When an irresistible force meets an immovable object: The interplay of agency and structure in the UK financial crisis

Journal of Business Research 67 (2014) 2671–2683

Contents lists available at ScienceDirect

Journal of Business Research

When an irresistible force meets an immovable object: The interplay of agency and structure in the UK financial crisis☆ Simon Ashby a,⁎, Linda D. Peters b, 1, James Devlin b, 2 a b

University of Plymouth, UK Nottingham University Business School, Jubilee Campus, Nottingham, NG8 1BB, UK

a r t i c l e

i n f o

Article history: Received 1 June 2010 Received in revised form 1 February 2011 Accepted 1 October 2011 Available online 12 April 2013 Keywords: Financial crisis Risk management Crisis management Structuration theory

a b s t r a c t The study sheds light on why certain financial institutions exposed themselves, and the financial system as a whole, to excessive risk. The study examines the human side of the crisis and its relationship to certain organizational and sector-wide practices dominant at the time. The study draws on pre-existing insights from the field of crisis management, and use structuration theory to explore the inter-relationships between the micro- and macro-factors that contributed to the crisis. Structuration theory allows exploration of how the irresistible force of human agency and the immovable object of situational imperatives together provide an understanding of how and why the crisis occurred. The study argues that the crisis was largely due to failures in the implementation of certain risk management processes. The research findings challenge the notion that greater regulatory prescription and capital requirements are required, or that simple solutions such as caps on bonus payments will prove effective. Rather, implementing enhancements in the risk management and governance practices of financial institutions and their regulators is necessary, together with facilitating mechanisms that support cultural change. © 2013 Elsevier Inc. All rights reserved.

1. Introduction Investigations into the causes of the financial crisis and the lessons the crisis teaches us are plentiful (Brunnermeier, 2009; Congressional Oversight Panel, 2009; de Larosière et al., 2009; Lewis, 2010; Turner, 2009). A common theme from these investigations is that the cause of the most recent financial crisis, like many before, was a series of macro-economic temptations. These temptations (i.e., low interest rates, booming housing and securities markets, and high levels of market liquidity) drove banks (and a few insurers) toward excessive market, credit, and liquidity risk. Examples include heavy exposures to collateralized debt obligations, excessive sub-prime lending, and an over-reliance on short-term funding. Financial institutions were unable to control these risks effectively because of weaknesses in their corporate governance frameworks (for example, excessive reliance on staff bonus payments which promoted moral hazard) and

☆ The authors thank the anonymous JBR reviewers, Christine Ennew (Nottingham University Business School) and Ken Starkey (Nottingham University Business School) for their comments on earlier drafts of the paper. ⁎ Corresponding author at: School of Management, University of Plymouth, Drake Circus, Plymouth, Devon, PL4 8AA, UK. Tel.: + 44 1752 585720. E-mail addresses: [email protected] (S. Ashby), [email protected] (L.D. Peters), [email protected] (J. Devlin). 1 Tel.: +44 115 84 66098. 2 Tel.: +44 115 95 15264. 0148-2963/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jbusres.2013.03.015

risk management arrangements (for example, holding insufficient levels of capital and liquidity). While such factors undoubtedly played their part, and are likely to do so again, these analyses contain a key flaw. Explanations of why some financial institutions exposed themselves to excessive levels of risk, while many others did not are imperfect. As Borio (2008, p. 14) states, “…while it is tempting to address the most conspicuous problems highlighted by the present turmoil, there is a risk of focusing too much on the symptoms, rather than the underlying causes.” Limited research exploring the different behaviors of financial institutions is a key barrier in understanding the many complexities of the crisis. Traditional economic theory currently employed lacks the necessary theoretical techniques addressing these behavioral aspects, as such theory uses mathematical deductivist modeling which, “…can provide limited insight at best into the workings of the economy (or any other part of social reality)” (Lawson, 2009, pp. 759–760). By contrast, more behaviorally oriented analyses of the crisis provide some interesting insights into the human foibles encouraging precipitation (Müßig, 2009; Tett, 2009; Walker, 2009). However, these lack the theoretical and ontological pedigree of the economics-based literature. To date, the literature includes no explicit attempt investigating the interplay between the macro-level factors that determined the structural features of the financial services sector prior to the crisis (for example, economic boom and high levels of liquidity) in conjunction with the social practices of financial institutions and the behaviors of their managers and directors (the agency of individuals).

2672

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

The study seeks to fill this void and contributes to the literature on the crisis in two ways. Firstly, the study draws on pre-existing insights from organizational crisis management. Most commentators on the current crisis do not apply this approach, but the method provides a wealth of information on the common underlying causes of crisis (financial or otherwise). Secondly, the study explores the inter-relationship between macro- and micro-factors (structure and agency) using structuration theory. The insights from structuration theory allow exploration of how the irresistible force of human agency and the immovable object of situational imperatives (organizational routine, structure, and the wider economic and regulatory environment, for example) together provide an understanding of how and why the crisis occurred. Using the insights that follow from structuration theory, coupled with data collected from interviews with 20 risk management and financial service professionals, the study indicates that the current financial crisis was largely self-inflicted and was due to managerial and cultural weaknesses within financial institutions and government regulators. Thus, the irresistible force of human agency (in both managerial action and decision-making) may not only be a positive or negative force in its own right in relation to the emergence of financial crises, but can also interact with the immovable object of structural features and situational factors, in terms of constraints such as regulation and competition. The study also provides evidence that structural features at the institutional and industry levels (for example, compensation arrangements and the prolonged economic boom) can hinder individual reflexivity and critical evaluation and help to reinforce the natural tendency of individuals to pursue their own self-interest by making extreme risks seem both acceptable and desirable. Following the introduction, Section 2 provides a critical review of the established literatures on organizational and financial crises in order to highlight the value of structuration theory as a means for understanding the evolution and management of the global financial crisis. Section 3 covers the research method. Section 4 presents the findings from the interview data. Section 5 expands the discussion's theoretical and practical implications. Section 6 concludes. 2. Understanding crisis 2.1. Organizational crises as a process Though established in the English vernacular, the notion of crisis is hard to define from an academic perspective (Boin, 2006; Roux-Dufort, 2007; Smith, 2006a). Many observers define organizational crises as extreme events that impose immense instability, uncertainty and cost on those caught up in them, whether financial, reputational or physical (cf., Gregory, 2005; Jaques, 2009). These attributes appear to define the global financial crisis very accurately. However, such crisis events (even multi-organization global financial crises) can also serve as valuable turning points, where those organizations prepared to learn from them can achieve beneficial outcomes, often by addressing previously unknown or underestimated weaknesses and inefficiencies to reduce the likelihood and severity of future crises. The development of a deeper conceptualization of crisis is a key theme in the literature on organizational crises and their management. An emerging view is that understanding and managing crises in a holistic manner, one that reflects the causal (before), operational (during), and recovery (after) phases, is necessary. Such a conclusion follows from understanding these phases, constituting the process through which crises form, manifest, and their rationalization (Coombs, 2001; Roux-Dufort, 2007; Smith, 2005; Smith & Elliott, 2006: Section 2; Toft & Reynolds, 2005). As a result, the literature seeks to move away from the traditional event-centered perspectives, whose focus is on business continuity planning and recovery. This

shift occurs because a narrow and reactive approach to managing the operational phase of crises alone can mean that organizations fail to appreciate their underlying causes and hence do little to learn from them. The trend towards a holistic view of crises is instrumental in developing a clearer picture of how and why organizational crises can emerge, and at times intensify their effect. One common element is weaknesses in the human-system interface. These typically emerge due to interrelations between the following: • Cultural and human factors (management risk perceptions, organizational safety and risk cultures, internal politics, and power dynamics). • Organizational design and structure (the complexity of an organization's structure and associated management systems). • Economic and strategic imperatives (external social, political, economic and competitive pressures). A more holistic view also reveals that the most visible and operational phase of any organizational crisis is usually the tip of the iceberg. This merely represents the manifestation of a long-standing incubation process through which causes develop and interact such that some form of major failure is inevitable. Neither is this phase the end, with the process of recovery and organizational learning continuing long after the operational phase. Although the literature on organizational crises provides some invaluable insights into crises and their management, a lack of robust theoretical models remains a constraint (Smith, 2006b). This observation results in calls for a regeneration of the study of organizational crises in order to address the under-theorization of the concept (Roux-Dufort, 2007). A resultant problem is that almost all of the research within the field of organizational crises involves the analysis of discrete crises and disasters, often relying on publicly available secondary sources, such as the results of public enquiries and accident investigations. Hence, the discipline became the (social) science of the exceptional, with little empirical evidence available to prove the general validity of its findings. The problem is especially acute in the new world of the trans-boundary crisis (Boin, 2009; Rosenthal, 2003), where crises are far more likely to be systemic in nature and cross both national and functional boundaries (as in the case of the most recent global financial crisis) rather than being restricted to a single organization and its stakeholders. 2.2. The study of financial crises While criticisms exist of the literature on organizational crises for its lack of theory building and empirical evidence, such criticisms do not apply to the literature on financial crises. This work possesses a long history of association with the discipline of Economics, drawing its legitimacy from a range of established economic theories and empirical methods. The financial crisis literature's ability to explain systemic events is a particular strength. Such events represent an Armageddon scenario for financial institutions and markets, which can lead to large-scale financial losses that not only cross over into non-financial markets (housing and manufacturing, for example), but also damage the economic wealth of nations on an international scale. The key driver for systemic risk is contagion, where financial interdependencies among institutions mean that the losses of one translate into losses for many others, or where a loss of confidence among investors means that they all try to withdraw their funds at the same time. Many studies are available into the causes and management of financial crises, and these studies benefit from comparisons among a wide range of different historical cases (Eichengreen, 2002; Goodhart & Illing, 2002; Kasuya, 2003; Laeven & Valencia, 2008). In the main, these studies found that financial crises are largely economic phenomena, with market-level factors as driving forces, such

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

as: macroeconomic policies and shocks (interest rates, etc.); credit booms; and the poor regulation of information asymmetries and balance sheet fragilities (for example, holding insufficient capital due to weak regulation). The implication is that individual financial institutions can do little to influence the onset and effects of such crises, and must simply ensure that they are prepared for the worst. This observation is contrary to the process-based literature on organizational crises, where organizations act as the agents of their own demise. The suggestion that the literature on financial crises completely ignores the actions of individual financial institutions is perhaps a little unfair, especially as more recent research into the global financial crisis pays considerably more attention to issues like corporate governance, risk management, and pay and bonus arrangements (Acharya & Richardson, 2009: Part 3; Dowd, 2008). Thus, a growing recognition that financial crises can (at least in part) be self-inflicted and may be due to excessive risk-taking on the part of profit-seeking institutions is present. Yet within the literature, investigations of the reasons behind such behavior are lacking, with researchers seemingly content to rely on the neo-classical concept of principal-agent theory and the problem of moral hazard for explanation, arguing that mis-aligned incentives, coupled with asymmetric information, create the motive and opportunity that make excessive risk-taking inevitable. In fact, excessive risk-taking is avoidable, as the financial crisis illustrates, where not all financial institutions (most insurers, for example) and even many banks (in the case of HSBC) chose not to participate in excessive risk-taking, despite the apparent incentives to do so. Of course from a neo-classical economic perspective such behavior does not make much sense, as rational decision makers will not pass up profitable opportunities however selfish and unethical those opportunities might seem. Yet in the real world the homo-economicus view of purely materialistic decision-making is not always correct, and a growing literature that challenges its dominance is available (Akerlof & Shiller, 2009; Michel-Kerjan & Slovic, 2010, for example). Some commentators (for example, Lawson, 2009) question the very foundations of the economic-based perspective of financial crises, arguing that the epistemology of economics is too inflexible and routed in mathematical models to be able to provide a useful lens for understanding their complexities. The literature on organizational crises already possesses a richer set of explanations for such excessive risk-taking in comparison to the literature in the field of economics. However, those researching financial crises ignore these insights (Boin, 2004). Neither does the literature on organizational crises pay much attention to financial crises (Boin, 2004). The reasons for such a lack of cross-fertilization are unclear. However, the different levels of analysis (predominantly micro versus predominantly macro) and philosophies (organization theory versus economics) of these two work streams may provide an explanation. A less charitable view is that many economists refuse to accept alternative world-views (Fullbrook, 2009). This raises the issue of reconciling the apparently disparate views of organizational versus financial crisis researchers to provide a more complete picture of the global financial crisis. One insightful approach for reconciliation is to use the lens of structuration theory. 2.3. The financial crisis as a relationship between agency and structure Structuration theory deals with the creation and maintenance of ideas and structures, as well as with change and continuity processes (Staber & Sydow, 2002). In the social sciences, the relationship between agency and structure is among the most pervading and difficult of issues in social theory (Pozzebon, 2004). Here, agency refers to the human capacity and capability to make unfettered choices of their own volition, whereas structure is the recurring arrangements, patterns, and system restraints that influence and limit individual choice and free will. Structuration theory focuses on the ways in

2673

which social interactions produce and reproduce social systems, views humans as role-taking and norm-fulfilling beings who act according to their images of what reality is, and considers all institutions and social practices as structures (Nonaka & Toyama, 2003). Structuration theory asserts that individuals are active agents – knowledgeable and reflexive individuals – with the capacity to transform their setting through action. However, by placing individuals within a social context the theory suggests that those contexts constrain individual actions, resulting in unintended consequences (Giddens, 1984). Thus organizations or institutions exemplify the duality (or dualism: Archer, 1995) of agency and structure, resulting in practices that stretch over time-space distances in the reproduction of social systems (Giddens, 1984). Institutions persist not only because of the adherence to routines across space and time, but also because those operating within them consent to create and re-create those routines. Thus, agents and structures are intertwined phenomena. The agency of the individuals within social systems recursively organizes the structural properties of these social systems through their continuous flow of reflexively monitored conduct (Boland, 1996). Giddens (1984) argues that the activity of knowledgeable individuals is primarily the result of their immersion in interactions, rather than a deliberate and conscious flow of pre-meditated actions (Parker, 2000). Giddens (1984) termed this flow of regularized and routine actions to create and maintain structures: instantiation. Criticisms of the notion of regularized and routine action include a highly compressed notion of temporality (Archer, 1995), that does not allow for reflectiveness and rational deliberation in human action (Parker, 2000). Nevertheless, regularized and routine action explains why apparently intelligent and capable individuals did not spot the gradual buildup of human-system and economic pressures that preceded the global financial crisis. Rather than exercise their agency to alter the structures that supported these economic pressures and help prevent the crises, they retreated to the safety of known and familiar routines. Indeed, the ability of individuals to negotiate strangeness and the role that structuration can play in helping to understand and manage this ability are identified in the groundbreaking research of Weick (1993, pp. 645-646) on the role of collapsed sense-making within organizational crises (cf. Weick, Sutcliffe, & Obstfeld, 2005, p. 417). As the pressure builds, trigger events can stop or change the normally reutilized, habitual, and taken-for-granted practices of individuals, causing what Giddens (1984) terms a critical situation in which a sense of insecurity results that impacts on institutional time and ontological security (Hardcastle, Usher, & Holmes, 2005). In so doing frameworks (roles, rules, and procedures) and meanings (shared interpretive schemes for understanding the world) can destroy rather than construct and reinforce one another, causing a deviation-amplifying causal loop (Weick, 1993) that leads to a loss of meaning and sense-making. Thus, a crisis (such as the global financial crisis) represents the impact of a critical situation on the pre-conditioned and routinized reality experienced by an individual. In the social system, their actions may spark unintended consequences and their loss of ontological and institutional security may lead to a consequential breakdown in their sense-making capabilities. For example, certain preconditions (such as, low interest rates, booming housing and securities markets, and high levels of market liquidity) encouraged specific patterns of risk-taking by the financial services sector. Sudden alteration to these conditions, due to the pressure exerted by trigger events (for example, the sudden lack of confidence between institutional lenders), resulted in unintended consequences (such as, the lack of availability of borrowing by businesses and homeowners) causing a crisis. A summary of this understanding of crisis appears in Fig. 1, which uses the three main factors identified from within the literature on organizational crises as its basis. In short, the value of structuration theory as a mechanism for understanding the global financial crisis is that the theory is consistent

2674

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

Fig. 1. Crisis as emergent phenomena. [Process flow diagram showing the interactions between the main preconditions of organizational crises and their link to the interplay between agency and structure, these causes build until a trigger event leads to the emergence of opportunity and/or loss].

with the concept of crises as emergent organizational processes, while at the same time providing a mechanism to understand their system-wide (trans-boundary) nature. In so doing the theory can provide greater insight into how pressures built up between agents and structures leading to actions that precipitated the crisis. To understand the buildup of micro- and macro-pressures (and how and why they led to the financial crisis), the study now explores the interplay of agency and structure. The study examines the reflexivity of individuals (the cultural and human factors), their social practices (the factors of organizational design and structure), and their access to resources (the economic and strategic imperatives). The next section outlines the methodology and data collection methods, followed by an analysis of the results. 3. Method 3.1. Approach: semi-structured and confidential interviews To explore the reasons behind the risk management failures that precipitated the current crisis the researchers conducted a series of semi-structured and confidential interviews with risk management professionals from a range of financial institutions. The interviews focused on the opinions of the respondent in relation to: the cause(s) of the current financial crisis; the role of risk management and its implementation; how organizational factors (i.e., culture and governance) contributed to events; and the participant's comments on the future in relation to sector regulation and the understanding and management of risk. The interview questions appear in Appendix 1. 3.2. Scope of analysis: the interviewees The study comprised twenty interviews within nineteen institutions from across the financial services sector. Conducting the interviews in the immediate aftermath of the initial phase of the crisis (July/August 2009), provided an early, first-hand, account of the causes. Appendix 2 contains pseudonyms for each interviewee and a brief description of their role, experience, and type of employing financial institution(s). Using a convenience sample of interviewees contacted via two established networks (a professional risk management institute and an industry-funded research forum), the researchers also included other respondents recommended by the initial sample. This technique proved successful in selecting interviewees from a broad mix of institutions (for example, large/small, commercial/investment, bank/insurer, independent/government owned). The individuals had considerable experience in the financial services sector having

worked for a range of institutions over their careers, including, in some cases, the UK regulator (the Financial Services Authority). The focus on interviewees who worked as or who had a background in risk management is significant since these individuals are likely to be the best placed within their institutions to comment on the causes of the financial crisis and lessons for the future. Risk managers benefit from a broad umbrella view of financial institutions, including their motivations and management practices (the typical risk manager, especially at the functional head or chief risk officer level, works across all areas of an institution). They are likely to adopt a more balanced view of the risk-taking behaviors of their institutions, since for the most part they do not get involved in front-line decision-making (trading, loan approvals), nor do performancerelated bonuses play a large part in their remuneration packages. Indeed the interviewees proved to be surprisingly critical, both of their profession and of the management of financial institutions in general — suggesting that interviewee response bias is not a significant problem within the interview analysis. 3.3. Method of analysis: examining the data Given the sensitive nature of the material, the researchers guaranteed confidentiality to respondents. With the exception of one respondent, the interviewees agreed to the recording of the interviews. For the one interview not recorded, the researcher made detailed notes at the time of the interview and added further reflections soon after. Recording and transcribing of all but one of the interviews allowed for more detailed data analysis using accurate quotations from the respondents that helped to convey their meaning and intentions more fully. The analysis uses pattern matching (Campbell, 1975; Denzin & Lincoln, 1994; Eisenhardt, 1989) by comparing empirical patterns between each of the participants interviewed. Initially, two of the researchers read each transcript to ensure a high level of data familiarity, and a holistic understanding of each transcript (King, 1994). Identification of key themes resulted from establishing similarities, dissimilarities and recurrent words and themes, and by noting patterns by volume or significance (Beech, 2000, p. 213). This process was firstly carried out on a within-case basis, and then on a between-case basis. Subsequent analysis proceeded by drawing together data concerned with similar themes from different cases, and by placing the emergent themes in three main data categories (cultural and human factors, organizational design and structure, and economic and strategic imperatives). Using the proportional reduction in loss method (Rust & Cooil, 1994), the reliability of the data categorization was 85%. The researchers presented the results in the form of a summary report to a convenience sample of financial services practitioners (including risk managers, government departments and financial regulators),

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

inviting further comments to allow confirmation of the validity of the findings. This approach is in line with recommendations by Carson, Gilmore, Perry, and Gronhaug (2001) and, importantly, helps to combat the possibility of researcher bias and enhances the validity of the findings. 4. Findings This section explores the underlying causes of the crisis using structuration theory coupled with the responses provided by the interviewees. The objective of the analysis is to explain why certain financial institutions exposed themselves to excessive amounts of risk, while others did not. Analysis of the interview data attributed the underlying causes of the crisis to six key factors. Termed the six Cs, the factors were: compensation, culture, complexity, communication, competition, and capital regime. Each of the six Cs relates to one of the common underlying preconditions identified in Fig. 1. Reflecting cultural and human factors, the two Cs of compensation and culture relate to the reflexivity of individuals. Reflecting organizational design and structure factors, the two Cs of complexity and communication relate to social practices. Reflecting economic and strategic imperatives, the two Cs of competition and capital regime relate to access to resources. The study now examines the six Cs in further detail. 4.1. Cultural and human factors: the reflexivity of individuals This section examines the two Cs of compensation and culture. According to structuration theory, individuals draw from pre-existing structures bounded by pre-existing rules. However, structuration does not imply that individuals are slaves of existing structures; individuals have the power to act otherwise and thus the interaction of individuals often instigates change. In this respect, human behavior is intentional and purposive and, through reflexivity, individuals may change the knowledge that they use to guide their action (Parker, 2000, p. 60). Reflexive individuals take action in order to modify and redefine structures in ways that will offer different possibilities for future actions. However, reflexivity does not imply that the motives, conditions, and the consequences of their actions are readily understood (Berends, Boersma, & Weggeman, 2003). This conclusion follows from the understanding that unacknowledged preconditions and the unintended consequences of action form the bounds of knowledgeability (Giddens, 1984) and play an important role in the production and reproduction of structure. In making sense of the communications and actions of oneself and others, individuals draw upon interpretative schemes that help them produce and reproduce structures of meanings. In structuration theory, Giddens (1984) terms this process of producing and reproducing structures of meanings as the rules of signification. Signification processes are the rules that help us to understand “how we do it in this organization” and the communication of those rules to others (Staber & Sydow, 2002, p. 412). Rules of signification create symbolic interpretative schemes that facilitate communication during interactions. These include how and what language is used, and the expectations regarding the use of other signs (clothing, symbols and tools, for example) to communicate. Signification represents how things appear to be (Callahan, 2004). Thus, systems of signification allow agents to communicate with each other through the application of interpretative schemes (Giddens, 1984). The study now assesses the data and how the cultural and human factors that contributed to the financial crisis (the two Cs of compensation and culture) relate to these reflexive and signification processes. 4.1.1. Compensation Blaming the crisis on greedy bankers is a popular thread in much of the publically available commentaries on the crisis (Congressional Oversight Panel, 2009; House of Commons Treasury Committee, 2009).

2675

However, not all of the interviewees agreed that the excessive risk-taking that preceded the crisis resulted from high salaries or bonuses alone. In fact, many of the interviewees do not receive large bonuses, and so have little self-interest in this area. Often, the respondents felt that the crisis emerged because of several factors combined, in particular remuneration regimes together with weak corporate governance and the complacency born of a boom economy. It [remuneration] was a brick in the foundations of what has happened rather than a cause. Remuneration on its own wouldn't drive the current turmoil but alongside weaknesses in corporate governance, the fact that risk management practices were not being used effectively and good times were being had by all so an element of complacency crept in — that along with remuneration all combined to create the issues we've got. [Finance director 2] Here, the key issues are threefold. First, incentive arrangements encouraged short-termist and sales-driven behavior. Second, these incentive arrangements did not claw back salary or bonuses when losses were incurred, meaning that key decision makers did not have an enduring financial stake in the decisions that they were making. Third, these incentive arrangements often failed to reinforce, or even contradicted, the governance arrangements of some financial institutions (for example, arrangements that did not incentivize managers to assess, monitor, or control risk properly). Fundamentally, the above flaws in compensation arrangements hindered reflexivity and critical evaluation in those involved, and helped to reinforce the natural tendency of agents to pursue their own self-interest. Thus, incentivizing agents in receipt of such compensations (for example, directors and senior managers) exposed their principals (typically shareholders) to an excessive amount of risk. These incentives supported interpretive schemes in individuals that made such risks seem both acceptable and desirable. Thus, the rules that applied to their compensation arrangements allowed these agents to share in the upside (profitable) states of the world that accompanied their decision-making, but not bear the full costs (or possibly even none of the costs) associated with loss-making states. I think this is really important actually because you've got people on mighty salaries running large organizations for relatively short periods of time and they don't have ‘skin in the game’. People have walked away with fortunes off the back of running enormous organizations without managing risk properly… If you are running a small business it's easier to understand it, if you've got skin in the game then I think you'll be more careful with it. I would say, and I'm no expert in this area, that oldstyle city investment banks which were based on partnerships were probably more acutely aware of the need to manage that business as their own, because it was their partnership, they really had skin in the game. Whereas when you get corporate, I just think there's more scope for mismanagement or a lack of attention to that kind of detail. [Head of operational risk 2] Thus, individuals interpreted these compensation schemes in ways that signified the moral acceptability of high-risk, short-termist, and personal gain. Hence, the redesign of compensation arrangements for some financial institutions is necessary. Further exploration of this idea appears in Section 5. 4.1.2. Culture The role of an organization's cultural weaknesses in causing crises is a major theme within much of the established research into crisis management. Prior research shows that a crisis-prone organization typically possesses a culture that is very different to that of a

2676

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

crisis-avoiding organization (Pauchant & Mitroff, 1988). Interviewees characterized these as two types of financial institution. Firstly, those that had aggressive sales- and market-share-oriented cultures, frequently run by dominant and egotistical executive directors (characterized as crisis-prone institutions that took excessive risks). Secondly, those that escaped significant losses did so by maintaining more traditional, prudent, and conservative cultures (characterized as crisis-avoiding institutions that exposed themselves to a more appropriate level of risk). The crisis-avoiding institutions also had directors and employees prepared to take a long-term view rather than striving for short-term growth. Expressing a crisis-prone culture: The culture of the whole organization was sales driven. The rewards and promotion were all internal and rapid. People were hugely rewarded, with lots of share options and bonuses. The risk people were disaggregated, not rewarded, so didn't have an integrated risk function.... There was no incentive for any of the risk people to stand up and challenge and there are stories of those who did challenge the groupthink that were heavily penalized as a result. [Chief risk officer 3] The prior observations regarding compensation highlighted an observable manifestation of this culture in the interpretive schemes that guided the rules and resources related to bonuses. Clearly, culture plays an important role in supporting the signification of more appropriate interpretive schemes over less appropriate ones. From the point of view of structuration theory, the drivers behind crisis-prone (excessive risk-taking) cultures are also significant. Further analysis of the data reveals that the immovable object of situational imperatives at the institutional, industry, and society-wide levels influenced what the authors term the irresistible force of human agency. At the institution level, respondents identified three core drivers. Firstly, the attitudes and risk perceptions of directors and senior managers. This manifested as a tone from the top; boards did not always pay sufficient attention to risk management considerations or were reluctant to challenge high levels of risk-taking while the profits were rolling in. Secondly, the presence of a blame culture where staff (including risk managers) were unlikely to challenge accepted signification processes (for example, cultural norms surrounding excessive risk-taking) because they did not feel able to “bring out their dead” (chief risk officer 6). Thirdly, the effectiveness of human resource practices where, for example, a failure to retain experienced staff can mean that an institution loses those individuals with experience of previous crises, who may be most likely to make sense of future ones. Equally, a failure to communicate risk management objectives in training and induction programs, or the absence of regular communication between risk management and other business functions can lead to a lack of understanding between these two groups, promoting conflict. Interestingly, many of the interviewees talked about the potential for conflict between risk management and other business functions. They commented that risk functions often adopted interpretive schemes that reflected “ivory tower” (head of operational risk 2) and/or officious and compliance-oriented cultures instead of business-focused perspectives devoted to supporting both strategic and operational decision-making. In so doing, risk management functions are much more likely to be subject to signification processes that leave them marginalized and ignored, meaning that attention to valuable messages about an institution's risk exposures are lacking. At the industry and society-wide levels, interviewees indicated that a number of long-term economic and political structural changes reinforced the behavior of the crisis-prone institutions. This finding is consistent with the view of Lawson (1997), who highlights the effect of structural changes within the macro-level environment

(market level and social changes, for example) on business norms and practices. In particular, the interviewees highlighted the increasing dominance of free market ideologies and laissez-faire government coupled with a prolonged economic boom. Just prior to the crisis these factors culminated in a very strong collective belief that the good times would continue almost indefinitely, a belief strengthened by ill-advised (at least with the benefit of hindsight) political statements from governments claiming to have conquered the traditional cycle of boom and bust. One of the interviewees likened the situation to a “drug crazed party” (finance director 2) including financial institutions, governments, and much of society (for example, homeowners). Thus, the wider industry environment (which held strong collective beliefs of invulnerability) discouraged reflexivity in the individuals engaged in these behaviors. Optimistic economic forecasts signified these beliefs to the industry, and legitimized them through the development of soft regulatory interventions. Chief risk officer 5 explained the situation particularly clearly: I suppose they [politicians] got wrapped up in their self-belief that they knew everything — Gordon Brown saying ‘I've ended boom or bust’ and making a statement like that lead everybody to believe it would go on forever. People were saying ‘we've taken these risks, nothing's happened, Gordon's saying no more boom or bust so we'll carry on’. Nobody thought another bust would come along, partly because he said so. The government rode the wave of the whole global upswing. We'd had a long period of boom but these things come in cycles. Everybody thought perhaps it's not going to be like that anymore. Perhaps the implicit and perceived protection enjoyed by financial institutions (that felt that they were either too big or too numerous to fail) helped to fuel the party. In fact, a common argument in the economics literature on financial crises is that the potential for government bailouts can help to promote moral hazard and risk-taking (for example, Dowd, 2008). However, the interviewees did not mention this argument. 4.2. Organizational design and structure: social practices This section examines the two Cs of complexity and communication. Structuration theory defines practices as the recurring and regularized actions of individuals situated within a social system that create and recreate that system (Berends et al., 2003). Thus, in structuration terms practices have a dual nature. On the one hand, practices are routinized activities that remain part of (and enable and constrain) the structuring of social systems. The study defines this structuring of social systems as the immovable object of situational imperatives. On the other hand, reflexive individuals (with their own perceptions and experiences) carry out these practices with what the study defines as the irresistible force of human agency. However, this force is not without constraints. As Archer (2000, p. 468) states, “…how it [social worlds] goes on is profoundly conditioned by the vested interests a given structure has distributed prior to current action sequences.” As individuals sanction routines, rules and practices they draw on norms or standards of morality, and thus maintain or modify social structures through what Giddens (1984) terms the rules of legitimation. Legitimation, defined as the rules and norms that help us to know that “what we should do and how to do it in this organization” (Staber & Sydow, 2002, p. 412), is fundamentally about how things should be, and creates an atmosphere in which things seem correct and appropriate (Callahan, 2004). Sanctions are the outcome of following, or not following, the rules of legitimation. Thus, individuals influence social practices through their collective sharing and interpretation of knowledge and their perceptions of organizational purpose and action. However, the same individuals draw upon the

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

rules and resources of the organization's wider social practices, shared behaviors, and norms to establish a common culture. Thus, systems of legitimation permit the sanctioning of interaction through the application of norms (Giddens, 1984). The study now explores how organizational design and structure (the two Cs of complexity and communication) relates to sanctions and legitimation processes. 4.2.1. Complexity Financial services are a complex business, becoming more complex over the past few years. Most notably lending represents a very diverse activity, with many specialist (and more risky) products becoming available, such as mortgages for so-called sub-prime customers who previously were unable to borrow money to buy a house. In addition, the sector witnessed significant growth in activities such as securitization (the packaging of loans as offerings and their sale around the financial system). This creates interdependencies between financial institutions on a level that crosses sectors and is much more global than before (for example, via the now infamous collaterized debt obligation, or CDO). Many analyses of the financial crisis attribute a major role to these structural changes, and to the role of complexity as both an underlying cause and an intensification mechanism that helped to make the crisis much worse than anyone expected (Brunnermeier, 2009; Congressional Oversight Panel, 2009; de Larosière et al., 2009). The argument is that this increase in complexity created the perfect environment for unforeseen correlations and extreme events, and institutions more easily exposed themselves to excessive amounts of risk. One key dimension is the significance of market liquidity; a phenomenon once taken for granted by many financial institutions who prior to the financial crisis assumed that whatever the risks they took a plentiful supply of cash would be available to finance their operations: In a way the leverage [access to finance] element of this is the ‘black swan’… In the 90's we had shed loads of capital and we didn't know what to do with it. I had an employee who spent a year learning credit derivatives and then came back and we started doing it. At the time leverage wasn't important it was the hedging aspect [via credit derivatives]. What none of us saw, and I put my hands up, was the leverage aspect. [Non-executive 1] These structuring rules and norms, which assumed easy liquidity and access to resources, and legitimized the assumption that this exonerated taking extreme risks, provided a situational imperative (the immovable object of structure) that drove behavior in the face of this complexity. While accepting that the financial services sector is highly complex and interdependent, most of the interviewees did not see the financial crisis as inevitable. Rather, they highlighted a failure on the part of some institutions to manage the complexities of their risk exposures in an effective manner. The following quote illustrates this sentiment: People forgot the basics of banking …‘Banking 101’ is ‘don't lend if you can't get it back’ and ‘Banking 102’ is ‘if you are going to lend long and borrow short you are taking a huge risk’. People forgot that one as well! I was taught this literally in my first week of banking. [Risk consultant 4] The quote is particularly illuminating, highlighting that even in complex environments certain basic rules or principles apply, with well-established deeper norms legitimized over a long period helping to reduce the associated risk exposures. This observation draws parallels with the work of Weick (1993) on the Mann Gulch disaster, who observes that when meaning becomes problematic core human traits such as intuition and wisdom become important drivers for the creation of new more effective meanings. Hence, while the complexity

2677

associated with the financial services sector played its part, the current crisis was far from inevitable. Rather, the ability (or lack of) management to cope with this complexity was key. Many of the interviewees indicated that failures in the management of systemic risk events were at fault, in particular weak stress testing and tame scenario analysis. Often, this involved tests and scenarios that were insufficiently extreme. In addition, management failed to consider the correlations between certain risk types (notably market and liquidity risk). Indeed, failure on the part of many institutions (not to mention quite a few regulators) to consider the effects of particular risk events on the financial sector as a whole, was in itself a systemic problem. The legitimation of weak sanctions in terms of risk management, with insufficient understanding of actions and consequences, left many in the industry dazed and confused when the crisis struck. 4.2.2. Communication Along with culture, weaknesses in communication are a key element in the incubation of many crises (Shrivastava, Mitroff, Millar, & Miglani, 1988; Toft & Reynolds, 2005; Weick, 1993). In the context of the financial crisis, two factors influenced the propensity for excessive risk-taking. Firstly, whether the right data are communicated to the right people (for example, whether key decision makers such as boards and senior managers receive the information they need to make effective strategic decisions and or avert an impending crisis). Secondly, the issue of whether those receiving this data are able to understand the meaning of the information. The interviewees had much to say on both points. For example: I can only speak for a few firms. I think they get on the whole accurate and reliable information, but is it the right information? [Chief risk officer 5] I don't think boards in general get sufficient information on [risk] exposures — because they either don't want or understand it. [Head of operational risk 3] I think it was happening in most places [risk reporting to boards] …. What the boards could not know or understand was how dubious some of the models were. [Non-executive 1] The interviewees also revealed that attributing all of the blame for excessive risk-taking to the boards of financial institutions is too simplistic. In some cases, management (including the risk management function) was to blame for not presenting meaningful information. This situation may be because management did not have the tools or expertise to do so, or because filtering mechanisms deliberately withheld information from boards, often legitimized, sanctioned, and promoted by executives and senior management eager to dilute potentially bad news. Thus, they diluted information as this information passed up the organizational hierarchy in order to minimize the potential impact: One of the big problems is whether boards have relevant and accurate info to be able to monitor a firm's risk exposures. Risk reports get diluted at every level as they go up until eventually something that is presentable, won't upset anyone, or rock the boat, is presented to the board. We've all seen that. [Risk consultant 2] What boards get may be accurate and viable but I don't think it tells the whole story. I think the editing is partly caused by the volume [of information], and people like reports that tell them what they like to hear and are comfortable with. An example of that would be in a project I've just undertaken, being told by

2678

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

senior and junior managers ‘we can't put bad news up the line, they don't want to hear it and if I take bad news to the board they're going to shoot me’. That affected reporting, how they managed risk… [Risk consultant 3] Management at all levels legitimized and sanctioned incomplete and insufficient information. Establishing social practices by sanctioning the presentation of incomplete and insufficient information thus legitimized the role of risk managers as information guardians. Reflexively, they began to see themselves as assessors rather than managers. The legitimized culture of many risk management functions was compliance oriented rather than business focused, which further reinforced this stance. This view led to a focus on the justifications for these assessments, along with their technical merit and ability to satisfy regulation, rather than the strategic and managerial implications of the risk level adopted: It's very difficult for any risk professional to be specialized in all these different areas … Then you've got the complications of all the derivatives and CDOs and all that — you've got to have a PhD to understand what they're doing. If there's someone who can understand what's being done and have that knowledge and convert it into a business language that the board can understand — I doubt that very rarely exists. [Chief risk officer 5] Interestingly, respondents recognized that a focus on mechanistic assessments led to the reverse problem, with some corporate boards overburdened with information, preventing them from seeing the wood for the trees. To counter this, a heavy reliance on tools such as mathematical models was common, which was in itself a problem (Turner, 2009), as overly complex models can provide a misleading indication of an institution's true risk exposure: I don't think the output of models was generally well understood at all. I think that is the domain of the ‘quants’ that build the models in the first place and are very clever and that is their reason for existence — to appear to be very clever. Do they serve any useful purpose in helping to run an organization? You only need look at today's battlefield to see the results of that — they clearly weren't. It's the danger of putting too much faith in models — models are only right some of the time and are only useful some of the time and people need to understand that at the outset. It's the application of a little common sense. It's like these people who use a sat-nav when they're in the middle of a pond — just look up once and you'll see the pond in front of you. [Risk consultant 2] In short, the use of potentially flawed models became routine with boards and senior management using the output from these models to make key strategic decisions about their institution's risk exposures without exercising sufficient critical evaluation or challenging their assumptions. Again, the immovable object of situational imperatives can drive the irresistible force of human agency in directions other than those intended. 4.3. Economic and strategic imperatives: access to resources This section examines the two Cs of competition and capital regime. Structure, in the structurationist perspective of Giddens (1984), consists of rules and resources. Individuals use appropriate rules and resources to give form to situations of action by interweaving interpretive meaning, normative sanctioning, and power. Rules and resources do not do anything, but exert their effect through individuals knowing and using them, exercising their agency within a situated context (Parker, 2000). According to Archer (1995), structures

are the outcomes of past agency and may emerge over time to become relatively autonomous and durable conditions of action, which is how individuals come to know and use them. Rules may be either interpretive (govern the way individuals interpret the world, that is, the cognitive aspect of social structure, termed signification) or normative (regulate the legitimization of actions), as already discussed. Resources on the other hand may be either authoritative (related to people and power dependency) or allocative (related to objects; tangible or economic). Individuals use these resources in a social system to exert power over other people and objects. While allocative resources relate to capabilities that generate command over objects, goods, or material phenomena, authoritative resources involve the organization of social time-space and the production and reproduction of relations between human beings in mutual association, and the organization of chances for self-development and self-expression (Fuchs, 2003). Power is the result of what people do with authoritative or allocative resources (Callahan, 2004) and accompanies action, thus implying that both the individual and structures are integral aspects in power analysis and are mutually dependent (Hardcastle et al., 2005). The way in which individuals utilize power in interactions through the ability to allocate material and human resources is a process that Giddens (1984) terms the creation, reinforcement, or change of structures of domination. Domination is the way in which control over resources is available to individuals, and the way in which they use facilities to mobilize available resources (Staber & Sydow, 2002). Thus, in structuration theory power is an aspect of structure that is subject to the process of domination by individuals exercising agency. Systems of domination enable individuals to affect each other's conduct via the exercise of power through the application of facilities such as rules and resources (Giddens, 1984). However, power might also accrue to individuals because of their positions in hierarchies or their membership of collectivities, and thus some variability of agency (Mouselis, 1999) in relation to power is possible. The relevant systems of domination that influenced the financial crisis are those of competition and regulation. 4.3.1. Competition Significant attention is devoted to the role of competition in causing the excessive risk-taking that preceded the financial crisis (Acharya & Richardson, 2009; de Larosière et al., 2009; Turner, 2009). One common argument is that low interest rates, coupled with a sustained economic and housing market boom, led to three interrelated phenomena: (1) rapid asset growth; (2) increased competition; and (3) reduced margins from traditional business avenues. As a result many financial institutions took progressively bigger risks (by growing their sub-prime and buy-to-let mortgage books or by purchasing credit securities and derivatives) in order to continue to grow and or maintain high levels of return. Perhaps unsurprisingly many of the interviewees made comments in relation to the issue of competitive pressures. A key theme was that blaming excessive risk-taking on greedy or irrational/incompetent managers (i.e., agency) was too simplistic an explanation. Instead, some very real and rational economic pressures (i.e., structure) often drove institutions to make risk management decisions: Some people say a bonus culture — I don't think it's bonus culture, I think it's a real desire for shareholders, stakeholders, and the analyst community to get returns and their consistent profits. You get very little credit for having a steady ship and being consistent, it's much more about differentiating from the rest of the market and getting the best margin. [Project director 1] Equally, increased competition drives some institutions to take short cuts in their risk management activities in order to save time

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

and money. Respondents observed that the opportunities for large profits and losses resulting from simple trading (taking on market risk) were giving way (with the near-perfect availability of market information) to extracting profits from the invention of more complex products and from more efficient processing of transactions. Thus, structures of domination tended to encourage and reward greater access to allocative resources by reducing the power of authoritative resources (managerial control) to manage risk. The interviewees did not overlook the role of human decisionmaking (the irresistible force of agency), pointing out that many financial institutions had taken on excessive amounts of risk with very little appreciation of even common sense limits, as one interviewee explained: I'm afraid that the old saying if it looks too good to be true it probably is. [Director of regulation 1] Thus, certain financial institutions (of their own volition and due to management incompetence), took on more risk than was required to satisfy the immovable object of the competitive environment that they faced. A further related observation was that while competitive pressures played a part in dictating the risk appetites (risk limits, targets, etc.) of many financial institutions, some had much better risk appetite frameworks than others, allowing them to control their risk exposures in a more effective manner. If I look at Dick Fuld [CEO of Lehman Brothers], for example, there was a man who'd been working for the same company for many, many years and he reckoned he knew pretty much what happened in his company. I would guess that Lehman's didn't actually have a heavy formalized structure for determining the corporate risk appetite, let alone link that with what's actually happening on the ground on a day-to-day basis. Whereas I would guess an organization like JP Morgan would have far more solid, identifiable systems in place, where they would effectively be back testing all the time as that's the only way you can get to know whether your risk appetite is working right. I think there are companies at both ends of the scale. [Director of regulation 1] This comment emphasizes the observation that not every financial institution followed the trend for increased risk-taking in the pursuit of higher rewards. The key features of such institutions were twofold, the quality of their risk appetite framework and associated risk-reporting structures, and an organizational culture that embodied the traditional financial services virtues of prudence and conservatism. Thus, where a financial institution implements an effective risk appetite framework and/or where a strong internal risk-aware culture is present, resisting external market pressure is possible. Hence, the more prudent processes of legitimation and signification can moderate the more extreme behaviors encouraged by structures of domination. 4.3.2. Capital regime While all of the interviewees accepted that many financial institutions must shoulder some of the blame for the current crisis, they also spoke about the structure of the current capital regime for banks and investment firms. Indeed, some of the interviewees were adamant that financial services regulation in general had failed, and had introduced system-wide weaknesses that promoted excessive risk-taking: Of course the regulators were appalling either at understanding what they were regulating or actually regulating, probably the latter because of the former. [Director of regulation 1]

2679

Two main points arise from the data; criticisms regarding the design of the current Basel-2-based prudential regime, and criticisms regarding the implementation of the regime by regulators such as the Financial Services Authority. The Basel 2 Accord forms the basis of the prudential regulation of banks across Europe and many other international states. The Basel Committee on Banking Supervision, whose membership at the time primarily comprised regulators from the G10 nations, completed the Accord in 2004. The Accord focuses on setting risk-based capital requirements via a three-pillar framework: minimum capital requirements; a supplementary supervisory review of the capital adequacy of each bank; and the market disclosure of information relating to these capital requirement/ adequacy assessments. In terms of the Basel 2 regime, the interviewees noted a bias towards quantitative risk assessment. This bias encouraged many financial institutions to become overly reliant on mathematical risk models at the expense of management judgment. The implications were: The point of risk management is to think about where the emerging risks are, and what the regulators do is focus on what can be measured, and they're not the same thing. [Chief risk officer 4] Interviewees also highlighted the capital myopia of the Basel 2 rules, which predominantly relate to the calculation of regulatory capital requirements, rather than the essential elements of sound frameworks for risk management and corporate governance. In addition, a real possibility exists that the regulatory emphasis on capital adequacy encouraged financial institutions to ignore risk management considerations. Institutions were able to increase their exposure to risk providing they had sufficient capital – since capital effectively operates as a substitute for good risk management in the Basel 2 Accord. This, of course, runs contrary to the conventional argument that risk-based capital requirements should help to incentivize good risk management. In terms of the competency of the regulatory agencies charged with implementing the Basel-2-based capital regime many of the interviewees criticized the skills and business-related experience of regulators, suggesting that their staff were not good enough to keep ahead of the institutions that they were regulating: Our supervisor at the FSA has probably just had his 14th birthday! I know that is a bit of an unkind statement, but they come in with their checklists and tick everything off, and we got a very good report, but they don't really understand what is going on underneath the surface because they are not streetwise enough and don't understand some of the people aspects — some of the characters on the board, some of the egos that fly around etc., which fundamentally I think is part of what's going on here. [Finance director 1] Respondents also noted that the UK Financial Services Authority implemented the Basel 2 Accord in a mechanistic way. The situation resulted from the difficulty in coordinating large numbers of regulatory supervisors and ensuring consistency in their work but meant that confirmation of compliance became little more than a box-ticking exercise that did not incentivize good risk management. These poorly developed processes of signification and legitimation caused an over-reliance on structures of domination to manage people and control their utilization of and access to resources. Hence, the situational imperatives of regulation and supervision (the immovable object of structure) dominated the irresistible force of human agency, preventing financial institutions from managing their risks using the frameworks that they saw fit. 5. Discussion The study here explores the underlying causes of the global financial crisis from the perspective of structuration theory. Drawing on the

2680

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

opinions of a diverse range of risk management professionals, the study provides a firsthand account of what went wrong, coupled with expert input into the action that the financial services sector and the sector regulators might take. This section now proceeds with a short discussion of the implications arising from the study.

the common good over personal gain, creating interpretive schemes that not only ensure that managers consider their institution's risk appetite when making business decisions, but also assess and report on risk issues in a timely and efficient way. Enforcing such arrangements requires the use of clear rules that highlight the kinds of risk-taking or reducing behaviors to reward or discourage.

5.1. Implications for crisis management theory With the exception of Weick (1993), research helping to explain the incubation, manifestation, and potential intensification of organizational crises does not use structuration theory. As Weick et al. (2005) observe, agency can be over emphasized in fields such as organizational crisis research (see also Maitlis & Sonenshein, 2010), while in the context of financial crises the opposite is the case (Lawson, 2009). As shown in the current study structuration theory can provide invaluable insights into the crisis process. The study demonstrates that structuration theory is a useful mechanism for explaining why decision-making in some financial institutions succumbed to the immovable object of structure (increased competition and complexity, coupled with mechanistic regulation), while decision-making in others was able to exercise the irresistible force of agency and withstand the macro-level temptations and pitfalls that were on offer. The study reinforces the importance of culture as a key factor influencing the susceptibility of organizations to crises, including financial crises. Those institutions that took a long-term view of their performance, promoted risk management, and encouraged open communication were able to withstand the crisis much better. Indeed, the crisis is unlikely to have ever occurred if all institutions had demonstrated such traits. All researchers should give this factor much more attention in the literature on financial crises, since culture is apparently as relevant here as in any other form of organizational crisis. The study also highlights the significance of weak management (including regulatory and political management) in causing the financial crisis, via the actions of incompetent, poorly informed and or self-interested decision makers who, whether knowingly or not, exposed (or allowed to be exposed) certain institutions and their stakeholders to excessive amounts of risk. This observation supports the perhaps uncomfortable notion that the crisis was largely self-inflicted rather than simply bad luck. However, the study also offers hope for the future, since the correction of these weaknesses should help to reduce (though probably not eliminate) the probability of future financial crises. This conclusion is consistent with Weick's notion that “crises are more controllable than we think” (Maitlis & Sonenshein, 2010, p. 554). The study suggests that researchers make greater use of structuration theory to investigate financial (and for that matter non-financial) crises. The theory provides a mechanism for integrating the predominantly macro-level focus of financial crisis research with the predominantly micro-level focus of the literature on organizational crises. Thus, using structuration theory brings the two literatures closer together. Structuration theory is likely to be especially important in the new age of the trans-boundary crisis (Boin, 2009) where crises no longer respect the disciplinary divisions of the current literature. 5.2. Implications for practice All of the interviewees accepted that learning opportunities from the financial crisis were available, for both financial institutions and their regulators. The study now presents the key implications raised by the research findings, structured according to the six Cs of the crisis. 5.2.1. Compensation Based on the current interview data, pay and bonus awards can be high, providing that they are earned on the basis of considered risk-taking that balances the interests of all major stakeholder groups (for example shareholders, consumers, employees and society at large). Financial institutions should design reward schemes that signify

5.2.2. Culture Developing an appropriate culture played a key part in reducing the susceptibility of certain financial institutions to the crisis. Boards and senior managers must pay closer attention to the development of appropriate organizational cultures, encouraging long-term thinking and greater levels of risk awareness coupled with high levels of openness and teamwork in order to promote the effective communication of risks and concerns. They must also become more reflexive decision makers who, when finding that their institutional exposure to risk is excessive, act quickly and decisively. The study also reveals that culture change is required in some risk management functions, risk managers should move away from their traditional compliance orientation, instead adopting a more enterprise-wide view that is less confrontational and more proactive. Achieving this, using relatively simple techniques (for example, internal placements where risk management staff work within front-line departments for a time or vice versa; and cross-department team-building projects) is possible and provides a good way to start breaking down barriers. 5.2.3. Complexity Even with the forced separation of commercial and investment banks within certain regimes, the level of complexity in the global financial services sector is unlikely to diminish imminently. However, this observation does not mean that institutions need to suffer from major crises. Effective management is the key to combating the problem of complexity, in particular via greater use of stress-testing and scenario analysis: I think scenario planning and modeling is a big lesson that's come out of this and there is a lot of work going on now in financial institutions who look at scenarios far more than they used to. [Non-executive 2] Stress testing and scenario analysis must become more judgmental (for example by incorporating softer risks, such as reputation risk) and extreme, reflecting genuine worst-case scenarios. In addition, discussion of the results from the study (right up to board level) could lead to improved corporate responses to complexity. It's qualitative, not quantitative, it's contingency planning on a larger scale almost. If you identify a scenario that potentially brings down your firm then you must, I would assume, conclude that it is an unacceptable risk for your business and you need to think about what you are going to do about it… There's been a significant improvement in the way they [financial institutions] are looking at their scenario analysis and stress testing. For a lot of these firms it's probably the most useful exercise because this will get the board's attention by talking about the headline risks that bother them and show them how they impact and how you mitigate against that to make sure they stay within risk appetite. [Risk consultant 2]

5.2.4. Communication Effective communication is vital to maintain the knowledge and reflexivity of managers. Managers cannot transform their setting through action and exercise their irresistible force as decision-making agents without timely, relevant, and accurate information.

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

In terms of solutions, recruiting more and better risk management staff may not always help. As pointed out by one interviewee (chief risk officer 2), many large banks have large and high quality risk teams. However, these may operate in isolation from key front-line decision makers and often their boards as well, meaning that they do not have the necessary influence (i.e., the power to impose effective sanctions that influence more desirable legitimation processes) that they need. Additionally, risk functions need to avoid excessive focus on the processes of collecting and reporting data. Greater consideration about how to use risk reports to support real-world decision-making is necessary. However, this action will require a fundamental change in the social practices of some financial institutions. In particular, social practices relating to risk management and to the perception of this activity by both risk management staff and other agents within the institution must change. Equally, more accurate and comprehensive reports may only be of limited benefit, not least because social practices, such as filtering, can attenuate their benefits. Rather, changes in the rules and norms that surround the supply and use of risk management information are necessary. These changes include receiving risk management information on a no-blame basis, and valuing this information not for its perceived complexity or ability to satisfy regulators, but for how such information can support strategic decision-making. 5.2.5. Competition Promoting effective competition is a key thread in many government reports on regulatory reform (Independent Commission on Banking, 2010; US Treasury, 2009). Properly focused competition can provide an effective constraint against overly powerful, self-seeking agents. Such competition provides a system of domination that ensures that decisions are in the interests of all major stakeholders. However, the effectiveness of the proposed governmental competition reforms is less clear; notably, the uncertain benefits of breaking up larger banks or segregating commercial and investment banking. Such competition reforms are inconclusive from a structuration perspective, as the irresistible force of agency (primarily managerial decision-making) can at times overcome these proposed structural constraints. For most of the interviewees the causes of the financial crisis were unrelated to the organizational structure of banks. While conglomerate banks may appear to be inherently risky, the interviewees emphasized that control of this risk is possible given effective management (i.e., by those who are competent and professional, and who are prepared to communicate with each other and work together) operating within an appropriate risk culture. Challenging the established structural features of the social system through competition might be possible through altering the structuration processes of domination and access to resources such as information. For example, one key issue raised by a number of the interviewees was market disclosure, an area they saw as too opaque and accounting biased: “Bland statements about numbers that don't mean a great deal” (Non-executive 1). In particular, to enhance the power of stakeholders and their ability to enact processes of domination, paying more attention to enhancing disclosure requirements is necessary. This includes reviewing the scope of information disclosed, along with its method of presentation; ideally the disclosure of risk management information that is less quantitative and accounting biased and more relevant to a wider range of stakeholders (notably, small shareholders and consumers). Disclosure documents should also contain more qualitative information about the future prospects of a bank and the design of its risk management, governance, and remuneration frameworks. 5.2.6. Capital regime Government initiatives on both a local and international scale (for example, Basel Committee on Banking Supervision, 2009) are

2681

already in place to enhance the capital requirement regulations for financial institutions, especially banks. As with competition reforms, designing such initiatives increases the influence of certain domination processes over financial institutions and so constrains their ability to act in self-seeking and systemically destructive ways. However based on the analysis in the study their ability to achieve this outcome is far from certain. Many of the interviewees expressed doubts over the current direction of regulatory capital initiatives, though for the most part they accepted the need for greater regulation (for example, greater prescription over governance and risk management arrangements). Clearly, capital requirements can be ineffective in constraining the activities of financial institutions whose management is weak, or who have crisis-prone cultures, and can in fact even accentuate problems. As finance director 2 commented: I think more capital decreases the likelihood of a bank going bust but doesn't necessarily improve risk management and you could argue that more capital could reduce the focus on risk management. Because there is more money to play with you can be more cavalier. Regulators should also attempt to reduce the mechanistic nature of their capital regimes, promoting management judgment alongside the use of quantitative models. In addition, they should enhance the quality of regulatory supervision to ensure that supervisors have the necessary skills and experience to monitor the activities of financial institutions and discipline them as necessary. As head of operational risk 3 said, “regulators should be feared.” 6. Conclusion …it will happen again. [Finance director 2] History shows that financial crises are inevitable. However, that does not mean that good risk management will not reduce their frequency or severity of impact. Learning from the current crisis and applying lessons and insights from the existing literatures on both financial and organizational crises is crucial. The study demonstrates that achieving this through the lens of structuration theory provides useful insights. What the interviewees revealed is that, while the causes of the financial crisis are many and varied, relatively few key factors explain why certain institutions chose to expose themselves to excessive amounts of risk. Specifically: • Human and cultural weaknesses at the institutional, industry, and/or society levels. • Communication weaknesses within some financial institutions, where boards and senior managers either did not receive the information that they needed, or failed to understand this information. • Weaknesses in the prudential regime for banks and investment firms, coupled with flawed supervision. Based on these findings, like many previous crises, the current financial crisis is largely self-inflicted; the causative agents were managerial and cultural weaknesses within both financial institutions and their regulators. The irresistible force of human agency (in both managerial action and decision-making) is not only a positive or negative force in its own right, but interacts with the immovable object of structural features and situational factors, in terms of constraints like regulation and competition. The study provides evidence that the structural features of financial institutions (for example, compensation arrangements) hindered individual reflexivity and critical evaluation and reinforced the natural tendency of individuals to pursue their own self-interest by making

2682

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683

extreme risks seem both acceptable and desirable. At the industry level, the observation that building strong collective beliefs of invulnerability was widespread supports this finding. Signification of these beliefs to the industry through optimistic economic forecasts, and their legitimation through the development of soft regulatory interventions, also supports the findings. In terms of the role of risk management, risk managers established social practices that legitimized their role as information guardians. They saw their role as assessors rather than managers. This view led to a greater focus on the justifications for their assessments rather than the managerial implications of the risk level adopted. Equally, increased competition encouraged short cuts in risk management activities, where structures of domination tended to encourage and reward greater access to allocative resources by reducing the power of authoritative resources (managerial control) to manage risk. Thus, poorly developed processes of signification and legitimation caused an over-reliance on structures of domination to manage people and control their utilization of and access to resources. A variety of lessons follow from these conclusions. In particular, the study challenges proposals for greater regulatory prescription and capital requirements, or suggestions that simple solutions such as caps on bonus payments will prove effective. Rather, a more permanent solution to the prevention of future financial crises should combine enhancements in the risk management and governance practices implemented by financial institutions and their regulators, together with mechanisms that support cultural change. This cultural change should be both organization wide and function specific, particularly in relation to moving the attitudes of risk functions away from compliance and towards a more business-like orientation where risk management staff and risk reports are used to support strategic decision-making. Adopting these lessons may prove hard for some, not least because they require a change of mindset for many within the financial sector. However, without change, similar crises will happen again.

Appendix 1

Interview agenda In your opinion what was the major underlying cause of the current banking crisis? Are traditional corporate risk management practices effective at assessing/controlling systemic risk? To what extent did boards/management have accurate and reliable information to allow them to monitor their firm's risk exposures and did they understand this information? Notably how reliant were they on mathematical models and did they understand the inputs/outputs to these models? What was the situation in your firm? To what extent did weaknesses in corporate governance contribute to the current crisis? For example, were warnings from chief risk officers/risk managers heard prior to the crisis and if not why not? In addition, were managers at all levels properly incentivised to manage risk effectively? Can you explain the situation in your own firm? Going forward, how should financial services firms use incentivisation to implement effective risk management frameworks? Is regulation that is more prescriptive the solution? Are alternatives available (e.g. improved disclosure, enhanced rating methodologies, changes to corporate governance rules)? In the light of the current banking crisis, what changes have you/are you making to your company's risk management framework(s)? Do you have any other comments?

Appendix 2. Respondents Risk consultant 1

Chief risk officer 1

Non-executive 1

Chief risk officer 2

Risk consultant 2

Finance director 1

Head of operational risk 1 Head of operational risk 2* Risk consultant 3

Head of control 1

Head of operational risk 3 Director of regulation 1 Non-executive 2

Chief risk officer 3

Chief risk officer 4* Finance director 2

Chief risk officer 5

Risk consultant 4

Chief risk officer 6

Project director 1

Large UK based, internationally active bank. A bank not badly affected by the crisis. The interviewee was head of a consulting subsidiary with over 30 years of experience in banking, insurance, and risk management. UK based investment bank. A bank not badly affected by the crisis. The interviewee possessed over 20 years of experience in business continuity and risk management. (a) UK bank; (b) wholesale financial market intermediary; (c) European subsidiary of an international bank. The interviewee possessed over 40 years of experience working for internationally active banks, many at director level. UK based mutual life insurer and investment provider. The crisis affected the institution, but they did not require government assistance. The interviewee possessed experience as a consultant for regulators as well as a risk manager. Professional services provider for asset management institutions. The individual possessed over 30 years senior-level experience in a range of financial institutions, including banks and insurers. UK based health insurance provider. The interviewee possessed over 30 years of experience having also worked for a large UK commercial bank. UK based bank and insurance conglomerate — not badly affected by the crisis. The interviewee possessed over 20 years of experience in risk management and IT roles, primarily in a large UK commercial bank. UK based life subsidiary of an international insurer. Prior to leading an operational risk corporate function, the individual worked as an auditor, consultant, and regulator. Freelance risk consultant. The individual worked for many UK and international financial institutions over approximately 30 years. This experience includes one of the UK banks most affected by the crisis. European, internationally active bank. The crisis badly affected the institution. The individual is an experienced accountant and risk manager with over 20 years of experience UK based life and property casualty insurer. The individual possessed around 15 years of experience as a consultant, regulator and group head of risk. Independent central counterparty (clearing house). The individual possessed over 30 years of experience as a risk manager and policy expert. UK based building society. The institution was not significantly affected affect by the crisis. The individual possessed over 40 years of experience, having worked as a director and non-executive for a range of both financial and non-financial institutions. UK based demutualised bank, badly affected by the crisis. The individual possessed over 30 years of experience in banking and risk management UK based life subsidiary of an international insurer. The individual possessed over 20 years of experience as an insurance professional and risk manager. UK based mutual insurer. The individual possessed over 30 years experience with the organization having worked up to director level. UK based building society — not badly affected by the crisis. The individual possessed over 20 years of experience in banking and especially bank lending, having worked up from branch level. Financial services consultant; previously at an international investment bank. The crisis did not badly affect the institution at the time of the crisis, though they are suffering some losses now. International life and property casualty insurer. The individual possessed over 20 years of experience in insurance and risk management. Large UK bank. The crisis badly affected the institution. The individual possessed over 15 years of experience having worked as a consultant and a senior risk manager.

*Note that these two interviewees worked for the same institution.

References Acharya, V., & Richardson, M. (2009). Restoring financial stability: How to repair a failed financial system. New York: Wiley Finance. Akerlof, A., & Shiller, R. (2009). Animal spirits: How human psychology drives the economy and why it matters for global capitalism. Princeton: Princeton University Press.

S. Ashby et al. / Journal of Business Research 67 (2014) 2671–2683 Archer, M. S. (1995). Realist social theory: A morphogenetic approach. Cambridge: Cambridge University Press. Archer, M. S. (2000). For structure: its reality, properties and powers: A reply to Anthony King. The Sociological Review, 48(3), 464–472. Basel Committee on Banking Supervision (2009). Enhancements to the Basel II framework. Basel, Switzerland: Bank for International Settlements. Beech, N. (2000). Narrative styles of management and workers: A tale of star crossed lovers. The Journal of Applied Behavioral Science, 36(2), 210–228. Berends, H., Boersma, K., & Weggeman, M. (2003). The structuration of organizational learning. Human Relations, 56(9), 1035–1056. Boin, A. (2004). On financial crises. British Journal of Management, 15(2), 191–195. Boin, A. (2006). Organizations and crisis: The emergence of a research paradigm. In D. Smith, & D. Elliott (Eds.), Key readings in crisis management: Systems and structures for prevention and recovery.. London: Routledge. Boin, A. (2009). The new world of crises and crisis management: Implications for policy making and research. Review of Policy Research, 26(4), 367–377. Boland, R. (1996). Why shared meanings have no place in structuration theory: A reply to Scapens and Macintosh Accounting. Organizations and Society, 21(7/8), 691–697. Borio, C. (2008). The financial turmoil of 2007–?: A preliminary assessment and some policy considerations. BIS Working Papers (No 251).. Basel, Switzerland: Bank for International Settlements. Brunnermeier, M. (2009). Deciphering the 2007–08 liquidity and credit crunch. Journal of Economic Perspectives, 23(1), 77–100. Callahan, J. L. (2004). Reversing a conspicuous absence: Mindful inclusion of emotion in structuration theory. Human Relations, 57, 1427–1448. Campbell, D. T. (1975). Degrees of freedom in the case study. Comparative Political Studies, 8(July), 178–193. Carson, D., Gilmore, A., Perry, C., & Gronhaug, K. (2001). Qualitative marketing research. London: Sage. Congressional Oversight Panel (2009). Special report on regulatory reform (January). Washington DC, USA: Congressional Oversight Panel, US Senate (http://cop. senate.gov/documents/cop-012909-report-regulatoryreform.pdf) Coombs, W. (2001). Teaching the crisis management/communication course. Public Relations Review, 27(1), 89–101. de Larosière, J., Balcerowicz, L., Issing, O., Masera, R., McCarthy, C., Nyberg, L., et al. (2009). The high-level group on financial supervision in the EU: Report (25th February), Brussels. Denzin, N. K., & Lincoln, Y. S. (1994). Handbook of qualitative research. Thousand Oaks, CA: Sage. Dowd, K. (2008). Moral hazard and the financial crisis. CRIS Discussion Paper, Series VI, University of Nottingham: Centre for Risk and Insurance Studies. Eichengreen, B. (2002). Financial crises (and what to do about them). Oxford: Oxford University Press. Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review, 14(4), 532–550. Fuchs, C. (2003). Structuration theory and self-organization. Systemic Practice and Action Research, 16(2), 133–167. Fullbrook, E. (2009). Lawson's reorientation. In E. Fullbrook (Ed.), Ontology and economics: Tony Lawson and his critics.. London and New York: Routledge. Giddens, A. (1984). The Constitution of society: Outline of the theory of structuration. Chicago: Polity Press. Goodhart, C., & Illing, G. (2002). Financial crises, contagion and the lender of last resort. Oxford: Oxford University Press. Gregory, A. (2005). Communication dimensions of the UK foot and mouth disease crisis. Journal of Public Affairs, 5, 312–328. Hardcastle, M. R., Usher, K. J., & Holmes, C. A. (2005). An overview of structuration theory and its usefulness for nursing research. Nursing Philosophy, 6, 223–234. House of Commons Treasury Committee (2009). Banking crisis: Reforming corporate governance and pay in the City. London: House of Commons. Independent Commission on Banking (2010). Issues paper call for evidence (September). London: Independent Commission on Banking.

2683

Jaques, T. (2009). Issue and crisis management: Quicksand in the definitional landscape. Public Relations Review, 35(3), 280–286. Kasuya, M. (2003). Coping with crisis: International financial institutions in the interwar period. Oxford: Oxford University Press. King, N. (1994). The qualitative research interview. In C. Cassell, & G. Symon (Eds.), Qualitative methods in organizational research (pp. 16–36). London: Sage. Laeven, L., & Valencia, F. (2008). Systemic banking crises: A new database. IMF Working Paper, (WP/08/224) Washington DC. Lawson, T. (1997). Economics and reality. London and New York: Routledge. Lawson, T. (2009). The current economic crisis: Its nature and the course of academic economics. Cambridge Journal of Economics, 33, 759–777. Lewis, M. (2010). The big short: Inside the Doomsday Machine. London: Allen Lane. Maitlis, S., & Sonenshein, S. (2010). Sensemaking in crisis and change: Inspiration and insights from Weick (1988). Journal of Management Studies, 47(3), 551–580. Michel-Kerjan, E., & Slovic, P. (2010). The irrational economist: Making decisions in a dangerous world. New York: Public Affairs. Mouselis, N. (1999). Back to sociological theory. Basingstoke: Macmillan. Müßig, A. (2009). The financial crisis: Caused by unpreventable or organized failures? International Journal of Economic Sciences and Applied Research, 2(1), 51–70. Nonaka, I., & Toyama, R. (2003). The knowledge-creating theory revisited: Knowledge creation as a synthesizing process. Knowledge Management Research and Practice, 1, 2–10. Parker, J. (2000). Structuration. Buckingham: Open University Press. Pauchant, T., & Mitroff, I. (1988). Crisis prone versus crisis avoiding organizations: Is your company's culture it own worst enemy in creating crises? Industrial Crisis Quarterly, 2, 53–63. Pozzebon, M. (2004). The influence of a structurationist view on strategic management research. Journal of Management Studies, 41(March), 247–272. Rosenthal, U. (2003). September 11: Public administration and the study of crises and crisis management. Administrative Science Quarterly, 35(2), 129–143. Roux-Dufort, C. (2007). Is crisis management (only) a management of exceptions? Journal of Contingencies and Crisis Management, 15(2), 105–114. Rust, R., & Cooil, B. (1994). Reliability measures for qualitative data: Theory and implications. Journal of Marketing Research, 31(Feb), 1–14. Shrivastava, P., Mitroff, I., Millar, D., & Miglani, A. (1988). Understanding industrial crises. Journal of Management Studies, 25(2), 283–303. Smith, D. (2005). What's in a name? The nature of crisis and disaster — A search for signature qualities. Working paper.. Liverpool: University of Liverpool Science Enterprise Centre. Smith, D. (2006a). Crisis management — Practice in search of a paradigm. In D. Smith, & D. Elliott (Eds.), Key readings in crisis management: Systems and structures for prevention and recovery.. London: Routledge. Smith, D. (2006b). Modelling the crisis management process: Approaches and limitations. In D. Smith, & D. Elliott (Eds.), Key readings in crisis management: Systems and structures for prevention and recovery.. London: Routledge. Smith, D., & Elliott, D. (2006). Key readings in crisis management: Systems and structures for prevention and recovery. London: Routledge. Staber, U., & Sydow, J. (2002). Organizational adaptive capacity: A structuration perspective. Journal of Management Inquiry, 11(4), 408–424. Tett, G. (2009). Fools gold: How unrestrained greed corrupted a dream, shattered global markets and unleashed a catastrophe. London: Little Brown. Toft, B., & Reynolds, S. (2005). Learning from disasters: A management approach (3rd ed.). UK: Perpetuity Press Ltd. Turner, A. (2009). The Turner Review: A regulatory response to the global banking crisis. London: Financial Services Authority. U.S. Treasury (2009). Financial regulatory reform, a new foundation: Rebuilding financial supervision and regulation. USA: Department of the Treasury. Walker, D. (2009). A review of corporate governance in UK banks and other financial entities: Final recommendations. London: HM Treasury. Weick, K. (1993). The collapse of sensemaking in organizations: The Mann Gulch Disaster. Administrative Science Quarterly, 38(4), 628–652. Weick, K., Sutcliffe, K., & Obstfeld, D. (2005). Organizing and the process of sensemaking. Organization Science, 16(4), 409–421.