How Culture Displaced Structural Reform: Problem Definition, Marketization, and Neoliberal Myths in Bank Regulation

Abstract

We use content analysis to show that the diagnosis of the financial crisis of 2007–2009 shifted significantly from a focus on the need for structural change in the banking industry to an emphasis on culture and reform at the organizational level. We consider four overlapping subsystems in which this shift in problem–solution clusters played out—political, regulatory, legal, and consulting—and show that the “structural reform agenda,”  which was initially strong and publicly prominent in the political arena, lost attention. Over time it was displaced by a neoliberal managerialist turn, which watered down or abandoned structural solutions and instead played up a new “culture and conduct reform agenda.” We explain this shift in terms of the marketization of regulation, which—following Mautner (Language and the market society, 1st ed. Routledge, 2010)’s model of interdiscursive alignment—we detect in the shifting language of financial-services reform across the four subsystems in scope. We argue that a neoliberal turn took place with a discursive closure that made the structural reform alternative gradually unsayable and, in the end, unthinkable. At the same time, the discourse turned to embrace the neoliberal agenda, built on the myth of self-regulating actors and markets, manifest in the culture problematic. This managerialist turn was able to mobilise, and be operationalised by, an industry of consultants, whereas structural change came to be seen by regulators as too risky to implement. We claim that these dynamics reveal how a form of “collective strategic ignorance,” based on powerful institutional myths, was systematically oriented to ignore and reject structural sources of crisis. Finally, we suggest that the observed pattern of displacement—whereby initial calls for structural change become later displaced by managerial and procedural solutions—is common to other social issues, such as audit reform and corporate social responsibility.

Introduction

The onset of the Great Financial Crisis of 2007–2009 was a shock to political and economic systems around the world. There was public outrage and manifest revulsion towards bankers, leading to public hearings and, eventually, to legal action against, and large fines imposed on, the major financial institutions at the centre of the collapse: Goldman Sachs, JP Morgan, Bank of America, and others. The actions brought by regulatory authorities were founded on a fundamental concept of conflict of interest and it was argued that financial institutions had prioritised their own interests and those of shareholders, over their fiduciary duties to one or more of their clients, and had failed to disclose such conflicts. In this period, to the middle of the 2010s, there were related calls for structural remedies to address the conflict-of-interest problem and the unethical behaviour of the large investment banks. In short, it was asserted that the big banks needed to be broken up and a full cultural adjustment (Turner, 1976) was imagined along the lines of deep, structural reform in banking.

If we skip forward to the end of the 2010s, a very different picture is evident. Rather than a widespread restructuring (or breaking up the big banks), the dominant solution to banks’ defective ethics and conflict-of-interest problems was framed in terms of reform to their operating cultures and values. On both sides of the Atlantic, structural separation measures were “rather limited” and watered down (Vickers, 2019). Cultural renewal, managed and demonstrated by the banks themselves, was promoted and normalized as a response to the behaviours which were perceived to have caused the crisis. This “cultural turn”, serviced by an army of advisers, included the creation of formal policy documents about the management of conflicts of interest, but was also much broader, focusing extensively on the renewal of risk management processes with increased attention to “behavioural risk” and corporate ethics.

How should we  understand this shift from a structurally grounded critique of investment banking to a critique focused on their “toxic” but reformable cultures? In this paper, we map the transition via an analysis of political hearings, and legal, regulatory, and consulting texts produced in the wake of the financial crisis. Specifically, we use content analysis to show that the diagnosis of the financial crisis of 2007–2009 shifted significantly from a focus on structural problems (manifest in the notion of “conflict of interest”) and the need for structural change in the industry to an emphasis on culture and ethical reform at the organizational—that is, managerial—level.

Drawing on Hilgartner and Bosk’s (1988) public arenas model and combining it with the literature on marketization and the myths of neoliberal capitalism (Mautner, 2010; Panitch & Konings, 2009), we seek to explain this shift in public discourse from a highly charged ethical understanding of conflict of interest to a more managerial understanding. We argue that this transition represents the ethical displacement and dilution of public concern about banking’s inherent and prevailing structural fault lines. Furthermore, we postulate that this ethical displacement was driven by prevailing and persistent thought patterns—the neoliberal myths of self-regulating organizations and markets.

These thought patterns are evident as linguistic traces to be found in the various arenas through which the conflict-of-interest problematic passes as it advances from public outrage, through the legal and regulatory domains, into the world of company management. We argue that the original insights into the structural roots of the crisis gradually degenerated into neoliberal institutional remedies, such as new disclosure requirements and prescriptions for behavioural risk management. Crucially, there was a deep-seated pragmatic settlement that (re)constructed financial organizations as self-regulating actors capable of internal interventions to prevent misconduct via cultural management.

In contrast to studies of conflict of interest at the individual level (Feldman & Halali, 2017; Fogel & Friedman, 2008; Gunz et al., 2009; Ishaque, 2019), we focus on the trajectory of conflict of interest in public discourses across multiple arenas and how these discourses reveal problem–solution clusters rather than ethical dilemmas in their pure form. We also contribute to ongoing conversations about ethics within the neoliberal paradigm by showing the force of marketization in the regulation and apparent reform of financial services after the Great Financial Crisis. While the literature locates the discourse of the marketization of financial-services reform in the media (Herzig & Moon, 2013; Czarniawska, 2012), we show, following Mautner’s (2010) model of interdiscursive alignment, that the neoliberal turn in banking reform also occurred in four additional subsystems: political, legislative, regulatory and, consulting.

We begin our analysis by briefly reviewing the literature on marketization and neoliberal institutional myths, including, but not restricted to, the case of banking (“Interpreting the financial crisis: neoliberal institutional myths”). We then turn to the conflict-of-interest problematic as an indicative case. While the literature examines conflicts of interest at the individual level, we focus on the trajectory of the category of “conflicts of interest” as a feature of the system-level diagnosis offered up by the political subsystem to financial regulators in the wake of the Great Financial Crisis. Drawing on Hilgartner and Bosk’s (1988) public arenas model, we ask how and why the original diagnosis (focussed on conflicts of interest) was displaced by diagnoses and policy solutions based on culture. We introduce Mautner’s (2010) model of interdiscursive alignment, which supports our analysis of the linguistic traces of marketization across subsystems (“Conflicts of interest, bank reform and interdiscursive alignment”). “Methodology” outlines our content analysis, which supports the displacement/dilution thesis. “Findings” reports our findings and provides a fuller explication of the conflict-of-interest problem–solution patterns across the four domains in scope. We distil three patterns—a conflict of interest emphasis; conflict of interest dilution; and a culture and conduct emphasis—which support the claim that the very category of conflict of interest was progressively displaced by the turn towards culture and conduct. Finally, we attempt to understand and explain this shift by drawing on the public arenas model, augmented by notions of neoliberal institutional myths and collective strategic ignorance.

Ultimately, we suggest that the turn towards culture as a remedy was shaped to avoid the high ethical stakes of the conflict-of-interest problematic and to “tame” and contain it within managerialist policies amenable to routine compliance monitoring. Detecting these neoliberal cultural myths by focusing on the linguistic repertoire across four subsystems allows us to conclude that no single social actor can be held responsible for the neoliberal turn in banking reform. Yet the model of interdiscursive alignment suggests that the power differential between the consulting and regulatory subsystems may explain why the political counter-discourse of structural reform was sidelined in the face of the managerialist pressure from the client-based consulting subsystem, leading the regulatory arena to adopt a culture-reform agenda. In conclusion, we propose that the culture-oriented diagnosis was able to dominate because it could draw upon and preserve important neoliberal institutional myths about the role of information in markets and the competence of organizations as self-regulating actors.

Literature Review

Interpreting the Financial Crisis: Neoliberal Institutional Myths

The Great Financial Crisis of 2007–2009 left many commentators and citizens baffled. According to Panitch and Konings’s (2009) early estimates, total losses ranged from $2.2 trillion (IMF) to $3.6 trillion (Nouriel Roubini). Mighty banks like Bear Sterns and Lehman Brothers disappeared.  Even the much-admired Goldman Sachs was sued for its role in the crisis: benefiting itself and certain privileged clients, while causing colossal losses to other clients.

Public reaction was resoundingly negative, exemplified by the media’s  “banker bashing” (Herzig & Moon, 2013). While many held the banking sector responsible for the financial crisis (Committee on Homeland Security & Governmental Affairs, 2010a; b), a parallel debate started about regulatory failure and the proper balance of  external regulation and self-regulation of financial services (Herzig & Moon, 2013).

We argue that this debate and the subsequent regulatory reform were actually just one chapter in the ongoing drama of marketization that characterizes our age of neoliberal capitalism. Following Mautner (2010, p. 16), we define marketization as a “shorthand for the process by which the laws of the marketplace are transferred to lifeworlds that were not originally organized along such lines.” It is well-documented how the practices, identity and public image of financial services—and regulatory agencies—changed after the “Big Bang” deregulation of the financial services industry in the UK in 1986 and the repeal of the Glass-Steagall Act in the US in 1999 (Panitch & Konings, 2009; Salz & Collins, 2013). Regulatory agencies such as the US Federal Reserve eagerly promoted the neoliberal myth of efficient markets, supposedly benefitting society most when regulators leave plenty of space for innovation and self-regulation. The neoliberal influence on financial services regulation was exemplified by former Federal Reserve chairman Alan Greenspan, who personally advocated against the regulation of complex derivatives transactions by declaring that “[r]egulation … hinders the efficiency of markets to enlarge the standard of living.”¹

Such myths are important to social change—or the lack of it. We define myths, in line with Lévi-Strauss (1963) and Lotesta (2019), as “narratives that order and make sense of the world around us” (Lotesta, 2019, p. 222).² The myths of neoliberal capitalism are manyfold. Apart from the myth of efficient markets, they also include the myth of the business friendly economy (Lotesta, 2019), the myths of “free” individuals and de-regulated, privatized businesses exercising agency (Wrenn, 2016).

Behind the widespread deregulation and marketization of financial regulation stood the neoliberal myths  of “efficient markets” and of the self-regulating capacity of autonomous market actors. Sociologist Barbara Czarniawska wittily referred to a line in Oliver Stone’s movie Wall Street 2: Money Never Sleeps: “Perhaps we in finance made a mistake or two, but we will be able to straighten it out without your help. Like any other family, we will resolve our problems, so stay away!” (Czarniawska, 2012, p. 771).

Following the Crisis, some expected the relentless march of marketization (Mautner, 2010) and the adjacent myths of neoliberal deregulation (Panitch & Konings, 2009) to be discredited and ousted as a dominant model. Indeed, widespread financial reform was called for—and started—on both sides of the Atlantic.

Discourse studies in the aftermath of the Crisis captured some of the counter-offensive that offered the tantalising prospect of an alternative model. A radical reformist discourse was reported in the media that talked of restructuring the banks (Herzig & Moon, 2013). Yet Herzig and Moon (2013) themselves weren’t so sure that such calls for radical reform would be heeded and both Czarniawska (2012) and Mautner (2010) doubted that the tide of marketization could be turned, given the extent to which regulators were in thrall to “market fundamentalism” (Czarniawska, 2012, p. 768).

We examine the discourses that emerged after the Crisis—in the political, regulatory, legal and consulting arenas—to help explain how that radical-structural counter-offensive was sidelined, allowing financial regulatory reform on both sides of the Atlantic to emphasise managerialist reform. This reform, although it nominally kept a focus on the professionalisation and improvement of market actors, represents a very much watereddown version of the original structural proposals. As Vickers (2019) summarised it, the banking industry was made “safer, but not safe enough.” We propose that central to this shift are the neoliberal myths of self-governing market actors, and that various market actors carried such myths across the domains relevant to financial regulation. We shine light on the crucial role that the client-oriented consulting industry played in promoting managerial solutions and the notion of autonomous financial actors capable of professionalization and self-improvement in response to external critique. At the same time, a cottage industry of consulting services emerged to offer banks those  very solutions, ranging from “risk culture” agendas to “compliance” and “conduct risk” management. Indeed, a larger trend of “riskification” further fuelled and legitimized this managerialist reform agenda focused on banks’ self-regulating capacities (Hardy & Maguire, 2016; Power, 2004, 2007).

Previous studies of the discourses around financial services reform tend to focus on the media discourse in the direct aftermath of the Great Financial Crisis (e.g., Czarniawska, 2012; Herzig & Moon, 2013; Panitch & Konings, 2009). We complement and extend this work by undertaking content analysis over an eight-year horizon across what we conceptualize as four inter-related subsystems implicated in financial reform: the political, legal, consulting and regulatory arenas. From demonstrating that there was a shift from an initial radical counter-discourse of structural reform to a managerialist reform agenda that watered down structural reform and played up the self-governing capacities of banks, we also seek to theorize and explain why this shift happened.

Conflicts of Interest, Bank Reform and Interdiscursive Alignment

At the heart of the outrage over the Crisis lay a fault line, built deep into the structure of financial services: conflict of interest. For a little while, this fault line became visible and was hotly debated. For example, Greg Smith, a Goldman Sachs employee, wrote a New York Times editorial on the day he resigned from the bank in March 2012. According to Smith, “to put the problem in the simplest terms, the interests of the client continue to be sidelined in the way the firm operates and thinks about making money” (Smith, 2012). Smith’s motivation may not have been entirely altruistic, but he recognized a real problem, namely conflict of interest.

For a generic definition of conflict of interest, applicable to multiple contexts, we draw on Moore et al. (2005), who examined business, law, medicine, and public policy settings. They define conflict of interest as a scenario in which “professionals face a conflict between their professional responsibilities to protect the interests of their constituents, shareholders, patients, clients, or students, and their own self-interest” (Moore et al., 2005, p. 1).

In the aftermath of the Great Financial Crisis (which by then included not only the US subprime mortgage debacle, but also the Eurozone credit crisis), the US and UK governments and the European Commission all sought high-level, independent expert advice. Three high-profile reports ensued (named, after their chairmen, the Volcker, Vickers, and Liikanen reports), all naming conflicts of interest as the source of the banking crisis. In particular, the reports note the failure to compartmentalize risky trading activities (Liikanen, 2012), the lack of separation between retail “utility” banking from investment banking and corporate finance activities (Vickers, 2011), and the presence of proprietary trading in the same banking group that serves retail customers (Sereix, 2022). All three reports suggest radical structural solutions as the basis of regulatory reform. Vickers, for example, recommends “full separation across the banking sector as a whole, meaning to enforce a complete structural separation across the sector” (Korotana, 2016, p.197) and Volcker explicitly demands that “banks are not [to be] allowed to engage in any trading activities which might incur material conflict of interest”³ (Sereix, 2022, p. 157).

Once the conflict-of-interest problematic was identified in the media and political realms (with the help of the Volcker, Vickers, and Liikanen reports) as a major source of the Crisis, it was possible to mobilize reform plans predicated on radical structural reform (Herzig & Moon, 2013). This is why the focal point of our study is the discourse around conflict of interest, and alternative diagnoses of the crisis.

In the literature on banking reform (Korotana, 2016; Sereix, 2022; Vickers, 2019), studies agree that the banking reform acts adopted in the US and UK are less radical than intended by the expert advisors and that the separation measures put in place are “not safe enough” (Vickers, 2019). For example, in the US, the Dodd-Frank Act did introduce prohibition of proprietary trading in universal banking groups. However, Congress embedded an exemption relating to “certain market-making-related activities.” This exemption blurred the line between market-making and proprietary trading and thus permitted market-making-related hedging transactions (Sereix, 2022). Four years later, it was this exemption that allowed four rogue traders in JPMorgan, under the disguise of hedging, to lose $6.2 billion in proprietary trades (Sereix, 2022; Thomas, 2016).

In the UK, the Vickers report recommended that universal banks “ring-fence” their retail operations within a separate legal entity to ensure that, in case of trouble, utility banking operations could continue and that bail-out costs for taxpayers would at least be contained. In order to prevent ring-fenced banks from engaging in proprietary trading, Vickers recommended not only separation, but also “electrifying” the ring-fence to keep utility banking and high-risk banking separated. Yet 10 years on, legal experts and Vickers himself agreed that such “structural separation measures have been rather limited” (Vickers, 2019, p.1), and that only “a watered-down version of electrification” had been enacted (Korotana, 2016, p. 197).

These and other studies generate our puzzle: how and why did this systematic “watering down” of reform occur?

We draw on two theoretical frameworks. First, we utilise Hilgartner and Bosk’s (1988)“public arenas” model of the process by which social problems rise and fall—that is, how they do or do not come  to be seen as important. In our study, we consider four relatively discrete public arenas that were implicated in regulatory reform and have been less studied in the neoliberal / marketization discourse literature (which hitherto focussed on the media). These four are the (1) political, (2) legislative, (3) regulatory, and (4) consulting arenas.⁴ The “public arenas” model postulates that in each such arena, social problems, and definitions of problems, compete for societal attention and that “linkages among public arenas produce feedback that drives the growth of social problems” (Hilgartner & Bosk, 1988, p.53). The model focuses on the power dynamics of the operatives who promote and attempt to control particular problems. These operatives cluster in distinct arenas (in our case, consulting, regulation, political hearings, and legislation), each having their own distinctive “carrying capacity” (which limits the number of problems that can gain their attention), and “principles of selection” (the institutional-cultural factors that influence their problem formulations). We read Mautner’s (2010) model of interdiscursive alignment as an implicit extension of Hilgartner and Bosk’s. It focuses on explaining the well documented “managerial assault” of the market logic, that is, its continuing reach into arenas where it had not been present before (Sandel, 2013). Mautner postulates that business and other subsystems are related in terms of their relative power and the dynamics of their mutual impact, which are detectable through the linguistic traces that different subsystems leave in others. As the economist Deidre McCloskey argues, changes in public discourse precede social transformation. In other words, “talk matters” because it “changes things” (Martin & Storr, 2012; McCloskey, 2010). In Mautner’s terms, talk matters because some subsystems allow others to colonize them by committing linguistic accommodation or, conversely, resist others by refusing to accept their discourses.

These linguistic acts are detectable in each subsystem’s discourse and can be surfaced by careful discourse and/or content analysis. Although Mautner (2010) examines a series of bilateral relationships between the business subsystem and others (such as higher education and public sector administration), she suggests that the model may be applicable to more complex constellations. All entities in this model have varying permeable boundaries—and the more permeable they are, the more   that linguistic accommodation will occur between the entities involved (Mautner, 2010), or as Hilgartner and Bosk (1988) would have it, the more permeable the boundaries, the more that feedback and synergies between subsystems can occur.

Mautner (2010) postulates that it is primarily business that exerts power over other subsystems, both overtly (for example, by controlling funding streams) and covertly (indirectly, informally, culturally). As members of the less-powerful subsystems orient themselves towards the more powerful one, they accommodate (adopt) the “talk” and linguistic repertoire legitimate in the dominant subsystem, which over time solidifies into discursive practice and a new normative logic in the less-powerful subsystem.

Linking Mautner’s (2010) theory of interdiscursive alignment to Hilgartner and Bosk’s (1988) public arenas model, we postulate and show that, among these four arenas, an interdiscursive alignment was happening, in the course of which certain problem-definitions by subsystem operatives (such as conflict of interest, “too big to fail”) and solutions (structural reform, “break up the big banks”) became unsayable and unthinkable. Meanwhile, other problem-definitions and solutions rose to prominence under the umbrella of improving bank culture (including “risk culture”,“conduct”, “compliance”, and “disclosure”), positing culture as both the problem and the solution. We contend that this “watering down” of banking reform left linguistic traces in the discourse around banking reform that took place not only in the political, regulatory and legal arenas, but also in an important and vocal fourth subsystem: that of consultants. In fact, we see the consulting domain as the dominant arena, capable of shaping and colonizing the reform discourse due to (a) its large “carrying capacity” and (b) an especially powerful “principle of selection”, namely the seductive institutional myth of the capable, autonomous, self-regulating market actor.

Methodology

Our initial objective is simply to establish an empirical phenomenon: the displacement of one type of public ethical discourse (explicitly centred on conflicts of interest) by another (focused on managerial practices, themselves focused on “culture” and its variants). Or, as the moral philosopher Mary Midgley would have it, we are interested in demonstrating the “degeneration of one dazzling insight” (the need for sweeping structural reform to address the conflict-of-interest problematic) into the kind of pragmatic settlement (Midgley, 2011) that turned the attention of financial institutions to their culture, leaving the structural fault lines largely intact. We had an initial sense of this shift as we followed political, regulatory, legislative and consulting discourses since the crisis began in 2007. Accordingly, our initial question had to be – ‘Is it really so?’ The financial crisis brought multiple possibilities of banking reform into focus and presents us with rich ‘strategic research material’ (Merton, 1987). We identified research sites and events (Merton, 1987) that we expected would help us  identify patterns in discourses about what we call “problem–solution definitions” in relation to banks’ egregious actions and defective ethics. In other words, we loosely follow Foucault’s notion of “problematization” by postulating that discourses of both problems and their solutions emerge together and are often co-dependent, especially in the diagnosis of public problems by subsystem operatives.

Our research materials include the following:

• The heated political discourse played out in public hearings before the US Senate Committee on Homeland Security and Governmental Affairs in 2010, followed by similar inquiries by the UK Parliament in 2012–2013. Transcriptions of the hearings are publicly available.

• Subsequent legal, regulatory, and advisory (consulting) reports, which pressured practitioners to overhaul banking practices.

• The US Dodd-Frank Act (2010) and the UK Financial Services Act (2013).

We trace the evolution of, and interactions among, these different discourses, to understand how the participants collectively constructed and reconstructed various diagnoses and remedies, or ‘problem–solution definitions’, for the financial crisis. We began our search for problem–solution clusters by using the documentation that Power et al. (2013) build on. We “cleaned” this text database by removing academic papers and media articles and complemented it by hand-collecting electronic reports from all major consulting firms and regulatory postmortems pertaining to the financial services industry during 2009–2017. Table 1 summarizes the corpus collected across the four arenas.⁵

Table 1 Textual resources for content analysis

We also conducted informal discussions with two senior executives at an international bank, one with the title Global Head of Regulatory Affairs, the other being the Global Head of Compliance (two meetings with each individual). We asked these executives to judge if our sample was representative. They shared internal review documents that reassured us that our database of documents was representative of the industry discourse and as comprehensive (and possibly broader) than that maintained by the regulatory and compliance function of a typical global bank.

Overall, we hand-collected, reviewed and coded 141 documents, containing 8527 pages, as our document sample (see Appendix 1 for the full list). Electronic text analysis (Adolphs, 2006) has long been used in the political-history literature to measure scope and typicality in language (see, for example, Blaxill, 2017; Blaxill & Saleh, 2021) and is increasingly applied to business ethics research, too (Lock & Seele, 2017). As Crane (1999, p. 243) notes, “content analysis can provide important insights into the salience of particular social issues” by coding and interpreting latent content or, to put it differently, by “empirically reading between the lines” (Lock & Seele, 2017, p. 158). Content analysis can be conducted both in a quantitative mode (such as word-counting, resonance analysis, and identifying word networks) and in a qualitative mode, the latter by examining texts more closely with a specific research question in mind. We focused on texts written explicitly to propose solutions to the predicament of the financial services sector.

Based on an initial reading of the introductory/summary sections of each document, we determined whether its problem-and-solution definition was structural or managerial. We took note of the wordings “giving away” these patterns. For structural definitions, the following terms were apparent: “too big to fail”; “conflicts-of-interest” (problem-definition); and “separation”; “break up the big banks”; “ring-fencing” or “consider structure” (solution-definition). As for the culture agenda, “toxic/bad culture” and “conduct/compliance” indicated a problem definition and “improving culture/conduct,” “tone from the top” and “conduct/compliance risk management” indicated a solution definition. Overall, such cultural solutions represent not only the “riskification” but also the “managerialization” and “marketization” (Mautner, 2010) of the governance and regulation of banks.

We used these linguistic traces in the next phase, a quantitative content analysis of the frequency (wordcount) and salience (wordcount/number of pages) of these wordings in each text.

First, we determined if a document pertained more specifically or less specifically to conflicts of interest by counting and interpreting each expression of that concept. We conducted content analysis to calculate the number of times the terms “conflicts of interest”, “too big to fail” and “culture” were used in comparison to the number of pages in the document. The frequency and salience counts revealed significant differences amongst documents in terms of their construction of the banking crisis. But we did not consider word counts to be definitive. We verified that these terms were used diagnostically by reading the context in which they appeared.

Next, we counted the frequency of the terms related to structural reform (such as “structure”, “separation” and “ring-fencing”) and cultural reform (such as “culture”, “conduct” and “compliance”). The absolute and relative frequency counts suggested the type of solution the documents conveyed. Again, we verified whether these terms were used as part of reform proposals by reading the relevant context in which they appeared.

Finally, based on the word-counting exercise and applying “human coding”, that is, our own judgment, we classified each text as follows:

Pattern 1 (conflict-of-interest): documents with a significant count of “conflict of interest” as the diagnosis and  with structural-solution terms as the solution.

Pattern 2 (diluted conflict-of-interest): documents that recognized “conflict of interest” as a problem in financial services but did not use any terms referring to structural reform and instead offered managerial solutions related to the culture agenda.

Pattern 3 (culture-and-conduct agenda): in these documents conflicts of interest were not mentioned at all. Instead, the roots of the banking problem were described in terms of cultural weaknesses (e.g. toxic/bad cultures and lack of compliance/risk culture) and the solutions were described in terms of improving or managing culture, conduct, or compliance and disclosure practices.

In sum, the wordcounts were only a first indication of these patterns; reading the context in which these terms occurred provided us with further confidence in the classification of the problematization patterns.

Our classification process involved three research assistants and one of the authors. Each document was reviewed by at least two people and classified as Political, Regulatory, Legislative or Consulting. For each document in which conflict of interest was a recognized diagnosis, we classified the problem–solution pattern as Pattern 1 if predominantly structural solutions were recommended or Pattern 2 if predominantly managerial solutions were prescribed. Each document was coded by two assistants and in cases of difference, one author acted as an arbiter. We were able to classify 128 of the 141 documents – the rest were impossible to classify due to lack of completeness or of relevance to our research question. The results of our word-counting exercise, the subsequent classification of problem–solution definitions, and the aggregate analyses of these variables across arenas were carried out in MS Excel. Table 2 and Appendix 2 illustrate the process by showing examples of our classification, step by step.

Table 2 Content analysis (quantitative and qualitative) and classification

Overall, our content analysis reveals different problem–solution configurations as they were advanced by various institutional actors, enabling us to track the degeneration of the “conflict-of-interest problematic” into the “culture agenda.”

Findings

Global Overview

First, we examine the distribution of the documents over time and over arenas (Political, Legislative, Regulatory, Consulting). Documents pertaining to the Great Financial Crisis in the political domain were published in 2009–2015 (peaking in 2011–2012); we found no more in 2016–2017. The consulting arena activated early (in 2009) and produced a steady and increasing flow of texts related to banking reform throughout our time horizon. Regulatory documents became more prevalent from 2012, taking in previous discourses and the US and UK legislative acts, that is, the Dodd-Frank Act and the Financial Services Reform Act 2013 (our two legislative documents). Figure 1 shows the distribution of our corpus across time.

Fig. 1

Document sample (N = 141) across arenas and time

We examined the ebb and flow in the explicit appearance of problem–solution patterns. To further differentiate them, we associated them with the arenas in which they had been originated. Table 3 summarizes the variation of problem–solution patterns across the four sectors.

Table 3 Problem–solution patterns across the four arenas, by number of documents

Next, Fig. 2 shows how these patterns varied across time and arenas:

Fig. 2

Number of documents exhibiting a particular problem–solution pattern

The timeline projected in Fig. 2 shows that, in our document database over 2009–2017, the discourse around conflicts of interest did begin in the banking industry in 2009 but disappeared (or was muted) by 2016. The political arena recognized the conflict-of-interest problematic and half of its documents suggested structural solutions (Pattern 1). However, over time, in the political arena the initial Pattern 1 (7 documents between 2009 and 2012) was complemented by Pattern 2, emphasizing toxic bank cultures as an important cause of bank failures. In 2012, the political arena presented all three patterns of problem–solution definition, suggesting the dilution of the initial structural problem–solution cluster. By 2013, Pattern 1 had disappeared from the political discourse and Pattern 3 had taken over; the banking crisis was discussed in terms of a cultural failure that required a cultural (managerial), not a structural, solution.

It is therefore not surprising that both the Dodd-Frank Act (2010) and UK Financial Services Act (2013) displayed Pattern 2; conflict of interest was recognized as the problem, but the solutions prescribed were a mixture of “watered down” structural safeguards (such as “ring-fencing” arrangements in UK banks) plus an avalanche of managerial measures (capital requirements for US banks and internal preoccupation with compliance in UK banks).

The regulatory discourse was thus divided between the diluted (Pattern 2) and the culture-focused (Pattern 3) problem–solution clusters. Two regulatory reports, the Salz Review (which addressed a failure at Barclays in 2013) and a Bank of England report in 2015 (on the failure of HBOS) stand out as displaying Pattern 1 in that they seem to capture the media outrage over these bank failures (Arnold, 2015). However, the regulatory discourse was dominated by a sense that the banking industry needed cultural reform (as also seen in Table 3).

Finally, consultants were quick to react, creating a cottage industry of risk-culture solutions. The first response came from a survey consultancy, Insync Surveys and PricewaterhouseCoopers (PwC). Both gave culture as the diagnosis and proposed a new product for the corporate governance market, the “risk culture survey” (Boards Insync, 2009; PwC, 2009). McKinsey & Co. and the other Big Four firms followed, urging managers to “rethink risk management” and giving them a framework with which to think about “risk culture breakdowns” (McKinsey & Co., 2010). Suddenly, executives and board members had to know : “what is your company’s risk culture?” (KPMG, 2010).

Interdiscursive Adjustments

According to Knoepfel et al. (2007), public policy plays out in a cycle of five phases: (1) problem emergence, (2) agenda setting, (3) policy formulation, (4) implementation and (5) evaluation. If so, one would expect financial services reform to start in the political arena, in which discourse (in a democratic state) reflects popular concerns (Knoepfel et al., 2007). Then legislators (endowed by parliamentary committees with a legislative function) become implicated. The legislature exercises its intervention authority when the political arena considers a given situation to be a public issue. Regulatory agencies promulgate standards to achieve the objective of the legal framework set by legislative powers. These agencies introduce concrete measures—regulations with legal force that companies from the private sector are expected to implement. Finally, consulting firms provide advisory services to help companies comply. Consulting firms and regulators also aim to determine the results and effects of policy.

According to our content analysis, the discourse on banking reform in the US and the UK indeed started in the political arena: in parliamentary and congressional hearings reflecting the public’s outrage (Bait, 2012; Czarniawska, 2012; Herzig and Moon, 2013). The perceived need for reform triggered requests and proposals for action. But the problem-definition and solution-finding processes did not only implicate state administrative and legislative actors; operatives in the consulting and regulatory arenas were also activated at the same time.

Initially, there were numerous public and expert calls for “breaking up the big banks” on the principle that “if a bank is too big to fail, it is too big to exist” (Jopson et al., 2016; Herzig and Moon, 2013). Philip Augar argued in the Harvard Business Review for a “better way to break up banks,” guided by the rule: “If you trade in markets, you cannot speak to clients” (Augar, 2010, p. 2). Thus, an expression of Pattern 1 emerged, as exemplified bu Augar:

“As well as reducing the scale and systemic threat of the investment banks, splitting them in this way would also put paid to conflict of interest, still one of the nastiest features of modern finance.” (Augar, 2010, p. 2).

Economist Nouriel Roubini went even further, suggesting that universal banks should be split up, too: “The only way to avoid [conflicts of interest] is to break up these financial supermarkets. When you have within the same firm commercial banking, investment banking, asset management, prime brokerage, insurance, underwriting, derivatives…there are no Chinese walls and there are massive conflicts of interest” (quoted in: Ro, 2012, online).

The implication was that banks that trade for themselves or for customers should be prohibited from giving advice. Investment banks thus would need to be split into advisory and trading firms. Similar recommendations were made by the Salz Review and the Bank of England review that diagnosed failings at two UK banks, Barclays and HBOS (Table 4).

Table 4 Examples of Pattern 1 discourse

But this problem–solution definition was not adopted by the other arenas.

We see two patterns emerging almost immediately in the aftermath of the Great Financial Crisis. On one hand, a dilutive discourse emerged, which we call Pattern 2 (the dilution of the conflict-of-interest agenda). This pattern identified conflicts of interest as a cause of the financial crisis—but often accompanied by other, “cultural” problems—and advocated softer remedies by calling on banks to strengthen their professionalism and improve their “risk-related attitudes, ethical values and standards of behaviour (i.e. risk culture)” (Deloitte (2017, p. 12.). Accordingly, Deloitte (2017) advocated that banks ought to invest in “technology that can proactively identify and manage conflicts.”

Table 5 further illustrates Pattern 2 discourse.

Table 5 Examples of Pattern 2 discourse

Meanwhile, Pattern 3 emerged, identifying toxic culture directly as a problem in itself and recommending managerial reform as the solution. This was strongly contrary to the initial structural problem–solution diagnosis (which briefly emerged in the political and the regulatory arena). The main force of Pattern 3 discourse was its silencing of the conflict-of interest discourse. In not one of the documents in which we identified Pattern 3 does the phrase “conflict of interest” surface at all. The problematic of “risk culture” (and its various guises as concern with compliance, conduct, organizational ethics, and “bad behaviours”) provided linguistic closure around managerial problem–solution patterns in a way that made structural fault lines—and the forceful and radical structural solutions meant to remedy them—not only unmentionable, but even unthinkable (Mautner, 2010).

Pattern 3 was particularly forcefully advocated by the Institute of International Finance (IIF), a self-regulatory association of the banking industry. Aiming at “strengthening practices for a more stable system”, the IIF issued a report in 2009, which emphasized defective risk culture as the root cause of the Crisis and management of culture as its solution:

“Most importantly, Boards and managements should realize that culture is not a given and can be changed; that fact, which is clear from many firms’ experiences, implies opportunity and also responsibility to do better at fostering a productive yet risk-sensitive and disciplined culture. Management will need to focus on it and governance processes will have to be designed to work against erosion of risk management standards and a risk-sensitive culture, especially as the next booms emerge either in product areas or generally” (IIF, 2009, p. 2, emphasis added).

Table 6 provides further illustration of Pattern 3 discourse.

Table 6 Examples of Pattern 3 discourse

Thus, over time, the conflict-of-interest discourse became diluted as actors in the legislative and regulatory arenas turned towards managerial solutions—which had been readily offered by the consulting subsystem—itself highly responsive to corporate client interests. We see that this dilution took place in the context of intensive marketing and lobbying by consulting firms and by the banks themselves (via their own self-regulatory bodies such as the IIF). As we see in Fig. 2, the regulatory and legislative arenas accommodated the diluted conflict-of-interest agenda (Pattern 2) in 2010–2013 and then also absorbed the culture agenda (Pattern 3) from 2014 onwards.

Risk culture has turned out to be a slippery notion. Power et al., (2013, p. 4) observe that, “risk culture is not a static thing but a continuous process, or processes, which repeats and renews itself.”  It is a mixture of formal and informal processes, mobilized to recover an organization’s ability to understand, make visible and manage the inevitable trade-offs between risk-taking and control. The risk-culture agenda potentially includes a multitude of managerial attempts to regain control over risks, from the centralization of risk management functions to incentive system (re)design and the “responsibilization” of senior managers (Power, 2007). Collectively, these efforts represent everything but the substantive legal and structural response to conflicts of interest.

Discussion

Based on content analysis, we have shown a progressive displacement of the conflict-of-interest agenda of structural reform for banking as a response to the Great Financial Crisis of 2007-2009 in favour of an institutional focus on the banks’ cultures and practices. We explain and theorize this shift by drawing on Hilgartner and Bosk’s (1988) public arenas model and by applying Mautner’s (2010) interdiscursive adjustment model to our four arenas of interest.

Interdiscursive Alignment Among Public Arenas

Social and economic problems (such as the structural fault lines and conflicts of interests revealed by the Great Financial Crisis) compete with each other not only for public attention but also for problem definitions—arising in various public and private arenas—aligned with institutional capacities for solution. Both Hilgartner and Bosk (1988) and Mautner (2010) propose an ecological model of social problems which highlights “the resource constraints that human actors face in constructing problems” (Hilgartner & Bosk, 1988, p. 56) and implicate social subsystems with their discourses, counter-discourses, and alignments oriented towards the more powerful social actors (Mautner, 2010).

Our data support this choice of framework. We find successive waves of overlapping and competing problem–solution definitions emerging, in the wake of the crisis, from different arenas. Although the crisis had no doubt been incubating for many years, its immediate onset was a time of public outrage and political shock, combined with emergency management by the monetary authorities. There was a “hot” initial period in which cool retrospective diagnosis was overshadowed by blame and moral revulsion towards bankers. Our analysis begins a little later, with the public hearings in the political domain. This arena sought to explicate the structural nature of conflicts of interest, which seemed to be inherent in modern financial services. Given this problematization of the financial crisis, breaking up banks and prohibiting activities that could give rise to conflicts of interest emerged as remedial options (Augar, 2010). During this period, politicians also called out banks for their lax professional and ethical standards and neglect of client needs.

This  seemingly ubiquitous conflict-of-interest discourse   was not, however, replicated in the legal arena, where the concept of conflict of interest surfaced explicitly only as part of the Dodd-Frank Reform Act (2010) and in the Financial Services Reform Act (2013) in the UK. In this legal arena, the problem–solution discourse focused on a trickle of watered-down structural measures overshadowed by a tsunami of culture-management processes and demands for capital requirements, adequate and faithful disclosure, risk management, and compliance. Hence the process of reforming bank regulation also involved a displacement and transformation of the discourse from something that politicians and legislatures did not have the capacity to directly address—structural conflicts of interest—to something that they could address—the problem of adequate managerial processes, disclosure, and compliance. It can be argued that the decline of the debate about “breaking up the big banks” reflects how the structural remedy properly belonged to, and stayed confined to, the realm of politics. However, the political arena also shifted its orientation. Politicians dropped the conflict-of-interest agenda and increasingly focused on “bad behaviours” and “bad cultures.” From this point onwards, our content analysis suggests, both politicians and regulators became more preoccupied with toxic cultures at the organization and field levels. The structural problem–solution discourse could not compete for public attention with the emerging focus on culture.

The problem –solution discourse of culture originated within the banking arena itself and was established by the IIF in 2009 as a swift response to the public’s moral outrage. It is plausible to see this as a defensive move to forestall the structural problematization of banking and its implications for reform. At that time, there were threats to break up the big banks in the UK and to bring back the Glass-Steagall Act in the US, which would have split up some of the largest banks in the world (Salz & Collins, 2013). We see, then, not only that there can be waves of successive problem definitions as Hilgartner and Bosk (1988) argue, but also, as our analysis suggests, these waves can  overlap as different problem–solution definitions emerge in reaction to others. We suggest that the competition between conflict of interest and culture softened as the latter problem-definition became more prominent. Over time, the political arena transitioned to a new discourse framing culture as both the problem in the financial services industry and its solution.

Hilgartner and Bosk (1988) also emphasize the networks of operatives who can reinforce problem definitions and build them into practice routines. Mautner (2010) argues that an interdiscursive alignment towards the most powerful operatives precedes the emergence of such new practice routines. In our analysis, consultants were quick to create advisory services relating to culture and conduct, noticeably devoid of specific worries about conflicts of interest. Conflict-of-interest policy documents at the organization level were subsumed into the culture/conduct reform agenda. Advice on culture was woven into many other services for management, reinforcing this problem–solution definition in all relevant arenas, so much so that even politicians and their publics have accepted it. The disappearance of the conflict-of-interests vocabulary by 2014 suggests that all the relevant actors had ceased to “call a spade a spade” as they had been doing quite loudly five years earlier. Ultimately, this new linguistic repertoire and the resulting discursive closure restricted the space for defining problems and solutions, leaving no room for the idea of structural reform, and gradually making it not only  unsayable but also unthinkable.

We observe that, initially, the conflict-of-interest agenda and the discourse of structural reform were able to gain attention across, and link, the arenas of politics, law and banking. Yet the capacities of these different arenas were not suited to sustain this attention, in part because structural reform was both drastic and risky in itself, but also because these arenas were focused on different remedies. The power or capacity of regulators to act on banks seems to influence their definition of, and attention to, the ethical issues. The emphasis on culture as a competing problem–solution definition became more prominent across these arenas, not least because consultants seemed capable of giving it operational life, fuelling the “culture explosion” in the discourse.

Rebirth of the Neoliberal Myth of the Capable, Self-Regulating Actor

While the public arenas model provides a compelling framework for our analysis of the shift in discourse from a conflict- to a culture-based problematization of the Great Financial Crisis, it does not yet fully explain why the turn to culture in the post-crisis period was so comprehensive. To do this, we draw on the ethics literature on the neoliberal paradigm, showing the power of marketization in the regulation and apparent reform of financial services. More precisely, we draw on the myths of neoliberal capitalism (Mautner, 2010; Panitch & Konings, 2009) to characterize the discursive   shift from an ethically highly charged and state-centered understanding of conflict of interest to a more managerial (self-regulatory) understanding. We argue that (a) this transition represents the ethical displacement and dilution of public concern with the inherent and prevailing structural fault lines in banking and that (b) this ethical displacement is driven by prevailing and persistent thought patterns—namely, the neoliberal myths of self-regulating organizations and markets.

Thus, conflicts of interest—a structural problem—acquired political prominence during the period of public moral outrage and public inquiries in the wake of the financial crisis. However, in the arena of legal-financial reform, the institutionalized and potent thought-pattern of law-making necessarily took over—one in which failures of disclosure (and other such managerial missteps) were defined as the problem, making disclosure and culture-reform the remedy. This thought pattern is deeply embedded in the reformist discourse on both sides of the Atlantic and has its roots in notions of how markets and information disclosure are efficient joint mechanisms for correcting failures of corporate governance and control. We designate such notions as “myths” because they project a long shadow and live on, even in the presence of contradictory evidence about their effectiveness. When there is competition for public attention,  dominant myths will favour some problem–solution definitions over others.

Midgley (2011) warns us that patterns of thought useful in one age or context can be highly problematic in another. If the neoliberal myth that the market is the arbiter of transparent pricing and coordinator of fair trade is used to frame the conflict-of-interest problematic, then it follows that the solution is the increased disclosure which will enable clients to be sufficiently informed to avoid unscrupulous financial actors. Yet a large body of research, drawing on psychology, business ethics, and behavioural economics, suggests that this “sub-myth” of disclosure is based on “misguided assumptions” and might therefore be counter-effective (Kinander, 2018).

Myths are often reshaped by or mingled with other thought-patterns (Midgley, 2011). Underlying our analysis, we propose that the neoliberal myth of the capable organization plays an important role in the observed shift in discourse. Hence the discourse of the Financial Crisis across each of the main arenas of concern needed ultimately to reconstruct and preserve a notion of competent organizational actorhood (Meyer & Jepperson, 2000). It needed to maintain the belief that self-reliant, adequately guided and incentivized financial institutions can in fact correct the faults of the market by carrying out their own internal behavioural interventions. The dramatic simplicity and optimism of these neoliberal myths is one of their chief attractions and explains why they dominate public arenas. Instead of framing the complex, messy structural and ethical fault lines underpinning conflicts of interest, these myths underwrite pragmatic solutions which unburden the state and its regulators and are relatively easy to legislate in the form of new disclosure and conduct requirements. This, in turn, mutes and obfuscates the original insight about the fault lines of financial capitalism; structural reform becomes diluted into an issue of corporate managerial change and its win–win logic (Lynn, 2021). From both Midgley’s and Meyer and Jepperson’s viewpoints, the emergence of culture as a distinctive problem–solution definition in the wake of the Crisis is itself an affirmation of capable organizational actorhood.

The issue for the structural problem–solution definitions of the Financial Crisis and their proponents can now be seen as much more than resource-constrained competition for public attention, as Hilgartner and Bosk argue. It is also deeply cultural in nature. Proponents of structural reform faced a large cultural disadvantage. Insofar as their discourses were neither animated by nor based on established institutional myths, their currency was always likely to be short-lived. Certainly, anti-trust and anti-monopoly values remain in play as features of the institutional myth of market competition, but the conflict-of-interest issue was not framed in these terms. In addition, “too big to govern” as a candidate for status as a myth that could support a structural problem–solution definition had only a brief public life because it could not attach itself to dominant prevailing myths. Regulatory bodies themselves also had little appetite for the complexities of structural reform, having already been blamed for being asleep on the watch. And consulting firms were always ready and willing carriers and curators of dominant institutional myths of purposeful and capable organizations.

Borrowing from Ungar (2008) and Lindsey (2007), we suggest that our case narrative also points to something even more fundamental; namely, a form of collective strategic ignorance embedded in dominant institutional myths about markets, information, and organizations. These collective ways of seeing and framing problems also become a way of not seeing, which keep specific problem–solution definitions from sustaining public attention. The “discursive competition” between conflict and culture as sources of problem definition may have seemed intense across diverse arenas, but it was institutionally loaded from the beginning. The gradual erasure of the ethically charged conflict-of-interest agenda was culturally inevitable in both the US and the UK.

To an anxious public, politicians and regulators in the end offered, as they had often done before, legislation and process improvements as a remedy and as an assurance that managers of financial institutions can be incentivized to reform bad behaviours. For example, in the UK, the Senior Managers and Certification Regime in financial services has emerged as a core feature of the conduct agenda. We do not analyse these more recent discursive shifts in regulatory arenas from culture to conduct, but it is plausible that they signal a further procedural specification of a culture-oriented reform agenda which might otherwise be too diffuse to sustain organizational attention. Bank regulators are likely to have an interest in privileging process remedies (such as risk culture and internal control reforms) over structural remedies. Breaking up the big banks would be opposed by the industry from the outset and would put regulators’ legitimacy at high risk. In contrast, process remedies have a curious feature, observed by Power (1994) in the context of audits: they are “remarkably invulnerable to their own failure” (Power, 1994, p. 7). Thus, the result of the intensification of process improvements, such as risk culture audits and incentive system reform, is that the underlying ethical problem (conflict of interest) and its structural solution are gradually displaced by a fuzzier problem (bad culture) which invites a plethora of possible managerial solutions. The original ethical fault lines are still there but, lacking support from institutional myths, they disappear from public sight.

Collective strategic ignorance in the face of ethical issues like conflict of interest, as we have identified it, may not be uncommon; in fact, the concept was originally developed in the climate-change literature to explain climate inaction (Oreskes & Conway, 2010, 2013). In the field of financial auditing, reform processes in the wake of corporate scandals have often called for structural change, given the obvious conflict of interest facing auditors paid and retained by the organizations they audit. Yet, such processes have historically resulted in new conduct rules and procedural reforms which leave the structure intact. It may also be that the present wave of enthusiasm for ESG as a solution to a range of prominent systemic issues, including climate change, fits the same model and results in systematic non-attention to structural sources of problems by blending easy-to-fix issues of governance with more contentious, trade-off-saturated issues of sustainability. We suggest that a public arenas model, modified to take account of the power of institutional myths, could have explanatory potential in these and many other settings.

In conclusion, we contend that our analysis of the transition from a political to a legalized and then to a more managerial understanding of conflict of interest reveals a dynamic of collective strategic ignorance resulting in the moral displacement of the concept of conflict of interest and its absorption into the problem of culture. Furthermore, it represents a shift in problem–solution definition from the structural level of the field of investment and universal banking to the organizational level in which critique is focused on “bad behaviours” by individuals operating in a given company’s “toxic culture.” It is a shift that alleviates and dilutes the moral burden on managers (Lynn, 2021), notwithstanding the ubiquitous mantra of “doing the right thing” which has accompanied the cultural turn in financial services.

Appendices

Appendix 1 Corpus Documents

Political Arena

• 1. The Commission of Investigation into the Banking Sector in Ireland (2011). Misjudging Risk: Causes of the Systemic Banking Crisis in Ireland.

• 2. Committee on Homeland Security and Governmental Affairs (2011). The Financial Crisis: Anatomy of a Financial Collapse.

• 3. Economic & Social Research Council (2014). Societal Dynamics and Trends in the UK: How can Financial Services Benefit from Understanding Society?

• 4. The Group of Thirty (2012). Toward Effective Governance of Financial Institutions.

• 5. The Group of Thirty (2015). Banking Conduct and Culture: A Call for Sustained and Comprehensive Reform.

• 6. Commonwealth of Australia House of Representatives (2009). The Global Financial Crisis and Regional Australia.

• 7. Independent Commission on Banking (2011). Vickers report.

• 8. Liikanen, E., Banziger, H., Campa, J.M., Gallois, L., Goyens, M., Krahnen, J.P., Mazzucchelli, M., Sergeant, C., Tuma, Z., Vanhevel, J. and Wijffels, H. (2012). High-level Expert Group on Reforming the Structure of the EU Banking Sector.

• 9. National Commission on the Causes of the Financial and Economic Crisis in the United States (2011). The Financial Crisis. Inquiry Report. Final Report of the National Commission on the Causes of the Financial and Economic Crisis in the United States.

• 10. The Parliamentary Commission on Banking Standards (2012). Changing Banking for Good. Report of the Parliamentary Commission on Banking Standards. Volume 1: Summary, and Conclusions and recommendations.

• 11. The Parliamentary Commission on Banking Standards (2012). Changing Banking for Good. Report of the Parliamentary Commission on Banking Standards. Volume 2: Chapter 1 to 11 and Annexes, together with formal minutes.

• 12. The Parliamentary Commission on Banking Standards (2013). 'An Accident Waiting to Happen': The Failure of HBOS.

• 13. The Parliamentary Commission on Banking Standards (2013). Banking Commission Publishes Report on Changing Banking for Good. (www.parliament.uk)

• 14. Science and Technology Select Committee (2011). Behaviour Change.

• 15. Select Committee on Economic Affairs (2009). Banking Supervision and Regulation.

• 16. Treasury Committee (2009). Banking Crisis: Dealing with the Failure of the UK Banks: Government, UK Financial Investments Ltd and Financial Services Authority Responses to the Seventh Report from the Committee.

• 17. Treasury Committee (2009). Banking Crisis: Reforming Corporate Governance and Pay in the City.

Legal Arena

• 18. Senate and House of Representatives of the United States of America in Congress (2010). The Dodd-Frank Act.

• 19. UK Parliament (2013) Financial Services (Banking Reform) Act.

Regulatory Arena

• 20. Association of Chartered Certified Accountants & Economic and Social Research Council (2014). Culture and Channelling Corporate Behaviour.

• 21. Association of Chartered Certified Accountants & Economic and Social Research Council (2014). Culture and Channelling Corporate Behaviour. Appendix 1: Review of the Academic Literature on Organisational Culture.

• 22. Association of Chartered Certified Accountants & Economic and Social Research Council (2014). Culture and Channelling Corporate Behaviour. Appendix 2: Findings of the ACA-ESRC Roundtable Discussions.

• 23. Association of Chartered Certified Accountants & Economic and Social Research Council (2014). Culture and Channelling Corporate Behaviour: Summary of Findings.

• 24. Bank of England (2012). The Bank of England, Prudential Regulation Authority. The PRA’s Approach to Banking Supervision.

• 25. Bank of England (2015). The Failure of HBOS plc (HBOS): A Report by the Financial Conduct Authority (FCA) and the Prudential Regulation Authority (PRA).

• 26. Central Bank of Ireland (2017). Is it Legal? A Question of Culture.

• 27. Chartered Institute of Internal Auditors (2014). Culture and the Role of Internal Audit. Looking Below the Surface.

• 28. Chartered Institute of Internal Auditors (2016). Organisational Culture. Evolving Approaches to Embedding and Assurance.

• 29. Chartered Insurance Institute (2016). Developing a Culture of Personal Responsibility in a Regulated Environment.

• 30. Chartered Management Institute (2014). The Moral DNA of Performance. Better Values, Better Decisions, Better Outcomes.

• 31. Committee on Homeland Security and Governmental Affairs (2013) JPMorgan Chase Whale Trades: A Case History of Derivatives Risks and Abuses.

• 32. CRO Forum (2015). Sound Risk Culture in the Insurance Industry.

• 33. European Securities and Markets Authority (2012). Guidelines: Guidelines on Certain Aspects of the MiFID Compliance Function. ESMA 2012/387

• 34. European Securities and Markets Authority (2012). Guidelines: Guidelines on Certain Aspects of the MiFID Compliance Function. ESMA 2012/388

• 35. European Securities and Markets Authority (2012). Guidelines on Certain Aspects of the MiFID Suitability Requirements. ESMA 2012/387

• 36. European Securities and Markets Authority (2012). Final Report: Guidelines on Certain Aspects of the MiFID Suitability Requirements. ESMA 2012/388

• 37. European Securities and Markets Authority (2012). Guidelines on Remuneration Policies. ESMA 2013/606

• 38. European Securities and Markets Authority (2015). Best Execution under MiFID. ESMA 2015/494

• 39. European Securities and Markets Authority (2015). Final Report on Complex Debt Instruments and Structured Deposits.

• 40. European Securities and Markets Authority (2015). Final Report on Guidelines for the Assessment of Knowledge and Competence.

• 41. European Securities and Markets Authority (2015). Guidelines: Guidelines on Cross-selling Practices.

• 42. European Securities and Markets Authority (2016). Final Report: Guidelines on Cross-selling Practices.

• 43. European Securities and Markets Authority (2016). Suitability Peer Review- Annex.

• 44. European Securities and Markets Authority (2016). Suitability Peer Review- Final Report.

• 45. Financial Conduct Authority (2015). Financial Crime: a Guide for Firms. Part 1: A Firm’s Guide to Preventing Financial Crime.

• 46. Financial Conduct Authority (2015). Financial Crime: a Guide for Firms. Part 2: A Firm’s Guide to Preventing Financial Crime.

• 47. Financial Conduct Authority (2016). Handbook (excerpt).

• 48. Federal Reserve Board of Governors (2017) The Role of the Board at Large Financial Firms.

• 49. Financial Stability Board (2013). Increasing the Intensity and Effectiveness of Supervision.

• 50. Financial Stability Board (2013). Thematic Review on Risk Governance: Peer Review Report.

• 51. Financial Stability Board (2014). A Culture Ratio is More Important than a Capital One.

• 52. Financial Stability Board (2017). Guidance on Supervisory Interaction with Financial Institutions on Risk Culture. A Framework for Assessing Risk Culture.

• 53. Financial Stability Board (2017). Implementation and Effects of the G20 Financial Regulatory Reforms.

• 54. Financial Stability Board (2017). Reducing Misconduct Risks in the Financial Sector: Progress Report to G20 Leaders.

• 55. Hong Kong Securities and Futures Commission (2013). Risk-focused Industry Meeting Series: G-SIFI Trends in Risk and Risk Mitigation.

• 56. Institute of International Finance (2009). Risk Culture. Reform in the Financial Services Industry: Strengthening Practices for a More Stable System. The Report of the IIF Steering Committee on Implementation (SCI).

• 57. Institute of Risk Management (2012). Risk Culture. Under the Microscope. Guidance for Boards.

• 58. International Monetary Fund (2012). Effects of Culture on Firm Risk-Taking: A Cross-Country and Cross-Industry Analysis.

• 59. Kay, J. (2012). The Kay Review of UK Equity Markets and Long-term Decision Making.

• 60. N.A. (2013). HBOS Review: Key Findings.

• 61. N.A. (2013). JP Morgan Whale Trading.

• 62. Prudential Regulatory Authority (2013). Operational Risk.

• 63. Prudential Regulatory Authority (2015). Internal Governance.

• 64. Prudential Regulatory Authority (2015). Remuneration.

• 65. Prudential Regulatory Authority (2015). The PRA Rulebook: Part 2.

• 66. Prudential Regulatory Authority (2015). The PRA Rulebook: Part 3.

• 67. Prudential Regulatory Authority (2015). The PRA Rulebook: Part 4.

• 68. Salz, A. & Collins, R. (2013). Salz Review: An Independent Review of Barclays' Business Practices.

Consulting Arena

• 69. Accenture (2013). Coming to Terms with Dodd-Frank. Balancing Strategic Considerations and Tactical Implications.

• 70. Accenture (2016). Ethics and Conduct Challenge for US banks: Learning from the UK experience.

• 71. Accenture (2016) New Rules for Culture Change.

• 72. AON Risk Solutions (2015). Risk Governance and Culture Frameworks for Banks.

• 73. Boston Consulting Group (2017). Purpose with the Power to Transform your Organization.

• 74. Boston Consulting Group (2017). Transforming Bank Compliance with Smart Technologies.

• 75. Board Insync (2009). Employee Risk Culture Survey.

• 76. Booz Allen Hamilton (2011) A Consultant View of Dodd-Frank.

• 77. Clifford Chance (2010). Dodd-Frank Wall Street Reform and Consumer Protection Act.

• 78. Corporate Research Forum (2010). Managing the People Dimension of Risk.

• 79. Corven Consulting (2012). Banking on Culture to Manage Risk: A Point of View by Corven Consulting.

• 80. Deloitte (2014). Dodd-Frank Act Push-out: Planning the Right Strategy.

• 81. Deloitte (2015). Corporate Culture: The Second Ingredient in a World-Class Ethics and Compliance Program.

• 82. Deloitte (2015). Cultivating a Risk Intelligent Culture. Understand, Measure, Strengthen, and Report.

• 83. Deloitte (2015). Getting the Right Outcome. Internal Audit and Retail Conduct Risk.

• 84. Deloitte (2016). Culture Shift. Changing Beliefs, Behaviours and Outcomes.

• 85. Deloitte (2016). Management Information on Culture: Connecting the Dots.

• 86. Deloitte (2016). The Fourth Element. How Addressing Culture can Improve Mission Performance in Government.

• 87. Deloitte (2017). Bringing it All Together. Financial Markets Regulatory Outlook 2018.

• 88. Deloitte (2017). Building World-Class Ethics and Compliance Programs: Five Ingredients to Meet Global Expectations.

• 89. Deloitte (2017). Compliance 2025: DNA Evolution in the Financial Services Industry.

• 90. Deloitte (2017). Managing Conduct Risk. Addressing Drivers, Restoring Trust.

• 91. Deloitte (2017). Would you Recognise the Warning Signs of a Toxic Culture?

• 92. EY (2010). Q&A on the Dodd-Frank Wall Street Reform and Consumer Protection Act.

• 93. EY (2011). Making Strides in Financial Services Risk Management.

• 94. EY (2011). Positioning for Change. US Financial Reform Six Months Later.

• 95. EY (2012). 4 Steps to Help Securitizers Comply With Dodd-Frank’s New Reporting Requirements.

• 96. EY (2014). Enhanced Prudential Standards for Foreign Banking Organizations.

• 97. EY (2015). Risk Culture. How Can you Create a Sound Risk Culture?

• 98. EY (2015). Non-Toxic: The Role of Bank Culture in Financial Reform.

• 99. EY (2015). The Challenges of Risk, Culture, Behaviour and Corporate Integrity in Financial Services. Findings from the MoralDNA TM of Culture and Conduct in Global Financial Services.

• 100. EY (2015). Understanding Risk Culture and its Challenges.

• 101. EY (2016). Cultural Assessments.

• 102. EY (2016). The Power of Culture.

• 103. EY (2017). Is your Board yet to Realise the True Value of Culture?

• 104. EY (2017). The Board's Role in Corporate Culture.

• 105. Independent Audit Limited (2011). It’s All About Behaviour: How Boards are Tackling Behaviour and Ethics.

• 106. IVEY Business Journal (2013). Leadership and Risk Culture.

• 107. KPMG (2010). Dodd-Frank for Foreign Banks.

• 108. KPMG (2010). What’s Your Company’s Risk Culture?

• 109. KPMG (2012). Dodd-Frank for Foreign Financial Institutions.

• 110. KPMG (2013). Parliamentary Committee on Banking Standards (PCBS) Report.

• 111. KPMG (2013). Parliamentary Committee on Banking Standards: Themes and Implications.

• 112. KPMG (2016). Organisations' Cultural Assessment and Transformation.

• 113. KPMG (2017). Compliance Transformation.

• 114. KPMG (2017). Driving corporate culture from the top.

• 115. KPMG (2017). Conduct Risk: Aligning Product, Customer, and Value.

• 116. MacQuarie University & CIFR (2014). Assessing Risk Culture and its Impact on Organizations.

• 117. Mayer & Brown (2010). Understanding the New Financial Reform Legislation.

• 118. McKinsey & Company (2010). After Black Swans and Red Ink: How Institutional Investors can Rethink Risk Management.

• 119. McKinsey & Company (2013). Managing the People Side of Risk.

• 120. McKinsey & Company (2014). Assessing and Enhancing your Risk Culture.

• 121. McKinsey & Company (2016). A Best-practice Model for Bank Compliance.

• 122. McKinsey & Company (2016). Taking Control of Organizational Risk Culture.

• 123. McKinsey & Company (2017). Culture for a Digital Age.

• 124. Morisson & Foerster (2011). SEC Proposes Dodd-Frank Conflicts of Interest Rules.

• 125. New City Agenda (2014). A Report on the Culture of British Retail Banking.

• 126. Poole College of Management (2013). Risk Culture's Critical Role in ERM.

• 127. PwC (2009). The Risk Culture Survey.

• 128. PwC (2014). US Regulatory Outlook: The Beginning of the End.

• 129. PwC (2015). Forging a Winning Culture.

• 130. PwC (2015). Stand out for the Right Reasons: Why you can’t Scare Bankers into Doing the Right Thing.

• 131. PwC (2016). Banking on Change. How to Respond to New Expectations for Audit Committees.

• 132. PwC (2016). Risk culture. Where to Go from Here.

• 133. PwC (2016). Stand out for the Right Reasons. The Senior Managers Regime: What it Really Takes to Sleep Soundly at Night.

• 134. PwC (2017). ERM Program Communication Not an Optional Activity.

• 135. PwC (2017). How your Board can Influence Culture and Risk Appetite.

• 136. PwC (2017). Mastering Culture and Conduct.

• 137. Thompson Reuters (2015). Tracing the True Origin of Bad Behaviour. New Ways to Predict Conduct Risk Exposure.

• 138. Thompson Reuters (2017). Mastering Culture and Conduct Risk 2017.

• 139. Thompson Reuters (2017). Top US Regulator Looks Abroad for Measurement Tools.

• 140. Towers Watson (2011). Measuring Risk Culture: A Powerful Approach to Demonstrate Embedding of Risk Management within an Organization.

• 141. WorkCover NSW (2012). Safety Culture Survey: Questionnaire.

Appendix 2. Methodology—Further Illustration

By identifying the causes and remedies proposed in each document, we were able to draw out patterns in the conflict-of-interest problematic.

Pattern 1 (conflict-of-interest agenda): the problem in banking is identified as “conflicts of interest” (“COI” or “too big to fail”); the suggested solution is structural (“structure” or “separation”).

Pattern 2 (dilution of the conflicts of interest agenda): the problem in banking is identified as “conflicts of interest”; the suggested solution is managerial in nature (‘culture’, ‘compliance’, or ‘conduct’).

Pattern 3 (culture-and-conduct agenda): the problem in banking is identified as managerial (e.g., bad or toxic “culture”); the solution is managerial (improving “culture”, “compliance”, “conduct,” or “disclosure”).

Document	Pages	Wordcounting				Qualitative content analysis		Classification of pattern
		“Conflict of interest” agenda (COI, TBTF)		“Culture” agenda (culture, conduct, compliance)		Diagnosis (example)	Solution (example)
		Frequency	Salience	Frequency	Salience
House of Lords (2009) Banking Supervision and Regulation	66	12	0.2	0	0.0	“there is an inbuilt conflict of interest” (p.40) “banks… can grow too big to fail” (p. 55).	“Policy-makers should grasp the chance to consider the appropriate structure of the financial system” (p. 55).	Pattern 1 (Conflict-of-interest agenda)
EY (2015) Understanding Risk Culture and its Challenges	7	0	0.0	90	12.9	“Of the issues currently bedevilling financial services firms, risk culture is one of the foremost” (p. 49).	“Reinforcing risk culture” (p. 51).	Pattern 3 (Culture-and- conduct agenda)
Accenture (2016) The Ethics and Conduct Challenge for US Banks: Learning from the UK Experience	12	7	0.6	58	4.8	“[C]ulture was a major contributor to misselling as well as the source of many of the problems that caused the market meltdown in 2008” (p.1). “conflicts of interest and ethics” (p.7).	“Invest in programs driving a strong culture” (p. 9).	Pattern 2 (Diluted conflict-of-interest agenda
Deloitte (2017) Building Worldclass Ethics and Compliance Programs_Five Ingredients to Meet Global Expectations	28	4	0.1	450	16.1	“[The 2008] meltdown exposed bribery and corruption, fraud, insider trading, conflicts-of interest, money-laundering, price-fixing, and Ponzi-schemes” (p. 3).	“A great ethics and compliance programme”. “Five ingredients”: (1) Tone at the top; (2) Corporate culture; (3) Risk assessments; (4) Chief Compliance Officer; (5) Testing and monitoring	Pattern 2 (Diluted conflict-of- interest agenda)
Deloitte (2017) Would you Recognize the Warning Signs of a Toxic Culture	12	0	0	159	13.25	“[O]rganizations involved in egregious or illegal activities, often as a result of toxic corporate cultures” (p. 1).	Culture indicators	Pattern 3 (Culture-and- conduct agenda)