Culture in financial institutions: it’s everywhere and nowhere

Speaker: Andrew Bailey[1], Chief Executive
Event: HKMA Annual Conference for Independent Non-Executive Directors, Hong Kong
Delivered: 16 March 2017
Note: this is the speech as drafted and may differ from the delivered version

Highlights

• A firm’s culture emerges in large part from inputs that are its responsibility.
• It is for firms to ensure that their desired culture is consistent with appropriate conduct outcomes, to identify the drivers of behaviour within the firm and control the risks that these drivers create.
• Culture change in itself is a challenge, and we know it takes time.

It is a great pleasure to be in Hong Kong and to have the opportunity to participate in this important discussion of a subject in which we all – financial institutions and regulators – have a strong interest. Thank you to the Hong Kong Monetary Authority for organising this opportunity to compare experiences.

The title of my speech today – ‘Culture in financial institutions: it’s everywhere and nowhere’ – is intended to recognise the somewhat elusive nature of the subject. I am going to start by defining what I mean by culture. Then, I will set out what I see as the respective roles of firms’ governing bodies and management and public authorities. Last, I will use two case studies – around remuneration and governance – to illustrate the changes that we are seeing. I will conclude by describing our role as regulators. Underlining that culture, well-defined, is at the heart of how we assess firms and look for improvements where these are needed.

What is culture in financial institutions?

There is an active, sometimes passionate, debate on the meaning of culture. It lies behind my choice of ‘everywhere and nowhere’. One form of this debate is to ask whether culture is an input to institutional behaviour, or is it a summary outcome? This is relevant because to my mind the former position – the input – requires the identification of a distinctive thing called culture. The second position – the outcome – puts culture more into the role of a summary indicator of the consequences, including for behaviour, of many inputs. The reference to ‘everywhere and nowhere’ intends to get across the message that almost everything that goes on in an institution affects its culture. The reference to ‘nowhere’ is in no way meant to diminish the importance of culture, but rather to be clear that I don’t think there is a distinctive and external ‘thing’ called culture that acts as an input to institutional behaviour. You can’t take institutional culture down from a shelf and seek to change it in some mechanical way. From this, you can tell that I see culture as an outcome more than an input.

When I describe culture as an outcome, you could easily respond by questioning whether that outcome is no more than a positive fact-based description of reality as it is, or whether there is an important normative part to culture – how things ought to be, a reflection of good and bad or right and wrong? The answer is yes, there is an important normative part to what is good culture, including good behaviours.

To illustrate, let me put this into context by drawing on another very closely related debate, namely how do we restore the public’s trust in the financial system in the light of all the problems that have been revealed in the last decade? Good culture and trust arguably go hand in hand – good culture makes it more likely that a firm and its people will be trusted. How then do we define trust? I think trust is most usefully defined as a set of beliefs about the honesty and veracity of commitments, and thus behaviour, looking forward, thereby reinforcing the predictability and reliability of future actions. Those commitments, which can also be described as expressions of intent, are made to customers of all sorts, to investors – creditors and shareholders – and to public authorities in respect of the public interest. To sum up this part of the argument, culture is an outcome of many elements of the behaviour of institutions, but also has an important normative part, namely of what good looks like.

Cultural outcomes are the product of a wide range of contributory forces: the structure and effectiveness of management and governance, including the well-used phrase ‘the tone from the top’; and the incentives they create; the quality and effectiveness of risk management; and the willingness of people throughout the organisation to enthusiastically adopt and adhere to the tone from the top. That adherence is crucial: actions really do speak louder than words and are at the heart of how we judge the intent towards good outcomes and thus culture.

Roles in shaping culture

Next, I want to develop the issue of how two important influences on shaping cultural outcomes actually work. These are the roles of governing bodies and management, and of public authorities.
It is appropriate to start with governing bodies and management as shapers of culture. The governance of firms has for the last forty years been seen largely through the lens of agency theory. One of the most formative articles in this area was published in 1976 by Jensen and Meckling^[1]. They introduced the article with a quotation from Adam Smith’s ‘The Wealth of Nations’, which bears repeating:

‘The directors of such [joint-stock] companies, however, being the managers rather of other people’s money than of their own, it cannot well be expected, that they should watch over it with the same anxious vigilance with which the partners in a private copartnery frequently watch over their own. Like the stewards of a rich man, they are apt to consider attention to small matters as not for their master’s honour, and very easily give themselves a dispensation from having it. Negligence and profusion, therefore, must always prevail, more or less, in the management of the affairs of such a company.’

Suffice to say, Smith cannot be labelled as an optimist on the prospects of good culture in the public company. In their development of agency theory, Jensen and Meckling defined an agency relationship as a contract where a principal engages an agent to perform a service on their behalf which involves delegating some decision making authority to the agent. If both parties are utility maximisers, the agent will not always act in the best interests of the principal. This was Smith’s point. The principal can limit divergence from their interest by creating appropriate incentives for the agent and by incurring monitoring costs to limit aberrant activities. As Jensen and Meckling also pointed out, there are some situations where it will pay the agent to incur some cost to not take action that harms the principal, or to ensure that the principal will be compensated if such actions are taken. They concluded from all of this that there are positive costs attached to ensuring optimal decisions are taken, and that there is an evident risk of divergence between the agent’s (manager’s) decisions and those decisions which would maximise the welfare of the principal (owner). The handling of these principal-agent relationships is an important influence on firm culture. Adam Smith used the term ‘negligence and profusion’. Over the last decade, we have had recourse to use the twenty first century equivalents of this term.

Agency theory has played an important part in shaping approaches for corporate governance. In the UK it led to formative work around twenty five years ago to shape modern corporate governance through the UK Corporate Governance Code produced by the committee chaired by Adrian Cadbury. The essence of the code was to establish the role of the Board of a company as to act in the best interests of the members of the company – thus on behalf of the principal. It follows from this that the role of the Board is to oversee the activities of the agent – management. With the enormous benefit of hindsight, it is noteworthy that while this description of the role of a Board is broad in scope, it was capable of being interpreted quite narrowly in terms of management oversight.

What has changed in light of the passage of twenty five years and a major financial crisis? One recent commentary on UK corporate governance has described the broadening of its role towards ‘everything that companies do’^[2]. What led to this change? Two causes stand out for me. First, the experience of practical application of the principal-agent framework is that it turns out to be more complex than envisaged. I will illustrate this in a moment with the case study on remuneration. Second, the scope of the public interest in the behaviour of firms has expanded substantially in the last twenty five years. Boards are expected to play their part in this responding to these changes.

I will now move on to the role public interest objectives.

Cultural outcomes in firms need to embody respect for public interest objectives, and there is no doubt that the definition of the public interest has expanded in the last twenty five years. 

The expansion of the scope of the role of corporate governance has included a much broader definition of where corporates are expected to act in the public interest. Often, but not always, the expectation of acting in the public interest comes with regulatory oversight. That does not, most emphatically, mean that Boards can outsource securing the achievement of public interest objectives by management (agents) to the regulator.

The job of the regulator is to ensure that, in combination, principal and agent are satisfying public interest objectives set out in legislation. Cultural outcomes in firms need to embody respect for public interest objectives, and there is no doubt that the definition of the public interest has expanded in the last twenty five years. The culture of firms therefore now includes the outcome of more public interest influences than used to be the case. In our world, this is most evident in the responses to the financial crisis, in both its prudential and conduct forms. The emphasis on financial stability, safety and soundness, market integrity, fair treatment of customers has been significantly strengthened post-crisis.

All of these objectives are public goods. The role of regulators cannot be separated from the pursuit of public goods. We can debate about the ways in which regulators pursue their objectives, but we should not in my view call into question these public goods. Unfortunately, history shows that how we pursue the objectives that represent the public goods can compromise the objectives themselves. They are underpinned by broad social norms. If we look at the years before the crisis we can see how these norms changed. There was, for instance, a greater advancement and admiration of strategies which brought success and wealth to banks and bankers before the crisis. The result was pressure for light touch regulation. The change to social norms and the impact on regulation had a decisive impact on the culture in firms and the regulator and compromised achieving the public interest objectives.

The response to the crisis has brought the public interest objectives into much sharper focus. The outcome of this emphasis should, and I believe does, influence the culture of firms. It is also therefore important for regulators to think through the likely consequences of their intended actions to understand the effects.

I now want to move onto the case studies which illustrate how we think about culture in firms and the influences on it.

Case Study: Bankers Remuneration

Bankers remuneration is a good example of three important points in the culture debate. First, there is no doubt that the approach towards remuneration will heavily influence the outcome of culture. Second, it is a good example of a virtual no-go area for public policy before the crisis which has changed substantially in the aftermath. Third, the broader history of social norms and attitudes towards remuneration is another important influence on outcomes and thus culture, but those outcomes will in turn also constantly shape the prevailing social norms.

I am going to start the case study with some history of broader corporate remuneration.

My starting point is the period from the end of the Second World War until the early 1970s. What is striking, viewed from the perspective of knowing what happened next, is the absence of emphasis on pay for performance, indeed the positive rejection of it along with a rejection of excessive executive remuneration (note that pay levels were closely associated with pay incentives). You can tell this just by the titles of some of the articles published at this time: ‘How to Ruin Motivation with Pay’, ‘Wealth Addiction’. But my favourite comes a little later, from the January 1988 edition of INC Magazine, ‘Incentives can be Bad for Business.’^[3] The argument by the way was that intrinsic interest in a task declines when someone is given an external reason for doing it. The author of the book ‘Wealth Addiction’ commented: ‘Getting people to chase money … produces nothing except people chasing money. Using money as a motivator leads to a progressive degradation in the quality of everything produced’.

At the time there was a broad cultural aversion to high pay. Paul Krugman has commented that ‘fear of outrage kept executive salaries in check’, pointing to a social norm^[4]. This social norm may have held, at least in the US, until the mid 1980’s. In 1984, the New York Times published an article with the title of ‘The Age of Me–First Management’ which noted concerns that top executives were ‘losing sight of moral standards in the new frenzy to get rich’.^[5]

A second point that stands out is a recognition, at least by some commentators, that if you incentivise someone to do something, they will tend to do just that.^[6]

And, third, this was a period in which there was no significant relationship between risk taking (as measured) and remuneration.^[7]

What happened next was quite different. Across corporate life, notably in the US, there was a rapid growth in the use of stock option plans to create a stronger link between pay and performance, or so it was presumed. In many banks, there was strong growth in the use of equity-based pay incentives, and studies of the US show that for banks this trend grew post de-regulation and was more pronounced in those with greater non-interest income. The consequence was that CEO wealth became more sensitive to change in the volatility of their bank’s share price and the evidence tends to suggest that those with higher option exposures chose riskier policies and that the distribution of options was highly coordinated to bank size.^[8] This set up the role of remuneration as a cause of the crisis.

The authors of the previous period would have predicted this in one important respect – they saw that when incentives are created, they are acted upon. Adam Smith saw this too. It was re-stated in 1972, by Alchian and Demsetz in their well-known article on economic organisation: ‘The classic relationship in economics that runs from marginal productivity to the distribution of income implicitly assumes the existence of an organization, be it the market or the firm, that allocates rewards to resources in accord with their productivity’.^[9] Two big questions arise from this: how is productivity measured in this context, and how do incentives affect it?

From some time in the late 1980s, the returns to management increased rapidly. The old social norms fell away. Agency theory would suggest that Boards should design compensation schemes to provide managers with incentives to maximise shareholder value. Remuneration can be a means to tackle the agency problem, but in doing so, the changes seen from the mid 1980s allowed managers to extract larger rents. It became part of the problem, and it had a distinct impact on the culture of firms. And, it also coincided with (I am being careful here not to make a casual statement) a change in broader social norms. It was the ‘Greed is Good’ era.

The trend towards managers extract rents begs the question of how to measure productivity. This is a crucial issue in many financial services including banks. The more uncertain the environment, the harder it is to measure the output. Why? This depends to what extent the output and returns are realised almost instantaneously, and to what extent they emerge over often long periods of time, consistent with the length of contracts. Older theory suggests that output-based remuneration is more likely to be observed where managers have more discretion in what they produce. This means that the returns to managers are not independent of the underlying riskiness of the environment.^[10] Managers have information not available readily to their Boards, and this was much more the case before the financial crisis when the role of Boards was less well developed. With financial services, the added component is that measuring output is much more difficult because of the time dimension – the length of time it takes to see the results. But a ‘solution’ was found by accounting standards, namely increasingly to book an estimated economic profit ahead of the contracts unfolding. The appeal to managers was obvious and the resulting risk looks obvious at least from today’s vantage point. There is to my mind no doubt that it influenced the culture of firms. What happened next is, as they say, history.

As financial regulators, we do not seek to control the level of pay, outside its impact on our public policy objectives.  But, the influence that we do have will affect the culture of firms.

What are we doing about it? Learning the lessons of history. A big one is the old one that incentives tend to work, but it is crucial to understand how they work. This one keeps coming up in most of the big problems, both prudential and conduct related, in our world. It is critical that Boards and regulators think through the consequences of structures and incentives. In bank remuneration in the UK we have emphasised the importance of deferring variable remuneration consistent with the observation that the risks and returns of activities evolve over a considerable time. And, during that time of deferral and after it, variable remuneration can be cancelled where problems or poor performance materialise. The objective here is to align structures with incentives, and create the culture that people have skin in the game. We are also acting to stop people evading this discipline by moving jobs and cashing in deferred remuneration – the problem known as rolling bad apples.

Second, we have used the regulatory system to establish clearly that variable remuneration cannot be paid if a firm does not have adequate capital resources. This is an obvious point you might say. Well, not so obvious that it was followed in the past.

In all of this, the approach is not to cap the level of remuneration, but rather to act on the structure of it and the incentives created. As financial regulators, we do not seek to control the level of pay, outside its impact on our public policy objectives. But, the influence that we do have will affect the culture of firms. That is intended.

Case study: Governance and Responsibility

My second case study is on Governance and Responsibility.

In the UK, one of the strong responses to the financial crisis, not least from our Parliament, was to ask who was responsible. In the case of firms, the critical question was why responsibility did not, in the formal sense of regulatory action, sit with senior managers, and in particular CEOs. This was sometimes portrayed as blood lust or a witch hunt, but in fact it was asking the reasonable question, ‘who is responsible and thus will be accountable for what happens’. Responsibility is a simple concept in many ways – parents teach it to their children at a young age – but in a more formal setting it has been elusive. In our regulatory regime for financial services it has been conflated with culpability. There is an important difference. Culpability is more about direct blame for an action whereas responsibility is about leadership. But direct culpability in banks during the crisis in terms of who took direct action that caused problems was often with others lower down. And with the absence of focus on responsibility, those at the top were seen to evade the consequences in a formal sense of regulatory action at least. Accountability is also a simple concept which sits closely with responsibility. But in the absence of the latter, accountability was clouded and the consequence was an obscuring of both key concepts.

In the UK, following the recommendation of the Parliamentary Commission on Banking Standards, we have introduced for banks and insurers, and will do so shortly for other financial services firms, the new Senior Managers and Certification Regime. The purpose of this is to embed responsibility and accountability. Senior managers should know what they are responsible for, as should key Board members and the map of responsibilities should go right across the firm. These individuals are approved by the regulators as fit and proper to carry out the responsibilities of their role.

The Certification element of the regime applies to more junior staff who are nonetheless risk takers in the full range of meaning of the terms. The process of Certification is overseen by senior managers in firms. It is therefore their responsibility to ensure it works, and the regulators take a keen interest in the working of this regime.

The key feature of all this for me is the simplicity of the concept of responsibility and accountability, and the powerful effects that comes from such simplicity. The effect on culture should be profound, and I say that in a good sense.

Our role as a Regulator

Culture in firms is an outcome of many inputs and the incentives that are created around those inputs. I have put more emphasis on what I might call more structural determinants of those incentives rather than adopting a more purely behavioural approach. Culture is characterised by a pattern of behaviours no doubt, but if we are to understand its causes, I think we need to get back to more structural features in, for instance, governance and risk taking.

At the FCA, we use a range of supervisory tools and methods to work with firms on issues relating to their culture, such as the firm’s stated purpose, ‘tone from the top’, incentive structures and the effectiveness of management and governance.

A firm’s stated purpose is an explanation of what a firm is trying to achieve – the definition of what constitutes success for a firm.

The tone from the top, or how the leadership (the Board, Executive, parent companies) of a firm behave and fulfil their responsibilities, drives the behaviour of staff and the outcomes they deliver. In turn, behaviour throughout the firm and outcomes also form a measure of the effectiveness of the tone from the top.

Incentive structures also drive the behaviour of staff, along with other people-related practices, such as recruitment and performance management. This is where tone from the top gets turned into real practice.
Finally, governance is the framework of responsibility that oversees the operations of a firm. It is essential that the leadership of firms identify what drives their culture.

For the FCA’s part as the UK’s financial conduct regulator, we seek to form judgements as to whether the inputs are producing appropriate culture and outcomes. For example, a question we seek to answer is whether practices such as recruitment, performance management, reward and capability drive positive behaviours and create a culture that works in the long term interests of the firm, its customers and market integrity.

A firm’s culture emerges in large part from inputs that are its responsibility. There is no such thing as an ideal organisational culture and, as a firm’s culture is influenced by many different factors, we cannot prescribe what it should be; nor do we believe that every firm should have the same culture.

It is for firms to ensure that their desired culture is consistent with appropriate conduct outcomes, to identify the drivers of behaviour within the firm and control the risks that these drivers create.

As a regulator we want to be transparent about the way we look at firms and share our expectations and view on culture so our stakeholders understand our approach and how it fits with our objectives.

Culture change in itself is a challenge, and we know it takes time. This is because culture comes from the past and is embedded in the legacy of a business. The culture of a firm is developed and reinforced over years, even decades, and is passed down from one generation to the next.

As a result, culture can be remarkably resilient in the face of attempts to change it. However, if culture is ignored then an opportunity is lost to tackle one of the major root causes of conduct failures. The answer is not to try to tackle the culture, but to act on the many things that determine it, of which governance and remuneration are important.