RESEARCH DOSSIER
Organizational Culture and Financial Misconduct
Abstract. This research dossier examines how organizational culture affects financial misconduct through enacted priorities, occupational identity, coworker influence, workforce sorting, the expression and withholding of concerns, and the operation of accountability. A structured narrative synthesis integrates selected experimental, archival, organizational, and official documentary evidence. Contrasting banking identity experiments establish the limits of occupational generalization; studies of financial-adviser peer exposure, mobility, and contextual socialization extend the inquiry to recorded professional conduct. Regulatory and institutional documentation concerning retail banking, foreign exchange markets, and nonfinancial risk governance illustrates failures of information, challenge, and oversight. The analysis distinguishes underlying misconduct from its detection and legal or administrative disposition, separates culture from climate and prudential risk from conduct outcomes, and defines the conditions required for causal evaluation of organizational interventions. International and offshore applications are developed as empirically testable questions concerning actual financial functions and institutional arrangements. The contribution is an interdisciplinary analytical synthesis of established evidence and its limits, rather than a new theoretical system or a pooled causal estimate.
Keywords: Organizational culture; financial misconduct; ethical climate; financial advisers; occupational identity; coworker influence; employee voice; risk culture; accountability; misconduct measurement; financial organizations; international finance
Research Object and Disciplinary Boundaries
Financial misconduct occupies the intersection of individual action and organized economic activity. It includes unauthorized dealings, deliberate misrepresentation, improper sales practices, manipulation, misuse of confidential information, and other violations of applicable duties involving financial products or transactions. The operative meaning in a particular study must be specified by its measurement instrument or governing authority. A recorded complaint, a disciplinary disclosure, an administrative finding, an admitted offense, and a criminal conviction are distinct observations. They cannot be pooled into one count of proven wrongdoing without an explicit coding rule.
Organizational culture is the historically developed pattern of shared meanings, expectations, and ways of acting through which members interpret their organizational environment. Organizational climate concerns members’ perceptions of the policies and practices that receive practical priority. The distinction follows Schneider, Ehrhart, and Macey’s (2013) review of climate and culture. Culture includes the learned interpretation of why certain practices are appropriate; climate concerns what members experience as expected, supported, and rewarded. The two are related, and their empirical measures need not coincide.
The present research object is the relationship between these collective patterns and misconduct in financial institutions and professional financial networks. Its central explanatory task is to establish how socially learned interpretations and enacted organizational priorities influence individual conduct, detection, correction, and recurrence. General organizational psychology supplies the conceptual foundation. Financial decision-making supplies the specificity: contractual complexity, delegation, fiduciary or professional obligations where applicable, customer-information asymmetry, measurable revenue, and specialized authority.
The dossier belongs to Consumer and Economic Psychology within Psychology by Bogdanova and the Global Offshore System research program. Corporate governance, financial law, professional ethics, and economic incentives are adjacent objects. Their rules and mechanisms enter where they alter psychological expectations or organizational behavior. The legality of offshore structures is determined through relevant law and facts. The geographical location of a transaction cannot itself establish misconduct. This boundary permits the study of organizational risk in international finance without treating cross-border activity as a proxy for wrongdoing.
Questions, Method, and Evidentiary Scope
The research addresses five connected questions. How do collective expectations become behaviorally consequential? Which findings support an influence of workplace context on financial misconduct? Why can the same organizational pressures produce different responses? How do reporting and enforcement affect recorded misconduct? Which interventions follow from the evidence, and what remains untested?
The method is a structured narrative synthesis of selected directly relevant scholarly studies, general organizational research, and primary regulatory and institutional documentation. It is not a systematic review or quantitative meta-analysis. The selection emphasizes explicit operational measures, contrasting results, identifiable study designs, and documented institutional cases. Scholarly findings are distinguished from regulatory findings, retrospective institutional interpretations, and the author’s analysis. No participant-level data were obtained, no new observations collected, and no effect sizes pooled. The evidence ledger specifies the actual access level of each source. When only the abstract or publication summary was accessible, the manuscript confines factual claims to that material. This is particularly important for econometric research: its identification strategy can be described at the level supported by the available source, while full coefficient verification and independent replication remain outside this study.
Four evidence classes structure interpretation. Experiments test the response to a specified manipulation within a defined sample. Archival and quasi-experimental studies examine recorded behavior and exploit variation that may support causal inference under assumptions. Organizational surveys and action research illuminate experienced practices and implementation processes. Official inquiries and enforcement documents establish particular institutional failures or admitted conduct, with the legal and inferential limits attached to each document.
The sources were examined in November 2026. Publication years and access versions are retained separately where appropriate. This research synthesizes available substantive evidence rather than asserting exhaustive coverage of all publications through that date. Its analytical contribution is the integration of different levels and outcome measurements within a common explanatory argument. The argument develops established organizational and behavioral mechanisms; it does not propose a new named theory or claim priority for the mechanisms.
Culture, Climate, and Enacted Priorities
Members learn organizational priorities through sequences of interaction. They observe which decisions receive approval, which failures attract scrutiny, which employees advance, what information reaches senior managers, and what happens when revenue conflicts with another duty. A written value statement is one piece of such evidence. The entire observed pattern may support a different interpretation from the statement.
A financial institution may formally endorse customer welfare, prudent risk taking, and compliance, while its local evaluation routines make short-term revenue the dominant determinant of status. Employees who repeatedly observe how competing objectives are resolved can develop shared expectations without ever receiving an explicit instruction to violate a rule. The resulting climate is practically consequential because it supplies information about the social and material consequences of alternative actions. Culture gives that information a durable interpretation: employees learn what counts as success and what kinds of accommodation are treated as ordinary work.
The organization is also internally differentiated. Business lines, offices, trading desks, seniority groups, and support functions can develop different patterns of expectation. A corporate mean therefore conceals potentially consequential local variation. Schneider and colleagues’ (2013) organizational research synthesis provides the basis for attending to levels of analysis and the strength of shared climate. In a unit with high agreement around problematic expectations, members may perceive social pressure with considerable clarity; agreement around protective expectations can support the opposite pattern.
A strong culture has no intrinsic ethical direction. Its consequences depend on the content of the shared expectations, the duties relevant to the activity, and the allocation of authority. Researchers must measure both consensus and the nature of the practices on which consensus exists. A single employee’s perception is informative about that employee’s experience, but evidence of a shared climate requires a defensible aggregation procedure and an account of within-unit agreement.
Financial Work and the Conditions of Misconduct
Financial work involves decisions made for or affecting others whose information about the process is often incomplete. Customers rely upon specialist explanations. Managers delegate decisions. Supervisors depend upon records generated by the institutions they oversee. The separation of authority, knowledge, and consequence makes the quality of organizational controls especially important.
The operational consequences of an act may become visible at different times. A sale can appear successful before subsequent repayment, surrender, complaint, reversal, or investigation reveals its effects. A trade may produce an immediate result while imposing risks that become apparent later. Recordkeeping and classification may mediate what the institution can see. When performance recognition is rapid and accountability is delayed, decision makers can experience conflicting incentives even where the written policy is clear.
A useful analytical distinction separates opportunity from authorization. Opportunity describes the technical and organizational capacity to perform an act; authorization describes whether the act is permitted under the applicable duties and delegated powers. Culture shapes the interpretation of ambiguous opportunities, while controls restrict capacities, establish records, and assign authorities. The persistence of misconduct ordinarily calls for an examination of both. An explanation based solely on personal dishonesty omits the organization’s role in creating or closing opportunities. An explanation based solely on pressure omits the individual action and potential responsibility.
International financial arrangements can extend these distances across entities, service providers, and jurisdictions. Different contractual roles and regulatory frameworks may distribute oversight unevenly. These structural features motivate inquiries into information flows and accountability; they do not establish an empirical relationship between cross-border structures and misconduct. Researchers must investigate the actual institutional arrangement.
Ethical Culture and the Quality of Organizational Signals
Kaptein’s (2011) ethical-culture study examined 341 triads consisting of managers and direct reports. The reported results associated six dimensions of ethical culture with less observed unethical behavior: congruence of management and supervisors, feasibility of ethical conduct, supportability, discussability, and sanctionability. Clarity and transparency were not statistically significant in the reported analysis. This result informs interpretation of culture as a configuration of practices rather than a unitary favorable or unfavorable attitude.
A policy can be clear while compliance is practically difficult. A manager can recognize ethical requirements while allocating inadequate time or resources to meet them. A reporting channel can exist while an employee lacks confidence that raising a concern will cause meaningful action. Observable reinforcement is particularly consequential when employees compare formal expectations with actual career and disciplinary consequences.
The Kaptein study concerns general organizational unethical behavior. Its measured associations support the relevance of these dimensions, but they do not isolate their causal effects in financial organizations. A researcher should also consider common determinants: competent management may improve both perceptions of ethical culture and conduct, and staff who observe wrongdoing may evaluate their workplace differently. Measurement of climate before a subsequent independently assessed outcome would improve temporal ordering.
For financial institutions, the practical implication is to investigate concrete decision situations. Whether a disputed sale can be paused; whether an adviser has realistic access to independent suitability review; whether a trader can challenge supervisory instructions; and whether customer corrections affect performance evaluation are empirically assessable questions. These practices reveal how ethical expectations enter the actual choice environment.
Incentives, Goal Pressure, and the Interpretation of Success
Financial organizations need performance measures, and revenue is often appropriate within a broader assessment of sustainable operations. The behavioral problem arises when the connection between a measured output and the substantive purpose of the activity weakens. A measure may retain numerical precision while losing its ability to distinguish valuable service from harmful activity.
In a system that rewards gross account openings, for example, the relevant distinction is between an authorized, useful account and an account counted without genuine customer authorization. If management can observe only the aggregate count, the indicator conflates these outcomes. If local employees observe that scrutiny is directed predominantly at missed targets, they can infer the relative practical priority of competing objectives. These are analytical consequences of the measurement design, rather than claims that targets invariably cause misconduct.
The Wells Fargo board’s (2017) sales practices investigation explicitly documents sales pressure and organizational failures in oversight. That particular history supports the importance of the mechanism as a research question. It does not identify the marginal causal effect of any single compensation formula independently of authority, staffing, controls, or leadership.
A well-designed empirical investigation would examine changes in targets, compensation, evaluations, and customer outcomes together. A discontinuity in an incentive schedule could provide useful variation if workers cannot freely manipulate assignment and if other conditions are continuous around the threshold. A staggered rollout could permit comparisons across units if adoption is plausibly independent of unobserved outcome trends or if those trends are credibly addressed. A comparison of annual targets with detected misconduct alone has weak explanatory power because enforcement may concentrate on high-pressure areas.
The substantive object of performance management is the quality of the financial activity, not the production of attractive metrics. Evaluating employee conduct requires information about both results and the process by which they were achieved.
Occupational Identity and Experimental Dishonesty
Cohn, Fehr, and Maréchal (2014) investigated whether making banking identity salient affected dishonest reporting in a task with privately observed random outcomes. Their bank identity experiment reported an increase in aggregate dishonest reporting following the occupational identity manipulation among employees of the bank studied. The authors also reported comparison evidence from other groups.
The design supplies a valuable distinction between a general population characteristic and a contextual response. Random allocation to an identity manipulation can identify the task-specific influence of the manipulation within the study population, subject to implementation and measurement assumptions. It does not identify the causal effect of choosing banking as an occupation, working for a particular bank over a career, or being exposed to an institution’s incentive system.
The outcome is aggregate statistical evidence of dishonest reporting under the task’s conditions. Private realizations cannot be used to identify which particular participant reported dishonestly. Moreover, the behavioral outcome differs from an unauthorized account, manipulation of a market, or breach of a contractual obligation. Extending the result to these activities requires additional evidence.
Consider the alternative interpretations of the prime. Participants may retrieve experiences of workplace norms, expectations of competition, stereotypes about their profession, or associations elicited by the experimental questions. An experimental effect supports a role for the manipulation; separating these psychological mediators requires a design that measures or separately manipulates them. The distinction is essential when an experiment becomes the basis for public claims about the ethical character of an occupation.
Heterogeneity and the Limits of Occupational Generalization
Rahwan, Yoeli, and Fasolo’s (2019) banker-culture investigation examined 1,282 participants in five populations across three continents and reported no significant overall effect of banking identity salience on dishonest reporting. The discrepancy with the earlier study is scientifically informative. It establishes that the earlier response should not be treated as a universal feature of banking employment.
The later study does not prove that every bank and occupational group would show the same response, or that identity salience can never influence behavior. Statistical nonsignificance is compatible with several possibilities, including genuinely small effects, heterogeneous effects, and uncertainty associated with particular samples. Institutional access and participation also delimit the populations represented. The available sources do not support a common pooled effect estimate in this dossier.
The two findings jointly redirect the inquiry toward conditions. Researchers should examine which identities are being activated, which competing norms are present, whether the experimental task is similarly understood, and how organizational context varies across samples. A study with prespecified subgroup comparisons and adequate cluster-level sampling could explore such heterogeneity without treating a post hoc difference as an established mechanism.
The inferential consequence is substantial for global financial research. Labels such as banking, private wealth management, or offshore services encompass approaches to work that differ across organizations, functions, and local settings. An identity-prime experiment in one setting provides a reason to investigate such differences, not a license to assign moral characteristics to people in another setting. Occupational classification is a sampling descriptor. Behavioral evidence must establish any further claim.
Coworker Exposure and the Diffusion of Conduct
An employee’s peers supply examples, practical information, social approval, and assessments of organizational tolerance. Coworkers can influence conduct directly through joint work, indirectly through expectations, and through the transmission of descriptions of previous events. These channels create a plausible relationship between local workplace composition and behavior.
Dimmock, Gerken, and Graham’s (2018) financial-adviser coworker study exploits changes in coworkers associated with mergers. The reported finding is that advisers exposed to coworkers with past misconduct become more likely to engage in recorded misconduct. Comparisons within merging firms across branches strengthen the authors’ case for peer influence relative to a simple comparison of firms.
The quasi-experimental logic has particular importance. Coworkers ordinarily sort into workplaces, so a positive association can arise even without interpersonal influence. Merger-related changes make some variation less dependent upon an adviser’s own workplace choice. A credible interpretation still requires that differences across branches are not driven by other merger-related changes that influence conduct, including customer mix, supervision, reporting, staffing, and business strategy. The abstract supports the authors’ interpretation of the design but cannot establish that every relevant identifying condition is satisfied.
Peer exposure also has multiple potential meanings. A coworker with prior recorded misconduct may influence the practice of colleagues, share a common supervisory environment with them, or be assigned to a branch where detection subsequently changes. The relevant empirical distinction separates the history of the exposed coworker from the behavior of the focal adviser and from the surveillance process used to record each.
The finding supplies direct financial-sector support for an organizational influence on misconduct. It does not show that exposure makes misconduct inevitable, or that every professional with a disciplinary history poses the same risk. The appropriate unit of policy is a specific combination of role, authority, documented history, supervision, and behavior.
Recruitment, Sorting, and Reemployment
Culture is reproduced through hiring, assignment, promotion, departure, and reemployment as well as through influence among existing employees. Firms can attract people with different experience and preferences, while workers choose among firms based upon conditions and opportunities. The resulting association between a firm and misconduct can reflect the composition of its workforce, the treatment of employees within it, or both.
Egan, Matvos, and Seru’s (2019) adviser misconduct research reports that 7% of advisers in its historical US data had misconduct records, that roughly one-third of advisers with such records were repeat offenders, and that about half lost their jobs following misconduct. Reemployment partly offset the disciplinary consequence, with some firms characterized by comparatively high recorded misconduct. These figures describe the authors’ historical sample and definitions; they are not contemporary prevalence estimates for finance globally.
The finding links labor-market accountability to organizational culture. Separation from one firm may remove an employee from that firm’s decision environment without removing the risk from the market. Conversely, excluding individuals solely on a crude historical flag can confuse allegation and finding, ignore differences in severity, and disregard evidence of remediation or elapsed time. A defensible hiring decision must comply with applicable rules and distinguish the content, disposition, and relevance of the record.
The interplay between sorting and influence creates an important empirical problem. A firm may develop a high recorded rate because of recruitment patterns, because its work organization increases the probability of misconduct, because its detection is stronger, or because of several processes together. The researcher should represent each channel explicitly rather than assigning unexplained firm effects to culture. Job transitions, exposure histories, and timing of record creation can improve such an investigation.
Learning Beyond the Firm and the Ecology of Norms
Organizational members have prior social histories. Childhood environments, professional education, earlier workplaces, and residential contexts can contribute to the expectations they bring to a financial organization. The research challenge is to separate contextual exposure, individual selection, and the interpretation of the measured proxy.
Clifford, Ellis, and Gerken (2025) report in their childhood exposure research a relationship between childhood exposure to local misbehavior and subsequent misconduct among financial advisers. The available advance-article abstract describes a stronger relationship for childhood than adulthood exposure. The finding motivates the investigation of durable socialization and its interaction with workplace context.
Pham and colleagues’ (2024) local corruption culture study examines 4,740,760 adviser-year observations of 692,820 advisers during 1999–2021. The authors use local political-corruption convictions as a measure of contextual corruption and report an association with financial-adviser misconduct. Such a proxy deserves close interpretation. Convictions depend upon conduct, investigation, prosecution, and the legal process; they can also reflect enforcement activity. The measure is therefore not an observation of the attitudes of each adviser or resident.
These studies expand the explanatory environment beyond the firm. They cannot establish a moral classification of communities, countries, childhood backgrounds, or nationalities. Individual inference from an aggregate geographical measure would constitute an ecological error. A local association can coexist with broad intra-local variation and with substantial differences in the institutional conditions affecting reporting.
The practical implication is to study prior experience and present organizational treatment using lawful, decision-relevant measures. Geographic proxies are potentially useful for research subject to serious identification and ethical limitations. They are inappropriate substitutes for evaluation of individual professional conduct. Organizational interventions should address the actual social practices under the organization’s influence.
Authority, Silence, and the Cost of Disagreement
Organizations distribute the right to make decisions and the capacity to challenge them. A frontline employee may possess direct knowledge of a problematic transaction without having authority to change the policy that generated it. A manager may receive evidence of a concern but have career incentives tied to maintaining the appearance of success. An independent control function may possess formal authority while lacking timely access to the relevant information.
Morrison’s (2023) employee voice and silence review establishes the broader research domain concerned with the expression and withholding of work-related concerns. Edmondson’s (1999) team psychological safety study reports an association between psychological safety and learning behavior in 51 manufacturing teams. Transferring these general findings to financial institutions motivates a specific hypothesis: employees are more likely to transmit troublesome information when expressing it is perceived as safe and practically worthwhile. It remains necessary to test how that expectation relates to validated conduct outcomes in the financial setting.
Silence has multiple interpretations. An absence of complaints can reflect safe and compliant activity, lack of knowledge, ambiguity of duties, limited opportunities to speak, or expectations that a concern will be ignored or punished. A high complaint count can reflect a deteriorating condition or an improving ability to report it. The count alone cannot distinguish those states.
Information flow also depends upon organizational language. Reclassifying repeated customer injuries as isolated operational exceptions can affect how patterns become visible to decision makers. Different functions may apply different thresholds to the same underlying event. The relevant governance issue is whether evidence reaches a decision-making body capable of evaluating its significance and acting upon it.
Psychological safety and accountability are compatible. Staff should be able to acknowledge difficulties, ask questions, and report concerns, while substantively justified action follows investigation of established wrongdoing. The distinction between reporting, adjudication, and sanction protects both institutional learning and procedural fairness.
Organizational Learning, Repetition, and Normalization
Repeated deviations change the environment in which later decisions are made. If a problematic act produces an attractive measured result and no meaningful correction, coworkers can acquire information about the organization’s practical tolerance. If a report triggers effective investigation and remediation, a different expectation becomes available to observers.
Normalization in this dossier describes the process through which a repeated practice becomes familiar, accepted, and incorporated into expectations. Familiarity has several possible behavioral effects. It can reduce the perceived distinctiveness of a questionable practice; it can supply social evidence that others participate; and it can alter beliefs about what management regards as permissible. These are analytical possibilities consistent with organizational learning, not measured effect sizes.
Institutional memory is central to the process. Recording and examining recurring failures across organizational boundaries allows an enterprise to identify a repeated pattern. If reports remain separated by office, product, employee, or classification, decision makers may observe many discrete exceptions without recognizing their common cause. Staff movement and leadership succession can further fragment such evidence.
The inverse process is also possible. Persistent attention to validated concerns, transparent correction of defective processes, and credible protection for legitimate disagreement can change expectations. Dürst and Kunz (2026) document the difficulties involved in embedding risk culture in a financial institution through participatory action research. Their process evidence suggests the importance of implementation and perception: the existence of a reporting or disciplinary arrangement may not guarantee that employees experience it as effective.
Organizational learning requires the transformation of individual incidents into reliable knowledge and organizational action. Assessing this transformation is a research undertaking distinct from counting training completions or recording the adoption of policies.
Retail Banking: Wells Fargo Sales Practices
The Wells Fargo case provides documented evidence of the interaction among performance expectations, supervision, and information flows. In September 2016 the Office of the Comptroller of the Currency issued an enforcement announcement imposing a $35 million civil money penalty and ordering restitution in connection with unsafe or unsound sales practices. The action concerned unauthorized account activity and associated failures in risk management.
The independent directors’ 2017 investigation report describes significant sales pressure, a decentralized management structure, resistance to scrutiny, and inadequacies in the transmission of information. It documents the investigation’s assessment of organizational factors and identifies limitations in the information available to senior decision makers. The report was commissioned by the company’s board and is retrospective institutional evidence. Its interpretation must be distinguished from judicial findings and from an independently controlled study.
In February 2020 the Department of Justice announced a criminal and civil resolution totaling $3 billion in connection with practices involving millions of accounts without customer authorization. The announcement describes admitted conduct spanning 2002 to 2016 and a deferred prosecution agreement. Corporate admissions and the resolution establish facts at that level; they do not adjudicate each employee’s conduct or mental state.
This sequence provides three different forms of evidence: agency action addressing unsafe practices, an institutional investigation explaining management failures, and a subsequent resolution with admissions. Their convergence supports a detailed account of the institutional problem. It cannot establish how many particular employees acted from a specific personal motive. Nor does it quantify the relative contribution of compensation, supervisory treatment, staffing, and other conditions.
The psychological importance of the case lies in the connection between visible success and the conditions of its production. A performance system can continue to generate favorable information after the substantive validity of some activities has deteriorated. An organization’s ability to recognize that discrepancy depends upon the independence and reach of its corrective information systems.
Wholesale Markets: Foreign Exchange Enforcement
The setting of wholesale foreign exchange trading differs from retail sales. Transactions concern market prices and counterparties, specialized teams possess sensitive information, and the social environment can cross formal institutional boundaries through professional interaction. Relevant duties depend upon roles and applicable rules.
The Federal Reserve’s May 2015 foreign exchange enforcement announcement described penalties exceeding $1.8 billion against six banking organizations for unsafe and unsound practices. The announcement distinguished institutional findings, including inappropriate sharing of information and failures to identify or address illegal agreements among traders in the institutions specified. It did not establish identical conduct or criminal responsibility for every organization and employee named.
The case illustrates how professional interaction and inadequately effective supervision can extend the relevant psychological environment beyond the formal organization. The peer group that informs a trader’s practical expectations may include market participants employed elsewhere. An institution’s internal values may therefore coexist with external professional patterns and incentives.
The research implication is that a study limited to the individual employing institution may miss important exposures. Empirical measurement can include legally accessible communications evidence, joint work patterns, role responsibilities, and supervisory records. Privacy, labor rights, data-protection obligations, and relevance constrain such measurement. The existence of communication ties is not evidence of improper coordination. The content and context of the observed activity matter.
Comparing wholesale and retail cases identifies a shared organizational problem without collapsing their legal distinctions: the practical conditions under which employees pursue success can detach from the substantive duties governing their particular activity. The nature of the activity determines what constitutes a breach and which protective mechanisms are appropriate.
Prudential Inquiry: Commonwealth Bank of Australia
The Australian Prudential Regulation Authority released its CBA prudential inquiry summary in May 2018 and accepted an enforceable undertaking. The inquiry identified governance, accountability, and nonfinancial risk-management weaknesses, with a particular concern that sustained financial success had contributed to complacency and insufficient challenge.
The inquiry is valuable because of its focus on the governance conditions surrounding risk, including operational, compliance, and conduct risk. It illustrates the analytical difference between financial performance and organizational capacity to identify weaknesses. Favorable financial outcomes can coexist with deficiencies in the management of nonfinancial risks.
The evidence has a distinct status from a finding of financial misconduct. A failure of governance or risk management does not itself establish that a particular financial duty was violated by a particular person. The inquiry informs the study of organizational receptiveness to concerns and of systemic conditions under which failures may persist. Each claim of actual misconduct would require its own evidentiary foundation.
The cross-case interpretation is necessarily conditional. The Wells Fargo documents identify sales-related practices and failures; the foreign exchange enforcement announcement addresses wholesale-market supervision; the Australian inquiry examines broader nonfinancial risk governance. Together these documents establish that the quality of internal challenge, accountability, and upward information can be central to financial institutions’ ability to recognize and control problems. They do not provide a single causal effect estimate across settings.
Risk Culture and Prudential Outcomes
The concept of risk culture connects collective expectations to the ways in which organizational members recognize, discuss, take, and manage risk. The Financial Stability Board’s (2014) risk culture guidance treats the object as a matter for supervisory assessment and monitoring. Its guidance emphasizes the practical qualities of a financial institution’s environment rather than prescribing one ideal culture for all organizations.
Suss, Bholat, Gillespie, and Reader (2021) developed quantitative measures from 20 indicators across six data sources in their Bank of England working paper on culture and bank risk. The paper reports a relationship between organizational culture and bank risk. This provides a useful measurement direction because administrative and external evidence can augment employee survey responses.
The distinction between prudential risk and misconduct remains fundamental. A financial institution can take significant legitimate risks with appropriate authorization and oversight. Conversely, misconduct can cause serious customer injury without immediately threatening the institution’s solvency. An index that combines fraud records, complaints, risk indicators, and other information can be useful for screening, but it may partly overlap with the outcome used to assess its predictive validity.
A research design should define separate variables for enacted culture, recorded conduct outcomes, and prudential outcomes. Where the same source contributes to more than one variable, it should document the overlap and test whether the relationship persists after that component is excluded. This guards against the mistaken presentation of a partly definitional association as independent empirical discovery.
Risk culture is relevant to misconduct insofar as it shapes how decisions are challenged, recorded, escalated, and corrected. The relevant strength of that relationship is an empirical question that varies with the activity and its oversight.
Recorded Misconduct and the Detection Process
Recorded misconduct is a jointly produced outcome. First an underlying event occurs. Then someone or some system detects it. The event is classified, investigated, and recorded according to applicable procedures. At later stages, it may result in correction, a regulatory finding, adjudication, or another disposition. Changes at any stage can change the final count.
An explicit representation clarifies the problem. Let an underlying event indicator be U and detection, classification, and recording indicators be D, C, and R. In a simplified setting with perfect specificity, the probability of a record is the probability of U multiplied by the conditional probability of D given U, then C given U and D, then R given the preceding stages. In practice, mistaken reports and disputed classifications require additional terms. The representation is an application of elementary probability, not an independent theory.
Suppose an organization improves surveillance and facilitates reporting. Its underlying event rate might remain constant or decline while recorded cases rise because the conditional probability of detection increases. A decline in recorded misconduct could in turn reflect fewer events or weaker detection. Evaluation must therefore observe both conduct and the processes through which conduct becomes visible.
This distinction applies directly to archival research on financial advisers. A historical misconduct disclosure records an administrative or professional event under specified reporting arrangements. The empirical outcome should be described using the study’s definition instead of reinterpreted as an unqualified measure of moral disposition. The same principle applies to counts of terminations, complaints, sanctions, and regulatory actions.
Researchers should also distinguish event time from discovery and disposition time. An intervention adopted after an incident but before its discovery can appear associated with an incident it could not have prevented. Analysts can use event histories and sensitivity analyses based on alternative assignments of uncertain dates to reduce this problem.
Measuring Culture at the Correct Level
A measurement instrument should identify the construct, the referent, the unit, and the observation period. An item asking whether senior leaders value ethical conduct differs from one asking whether the respondent’s immediate manager supports challenge. A behavioral indicator of completed customer remediation differs from both. Combining them without explicit justification obscures what is being measured.
Survey design can distinguish beliefs about expectations, observed conduct, actual experiences, and intended responses to hypothetical situations. Where respondents are nested in work groups, analyses should estimate the distribution within as well as between groups. Aggregation requires evidence that the items have a shared referent and that group-level comparisons are meaningful. Agreement is itself information, and low consensus should not be concealed by a precise group mean.
The approach follows the level distinctions in Schneider and colleagues’ (2013) culture and climate review. Independent data sources can further reduce dependence on a single perception measure, as illustrated by Suss and colleagues’ (2021) multiple-indicator approach. The analyst must nevertheless verify that an apparently independent measure is not largely generated by the same underlying incident or reporting process.
Comparisons over time and across languages require attention to measurement invariance. An item about whether formal challenge is encouraged can have different practical implications across organizations with different authority structures. Translation alone does not establish equivalence. Qualitative cognitive interviewing and tests of the measurement structure can aid interpretation where the design and sample support them.
Employee survey responses should be collected under credible privacy arrangements. Managers should not receive information that allows the identification of individual respondents from small-cell reports. A research environment that makes participation risky can distort the measured construct by encouraging socially acceptable answers.
Causal Inference and Competing Explanations
The causal question must specify the exposure and contrast. The effect of making an occupational identity salient in a laboratory task is different from the effect of moving to a team with different peer histories; both differ from the effect of changing compensation, introducing a reporting channel, or replacing a manager. The term culture cannot replace specification of those exposures.
Randomized experiments can identify the effect of a defined treatment within the sample when allocation and outcome measurement are valid. The experiments of Cohn and colleagues (2014) and Rahwan and colleagues (2019) are appropriately read as tests of identity salience under particular conditions: their initial experimental result and later heterogeneous evidence make generalization an empirical question.
Quasi-experimental studies can address effects of changes that would be difficult or inappropriate to manipulate experimentally. Dimmock and colleagues’ (2018) merger-based research design is important in this respect. The relevant assumptions concern how workplace changes arise and whether other changes explain the comparison. Firm and time controls do not automatically solve time-varying differences in surveillance, customers, or managerial behavior.
A useful causal diagram for planning would distinguish prior worker characteristics, prior workplace assignment, current exposure, incentive and control conditions, underlying acts, and detection. It would also distinguish a potential mediator such as perceived tolerance from a confounder such as managerial competence. Conditioning on an outcome-generated complaint measure can create bias if the measure is affected by both misconduct and reporting behavior.
The design should prespecify the estimand, population, relevant time horizon, and handling of missing observations and job transitions. Analysts should avoid implicitly estimating one effect while describing another. If the outcome is recorded misconduct, the estimand concerns records under the study’s measurement conditions unless the design separately identifies underlying acts. This precision is indispensable for practical intervention decisions.
Comparative Research Across Financial Institutions
Financial institutions differ in their products, distribution channels, customer relationships, authority structures, and applicable duties. A comparison requires common conceptual definitions and appropriate adaptation of the observable measures. Comparable concepts do not require identical institutional procedures.
Consider retail sales, investment advice, and institutional trading. The first can involve a substantial volume of standardized customer interactions; the second often involves recommendations where customers depend on specialized expertise; the third involves market conduct, execution, information flows, and counterparties. Each setting offers distinct ways to observe the connection between output measures and substantive duties. Each also requires a definition of the underlying event grounded in the relevant activity.
A multi-institution study should sample sufficient organizational units rather than relying on a large number of employees in a small number of firms. The effective sample for a firm-level exposure depends upon variation at the firm level. A design with thousands of employee observations but only two firms cannot reliably separate a general firm-culture effect from all other differences between those firms.
Longitudinal observation helps distinguish enduring patterns from short-lived reactions to an enforcement event or leadership change. Mixed methods can connect survey scores to concrete situations and independently verified decision records. Qualitative work can reveal the meaning of particular practices and assist in the construction of quantitative variables. Inferences about frequencies and causal magnitude require designs suited to those questions.
Interinstitutional data sharing should use authorized and appropriately minimized records. Any cross-border comparison must account for differences in reporting duties, enforcement intensity, legal categories, and the availability of public records. A ranking based on raw recorded cases would otherwise conflate the underlying activity with its visibility.
International Finance and Offshore Organizational Contexts
The application of this research to the Global Offshore System concerns the organizational psychology of financial services and decision-making across jurisdictions. International financial arrangements can have legitimate economic, administrative, investment, or risk-management purposes. Their structure may also change where knowledge, delegated authority, and oversight reside. The psychological research object is the effect of these concrete organizational conditions on decision behavior and accountability.
A cross-border service chain may involve a client-facing adviser, a specialist administrator, independent fiduciaries, financial institutions, and compliance personnel. The formal allocation of duties can differ from the operational allocation of information. An employee who observes one part of the transaction may not possess the information needed to evaluate another. Where a concern depends on evidence distributed among multiple organizations, the reliability of lawful information transfer becomes especially consequential.
These observations motivate empirical investigation. They do not establish that offshore finance has a general misconduct rate, or that its organizational participants exhibit a distinctive ethical culture. The verified studies examined here concern particular banking populations, financial advisers, general organizational research, and documented institutional episodes. No direct comparative estimate for offshore versus onshore organizational cultures was verified in this source set.
A scientifically defensible application would compare matched activities subject to explicit institutional and legal definitions, measure authority and challenge at the level where work occurs, and separate the detection process from event occurrence. Researchers should also document differences in client type, business function, firm size, staffing, and supervisory access. Jurisdiction should enter as a specified institutional context rather than as a presumed moral characteristic.
The contribution of psychology is to explain how people acquire and act upon expectations within organized financial activity. Economic and legal research can then clarify the incentives, duties, and consequences that constitute the environment in which those expectations are formed.
Governance, Accountability, and Supervisory Guidance
The Financial Stability Board’s (2018) misconduct risk toolkit presents 19 tools directed toward cultural drivers, individual accountability, and the movement of employees with histories of misconduct. The toolkit is supervisory and governance guidance. It supplies a structured set of policy options; it is not an experimental evaluation of their effectiveness.
Accountability requires a sufficiently clear relationship between authority, information, duty, and decision. A person who is assigned responsibility must have access to the information and resources necessary to discharge it. An organization must be able to establish who made, authorized, reviewed, or had a duty to review significant decisions. Documented roles support investigation and corrective action, provided records are accurate and appropriate procedural safeguards apply.
At the psychological level, governance changes the information available to organizational members about expected conduct and the consequences of choices. Repeated evidence that significant concerns reach independent decision makers can affect confidence in reporting. Repeated evidence of superficial investigations can affect confidence in the opposite direction. These are plausible behavioral pathways that require measurement if an organization asserts that a particular reform has changed culture.
Institutional accountability also extends to the conditions under which workers act. Investigations should ask whether procedures, staffing, incentives, or supervisory instructions contributed to the problem. Individual responsibility and organizational responsibility can coexist. Establishing one does not exhaust the other.
The form of accountability should vary with the decision, the applicable duties, and the evidence. A standardized accountability label without substantive investigation may assign responsibility inaccurately; excessive concentration of responsibility at low organizational levels may obscure the role of policy and oversight. Effective review requires access to evidence at each relevant level.
Intervention Design and Behavioral Implementation
An intervention should begin with a specified problem and a plausible causal pathway. If the identified problem is distorted sales measurement, the intervention should address the substantive validity of the measured output and the incentives tied to it. If control personnel cannot obtain timely information, training alone cannot remove the information barrier. If employees expect retaliation for raising concerns, the existence of a written reporting policy is insufficient evidence that the barrier has changed.
Dürst and Kunz’s (2026) financial institution action research documents a participatory effort to embed risk culture and the practical difficulties encountered in implementing it. The study provides process knowledge about the translation of intentions into organizational practices. Without an independently identified counterfactual, it cannot establish the size of any reduction in misconduct attributable to the intervention.
An institution could evaluate a change through a carefully designed implementation sequence. Baseline measures would distinguish decision quality, verified incidents, reporting accessibility, employee perceptions, and actual investigation outcomes. The organization would document how the change is implemented at the level of individual units and whether those units receive comparable resources. Subsequent outcomes would be compared against an appropriate reference where feasible.
A randomized or staggered introduction may be ethical for changes in training or communication that can lawfully be phased. Mandatory protective requirements and immediate correction of known harmful practices cannot be withheld merely to create a research contrast. Where random assignment is inappropriate, interrupted time series, comparison units, or independently specified process evaluation can contribute evidence with their respective limitations.
Interventions should preserve internal challenge, procedural fairness, and customer protection. Success is demonstrated through the substantive conduct and decision outcomes the intervention was intended to improve, alongside evidence about its implementation and detection effects. Policy adoption and employee attendance establish exposure to an initiative, not its effectiveness.
Reporting, Remediation, and the Interpretation of Improvement
A reporting system performs several functions. It supplies institutions with otherwise inaccessible information, supports investigation, enables corrective action, and provides evidence about the performance of controls. It also has psychological consequences because employees form expectations from observed responses to reported concerns.
Evaluation should distinguish receipt, triage, investigation, disposition, and remediation. Delays and abandonment can arise at different stages. A single closure count cannot show whether a case was appropriately resolved. Independent review of representative samples, subject to confidentiality and procedural protections, can improve interpretation of administrative records.
Customer remediation provides another source of evidence. The restoration of an affected customer’s position and the correction of the process that generated the problem are distinct goals. A system may complete individual remediation while leaving the originating organizational condition unchanged. The recurrence of a validated problem under comparable activity and detection conditions is relevant to assessing the adequacy of the response.
Confidentiality can create a difficult balance. Individual disciplinary matters may properly remain confidential, while employees need credible information that established violations receive appropriate attention. Organizations can communicate anonymized process-level information and procedural expectations without disclosing protected personal details. The action research by Dürst and Kunz (2026) on risk culture implementation is pertinent to the practical difficulty of making formal processes perceptible.
Improvement consequently requires more than a favorable direction in a single indicator. A rise in early reporting accompanied by more effective correction and a subsequent decline in independently assessed harms can be consistent with improved detection and prevention. The pattern is more informative than an isolated falling or rising count, while its causal interpretation still depends on the study design.
Analytical Synthesis: The Relationship Between Culture and Conduct
The reviewed evidence supports a differentiated account of organizational influence. Culture matters through the ways in which organizational members interpret available choices, anticipated consequences, and substantive obligations. It is reproduced through interaction, selection, institutional memory, and the repeated observation of decisions. Its observable effects depend upon the activity, the organization, and the measurement process.
The experimental literature supplies evidence about the conditional activation of occupational identity. Its contrasting findings rule out a simple inference from occupational membership to a universal response. The merger-based adviser research supplies a different form of evidence about exposure to coworkers, bringing the inquiry closer to recorded professional misconduct. The labor-market research shows that hiring and reemployment can contribute to the distribution and persistence of such records. Research on childhood and local context expands the temporal and geographical scope of inquiry while raising problems of ecological inference and selection.
The institutional cases demonstrate that financial misconduct and deficient risk governance can involve failures across different levels of authority and information. Their documented patterns support examining management practices, organizational challenge, and the quality of control information. Retrospective investigations and enforcement actions do not isolate the effects of individual cultural factors; their role is documentary and explanatory.
The synthesis yields a central analytical conclusion: culture should be evaluated through the specific mechanisms and decisions by which it becomes consequential. A psychological explanation is strongest when it specifies how relevant information and expectations change the practical assessment of alternatives and when the resulting behavior is measured independently of the explanation.
This conclusion also places limits on culture-based claims. An unexplained difference between firms does not establish a cultural cause. A recorded difference between regions does not establish a difference in individual integrity. An experiment with an occupational prime does not establish the effect of an entire occupational career. A supervisory recommendation does not establish an achieved outcome. These distinctions strengthen the study of financial misconduct by clarifying what each form of evidence can establish.
Research Agenda and Empirical Standards
Future research should first improve construct specification. Studies should compare experienced priorities, shared practices, perceived tolerance, and observed managerial decisions in sufficient detail to identify the exposure. A general index is useful for exploration, but a causal intervention requires a specific actionable condition.
A second priority is separating organizational effects from worker selection. Longitudinal studies of job mobility and exogenous or plausibly exogenous changes in peer environments can contribute to this task. They should also account for changes in reporting, supervision, case mix, and job role. A professional’s history should be represented with the appropriate distinctions among allegations, disposition, severity, and timing.
A third priority is measuring underlying conduct independently from its detection. Where ethically and legally possible, carefully designed audits of representative transactions can augment administrative counts. Researchers should distinguish discovery from event dates, specify sampling and classification procedures, and report uncertainty. Data on correction and remediation can complement but cannot replace assessment of the original activity.
A fourth priority is testing interventions. Changes in performance measures, management expectations, independent review, reporting, and response should be evaluated separately or within designs that make the combined treatment explicit. Implementation fidelity and possible unintended effects belong in the analysis. Research should examine customer outcomes as well as employee perceptions and organizational records.
Finally, international and offshore financial research needs genuinely comparative evidence. Comparable financial functions should be studied in their actual institutional settings with transparent definitions and measurement procedures. Interdisciplinary collaboration is particularly useful where the relevant legal duty, economic incentive, and psychological expectation are defined in different research traditions. The research agenda preserves the independence of those explanatory objects while connecting their empirical evidence.
Conclusion
Organizational culture influences financial misconduct through socially learned expectations, practical interpretations of authority and duty, coworker exposure, and the observed consequences of organizational decisions. The empirical literature supports context-dependent influence rather than a uniform ethical characteristic of financial occupations. Experimental evidence, archival studies, and documented institutional cases contribute different kinds of knowledge, and their explanatory strength depends upon the question asked.
Individual decisions take place within organizations that determine opportunities, access to information, performance recognition, and the conditions of challenge. Those conditions can facilitate responsible conduct or allow harmful practices to persist. The organization’s responsibility for its arrangements is compatible with the individual’s responsibility for established conduct. Accurate assignment of each requires evidence of the relevant act, duty, authority, and information.
Recorded misconduct is also a function of visibility, investigation, and classification. Culture research must therefore separate behavior from the processes through which behavior becomes known. An organization that improves reporting may initially reveal more problems, and an organization with few recorded cases may have weak detection. Evaluating prevention demands a substantive assessment of decisions, customer effects, reporting processes, and corrective action.
For the Global Offshore System, the findings establish a disciplined research direction: investigate the actual organizational conditions of international financial activity, distinguish lawful structures from proven violations, and connect psychological mechanisms to the legal and economic properties of specific financial functions. The source set does not establish a comparative misconduct rate for offshore finance. Such a conclusion requires its own empirical investigation.
The primary contribution of this dossier is the integration of evidence on organizational culture, occupational identity, peer influence, labor-market sorting, information flows, and accountability within a causally differentiated account of financial misconduct. Its practical conclusion follows directly: the quality of financial conduct is an organizational achievement sustained through substantive performance evaluation, credible challenge, accurate records, and effective correction, and its improvement must be demonstrated with evidence at the level where financial decisions occur.
References
- Australian Prudential Regulation Authority. (2018, May 1). APRA releases CBA Prudential Inquiry Final Report and accepts Enforceable Undertaking from CBA. Source record
- Board of Governors of the Federal Reserve System. (2015, May 20). Federal Reserve announces fines totaling more than $1.8 billion against six major banking organizations for their unsafe and unsound practices in the foreign exchange (FX) markets. Source record
- Clifford, C. P., Ellis, J. A., & Gerken, W. C. (2025). Childhood exposure to misbehavior and the culture of financial misconduct. Review of Financial Studies, advance article hhaf075. DOI: 10.1093/rfs/hhaf075. Source record
- Cohn, A., Fehr, E., & Maréchal, M. A. (2014). Business culture and dishonesty in the banking industry. Nature, 516, 86–89. DOI: 10.1038/nature13977. Source record
- Dimmock, S. G., Gerken, W. C., & Graham, N. P. (2018). Is fraud contagious? Coworker influence on misconduct by financial advisors. Journal of Finance, 73(3), 1417–1450. DOI: 10.1111/jofi.12613. Source record
- Dürst, N., & Kunz, J. (2026). Embedding risk culture in a financial institution: An action research perspective. Review of Managerial Science, 20, 2581–2616. DOI: 10.1007/s11846-025-00946-2. Source record
- Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. DOI: 10.2307/2666999. Source record
- Egan, M., Matvos, G., & Seru, A. (2019). The market for financial adviser misconduct. Journal of Political Economy, 127(1), 233–295. DOI: 10.1086/700735. Source record
- Financial Stability Board. (2014, April 7). Guidance on supervisory interaction with financial institutions on risk culture: A framework for assessing risk culture. Source record
- Financial Stability Board. (2018, April 20). Strengthening governance frameworks to mitigate misconduct risk: A toolkit for firms and supervisors. Source record
- Independent Directors of the Board of Wells Fargo & Company. (2017, April 10). Sales practices investigation report. Wells Fargo & Company; filed with the U.S. Securities and Exchange Commission. Source record
- Kaptein, M. (2011). Understanding unethical behavior by unraveling ethical culture. Human Relations, 64(6), 843–869. DOI: 10.1177/0018726710390536. Source record
- Morrison, E. W. (2023). Employee voice and silence: Taking stock a decade later. Annual Review of Organizational Psychology and Organizational Behavior, 10, 79–107. DOI: 10.1146/annurev-orgpsych-120920-054654. Source record
- Office of the Comptroller of the Currency. (2016, September 8). OCC assesses penalty against Wells Fargo, orders restitution for unsafe or unsound sales practices. News Release 2016-106. Source record
- Pham, M. H., Nguyen, H., Young, M., & Dao, A. (2024). Who keeps company with the wolf will learn to howl: Does local corruption culture affect financial adviser misconduct? Journal of Business Ethics, 194, 185–210. DOI: 10.1007/s10551-024-05618-x. Source record
- Rahwan, Z., Yoeli, E., & Fasolo, B. (2019). Heterogeneity in banker culture and its influence on dishonesty. Nature, 575, 345–349. DOI: 10.1038/s41586-019-1741-y. Source record
- Schneider, B., Ehrhart, M. G., & Macey, W. H. (2013). Organizational climate and culture. Annual Review of Psychology, 64, 361–388. DOI: 10.1146/annurev-psych-113011-143809. Source record
- Suss, J., Bholat, D., Gillespie, A., & Reader, T. (2021). Organisational culture and bank risk. Bank of England Staff Working Paper No. 912. Source record
- U.S. Department of Justice. (2020, February 21). Wells Fargo agrees to pay $3 billion to resolve criminal and civil investigations into sales practices involving the opening of millions of accounts without customer authorization. Source record