Experts in this field often cite ‘bad’ or ‘toxic’ cultures as the root cause of major prudential and conduct failings in financial services in recent years, from the financial crisis of 2008 to Libor rate-fixing, PPI mis selling, and rogue trading. However, while the link between poor organisational culture and bad banking outcomes is often discussed, there is surprisingly little research that has investigated this link empirically. This is predominantly due to measurement difficulties – identifying a bad banking culture before a crisis presents considerable challenges.
Culture is an amorphous. Culture shapes the way staff act, and don’t act, on a daily basis and it can be shaped itself by influential people inside and outside the organization. Getting culture right may not be a panacea to banks’ many ailments. However, an effective culture can serve as glue, it binds together elements such as governence, risk management,compliance,high levele systmes and controls, and makes it strong. The evidence of a cultural problem can be seen in the financial and reputational damage the industry has inflicted on itself. Excessive risk taking inflicted catastrophic loses on the industry. Aggressive selling leads to various regulatory and compliance limits.
Scholars and regulators typically rely on staff surveys and interviews with board members and senior executives to assess organisational culture. Research in the behavioural sciences suggests that these sources are generally unreliable. There are a number of reasons for this. First, surveys and interviews lend themselves to impression management, i.e. deliberate attempts by those under scrutiny to create a favorable perception that may be at odds with reality. Second, and less intentionally, employees embedded in particular organisations tend to take for granted their cultural context as ‘normal’ when, in fact, it may be anomalous and specific to that firm. This can make employees unreliable witnesses even if they are committed to revealing the truth about their firm’s culture. Indeed, because culture shapes how people respond to surveys, it is not clear whether responses are perceptions or products of culture. Relatedly, employees who engage in unethical or risky acts rarely consider themselves be doing so, with people justifying such acts (e.g., inflating results) as being helpful to an institution. Finally, staff surveys and interviews tend to be non-representative of the entire population of staff – important swathes of an organisation, e.g. senior executives or employees who are disengaged, usually refrain from self-reporting, potentially biasing results.
Given these problems with self-reported measures of organisational culture, a nascent literature has emerged which uses data gleaned unobtrusively. For example, recent studies have analysed online employee workplace reviews (Corritore et al., 2019; Moniz and Jong, 2014), internal email communications (Goldberg et al., 2016; Srivastava et al., 2018), annual report publications (Fiordelisi and Ricci, 2014; Gupta and Owusu, 2019; Nguyen et al., 2019), and other textual disclosures, e.g., company websites (Grennan, 2019a). Although these studies represent advances in the assessment of organisational culture, they are not without limitations. Organisational culture, being distributed and pervasive, is unlikely to be captured by a single isolated measure. For example, detailed insight on the values of specific units or types of behaviour typically studied through surveys may be difficult to detect with data derived from a single source (e.g., earning call transcripts, or employee online reviews which only capture a small sample of an organisation’s population). Instead, what is needed is a battery of measures probing different aspects and levels of the organisation.
The term ‘unobtrusive measure’ was coined by Webb et al. (1966) to describe the value of using non-reactive methodologies – where data is collected and analysed without engaging participants. The key benefit of unobtrusive measures is that, compared to methodologies such as surveys where “the processes involved in measurement affect the value obtained for the variable” (Sechrest and Phillips, 1979, p .3), issues such as the social desirability in responses and observer effects can be addressed (Webb et al., 1966). Drawing on this foundation, Reader et al. (2020) outlines the construct of an ‘unobtrusive indicator of culture’ (UIC) and conduct a systematic review of the academic literature that use unobtrusive measures. A UIC refers to a single measure of organisational culture based on data collected without engaging employees. Measuring culture in this way addresses the social psychological finding that, just as attitudes do not necessarily correspond to behaviour (Wicker, 1969), the values and norms espoused by institutional members may not necessarily correspond to practices (Hill et al., 2014). Through drawing on naturally occurring data, which is often collated anonymously from across an institution (e.g., employee online reviews), extemporaneous (e.g., executives responding to questions during an earnings call), and revealing of values (e.g., institutional reward systems), analyses of culture are rooted in instantiations of organisational values rather than assessments. This, intuitively, seems more useful for capturing data on practices that are associated with poor outcomes. For instance, in an organisation where there are conditions that create risk (e.g., acceptance of unethical conduct, poor workforce management), gaining access to undertake a comprehensive survey of employees may be challenging, and reporting biases will likely influence the data collected.
Some of the examples of UICs are: i) data quality metrics (DQMs) derived from regulatory return submissions; ii) diversity data from the Approved Persons database; iii) customer complaint reports; iv) whistleblowing referrals obtained authority intelligence; v) reports of internal fraud cases and costs; and vi) balance sheet and capital requirement information.
Culture measurement in financial services presumably has similar confounds, for instance in organisations with conduct problems, and thus may be particularly useful for detecting problematic bank cultures. As a result, UICs may have practical implications for bank regulators, among other stakeholders and market observers. Supervisors are often interested in monitoring organisational culture as a barometer of their safety and soundness, and so UICs are likely to be promising.
Policymakers, financial regulators, and market observers have, for some time, believed organisational culture to be an important factor driving prudential outcomes. The challenge has been how to measure culture effectively. Traditionally, information on bank culture has been gathered from employee self-reports. However, these are known to have limitations, as noted earlier. UIC have helped to marry up a unique conceptual approach with multiple data sources to measure organisational culture unobtrusively. Rather than interrogate bank employees, asking them to ‘show us’ their bank’s culture, this method instead interrogated data that provides insight on different dimensions of culture, from the quality and lateness of their regulatory reporting, to the volume and handling of customer complaints. Importantly, the sources that have been examined in this method are more directly related to their respective cultural dimension than is typical in prior studies using unobtrusive data. For our jurisdiction if we can get such data, it can be easily found out that poor culture leads to greater bank risk as hypothesized.
Contributed by –
Dr.Chiragra Chakrabarty
CEO-KATIC CONSULTING LTD
*The views expressed are personal.
