as “practical ideas and techniques through which state and non-state entities, institutions, organisation, communities and individuals seek to understand, govern, and change the world”. Insurance can be seen as a specific regime of government which conceptualises risk in a way that enables it to be governable. Ericson and Doyle (2004b: 5) argue that “insurance is a key institution in which to understand knowledge of risk”. The definition proposed by O’Malley (2004) has been chosen because it is able to account for the way that insurers attribute meaning to the abstract concept of risk. Insurers view risk in a specific way in order for it to be rendered governable, thus, insurance is a particular way to approach risk.
Governance has been chosen as a theoretical approach to understand insurance because it is able to account for the wide range of techniques, mechanisms and rationalities which enable a risk to be insured. It is important to recognise “the diversity of techniques” that insurers use “for governing the future, rather than conflating them into a binary of ‘risk’ and the ‘incalculable’ other of estimation” (O’Malley 2003: 277). Ericson and Doyle (2004a: 139) are in agreement as they present the argument that “insurance based governance of threats involves diverse elements assembled in different ways”. Understanding insurance as a technique of governance draws attention towards the broad array of methods used by insurers to enable a risk to become insurable. In his discussion of O’Malley’s (2004) work, Ericson (2005: 661) argues that O’Malley (2004) sees insurance as a way to “evoke specific practices in relation to risk”. Ericson (2005: 663-7) elaborates as he argues that O’Malley (2004) sees insurance as a technology of “government that direct(s) conduct in particular ways”. Governance is the result of of the way in which insurers approach risk. By viewing insurance as a governance technique, insight can be gained into the specific practices and conducts which enable a risk to become insurable.
This thesis conceptualises governance according to the model proposed by Miller & Rose (1990). The reason for choosing this model is because:
It draws attention to the fundamental role that knowledges play in rendering aspects of existence thinkable and calculable, and amenable to deliberated and plentiful initiatives: a complex intellectual labour involving not only the invention of new forms of thought, but also the invention of novel procedures of documentation, computation and evaluation (Miller & Rose 1990: 3).
The governance model of Miller & Rose (1990) is insightful in the way that it prescribes particularly importance to knowledge. By highlighting the role of knowledge the model is able to account for a diverse range of techniques that insurers use for the purpose of governing risk. These techniques include calculation, as well as quantitative methods. Further, by concentrating on knowledge, the model draws attention to the role of expertise and expert knowledge within governance. Thus, although the governance models of O’Malley (2003) and Ericson and Doyle’s (2004a) provide a comfortable starting point for understanding the relation between insurance and risk, the model of Miller & Rose (1990) was preferred due to its concentrate on the role of knowledge and expertise.
Miller & Rose’s (1990) Model of Governance
Miller and Rose (1990: 3) argue that governance highlights “the diversity of powers and knowledge entailed in referring fields practical and amenable to intervention”. Building on the work of Foucault (1978: 102) who argues that governance is a “complex form of power”, Miller and Rose (1990) state that governance is a form of power exerted for the purpose of intervention. In relation to insurance, risk is the field which is amenable to intervention.
Governance is able to make fields practical and amenable to intervention by exerting “technologies of government” (Miller and Rose 1990: 8), these refers to:
Mundane mechanisms which appear to make it possible to govern: techniques of notation, computation and calculation; procedures of examination and assessment: the invention of devices such as surveys and
presentational forms such as tables; the standardisation of systems for training and the insulation of habits; the inauguration of professional specialisms and vocabularies… the list of heterogeneous and is, in principle, unlimited. Miller and Rose (1990: 8)
The ‘technologies of government’ listed above are the ways in which insurance intervenes in risk and renders it practicable. Miller and Rose (Ibid.) build on Foucault’s (1978) idea relating to how governance is achievable as a result of specific forms of knowledge. Governance is:
Always dependent on knowledge, and proponents of diverse programmes seek to ground themselves in a positive knowledge of that which is to be governed, ways of reasoning about it, analysing it and evaluating it, identifying its problems and devising solutions. Miller & Rose (1990: 7).
Knowledge is the factor that enables an object to be conceptualised in a fashion which allows a set of processes and relations to be developed for the purpose of governance. Miller & Rose’s (1990) concentration on knowledge means that they also highlight the importance of expertise in governance. Expertise is defined as “the social authority ascribed to particular agents and forms of judgment on the basis of their claims to possess specialised truths” (Idem 2). Expertise encompass “the complex amalgam of professionals, truth claims and technical procedures (which) provide versatile mechanisms for shaping and normalising the ‘private’ enterprise” (Idem 8). Miller and Rose (1990) see expertise, something which is socially prescribed, as the factor which ultimately enables governance.
In application to insurance, expertise are the commodity which enable insurers to conceptualise risk in a way that enables governance.
‘Knowing’ an object in such a way that it can be governed is more than a purely speculative activity: it requires the invention of procedures of annotation, ways of collecting and presenting statistics, the transportations of these to centres where calculations and judgments can be made and so forth. Miller & Rose (1990: 5)
Insurers posses unique knowledge which enable them to understand risk in a certain fashion. This understanding transcends into the construction of technologies of government which intervene in that risk and enable it to become governable.
The below table was created in order to demonstrates how the governance model of Miller and Rose (1990) is applicable to the relationship between risk and insurance. Insurance is a technique of governance which enables intervention in risk. Intervention is achieved via technologies of government which consist of quantitative and qualitative mechanisms which rely on expertise.
The model of governance offered by Miller & Rose (1990) provides a comprehensive approach to understanding the relation between insurance and risk, however, it is particularly useful in application to terrorism. The model of governance provided by Miller & Rose (1990) is productive because it is able to account for a broad range of methods, implemented by insurers, which enable terrorism to be insured. The model offers a richer framework relative to Beck’s (2002) approach, which is based on scientific rationality and has a particularly strong emphasis on quantitative risk calculation.
As covered in the previous Chapter, terrorism is considered by insurers as a particularly complex risk. This is due to the lack of historical loss data, combined with the human element, which makes the frequency and severity of a future attack difficult to predict. Thus, given the absence of sufficient data, scientific rationality proves difficult to achieve. The unpredictable nature of terrorism thus complicates its insurability and has pushed insurers to apply innovative solutions to enable its insurability. The model of Miller & Rose (1990) is able to account for these solutions, which include both quantitive and qualitative techniques. Most importantly, by concentrating on the importance of knowledge and expertise, the model is also able to account for the role of human judgment in risk decision making. The ability of Miller & Rose’s (1990) model to encompass a diverse range of mechanisms thus makes it useful in answering the question of how insurance governs the risk of terrorism.
The complexities involved in insuring terrorism raise the important question of why companies have an interest in governing it. In order to answer this question attention will be drawn to the work of Ericson and Doyle (2004) and Ericson, Barry and Doyle (2000) who argue that insurance companies govern risk for the the dual purpose of economic gain and societal risk reduction.
Governing for economic gain
This thesis is in line with the argument that insurers govern risk for the purpose of achieving economic gain. Profits are the reason for why insurance companies “assume liabilities of the unknown” (Ericson and Doyle 2004b: 5). In the instance of a no loss scenario, “insurers can thrive on conditions of extreme uncertainty” (Ericson and Doyle 2004a: 148). Insurers turn a profit by gaining premium payment in return for issuing a policy that insures against risks, such as terrorism. The argument that risk governance is driven by economic initiative is supported by Foucault’s (1978) model of governmentality which refers to:
The ensemble formed by the institutions, procedures, analyses and reflections, the calculation and tactics that allow the exercise of this very specific albeit complex form of power, which has at its target population, as its principal form of knowledge political economy, and as its essential technical means apparatuses of security. Foucault (1978: 102)
Governmentality is similar to governance in that it can be understood as the organised practices which enable an object to be governed. Foucault (Idem 92) argues that the art of governmentality “is concerned with answering the question of how to introduce economy…to govern a state will therefore mean to apply economy”. Foucault (1978) elaborates:
The new science called political economy arises out of the perception of new networks of continuous and multiple relations between people, territory and wealth; and this is accompanied by the formation of a type of intervention characteristic of government, namely intervention in the field of economy and population. Foucault (1978: 101)
In this light, political economy is seen as ‘“the science and the technique of intervention of the government in that field of reality” (Idem 102). By applying Foucault’s (1978) approach to governance it can be argued that insurance intervenes in risk because it is a means to economically organise the population. This argument is best explained by viewing insurance as a security commodity.
Insurance can be understood as taking advantage of the risk of terrorism being present by providing a product which is able to offer security. “The risk society becomes intertwined with the security society” (Davoudi 2015: 466). The security society can be understood as a response to the risk society. “Concerns about risk usher in deep anxieties about security (Idem). The ambivalence created by the risk society, which sees an increase in “new risks and side-effects” (Beck, Bonss, Lau 2003: 17) has produces “the desire for neutralisation or avoidance of risk” which has at its core “the rationale for relentless pursuit of security (Davoudi 2015: 466). The fact that the amount of terrorism insurance policies purchased globally increased from 27% in 2003 to 58% in 2005 (Insurance Information Institute: 2016) suggest that the risk society has motivated the quest for security products that are able to diminish risk.
Terrorism is one of the axes of evil that make up the risk society (Beck 1992). In response to the existence of terrorism, insurers sell a product that is able to offer protection against the financial burden that an attack can inflict. Through the provision of terrorism insurance, companies are intervening in both the economy and population. The population has an interest in purchasing terrorism insurance because they are concerned with the economic damage that an act of terrorism can cause. Similarly, insurers have an interest in selling policies because they perceive that the amount of premium payments received will outweigh the cost of the number of potential claims caused by acts of terrorism. Thus, insurance is a governance mechanism that is able to capitalise on the risk of terrorism being present by offered the population a security product that can diminish the impact of the risk.
Insurance as a form of security governance
This thesis is not suggesting that insurers govern risk for economic gain alone. This paper does support the view that insurance is a key institute in the provision of risk reduction initiatives against terrorism. Insurance companies govern risk because they have assumed a role as security provides (Ericson and Doyle 2004b). “Insurers have an enormous social responsibility to assess and act upon risk” (Ibid: 5), they are “ a central institution of governance beyond the state” (Ericson, Barry and Doyle 2000: 533). This view is echoed by Peterson (2008: 403) who argues that:
In today’s fight against terrorism, national security concerns cannot be reduced to external threats and handled by diplomatic or military means alone. Counter- terrorism demands the active involvement of civil society, including private companies.
Specifically in relation to insurance, “due to national security concerns demands have been made for …insurance companies to provide terrorism risk coverage” (Ibid). “The importance of insurance as a global security technology cannot be overstated” (Lobo-Guerrero 2011: 1). Insurance has an important role to play in both risk mitigation and risk transfer.
Application of theory to research - how does insurance govern the risk of terrorism?
The thesis has now presented the argument that insurance is a technique of governance implemented for the dual objective of economic gain and societal risk reduction. The next phase of the thesis will involve answering the question of how insurance governs the risk of terrorism. In order to achieve this, this thesis will empirically analyse two technologies of government, as well as expertise.
The first technology of government to be analysed will be calculation. Calculation is one of Miller & Rose’s (1990) prescribed quantitative technologies of governance. Calculation has been chosen for study because there is a consensus within academic circles (Beck 2002, Bougen 2003, O’Malley 2003, Ericson and Doyle 2004a, Ericson and Doyle, A. 2004b, Ericson 2005, Collier 2008) that this is a key technique used by insurers for the purpose of governing risk. Calculation govern risk by making use of statistical information in order to predict frequency and severity. Although there is a consensus among the scholars listed that calculation is an important tool used by insurers to govern risk, none of the scholars empirically analyse how this is done. Thus, the aim of this thesis is to show how calculation is put into practice.
The second technology of government will be risk mitigation. Risk mitigation translates into a technology of government because it can be viewed as a combination of procedures implemented in response to the examination and assessment of terrorism risk. Ericson and Doyle (2004a), Ericson and Doyle (2004b), Ewald (2002) and O’Malley (2003) argue that when a risk, such as terrorism, exceeds calculability, due to its low frequency and high severity, risk mitigation solutions need to be implemented as a form of governance. Risk mitigation is a preemptive security technique which is based off the precautionary principle that aims to eliminate the threat before it materialises. Similarly with calculation, the scholars who list risk mitigation as a governance technique do not elaborate on what precisely these security initiatives entail. This thesis will thus aim to explore what these technique involves. In addition, this thesis also wants to investigate how effective insurers see risk mitigation as a method to govern terrorism.
The third criteria that this thesis will empirical approach is expertise. Expertise have been prescribed by Miller & Rose (1990) as the element which enable technologies of government to be practiced. Expertise are thus the cornerstone of all techniques used by insurers to govern risk. “Governing involves not just the ordering of activities and processes. Governing operates through subjects” (Idid 1990: 18). It is thus important to assess the knowledge and expertise of insurers. In order to achieve this objective this thesis will analyse how insurers gain their expertise. In addition, an analysis will be conducted into the ways that insurers understand the risk of terrorism and practically act upon it.
In addition to the two technologies of governance and expertise, the thesis will also aim to confirm the idea that insurance governs terrorism for the dual objective of economic gain and protecting the population.
In order to achieve the empirical project set out, the thesis will use expert interviews and content analysis as the methodological approach. These methodological approaches will be outlined in the following Chapter.
Methodology
This Chapter will first of all explain why the London insurance market has been chosen as the subject for empirical analysis. Secondly, this Chapter will justify why expert interviewing and content analysis were chosen as the methodological approaches to study the three practices under investigation: calculation, risk mitigation and expertise.
Part 1: London insurance market
The London insurance market has been selected for empirical investigation because it has a reputation as being the global epicentre of insuring complex risks. Ericson and Doyle (2004: 18) affirm that “the London market remains the locus for insuring risks that have proven too difficult to insure elsewhere”. Outside of London, the global insurance hubs consist of the Americas, Europe, Middle East and Africa (EMEA), as well as Asia-Pacific (APAC). The global prominence of the London market as a specialty insurer has awarded it the title of being acknowledged as a separate hub from the rest of Europe. In fact, Lloyds of London, the platform through which the majority of complex risks are insured, labels itself as “the world’s specialty insurance and reinsurance market” which “specialises in unusual risks” (Lloyds of London: 2016). Lloyds claims to possess “an unrivalled concentration of specialist underwriting expertise” (Ibid). The expertise possessed by insurance professionals within Lloyds means that the London market has a “reputation for product innovation” which is achieved through “analytics and technology” and consequently provides the market with a “sustainable competitive advantage” over other insurance hubs (Ibid.).
Terrorism - Specialty Risk
The reputation of London as the global specialty insurer is important to acknowledge when empirically analysing how terrorism is governed through insurance. Unlike traditional insurable risks such as property, terrorism is considered a ‘specialty risk’ within both the global and the London insurance market. Specialty risks can be defined as “types of risks not underwritten by most insurers” (International Risk Management Institute: 2016). The reason that such risks are commonly not underwritten is because “the risk is either considered too great to underwrite or too correlative to effectively
spread” (Rhee 2008: 3). Put simply, risks such as terrorism are considered too complex and high risk by most insurance markets.
The tragic events of 9/11 are a consistent reminder that terrorism is capable of achieving the status of both a systemic and a catastrophic risk, it “has the power to not only change the course of political history, but also the economic order” (Rhee 2005: 437). The major losses incurred by insurers following 9/11 demonstrated that the low frequency and high severity potential of a terrorist attack makes a loss scenario difficult to accurately predict and very expensive to pay off. As a result of the difficulty in predicting the likelihood and cost of a potential claim, many insurance companies do not offer terrorism insurance coverage. Thus, relative to traditional lines of insurance, such as property, the number of insurers offering coverage for terrorism is smaller and hence terrorism is considered a specialty risk. Given that the London insurance market prides itself as the hub which is at the forefront of insuring specialty risks, this thesis has specifically chosen it as the case study for analysis.
Methodology
The first step of the investigation consisted of defining key theoretical premises which were linked to the three insurance practices under enquiry. The purpose of defining each element was to demonstrate how calculation, risk mitigation and expertise can be conceptualised as mechanisms of governance. The definitions were acquired by applying each practice to the governance model prescribed by Miller & Rose (1990). Calculation and risk mitigation were defined as technologies of government, whilst expertise were the factor that enabled these technologies to be practiced. Once definitions had been acquired, each practice was directly linked to insurance. The purpose of creating this link was to highlight how each practice can be seen as a technique used by insurers to govern risk.
The second step involved composing empirical questions that would generate information relating to the theoretical premises. The purpose of posing the questions was to gain key insights and acquire empirical findings that could contribute and build on the already existing literature. Academic work has already researched and presented the argument that calculation, risk mitigation and expertise are techniques used by insurers to govern terrorism. However, this research lacks detail and is missing concrete examples
that are able to demonstrate exactly how these practices work. Further, the majority of insurance terrorism literature concentrates on the US insurance market. Thus, this thesis would like to provide an alternative viewpoint by exploring the functionalities of the London market specifically.
The below table was created for a dual purpose. Firstly, to demonstrate which theoretical definitions were chosen and how these were applicable to insurance. Secondly, to connect the theoretical elements described to concrete empirical questions.
Part 2: Expert Interviewing
Face-to-face systematising expert interviews were chosen as a methodological approach because they allow direct access to expert knowledge which offers an ideal means by which to gain answers to the empirical questions posed. Systematising interviews are defined by Mauser and Nagel (2009: 46) as those “oriented towards gaining access to exclusive knowledge possessed by the expert”. The expert here is treated as “a guide who possesses certain valid pieces of knowledge and information, as someone with a specific kind of specialised knowledge that is not available to the researcher”. (Ibid. 47) Face-to-face interviews allow personal contact between interviewer and interviewee and hence may facilitate increased information sharing relative to non personal means of data collection. Interviewing allows questions to be pre structured and moulded around the specific area of interest. In addition, the close proximity to the interviewee allows the opportunity to present follow up questions, which can significantly enrich data.
Expert interviews were also chosen because “they lend themselves to those kinds of situations in which it might prove difficult or impossible to gain access to a particular social field” Bogner, Littig and Menz (2009: 2). Retrieving answers to the empirical questions presented would prove especially difficult via other means of data collection because the methods used by insurance companies to calculate and manage risks are highly confidential . As previously mentioned, the competitive nature of the insurance market means that insurers are secretive about their underwriting strategies. Thus, in order to gain this information, it is most productive to engage in face-to-face interviews which allow direct access to experts working within the field.
A further reason for why expert interviews were chosen is due to my privileged position as an insurance insider. I have over three years experience working within the London market, this includes exposure to terrorism insurance. At the time of deciding upon methodological approaches I was an employee of one of London’s most prominent insurance companies - XL Catlin. The company is well established within the London market, it was originally established as a Lloyds of London syndicate and has since prided itself as a specialty insurer (XL Catlin: 2016). I have reason to believe that employment at XL Catlin placed me well for engaging in expert interviews.
Firstly, my employment assisted me in overcoming the major problem of “gaining access to the experts” (Littig 2009: 98). As one way to distinguish themselves, experts commonly set up access barriers, “the following rule seemingly applies: the higher the social class, the more difficult access becomes” (Ibem: 104). I believe my employment at XL Catlin increased the willingness of experts to meet with me. “Acceptance of the interviewer as co-expert can be seen as a preliminary contribution made by the expert at the beginning of the interview. It is based initially on vague indications and impressions, and these are implicitly checked as the conversation develops” (Bogner and Menz (2009: 58).
Secondly, my employment enabled me to be in a position whereby I knew who would be useful for an interview. “The success of interview-based investigations considerably depends on the ‘quality’ of the interviewees, that is on the extent to which they meet our expectations of the interview situation” (Glaser and Laudel 2009: 117). My position as an insider meant that I already had a good idea of who would be able to answer my empirical questions. In addition, I had a network in place that would help me access relevant experts.
In addition to increasing my meeting chances, I believe my employment at XL Catlin enriched the data acquired. This is likely to have been because the interviewee and I already shared a foundation of knowledge upon which we were able to base the discussion on. Bogner and Menz (2009: 59) argue that:
Acceptance of the interviewer as co-expert can turn out to be a specific advantage. If the researcher demonstrates his or her specialist interest in the subject, makes use of his or her knowledge and engages in a lively discussion on this basis, the interviewee will be prepared to do the same and to reveal information and knowledge she or he would unlikely to make available if the interviewer’s role were assessed differently and competence attributed differently.
My employment, combined with my specific interest in terrorism insurance meant that I was well prepared for each interview. “For the interviewer it is a must to prepare the interview topics thoroughly and build up a knowledge base of the field the experts are moving in” (Meuser and Nagel 2009: 31). It is important to appreciate that “expert
